Method of refilling ampoule

Abstract

Systems, methods, and software for manufacturing processes of an automatic replenishment system for ampoules, the ampoules including one or more sensors configured to determine one or more liquid level heights within the ampoules. The automatic replenishment system has a state machine configured to control the automatic replenishment system, the state machine having one or more states for replenishing the ampoules.

Description

Technical Field

[0001] This invention discloses a method for replenishing ampoules used in semiconductor manufacturing processes to improve system throughput. Specifically, the invention relates to an automated replenishment system for controlling liquid delivery that can be adjusted for different applications. Prior Technology

[0002] Chemical vapor deposition (CVD) is a type of deposition process used to deposit layers on a substrate. Physical vapor deposition (PVD) is another type of deposition process in which the precursor changes from a condensed phase to a gaseous phase and then returns to the thin film condensed phase. Atomic layer deposition (ALD), also known as cyclic deposition, uses chemisorption technology to deliver precursor molecules to the substrate surface in continuous cycles. Each cycle exposes the substrate surface to a first precursor vapor, a rinsing gas, a second precursor vapor, and a rinsing gas.

[0003] Precursor vapors (e.g., organometallic precursor vapors) are commonly used in film deposition processes, involving the thermal deposition of one or more precursor vapors in a processing chamber. Precursors (such as organometallic precursors) are generally in liquid or solid form. Precursor vapors are typically generated thermally inside a closed container or ampoule. The precursor molecules are then transported through a gas delivery conduit to the substrate surface inside the processing chamber. To prevent the vapor from reverting to its bulk form, the gas delivery conduit is typically thermally controlled above the dew point of the specific precursor. In direct liquid injection processes, liquid is injected into the chamber and evaporates at high temperatures.

[0004] The evaporation of precursors and the deposition of thin films in a reactive gas environment are sensitive to gas flow, and precise gas flow control is required in methods and apparatus for conveying precursors to the film deposition chamber. Gas flow variability leads to variability between chambers and product variability.

[0005] In multiple cycles, the sealed container or ampoule must undergo replenishment processes, during which the pressure and temperature within the sealed container or ampoule can vary. Due to the lower temperature of the fresh liquid, the temperature of the liquid in the ampoule drops, causing process drift. This is conventionally controlled by preheating the precursor or by including heating elements within the sealed container or ampoule, both of which increase the total footprint of the sealed container or ampoule, increase system complexity, and increase overall manufacturing costs.

[0006] Therefore, there is a need for supplementary processes and methods that mitigate the effects of pressure and temperature on sealed containers or ampoules during the supplementary process. Summary of the Invention

[0007] One aspect of this disclosure relates to a method for replenishing ampoules, comprising controlling the operation of an automatic replenishment system. The automatic replenishment system has a plurality of liquid level sensors. Each of the plurality of liquid level sensors is positioned at a known location perpendicular to the height of the inner surface of the ampoule. Each of the liquid level sensors is configured to send a signal to a sensor controller for detecting the liquid level in the ampoule. The automatic replenishment system includes a sensor controller and a gas delivery controller. Signals are received by the sensor controller from a first liquid sensor of the plurality of liquid level sensors. The first liquid sensor is positioned at a first distance from the bottom of the ampoule and is configured to determine the liquid level L of the ampoule. Signals from a second liquid sensor of the plurality of liquid level sensors are received by the sensor controller. The second liquid sensor is positioned at a second distance from the bottom of the ampoule and is configured to determine the liquid level H of the ampoule. Replenishment operation is determined for each of the plurality of sensors. The sensor controller sends a signal to the gas delivery controller. The signal from the first liquid sensor is configured to prompt the gas delivery controller to activate the main valve, and the signal from the second liquid level sensor is configured to prompt the gas delivery controller to deactivate the main valve. The gas delivery controller is prompted to activate the main valve when the liquid level is at a low liquid level L, and to deactivate the main valve when the liquid level is at a filling liquid level.

[0008] Additional embodiments disclosed herein relate to a computer program product containing computer-executable code embodied in a non-transitory computer-readable medium, which, when executed on one or more computing devices, performs the following steps: controlling the operation of an automatic replenishment system having a plurality of liquid level sensors, each of the plurality of liquid level sensors being positioned at a known location perpendicular to the height of the inner surface of an ampoule, each of the liquid level sensors being configured to detect a liquid level in a reservoir; receiving signals from a first liquid sensor and a second liquid sensor of the plurality of liquid level sensors, the signals from the first liquid sensor and the second liquid sensor corresponding to a replenishment function of a component of the automatic replenishment system, the automatic replenishment system including a main valve; determining a replenishment function assigned to each of the plurality of sensors, the first liquid sensor being assigned to enable the main valve, and the second liquid sensor being assigned to deactivate the main valve; enabling the main valve when the liquid level is at the first liquid sensor; and deactivating the main valve when the liquid level is at the second liquid sensor.

[0009] Another embodiment disclosed herein relates to a method for replenishing an ampoule. The method includes monitoring a liquid level in the ampoule using at least one liquid level sensor spaced at known heights within the ampoule. A plurality of wafers are processed until the liquid level in the ampoule decreases to change the activation state of the liquid level sensor from a first state to a second state. Processing of the plurality of wafers continues after the activation state change, and wafers are counted after the activation state change. A wafer change event is determined to be enabled, and if a predetermined number of wafers have been processed after the activation state change and the wafer change event is enabled, the liquid level in the ampoule is replenished until the activation state of the liquid level sensor changes back to the first state. Simple Explanation of the Diagram

[0010] Figure 1A illustrates a cross-sectional view of an ampoule according to one or more embodiments of the present disclosure;

[0011] Figure 1B illustrates a schematic diagram of a gas delivery system according to one or more embodiments of the present disclosure;

[0012] Figure 2 illustrates a schematic diagram of an automatic replenishment system according to one or more embodiments of the present disclosure;

[0013] Figure 3A illustrates a schematic diagram of an automatic replenishment system according to one or more embodiments of the present disclosure;

[0014] Figure 3B illustrates a schematic diagram of the main reservoir of an automatic replenishment system according to one or more embodiments of the present disclosure;

[0015] Figure 4 illustrates a method for producing a supplementary ampoule according to one or more embodiments of this disclosure;

[0016] Figure 5 illustrates a cross-sectional view of an ampoule according to one or more embodiments of the present disclosure;

[0017] Figure 6 illustrates a condition table for a state machine according to one or more embodiments of this disclosure;

[0018] Figure 7 illustrates a condition table for a state machine according to one or more embodiments of this disclosure;

[0019] Figure 8A illustrates a schematic diagram of a state machine according to one or more embodiments of the present disclosure;

[0020] Figure 8B illustrates a schematic diagram of a state machine according to one or more embodiments of the present disclosure;

[0021] Figure 9 illustrates a schematic diagram of a control system and neural network 700 according to one or more embodiments of the present disclosure;

[0022] Figure 10 illustrates a manufacturing process for controlling an automatic replenishment system having a finite state machine and a neural network, according to one or more embodiments of this disclosure;

[0023] Figure 11 illustrates a schematic diagram of a data structure according to one or more embodiments of the present disclosure;

[0024] Figure 12 illustrates a schematic diagram of a system according to one or more embodiments of the present disclosure;

[0025] Figure 13 illustrates a schematic diagram of a software system according to one or more embodiments of the present disclosure; and

[0026] Figure 14 illustrates a schematic diagram of a network according to one or more embodiments of the present disclosure. Implementation

[0027] Before describing several exemplary embodiments of this disclosure, it will be understood that this disclosure is not limited to the details of the construction or process steps set forth in the following description. This disclosure can have other embodiments and can be practiced or carried out in various ways.

