An automated ai based film measurement for microleds

Abstract

The present disclosure provides an automated film measurement of microLED substrates system in which an automated film measurement inspection module is used to determine the quality of incoming processing material, and an automated film measurement process module is used to process the processing material, an automated film measurement post-process module is used to inspect the resultant process material for acceptability, and an automated film measurement historical module is used to analyze the post process results of the most recent process, against all historical data to find enhanced process changes in the automated film measurement process module and an automated film measurement integration module to update the automated film measurement process module based upon the results of the automated film measurement historical module.

Description

[0001] AN AUTOMATED Al BASED FILM MEASUREMENT FOR MICROLEDS

[0002] BACKGROUND AND FIELD OF THE INVENTION

[0003] [1] The present disclosure is generally related to an automated film measurement process of a microLED chips system.

[0004] [2] Currently, maintaining the stringent requirements of cleanroom environments for microLED fabrication presents a multifaceted challenge. Continuous monitoring and control systems struggle to keep particulate levels within required limits, risking compromised fabrication processes and reduced yields. Additionally, achieving uniform environmental conditions such as temperature and humidity across large cleanroom facilities proves challenging, impacting material properties and process outcomes. Even with advanced filtration and airflow management systems, ensuring comprehensive coverage of air purity remains elusive, further complicating microLED fabrication. Also, the primary challenge during film measurement lies in achieving high accuracy and resolution within the small dimensions of microLEDs. Existing measurement equipment lacks the necessary resolution and noise reduction capabilities to accurately measure film thickness and optical properties at the microscale, compromising device performance.

[0005] [3] Moreover, striking a balance between measurement speed and precision is essential for high-volume production but remains elusive, often resulting in sacrifices in accuracy to expedite processes. Lastly, complex data analysis and interpretation after film measurements pose significant hurdles in microLED fabrication. The complexity of measurement data requires advanced analysis software and expertise to ensure accurate interpretation and utilization for process adjustments or quality control. Additionally, precise correction mechanisms are needed to address identified issues with film uniformity or thickness post-measurement. Integrating measurement data into the production process for real-time adjustments remains a challenge, hindering advancements in production efficiency and throughput. Thus, there is a need in the prior art for an automated film measurement process of microLED chips system.

[0006] SUMMARY

[0007] [4] An automated film measurement of microLED substrates system, comprising an automated film measurement inspection module, an automated film measurement process module, an automated film measurement post-process module, an automated film measurement historical module and an automated film measurement integration module; wherein, an automated film measurement inspection module is used to determine the quality of incoming processing material and, an automated film measurement process module is used to process the processing material, an automated film measurement post-process module is used to inspect the resultant process material for acceptability, and an automated film measurement historical module is used to analyze the post process results of the most recent process, against all historical data to find enhanced process changes in the automated film measurement process module and an automated film measurement integration module to update the automated film measurement process module based upon the results of the automated film measurement historical module.

[0008] DESCRIPTIONS OF THE DRAWINGS

[0009] [5] FIG. 1 : Illustrates an automated Al-based film measurement for microLEDs, according to an embodiment.

[0010] [6] FIG. 2: Illustrates a Base Module, according to an embodiment.

[0011] [7] FIG. 3 : Illustrates an FM Inspection Module, according to an embodiment.

[0012] [8] FIG. 4: Illustrates an FM Process Module, according to an embodiment.

[0013] [9] FIG. 5: Illustrates an FM Post-Process Module, according to an embodiment.

[0014]

[0010] FIG. 6: Illustrates an FM Historical Module, according to an embodiment.

[0015]

[0011] FIG. 7: Illustrates an FM Integration Module, according to an embodiment.

[0016]

[0012] FIG. 8: Illustrates an FM Inspection Database, according to an embodiment.

[0017]

[0013] FIG. 9: Illustrates an FM Process Database, according to an embodiment.

[0018]

[0014] FIG. 10: Illustrates an FM Post-Process Database, according to an embodiment.

[0019] DETAILED DESCRIPTION

[0020]

[0015] Embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings in which numerals represent like elements throughout the several figures, and in which example embodiments are shown. Embodiments of the claims may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. The examples set forth herein are non-limiting examples and are merely examples among other possible examples.

[0021]

[0016] FIG. 1 illustrates a system for an automated Al-based film measurement for microLEDs. This system comprises a 3rd party film measurement Al network 102, which integrates pre-quality inspection, processing, post-processing, historical analysis, and continuous improvement in a film measurement system. The FM inspection module 110 measures essential parameters and calculates outputs to guide the subsequent film measurement process. The FM process module 112 acts as the core controller, utilizing inputs from the FM inspection module 110 to govern the film measurement process to ensure optimal and accurate speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc. The FM post-process module 114 conducts inspections to assess the execution of parameters and identify potential deviations. The FM historical module 116 analyzes data collected across various stages, allowing for continuous learning and pattern recognition. The FM integration module 118 then employs historical insights to update and optimize the process, enabling adaptive improvements over time.

