Thin film deposition system and method thereof

Abstract

A thin film deposition system and method thereof, the thin film deposition system includes a top plate positioned above a wafer and used to generate plasma during a thin film deposition process. The thin film deposition system includes a sensor, the sensor is used to generate a sensor signal indicating the life of a component of the thin film deposition system, the characteristics of a thin film deposited by the thin film deposition system, or the characteristics of a process material flowing into the thin film deposition system. The thin film deposition system includes a control system, the control system is used to adjust the relative position of the top plate of the thin film deposition system relative to the position of the wafer in the thin film deposition system during the thin film deposition process in response to the sensor signal.

Description

Technical Field

[0001] The present disclosure relates to semiconductor processing technology and more particularly to thin film deposition processes. Background Art

[0002] The manufacture of integrated circuits is typically accomplished by performing a large number of processing steps on a semiconductor wafer. The processing steps typically result in the formation of a large number of transistors in a highly complex configuration in conjunction with a semiconductor substrate. The processing steps also result in the formation of dielectric layers, metal interconnects, vias, plugs, and other integrated circuit structures and components.

[0003] For optimal device performance and production yield, the thin film deposited during the thin film deposition process preferably exhibits uniform properties. When the deposited thin film does not exhibit uniform properties, non-compliant devices are produced that may need to be discarded. In addition, time must be spent adjusting the process to minimize the number of non-compliant devices produced. Summary of the invention

[0004] According to some embodiments of the present disclosure, a method for depositing a thin film includes: performing a thin film deposition process on a wafer using a thin film deposition system; generating a plurality of sensor signals using sensors, the sensor signals indicating a lifetime of components of the thin film deposition system, characteristics of a thin film deposited by the thin film deposition system, or characteristics of a process material flowing into the thin film deposition system during the thin film deposition process; and adjusting a relative position of a top plate of the thin film deposition system relative to a position of a wafer in the thin film deposition system in response to the sensor signals during the thin film deposition process.

[0005] According to some embodiments of the present disclosure, a thin film deposition system includes a thin film deposition chamber, a wafer support, a top plate, a sensor, and a control system. The wafer support is positioned in the thin film deposition chamber and is used to support a wafer in the thin film deposition chamber. The top plate is positioned above the wafer support and is used to generate plasma in the thin film deposition chamber during a thin film deposition process. The sensor is used to generate a plurality of sensor signals indicating the life of a component of the thin film deposition system, the characteristics of a thin film deposited by the thin film deposition system, or the characteristics of a process material flowing into the thin film deposition system during a thin film deposition process. The control system is used to receive the sensor signal and adjust the relative position of the top plate of the thin film deposition system relative to the position of the wafer in the thin film deposition system in response to the sensor signal.

[0006] According to some embodiments of the present disclosure, a method for depositing a thin film includes: performing a plasma enhanced chemical vapor deposition process on a wafer in a thin film deposition chamber; generating plasma using a top plate positioned above the wafer during the plasma enhanced chemical vapor deposition process; generating a plurality of sensor signals indicating lifetimes of components of the thin film deposition system, properties of a thin film deposited by the thin film deposition system, or properties of process materials flowing into the thin film deposition system using sensors during the plasma enhanced chemical vapor deposition process; analyzing the sensor signals using a control system during the plasma enhanced chemical vapor deposition process; and adjusting a position of a top plate of the thin film deposition system relative to a position of a wafer in the thin film deposition system using a control system based on the sensor signals during the plasma enhanced chemical vapor deposition process. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 is a block diagram of a thin film deposition system according to one embodiment;

[0008] Figure 2 is a diagram of a thin film deposition system according to one embodiment;

[0009] Figure 3 A graph containing thin film deposition rates for two different wafers according to one embodiment;

[0010] Figure 4A is a block diagram of a control system according to one embodiment;

[0011] Figure 4B According to an embodiment Figure 4A The block diagram of the analytical model of the control system;

[0012] Figure 5 is a flow chart of a method for performing a thin film deposition process according to one embodiment;

[0013] Figure 6 is a flow chart of a method for performing a thin film deposition process according to one embodiment.

[0014]

Explanation of symbols

[0015] 100: Thin film deposition system

[0016] 101: Counter

[0017] 102: Thin film deposition chamber

[0018] 103: Process Material Sensor

[0019] 104: Wafer

[0020] 106: Wafer support / wafer support plate

[0021] 108: Top plate

[0022] 110: Gap

[0023] 112: Thin film sensor

[0024] 112a: Transmitter

[0025] 112b: Receiver

[0026] 113: Gap sensor

[0027] 113a: Radiant emitter

[0028] 113b: Radiation Sensor

[0029] 114: Control System

[0030] 116: Heater

[0031] 118: Bottom plate

[0032] 122: RF Power Supply

[0033] 124: Process material source / gas source

[0034] 126: Manifold

[0035] 128: Porosity

[0036] 130: Thin film deposition gas

[0037] 132: Exhaust channel

[0038] 134: Exhaust fluid

[0039] 136: Radiation

[0040] 136a: Radiation beam

[0041] 136b: Radiation / Radiation Amount

[0042] 137a~137b: Signal

[0043] 300:Charts

[0044] 302: Charts

[0045] 402:Analysis Model

[0046] 404: Training module

[0047] 406: Training set data

[0048] 408: Historical Film Data

[0049] 410: Historical sensor data

[0050] 412: Processing Resources

[0051] 414:Memory Resources

[0052] 416: Communication Resources

[0053] 452: Sensor signal vector

[0054] 454:Data Field

[0055] 456a: First neural layer

[0056] 456b: Second neural layer

[0057] 456c: Third neural layer

[0058] 456d: Neural layer

[0059] 456e: Final neural layer

[0060] 458: Node

[0061] 468:Data value

[0062] 470:Data value

[0063] 472: Error value

[0064] 500:Method

[0065] 502~508: Steps

[0066] 600: Method

[0067] 602~610: Steps

[0068] D: Distance DETAILED DESCRIPTION

[0069] In the following description, many thicknesses and materials of various layers and structures within the integrated circuit die are described. For various embodiments, specific dimensions and materials are given by way of example. Based on this disclosure, one of ordinary skill in the art will recognize that other dimensions and materials can be used in many cases without departing from the scope of this disclosure.

[0070] The following disclosure provides many different embodiments, or examples, for implementing the different features of the provided subject matter. Specific examples of components and configurations are described below to simplify the disclosure. Of course, these are merely examples and are not intended to be limiting. For example, in the following description, the formation of a first feature above or on a second feature may include an embodiment in which the first feature is formed in direct contact with the second feature, and may also include an embodiment in which an additional feature may be formed between the first feature and the second feature so that the first feature and the second feature may not be in direct contact. In addition, the disclosure may repeat reference numbers and / or letters in various examples. This repetition is for the purpose of simplicity and clarity, and does not itself indicate the relationship between the various embodiments and / or configurations discussed.

[0071] Additionally, for ease of description, spatially relative terms, such as "below," "beneath," "lower," "above," "upper," and the like, may be used herein to describe the relationship of one element or feature to another element or feature(s) illustrated in the figures. The spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. The device may be otherwise oriented (rotated 90 degrees or at other orientations), and the spatially relative descriptors used herein should be similarly interpreted accordingly.

[0072] In the following description, certain specific details are set forth to provide a thorough understanding of various embodiments of the present disclosure. However, one of ordinary skill in the art will appreciate that the present disclosure may be practiced without these specific details. In other instances, well-known structures associated with electronic components and manufacturing techniques are not described in detail to avoid unnecessarily obscuring the description of the embodiments of the present disclosure.

[0073] Unless the context requires otherwise, throughout the specification and the appended claims, the word "comprise" and variations such as "comprises" and "comprising" are to be interpreted in an open and inclusive sense, that is, to mean "including, but not limited to."

[0074] The use of ordinal numbers such as first, second, and third does not necessarily imply an ordering meaning, but may simply distinguish multiple instances of an action or structure.

[0075] References in this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least one embodiment. Therefore, the phrases "in one embodiment" or "in an embodiment" appearing in various places in this specification do not necessarily refer to the same embodiment. Furthermore, in one or more embodiments, particular features, structures, or characteristics may be combined in any suitable manner.

[0076] As used in this specification and the appended patent applications, the singular forms "a", "an", and "the" include plural references unless the content clearly dictates otherwise. It should also be noted that the meaning of the term "or" generally includes "and / or" unless the content clearly dictates otherwise.

[0077] Embodiments of the present disclosure provide a thin film deposition system with improved performance. The thin film deposition system includes a top plate positioned above a semiconductor substrate (e.g., a wafer, held by an electrostatic chuck or a wafer stage or pedestal). Embodiments of the present disclosure monitor the relative position of the top plate relative to the wafer and automatically adjust the top plate position or the wafer position or both in situ during the deposition process. This provides a stable deposition rate so that the deposited film has uniform properties from wafer to wafer and from batch to batch. As a result, fewer wafers are scrapped, downtime for scheduled maintenance is reduced, downtime for equipment adjustments is reduced, and the film is formed correctly.