[0028] As used herein, the terms "ampoule," "reservoir," and "closed container" are used interchangeably and refer to precursor ampoules in low- and high-pressure applications. Precursor ampoules can be of the bubbler type, in which inlet carrier gas enters a tube immersed in the precursor; and cross-flow ampoules, in which carrier gas purges the top space of the ampoule.

[0029] Referring to the diagrams described below, a linear arrangement is depicted; however, any suitable layout of components and hardware can be usefully employed depending on the specific process design. Additionally, tools such as valves, gas delivery devices, controllers, ampoules, and the like, along with sensors (such as pressure sensors, optical sensors, contamination sensors, etc.), timers, switches, actuators, repeaters, motors, etc., can be suitably used to control or monitor the process. All such variations are intended to fall within the scope of the system described herein.

[0030] Figure 1 illustrates an ampoule 100 used with semiconductor manufacturing reagents. The ampoule 100 includes a container 110 having a bottom 112, sidewalls 114, and a cap 116 enclosing a volume 101. An inlet port 120 and an outlet port 130 are in fluid communication with the volume 101. In some embodiments, the ampoule 100 includes more than one inlet port 120 for delivering a carrier gas, precursor gas, or precursor liquid.

[0031] The inlet port 120 is generally configured to allow connection to a gas source and may have a suitable threaded or sealed connection. The inlet port 120 has a passage 121 with an inner diameter defining the cross-sectional width of the passage 121. The passage 121 has a bottom end 125, which may be curved, tapered, or flat.

[0032] In some embodiments, the inlet port 120 has a spray head at an end of the inlet port 120 located within the volume 101. The spray head is configured to guide the flow of gas or liquid, and in some embodiments, includes one or more nozzles.

[0033] The outlet port 130 is also in fluid communication with volume 101 in container 110. The outlet port 130 is generally configured to be connected to a line to allow gas leaving container 110 to flow into the processing chamber (or other component). The outlet port 130 may have a threaded connection to allow connection of a gas line.

[0034] In some embodiments, as shown in Figure 1, the cap 116 is a component separate from the bottom 112 and the sidewall 114. The cap 116 can be attached to the sidewall 114 of the container 110 using a removable bolt through a suitably shaped opening 160. The opening 160 may have a threaded portion to allow for easy connection of a threaded bolt. The removable bolt allows the cap 116 to be removed from the container 110, making it possible to change or add precursors in the container 110. In some embodiments, the container 110 includes an O-ring 162 positioned between the cap 116 and the sidewall 114 to form a fluid-tight seal. In some embodiments, the ampoule is configured to contain a solid precursor.

[0035] Ampoule 110 further includes a plurality of liquid level sensors positioned along the inner surface of sidewall 114. As shown in Figure 1, at least four liquid level sensors are present. Each of the at least four liquid level sensors is positioned vertically along the height of the inner surface of the ampoule. Bottom liquid level sensor 150 is positioned at a distance from the bottom 112 of ampoule 100. Top liquid level sensor 152 is positioned at a distance from cap 116. Two precision liquid level sensors are located between bottom liquid level sensor 150 and top liquid level sensor 152. Specifically, first liquid level sensor 154 is positioned above bottom liquid level sensor 150 and below top sensor 152, and second liquid level sensor 156 is positioned above first liquid level sensor 156 and below top sensor 152. In some embodiments, top liquid level sensor is positioned below vent (opening 160).

[0036] Each of the bottom liquid level sensor 150, the top liquid level sensor 152, and two precision liquid level sensors (first liquid level sensor 154 and second liquid level sensor 156) is connected and controlled by a sensor controller 170. In embodiments in which more than four sensors are implemented, each of the more than four sensors is connected and controlled by the sensor controller 170. The sensor controller 170 is configured to send and receive liquid level signals from one or more of the liquid level sensors (150, 152, 154, 156) to indicate when a certain liquid level has been reached. In some embodiments, as explained in further detail below, the sensor controller 170 is configured to send liquid level signals to a gas delivery controller (not shown).

[0037] As explained in further detail below, in this embodiment, the bottom liquid level sensor 150 is configured to indicate the lowest liquid level; the top liquid level sensor 152 is configured to indicate the highest liquid level. It will be understood that, depending on the batch quantity or chemical characteristics of the precursor rather than the minimum and maximum precursor volumes that can be contained within the ampoule 100, in some embodiments the lowest and highest liquid levels are set to indicate the minimum and maximum desired liquid levels.

[0038] As explained in further detail below, in this embodiment, two precise liquid level sensors (first liquid level sensor 154 and second liquid level sensor 156) are configured to indicate the start and stop levels of the precise automatic replenishment system 200. The first liquid level sensor 154 is configured to indicate that the ampoule 100 needs to be replenished, thus prompting the controller 170 to enable the automatic replenishment system 200. The second liquid level sensor 156 is configured to indicate that the ampoule 100 has been replenished, thus prompting the controller to deactivate the automatic replenishment system 200.

[0039] According to one or more embodiments, the sensor controller 170 includes a processor 172, memory 174 coupled to the processor, input / output devices coupled to the sensor controller 170, and support circuitry for providing communication between different components of a system or device, and operation of one or more liquid level sensors (150, 152, 154, 156). In some embodiments, the sensor controller 170 communicates with a gas delivery controller 291 and is configured to send signals to the gas delivery controller 291.

[0040] Figures 1B, 2, and 3 illustrate schematic representations of the precursor delivery system 50 and the automatic replenishment system 200. The precursor delivery system 50 is configured to deliver precursor vapor or liquid to the ampoule 202, and the automatic replenishment system 200 is configured to replenish the ampoule 202 during various stages of substrate processing and deposition, as explained in further detail below. The automatic replenishment system is connected to a gas delivery system 229, which is configured to deliver a driving gas into the ampoule 202. Figure 1B illustrates a conventional precursor delivery system 50 for delivering precursor vapor or liquid to the ampoule 202.

[0041] The precursor delivery system 50 includes a precursor supplier 52 connected to an inlet conduit 54. The precursor supplier 52 supplies liquid or gas to the ampoule 202 via the inlet conduit 54 connected to an inlet port 55 of the ampoule 202. In some embodiments, a pressure sensor 56 is located downstream of the precursor supplier 52 and configured to measure the pressure of the precursor entering the ampoule 202. In some embodiments, the precursor delivery system 50 further includes a mass flow control or volumetric flow controller located downstream of the precursor supplier 52 to regulate the flow rate from the precursor supplier 52 to the inlet conduit 54. In some embodiments, upstream of the mass flow controller, the precursor delivery system 50 further includes a heater configured to heat the precursor supplier 52 to a desired temperature. In some embodiments, the actual gas supply source is a separate component and the precursor supplier 52 is any suitable inlet connector that can be connected to the gas supply source, as will be familiar to those skilled in the art.

[0042] In some embodiments, the inner end 58 of the inlet conduit 54 is immersed in the liquid precursor 211 during substrate processing, wherein precursor vapor is delivered to the processing chamber 283 during the film formation process. In other embodiments, the inner end 58 of the inlet conduit 240 is not immersed in the liquid precursor 211 during substrate processing, wherein precursor vapor is delivered to the processing chamber 283 during the film formation process. In some embodiments, the inner end 58 of the inlet conduit 240 has components for redirecting or diffusing the flow of carrier gas through the inlet conduit 54. In some embodiments, an injector 60 is positioned on the inner end 58 of the inlet conduit 54. The injector 60 is in fluid communication with the inlet conduit 55 to allow gas to flow through the inlet conduit 54 to bubble through the liquid precursor 211 via the injector 60.

