APPARATUS AND METHOD FOR DELIVERY OF VAPOR

FIELD OF THE INVENTION

[0001] A vaporizing device and method for operating same is provided for control, delivery, and/or purification of a vapor of a liquid, e.g., for use in microelectronics and other critical process applications.

BACKGROUND OF THE INVENTION

[0002] Vapors of water and other liquids are used in many applications in current advanced technology processes, including processes employed in microelectronics manufacturing, as well as in clean room and medical applications. Desirably, such vapors should introduce minimal impurities into a process, such that they can be used, e.g., for the oxidation of silicon, in the production of thin gate oxides (e.g., in the electronics and micro-electronics industries), and in ultra-high purity cleaning processes. However, delivery of liquid vapor (such as, e.g., water, alcohols, amines, organic compounds such as organometallic compounds or semiconductor compounds, or the like) to a process using conventional methods is difficult, and the choices limited to direct liquid injection (direct liquid injection) or bubblers, each of which tends to introduce contaminants along with liquid vapor, or, in the case of bubblers, may deliver microdroplets in an irregular fashion.

[0003] In the medical industries, water vapor can be used for humidification. Such water vapor when produced using conventional systems can yield a product having, e.g., prions, viruses, allergens, proteins, bacteria, and other biologically active macromolecules or substances present. Additionally, inorganic substances, such as borates and silicates or metallic substances such as iron, nickel, chromium, copper, and other toxic metals can be present.

[0004] Typically, water for micro-electronics applications is produced by reacting gaseous hydrogen and oxygen to yield water vapor. The production of pure water vapor is practically impossible due to the presence of residual oxygen and/or hydrogen remaining in the product water vapor. Removing these components often requires additional expensive and complex separation processes. Additionally, high concentrations of gaseous hydrogen are often required for the synthesis reaction with oxygen, which is conducted at high temperatures well above the explosive limit of hydrogen (approximately 8 % at a pressure of approximately 100 kPa).

[0005] The simple boiling of high purity de- ionized water to yield water vapor can avoid the problems and dangers inherent in the direct reaction of hydrogen and oxygen to yield steam. However, removing dissolved gases can be difficult and often requires multiple boiling/condensation cycles in a hermetically sealed environment, which can be expensive. Moreover, aerosols containing materials that are not normally volatile, such as salts or metals, can be produced during the boiling process and can add unwanted impurities.

[0006] For delivery, e.g., in semiconductor processing, of precursor gases, such as phosphine (PH ₃ ), hydrogen sulfide (H ₂ S), and hydrogen selenide ( bSe) the conventional delivery method is to use a cylinder containing the correct composition of the precursor(s) in conjunction with a thermal mass flow controller (MFC), to control the vapor phase concentration. The main criteria for successful and reproducible use are that the inlet pressure is greater than ca. 50 torr, the operating temperature at the MFC is less than 60°C, and the flow rate is greater than 0.1 seem. The vapor can also originate from a liquid source as long as these conditions are met. Special liquid mass flow controllers (LMFC) are available, for the direct control of liquid delivery. The liquid is fed from the LMFC to a vaporizer to produce a vapor of the precursor at the point of use. A further variation in MFC design is the pressure-based MFC, used to deliver precursors from low-pressure sources. Examples of their use include the control of tetraethoxy silane and hexamethyl disilazane for inter-level dielectrics, parylene for thermal polymerization, and metal-organic reactants for metal organic chemical vapor deposition (MOCVD) of Π-VI and ΠΙ-V semiconductor compounds.

[0007] As noted above, the delivery of liquid precursors has been dominated by the use of so-called "bubblers". Typically, these are passivated stainless steel Dreschler bottles containing the precursor, ideally a liquid with a significant vapor pressure. The precursor is delivered to the CVD reactor by entrainment with a carrier gas that is fed through the liquid and should become fully saturated with the precursor on passing through the bubbler. There are few problems with the use of modestly volatile liquids, i.e., those with equilibrium vapor pressures roughly in the range of 5-30 torr at room temperature. There are, however, problems when using materials with very low or high vapor pressures, and with the use of solids. To overcome these difficulties, liquids with low vapor pressures can be heated, using oil baths or thermal tape wrapped around the bubbler to increase the equilibrium vapor pressure within the bubbler. However, there are two problems associated with this technique. Firstly, to prevent condensation of the precursor in other parts of the CVD system, it may become necessary to heat the entire network of feed lines up-stream of the substrate. Secondly, for delivery to be successful, the precursors must have long-term stability at the elevated temperature used. Precursors with high vapor pressures can be difficult to control and in such cases it may be necessary to cool the bubbler. For highly volatile precursors, such as dimethyl zinc, a slight temperature fluctuation can have a significant effect on the gas phase concentration. Most solid precursors have very low volatilities and thus produce similar problems to those of low volatile liquid precursors. Additionally, the surface area of solids in contact with the entraining gas may be insufficient for equilibrium to be established on the passage of the gas. As the solid sublimes, there will be a variation in the surface area leading to a variation in the rate of mass transport over time, so it is more difficult for equilibrium to be established under the dynamic gas flow within a bubbler. One approach which has been used to overcome this problem, notably with trimethyl indium, is to use two bubblers in series. This method allows more contact between the carrier gas and the precursor so that the carrier gas has sufficient contact time to become fully saturated and equilibrium established before exiting the second bubbler.

[0008] The problem of the carrier gas not becoming fully saturated before exiting the bubbler is common when the amount of precursor remaining in the bubbler is low. One approach to this problem, for liquid precursors, is to use a bubbler with a self- metering reservoir (SMR). Because of real-time replenishment of the precursor, the volume of precursor stays at a constant level within the bubbler, and so at constant gas flow and temperature, the vapor phase concentration of precursor should also remain constant. When using this conventional approach, only volatile compounds can be considered as precursors. Although these compounds are volatile they do not necessarily lead to clean deposition.

[0009] Most of the current systems in use for delivering precursors or other substances can be classified within the categories of liquid injection systems (LISs) where a precursor is vaporized directly from a solution, solid delivery systems (SDSs) where the precursor is vaporized directly from a solid, or aerosol-assisted delivery systems (AADSs) where an aerosol containing the precursor is formed prior to delivery. Direct Liquid Injection-MOCVD (DLI-MOCVD) can be used for liquid precursors or a solution containing precursors. The solution is transported, most often via a syringe, to a vaporization chamber adjacent to the CVD reactor from where it is swept, using a carrier gas, into the reactor. It is the speed at which the syringe is depressed that determines the gas phase concentration of precursor. The precursor is maintained at room temperature until just before use, thus reducing the potential for premature decomposition. DLI has been used to grow thin films of single metal oxides of nickel, zinc, and iron along with the binary ferrites (nickel ferrite, zinc fer- rite), and the ternary nickel zinc ferrite, from the (β-diketonate precursors, Ni(tmhd)2, Zn(tmhd) ₂ , and Fe(tmhd) ₃ , dissolved in THF. The CVD of lead-scandium-tantalate has also been grown using DLI from Pb(tmhd) ₂ , Sc(tmhd) ₃ , and Ta(OEt) ₂ , wherein tmhd is 2,2,6,6,-tetramethylheptane-3,5-dione and Et is ethyl.

[0010] One type of DLI system consists of a high precision two-stage syringe-type liquid pump, a vaporizer with variable temperature zones made up from an array of stacked thin metal disks, and an electronic control unit to regulate the operation of the pump and temperature of the vaporizer. A pressurized liquid precursor is fed into the multistage pump and pumped at high pressure into the evaporator/vaporizer, thus a constant flux of material can be delivered without the use of a solvent. However, as in a conventional bubbler, the precursor must be held above its melting point; thus, in some cases, there are problems with decomposition of the precursors prior to delivery.