[0078] Figure 1 1 is a block diagram of a thin film deposition system 100 according to one embodiment. The thin film deposition system 100 includes a thin film deposition chamber 102. During a thin film deposition process, a wafer 104 is supported by a wafer support plate 106 in the thin film deposition chamber 102. Process materials are provided to the thin film deposition chamber 102. In a plasma enhanced deposition process, plasma is generated in the thin film deposition chamber 102. The thin film deposition system 100 deposits a thin film on the wafer 104 during the thin film deposition process.

[0079] The thin film deposition system 100 includes a top plate 108 positioned above the wafer 104. The top plate 108 can assist in the thin film deposition process. For example, the top plate 108 can assist in generating a voltage or electric field within the thin film deposition chamber 102. The voltage or electric field can promote the deposition of a selected type of thin film on the wafer 104.

[0080] Wafer 104 is separated from top plate 108 by gap 110. Gap 110, i.e., the distance between the top surface of wafer 104 and the bottom surface of top plate 108, can have an impact on the thin film deposition process. If the gap is too large or too small, the thin film may not be deposited properly, or may have an undesirable deposition rate, whether it is too low or too high. When the thin film is not deposited properly, the deposited thin film may not have uniform properties across the wafer and / or the deposited thin film may not have uniform properties from wafer to wafer.

[0081] According to some embodiments of the present disclosure, the thin film deposition system 100 includes a thin film sensor 112. The thin film sensor 112 is used to generate a sensor signal based on the characteristics of the thin film on the wafer 104. According to other embodiments, the thin film sensor 112 detects the characteristics of the thin film deposited on the surface of the deposition chamber 102 of the thin film deposition system 100. Such a film thickness may be sensed during the processing of a single batch of wafers, or may be collected for the operation of different batches of wafers or for different operations of a portion of a single batch of wafers. The sensor signal from the thin film sensor 112 may also indicate an area on the surface of the wafer from which the characteristics are detected. Examples of different areas 1 to 4 are described below. The sensor signal indicates the characteristics of the thin film on the wafer 104. For example, the sensor signal indicates one or more of the following: film thickness, film composition, film crystal structure, film hardness, film thermal stability, film surface roughness, film band gap, or residual stress in the film. Sensors capable of generating sensor signals indicating these film properties are known and are not described in more detail here. Thin film sensors according to embodiments of the present disclosure are not limited to sensors capable of generating the above-described sensor signals. In other embodiments, thin film sensor 112 may be a sensor capable of generating sensor signals indicative of thin film properties other than the above-described properties. As described in more detail below, such sensor signals are used to determine where top plate 108 should be located, where wafer 104 should be located, or both.

[0082] The thin film deposition system also includes a gap sensor 113. The gap sensor 113 is used to generate a sensor signal based on the gap 110, for example, the size of the gap 110. The sensor signal from the gap sensor 113 may indicate the magnitude of the gap 110 or the relative size of the gap 110. In any case, the sensor signal from the gap sensor 113 may be used to determine whether the size of the gap 110 should be adjusted.

[0083] exist Figure 2 In one embodiment shown in , the gap sensor 113 may include a radiation emitter 113a that emits electromagnetic radiation 136a to the gap 110. The gap sensor 113 may also include a radiation sensor 113b that senses the radiation 136b emitted through the gap 110. The amount of radiation sensed by the radiation sensor 113b of the gap sensor 113 may indicate the size of the gap 110. The radiation sensor 113b may provide a sensor signal indicating the size of the gap 110. The radiation emitter 113a may emit electromagnetic radiation in the visible spectrum, infrared spectrum, ultraviolet spectrum, or another spectrum. The radiation emitter 113a may emit radiation in a laser beam or in another manner.

[0084] In one embodiment, the gap sensor 113 includes an image capture device that captures an image of the gap 110. The image capture device may capture visible light images, infrared images, ultraviolet images, or other types of images. The size of the gap 110 may be measured from the captured image. Various other types of gap sensors 113 may be used without departing from the scope of the present disclosure.

[0085] exist Figure 1 and Figure 2 , the gap sensor 113 is shown as being positioned within the thin film deposition chamber 102. However, in some embodiments, the gap sensor 113 may be positioned outside the thin film deposition chamber 102. For example, the gap sensor 113 may sense parameters associated with the gap 110 through one or more apertures or windows in the wall of the thin film deposition chamber 102. Various locations of the gap sensor 113 may be utilized without departing from the scope of the present disclosure.

[0086] In one embodiment, the thin film sensor 112 may include an ellipsometer, an x-ray photoelectron spectrometer (XPS), an ultrasonic sensor, an x-ray fluorescence (XRF) spectrometer, an optical emission spectrometer, an interferometric sensor. Such devices generally include a transmitter 112a that transmits a signal 137a that interrogates the thin film on the wafer 104 to determine the properties of the thin film on the wafer 104. The thin film sensor 112 may also include a receiver 112b that senses a signal 137b from the thin film. The signal 137b received by the receiver 112b of the thin film sensor 112 may indicate the above-mentioned properties of the thin film. The receiver 112b may provide a sensor signal indicative of such properties.

[0087] In one embodiment, the thin film sensor 112 includes an image capture device that captures an image of the deposited thin film surface. The image capture device can capture visible light images, infrared images, ultraviolet images, or other types of images. The captured images can be used to evaluate certain characteristics of the thin film. Various other types of thin film sensors 112 can be used without departing from the scope of the present disclosure.

[0088] exist Figure 1, the thin film sensor 112 and its transmitter 112a and receiver 112b are shown as being positioned at a particular location within the thin film deposition chamber 102. However, the transmitter 112a and the receiver 112b may be positioned at different locations or different angles within the thin film deposition chamber 102. Furthermore, in some embodiments, one or more of the transmitter 112a and the receiver 112b may be located outside the thin film deposition chamber 102. For example, the transmitter 112a or the receiver 112b or both may sense parameters associated with the thin film through one or more apertures or windows in the wall of the thin film deposition chamber 102. Various locations of the transmitter 112a and the receiver 112b may be utilized without departing from the scope of the present disclosure.

[0089] exist Figure 1 and Figure 2 In the illustrated embodiment, the thin film deposition system 100 includes a control system 114. The control system 114 is coupled to the thin film sensor 112, the gap sensor 113, the top plate 108, and the wafer support 106. The control system 114 receives sensor signals from the thin film sensor 112 and the gap sensor 113. The control system 114 can determine whether the position of the top plate 108, the wafer 104, or both should be adjusted based on the sensor signals from the thin film sensor 112 and the gap sensor 113. The control system 114 can adjust the relative position of the top plate 108 relative to the position of the wafer 104 by raising or lowering the top plate 108 relative to the wafer 104. Alternatively, the control system 114 can adjust the relative position of the top plate 108 relative to the position of the wafer 104 by raising or lowering the wafer support 106 relative to the top plate 108. In other embodiments, the relative position of the top plate 108 relative to the position of the wafer 104 may be adjusted by moving the top plate 108 laterally in a direction perpendicular to the direction in which the top plate 108 or wafer support 106 is raised or lowered. In other embodiments, the relative position of the top plate 108 relative to the position of the wafer 104 may be adjusted by moving the wafer support 106 laterally in a direction perpendicular to the direction in which the top plate 108 or wafer support 106 is raised or lowered.

[0090] In one embodiment, the control system 114 receives analog sensor signals from the film sensor 112 and the gap sensor 113. The control system 114 may convert the analog sensor signals into digital form by performing an analog-to-digital conversion. The control system 114 may then analyze the digitized sensor signals to determine the properties of the film and the absolute or relative magnitude of the gap 110. In one example, the control system 114 may compare the digitized sensor signals to reference sensor signals stored in one or more memories associated with the control system 114. The control system 114 may then adjust the relative position of the top plate 108 relative to the wafer 104, for example, by controlling a motor that raises or lowers the top plate 108, controlling a motor that raises or lowers the wafer support 106, controlling a motor that moves the top plate 108 laterally, or controlling a motor that moves the wafer support 106 laterally.

[0091] In one embodiment, the control system 114 receives digital sensor signals from the film sensor 112 and the gap sensor 113. The control system 114 may process and analyze the digital sensor signals received from the film sensor 112 and the gap sensor 113. The control system 114 may then determine the characteristics of the film and the magnitude of the gap 110, and determine whether the relative position of the top plate 108 with respect to the wafer 104 should be adjusted based on the analysis of the sensor signals.