[0022]

[0017] Further, embodiments may include a communication interface 104, which may be a hardware or software component that enables communication between the 3rd party film measurement Al network 102 and the film measurement pre-processl28, film measurement process 134, film measurement post-process 150. The communication interface 104 may include a set of protocols, rules, and standards that define how information is transmitted and received between the devices. The communication interface 104 may be a physical connector, wireless network, or software application and may include components such as drivers, software libraries, and firmware that may be used to control and manage the communication process. In some embodiments, the communication interface 104 may be compatible with USB, Bluetooth, or WiFi. The communication interface 104 may communicate with a network. Examples of networks may include, but are not limited to, the Internet, a cloud network, a Wireless Fidelity (Wi-Fi) network, a Wireless Local Area Network (WLAN), a Local Area Network (LAN), a telephone line (POTS), Long Term Evolution (LTE), and / or a Metropolitan Area Network (MAN).

[0023]

[0018] Further, embodiments may include a memory 106, which may include suitable logic, circuitry, and / or interfaces that may be configured to store a machine code and / or a computer program with at least one code section executable by a processor. Examples of implementation of the memory 106 may include, but are not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Hard Disk Drive (HDD), and / or a Secure Digital (SD) card.

[0024]

[0019] Further, embodiments may include a base module 108, which initiates the FM inspection module 110, the FM process module 112, the FM post-process module 114, the FM historical module 116, and the FM integration module 118.

[0025]

[0020] Further, embodiments may include an FM inspection module 110, which begins by being initiated by the base module 108. The FM inspection module 110 connects to the film measurement pre-process 130. The FM inspection module 110 collects the data from the film measurement pre- process 130. The FM inspection module 110 performs the inspection algorithm. The FM inspection module 110 sends the determined process data from the inspection algorithm to the FM process module 112. The FM inspection module 110 stores the data in the FM inspection database 120. The FM inspection module 110 returns to the base module 108.

[0026]

[0021] Further, embodiments may include an FM process module 112, which begins by being initiated by the base module 108. The FM process module 112 receives the process data, such as the data on the microLED processed wafer, from the FM inspection module 110. The FM process module 112 connects to the film measurement process 140. The FM process module 112 collects the data from the film measurement process 140. The FM process module 112 performs the control algorithm. The FM process module 112 adjusts the control parameters of the film measurement process 140. The FM process module 112 executes the film measurement process. The FM process module 112 stores the data in the FM process database 122. The FM process module 112 returns to the base module 108.

[0027]

[0022] Further, embodiments may include an FM post-process module 114, which begins by being initiated by the base module 108. The FM post-process module 114 connects to the film measurement post-process 150. The FM post-process module 114 collects the post-process data. The FM post-process module 114 extracts the data from the FM inspection database 120 and FM process database 122. The FM post-process module 114 performs the post-process algorithm. The FM post-process module 114 performs the post-process inspection. The FM post-process module 114 stores the data in the FM post-process database 124. The FM post-process module 114 returns to the base module 108.

[0028]

[0023] Further, embodiments may include an FM historical module 116, which begins by being initiated by the base module 108. The FM historical module 116 connects to the film measurement pre-process 130, the film measurement process 140, and the film measurement post-process 150. The FM historical module 116 aggregates the data from the various types of processes and stores the data in the FM network database 126. The FM historical module 116 performs the historical machine learning algorithm on the historical data stored in the FM network database 126. The FM historical module 116 sends the process adjustments to the FM inspection module 110, the FM process module 112, and the FM post-process module 114. The FM historical module 116 returns to the base module 108.

[0029]

[0024] Further, embodiments may include an FM integration module 118, which begins by being initiated by the base module 108. The FM integration module 118 performs the integration machine learning algorithm. The FM integration module 118 connects to the film measurement pre-process Al module 136, the film measurement process Al module 146, and the film measurement postprocess Al module 156. The FM integration module 118 sends the process adjustments to the film measurement pre-process Al module 136, the film measurement process Al module 146, and the film measurement post-process Al module 156. The FM integration module 118 returns to the base module 108.

[0030]

[0025] Further, embodiments may include an FM inspection database 120, which provides an example of the results of the inspection algorithm performed in the FM inspection module 110. The FM inspection database 120 may contain the experiment ID, the relative humidity outputted by the inspection algorithm, the final insulator thickness controlled by the inspection algorithm, etc. The first example illustrates the results of 10 experiments conducted to measure the insulator thickness on microLED process substrates using a film measurement device in an automated manner. Each row represents a distinct experiment, detailing the experiment ID, the humidity level in relative humidity or RH during the measurement, and the final insulator thickness measured on a control sample from a known specification in a reference area. The humidity levels show a normal variation with current controls ±10% range of a baseline of 50% RH, and the resulting insulator thickness is an example measurement. For example, in experiment 1, with a humidity level of 60 RH, the final insulator thickness measured was 1.05 microns, indicating an increase in thickness correlated with higher humidity levels. Experiment 3, conducted at a lower humidity level of 44 RH, resulted in a thinner insulator layer of 0.97 microns. These variations illustrate the sensitivity of the final insulator thickness measurements to changes in environmental humidity, illustrating the importance of maintaining controlled humidity levels for precise and accurate measurements in the microLED fabrication process.