[0011] Another liquid delivery system consists of a large reservoir containing the precursor(s) dissolved in a suitable solvent. The mixture is vaporized by a flash evaporator matrix. This system allows a constant flux of precursor to be delivered, but only if all the components within the precursor mixture are totally soluble and remain fully miscible. Provided no adverse pre -reactions occur between the precursors and/or the chosen solvent, this system is useful in that it eliminates the need to keep the solution hot, so problems of thermal decomposition are minimized and heating of the feed lines becomes unnecessary. A number of disadvantages are associated with this system. Namely, there is a potential for reactions between precursors in solution, e.g., the hydrolysis of an alkoxide, so both solvent and precursor need to be of high purity. If there is more than one precursor in the solution, any reactions in the source solution between the precursors or solvent can result in the formation of new compounds with different physical properties, which could adversely affect the behavior of the precursors and their delivery. There are also potential problems with the high volume of carrier solvent utilized, which may cause a build-up of flammable vapor within a CVD reactor.

[0012] Other liquid delivery systems (LDS) include pulsed liquid-injection CVD whereby micro amounts of a precursor dissolved in a solvent are sequentially injected into an evaporator where flash volatilization occurs. The amount of precursor to be deposited is dependent on the volume of precursor evaporated at any given time and this is, in turn, dependent on the length of each injection cycle or "pulse width". Using this method, CVD growth can be controlled by a computer.

[0013] There is a similar, modified LDS employing a positive displacement pump, again designed to improve accuracy in terms of the amount of precursor delivered at a given time. In this method, multiple precursors are mixed just prior to evaporation. A computer-controlled amount of each precursor solution is delivered into the low-pressure side of a dual-piston pump. When the low-pressure valve of the pump is filled with the required amount of each precursor (as controlled by the computer), the mixed solution is transferred to the high-pressure side where delivery into a vaporizer occurs.

[0014] The widely used aerosol-assisted (AA) CVD has many advantages similar to those of the newer liquid injection systems. Essentially, as in liquid delivery, the method is based on flash evaporation but in this case of an aerosol. A sweep of carrier gas is used to transfer the aerosol of a precursor-containing solution into the hot zone of a reactor. Again, this method has been designed for compounds with low volatility and thermal stability. The precursor(s) are dissolved in a suitable solvent and the solution is atomized or vaporized into a carrier gas stream. A method of producing an aerosol is to use an ultrasonic technique whereby the precursor-containing solution is placed within a vessel fitted with a piezoelectric transducer. The advantages and disadvantages of the method are essentially similar to those discussed above for liquid injection. However, when using this method, it is also important that the rate of transfer of the aerosol to the evaporator remains constant, and that the aerosol is of constant composition. One example of the use of AACVD is in the growth of the superconductor YBCO, from the Ba, Y, and Cu precursors, Ba(tmhd)2, Y(tmhd)3, and Cu(tmhd)2, which are dissolved together in a suitable solvent (butylacetate, THF, toluene, decane, or supercritical CO ₂ ) prior to use. [0015] Solid delivery systems have also been developed that allow for the delivery of solid state precursors, which may easily decompose if heated for prolonged periods. The precursors are essentially flash- sublimed into the growth chamber of a CVD reactor. The superconductor YBCuO has also been prepared from Ba(tmhd) ₂ , Y(tmhd) ₃ , and Cu(tmhd) ₂ by this method. However, unlike AACVD, the precursors are used as solids. They are mixed and placed in a Pyrex tube which is then passed through a large temperature gradient (25-300°C) over a very small distance (2 mm). This initiates vaporization of the precursors, and they are subsequently swept to the substrate by a carrier gas.

[0016] A method combining the principles of both liquid and solid state delivery has been developed whereby droplets of precursor-containing solutions are sequentially injected onto an inert porous conveyer belt. The solvent is removed at room temperature by evaporation into a flowing gas stream (away from the substrate), which is removed from the system via a cold trap, The as-deposited solid is transported, on the porous conveyer belt, into the hot evaporation zone where rapid flash volatilization into an appropriate carrier gas occurs, This then transports the precursor to the substrate.

SUMMARY OF THE INVENTION

[0017] A method of producing liquid vapor without the need for a carrier gas is desirable. Such liquid vapor can contain reduced levels of impurities, and can be delivered in a controlled manner (e.g., at a specific flow rate, or in a specific amount) over a variety of conditions is desirable. The vaporizer systems, devices, and methods of the preferred embodiments achieve these goals.

[0018] In a first aspect, a vaporizer device configured for delivery of a vapor into a vacuum in an absence of a carrier gas is provided, the device comprising: a vacuum chamber; and a membrane assembly comprising a membrane in fluid communication with a vacuum chamber on a first side of the membrane and a chamber configured to contain a source liquid on a second side of the membrane, wherein the membrane is configured to permit passage therethrough of the source liquid, whereby the source liquid passes through the membrane assembly to yield a source vapor in the vacuum chamber, and wherein the membrane assembly is provided with a heater, a temperature sensor configured for measurement of a temperature of the source liquid in the chamber, and a sensor for measuring current and voltage in the heater, wherein the vaporizer device is configured such that a temperature at which the membrane assembly operates is calibrated using theoretical mass transfer information.

[0019] In an embodiment of the first aspect, the vaporizer device further comprises a control loop which employs data from the temperature sensor and the sensor for measuring current and voltage in the heater to adjust the power to the heater, with temperature correction via power adjustment provided by a comparison of actual power consumption to calculated power consumption providing a correction to a temperature set point determined by an initial calibration of the system.

[0020] In an embodiment of the first aspect, the vaporizer device further comprises a pressure transducer for measuring pressure in the vacuum chamber.

[0021] In an embodiment of the first aspect, the membrane assembly comprises an ionic membrane.

[0022] In an embodiment of the first aspect, the membrane assembly comprises an ion exchange membrane comprising a perfluorinated ionomer.

[0023] In an embodiment of the first aspect, the membrane assembly comprises a fluoropolymer membrane containing sulfonic acid groups.

[0024] In an embodiment of the first aspect, the vaporizer device is configured to generate water vapor, and wherein the membrane assembly comprises a substantially gas -impermeable membrane having a leak rate of gases other than water vapor of less than about 10 - ^" 3 cm 3 /cm 2 /s under standard atmosphere and pressure.

[0025] In an embodiment of the first aspect, the membrane assembly comprises a substantially gas -impermeable membrane that is more permeable to water vapor than to other gases by a ratio of at least about 100:1.

[0026] In an embodiment of the first aspect, the temperature sensor is selected from the group consisting of a thermocouple, a thermistor, a resistive thermal device, and an optical temperature sensor.

[0027] In an embodiment of the first aspect, the vaporizer device is configured to generate an organometallic compound vapor.

[0028] In an embodiment of the first aspect, the vaporizer is configured to deliver source vapor at a controlled rate of about 1 micrograms per minute to about 10 grams per minute.