[0092] exist Figure 1 and Figure 2In the embodiment shown in , the thin film deposition system 100 includes sensors in addition to or as a substitute for the thin film sensor 112. For example, the thin film deposition system 100 includes a counter 101 that tracks and determines the number of wafers that have been processed in the thin film deposition chamber 102 and generates a sensor signal indicating the number of wafers that have been processed. The sensor signal from the counter 101 indicating the number of wafers that have been processed in the thin film deposition chamber 102 represents the life of the thin film deposition system component. In other words, the counter 101 is an example of a sensor that can generate a sensor signal indicating the life of a component of the thin film deposition system. For example, the sensor signal from the counter 101 may indicate the life of the following components: the showerhead (e.g., how long the showerhead has been operated since the last inspection, scheduled maintenance, or replacement), the heater, the electrostatic chuck, the RF generator, the processing chamber, or other components of the thin film deposition system 100. Monitoring the number of wafers that have been processed will also indicate the length of time the chamber has been performing a process, that is, the chamber process time. In addition, sensors rather than counters may be used to monitor the life of components of a thin film deposition system. For example, a timer that counts the operating time and downtime of a thin film deposition system is another example of a sensor capable of generating a sensor signal indicating the life of a component of a thin film deposition system. For example, a timer may be used to monitor the number of operating hours of an RF generator. Such a timer may also be used to monitor the length of time a chamber runs a process. Sensors capable of generating a sensor signal indicating the life of a component of a thin film deposition system are not limited to the counters or timers described above. According to an embodiment of the present disclosure, sensors capable of generating a sensor signal indicating the life of a component of a thin film deposition system include sensors other than the counters or timers described above.

[0093] exist Figure 2 In the embodiment shown in , the thin film deposition system 100 also includes a process material sensor 103, which is capable of generating a sensor signal reflecting the characteristics of the process material flowing into the thin film deposition chamber 102. The process material characteristics reflected by the sensor signal generated by the process material sensor 103 include the composition of the process gas, the flow rate of the process gas, the temperature of the process gas, and the pressure of the process gas. Sensors that can generate an indication of the composition of the process gas, the flow rate of the process gas, the pressure of the process gas, or the temperature of the process gas are known and are not described in more detail herein. The process material sensor 103 is not limited to detecting the aforementioned characteristics of the process gas. Process gas sensors that can detect other characteristics of the process gas are within the scope of the present disclosure.

[0094] exist Figure 1 and Figure 2In the embodiment shown in , the thin film deposition system 100 is illustrated as including a thin film sensor 112, a counter 101, and a process material sensor 103; however, embodiments according to the present disclosure are not limited to a thin film deposition system 100 including all three sensors. In other embodiments, the thin film deposition system 100 includes less than all three sensors. Furthermore, in other embodiments, the thin film deposition system 100 includes sensors other than the thin film sensor 112, the counter 101, and the process material sensor 103. For example, the thin film deposition system 100 may include a sensor capable of generating a sensor signal indicating a tilt angle of a wafer on the support 106. Sensors capable of generating a sensor signal indicating a tilt angle of a wafer on the support 106 are known and are not described in further detail herein.

[0095] Figure 2 1 is a diagram of a thin film deposition system 100 according to one embodiment. The thin film deposition system 100 includes a thin film deposition chamber 102. The thin film deposition chamber 102 includes a wafer 104 supported on a wafer support 106. The thin film deposition system 100 deposits a thin film on a surface of the wafer 104 during a thin film deposition process.

[0096] In the example Figure 2 In the embodiment of the present invention, the thin film deposition system 100 is a plasma enhanced chemical vapor deposition (PECVD) system. The plasma enhanced chemical vapor deposition system can be used to deposit a dielectric layer on the wafer 104. The dielectric layer can include silicon oxide, silicon nitride, or other types of dielectric layers. The thin film deposition system 100 can be used to deposit other types of layers or perform other types of semiconductor processes other than those described above without departing from the scope of the present disclosure.

[0097] Wafer support 106 supports wafer 104. Wafer support 106 may include heater 116 or may function as a heater to heat wafer 104 during a thin film deposition process. Heating wafer 104 may facilitate deposition of a thin film on the surface of wafer 104. Heater 116 may include one or more heating coils that generate thermal energy that is transferred to wafer 104. Alternatively, heater 116 may include other types of heating elements for generating thermal energy and heating wafer 104. In some embodiments, support 106 includes a cooling element for removing thermal energy from wafer 104. More details regarding the heating function of wafer support 106 will be described in more detail below.

[0098] The wafer support 106 may also include a bottom plate 118. The bottom plate 118 may be a bottom electrode to facilitate the generation of plasma in the film deposition chamber 102, as described in more detail below. In one embodiment, the wafer 104 rests directly on the bottom plate 118. In other embodiments, the wafer 104 rests on pins attached to the bottom plate 118. The bottom plate 118 may include a conductive material, such as a metal or a conductive ceramic. In one embodiment, the heater 116 and the bottom plate 118 may be integrated together. In one embodiment, the bottom surface of the support 106 is connected to a pedestal, which includes a mechanism for raising, lowering, or laterally moving the wafer support 106. When the pedestal is raised, the wafer 104 is brought closer to the top plate 108. When the pedestal is lowered, the proximity of the wafer 104 relative to the top plate 108 increases. When the pedestal is moved laterally, the lateral position of the wafer 104 relative to the top plate 108 changes.

[0099] The thin film deposition system 100 includes a top plate 108 and a radio frequency power source 122. The top plate 108 acts as a top electrode for generating plasma during a thin film deposition process. The top plate 108 may include a conductive material, such as a metal or a conductive ceramic material.

[0100] During the thin film deposition process, the RF power source 122 provides an RF voltage between the bottom plate 118 and the top plate 108 . Figure 2 The electrical connections between the RF power source 122 and the bottom plate 118 and the top plate 108 are not shown. The RF power source 122 is also coupled to the control system 114. The control system 114 can control the functions of the RF power source 122. During the thin film deposition process, the control system 114 can cause the RF power source 122 to apply an RF voltage between the top plate 108 and the bottom plate 118.

[0101] The thin film deposition system 100 includes a process material source 124 and a manifold 126. The process material may be a gas, which includes a material to be deposited as a thin film. During the thin film deposition process, the process material source 124 supplies a deposition gas 130 to the thin film deposition chamber 102 through the manifold 126. The thin film deposition chamber 102 may include an aperture 128, which enables the deposition gas 130 to flow from the gas source 124 into the thin film deposition chamber 102. The thin film deposition gas 130 flows into the thin film deposition chamber 102, and in combination with other factors, causes a thin film to be deposited on the top surface of the wafer 104. The thin film deposition system 100 includes an exhaust channel 132, through which an exhaust fluid 134 is exhausted from the thin film deposition chamber 102.

[0102] In one embodiment, the thin film deposition system 100 includes a top cover. The top cover may include apertures 128 through which a deposition gas 130 flows into the thin film deposition chamber 102 during a thin film deposition process.

[0103] As mentioned above Figure 1 As stated, Figure 2 The thin film deposition system 100 shown in FIG. 1 also includes a process material sensor 103 for generating a sensor signal indicative of a characteristic of a process material delivered to the thin film deposition chamber 102 by a process material source 124 . Figure 2 The thin film deposition system 100 shown in FIG. 1 also includes the thin film deposition system 100 described above with reference to FIG. Figure 1 The counter 101.

[0104] exist Figure 2 In the example of FIG. 1 , the top plate 108 is shown as being separated from the components that allow the deposition gas 130 to flow into the thin film deposition chamber 102. However, in one embodiment, the top plate 108 may be a showerhead-type top electrode, which includes a plurality of pores in fluid communication with the gas source 124. In this case, the deposition gas 130 is output from the gas source 124 to the thin film deposition chamber 102 through the pores in the top plate 108. Other types of systems and components can be used to allow the deposition gas 130 to flow into the thin film deposition chamber 102 without departing from the scope of the present disclosure. As more and more wafers are processed through the thin film deposition system, the pores of the showerhead-type top electrode tend to become smaller. Because the process materials are deposited on the surface of the showerhead and / or in the pores in the showerhead, the pores tend to become smaller. Therefore, regular maintenance must be performed on the showerhead in order to clean the showerhead surface and remove the materials that cause the showerhead pores to become smaller. During the regular maintenance, the reduction in the size of the smaller pores in the showerhead must be compensated by increasing the flow rate of the process material.

[0105] During the PECVD process, a deposition gas 130 is introduced from a gas source 124 into the interior volume of the thin film deposition chamber 102. The RF power source 122 generates plasma from the deposition gas 130 within the interior volume by applying a voltage between the bottom plate 118 and the top plate 108. Specifically, a plasma is generated between the top surface of the wafer 104 and the bottom surface of the top plate 108. The plasma enhances the deposition of the thin film on the wafer 104. Embodiments according to the present disclosure are not limited to plasma enhanced CVD processes. For example, embodiments according to the present disclosure are useful for non-plasma enhanced CVD processes.

[0106] In one example, the RF power source 122 applies an AC voltage signal having a frequency between 5 MHz and 15 MHz. In one example, the RF power source 122 applies an RF voltage having a magnitude between 100 V and 500 V. Other frequencies and voltages may be used without departing from the scope of the present disclosure. In addition, in some examples, the plasma may be generated by applying a DC voltage. In some examples, the bottom plate 118 is not present, and the plasma is generated primarily by applying a voltage to the top plate 108.