[0031]

[0026] The second example illustrates the results of 10 experiments aimed at measuring insulator thickness in microLED processes utilizing the inspection algorithm and enhanced controller to minimize variations in relative humidity and the related insulator thickness variations. Each row represents an individual experiment, detailing the experiment ID, the more consistently controlled humidity level through the inspection algorithm, and the final insulator thickness measured on a control sample. For example, experiments 1 and 2, with humidity levels precisely controlled at 50 RH, resulted in a consistent insulator thickness of 1.000 microns, demonstrating the effectiveness of the inspection algorithm in maintaining target environmental conditions. Experiments 3 and 5, where the humidity slightly increased to 52 RH, saw a slight increase in insulator thickness to 1.005 microns. Experiments 6, 7, and 8, with a reduced humidity level of 48 RH, showed a minor decrease in thickness to 0.995 microns. The second example illustrates the inspection algorithm’s ability to reduce variability in environmental conditions, leading to more consistent and predictable insulator thickness measurements in microLED fabrication processes. In some embodiments, the FM inspection database 120 may contain parameters, optimized parameters, parameter adjustments from the inspection algorithm, etc.

[0027] Further, embodiments may include an FM process database 122, which provides an example of the results of the control algorithm performed in the FM process module 112. The FM process database 122 may contain the experiment ID, the relative humidity, the final insulator thickness, the optimized speed of the measurement, etc. The first example illustrates the speed of measurement in seconds, which may be based on the variability in RH from the baseline of 50%. The deviation from the baseline shows measurement time changing, starting from a setpoint speed of 5 seconds under optimal conditions to the actual time it takes to get the measurement through the optical measurement tool. In the context of optical measurement devices, settling refers to the process by which the instrument stabilizes and reaches a consistent state after being exposed to changes in its environment or operating conditions. When a measurement device settles, it has adjusted to any fluctuations in temperature, humidity, or other environmental factors and is ready to provide accurate and reliable measurements. The settling time of an optical instrument is the duration it takes for the device to achieve this stable state. During this time, the instrument may undergo internal automatic adjustments or calibration procedures to compensate for changes in its surroundings. Settling time is crucial for ensuring the accuracy and repeatability of measurements. For example, in optical thickness measurements of insulator films for microLED fabrication, the optical tool may require a settling period to adapt to variations in humidity levels. This settling time allows the instrument to compensate for any changes in the optical properties of the insulator material due to moisture absorption, ensuring that subsequent measurements are performed under consistent conditions. The first example illustrates that as the humidity levels deviate further from the optimal 50% RH, the time required for measurements increases, reflecting the additional complexity in ensuring accurate film thickness readings.

[0032]

[0028] For example, experiment 1, with a humidity level of 60% RH, displays a higher measurement time of 7 seconds, which indicates the potential need for additional calibration or adjustments due to the higher humidity. Measurements closer to the optimal humidity condition, such as experiments 4, 8, and 10, are quicker, at 5.4 seconds, demonstrating the efficiency of maintaining controlled environmental conditions during the film thickness measurement process. Focusing solely on optical film thickness measurements for insulators in the context of microLED fabrication, the justification for the speed of measurements being affected by relative humidity (RH) changes is nuanced. Optical measurement techniques, such as spectroscopic reflectometry or ellipsometry, primarily depend on the interaction of light with the material's surface to gauge thickness. These methods are inherently fast, but the overall process efficiency and speed may be indirectly influenced by RH through hygroscopic effects, condensation on the surface of the insulator, calibration due to environmental factors, requiring data analysis techniques to isolate and remove the effects of the environmental conditions, etc. In the second example, the columns are labeled to indicate the Al adjustments by the control algorithm to humidity measurements and the resulting enhancement in the speed of measurements for insulator thickness in microLED fabrication. In this example, the control algorithm may be a linear regression model or a decision tree algorithm to control humidity levels and optimize measurement speed. In some embodiments, the FM process database 122 may store other parameter data collected or calculated from the FM process module 112, such as humidity, temperature, surface roughness, film uniformity, film composition, etc.