[0029] In an embodiment of the first aspect, the membrane assembly comprises a polymeric membrane, wherein the polymer is selected from the group consisting of rubbery membranes, glassy membranes, nitrile rubber, neoprene, silicone rubbers, polydimethylsiloxane, chloro sulfonated polyethylene, polysilicone-carbonate copolymers, fluoroelastomers, fluorinated ethylene propylene copolymers, chloro sulfonated polyethylene, polylvinylchloride, polyurethane, cis-polybutadiene, cis- polyisoprene, poly(l-butene) polystyrene-butadiene copolymer, styrene/ethylene/butylenes block copolymers, styrene/butadiene/styrene block copolymers, polymers of perfluoro-2,2-dimethyl-l,3-dioxole, and substituted polyacetylenes of the general structural formula :

wherein Ri and R ₂ are independently selected from the group consisting of hydrogen, halogen, C ₆ H5, linear alkyl, and branched alkyl group.

[0030] In a second aspect, amethod for growing a semiconductor film is provided, comprising: providing an organometallic vapor from the vaporizer device of the first aspect; and depositing a semiconductor film from the organometallic vapor onto a substrate.

BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 provides a schematic depiction of a vaporizer device of a preferred embodiment.

[0032] Figure 2 provides a schematic depiction of a feedback loop employed with a vaporizer device of a preferred embodiment.

[0033] Figure 3 provides a graph of vapor pressure and flow rate as a function of temperature for water.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

[0034] A pervaporation device is provided for the control, delivery, and/or purification of vapors of liquids, including but not limited to water, isopropanol (also referred to as isopropyl alcohol), other common solvents (e.g., acetone, methanol, ethanol, N-methyl-2-pyrrolidone, ethylene glycol), organometallic compounds, semiconductor compounds (e.g., triethyl gallium, trimethyl gallium, trimethyl indium, solution trimethyl indium, trimethyl aluminum, diethyl zinc, tetraethyl tin, biscyclopentadienyl magnesium, biscyclopentadienyl magnesium, trimethyl antimony, dimethyl aluminum hydride), small inorganic dopant molecules (e.g., trimethyl boron, trialkyl boron, trialkyl phosphorous), compounds that are liquids at room temperature, and other substances that are vapors or can be converted to vapors. The vapors can be delivered without the need for a carrier gas stream. The apparatus and methods of preferred embodiments are particularly useful in micro-electronics applications. The vaporizer enables the controlled transfer and/or purification of water, alcohols, organometallic compounds, semiconductors, semiconductor compounds, and/or mixtures thereof directly into a vacuum via a non- porous membrane. The vapor to be transferred is referred to as the "source." By utilizing information such as temperature, energy consumption, membrane thickness, membrane surface area, and diffusion rate across the membrane, calibration standards can be determined and vapor at a particular flow rate (or in a particular amount) can be precisely delivered.

[0035] The systems of preferred embodiments utilize a non-porous membrane to provide a barrier between the liquid source and the vacuum chamber. The molecules from the liquid source permeate across the membrane and into the vacuum chamber. The source is diffusion driven by a concentration gradient between the liquid and the vacuum chamber. Delivery of water vapor and other low vapor pressure materials across polymeric membranes can difficult when relying only upon pressure and temperature control. The methods of preferred embodiments achieve controlled vapor delivery through the use of calculations of the theoretical amount of energy consumed for mass transfer of the vapor. This data is subsequently utilized as feedback input to temperature and pressure control components within the vapor delivery system. Calibration data is acquired by means of a vapor pressure measurement apparatus (see, e.g., Figure 1) to obtain known parameters for mass transport across a given membrane. Temperature requirements can then be developed for specific vapor materials with respect to vapor mass transport across a defined membrane thickness and surface area. The mass transport and temperature data can be used to calculate the theoretical energy requirements for precise mass delivery and subsequently used as a calibration standard. The actual energy used in a given system can then be measured and used as feedback input with reference to known calibration standards for that system. Consequently, more precise input parameters for temperature and pressure control may be utilized and adjusted with regard to a particular system and individual vapor material requirements.

[0036] The membrane employed in the vaporizer of preferred embodiments is preferably designed specifically to select only the source molecules - other contaminants in the liquid source preferably cannot permeate across the membrane and cannot then enter into the vacuum. The membrane excludes particles, micro-droplets, volatile gases, and other opposite-charged species. For example, a hydrophilic membrane can be employed to exclude hydrophobic gases, and a hydrophobic membrane can be employed to exclude hydrophilic gases.

[0037] In a preferred embodiment, the device includes the following basic components: a vaporizer; and a vacuum chamber. The vaporizer typically includes, e.g., a conduit or passageway to a container or other source of source liquid, a membrane assembly connected to the source of source liquid, heating and/or cooling unit(s) to control the temperature of the membrane assembly, and a temperature sensor (e.g., a thermocouple, RTB device, or thermistor) to measure the temperature in the membrane assembly. The vaporizer can optionally include a level detection device, pump, or other device for monitoring and/or controlling the amount of source liquid in the source container. The vaporizer can also optionally include a pressure transducer, humidity detector or temperature transducer for feedback to any temperature control devices. The vacuum chamber typically employs a temperature sensor and pressure transducer, as well as a downstream vacuum pump and vent. Valves can be advantageously employed throughout the system. Figure 1 provides a schematic depiction of a vaporizer system of a preferred embodiment.

[0038] In a particularly preferred embodiment, the vaporizer device is configured for the transfer of water vapor. In this embodiment, a fluoropolymer membrane which contains sulfonic acid groups is preferably employed. Such membranes are commercially available under the tradename of NAFION® by E. I. du Pont de Nemours and Company, Wilmington, DE. Such membranes allow for the rapid transfer of water vapor into a vacuum. The system can also be used for delivery of other vapors, e.g., alcohols such as methanol, ethanol, isopropanol, etc., semiconductor liquids (e.g., trimethyl gallium, diethyl zinc, trimethyl indium and other such liquids as are enumerated in O'Brien et al., Chem. Vap. Deposition 2002, 8, No. 6, pp. 237-249, the contents of which are hereby incorporated by reference in its entirety and are hereby made a portion of this specification). For delivery of organometallic compounds, it is generally preferred to employ a hydrophobic membrane, e.g., a polyolefinic membrane.

[0039] The devices and systems of the preferred embodiments offer numerous advantages over prior art systems. Delivery of water, isopropanol to a process using conventional methods is difficult, and the choices are limited to, e.g., direct liquid injection or bubblers. A description of conventional methods for delivery of vapors is provided in O'Brien et al., Chem. Vap. Deposition 2002, 8, No. 6, pp. 237-249, the contents of which are hereby incorporated by reference in its entirety and are hereby made a portion of this specification. Direct liquid injection is costly and problematic for use with different flow rates. At low flow rates control has limited accuracy, and at high flow rates direct liquid injection is susceptible to bubbles in the liquid which generate erratic flow rate values. Direct liquid injection utilizes a metallic vaporizer or a metal hot plate to convert the liquid to gas, and can vaporize only limited quantities of liquid due to thermal transfer rates. Other disadvantages to direct liquid injection are that contamination can build up on the hot plate, and there is a potential for chemical decomposition of certain liquids. The greatest disadvantage of direct liquid injection, however, is that it cannot provide any purification of the liquid being vaporized, since everything in the liquid, including contaminants and impurities, is vaporized and introduced into the process. In addition, the hot metal vaporization surface may contribute metal contaminants to the vapor stream.

[0040] Bubblers have the advantage of low cost. However, bubblers are inaccurate and imprecise, due to poor control of the temperature of the gas, the temperature of the liquid, the operating pressure, the liquid level, and thermal droop. Bubblers offer somewhat better performance than direct liquid injection as to entrainment of contaminants, since bubblers leave behind some contamination during the vaporization process, but bubblers cannot prevent entrainment of dissolved gases, volatile molecular contaminants, and micro-droplets which can carry particulate and ionic molecular contaminants.