[0043] The gas delivery system 229 further includes an outlet conduit 250 connected to an outlet port of ampoule 202 and configured to allow gas to flow out of ampoule 202. In some embodiments, the inner end 58 of inlet conduit 54 is attached above the liquid level surface 211a of liquid precursor 211. In one or more embodiments, the inner end 58 of inlet conduit 54 and the inner end 252 of outlet conduit 250 do not contact liquid precursor 211. In such embodiments, vapor of the precursor in the top space above liquid precursor 211 is carried through outlet conduit 250 as a carrier gas leaving the ampoule mixed with the vapor of precursor 205, and this carrier gas is delivered to processing chamber 283.

[0044] In 216, it is far enough to reduce the amount of precursor transported.

[0045] In the illustrated embodiment, the outlet is connected to a processing chamber 283, such as a film forming chamber, into which precursor vapor entrained in the carrier gas is delivered for the film deposition process. The processing chamber can be any suitable processing chamber known to those skilled in the art, including but not limited to atomic layer deposition (ALD) chambers, plasma-enhanced atomic layer deposition (PEALD) chambers, chemical vapor deposition (CVD) chambers, and plasma-enhanced chemical vapor deposition (PECVD) chambers.

[0046] In some embodiments, the gas delivery system 229 can be adapted to existing or conventional ampoules. Some embodiments relate to apparatus or methods that include a gas delivery system 229 configured to be adapted to existing or conventional ampoule caps.

[0047] Figures 2, 3A, and 3B illustrate a schematic representation of an automatic replenishment system 200 having an ampoule 202 connected to a gas delivery system 229. The automatic replenishment system 200 includes an exemplary ampoule 202 having a volume 201 configured to contain fluid within its internal volume, defined by sidewalls 214 and an inner surface 218.

[0048] As shown in Figures 3A and 3B, the gas delivery system 229 includes a gas supplier 279 that supplies propelled gas to the ampoule 202 via an inlet conduit 240 connected to an inlet port 220. In some embodiments, the gas supplier 279 contains a carrier gas 203, such as argon, helium, or nitrogen. In some embodiments, the carrier gas 203 is liquefied before entering the ampoule 202. The carrier gas 203 propels a precursor through an outlet conduit 250 for deposition in the processing chamber 283. In some embodiments, the actual gas supply source is a separate component and the gas supplier 279 is any suitable inlet connector that can be connected to the gas supply source, as will be familiar to those skilled in the art.

[0049] As shown in Figure 3B, the outlet port 241 of the inlet conduit 240 is below the liquid level surface 211a of the liquid precursor 211, such that the carrier gas 203 carries the liquid precursor 211 and pushes it into the outlet conduit 250. As shown in Figure 3B, in some embodiments, the gas supply 279 includes a main reservoir 282 located outside the processing chamber 283. The main reservoir 282 has a large volume ranging from 20 liters to 100 liters. A pushing pressure is maintained at the inlet 281 of the main reservoir 282, such that the carrier gas remains liquefied within the main reservoir 282, and the reservoir-side inlet 245 of the inlet conduit 240 is maintained below the carrier gas liquid level 247 within the main reservoir 282.

[0050] Referring back to Figure 3A, the gas delivery system 229 connected upstream of the gas supplier 279 can utilize a mass flow controller or a volume flow controller to regulate the flow rate of carrier gas 203 from the gas supplier 279 to the inlet pipe 240. In this embodiment, a main valve 224 is utilized. Upstream of the main valve 224, the gas delivery system 229 may further include a heater 226 configured to heat the gas supplier 279 to a desired temperature.

[0051] In some embodiments, the gas delivery system 229 further includes a converter or gas pressure sensor 228 configured to monitor gas pressure within the gas delivery system 229. Any suitable gas pressure sensor (such as a pressure gauge or pressure converter) is suitable for use in chemical vapor deposition or atomic layer deposition systems. In some embodiments, the gas delivery system 229 further includes a flow restrictor in the form of a needle valve 265 downstream of the ampoule 202, configured to variably adjust the gas pressure to a predetermined gas pressure value. The needle valve controls the gas / vapor flow rate to the processing chamber. Common types of flow restrictors include VCR orifice gasket flow restrictors, porous metal flow restrictors, and needle valves. However, it has been determined that only needle valves can be used to reliably deliver precursors to the membrane formation chamber, particularly for precursors that are prone to particle formation in gas delivery systems.

[0052] In one or more embodiments, the gas delivery system 229 includes a gas delivery controller 291 that communicates with a main valve 224, a heater 226, a gas pressure sensor 228, a needle valve 265, and a gas supply unit 279. In some embodiments, the gas delivery controller 291 is configured to send and receive liquid level signals from one or more liquid level sensors (150, 152, 154, 156) of the signal controller 170, and to enable one or more components of the gas delivery system 229. The gas delivery controller 291 is also configured to send and receive pressure signals from the gas pressure sensor 228, the pressure signals indicating the pressure of the gas delivery system 229.

[0053] In some embodiments, the gas delivery controller 291 is configured to enable or disable the gas supply 279. In some embodiments, the gas delivery controller 291 is configured to send a signal to the gas supply 279, and the gas delivery controller 291 is configured to increase or decrease the gas supply 279. Similarly, in some embodiments, the gas delivery controller 291 is configured to send a signal to the main valve 224, and the gas delivery controller is configured to enable or disable the main valve 224, and in some embodiments, the gas delivery controller 291 is configured to increase or decrease the main valve 224.

[0054] In some embodiments, the gas delivery controller 291 is configured to send a signal to adjust the needle valve 265 to change the gas pressure in the gas delivery system 229. In one or more embodiments, the needle valve 265 includes manual control, which includes, for example, a vernier handle configured to provide fine adjustment of the airflow through the needle valve. In some embodiments, the needle valve with the vernier handle provides fine control to allow precise adjustment of the airflow and pressure in the gas delivery system 229. In some embodiments, the vernier handle may include a scale (not shown) on the handle, which can be used to visually detect movement and adjustment of the vernier handle to allow precise adjustment of the gas pressure in the gas delivery system 229. In other embodiments, the needle valve 265 includes a motor control valve. Examples of motor control valves include motor-controlled actuator valves, such as pneumatic control valves, electro-controlled valves, hydraulic control valves, or piezoelectric controlled linear actuator valves. In one or more embodiments, the gas delivery controller 291 controls the operation of the motor control valve.

[0055] According to one or more embodiments, a gas delivery controller 291 includes a processor 293, a memory 295 coupled to the processor, input / output devices coupled to the processor 293, and support circuitry for communication between different components of the system or device, operation of the main valve 224, heater 226, needle valve 265, gas supply 279, and gas flow supply to the processing chamber 283. The gas delivery controller 291 communicates with a gas pressure sensor 225 and the needle valve 265 to regulate the gas pressure in the gas delivery system 230. In some embodiments, the gas delivery controller 291 communicates with the gas pressure sensor 225 and the needle valve 265, and the gas delivery controller 291 is configured to send a signal to adjust the needle valve 265 to change the gas pressure in the gas delivery system 230. In embodiments where the needle valve 265 is a manually controlled needle valve, the signal may be an alarm to indicate that the needle valve 265 needs adjustment to regulate or change the gas pressure in the gas delivery system. In a specific embodiment where the needle valve 265 includes a motor control valve, the gas delivery controller 291 adjusts the opening and closing of the needle valve 265. In this embodiment, the gas delivery controller 291 is configured to automatically operate the needle valve 265 to regulate or change the gas pressure in the gas delivery system 230, for example, to regulate or change it to a predetermined gas pressure value.