[0033]

[0029] Further, embodiments may include an FM post-process database 124, which provides an example of the results of the post-process algorithm performed in the FM post-process module 114. The example data illustrates the lot number or ID, the initial moisture percentage, temperature, time, final moisture percentage, moisture reduction percentage, etc. The first example illustrates that rows 1 and 2 have speeds of 7.0 and 6.6 seconds, respectively, which exceed the specified threshold of 6.0 seconds. Therefore, these rows would require a rescan or retake to ensure that the insulator thickness measurements meet the required specifications for speed. Rows 3 to 10 have speeds within the acceptable range and would not require a rescan based on the specified threshold. The second example illustrates improved RH control by the post-process algorithm, and the film thickness measurements are more consistent, and only one row, experiment ID 11, would require a re-scan. The post-process algorithm improved humidity control leads to more reliable and precise thickness measurements and reduces the need for rescanning or retaking measurements. Therefore, most of the rows do not require a re-scan, indicating the effectiveness of the post-process algorithm in improving measurement accuracy and ensuring that film thickness remains within specification limits. In the second example, the post-process algorithm controls humidity levels by adjusting environmental parameters such as airflow and temperature. For example, in experiments where the humidity level is higher than optimal, such as experiment 11 with 60% RH, the post-process algorithm may trigger actions to increase airflow or activate dehumidifiers to reduce RH to the desired level, thereby improving measurement speed. In experiments with lower humidity levels, such as experiment 13 with 44% RH, the post-process algorithm may adjust the temperature or introduce moisture into the air to raise RH to the optimal range. These proactive adjustments ensure that the insulator thickness measurements can be performed more rapidly and consistently, which contributes to increased productivity and efficiency in microLED fabrication processes. In some embodiments, the FM post-process database 124 may store other parameter data collected or calculated from the FM post-process module 114, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc.

[0034]

[0030] Further, embodiments may include an FM network database 126, which may contain the historical data from the various processes performed by the film measurement pre-process 130, film measurement process 140, and film measurement post-process 150. The FM network database 126 may contain the data parameters collected during the inspection process, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, and control parameters of the process, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc., and data parameters collected during post-processing, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc.

[0035]

[0031] Further, embodiments may include a cloud 128, which may be a network of remote servers that provide on-demand computing resources and services over the Internet. The cloud 128 may consist of a collection of servers, storage devices, and networking equipment. Users may access the cloud 128 through a variety of devices, such as computers, smartphones, and tablets, using internet connectivity. In some embodiments, the architecture of the cloud 128 may be based on a distributed computing model, with multiple servers working together to provide services to users.

[0036]

[0032] Further, embodiments may include film measurement pre-process 130, which may be the processes, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process. The pre-process may involve collecting a plurality of parameter data, including humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. In some embodiments, the film measurement pre-process 130 may send and receive data from the 3rd party film measurement Al network 102, including parameter adjustments, control adjustments, material or workpiece data, decisions on material inspections, inputs to be used by the film measurement pre-process Al module 136, etc.

[0037]

[0033] Further, embodiments may include a communication interface 132, which may be a hardware or software component that enables the communication between the film measurement pre-process 130 and the 3rd party film measurement Al network 102. In some embodiments, the film measurement pre-process 130 may communicate with the film measurement process 140 and film measurement post-process 150. The communication interface 132 may include a set of protocols, rules, and standards that define how information is transmitted and received between the devices. The communication interface 132 may be a physical connector, wireless network, or software application and may include components such as drivers, software libraries, and firmware that may be used to control and manage the communication process. In some embodiments, the communication interface 132 may be compatible with USB, Bluetooth, or Wi-Fi. The communication interface 132 may communicate with a network. Examples of networks may include, but are not limited to, the Internet, a cloud network, a Wireless Fidelity (Wi-Fi) network, a Wireless Local Area Network (WLAN), a Local Area Network (LAN), a telephone line (POTS), Long Term Evolution (LTE), and / or a Metropolitan Area Network (MAN).

[0038]

[0034] Further, embodiments may include a memory 134 may include suitable logic, circuitry, and / or interfaces that may be configured to store a machine code and / or a computer program with at least one code section executable by a processor. Examples of implementation of the memory 134 may include, but are not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Hard Disk Drive (HDD), and / or a Secure Digital (SD) card.

[0039]

[0035] Further, embodiments may include a film measurement pre-process Al module 136 in which a predictive model may be performed to predict the final product of the film measurement process based upon the data collected from the film measurement pre-process 130, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. Predictive modeling involves the systematic analysis of historical data to identify patterns, correlations, and trends that can be used to build models capable of making accurate predictions about future outcomes. In some embodiments, the film measurement pre-process Al module 136 may receive a predictive model from the FM integration module 118. In some embodiments, the film measurement pre-process Al module 136 may receive and send data to the FM inspection module 110. In some embodiments, the data received from the FM inspection module 110 may be inputted into the predictive model to determine if any parameters of the pre- process should be adjusted, such as humidity, temperature, etc.

[0040]

[0036] Further, embodiments may include a film measurement pre-process database 138, which may include data parameters collected from the film measurement pre-process 130, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. In some embodiments, the data stored in the film measurement pre-process database 138 may be sent to the FM inspection module 110 and / or the FM historical module 116.

[0041]

[0037] Further, embodiments may include film measurement process 140, which may be systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected. The film measurement process 140 may involve collecting parameter data, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc. The film measurement process 140 may include equipment, such as an ellipsometer, spectrophotometer, atomic force microscope, scanning electron microscope, x-ray photoelectron spectroscopy, four-point probe, profilometer, Raman spectrometer, quartz crystal microbalance, reflectometer, etc. In some embodiments, the film measurement process 140 may send and receive data from the 3rd party film measurement Al network 102, including parameter adjustments, control adjustments, inputs to be used by the film measurement process Al module 146, etc.