[0107] The PECVD process follows a specific recipe based on the specifics of the PECVD system, the film to be deposited, and other factors. One factor in the PECVD recipe is the distance D between the top surface of the wafer 104 and the bottom surface of the top plate 108. The distance D corresponds to the magnitude of the gap 110. Therefore, the gap 110 corresponds to the distance D between the top surface of the wafer 104 and the bottom surface of the top plate 108. If the magnitude of the gap 110 is outside the selected range, the film deposition process may not be performed correctly. As will be described in more detail below, insufficient uniformity in the deposited film, including insufficient uniformity with respect to the film thickness, the composition of the film, or other characteristics of the film, are all possible results of the gap 110 being outside the selected range according to the specific PECVD process recipe. Therefore, the film deposition process is sensitive to the magnitude of the gap 110.

[0108] For some PECVD processes or systems, the required gap 110 may be between 5 mm and 20 mm. In other cases, the required gap 110 may be significantly smaller. For example, in one embodiment, the PECVD thin film deposition system 100 requires the gap 110 to be between 0.2 mm and 0.7 mm based on the specific process and other factors. The actual tolerance range may be less than 0.1 mm. For example, the PECVD system may require the gap 110 to be 0.5 mm, with a tolerance range of ±0.1 mm beyond this value. In other words, the recipe may require the gap 110 to be between 0.4 and 0.6 mm.

[0109] When a PECVD process is performed on the wafer 104, a film with non-uniform properties may be formed on the wafer 104. This non-uniformity may be the result of the gap 110 not being within the PECVD process specification. In the past, when such non-uniformity was detected, the thin film deposition system 100 had to be shut down so that the position of the bottom surface of the top plate 108, the top surface of the wafer 104, or both could be adjusted to return the distance D to the process specification. After such adjustments and restarts, the uniformity of the deposited film will be monitored and checked to see if it meets the requirements. If the adjustments do not result in the deposition of a film within the required uniformity specification range, the system needs to be shut down again and further adjustments made. Shutting down the thin film deposition system 100 results in an undesirable reduction in productivity, thereby increasing the cost of each wafer. The thin film deposition process may be adversely affected by the gap 110 being outside the desired range in a manner other than the uniformity of the deposited film. For example, the deposition rate, that is, how fast the film is formed during the thin film deposition process, can be affected by the magnitude of the gap 110. If the gap 110 is too wide, the deposition rate may be slowed. If the deposition rate is slowed, the film may not be deposited correctly. For example, if the silicon dioxide layer is intended to have a final thickness of 20 nm, the duration of the film deposition process may need to be carefully timed based on the expected deposition rate. If the deposition rate is slower than expected, the film may not be as thick as expected. This may result in the integrated circuit produced from the wafer 104 not functioning properly, or failing completely. In addition, if the gap 110 is not within a selected range, the film may not have the intended composition.

[0110] During the thin film deposition process, there are various factors that can cause undesirable fluctuations in the gap 110. For example, as previously described, the heater 116 heats the wafer 104 during the thin film deposition process. Due to the heat transfer between the wafer 104 and the top plate 108, the heating of the wafer 104 may also cause some heating of the top plate 108. When the top plate 108 is heated, the top plate 108 will experience thermal expansion. The higher the temperature, the greater the thermal expansion. When the gap 110 is predetermined to be slightly smaller, even a small expansion of the top plate 108 may cause the gap 110 to shrink beyond the predetermined range. In one example, the heater 116 heats the wafer 104 to approximately 450°C. Depending on the deposition conditions, the temperature of the top plate 108 may be correspondingly increased to an unknown degree, resulting in an unknown amount of thermal expansion and the resulting unknown change in the size of the gap 110.

[0111] Another factor that may lead to different gap sizes is the different thicknesses of the wafers 104. For example, some wafers have a thickness of about 770 μm. Other wafers have a thickness of about 720 μm. When the size of the gap 110 is already very small, this 50 μm difference in wafer thickness may result in very different film deposition rates. Therefore, if the thickness of the wafer 104 is unknown, not calculated, or miscalculated, the gap 110 may have a magnitude outside the desired range.

[0112] The thin film deposition system 100 includes a gap sensor 113 in the form of a through-beam sensor. The through-beam sensor includes a radiation emitter 113a and a radiation sensor 113b. The radiation emitter 113a emits radiation 136a through the gap 110 between the wafer 104 and the top plate 108. The radiation sensor 113b is positioned on the side of the wafer 104 and the top plate 108 opposite the radiation emitter 113a. The radiation sensor 113b is aligned to receive the radiation 136b after the radiation 136a has passed through the gap 110. The amount of radiation 136b received by the radiation sensor 113b indicates the size of the gap 110.

[0113] In one example, radiation emitter 113a emits radiation beam 136a. Figure 2 10 is not illustrated to scale, but the diameter of the beam is greater than the predetermined magnitude of the gap 110. Therefore, a portion of the radiation beam 136a is blocked by the sides of the top plate 108 and the sides of the wafer 104, the bottom plate 118, and the heater 116. If the gap 110 is wider, less of the beam is blocked, and accordingly a larger portion of the radiation beam 136a is received by the radiation sensor 113b. If the gap 110 is narrower, more of the beam is blocked, and accordingly a smaller portion of the radiation amount 136b of the radiation beam 136a is received by the radiation sensor 113b. Therefore, the amount of radiation 136b received by the radiation sensor 113b is indicative of the size of the gap 110.

[0114] In one embodiment, radiation sensor 113b generates a voltage based on the amount of radiation 136b received by radiation sensor 113b. In one example, radiation sensor 113b generates a voltage between 3.4V and 3.8V depending on the size of gap 110. The higher the voltage, the larger the gap 110.

[0115] The radiation sensor 113b outputs a sensor signal to the control system 114. The sensor signal may be an analog signal or a digital signal. The sensor signal may include a voltage generated by the radiation sensor 113b from the received radiation 136b. The sensor signal may include an analog or digital representation of a voltage generated by the radiation sensor 113b from the received radiation 136b. The sensor signal may indicate an absolute magnitude of the gap 110 or a relative magnitude of the gap 110.

[0116] In one embodiment, the radiation emitter 113a emits an infrared laser beam. The laser beam may have a diameter between 0.75 cm and 1.5 cm. The diameter of the laser beam may be selected so that under normal operating conditions, the gap 110 does not exceed the diameter of the laser beam. The radiation emitter 113a may emit a laser beam in the ultraviolet or visible spectrum.

[0117] A specific example of the gap sensor 113 is Figure 2. However, many other types of gap sensors 113 may be implemented without departing from the scope of the present disclosure. For example, the gap sensor 113 may include an array of image capture devices that capture images of the gap 110 and output images or other signals indicating the size of the gap 110. Many other types of gap sensors may be used without departing from the scope of the present disclosure.

[0118] In addition to receiving sensor signals from the gap sensor 113, the control system 114 also receives sensor signals from one or more of the counter 101, the process material sensor 103, and the thin film sensor 112. The control system 114 analyzes the sensor signals from the gap sensor 113, analyzes the sensor signals from one or more of the counter 101, the process material sensor 103, and the thin film sensor 112, and determines whether the gap 110 should be adjusted. If the control system 114 determines that the gap 110 should be adjusted based on the sensor signals, the control system 114 can control a motor coupled to the top plate 108, or coupled to a mechanism that suspends the top plate 108. The control system 114 can cause the motor to raise or lower the top plate 108 relative to the wafer 104 to adjust the gap 110. Raising the top plate 108 relative to the wafer 104 increases the gap 110. Lowering the top plate 108 relative to the wafer 104 decreases the gap 110. Alternatively, the control system 114 may control a motor coupled to a pedestal supporting the heater 116, the bottom plate 118, and the wafer 104, and cause the motor to raise or lower the pedestal relative to the wafer 104 to adjust the gap 110. Raising the pedestal relative to the top plate 108 will reduce the size of the gap 110. Lowering the pedestal relative to the top plate 108 will increase the size of the gap 110. In other embodiments, the control system 114 may control a motor coupled to the top plate 108, a mechanism coupled to the top plate 108 or the pedestal, and cause the motor to move the top plate 108, a mechanism coupled to the top plate 108 or the pedestal in a lateral direction as described above.

[0119] In one embodiment, the control system 114 stores reference data. The reference data may include sensor signals known to correspond to a particular size of the gap 110, the lifetime of the component(s) of the thin film deposition system 100, the properties(s) of the thin film deposited by the thin film deposition system 100, or the properties(s) of the process material flowing into the thin film deposition system 100. When the control system 114 receives sensor signals from the gap sensor 113 and one or more of the counter 101, the process material sensor 103, or the thin film sensor 112 described above, the control system 114 compares the sensor signals to the reference data. Based on this comparison of the reference data to the individual sensor signals, the control system 114 causes the motor(s) to adjust the relative position of the top plate 108 relative to the position of the wafer 104, for example, to adjust the size of the gap 110. If the comparison of the sensor signal to the reference data indicates that the sensor signal corresponds to a gap 110 that is too large, the control system 114 causes the motor(s) to make the gap 110 smaller, for example, by lowering the top plate 108 or raising the wafer 104. If the comparison of the sensor signal with the reference data indicates that the sensor signal corresponds to a gap 110 that is too small, the control system 114 causes the motor to enlarge the gap 110 , for example, raising the top plate 108 or lowering the wafer 104 and increasing the gap 110 .