[0041] Conventional membrane contactors can also be used to permit gas transfer between a liquid and gas; however, such contactors suffer from certain disadvantages. Conventional membrane contactors are typically constructed from hollow fiber membranes which are porous. This configuration allows for the simultaneous transfer of the gas into the liquid and the liquid into the gas. Most hollow fiber membranes are hydrophobic and thus must be modified so as to be suitable for use with hydrophilic molecules. Some hollow fibers can be rendered hydrophilic by chemical modification, but in the case of surface treatments, such modification may only be partially effective, and may wet out to allow undesirable direct liquid transfer across the membrane. Because such membranes are porous, they have limited ability to provide purification or microdroplet permeation. Conventional membrane contactors are not specific to which gases can permeate, and thus have no purification capability.

[0042] In contrast to the prior art methods, the vaporizer devices of preferred embodiments permit purified vapor to be provided in a flow rate-controlled, mass- controlled, and/or volume-controlled fashion, into a vacuum (or, in certain embodiments into a carrier gas). Flow rates of from about 1 μg/min or less to about 1 g/min, 2 g/min, 3 g/min, 4 g/min, 5 g/min, 6 g/min, 7 g/min, 8 g/min, 9 g/min, 10 g/min, 20 g/min, 30 g/min, 40 g/min, 50 g/min, or 100 g/min or more can be achieved using the systems of preferred embodiments The non-porous membrane prevents dissolved gases, most volatile contaminants, particles, and microdroplets from being transferred, such that the vapor product is more consistent in composition and significantly more pure.

The Membrane

[0043] In a preferred embodiment, water or other vapor is introduced into a vacuum through a substantially gas -impermeable ionic exchange membrane. The term "substantially gas -impermeable membrane" as used herein is a broad term, and is to be given its ordinary and customary meaning to a person of ordinary skill in the art (and is not to be limited to a special or customized meaning), and refers without limitation to a membrane that is permeable to water vapor or another vapor to be provided, but relatively impermeable to other gases such as, but not limited to, hydrogen, nitrogen, oxygen, carbon monoxide, carbon dioxide, hydrogen sulfide, hydrocarbons (e.g. , ethylene), volatile acids and bases, refractory compounds, and volatile organic compounds. Gas impermeability can be determined by the "leak rate" of the membrane. The term "leak rate" as used herein is a broad term, and is to be given its ordinary and customary meaning to a person of ordinary skill in the art (and is not to be limited to a special or customized meaning), and refers without limitation to the volume of a particular gas that penetrates the membrane surface area per unit of time. For example, a substantially gas- impermeable membrane could have a low leak rate of gases other than water vapor, such

-3 3 2

as a leak rate of less than about 10 ^" cm /cm /s under standard atmosphere and pressure. Alternatively, a "substantially gas -impermeable" membrane can be identified by a ratio of the permeability of water vapor compared to the permeability of other gases. Preferably, the substantially gas-impermeable membrane is more permeable to water vapor than to other gases by a ratio of at least about 1000:1, such as a ratio of at least about 2,000:1, 3,000:1, 4,000:1, 5,000:1, 6,000:1, 7,000:1, 8,000:1, 9,000:1 or a ratio of at least about 10,000:1, 20,000:1, 30,000:1, 40,000:1, 50,000:1, 60,000:1, 70,000:1, 80,000:1, 90,000:1 or even a ratio of at least about 100,000:1. However, in other embodiments, other ratios that are less than 1,000:1 can be acceptable, for example, 1.5:1, 2:1, 3:1, 4:1, 5:1, 6:1, 7:1, 8:1, 9:1, 10:1; 50:1, 100:1, 500:1, 1,000:1, or 5,000:1 or more.

[0044] The term "ion exchange membrane" as used herein is a broad term, and is to be given its ordinary and customary meaning to a person of ordinary skill in the art (and is not to be limited to a special or customized meaning), and refers without limitation to a membrane comprising chemical groups capable of combining with ions or exchanging ions between the membrane and an external substance. Such chemical groups include, but are not limited to, sulfonic acid, carboxylic acid, phosphoric acid, phosphinic acid, arsenic groups, selenic groups, phenol groups, and salts thereof. The chemical groups can be in a salt form or an acid form where the cations or protons are exchangeable with other cations from an external source, e.g., a solution or gas. Ion exchange membranes can be provided in acid form and converted to salt forms by pretreating the membrane with a base, such as an alkali metal base, e.g., sodium hydroxide, potassium hydroxide, sodium carbonate, sodium hydrogen carbonate, potassium carbonate or potassium hydrogen carbonate lithium hydroxide, or an alkaline earth metal bases, e.g., calcium hydroxide, calcium oxide, magnesium hydroxide or magnesium carbonate.

[0045] In one embodiment, the ion exchange membrane is a resin, such as a polymer containing exchangeable ions. Preferably, the ion exchange membrane is a fluorine-containing polymer, e.g., polyvinylidenefluoride, polytetrafluoroethylene (PTFE), ethylene tetrafluoride-propylene hexafluoride copolymers (FEP), ethylene tetrafluoride- perfluoroalkoxyethylene copolymers (PFE), polychlorotrifluoroethylene (PCTFE), ethylene tetrafluorideethylene copolymers (ETFE), polyvinylidene fluoride, polyvinyl fluoride, vinylidene fluoride-trifluorinated ethylene chloride copolymers, vinylidene fluoride-propylene hexafluoride copolymers, vinylidene fluoridepropylene hexafluoride- ethylene tetrafluoride terpolymers, ethylene tetrafluoride-propylene rubber, and f uorinated thermoplastic elastomers. Alternatively, the resin comprises a composite or a mixture of polymers, or a mixture of polymers and other components, to provide a contiguous membrane material. In certain embodiments, the membrane material can comprise two or more layers. The different layers can have the same or different properties, e.g. , chemical composition, porosity, permeability, thickness, and the like. In certain embodiments, it can also be desirable to employ a layer (e.g. , a membrane) that provides support to the filtration membrane, or possesses some other desirable property.

[0046] The ion exchange membrane is preferably a perfluorinated ionomer comprising a copolymer of ethylene and a vinyl monomer containing an acid group or salts thereof. Exemplary perfluorinated ionomers include, but are not limited to, perfluoro sulfonic acid/tetraafluoroethylene copolymers ("PFSA-TFE copolymer") and perfluorocarboxylic acid/tetrafluoroethylene copolymer ("PFCA-TFE copolymer"). These membranes are commercially available under the tradenames NAFION® (E.I. du Pont de Nemours & Company), FLEMION® (Asahi Glass Company, Ltd.), and ACIPLEX® (Asahi Chemical Industry Company).