[0056] The process for operating the automatic replenishment system 200 can typically be stored as a software routine in memory such as the following: sensor controller 170, gas delivery controller 291, or other controllers. When executed by a processor, this software routine causes the automatic replenishment system 200 to perform the methods described in this disclosure. The software routine can also be stored and / or executed by a second processor (not shown) located at a hardware remotely controlled by the processor. Some or all of the methods disclosed herein can also be executed by hardware. Therefore, the methods described in this disclosure are implemented in software and executed by hardware using a computer system, as, for example, application-specific integrated circuits or other types of hardware implementations, or as a combination of software and hardware. When executed by a processor, the software routine converts a general-purpose computer into a dedicated computer (controller) that controls the chamber operation, enabling the process to be executed.

[0057] In some embodiments, the sensor controller 170 and the gas delivery controller 291 are a single controller having a processor and memory coupled to the processor, input / output devices coupled to the processor, and support circuitry for providing communication between different components of the automatic replenishment system 200.

[0058] Some or all of the methods disclosed herein can also be implemented in hardware. Therefore, the methods described herein are implemented in software and executed by a computer system using hardware, as, for example, application-specific integrated circuits or other types of hardware implementations, or as a combination of software and hardware. When executed by a processor, the software typically converts a general-purpose computer into a dedicated computer (controller) that controls the chamber operation, enabling the process to be executed.

[0059] The memory 174 of the sensor controller 170 and the memory 294 of the gas delivery controller 291 in one or more embodiments include one or more of the following: temporary memory (e.g., random access memory) and non-temporary memory (e.g., storage). The processor's memory may be one or more of the following: readily available memory, such as random access memory (RAM), read-only memory (ROM), floppy disk, hard disk, or any other form of digital storage, local or remote. The memory may store an instruction set that can be operated by the processor to control system parameters and components. Support circuitry is coupled to the processor to support it in a conventional manner. For example, the circuitry may include cache memory, a power supply, clock circuitry, input / output circuitry, subsystems, and the like.

[0060] In one or more embodiments, the sensor controller 170 and the gas delivery controller 291 execute instructions to deliver the flow rate of the precursor or reactant gas from the ampoule 202 to the processing chamber 283.

[0061] As shown in Figure 2, the automatic replenishment system 200 further includes control software 190. The control software 190 performs various tasks associated with replenishing the ampoules within the automatic replenishment system 200. For example, and not limited to, the control software 190 can control the operation of sensor controller 190, gas delivery controller 291, and one or more components of the automatic replenishment system 200, including liquid level sensors (150, 152, 154, 156), body valve 224, heater 226, gas pressure sensor 228, needle valve 265, and gas supply unit 279. Each of these hardware items may have a dedicated or open-source interface, and the control software 190 may also, or alternatively, manage communication with these hardware items, such as by interpreting data from the hardware or providing control signals to the hardware. At a more abstract level, control software 190 can coordinate various components of the automatic replenishment system 200 to enable or disable components such as the liquid level sensors (150, 152, 154, 156), main valve 224, heater 226, gas pressure sensor 228, needle valve 265, or gas supply 279, or more thereof. Control software 190 may also provide an external programming interface for controlling the automatic replenishment system 200 and the heater 226, needle valve 265, gas supply 279, and the flow rate of the gas being processed, and may or may not provide information to a wider computer infrastructure, such as event logs, status information, and the like. As will be described in more detail below, control software 190 may employ neural networks to calculate the liquid level of the precursor within the ampoule and to calculate the state of the state machine. Control software 190 may also provide a graphical user interface for users to interact with the automatic replenishment system 200 and related process data. More generally, the control software 190 may support any software functions associated with the status, monitoring, maintenance, evaluation, programming, control, and / or operation of the automatic replenishment system 200, whether relating to a specific device or component of the automatic replenishment system 200 or to the automatic replenishment system 200 as a whole.

[0062] Some embodiments disclosed herein relate to a method of controlling an automatic replenishment system 200 for precisely replenishing ampoules. Referring now to Figure 4, in one or more embodiments, at operation 302, method 300 includes controlling the operation of an automatic replenishment system having a plurality of liquid level sensors. In some embodiments, each of the plurality of liquid level sensors is positioned at a known location perpendicular to the height of the inner surface of the ampoule, each liquid level sensor being configured to detect a liquid level in the ampoule; each of the plurality of sensors corresponds to a replenishment function of a component of the automatic replenishment system. In some embodiments, the automatic replenishment system includes one or more sensor controllers and mass flow controllers of the embodiments described above.

[0063] At operation 304, the method further includes receiving a signal from a first liquid sensor of a plurality of liquid level sensors to a sensor controller, and at operation 306, receiving a signal from a second liquid sensor of the plurality of liquid level sensors to the sensor controller. The first liquid sensor is positioned at a first distance from the bottom of the ampoule and is configured to determine a low liquid level state of the ampoule. The second liquid sensor is positioned at a second distance from the bottom of the ampoule and is configured to determine a full liquid level state of the ampoule.

[0064] At operation 308, the method further includes a supplementary operation of determining each of the plurality of sensors, wherein the sensor controller sends a signal to the gas delivery controller, the signal of the first liquid sensor is configured to prompt the gas delivery controller to enable the mass flow controller, and the signal of the second liquid sensor is configured to prompt the gas delivery controller to deactivate the mass flow controller. A mass flow controller (MFC) is used to describe the operation of the various embodiments of this disclosure. However, those skilled in the art will recognize that the scope of this disclosure is not limited to MFC and other gas delivery systems (e.g., diaphragms, ALD valves) can be used. At operation 310, the method further includes prompting the gas delivery controller to enable the mass flow controller when the liquid level is at a low liquid level L, and at operation 312, the method further includes prompting the gas delivery controller to enable the mass flow controller when the liquid level is at a liquid level H.

[0065] In some embodiments, the sensor controller is configured to send and receive liquid level signals from one or more liquid level sensors to indicate when a certain liquid level has been reached. A first liquid level sensor and a second liquid level sensor are configured to indicate the activation and deactivation of a mass flow controller. The first liquid level sensor is configured to indicate a signal regarding the need to replenish an ampoule and sends that signal to the sensor controller, thus prompting the controller to activate the mass flow controller, and the second liquid level sensor is configured to indicate a signal regarding the replenishment of an ampoule and send that signal to the sensor controller, thus prompting the sensor controller to deactivate the mass flow controller.

[0066] In some embodiments, the method further includes operating the automatic replenishment system with a state machine. In some embodiments, the state machine includes a plurality of states associated with a plurality of operations of the sensor controller. Referring to Figures 5 and 6, the plurality of states of the state machine correspond to the liquid level of the ampoule.

[0067] Figure 5 illustrates a schematic representation of at least the liquid level sensors (150, 152, 154, 156) of Figure 1 having corresponding liquid level heights within volume 401 of the exemplary ampoule 400. Figure 6 illustrates a plurality of states of a state machine corresponding to the liquid level height within volume 401 relative to the liquid level sensors (150, 152, 154, 156).

[0068] As shown in Figure 5, ampoule 400 includes a sidewall 414 having an inner surface 418. Liquid level sensors (150, 152, 154, 156) are positioned at vertical heights (H1, H2, H3, and H4) along the inner surface 418 of the sidewall 414. As used in this manner, the term "vertical" means a direction perpendicular to the bottom of the ampoule. Vertical height H1 corresponds to a liquid level at or below the bottom liquid level sensor 150. If the liquid level has reached or is below vertical height H1, the bottom liquid level sensor 150 is configured to send a signal to sensor controller 170. Vertical height H2 corresponds to a liquid level at or below the first liquid level sensor 154. If the liquid level has reached or is below vertical height H2, the first liquid level sensor 154 is configured to send a signal to sensor controller 170. Vertical height H3 corresponds to a liquid level at or above the second liquid level sensor 156. In some embodiments, the second liquid level sensor 156 at vertical height H3 is configured as a production control sensor, wherein, if enabled, it initiates a dump process, as explained in further detail below. If the liquid level has reached or is above vertical height H3, the second liquid level sensor 156 is configured to send a signal to the sensor controller 170. Vertical height H4 corresponds to a liquid level at or above the top liquid level sensor 152. If the liquid level has reached or is above vertical height H1, the top liquid level sensor 152 is configured to send a signal to the sensor controller 170. In some embodiments, the top liquid level sensor 152 at vertical height H4 is configured as a boundary sensor, wherein, if enabled, it initiates a dump and warning process, as explained in further detail below.