[0042]

[0038] Further, embodiments may include a communication interface 142, which may be a hardware or software component that enables communication between the film measurement process 140 and the 3rd party film measurement Al network 102. In some embodiments, the film measurement process 140 may communicate with the film measurement pre-process 130 and film measurement post-process 150. The communication interface 142 may include a set of protocols, rules, and standards that define how information is transmitted and received between the devices. The communication interface 142 may be a physical connector, wireless network, or software application and may include components such as drivers, software libraries, and firmware that may be used to control and manage the communication process. In some embodiments, the communication interface 142 may be compatible with USB, Bluetooth, or Wi-Fi. The communication interface 142 may communicate with a network. Examples of networks may include, but are not limited to, the Internet, a cloud network, a Wireless Fidelity (Wi-Fi) network, a Wireless Local Area Network (WLAN), a Local Area Network (LAN), a telephone line (POTS), Long Term Evolution (LTE), and / or a Metropolitan Area Network (MAN).

[0043]

[0039] Further, embodiments may include a memory 144 may include suitable logic, circuitry, and / or interfaces that may be configured to store a machine code and / or a computer program with at least one code section executable by a processor. Examples of implementation of the memory 144 may include, but are not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Hard Disk Drive (HDD), and / or a Secure Digital (SD) card.

[0044]

[0040] Further, embodiments may include a film measurement process Al module 146 in which a predictive model may be performed to predict the final product of the film measurement process based upon the data collected from the film measurement process 140, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc. Predictive modeling involves the systematic analysis of historical data to identify patterns, correlations, and trends that can be used to build models capable of making accurate predictions about future outcomes. In some embodiments, the film measurement process Al module 146 may receive a predictive model from the FM integration module 118. In some embodiments, the film measurement process Al module 146 may receive and send data to the FM process module 112. In some embodiments, the data received from the FM process module 112 may be inputted into the predictive model to determine if any parameters of the film measurement process 140 should be adjusted or rescanned, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc.

[0041] Further, embodiments may include a film measurement process database 148 which may include data parameters collected from the film measurement process 140, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc. In some embodiments, the data stored in the film measurement process database 148 may be sent to the FM process module 112 and / or the FM historical module 116.

[0045]

[0042] Further, embodiments may include film measurement post-process 150, which may be the processes and methods after the completion of the film measurement process 140, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc. The post-process may involve parameters such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. In some embodiments, the film measurement post- process 150 may send and receive data from the 3rd party film measurement Al network 102, including parameter adjustments, control adjustments, inputs to be used by the film measurement post-process Al module 156, etc.

[0046]

[0043] Further, embodiments may include a communication interface 152, which may be a hardware or software component that enables communication between the film measurement postprocess 150 and the 3rd party film measurement Al network 102. In some embodiments, the film measurement post-process 150 may communicate with the film measurement pre-process 130 and film measurement process 140. The communication interface 152 may include a set of protocols, rules, and standards that define how information is transmitted and received between the devices. The communication interface 152 may be a physical connector, wireless network, or software application and may include components such as drivers, software libraries, and firmware that may be used to control and manage the communication process. In some embodiments, the communication interface 152 may be compatible with USB, Bluetooth, or Wi-Fi. The communication interface 152 may communicate with a network. Examples of networks may include, but are not limited to, the Internet, a cloud network, a Wireless Fidelity (Wi-Fi) network, a Wireless Local Area Network (WLAN), a Local Area Network (LAN), a telephone line (POTS), Long Term Evolution (LTE), and / or a Metropolitan Area Network (MAN).

[0044] Further, embodiments may include a memory 154 may include suitable logic, circuitry, and / or interfaces that may be configured to store a machine code and / or a computer program with at least one code section executable by a processor. Examples of implementation of the memory 154 may include, but are not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Hard Disk Drive (HDD), and / or a Secure Digital (SD) card.

[0047]

[0045] Further, embodiments may include a film measurement post-process Al module 156 in which a predictive model may be performed to optimize the future final products based upon the data collected from the film measurement post-process 150, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. Predictive modeling involves the systematic analysis of historical data to identify patterns, correlations, and trends that can be used to build models capable of making accurate predictions about future outcomes. In some embodiments, the film measurement postprocess Al module 156 may receive a predictive model from the FM integration module 118. In some embodiments, the film measurement post-process Al module 156 may receive and send data to the FM post-process module 114. In some embodiments, the data received from the FM postprocess module 114 may be inputted into the predictive model to determine if any parameters of the film measurement post-process 150 should be adjusted, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc.

[0048]

[0046] Further, embodiments may include a film measurement post-process database 158, which may include data parameters collected from the film measurement post-process 150, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. In some embodiments, the data stored in the film measurement post-process database 158 may be sent to the FM post-process module 114 and / or the FM historical module 116.