■(CF ₂ - CF ₂ }x - (CF2 - CF) _y -

O - [C F ₂ - CF - 0]m -C F ₂ - CF ₂ - SO3H CF ₃

-TFE- Vinyl Ether

Tetrafl uoroethylene

Pe rfl u 0 ro (4 /- M ethy 1-3 , 6-D i coca- 7- Octene-1 Sulfonic Acid)

PFSA-TFE Copolymer in the Hydrolyzed Sulfonic Acid Form

[0047] A PFSA-TFE copolymer contains a tetrafluoroethylene (TFE) "backbone" to which perfluoro sulfonic acid (perfluoro(4-methyl-3,6-dioxa-7-octene-l- sulfonic acid)) groups are attached. There can be one, two, three, four, five, or six perfluoro sulfonic acid groups for every six TFE backbone units. Any suitable molecular weight polymer can be employed. Preferably, a polymer having a molecular weight from about 500 MW or less to about 2000 MW or more or more is employed. The molecular weight can also be from about 600, 700, 800, or 900 MW to about 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, or 1900 MW. The molecular weight can even be from about 910, 920, 930, 940, 950, 960, 970, 980, or 990 MW to about 1000, 1010, 1020, 1030, 1040, 1050, 1060, 1070, 1080 or 1090 MW. A single copolymer can be employed, or combinations of two or more different copolymers (e.g., varying in chemical composition, molecular weight, or other property) can be employed. A copolymer having a single peak molecular weight is generally preferred; however, in certain embodiments it can be preferred to employ a polymer with a bimodal or multimodal molecular weight distribution, with varying amounts of polymer chains having different molecular weights. The copolymers can be in any configuration, e.g. , block, tapered, random, linear, branched, and/or crosslinked.

-(CF ₂ - CF ₂ )x - (C F 2 - CF)y -

0 - [CF ₂ - CF - 0] _rn -CF ₂ - CF ₂ - C0 ₂ H

CF:

-TFE- Vinyl Ether

TetrafI uoro ethylene

Pe rfl u o ro (4 /- M ethy 1-3 , 6-D i oxa- 7- Octene-1 Carboxylic Acid)

Chemical Structure of a PFCA-TFE Copolymer in the Hydrolyzed Carboxylic Acid Form

[0048] PFCA-TFE copolymers contain a tetrafluoroethylene (TFE) "backbone," to which the perfluorocarboxylic acid (perfluoro(4-methyl-3,6-dioxa-7- octene-1 -carboxylic acid)) groups are attached. PFSA-TFE copolymers and PFCA-TFE copolymers can be converted to the salt form by pretreatment with a suitable base, such as an alkali metal base (e.g. , as described above). Such pretreatment processes of ion exchange membranes are well known in the art and can be performed, for example, in accordance with the manufacturer's recommendations. Depending upon the nature of the source (e.g., impurities, impurity levels) and the resulting water vapor desired, the pretreatment conditions can be adjusted to yield an optimized membrane. For example, the selection of base, solvents used, temperature, exposure time, rinse conditions, extent of ion exchange (e.g. , 10% or less to 90% or more) can be adjusted. It can also be desirable to adjust the hydrophilicity of the resulting membrane by crosslinking it with a hydrophilic agent, or co-casting the polymer with a hydrophilic component. In such embodiments, the polymer already includes crosslinkable groups, or is functionalized to include crosslinkable groups. Other forms of pretreatment can also be employed (e.g., reaction with agents to modify the surface morphology of the polymer (roughen, increase or decrease porosity, etc.), without modifying the surface chemistry.

[0049] Preferably, the membrane is a substantially gas -impermeable perfluorinated ionomer (e.g., a NAFION® membrane). The permeability of water vapor is greater than two orders of magnitude larger than the permeability of C0 ₂ or CO through a NAFION® membrane, and approximately two orders of magnitude greater than the permeability of oxygen or nitrogen. Hydrogen diffusion can be effectively suppressed through a NAFION® membrane, while permitting passage of water vapor. The substantially gas -impermeable membrane is preferably substantially nonporous. The substantially gas impermeable membranes can suppress the diffusion of gases and other materials, such as particles, aerosols, viruses, bacteria, prions, metals, ions, and other airborne molecular contaminants.

[0050] For the delivery of semiconductor organometallics vapors, membranes typically used for separation of condensable organic components from gases may be used. These may be of the rubbery or glassy type where the organic vapor is the preferential permeating component. Examples include, nitrile rubber, neoprene, silicones rubbers including polydimethylsiloxane, chloro sulfonated polyethylene, polysilicone-carbonate copolymers, fluoroelastomers including fluorinated ethylene propylene copolymers, chloro sulfonated polyethylene, polylvinylchloride, polyurethane, cis-polybutadiene, cis- polyisoprene, poly(l-butene) polystyrene-butadiene copolymer, styrene/ethylene/butylenes block copolymers, styrene/butadiene/styrene block copolymers, polymers of perfluoro-2,2-dimethyl-l,3-dioxole, polyhexenes, polyoctenes, or similar polyolefins. In addition, substituted polyacetylenes of the general structural formula:

where Rl is hydrogen, a halogen, C6H5 or a linear or branched alkyl group may be used.

[0051] In preparing a vapor, a source is passed through the membrane. The term "passing a source through a membrane" as used herein is a broad term, and is to be given its ordinary and customary meaning to a person of ordinary skill in the art (and is not to be limited to a special or customized meaning), and refers without limitation to contacting a first side of a membrane with the source, such that the source' s molecules (water or another liquid' s molecules) pass through the membrane, and obtaining a source's vapor on the opposite side of the membrane. The first and second sides can have the form of substantially flat, opposing planar areas, where the membrane is a sheet. Membranes can also be provided in tubular or cylindrical form where one surface forms the inner portion of the tube and an opposing surface lies on the outer surface. One of ordinary skill in the art can readily appreciate that the membrane can take any form, so long as a first surface and an opposing second surface sandwich a bulk of membrane material. Depending upon the processing conditions, nature of the source, volume of liquid vapor to be generated, and other factors, the properties of the membrane can be adjusted. Properties include, but are not limited to physical form (e.g., thickness, surface area, shape, length and width for sheet form, diameter if in fiber form), configuration (flat sheet(s), spiral or rolled sheet(s), folded or crimped sheet(s), fiber array(s)), fabrication method (e.g. , extrusion, casting from solution), presence or absence of a support layer, presence or absence of active layer (e.g., a porous prefilter to adsorb particles of a particular size, a reactive prefilter to remove impurities via chemical reaction or bonding), and the like. It is generally preferred that the membrane be from about 0.5 microns in thickness or less to 2000 microns in thickness or more, preferably from about 1, 5, 10, 25, 50, 100, 200, 300, 400, or 500 microns to about 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800 or 1900 microns. Membrane thickness is most preferably from about 25 microns to 250 microns. For generation of water vapor, a membrane thickness of about 125 microns is generally desirable. When thinner membranes are employed, it can be desirable to provide mechanical support to the membrane (e.g. , by employing a supporting membrane, a screen or mesh, or other supporting structure), whereas thicker membranes may be suitable for use without a support. The surface area of the membrane can be selected based on the mass of vapor to be produced. Membranes in fiber form are generally preferred over sheet form, due to their ability to provide greater exposed membrane surface area in a fixed volume when fibers are employed. Preferably, the membranes comprise hollow fibers, the fibers being fixed at each end in a header. The fibers are normally sealed at the lower end and open at their upper end to allow removal of vapor; however, in some arrangements, the fibers can be open at both ends to allow removal of vapor from one or both ends. Alternatively, source can be provided to the interior of the membrane fibers, and vapor removed from a space surrounding the exterior of the fibers.