[0120] In one example, the deposition recipe requires a gap 110 of approximately 0.35 mm. Reference data stored by the control system 114 indicates that a gap 110 of 0.35 mm corresponds to a sensor signal having or indicating a voltage of approximately 3.6 V. If the gap 110 is too small, the sensor signal may have or indicate a voltage of approximately 3.5 V. The control system 114 adjusts the gap 110 by raising the top plate 108 during the thin film deposition process until the sensor signal has or indicates a voltage of approximately 3.6 V. If the gap 110 is too large, the sensor signal may have or indicate a voltage of approximately 3.7 V. The control system 114 lowers the top plate 108 until the sensor signal has or indicates a voltage of approximately 3.6 V. The above voltages and distances are given as examples only. Other distances, voltages, and types of sensor signals may be utilized without departing from the scope of the present disclosure.

[0121] The thin film deposition system 100 can adjust the gap 110 in situ during the thin film deposition process. For example, during the thin film deposition process, the gap sensor 113 can continuously generate a sensor signal indicating the gap 110. One or more of the counter 101, the process material sensor 103, and the thin film sensor 112 will continuously generate a sensor signal indicating the life of the thin film deposition system components, the characteristics of the process material flowing into the thin film deposition system, or the characteristics of the thin film deposited by the thin film deposition system. The sensor signal may initially indicate that the gap 110 has a desired value. However, as the deposition process continues, comparison of the sensor signals from other sensors may indicate that the gap 110 no longer has the desired value. The gap 110 that no longer has the desired value may be the result of multiple factors. For example, the top plate 108 may be heated and expand due to thermal expansion, thereby reducing the gap 110. The sensor signal from the gap sensor 113 will indicate this change in the gap 110 during the thin film deposition process. The control system 114 can adjust the gap 110 by moving the top plate 108 or the wafer 104. Therefore, not only can the thin film deposition system 100 adjust the gap 110, but the thin film deposition system 100 can also adjust the gap 110 in real time during the thin film deposition process in response to changing conditions during the thin film deposition process.

[0122] Figure 3 Graphs 300 and 302 are included that indicate film deposition rates for two types of wafers according to one embodiment. Graph 300 illustrates that, without Figure 1 and Figure 2 In the case of the in-situ gap adjustment, the deposition rate of the wafer with a thickness of 720 μm is different from the deposition rate of the wafer with a thickness of 770 μm. For wafers with different outer diameters, the deposition rate of the 770 μm wafer is different from that of the 720 μm thick wafer. This is because the gaps 110 of these wafers are different due to their different thicknesses. The different deposition rates cause the 720 μm thick wafer to have a film thickness profile in the radial direction, which is different from, that is, inconsistent with, the thickness profile of the film on the 770 μm thick wafer in the radial direction.

[0123] Diagram 302 illustrates a situation where Figure 1 and Figure 2 As described above, in-situ gap control is performed. In this case, a wafer with a thickness of 770 μm has the same deposition rate as a wafer with a thickness of 720 μm. This is because the gap sensor 113, the control system 114, and one or more of the other sensors 101, 103, and 112 cooperate to monitor the film process and adjust the gap 110 in real time, such as Figure 1 and Figure 2 Therefore, the two types of wafers have the same gap 110 and the same deposition rate, which results in the two types of wafers having a consistent thickness profile in the radial direction.

[0124] Figure 4A According to an embodiment Figure 1 and Figure 2 A block diagram of a control system 114 is shown. According to one embodiment, Figure 4A The control system 114 is used to control the operation of the thin film deposition system 100 to deposit a thin film on the wafer 104 and adjust the gap 110. In another embodiment, the control system 114 controls the flow rate of the process material through the top plate 108. The control system 114 uses machine learning to learn to appropriately adjust the gap 110 in response to sensor signals received from the gap sensor 113 and one or more of the counter 101, the process material sensor 103, and the film sensor 112. The control system 114 can adjust the gap 110 in situ during the thin film deposition process. The control system 114 can also adjust the flow rate of the process material through the top plate 108 during the thin film deposition process.

[0125] In one embodiment, the control system 114 includes an analysis model 402 and a training module 404. The training module 404 uses a machine learning program to train the analysis model 402. The machine learning program trains the analysis model 402 to adjust the gap 110 to produce a film with selected properties. Although the training module 404 is shown as being separate from the analysis model 402, in practice, the training module 404 can be part of the analysis model 402.

[0126] The control system 114 includes, or stores, training set data 406. The training set data 406 includes historical film data 408 and historical sensor data 410. The historical film data 408 includes data related to a thin film produced by a thin film deposition process. The historical sensor data 410 includes data related to sensor signals during the thin film deposition process that produced the thin film. As will be described in more detail below, the training module 404 utilizes the historical film data 408 and the historical sensor data 410 to train the analysis model 402 using a machine learning process.

[0127] In one embodiment, the historical film data 408 includes data related to the thickness and composition of historical deposited films. For example, during the operation of a semiconductor manufacturing facility, thousands or millions of semiconductor wafers may be processed over a period of months or years. Each of the semiconductor wafers may include a thin film deposited by a thin film deposition process. After each thin film deposition process, the thickness of the film is measured as part of a quality control process. The historical film data 408 includes the thickness of each of the thin films deposited by the thin film deposition process. The historical film data 408 may include the thickness of each of the thin films deposited by the thin film deposition process at different regions of the wafer. For example, the film thickness in the central region of zone 1 of the wafer, the annular region of zone 2 outside zone 1, the annular region of zone 3 outside zone 2, and the annular region of zone 4 extending to the periphery of the wafer outside zone 3. Therefore, the historical film data 408 may include the thickness data of a large number of thin films deposited at different locations on the wafer by the thin film deposition process.

[0128] In one embodiment, the historical sensor data 410 includes various sensor signals or sensor data generated by the gap sensor 113 and one or more of the counter 101, the process material sensor 103, and the thin film sensor 112 during the thin film deposition process that deposits the thin film associated with the historical thin film data 408. Thus, for each thin film having data in the historical thin film data 408, the historical sensor data 410 may include sensor signals generated by the gap sensor 113 and one or more of the counter 101, the process material sensor 103, and the thin film sensor 112 during the thin film deposition. For example, the historical sensor data 410 may be a voltage generated by the receiver 112b of the thin film sensor 112.

[0129] In one embodiment, the training set data 406 links the historical film data 408 with the historical sensor data 410. In other words, the film properties associated with the film in the historical film data 408, such as thickness, material composition, crystal structure, hardness, thermal stability, surface roughness, bandgap, or residual stress, are linked to the sensor signal data associated with the deposition process. As will be described in more detail below, the labeled training set data can be used during a machine learning process to train the analysis model 402 to predict semiconductor sensor signals that will result in the correct formation of a film.

[0130] In one embodiment, the control system 114 includes processing resources 412, memory resources 414, and communication resources 416. The processing resources 412 may include one or more controllers or processors. The processing resources 412 are used to execute software instructions, process data, make thin film deposition control decisions, perform signal processing, read data from memory, write data to memory, and perform other processing operations. The processing resources 412 may include physical processing resources located at a location or facility of the thin film deposition system 100. The processing resources 412 may include virtual processing resources at a location remote from the thin film deposition system 100 or at a facility where the thin film deposition system 100 is located. The processing resources 412 may include cloud-based processing resources, which include processors and servers accessed through one or more cloud computing platforms.

[0131] In one embodiment, the memory resource 414 may include one or more computer-readable memories. The memory resource 414 is used to store software instructions associated with the functions of the control system and its components (including but not limited to the analysis model 402). The memory resource 414 may store data associated with the functions of the control system 114 and its components. The data may include the training set data 406, the current sensor signal data, and any other data associated with the operation of the control system 114 or any of its components. The memory resource 414 may include physical memory resources located at a site or facility of the thin film deposition system 100. The memory resource may include virtual memory resources located at a site or facility remote from the thin film deposition system 100. The memory resource 414 may include cloud-based memory resources accessed through one or more cloud computing platforms.

[0132] In one embodiment, the communication resources 416 may include resources that enable the control system 114 to communicate with devices associated with the thin film deposition system 100. For example, the communication resources 416 may include wired and wireless communication resources that enable the control system 114 to receive sensor signals and control the RF power source 122, the gas source 124, the heater 116, and the motor that raises, lowers, or laterally moves the top plate 108 or the motor that raises, lowers, or laterally moves the base.