[0069] The vertical heights (H1, H2, H3, H4) of the liquid level sensors (150, 152, 154, 156) can be at any height along the sidewall of the ampoule. Various optimizations of the precursor chemical can be investigated as the liquid level changes to determine the vertical heights (H1, H2, H3, H4). In some embodiments, the vertical heights are selected to minimize precursor chemical decomposition by maintaining a minimum amount of precursor chemical in the ampoule. In some embodiments, the vertical heights (H1, H2, H3, H4) can be determined based on the growth per cycle of the precursor chemical as the temperature of the precursor chemical within the ampoule changes. In some embodiments, the vertical height H1 is selected to have a minimum amount of decomposition.

[0070] As shown in Figure 6, each of the previously mentioned liquid level heights in Figure 5 corresponds to a state of state machine 500. Figure 6 illustrates the state of state machine 500 under an exemplary single batch. A single batch is defined as the number of substrates processed under a complete process recipe to be run. Specifically, the number of substrates processed under a complete process recipe can be any number of substrates processed during a continuous substrate manufacturing process for a specific process recipe. Multiple batches can have the same or different process recipes. Multiple batches can comprise consecutive and uninterrupted batches of the same recipe with or without downtime. The ampoules disclosed herein can be supplemented before, after, between, or during single or multiple batches.

[0071] For a single batch, the state machine 500 can be in the following states, and the state machine 500 can prompt one or more components of the automatic replenishment system 200 to perform an action. The prompt can occur during, before, or after a single batch. When the liquid level is at or below the bottom liquid level sensor 150 (vertical height H1), the state machine 500 receives a signal liquid level LL and sets the state machine 500 to a low-low state. When the liquid level is below the first liquid level sensor 154 (vertical height H2), the state machine 500 receives a signal liquid level L and sets the state machine 500 to a fill-in state. When the liquid level is at or above the second liquid sensor 156 (vertical height H3), the state machine receives a signal liquid level H and sets the state machine 500 to a stop-fill state. When the liquid level is at or above the top liquid level sensor 152 (vertical height H4), the state machine receives a signal liquid level HH and sets the state machine 500 to a high-high state.

[0072] In a low-low state, in some embodiments, state machine 500 prompts gas delivery controller 291 to activate or increase one or more of the gas supply 279 and main valve 224 shown in Figures 2 and 3, thus replenishing ampoule 400. In some embodiments, in a low-low state, state machine 500 prompts gas delivery controller 291 to activate or increase one or more of the gas supply 279 and main valve 224 until the liquid level reaches a stop-fill state. In a fill-required state, state machine 500 prompts gas delivery controller 291 to activate or increase one or more of the gas supply 279 and main valve 224. If the liquid level has exceeded the vertical height H2, the state machine prompts gas delivery controller 291 to deactivate one or more of the gas supply 279 and main valve 224. Due to the time delay in the termination of flow of gas supply 279, in some embodiments, the liquid level exceeds height H2. In some embodiments, during the filling-in state, state machine 500 prompts gas delivery controller 291 to activate or increase one or more of gas supply units 279 and main valves 224 until the liquid level reaches the stop filling state. During the stop filling state, state machine 500 prompts gas delivery controller 291 to deactivate one or more of gas supply units 279 and main valves 224.

[0073] As explained in further detail below, in some embodiments, machine 500 prompts one or more of the sensor controller 170 or gas delivery controller 291 to warn, notify, or display a prompt indicating the state or liquid level of state machine 500 to the user interface. In some embodiments, in a low-low state, state machine 500 prompts one or more of the sensor controller 170 or gas delivery controller 291 to warn, notify, or display a prompt indicating that ampoule 400 is below the critical liquid level to the user interface. In some embodiments, in a high-high state, state machine 500 prompts one or more of the sensor controller 170 or gas delivery controller 291 to warn, notify, or display a prompt indicating that ampoule 400 is above the critical liquid level to the user interface.

[0074] In some processing scenarios, the automatic replenishment system 200, including state machine 500 and control software 190, simultaneously replenishes and monitors the liquid levels of multiple ampoules. In some embodiments, depending on the batch of substrates processed within processing chamber 283, the automatic replenishment system 200, including state machine 500 and control software 190, implements and activates method 300 and state machine 500. Therefore, as explained in further detail below, the automatic replenishment system 200, including state machine 500 and control software 190, activates method 300 and state machine 500 before, during, or in the middle of a predetermined batch of substrates being processed.

[0075] As shown in Figure 7, each of the previously mentioned liquid level heights in Figure 5 corresponds to a state of state machine 600, which integrates the machine states of state machine 500 in Figure 6. Figure 7 illustrates the machine states of state machine 600 under multiple batches during multiple substrate manufacturing processes in processing chamber 283. Multiple batches may have the same or different process formulations. Multiple batches may comprise consecutive and uninterrupted batches of the same formulation with or without downtime. The ampoules disclosed herein may be supplemented before, after, between, or during single or multiple batches. Processing chamber 283 is configured to process substrates, transfer substrates, start or stop a specific batch formulation, and process one or more substrates for a total batch count.

[0076] For multiple batches with a total batch count, the state machine 600 can be in the following states, and the state machine 600 can prompt one or more components of the automatic replenishment system 200 to perform an action. The prompt can occur during, before, or after one or more periods of the multiple batches, for a duration equal to the total batch count.

[0077] If state machine 500 is in a low-low state (liquid level LL, vertical height H1) and processing chamber 283 is transferring the substrate, then state machine 600 is placed in the batch start filling state. Alternatively, if state machine 500 is in the fill-required state (liquid level L, vertical height H2) and processing chamber 283 is transferring the substrate and processing chamber 283 is at least partially completed with a predetermined batch counter, then state machine 600 is placed in the batch start filling state. The batch start filling state of state machine 600 prompts gas delivery controller 291 to activate or increase one or more of gas supply units 279 and main valve 224 until the liquid level has reached the stop filling state (above or at liquid level H) or has reached the batch stop filling state. In some embodiments, if state machine 500 is in a low-low state, gas delivery controller 291 will send a warning indicator to the user interface, but processing of processing chamber 283 will continue uninterrupted. In some embodiments, the low-low state prompts gas delivery controller 291 to send a warning indicator to the user interface and terminate processing of 283.

[0078] If state machine 500 is between liquid level L and liquid level H (between vertical height H2 and vertical height H3), then state machine 600 is placed in the batch stop filling state. Alternatively, if processing chamber 283 is starting but formulation has not yet begun, then state machine 600 is placed in the batch stop filling state. Alternatively, if state machine 500 is between liquid levels L1 and above liquid level L (between vertical height H1 and above vertical height H2), then state machine 600 is placed in the batch stop filling state. The batch stop filling state of state machine 600 prompts gas delivery controller 291 to deactivate or reduce one or more of gas suppliers 279 and main valves 224.

[0079] In some embodiments, if the liquid level exceeds the vertical height H3, or if the sensor at the vertical height H3 is activated, state machine 500 is placed in a dump state, wherein state machine 500 prompts gas delivery controller 291 to discharge liquid from ampoule 100 and gas delivery controller 291 sends a warning indicator to the user interface, but processing in processing chamber 283 continues uninterrupted. In some embodiments, the dump state sends a warning indicator to the user interface and terminates processing in 283.