[0052] In a particularly preferred embodiment, the membrane is provided in a form of a flat sheet or tubular membrane overlying a tubular support. While a tubular support is particularly preferred, other configurations can also be employed, e.g., flat sheet overlying a supporting plate. The support preferably includes a plurality of small orifices (e.g., randomly or regularly distributed circular holes) in contact with the overlying membrane. The number and size of the orifices in the support can be controlled such that the exposed surface area of the membrane is known with precision. In preferred embodiments the support is, e.g., a 1/8 inch diameter tube provided with a plurality of small holes (e.g., 25/1000 inches in diameter). The supported membrane is provided in a chamber or other container, with the membrane side exposed to the liquid to be vaporized, with vapor drawn from the support side. A single supported membrane can be provided, or a plurality of supported membranes can be provided, depending upon the amount of vapor to be delivered or other factors. A temperature sensor is submerged in the liquid to be vaporized to provide temperature information. A temperature sensor is also provided in the vacuum chamber into which the vapor is drawn. A pressure transducer in the vacuum chamber is employed to monitor the pressure. In a simple system as illustrated in Figure 1, a particular vapor flow rate is selected based on system needs, and the membrane properties (thickness, exposed surface area) are selected so as to deliver a flow rate that generally meets this requirement. Temperature adjustment can then be employed for fine control of the flow rate.

[0053] Figure 2 illustrates a control system for achieving a flow rate with a high degree of control, e.g., in the vaporizer system of Figure 1. The power required for mass transfer based on enthalpy of evaporation (see Figure 2) is a calculated value based on the heat of vaporization (AH _vap ), which is well defined for all liquids. By measuring the change in pressure in the vacuum chamber while holding the temperature constant, the moles of gas transferred can be calculated using the ideal gas law (PV=nRT). A calibration curve can be constructed by performing this calculation at various temperatures (see Figure 3). This calibration curve can be used to convert the mass transfer set point to an uncorrected temperature set point (see Figure 2). The calculated power required to vaporize a given amount of liquid (theoretical power requirement) can be compared to the actual power drawn by the vaporizer heater assembly (determined by measuring current and voltage going to the heater device). The results of the comparison are used by the power controller to provide a temperature set point correction. The temperature set point correction is compared to the uncorrected temperature set point to yield a corrected temperature set point, which is compared to the actual temperature measured by the temperature sensor in the vaporizer. If a temperature error is noted, the temperature controller adjusts the power to the vaporizer heaters. The process is conducted iteratively, with temperature and power measurements being obtained and processed on a regular basis. In this way, power consumption by the vaporizer heater and temperature in the vaporizer can be used to provide accurate flow rates. The system offers advantages, in that one can monitor the energy consumption to detect failure in a temperature sensor or other system breakdown or maintenance failure.

[0054] Figure 3 provides a graph of water vapor pressure and flow rate as a function of water temperature. Vapor is typically delivered into the vacuum chamber at a pressure of 600 torr or less, preferably from about 0 to 100 torr.

Organometallic Compounds

[0055] The devices and methods of preferred embodiments are particularly well-suited for the delivery of vapor of organometallic compounds, e.g., organometallic compounds employed in preparing semiconductor films, e.g., Group ΠΙ-V films and Group II- VI films, e.g., by metal organo chemical vapor deposition (MOCVD). Such compounds include gallium nitride, indium phosphides, gallium arsenide, aluminum gallium nitride, and transparent conducting oxide films such as indium tin oxide, tin oxide, and zinc oxide.

[0056] For diethyl zinc, slight temperature fluctuations can have a great effect on the amount of compound delivered when a bubbler is employed because of the carrier gas. The methods and devices of preferred embodiments which do not employ a carrier gas offer significant advantages over the use of a bubbler.

Precursors for Use in Manufacturing Semiconductor Devices

[0057] The devices and methods of preferred embodiments are particularly well- suited for the delivery of metal-containing precursors for use in chemical vapor deposition (CVD) processes. As discussed in this section, such precursors encompass a wide range of chemical compounds, from gases through volatile liquids, sublimable solids to relatively involatile solids. Those commonly used in the crystal growth of thin films of functional materials include o-bonded metal-alkyl compounds such as the pyrophoric volatile liquid dimethyl zinc (ZnMe ₂ ) and the pyrophoric solid trimethyl indium (InMe ₃ ), and π-bonded organometallic compounds, most notably used for otherwise difficult metals such as magnesium or manganese. Coordination compounds such as acetylacetonates and other β-diketonate derivatives have also been used, particularly in the deposition of oxides. To be successful precursors, compounds have properties such as high purity, including the exclusion of any extrinsic impurities acquired during precursor synthesis, especially if they can act as dopants, clean decomposition on the substrate surface, without the incorporation of unwanted intrinsic impurities. Impurities can include elements present within the ligands e.g., carbon from an alkyl group or fluoride from a substituted β-diketonate derivative. The compounds also have moderate vapor pressures, volatilize quantitatively on heating and, at the temperature used for volatilization, remain stable for an adequate period of time in the gas phase, but also decompose cleanly at the temperature of the substrate (this may only be a little higher then the temperature needed to volatilize the precursor).

[0058] Forming mixtures with other precursors can modify the physical properties of a precursor. This, in turn, can have an advantageous effect on the delivery of the precursor(s) to the CVD reactor. The phase rule is often written in the following simple form (for systems composed of reactive components): F=(N-R)+2 - P or F = C+ 2 - P, where F is number of degrees of freedom or variance of the system (the number of variables which can be chosen by the experimenter, and must be chosen to define the system), C = (N-R) is the number of distinct chemical entities (N) minus the number of independent reactions between them (R) (it is the minimum number of substances that must be mixed to form the system), and P is number of phases present (liquid, gas, etc.). Conclusions can be drawn from this law, e.g., the number of phases available cannot exceed the number of components by more than two. This is classically illustrated by the triple point of water: there is a specific temperature and pressure at which the three phases coexist (ice, water, steam), so there can be no degrees of freedom. This approach can be applied to the delivery of a precursor vapor to the reactor chamber of a CVD system. The consequences of the phase rule, as it governs the separation of liquids by distillation, are familiar to all chemists. Many mixtures distill with constant composition (azeotrope), with perhaps the most familiar example being that of water-ethanol. The metal-alkyl Lewis base adducts Me ₂ Zn NEt ₃ and Me ₂ Cd S(CH ₂ ) ₄ are commercially available precursors, and have several advantages over the base-free metal-alkyls, including greater ease of handling (the adducts are not pyrophoric liquids), fewer premature homogeneous reactions between the precursors (up stream of the substrate), and greater purity. In addition, lower vapor pressures make the flow rates into the reactor easier to control (particularly important when such compounds are used as dopants). In many cases, the presence of a Lewis base compound also improves the quality of the semiconductor layers grown. Lewis base adducts of dimethyl zinc are particularly useful in the doping of ΙΠ/V materials with zinc. Many monodentate amines "bis" coordinating adducts in the solid state with group 12 alkyl compounds, similar to the dimethyl zinc bistriazine adduct. Dimethyl zinc has a high vapor pressure and can be used at low temperatures. The commonly used mixture of dimethyl zinc-triethylamine (DMZ-TEAM) adduct is a liquid at room temperature but in the vapor phase the components are fully dissociated. Moreover, one important feature of the system is that the vapor above a 2:1 mixture of DMZ and TEAM consists mainly of TEAM, with little or no evidence (from vapor phase infrared spectra) of the presence of DMZ. In fact the mixture is effectively air stable, suggesting very low volatility for any zinc-containing species. This indicates that the vapor pressure of TEAM is very much greater than that of any adducts with DMZ. Indeed, the vapor pressure of a 2:1 mixture of TEAM/DMZ very closely resembles that of the pure amine. A second important feature of the chemistry of zinc-alkyl compounds is their relatively weak Lewis acidity. As a consequence, they have a low degree of association with Lewis base compounds such as amines.