[0133] Figure 4B is a diagram illustrating a method according to an embodiment Figure 4A The block diagram of the operation and training state of the analysis model 402 is shown in FIG. Figure 1 and Figure 2As described above, the analytical model 402 can be used to select adjustments to the gap 110 between the wafer 104 and the top plate 108. As previously described, the training set data 406 includes data related to a plurality of previously executed thin film deposition processes. Each previously executed thin film deposition process deposited a thin film having a specific thickness and associated with a specific sensor signal based on the gap 110. The sensor signal for each previously executed thin film deposition process is formatted as an individual sensor signal vector 452. The sensor signal vector 452 includes a plurality of data fields 454. Each data field 454 corresponds to a sensor signal at a specific time during the deposition process.

[0134] The analysis model 402 includes a plurality of neural layers 456a-e. Each neural layer includes a plurality of nodes 458. Each node 458 may also be referred to as a neuron. Each node 458 from the first neural layer 456a receives a data value of each data field 454 from the sensor signal vector 452. Figure 4B In the example of FIG. 4 , each node 458 from the first neural layer 456a receives nine data values ​​because the sensor signal vector 452 has nine data fields 454. Each node 458 includes Figure 4B 4(x). Each node 458 of the first neural layer 456a generates a scalar value by applying the internal mathematical function F(x) to the data value of the data field 454 from the sensor signal vector 452. Further details about the internal mathematical function F(x) are provided below.

[0135] Each node 458 of the second neural layer 456b receives the scalar value generated by each node 458 of the first neural layer 456a. Figure 4B In the example of , each node of the second neural layer 456b receives four scalar values ​​because there are four nodes 458 in the first neural layer 456a. Each node 458 of the second neural layer 456b generates a scalar value by applying a respective internal mathematical function F(x) to the scalar value from the first neural layer 456a.

[0136] Each node 458 of the third neural layer 456c receives the scalar value generated by each node 458 of the second neural layer 456b. Figure 4B In the example of FIG. 4 , each node of the third neural layer 456 c receives five scalar values ​​because there are five nodes 458 in the second neural layer 456 b. Each node 458 of the third neural layer 456 c generates a scalar value by applying a respective internal mathematical function F(x) to the scalar values ​​from the nodes 458 of the second neural layer 456 b.

[0137] Each node 458 of the neural layer 456d receives a scalar value generated by each node 458 of the previous neural layer (not shown). Each node 458 of the neural layer 456d generates a scalar value by applying a respective internal mathematical function F(x) to the scalar value from the node 458 of the previous neural layer.

[0138] For example, the final neural layer includes only a single node 458. The final neural layer receives scalar values ​​generated by the respective nodes 458 of the previous neural layer 456d. The nodes 458 of the final neural layer 456e generate data values ​​468 by applying a mathematical function F(x) to the scalar values ​​received from the nodes 458 of the neural layer 456d.

[0139] exist Figure 4B In the example of , data value 468 corresponds to a predicted thickness of the film generated by the sensor signal data, which corresponds to the value contained in the sensor signal vector 452. In other embodiments, the final neural layer 456e may generate multiple data values, each corresponding to a specific film characteristic, such as film crystal orientation, film uniformity, or other characteristics of the film described above. The final neural layer 456e will include a separate node 458 for each output data value to be generated. In the case of predicted film thickness, the engineer or user can provide a constraint to specify that the predicted film thickness 468 is within a selected range, such as, in one example, between 0 nm and 50 nm. The analytical model 402 will adjust the internal function F(x) to ensure that the data value 468 corresponding to the predicted film thickness will fall within the specified range.

[0140] During the machine learning procedure, the analysis model 402 compares the predicted thickness in the data value 468 with the actual thickness of the film indicated by the data value 470. As previously described, for each historical sensor signal data set, the training set data 406 includes film characteristic data indicating the film characteristics produced by the historical film deposition process. Therefore, the data field 454 includes the actual thickness of the film produced by the deposition process reflected in the sensor signal vector 452. The analysis model 402 compares the predicted thickness from the data value 468 with the actual thickness from the data value 470. The analysis model 402 generates an error value 472, which indicates the error or difference between the predicted thickness from the data value 468 and the actual thickness from the data value 470. The error value 472 is used to train the analysis model 402.

[0141] The training of the analytical model 402 may be more fully understood by discussing the internal mathematical function F(x). Although in one embodiment, all nodes 458 are labeled with an internal mathematical function F(x), the mathematical function F(x) of each node is unique. In one example, each internal mathematical function has the following form:

[0142] F(x)=x1*w1+x2*w2+…x n *w1+b.

[0143] In the above equation, the values ​​x1 to x n corresponds to the data value received from the node 458 in the previous neural layer, or, in the case of the first neural layer 456a, the respective values ​​x1-x n corresponds to the corresponding data value of data field 454 from sensor signal vector 452. Therefore, n for a given node is equal to the number of nodes in the previous neural layer. n is the scalar weight value associated with the corresponding node from the previous layer. The analysis model 402 selects the weight values ​​w1~w n The constant b is a scalar bias value, which can also be multiplied by a weight value. The value generated by node 458 is based on the weight values ​​w1~w n Therefore, each node 458 has n weight values ​​w1~w n Although not shown above, each function F(x) may also include an activation function. The sum in the above equation is multiplied by the activation function. Examples of activation functions may include a rectified linear unit (ReLU) function, a sigmoid function, a hyperbolic tension function, or other types of activation functions.

[0144] After calculating the error value 472, the analysis model 402 adjusts the weight values ​​w1-w2 of each node 458 of each neural layer 456a-456e. n In the analysis model 402, the weight values ​​w1 to w n Thereafter, the analytical model 402 again provides the sensor signal vector 452 to the input neural layer 456a. Since the weight values ​​of the various nodes 458 of the analytical model 402 are different, the predicted thickness 468 will be different from the thickness in the previous iteration. The analytical model 402 again generates an error value 472 by comparing the actual thickness from the data value 470 with the predicted thickness from the data value 468.

[0145] The analysis model 402 again adjusts the weight values ​​w1-w associated with each node 458. n The analytical model 402 processes the sensor signal vector 452 again and generates a predicted thickness 468 and an associated error value 472. The training process includes adjusting the weight values ​​w1-w n , until the error value 472 is minimized.

[0146] Figure 4BA single sensor signal vector 452 is shown being passed to the analysis model 402. In practice, the training process includes passing a large number of sensor signal vectors 452 through the analysis model 402, generating a predicted thickness 468 for each sensor signal vector 452, and generating an associated error value 472 for each predicted thickness. The training process may also include generating an aggregate error value that indicates the average error of all predicted thicknesses for a batch of sensor signal vectors 452. The analysis model 402 adjusts the weighting values ​​w1-w2 after processing each batch of sensor signal vectors 452. n In one embodiment, the training process continues until the average error over all sensor signal vectors 452 is less than a selected threshold tolerance. When the average error is less than the selected threshold tolerance, the training of the analysis model 402 is complete, and the analysis model 402 is trained to accurately predict the film thickness based on the sensor signal. The analysis model 402 can then be used to select the sensor signal value that will produce the desired film thickness. Thus, the analysis model 402 can receive the current sensor signal, determine the sensor signal value that will produce the desired film thickness, and then adjust the gap 110 to adjust the sensor signal.

[0147] A specific example of the neural network based analytical model 402 is described in detail in Figure 4B However, other types of analysis models based on neural networks, or other types of analysis models other than neural networks, may be used without departing from the scope of the present disclosure. In addition, the neural network may have a different number of neural layers (which have a different number of nodes) without departing from the scope of the present disclosure.

[0148] Figure 5 is a flow chart of a method 500 for performing a thin film deposition process. The method 500 may be directed to Figures 1 to 4B At 502, method 500 includes performing a thin film deposition process on a wafer using a thin film deposition system. An example of a thin film deposition system is Figure 1 At 504, the method 500 includes using a gap sensor to generate a sensor signal indicating a gap between a top plate of the thin film deposition system and a wafer during a thin film deposition process. An example of a gap sensor is Figure 1 An example of a top plate is Figure 1 An example of a wafer is Figure 1 of wafer 104. An example of a gap is Figure 1 At 506, method 500 includes generating a sensor signal indicative of a lifetime of a component of the thin film deposition system, a property of a thin film formed by the thin film deposition system, or a property of a process material provided to the thin film deposition system using a counter, a process material sensor, or a thin film sensor. An example of a counter is Figure 1 and Figure 2 An example of a process material sensor is Figure 2 A process material sensor 103. An example of a thin film sensor is Figure 1 At 508, method 500 includes adjusting the gap by moving a top plate or a pedestal in response to the sensor signal during the thin film deposition process. An example of a top plate is Figure 1 The top plate 108 in.