[0080] As previously mentioned, if the liquid level exceeds the vertical height H2, the state machine prompts the gas delivery controller 291 to deactivate one or more of the gas supply 279 or the main valve 224. Due to the time delay in the termination of flow at the gas supply 279, in some embodiments, the liquid level exceeds the vertical height H2. In some embodiments, if the processing chamber 283 is processing a substrate and the liquid level is at height H2, the state machine 500 is placed in a batch intermediate fill state. The batch intermediate fill state of the state machine 600 prompts the gas delivery controller 291 to deactivate one or more of the gas supply 279 and the main valve 224, but does not terminate the processing chamber 283 from processing the substrate. In some embodiments, the batch intermediate fill state of the state machine 600 further initiates a batch count stored in a data structure to store precursor-specific data to count the number of substrates processed until the liquid level has exceeded the vertical height H2. In some embodiments, as the liquid level decreases due to continuous processing in the processing chamber 283, if the liquid level falls below the vertical height H2, the state machine 600 prompts the gas delivery controller 291 to deactivate one or more of the gas supply 279 and the main valve 224. In some embodiments, if the intermediate fill state of a batch is reached more than three times consecutively, the gas delivery controller 291 will send a warning indicator to the user interface, but processing in the processing chamber 283 will continue uninterrupted. The batch count of the intermediate fill state provides a count for determining the number of substrates processed between intermediate fills in the intermediate fill state of a batch.

[0081] Figures 8A and 8B illustrate state machines 500 and 600, respectively. It will be understood that the illustrated state machines (500, 600) are simplified representations, and that state machines for real-time control of semiconductor processes are more complex and have significantly more states and transitions compared to those shown in the figures. As previously shown in Figure 6, state machine 500 includes several states, including a low-low state, a fill-in-demand state, a fill-out-stop state, and a high-high state. Similarly, state machine 600 integrates at least one state of state machine 500, and state machine 600 includes a batch start fill-in state and a batch stop fill-in state. In some embodiments, state machine 600 further includes an optional batch intermediate fill-in state.

[0082] For example, and as previously explained, a state can represent one or more states of the liquid level sensors (150, 152, 154, 156), the operation of the main valve 224, heater 226, needle valve 265, gas supply 279, and the flow rate of gas to the processing chamber 283. Each change from one state to another occurs via a transition, such as the first transition 510 of state machine 500 or the first transition 610 of state management 600. Note that each state can have one or more transitions into and out of that state. This can be, for example, by a hardware item triggering a response to a control signal or sensor output to transition to a different state.

[0083] Figures 8A and 8B illustrate a simple state machine; however, other techniques can be used to represent such state machines. For example, state machines 500 and 600, represented by the state tables in Figures 6 and 7, respectively, have states and conditions presented in tabular form. In some embodiments, state machines 500 and 600 provide binary conditions for transitions, while, for example, state machine 600 provides a generalized expression for evaluating state changes. Other commonly used representations of state machines in software implementations include algorithmic state machines, Unified Modeling Language (UML) state diagrams, directed graphs, and so on.

[0084] Figure 9 illustrates a control system 750 including a neural network 700 configured to receive input from and send output to processing hardware 752. In some embodiments, processing hardware 752 includes one or more of a sensor controller 170, liquid level sensors (150, 152, 154, 156), a gas delivery controller 291, a main valve 224, a heater 226, a gas pressure sensor 228, a needle valve 265, and a gas supply 279. In some embodiments, neural network 700 is further configured to receive user input from a user input interface 198. For example, user input may include precursor-specific parameters, process recipes, and batch counts. In some embodiments, neural network 700 is configured to send output to a user output interface 199. For example, user output may include a graphical user interface for human monitoring and control. It will be understood that the depicted neural network is a generalized representation. In normal operation, the processing hardware 752 provides input 754 to the neural network 700, and the neural network 700 then calculates one or more states of one or more finite state machines (state machine 500, state machine 600) of the previous embodiments. These states then provide control signals to the processing hardware 752 via output 756 from the neural network to the processing hardware 752.

[0085] Input 754 to neural network 700 may include any data derived from processing hardware 752, such as sensor data from optical sensors, pressure gauges, switches, etc. In some embodiments, input 754 may also include robotic or processing data from processing chamber 283, such as the recipe being processed and batch counts. In some embodiments, the processing chamber is a component of processing hardware 752. In some embodiments, input 754 may also include processed sensor data, wherein data processing occurs by separate components of processing hardware 752. For example, converting separate components of processing hardware 752 into Boolean representations of component states. In some embodiments, input 754 to neural network 700 includes system information, such as pick-up time, processing state, substrate transition time, or substrate batch, etc. Broadly speaking, input 754 is in the form of any raw or processed data available from processing hardware 752.

[0086] In some embodiments, the inputs 754 and outputs 756 of the control system may include vectors, real numbers, complex numbers, and the data may be represented as integers, floating-point values, or any other numerical representation suitable for use with the neural network 700. In some embodiments, the inputs 754 and outputs 756 are asynchronous 608.

[0087] As shown, neural network 700 includes a three-layer network of objects. Specifically, at the input layer, the top level or top layer 702 of neural network 700 receives input data or input signals from sensor controller 170, and in some embodiments, from user input interface 198. Output can be generated at the output layer by means of the bottom level and sent to gas delivery controller 291, and in some embodiments, to user output interface 199.

[0088] In embodiments, each of the objects may be implemented on: a single-processor computer; multiple single-processor computers; a single multiprocessor computer; multiple multiprocessor computers, etc. In some embodiments, neural network 700 provides parallel computation of input vectors to output vectors across multiple objects. In some embodiments, object nodes may use algorithms for evaluation or computation. Evaluation can be performed locally in parallel or sequentially on a single computer or distributed across multiple computers. The architecture may employ multi-threaded, multi-tasking, distributed computing, batch computing, time-sharing computing, remote procedure call, client-server computing, peer computing, SIMD computing, MIMD computing, SISD computing, MISD computing, service-oriented architecture, n-tier architecture, monolithic software architecture, modular software architecture, or other architectures known in the art.

[0089] Each level (or layer) contains objects, which may contain any number of artificial neurons. In some embodiments, any number of objects are present in the input layer 702, the output layer 706, and the intermediate layer 704. In some embodiments, the number of objects at the input layer 702 corresponds to the number of input components, and the number of objects at the output layer 706 corresponds to the number of output components. In some embodiments, for example, the number of objects in the intermediate layer 704 is the average number of objects in the input layer 702 and the output layer 706.

[0090] In one or more embodiments, each object in one or more layers (702, 704, 706) of the neural network 700 is fully connected to objects in adjacent layers. In other words, the output from an individual object in input layer 702 is provided as input to each and every object in intermediate layer 704. Similarly, the output from each and every object in intermediate layer 704 is provided as input to each and every object in output layer 706.

[0091] In some embodiments, neural network 700 will provide prompts to finite state machines, such as state machines 500 and 600 of the previously described embodiments. In some embodiments, neural network 700 is a neural network scheduler. In some embodiments, the neural network scheduler operates to process inputs and compute the states of one or more of state machines 500 and 600. The states of one or more of state machines 500 and 600 may employ any current value, signal, or data from the neural network scheduler to determine whether conditions are required for a transition from one state to another, and generate any suitable control signals to the processing hardware 752.

[0092] In some embodiments, the state machine disclosed herein may employ multiple parallel states. In some embodiments, the state machine disclosed herein includes multiple sub-state machines for independent components of the automatic replenishment system 200, the sub-state machines evaluating one or more tasks of state machine 500 or state machine 600 for discrete grouping of hardware.