[0059] Mixtures of triisopropyl gallium with TEAM in metal organic molecular beam epitaxy (MOMBE) growth, or ethyldimethyl indium (EDMIn) may be employed for delivery of oxides. Heteroleptic metal-alkyls such as EDMIn, EtMe ₂ In, as a liquid precursor for the delivery of indium in thin film CVD techniques can be employed. The composition of the system will be close to the stoichiometry EtMe ₂ In, with the formation of trimethyl indium (TMIn) and diethymethyl indium (DEMIn). Neglecting the formation of TEIn, an azeotrope will form if there is a minimum in the vapor pressure composition curve.

[0060] These systems sometimes give poor growth results, with poor elemental stoichiometry in the as-grown semiconductor layers. This has been attributed to ligand exchange. Ligand exchange produces new compounds (TMIn, EDMIn, and TEIn in the case of EDNIn), that have quite different properties from the originally designed precursors. Moreover, due to ligand exchange being a common phenomenon in main group metal-alkyls, when a heteroleptic metal-alkyl, or a number of metal-alkyls possessing different alkyl groups are used, after equilibrium is established, a complex mixture of labile (in this case indium) alkyl compounds are formed. Indium alkyl compounds possess different physical properties from one another, including the temperatures at which they decompose. These different species decompose at different temperatures via different mechanistic pathways. When preparing ternary main group semiconductors this can become a severe problem.

[0061] Thin film metal-organic vapor phase epitaxy (MOVPE) growth of InGaAs from a mixture of the metal-alkyls, triethyl gallium (GaEt ₃ ), trimethyl indium (InMe ₃ }, and tert-butyl arsine (As ^£ Bu ₃ ), as attained from on-line IR monitoring, shows ligand exchange between the three precursors in the feed lines of the reactor, upstream of the substrate. The composition of the resulting product consists of trimethyl gallium (GaMe ₃ ), dimethylethyl gallium (Me ₂ GaEt), methyldiethyl gallium (MeGaEt ₂ ), dimethyl- ethyl indium (Me ₂ InEt), and methyldiethyl indium (MeInEt ₂ ). Ligand exchange has a profound effect on the composition of precursors in the reactor feed, with the relative concentrations of the exchange products being dependent on the initial ratio of precursors used. The thermal stability of these resulting compounds varies over a wide temperature range, with the indium species decomposing in the order MDEIn > DMEIn > TMIn. The gallium species tend to decompose at higher temperatures, but in a similar order. This, in turn, can result in inhomogeneous element distribution (indium and gallium) in the as- grown InGaAs layers, with the film near the reactor inlet being rich in indium, while that near the reactor outlet is rich in gallium.

[0062] For main group alkyl precursors, ligand exchange appears to be a common occurrence and has been observed in hydrocarbon solutions of trimethyl gallium, trimethyl indium, and trimethyl aluminum at temperatures as low as -65°C. Rapid ligand exchange was also observed in cold (-60°C and -85°C) hydrocarbon solutions, between trimethyl gallium and triethyl gallium, along with mixtures of group 12 alkyls and trimethyl indium or trimethyl gallium. Ligand exchange can occur in the gas phase, e.g., when growing CdZnTe in a MOVPE reactor, wherein the precursors diethyl zinc and dimethyl cadmium rapidly exchange their alkyl ligands to produce methylethyl zinc, dimethyl zinc, methylethyl cadmium, and diethyl cadmium. This is also the case for the gas phase mixing of trimethyl gallium, trimethyl indium, and trimethyl amine alane, [A1H ₃ -N(CH ₃ )3] which produced methyl aluminum, alkyl gallium hydride, and alkyl indiumhydride compounds in the feed lines of a MOVPE reactor. The use of sterically bulky ligands, such as isobutyl groups and o-donor coordinating complexes to form Lewis acid-base mixtures may reduce or even inhibit ligand exchange between these compounds, which may be a factor in why precursor-amine mixtures, in general, give better growth results.

[0063] TMIn is probably the most commonly used solid precursor in semiconductor deposition. A saturated solution of TMIn in a non-volatile coordinating solvent that is in contact with the solid precursor can be vaporized. Solvent-TMIn combinations include Ν,Ν-dimethyldodecylamine (dmda) and hexadecane

[0064] Compounds containing the β-diketonate ligand tmhd, or its derivatives have been widely used as precursors in CVD applications. Many commercially used derivatives contain fluorinated ligands as these have the effect of increasing volatility. However, because of the presence of fluorine, thermally stable fluorides such as BaF ₂ can be formed on deposition, and this, in turn, can lead to fluoride being incorporated into the growing thin films, or etching of the films. To prevent the formation of fluorides, water vapor can added to the reactor chamber and/or a higher temperature used, however, for many materials this is undesirable. Ammonia gas, NH ₃ , can be utilized to increase the transport of non-fluorinated β-diketonate compounds, e.g., [Ba ₄ (tmhd)8] , to the reactor chamber and substrate. This approach allows the use of less volatile precursors such as [Ba(acac) ₂ ] _n , wherein acac is acetylacetonate, to deposit thin films of BaO. As no functional substitutions are required on the diketonate ligand, there is no potential for contamination in the resultant films, especially crucial in superconducting structures. One drawback is that this method can produce a buildup, within the CVD system, of NH ₃ gas, which is extremely reactive and corrosive, and has, in some cases, resulted in spectacular explosions.

[0065] Films of the high-Z, superconductor Bi ₂ Sr ₂ CaCu ₂ O _x from Sr(tmhd) ₂ , Ca(tmhd) ₂ , Cu(acac) ₂ , and triphenyl bismuth [Bi(Ph) ₃ ] can be prepared by low-pressure CVD. The strontium and copper precursors attain sufficient volatility at temperatures in excess of 200°C. Enhanced source volatility for the copper precursor is observed when in the presence of pyridine.

[0066] To attain a fully saturated coordination sphere at the central metal atom, precursors such as Sr(tmhd) ₂ , Ca(tmhd) ₂ , and Cu(acac) ₂ are polynuclear in the solid state. Heating at or before volatilization leads to dissociation into molecular monomers. Thus equilibrium is established at the surface of the solid or the melt, between monomers in the gas phase and higher oligomers in the bulk solid. In the gas phase, when monomer dimerization occurs, the resulting oligomers precipitate out. In the presence of ammonia, ammonia gas may coordinately saturate the metal center, via nitrogen-metal Lewis acid- base bonds. This reaction may have the effect of inhibiting, or at least slowing down, gas- phase oligomerization, shifting the overall equilibrium, thus increasing the volatility of the precursors. One method of achieving monomeric calcium compounds, albeit in the solid phase, is to use additional o-donor compounds, such as chelating amines, to prevent tmhd ligands from inter- molecular coordination, as in the case of [Ca(tmhd) ₂ (tetraen)], wherein tetraen is tetraethylenepantamine, where the chelating amine completes the coordination sphere about the metal center (Fig. 15b). The adduct molecule has a higher vapor pressure than the non-adducted parent complex [Ca(tmhd)6] .

[0067] The methods and devices of preferred embodiments offer advantages over conventional methods of delivery such as bubblers. Bubblers typically exhibit condensation along lines of delivery, which is undesirable, and the organometallic compounds can decompose on the bubbler heat source and then catalyze more decomposition.

Micro-electronics Processing [0068] The vaporizer devices of preferred embodiments can be configured to function as ultrapure humidifiers capable of meeting micro-electronics and other critical process requirements for delivery of water vapor or other vapors.