[0149] Figure 6 is a flow chart of a method 600 for performing a thin film deposition process. The method 600 may be directed to Figures 1 to 5 The system, components, and processes described herein are performed. At 602, method 600 includes performing a plasma enhanced chemical vapor deposition process on a wafer in a thin film deposition chamber. An example of a thin film deposition chamber is Figure 2 A thin film deposition chamber 102. An example of a wafer is Figure 2 At 604, method 600 includes generating a plasma using a top plate positioned above the wafer during a plasma enhanced chemical vapor deposition process. An example of a top plate is Figure 2 At 606, method 600 includes generating a sensor signal indicative of a gap between the wafer and the top plate using a gap sensor during a plasma enhanced chemical vapor deposition process. An example of a gap sensor is Figure 2 An example of a gap is Figure 2 At 607, the method includes generating a sensor signal indicating a lifetime of a component of the thin film deposition system, a property of a thin film deposited by the thin film deposition system, or a property of a process material delivered to the thin film deposition system using a counter, a process material sensor, or a thin film sensor during the plasma enhanced chemical vapor deposition process. An example of a counter is Figure 1 and Figure 2 An example of a process material sensor is Figure 2 A process material sensor 103. An example of a thin film sensor is Figure 1 At 608, the method 600 includes analyzing a sensor signal from the gap sensor and a sensor signal from one or more of a counter, a process material sensor, or a thin film sensor during a plasma enhanced chemical vapor deposition process using a control system. An example of a control system is Figure 2 The control system 114. At 610, the method 600 includes adjusting the gap using the control system based on the sensor signal during the plasma enhanced chemical vapor deposition process.

[0150] In one embodiment, method 600 is implemented using sensor signals indicating film uniformity at various regions of the wafer (from film sensor 112), process gas flow rate into the deposition chamber (from process material sensor 103), and gap size (from gap sensor 113). In another embodiment, method 600 is implemented using sensor signals indicating film uniformity at different regions of the wafer (from film sensor 112), process gas composition (from process material sensor 103), and gap size (from gap sensor 113). In another embodiment, method 600 is implemented using sensor signals indicating film uniformity at different regions of the wafer (from film sensor 112), process gas pressure (from process material sensor 103), and gap size (from gap sensor 113). In another embodiment, method 600 is implemented using sensor signals indicative of film uniformity at different regions of the wafer (from film sensor 112), process gas temperature (from process material sensor 103), and gap size (from gap sensor 113). In another embodiment, a sensor signal indicative of wafer tilt angle from a tilt angle sensor associated with pedestal or wafer support 106 and combined with the sensor signal of the embodiments described above in performing method 600 is used.

[0151] In one embodiment, method 600 is performed using sensor signals indicative of one or more of the following: film thickness, film composition, film crystal structure, film hardness, film thermal stability, film surface roughness, film bandgap, and residual stress in the film at various regions of the wafer (from film sensor 112), process gas flow rate into the deposition chamber (from process material sensor 103), and gap size (from gap sensor 113). In another embodiment, method 600 is performed using sensor signals indicative of one or more of the following: film thickness, film composition, film crystal structure, film hardness, film thermal stability, film surface roughness, film bandgap, or residual stress in the film at different regions of the wafer (from film sensor 112), process gas composition (from process material sensor 103), and gap size (from gap sensor 113). In another embodiment, the method 600 is implemented using sensor signals indicating one or more of the following: film thickness at different regions of the wafer, film composition, film crystal structure, film hardness, film thermal stability, film surface roughness, film bandgap or residual stress in the film (from the film sensor 112), process gas pressure (from the process material sensor 103), and gap size (from the gap sensor 113). In another embodiment, the method 600 is implemented using sensor signals indicating one or more of the following: film thickness at different regions of the wafer, film composition, film crystal structure, film hardness, film thermal stability, film surface roughness, film bandgap or residual stress in the film (from the film sensor 112), process gas temperature (from the process material sensor 103), and gap size (from the gap sensor 113). In another embodiment, a sensor signal indicative of a wafer tilt angle from a tilt angle sensor associated with the pedestal or wafer support 106 and combined with a sensor signal of the embodiments described above in performing method 600 .

[0152] In other embodiments of performing the method 600 , the sensor signal indicative of the gap size (from the gap sensor 113 ) is not used.

[0153] In one embodiment, method 600 is implemented using sensor signals indicative of one or more of the following: thickness of a film on the wafer at one or more regions of the wafer, composition of the film on the wafer, crystal structure of the film on the wafer, hardness of the film on the wafer, thermal stability of the film on the wafer, surface roughness of the film on the wafer, bandgap of the film on the wafer, and residual stress in the film on the wafer (from film sensor 112), and one or more of the following: lifetime of the showerhead (from counter 101 or timer), length of time the process chamber has been in operation since the start of the current run or the last scheduled maintenance (from counter 101 or timer), run-to-run variation in deposited material accumulation on the walls of the process chamber (from film sensor), and flow rate of process gas into the process chamber (from process material sensor 103). The discloser has observed that adjusting the distance D of the gap 110 based on consideration of a combination of the specific sensor signals described in this paragraph is particularly effective in producing a desired uniformity of film on the wafer.

[0154] In one embodiment, a method of depositing a thin film includes: performing a thin film deposition process on a wafer using a thin film deposition system, generating a plurality of sensor signals using a sensor, the sensor signals indicating the lifetime of a component of the thin film deposition system, a property of a thin film deposited by the thin film deposition system, or a property of a process material flowing into the thin film deposition system during the thin film deposition process. The method further includes adjusting a relative position of a top plate of the thin film deposition system relative to a position of a wafer in the thin film deposition system in response to the sensor signals during the thin film deposition process.

[0155] In one embodiment, the sensor is a counter for generating a sensor signal indicative of a number of the plurality of wafers processed by the thin film deposition system.

[0156] In one embodiment, the sensor is used to generate a sensor signal that indicates the thickness of a thin film deposited by a thin film deposition system on a wafer or on a surface of a chamber in which a thin film deposition process is performed on the wafer.

[0157] In one embodiment, the sensor is used to generate a sensor signal indicative of an area of ​​the wafer, from which a sensor signal indicative of a thickness of a thin film deposited on the wafer by a thin film deposition system is generated.

[0158] In one embodiment, the sensor is used to generate a sensor signal indicative of a flow rate of a process material entering a chamber in which a thin film deposition process is performed on a wafer.

[0159] In one embodiment, adjusting includes adjusting a distance between the top plate and the wafer.

[0160] In one embodiment, adjusting includes moving the top plate.

[0161] In one embodiment, adjusting includes moving the wafer.

[0162] In one embodiment, the method further includes: transmitting the sensor signal to a control system; determining, using the control system, an adjustment to be made to the gap between the top plate and the wafer based on the sensor signal; and adjusting the gap by moving the top plate, moving the wafer, or both according to the determined adjustment.

[0163] In one embodiment, the method further comprises: generating a plasma using a top plate during the thin film deposition process.

[0164] In one embodiment, adjusting the relative position of a top plate of a thin film deposition system relative to a position of a wafer includes moving the wafer upward on one side relative to the top plate.

[0165] In one embodiment, a thin film deposition system includes a thin film deposition chamber, a wafer support positioned in the thin film deposition chamber and used to support a wafer in the thin film deposition chamber, and a top plate positioned above the wafer support and used to generate plasma in the thin film deposition chamber during a thin film deposition process. The thin film deposition system includes a sensor for generating a sensor signal indicating a lifetime of a component of the thin film deposition system, a property of a thin film deposited by the thin film deposition system, or a property of a process material flowing into the thin film deposition system during the thin film deposition process. The thin film deposition system also includes a control system for receiving the sensor signal and adjusting a relative position of a top plate of the thin film deposition system relative to a position of a wafer in the thin film deposition system in response to the sensor signal.

[0166] In one embodiment, a control system is used to adjust a relative position of a top plate of a thin film deposition system relative to a position of a wafer during a thin film deposition process.

[0167] In one embodiment, the sensor is used to generate a sensor signal indicative of a flow rate of a process material entering a thin film deposition chamber where a wafer is processed in the thin film deposition system.

[0168] In one embodiment, the sensor is used to generate a sensor signal indicative of a number of the plurality of wafers processed by the thin film deposition system.

[0169] In one embodiment, the sensor is used to generate a sensor signal indicative of a thickness of a thin film deposited by a thin film deposition system on a wafer or a surface of a thin film deposition chamber.

[0170] In one embodiment, the thin film deposition system further includes a sensor configured to generate a sensor signal indicative of an area of ​​the wafer from which a signal indicative of a thickness of a thin film deposited on the wafer by the thin film deposition system is generated.

[0171] In one embodiment, a method for depositing a thin film includes: performing a plasma enhanced chemical vapor deposition process on a wafer in a thin film deposition chamber. The method includes generating plasma using a top plate positioned above the wafer during the plasma enhanced chemical vapor deposition process, and generating sensor signals indicating the lifetime of a component of the thin film deposition system, the characteristics of a thin film deposited by the thin film deposition system, or the characteristics of a process material flowing into the thin film deposition system using a sensor during the plasma enhanced chemical vapor deposition process. The method further includes analyzing the sensor signals using a control system during the plasma enhanced chemical vapor deposition process. During the plasma enhanced chemical vapor deposition process, based on the sensor signals, the control system is used to adjust the position of the top plate of the thin film deposition system relative to the position of the wafer in the thin film deposition system.