[0049]

[0047] FIG. 2 illustrates the base module 108. The process begins with the base module 108 initiating, at step 200, the FM inspection module 110. For example, the FM inspection module 110 begins by being initiated by the base module 108. The FM inspection module 110 connects to the film measurement pre-process 130. The FM inspection module 110 connects with the film measurement pre-process 130, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process. In some embodiments, the FM inspection module 110 may transmit and receive data from the film measurement pre-process 130. In some embodiments, the film measurement pre-process 130 may measure humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. In some embodiments, the FM inspection module 110 may control or send inputs to control the film measurement pre-process 130. The FM inspection module 110 collects the data from the film measurement pre-process 130. The FM inspection module 110 collects the pre-processing data, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. The FM inspection module 110 performs the inspection algorithm. For example, the inspection algorithm may minimize variations in relative humidity and the related insulator thickness variations. For example, the effect of humidity on the thickness of insulator layers may vary due to the different physical and chemical properties of the materials. Polymers may absorb moisture from the environment, which may affect their physical properties, including dimensional changes or swelling, which may lead to variations in the thickness of layers or coatings made from these materials.

[0050]

[0048] The interaction between moisture and the polymer matrix may lead to hydrolysis reactions or plasticization, which may modify the material's structure and its physical dimensions, including thickness. Insulators generally do not absorb water in the same way polymers do, but the presence of water molecules may still affect the growth or formation of insulator layers, primarily through mechanisms such as the facilitation of oxidation reactions or affecting the surface energy and dynamics of the insulator layer formation. In some embodiments, the impact of humidity on the thickness of insulator versus polymer layers may depend on the material's composition, the environmental conditions, and the mechanisms through which moisture interacts with the material. Higher relative humidity, or RH, may lead to thicker insulator layers on surfaces due to enhanced oxidation reactions or other surface chemistry effects. The impact on polymer insulators may be related to the physical absorption of moisture and its interaction with the polymer matrix. The inspection algorithm may control humidity levels and result in more consistent insulator thickness measurements by utilizing a precision environmental control system. The inspection algorithm may perform a combination of predictive modeling and real-time adjustments to optimize the environmental conditions within the film measurement tool. The inspection algorithm may be a machine learning model, such as a random forest or gradient boosting machine, that may predict the impact of various environmental factors, including humidity, on the insulator layer thickness with high accuracy. The inspection algorithm may be trained on historical data, including a wide range of environmental conditions and their corresponding effects on insulator thickness measurements. The data set may include variables such as temperature, humidity levels, substrate material properties, and previous insulator thickness measurements.

[0051]

[0049] The trained inspection algorithm may predict the optimal humidity levels required to achieve a target insulator thickness, taking into account the current environmental conditions and material properties. The inspection algorithm may incorporate a feedback loop using sensors to continuously monitor the actual humidity levels and the insulator thickness measurements, which may be fed back into the inspection algorithm to adjust predictions as needed.

[0052]

[0050] In some embodiments, actuators within the environmental control system may adjust the conditions inside the measurement tool based on the inspection algorithm’s recommendations, which may involve controlling dehumidifiers, humidifiers, or air conditioning systems to maintain the desired humidity level.

[0053]

[0051] In some embodiments, the control system may be adaptive, capable of adjusting its parameters in real-time based on the inspection algorithm’s output and the feedback received from the environment to ensure the system can respond to any unforeseen changes or drifts in environmental conditions, maintaining the optimal humidity level for consistent insulator thickness measurements.

[0054]

[0052] In some embodiments, the inspection algorithm may incorporate optimization techniques, such as genetic algorithms or simulated annealing, to find the best control strategy that minimizes variability in insulator thickness while considering operational constraints and efficiency.

[0055]

[0053] In some embodiments, the inspection algorithm may utilize neural networks, such as deep learning models, which may learn from historical data comprising various environmental conditions and their impact on insulator thickness to predict the exact adjustments needed in the environmental control systems to achieve the desired humidity levels that result in optimal insulator thickness, accounting for complex interactions between different factors.

[0054] In some embodiments, the inspection algorithm may utilize support vector machines for regression, which may be used to predict the optimal environmental conditions necessary for achieving the target insulator thickness on substrates in the microLED process by identifying adjustments needed in the environmental control system to maintain desired conditions by being trained on historical data, including humidity levels and corresponding insulator thickness measurements. The FM inspection module 110 sends the determined process data from the inspection algorithm to the FM process module 112. For example, the FM inspection module 110 may send the relative humidity and final insulator thickness to the FM process module 112 to optimize the film measurement process.

[0056]

[0055] In some embodiments, the FM inspection module 110 may send temperature, surface roughness, film uniformity, film composition, etc. The FM inspection module 110 stores the data in the FM inspection database 120. The FM inspection module 110 stores the data outputted from the inspection algorithm in the FM inspection database 120, such as the experiment ID, the relative humidity outputted by the inspection algorithm, the final insulator thickness controlled by the inspection algorithm, etc.