[0093] Figure 10 illustrates a process 800 for controlling an automatic replenishment system 200 using a finite state machine 801 and a neural network 700 of one or more previously described embodiments. In some embodiments, the finite state machine 801 is one or more state machines (500, 600) of the previously described embodiments.

[0094] As shown in step 802, neural network 700 receives data from automatic replenishment system 200 and applies the data input to 700. As shown in step 804, neural network 700 then processes nodes to calculate outputs, as shown in step 806. As shown in step 808, finite state machine 801 evaluates the current state to determine whether to transition to another state using the output from neural network 700, as shown in step 810. If, at step 810, it is determined that the transition is appropriate, finite state machine 801 then proceeds to change to the new state, as shown in step 812, and generates a corresponding control signal for output to system 730. If, at step 810, the transition is inappropriate, finite state machine 801 returns to re-evaluate the current state at step 808. In some embodiments and as previously described, the transition from one state to another may be a conditional transition, and therefore the state may conditionally transition to another state when the pressure in the internal chamber of the processing tool is below a threshold, wherein the threshold is represented by weights from the neural network. Similarly, in some embodiments, transitions are Bollinger values ​​and / or expressions or sensor data or signals. In other embodiments, weights may represent values ​​assigned to several possible transitions from a state.

[0095] In some embodiments, at step 814, the automatic replenishment system 200 receives a control signal generated and transmitted by state machine 801. In some embodiments, at step 816, the automatic replenishment system 200 continuously executes the control signal, and in some embodiments, continuously generates data output 818. It will be understood that the timing of these steps may be continuous or periodic and synchronous or asynchronous.

[0096] Figure 11 illustrates a data structure 900 for storing precursor-specific data. The data structure 900 may generally include a precursor identifier 910, a precursor pressure map 915, a process history 920, a recipe identifier 930, a batch identifier 940, a batch counter 945, and real-time data 950.

[0097] As described herein, data structure 900 can be implemented as an XML database, a relational database, an object relational database, or any other database or data structure suitable for storing information. Data structure 900 can be embodied in volatile memory (such as memory allocated for executing control programs), or in non-volatile memory (such as a manufacturing facility data repository), or a combination thereof. Storing wafer data in a single data structure offers significant advantages, including easier access to wafer-specific data and greater portability of wafer data, including, for example, the data described below.

[0098] In some embodiments, data structure 900 includes a precursor identifier 910 and a precursor pressure map 915, which identify and / or describe the precursor supplied by gas supplier 279. This may include numbers or other codes that uniquely identify the wafer globally or in a particular manufacturing facility, wafer set, or other suitable context. The precursor identifier 910 may also include descriptive information, such as the chemical properties of the precursor.

[0099] In some embodiments, pressure data can be used to match chamber-to-chamber chemical delivery across multiple chambers for a given precursor. Specifically, pressure profile 915 can be applied in batches to one or more gas delivery controllers 291 and processing chambers 283. For example, and not limited to, in two exemplary processing chambers with different precursor gases, different flow restrictors, and gas line lengths, matching a given pressure profile 915 with the parameters and the amount of chemical delivered from gas supplier 279 and mass flow controller 244 establishes uniformity between the two processing chambers with different precursor gases. In other words, by matching different pressures across different precursors above gas pressure sensor 228, the amount of chemical delivered from gas supplier 279 and mass flow controller 244 can be adjusted to achieve uniformity and control of the multiple processing chambers.

[0100] In some embodiments, data structure 900 includes process history 920, which contains event logs, recipes, process steps, and the like for one or more processing chambers. Therefore, the complete or partial history of one or more processing chambers is stored in data structure 900 for subsequent review or archiving.

[0101] In some embodiments, data structure 900 includes a formulation 930 for processing one or more substrates. This is a current or intended formulation suitable for a particular substrate or chamber process. Where the substrate is one of several substrates intended for batch or other grouped processes, formulation 930 may contain a reference to the location of external (from the data structure) data containing the formulation, such that a single formulation can be shared among several processing chambers.

[0102] In some embodiments, data structure 900 includes batch identifier 940 and batch counter 945 storing batch process or quantity or batch given process.

[0103] In some embodiments, data structure 900 includes real-time data 950 of precursor properties stored in precursor identifier 910. This data may include several fields for various data types, such as temperature, pressure, replenishment amount during a given batch or formulation, such as in a related table.

[0104] It will be understood that other fields are possible, and each field may contain a single data item, a series of data items, or a reference to a table or external data source containing the corresponding data. It will be further understood that data may be embodied in object-oriented data structures, XML data structures, relational databases, or any other form suitable for storing and retrieving data from its fields. As a significant advantage, data structure 900 can encapsulate any relevant data about the wafer and can be easily ported to any other environment. Therefore, data structure 900 can be stored in a manufacturing facility data repository, where it can be used in the current process and any subsequent processes. Data structure 900 can be transported with the wafer to other manufacturing facilities and used by various programs to analyze the wafer, including wide-facility programs, to calculate yield, throughput, or similar, and the specific handling / processing system in which the wafer will be processed. Other advantages will become clear from the above, including the ability to maintain a physical model of the wafer, which allows for more efficient and / or accurate processing decisions during the manufacturing process.

[0105] Figure 12 illustrates a system 1000 for controlling an ampoule replenishment process using precursor and processing information. In system 1000, an automatic, semi-automatic, or manual process, or a combination thereof, guides a gas delivery system 1002 through one or more replenishment components 1004. Data can be collected throughout the process and forwarded to a database 1010, where the data can be stored together with or otherwise indexed by a data structure associated with the gas delivery system 1002. In some embodiments, data such as vertical liquid levels, the status of one or more components, pressure, temperature, gas delivery volume, and the like are acquired from sensors within system 1000. In some embodiments, this data is forwarded to database 1010 and associated with other data from the gas delivery system 1002.

[0106] Each supplement 1004 can receive data and / or apply control signals (e.g., to control a robot or processing module) to the gas delivery system 1002. Records, objects, or other data structures can be generated in the database 1010 via the status of the gas delivery system 1002 or the supplement 1004. These data structures can be associated with or indexed by data from the data structure 900 described above.

[0107] In some embodiments, actions or measurements performed by the gas delivery system 1002, along with any instances of related data, relay data, or sensor data, are recorded in the database 1010. Any or all of these values, data, and relay data may be stored in the database 1010 associated with the gas delivery system 1002. It will be understood that although a linear system is depicted, the gas delivery system 1002 can deliver data from one supplement to another in any order.

[0108] In one configuration, database 1010 can be maintained in real time. Database 1010 may contain information about the status of supplementary component 1004 and / or the gas delivery system 1002 itself, and any other relevant data. Database 1010 can be used in a user interface or supplementary management system. Database 1010 is readily available. In embodiments, database 1010 may be stored using an SQL database management system, an XML database management system, or any other database management system or data storage system. In any case, the database may be stored off-site and remote from the system (database 1010).

[0109] Figure 13 illustrates an exemplary software system 1100, which includes a graphical user interface (GUI) 1102, a plurality of interfaces 1104, a plurality of black-box loggers 1108, a plurality of application logic 1110, and a plurality of APIs 1112. As shown, the software modules are arranged in a stack. In some embodiments, the modules are grouped together to form one or more applications 1114. The applications 1114 can communicate via the software modules of interface 1104.