[0172] In one embodiment, analyzing the sensor signal includes comparing the sensor signal with a plurality of reference data.

[0173] In one embodiment, the method further includes: using a machine learning program to train an analytical model of a control system; using the analytical model to analyze sensor signals; using the analytical model to determine adjustments to be made to the gap based on the sensor signals; and using the control system to adjust the position of a top plate of the thin film deposition system relative to a wafer in the thin film deposition system based on the analytical model.

[0174] Embodiments of the present disclosure provide a thin film deposition system with improved performance. A thin film deposition system includes a top plate positioned above a wafer. Embodiments of the present disclosure monitor the gap between the top plate and the wafer and automatically adjust the gap in situ during the deposition process. This provides a stable deposition rate and produces a more uniform film on a given wafer and from wafer to wafer. The result is fewer scrapped wafers, less frequent equipment changes, and correctly formed thin films.

[0175] The various embodiments described above can be combined to provide further embodiments. Aspects of the embodiments can be modified as needed to adopt the concepts of various patents, applications and publications to provide further embodiments.

[0176] These and other changes can be made to the embodiments in light of the above detailed description. In general, in the appended claims, the terms used should not be construed to limit the claims to the specific embodiments disclosed in the specification and claims, but should be construed to encompass all possible embodiments and all equivalents to which the claims are entitled. Therefore, the claims are not limited by the disclosure.

Claims

1. A method for depositing a thin film, characterized in that: Include: Performing a thin film deposition process on a wafer using a thin film deposition system; generating a plasma using a top plate on the wafer during the film deposition process; generating a plurality of sensor signals using a plurality of sensors, the plurality of sensor signals of the plurality of sensors indicating a lifetime of a component of the thin film deposition system or a characteristic of a thin film deposited by the thin film deposition system; A through-beam sensor is used during the thin film deposition process to generate a sensor signal indicative of a gap, the gap being a distance between a bottom surface of the top plate and a top surface of the wafer, wherein: The beam sensor includes a radiation emitter and a radiation sensor. The radiation emitter emits a radiation beam through the gap between the wafer and the top plate, wherein a diameter of the radiation beam is greater than a predetermined value of the gap. A portion of the radiation beam is blocked by the sides of the top plate and the sides of the wafer, a bottom plate and a heater, The radiation sensor is positioned on a side of the wafer and the top plate opposite the radiation emitter, the radiation sensor is aligned to receive the radiation beam after the radiation beam has passed through the gap, and an amount of radiation received by the radiation sensor indicates the size of the gap; and In response to one or more of the plurality of sensor signals of the plurality of sensors and the sensor signal of the through-beam sensor during the thin film deposition process, a relative position of the top plate of the thin film deposition system relative to a position of the wafer in the thin film deposition system is adjusted.

2. The method according to claim 1, characterized in that: One of the plurality of sensors is a counter for generating a sensor signal indicative of a number of wafers processed by the thin film deposition system.

3. The method according to claim 1, characterized in that One of the plurality of sensors is used to generate a sensor signal indicative of a thickness of the thin film deposited by the thin film deposition system on the wafer or on a surface in a chamber in which the thin film deposition process is performed on the wafer.

4. The method according to claim 3, characterized in that One of the plurality of sensors is used to generate a sensor signal indicative of an area of ​​the wafer, from which the sensor signal indicative of the thickness of the film deposited by the thin film deposition system on the wafer is generated.

5. The method according to claim 1, characterized in that Further including: Using a machine learning program to train an analytical model of a control system; analyzing one or more of the plurality of sensor signals of the plurality of sensors and the sensor signal of the through-beam sensor using the analytical model; utilizing the analytical model to determine an adjustment to be made to the gap based on one or more of the plurality of sensor signals of the plurality of sensors and the sensor signal of the through-beam sensor; and The control system is used to adjust the position of the top plate of the thin film deposition system relative to the wafer in the thin film deposition system based on the analytical model.

6. The method according to claim 1, characterized in that The adjustment includes adjusting the distance between the top plate and the wafer.

7. The method according to claim 6, characterized in that The adjusting includes moving the top plate.

8. The method according to claim 6, characterized in that The adjusting includes moving the wafer.

9. The method according to claim 8, characterized in that Further including: transmitting one or more of the plurality of sensor signals of the plurality of sensors and the sensor signal of the through-beam sensor to a control system; utilizing the control system to determine an adjustment to be made to the gap based on one or more of the plurality of sensor signals of the plurality of sensors and the sensor signal of the through-beam sensor; and Based on the determined adjustment, the distance is adjusted by moving the top plate, moving the wafer, or both.

10. The method according to claim 1, characterized in that Further including: A control system is used to analyze one or more of the multiple sensor signals of the multiple sensors and the sensor signal of the through-beam sensor during the thin film deposition process, wherein analyzing the sensor signal of the through-beam sensor includes comparing one or more of the multiple sensor signals of the multiple sensors and the sensor signal of the through-beam sensor with multiple reference data.

11. The method according to claim 1, characterized in that: The step of adjusting the relative position of the top plate of the thin film deposition system relative to the position of the wafer comprises: The wafer is moved upward on one side relative to the top plate.

12. A thin film deposition system, characterized in that: Include: a thin film deposition chamber; a wafer support positioned in the thin film deposition chamber and used to support a wafer in the thin film deposition chamber; a top plate positioned above the wafer support and used to generate a plasma in the film deposition chamber during the film deposition process; a through-beam sensor for generating a sensor signal indicating a distance between a bottom surface of the top plate and a top surface of the wafer, wherein the through-beam sensor comprises a radiation emitter and a radiation sensor, the radiation emitter being configured to emit a radiation beam having a diameter greater than the distance between the bottom surface of the top plate and the top surface of the wafer, and the radiation sensor being configured to receive a portion of the radiation beam having passed between the bottom surface of the top plate and the top surface of the wafer; and A control system is used to receive the sensor signal of the through-beam sensor and adjust a relative position of the top plate of the thin film deposition system relative to a position of the wafer in the thin film deposition system in response to the sensor signal of the through-beam sensor.

13. The thin film deposition system according to claim 12, characterized in that: The control system is used to adjust the relative position of the top plate of the thin film deposition system relative to the position of the wafer during the thin film deposition process.

14. The thin film deposition system according to claim 12, characterized in that: Further including: A process material sensor is used to generate a sensor signal indicative of a flow rate of the process material entering the thin film deposition chamber in which the wafer is processed in the thin film deposition system.

15. The thin film deposition system according to claim 12, characterized in that: Further including: A counter is used to generate a sensor signal indicating a number of wafers processed by the thin film deposition system.

16. The thin film deposition system according to claim 12, characterized in that: Further including: A thin film sensor is used to generate a sensor signal indicative of a thickness of the thin film deposited by the thin film deposition system on the wafer or on a surface of the thin film deposition chamber.

17. The thin film deposition system according to claim 16, characterized in that: The thin film sensor is further configured to generate a sensor signal indicative of an area of ​​the wafer, from which the sensor signal indicative of the thickness of the thin film deposited on the wafer by the thin film deposition system is generated.

18. A method for depositing a thin film, characterized in that: Include: Performing a plasma enhanced chemical vapor deposition process on a wafer in a thin film deposition system; During the plasma enhanced chemical vapor deposition process, generating a plasma using a top plate positioned above the wafer; During the PECVD process, a through-beam sensor is used to generate a sensor signal indicative of a gap, the gap being a distance between a bottom surface of the top plate and a top surface of the wafer, wherein: The beam sensor includes a radiation emitter and a radiation sensor. The radiation emitter emits a radiation beam through the gap between the wafer and the top plate, wherein a diameter of the radiation beam is greater than a predetermined value of the gap. A portion of the radiation beam is blocked by the sides of the top plate and the sides of the wafer, a bottom plate and a heater, The radiation sensor is positioned on a side of the wafer and the top plate opposite the radiation emitter, the radiation sensor is aligned to receive the radiation beam after the radiation beam has passed through the gap, and an amount of radiation received by the radiation sensor indicates the size of the gap; Analyzing the sensor signal using a control system during the PECVD process; and During the plasma enhanced chemical vapor deposition process, the control system is used to adjust a position of the top plate of the thin film deposition system relative to a position of the wafer in the thin film deposition system based on the sensor signal.

19. The method according to claim 18, characterized in that Analyzing the sensor signal includes: The sensor signal is compared with a plurality of reference data.

20. The method according to claim 18, characterized in that Further including: Using a machine learning program to train an analytical model of the control system; Analyzing the sensor signal using the analysis model; utilizing the analytical model to determine an adjustment to be made to the distance between the bottom surface of the top plate and the top surface of the wafer based on the sensor signal; and The control system is used to adjust the position of the top plate of the thin film deposition system relative to the wafer in the thin film deposition system based on the analytical model.

Images