[0057]

[0056] In some embodiments, the FM inspection database 120 may include other parameter data collected or calculated during the FM inspection module 110 process, such as temperature, surface roughness, film uniformity, film composition, etc. The FM inspection module 110 returns to the base module 108. The base module 108 initiates, at step 202, the FM process module 112. For example, the FM process module 112 begins by being initiated by the base module 108. The FM process module 112 receives the pre-process data, such as the data on the microLED processed wafer, from the FM inspection module 110. The FM process module 112 receives the outputted data from the inspection algorithm performed in the FM inspection module 110, such as the relative humidity, allowing the film measurement process 140 to determine the optimal control parameters to optimize the film measurement process.

[0058]

[0057] In some embodiments, the FM process module 112 may receive temperature, surface roughness, film uniformity, film composition, etc. The FM process module 112 connects to the film measurement process 140. The FM process module 112 connects with the film measurement process 140, such as systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected. In some embodiments, the FM process module 112 may transmit and receive data from the film measurement process 140. The FM process module 112 collects the data from the film measurement process 140. The FM process module 112 collects the processing data, such as humidity, temperature, surface roughness, film uniformity, film composition, etc. The FM process module 112 performs the control algorithm. The control algorithm may optimize the measurement speed by controlling the humidity levels. For example, changes in humidity may affect the measurement speed as these methods are typically rapid and based on the interaction of light with the film surface. Humidity may influence the accuracy and the effective speed of the process indirectly by necessitating additional steps or corrections. Humidity may alter the refractive index of hygroscopic insulator films by changing their composition through moisture absorption. Since optical thickness measurement methods, such as ellipsometry or reflectometry, rely on changes in light polarization or reflectance that depend on the film's refractive index, variations in humidity can lead to inaccuracies in the measured film thickness if not accounted for. Some materials may physically swell or contract, with changes in humidity, affecting the actual film thickness, and identifying and correcting for these physical changes to ensure accuracy may add time to the overall measurement procedure.

[0059]

[0058] In environments with high humidity, condensation may form on the surfaces to be measured, scattering or absorbing the incident light and interfering with optical measurements, which may require the substrates to be dried or conditioned before measurement, indirectly slowing down the process. The control algorithm may be a linear regression model or a decision tree algorithm utilized to control humidity levels and optimize measurement speed. For example, the control algorithm may analyze historical humidity data and corresponding measurement times to identify linear relationships between RH levels and measurement speed. By extrapolating this relationship, the control algorithm may predict the optimal RH level for minimizing measurement time in real-time.

[0060]

[0059] In some embodiments, the control algorithm may analyze the data to identify threshold values of humidity at which measurement speed tends to decrease, and based on these thresholds, the control algorithm may make decisions on when to adjust environmental conditions to maintain RH within an optimal range for efficient measurements, the control algorithm may control humidity levels by adjusting environmental parameters such as airflow and temperature. For instance, in experiments where the humidity level is higher than optimal, the control algorithm may trigger actions to increase airflow or activate dehumidifiers to reduce RH to the desired level to improve measurement speed.

[0061]

[0060] In experiments with lower humidity levels, the control algorithm may adjust the temperature or introduce moisture into the air to raise RH to the optimal range. These adjustments from the control algorithm ensure that the insulator thickness measurements may be performed more rapidly and consistently, contributing to increased productivity and efficiency in microLED fabrication processes.

[0062]