[0110] As shown, in some embodiments, there exists a GUI 1102 application 1114, a logic application 1114 containing application logic 1110, and an application 1114 containing application logic 1110 and one or more APIs 1112. One or more black-box loggers 1108 generate data 1118. In some embodiments, the data 1118 includes information about the state of the software module. In some embodiments, the information corresponds to data or information sent and received between modules. In some embodiments, the information is in the form of alarms, error messages, status messages, or other messages about the operation of the software module. In some embodiments, one or more black boxes 1108 access the software module and retrieve data or information from the software module. In some embodiments, this is implemented as a method call to the software module. In embodiments, the black box 1108 and the software module are implemented as states in a state-oriented programming algorithm. The data 1118 can be used to help determine the state of the system.

[0111] Figure 14 illustrates a network for interconnecting processing hardware. System 1200 includes a plurality of controllers 1204, which in some embodiments employ any of the control software and / or data structures described in previous embodiments. In some embodiments, controllers 1204 are connected to one or more components 1202 (e.g., controllers, valves, etc.) or sensors described herein, including other processing hardware using a shared physical medium 1206. In some embodiments, for example, the shared medium 1206 is hardwired or serially connected, such that the controllers and / or components 1202 are interconnected using a single communication bus.

[0112] Throughout this specification, references to "one embodiment," "some embodiments," "various embodiments," "one or more embodiments," or "an embodiment" mean that a particular feature, structure, material, or characteristic described in connection with an embodiment is included in at least one embodiment of this disclosure. Therefore, the appearance of phrases such as "in one or more embodiments," "in some embodiments," "in various embodiments," "in one embodiment," or "in one embodiment" in various places throughout this specification does not necessarily refer to the same embodiment of this disclosure. Furthermore, a particular feature, structure, material, or characteristic may be combined in any suitable manner in one or more embodiments.

[0113] Although the disclosure herein provides a description with reference to specific embodiments, it will be understood that these embodiments are merely illustrative of the principles and applications of this disclosure. Those skilled in the art will readily recognize that various modifications and variations can be made to this disclosure without departing from its spirit and scope. Therefore, this disclosure is intended to include modifications and variations within the scope of the appended claims and their equivalents.

[0114] 52: Precursor Supply 54: Inlet pipe 55: Entrance Port 56: Pressure sensor 58: Inner End 60: Injector 100: Ampoule 101: Volume 110: Container 112: Bottom 114: Sidewall 116: Cover 120: Entrance Port 121:Pathway 130: Export Port 150: Bottom Liquid Level Sensor 152: Top Liquid Level Sensor 154: First liquid level sensor 156: Second liquid level sensor 160: Opening 162: O-ring 170: Sensor Controller 172: Processor 174: Memory 190: Control software 198: User Input Interface 199: User Output Interface 200: Automatic Replenishment System 202: Ampoule 211: Liquid precursors 211a: Liquid level surface 214: Sidewall 218: Inner surface 220: Entrance Port 224: Main valve 225: Gas pressure sensor 226: Heater 228: Gas pressure sensor 229: Gas delivery system 230: Gas delivery system 240: Inlet pipe 245: Tank side inlet 247: Carrier Gas Liquid Level 250: Export Pipeline 265: Needle valve 279: Gas Supply Unit 281: Entrance 282: Main storage tank 283: Processing Chamber 291: Gas delivery controller 293: Processor 295: Memory 300: Method 302: Operation 304: Operation 306: Operation 308: Operation 310: Operation 312: Operation 400: Ampoule 401: Volume 414: Sidewall 418: Inner surface 500: State Machine 510: The First Transformation 600: State Machine 610: The First Transformation 700: Neural Networks 702: Input Layer 704: Intermediate Layer 706: Output Layer 750: Control System 752: Hardware handling 754: Input 756: Output 800: Process 801: Finite State Machine 802: Steps 804: Steps 806: Steps 808: Steps 810: Steps 812: Steps 814: Steps 816: Steps 818: Data Output 900: Data Structure 910: Precursor Identifier 915: Precursor Pressure Diagram 920: Process History 930: Recipe Identifier 940: Batch Identifier 945: Batch Counter 950: Real-time Information 1000: System 1002: Gas delivery system 1004: Supplementary Part 1010: Database 1100: Software System 1101: 1104: Interface 1108: Black Box Recorder 1110: Application Logic 1114: Application 1128: 1200: System 1202: Component 1204: Controller 1206: Media H1: Vertical height H2: Vertical height H3: Vertical height H4: Vertical height

[0115] Domestic storage information (please note in order of storage institution, date, and number) none Overseas storage information (please note in the order of storage country, institution, date, and number) none

Claims

1. A method for replenishing an ampoule, the method comprising the steps of: controlling the operation of an automatic replenishment system having a plurality of liquid level sensors, each of the plurality of liquid level sensors being positioned at a known location perpendicular to a height along an inner surface of the ampoule, each of the plurality of liquid level sensors being configured to send a signal to a sensor controller for detecting a liquid level in the ampoule; the automatic replenishment system comprising the sensor controller and a gas delivery controller; receiving a signal from a first liquid sensor of the plurality of liquid level sensors to the sensor controller, the first liquid sensor being positioned at a first distance from a bottom of the ampoule, the first liquid sensor being configured to determine a liquid level (L) of the ampoule; receiving a signal from a second liquid sensor of the plurality of liquid level sensors to the sensor controller, the second liquid sensor being positioned at a second distance from the bottom of the ampoule, the second liquid sensor being configured to determine a liquid level (H) of the ampoule. The system determines a replenishment operation for each of the plurality of liquid level sensors. The sensor controller sends a signal to the gas delivery controller, wherein the signal from the first liquid sensor is configured to prompt the gas delivery controller to activate a main valve, and the signal from the second liquid sensor is configured to prompt the gas delivery controller to deactivate the main valve; the main valve is activated when the liquid level is at the liquid level (L); and the main valve is deactivated when the liquid level is at a full liquid level. The operation of the automatic replenishment system is controlled by a state machine having one or more states, including a batch start filling state, a batch stop filling state, and a batch intermediate filling state. The batch start filling state corresponds to one or more of the following: a liquid level at a vertical height (H1) and a processing chamber connected to an outlet of the ampoule is transferring a substrate; a liquid level at a vertical height (H2) and a processing chamber connected to an outlet of the ampoule is transferring a substrate, and the processing chamber is at least partially completed using a predetermined batch counter.

2. The method as described in claim 1, wherein at least one of the one or more states represents a state of an item of hardware within the automatic replenishment system.

3. The method as described in claim 1, wherein at least one of the states represents a signal sent to the gas delivery valve or the main valve.

4. The method as described in claim 1, wherein at least one of the one or more states has a condition that depends on a liquid level within the ampoule.

5. The method as described in claim 1, wherein the state machine has a high-high state corresponding to a liquid level (H4), a stop-fill state corresponding to a vertical liquid level (H3), a fill-required state corresponding to a vertical liquid level (H2), and a low-low state corresponding to a vertical liquid level (H1), the fill-required state prompting the gas delivery controller to activate the main valve, and the stop-fill state prompting the gas delivery controller to deactivate the main valve.

6. The method as described in claim 1, wherein the batch start filling status prompts the gas delivery controller to activate the main valve until the liquid level is above or at the liquid level (H).

7. The method as described in claim 1, wherein the batch stop filling state corresponds to a liquid level between vertical height (H2) and vertical height (H3).

8. The method as described in claim 7, wherein the batch stop filling prompts the gas delivery controller to deactivate the main valve.

9. The method as described in claim 1, wherein the batch stop filling state corresponds to the start of a processing chamber, but a formulation has not yet been started.

10. The method of claim 1, wherein the batch stop filling state corresponds to a liquid level between and above the vertical height (H2).

11. The method of claim 1, wherein the intermediate batch filling state corresponds to a liquid level below a vertical height (H2) and a processing chamber is not transferring a substrate, and the intermediate batch filling prompts the gas delivery controller to deactivate the main valve.

Images