[0069] Preferably, the portions of the device in contact with vapor are constructed of either all fluoropolymer components, or fluoropolymer and 316L stainless steel components, depending on the application.

[0070] The vaporizer devices of preferred embodiments are configured for use over a variety of ambient conditions. The thermal response and control of the vaporizer device are very fast, accurate, and capable of high vaporization rates by using thick or thin film resistive heating elements bonded directly to the quartz or stainless steel vaporizer housing. The fittings are selected based on the application, and typically include face seal, compression, flare or tube stub fittings. Such fittings are commercially available under the trademarks VCR®, SWAGELOK®, and FLARETEK®. In preferred embodiments, the vaporizer is constructed from either PFA (a polymer of tetrafluoroethylene and perfluorovinylether), polyvinylidene fluoride (PVDF), quartz, or stainless steel; however, other materials can also be suitable for use. The vaporizer device is advantageously provided with heaters, coolers, and/or temperature controllers.

[0071] The vaporizer devices of preferred embodiments are particularly well suited for use in delivering water vapor to micro-electronics processes. Water vapor is used, e.g., in rapid thermal processing (RTP), atomic layer deposition (ALD), plasma stripping, and diffusion. For ALD, RTP, and diffusion, water vapor is employed to grow oxides. Water is often generated in a pyrolitic process to ensure the purity of the water vapor. Difficulties with such pyrolitic processes include the need for combusting oxygen and hydrogen, the need for external torches on the diffusion chambers to prevent the 800°C heat necessary for the pyrolitic process from changing the thermal profile of the tool, particulation of the torch tip, long startup and shut down times, safety issues associated with hydrogen use, and problems associated with excess hydrogen in the chamber. In addition, such systems have difficulty operating properly with very low flow rates or low water vapor to hydrogen ratios.

[0072] For RTP, high flow rates of water vapor for short periods of time are needed. Conventional catalytic systems for generating water vapor are metallic, expensive, and have similar safety issues as torches. They also have problems with rapid cycle times and are not easily scalable due to thermal buildup in the catalytic combustion cell.

[0073] ALD requires very small amounts of water vapor for High K film formation. The purity of water vapor is critical for good film formation. Bubblers cannot control water vapor purity, and thus are unsuitable for use in ALD processes.

[0074] The vaporizer devices of preferred embodiments offer advantages over the conventional water vapor delivery methods used in RTP and ALD.

[0075] In plasma stripping, the process is more effective when water vapor is used to help lift the film of the wafer surface. The vaporizer devices of preferred embodiments can be employed to provide high flow rates of pure water vapor to a plasma stripping process in a controlled fashion. Lithography needs high flow rates of humid clean dry air. The vaporizer devices of preferred embodiments are capable of humidifying gas streams of up to and exceeding 1000 slm.

[0076] Wafer cleaning has significant challenges as device feature sizes are scaled down, as particle sizes get smaller, and as acceptable contamination levels are made lower. Water in liquid form does not contain enough thermal energy to remove contamination from the wafer surfaces. However, by contacting the wafer with pure steam, the high thermal energy water vapor can remove contamination, thus overcoming many of the challenges inherent in the use of liquid water.

[0077] All references cited herein are incorporated herein by reference in their entirety. To the extent publications and patents or patent applications incorporated by reference contradict the disclosure contained in the specification, the specification is intended to supersede and/or take precedence over any such contradictory material.

[0078] Unless otherwise defined, all terms (including technical and scientific terms) are to be given their ordinary and customary meaning to a person of ordinary skill in the art, and are not to be limited to a special or customized meaning unless expressly so defined herein.

[0079] Terms and phrases used in this application, and variations thereof, especially in the appended claims, unless otherwise expressly stated, should be construed as open ended as opposed to limiting. As examples of the foregoing, the term 'including' should be read to mean 'including, without limitation,' 'including but not limited to,' or the like; the term 'comprising' as used herein is synonymous with 'including,' 'containing,' or 'characterized by,' and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps; the term 'having' should be interpreted as 'having at least;' the term 'includes' should be interpreted as 'includes but is not limited to;' the term 'example' is used to provide exemplary instances of the item in discussion, not an exhaustive or limiting list thereof; adjectives such as 'known', 'normal', 'standard', and terms of similar meaning should not be construed as limiting the item described to a given time period or to an item available as of a given time, but instead should be read to encompass known, normal, or standard technologies that may be available or known now or at any time in the future; and use of terms like 'preferably,' 'preferred,' 'desired,' or 'desirable,' and words of similar meaning should not be understood as implying that certain features are critical, essential, or even important to the structure or function of the invention, but instead as merely intended to highlight alternative or additional features that may or may not be utilized in a particular embodiment of the invention. Likewise, a group of items linked with the conjunction 'and' should not be read as requiring that each and every one of those items be present in the grouping, but rather should be read as 'and/or' unless expressly stated otherwise. Similarly, a group of items linked with the conjunction 'or' should not be read as requiring mutual exclusivity among that group, but rather should be read as 'and/or' unless expressly stated otherwise.

[0080] With respect to the use of substantially any plural and/or singular terms herein, those having skill in the art can translate from the plural to the singular and/or from the singular to the plural as is appropriate to the context and/or application. The various singular/plural permutations may be expressly set forth herein for sake of clarity.

[0081] It will be further understood by those within the art that if a specific number of an introduced claim recitation is intended, such an intent will be explicitly recited in the claim, and in the absence of such recitation no such intent is present. For example, as an aid to understanding, the following appended claims may contain usage of the introductory phrases "at least one" and "one or more" to introduce claim recitations. However, the use of such phrases should not be construed to imply that the introduction of a claim recitation by the indefinite articles "a" or "an" limits any particular claim containing such introduced claim recitation to embodiments containing only one such recitation, even when the same claim includes the introductory phrases "one or more" or "at least one" and indefinite articles such as "a" or "an" (e.g., "a" and/or "an" should typically be interpreted to mean "at least one" or "one or more"); the same holds true for the use of definite articles used to introduce claim recitations. In addition, even if a specific number of an introduced claim recitation is explicitly recited, those skilled in the art will recognize that such recitation should typically be interpreted to mean at least the recited number (e.g., the bare recitation of "two recitations," without other modifiers, typically means at least two recitations, or two or more recitations). Furthermore, in those instances where a convention analogous to "at least one of A, B, and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B, and C" would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.). In those instances where a convention analogous to "at least one of A, B, or C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B, or C" would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and/or A, B, and C together, etc.). It will be further understood by those within the art that virtually any disjunctive word and/or phrase presenting two or more alternative terms, whether in the description, claims, or drawings, should be understood to contemplate the possibilities of including one of the terms, either of the terms, or both terms. For example, the phrase "A or B" will be understood to include the possibilities of "A" or "B" or "A and B."

[0082] All numbers expressing quantities of ingredients, reaction conditions, and so forth used in the specification are to be understood as being modified in all instances by the term 'about.' Accordingly, unless indicated to the contrary, the numerical parameters set forth herein are approximations that may vary depending upon the desired properties sought to be obtained. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of any claims in any application claiming priority to the present application, each numerical parameter should be construed in light of the number of significant digits and ordinary rounding approaches.

[0083] Furthermore, although the foregoing has been described in some detail by way of illustrations and examples for purposes of clarity and understanding, it is apparent to those skilled in the art that certain changes and modifications may be practiced. Therefore, the description and examples should not be construed as limiting the scope of the invention to the specific embodiments and examples described herein, but rather to also cover all modification and alternatives coming with the true scope and spirit of the invention.