[0061] In some embodiments, the control algorithm may utilize support vector machines to optimize measurement speed, in which the control algorithm may analyze historical data of humidity levels and corresponding measurement times. The control algorithm may learn to classify or regress on the data to determine the optimal humidity range for faster measurements. In some embodiments, the control algorithm may utilize random forests to analyze historical humidity data along with other environmental variables to predict the optimal conditions for faster measurement speed. By aggregating the predictions of multiple decision trees, the control algorithm may provide more accurate and reliable recommendations for adjusting environmental parameters to maintain optimal humidity levels. The FM process module 112 adjusts the control parameters of the film measurement process 140. In some embodiments, the FM process module 112 may optimize the speed of the measurements by optimizing the humidity levels. In some embodiments, the FM process module may optimize the temperature. The FM process module 112 executes the film measurement process. The film measurement process may include collecting baseline measurements, data acquisition, data analysis, parameter calculation, interpretation and validation, etc. For example, some measurement techniques, such as ellipsometry or reflectometry, may require a baseline measurement of the bare substrate without the thin film and this is used as a reference for comparing measurements with the thin film present. The measurement process begins, and data is acquired according to the selected technique, which may involve scanning the sample surface, illuminating it with light of specific wavelengths, applying electrical probes, etc., depending on the measurement method. The acquired data is analyzed to extract relevant information about the thin film properties, which may involve mathematical modeling, fitting experimental data to theoretical models, performing calculations based on measurement results, etc. Using the analyzed data, various parameters of interest are calculated. For example, in ellipsometry, parameters such as film thickness, refractive index, and extinction coefficient are determined from the measured ellipsometry angles. The calculated parameters are interpreted in the context of the thin film's intended application and manufacturing process. Any deviations from expected values are investigated to ensure measurement accuracy and reliability. The FM process module 112 stores the data in the FM process database 122. The FM process module 112 stores the data in the FM process database 122, such as the data collected from the film measurement process 140, the output of the inspection algorithm, and the output of the control algorithm. The FM process database 122 may contain the experiment ID, the relative humidity, the final insulator thickness, the optimized speed of the measurement, etc. The FM process module 112 returns to the base module 108. The base module 108 initiates, at step 204, the FM post-process module 114. For example, the FM post-process module 114 begins by being initiated by the base module 108. The FM post-process module 114 connects to the film measurement post-process 150. The FM postprocess module 114 may connect to the film measurement post-process 150, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc. The FM post-process module 114 collects the post-process data. The FM post-process module 114 may collect the parameter data, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. The FM post-process module 114 extracts the data from the FM inspection database 120 and FM process database 122. The FM post-process module 114 extracts the data, such as material data, parameter data, etc., of the pre-process and film measurement process. The FM post-process module 114 performs the post-process algorithm. The post-process algorithm may optimize the environmental conditions to minimize the need for rescanning or retaking measurements. For example, in film measurement systems, there may be a mechanism for automated rescanning or retaking measurements if the initial measurement falls outside the specified tolerances or specifications to ensure that measurements are accurate and reliable. The insulator may be rescanned by determining if a measurement is outside a specification threshold. Once the threshold has been exceeded, the system automatically triggers the need for a rescan, then the system performs another measurement, and the new measurement is compared against the specific tolerances or specifications. Once a valid measurement is obtained, the results are documented and reported. The measured data may change through sample variation, environmental changes, instrument drift, surface contamination, etc. The post-process algorithm may analyze historical humidity data and corresponding measurement times to identify linear relationships between RH levels and measurement speed and by extrapolating this relationship, the post-process algorithm may predict the optimal RH level for minimizing measurement time in real-time. In some embodiments, the post-process algorithm may utilize decision tree algorithms, which may analyze the data to identify threshold values of humidity at which measurement speed tends to decrease, and based on these thresholds, the postprocess algorithm may make decisions on when to adjust environmental conditions to maintain RH within an optimal range for efficient measurements. The post-process algorithm may control humidity levels by adjusting environmental parameters such as airflow and temperature. For example, when the humidity level is higher than optimal, the post-process algorithm may trigger actions to increase airflow or activate dehumidifiers to reduce RH to the desired level to improve measurement speed. When the humidity level is lower than optimal, the post-process algorithm may adjust the temperature or introduce moisture into the air to raise RH to the optimal range. The adjustments by the post-process algorithm may ensure that the insulator thickness measurements can be performed more rapidly and consistently, which may contribute to increased productivity and efficiency in microLED fabrication processes. In some embodiments, the post-process algorithm may utilize support vector machines, which may be used to analyze historical data of humidity levels and corresponding measurement times and may learn to classify or regress on the data to determine the optimal humidity range for faster measurements. In some embodiments, the post-process algorithm may utilize random forests, which may analyze historical humidity data along with other environmental variables to predict the optimal conditions for faster measurement speed, and by aggregating the predictions of multiple decision trees, the post-process algorithm may provide more accurate and reliable recommendations for adjusting environmental parameters to maintain optimal humidity levels. The FM post-process module 114 performs the post-process inspection. For example, the film measurement post-process may include data analysis and interpretation, feedback loops for process adjustments, quality control and assurance processes, integration with device performance testing, etc. Once the film thickness and optical properties data have been collected, software tools may compare the measured values against predefined standards and specifications to identify any deviations. The data obtained from the film measurements may be fed back into the film measurement process to make real-time adjustments to reduce defects. The film measurement data is post-processed by being analyzed to identify trends and patterns that may indicate underlying process issues, which may require corrective measures to be implemented. Then, the film measurement data is correlated with the performance characteristics of the microLED to understand how variations in film properties affect the performance of the device. The FM post-process module 114 stores the data in the FM post-process database 124. The FM post-process module 114 stores the data in the FM post-process database 124, such as the data collected from the film measurement post-process 150, the output of the postprocess algorithm, etc. The FM post-process database 124 may contain the experiment ID, the relative humidity, the final insulator thickness, the speed of the measurement, if a rescan is necessary, etc. The FM process database 122 may contain the data for each experiment that is being processed by the film measurement post-process 150. The FM post-process module 114 returns to the base module 108. The base module 108 initiates, at step 206, the FM historical module 116. For example, the FM historical module 116 begins by being initiated by the base module 108. The FM historical module 116 connects to the film measurement pre-process 130, the film measurement process 140, and the film measurement post-process 150. The historical module 116 connects to the film measurement pre-process 130, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process, the film measurement process 140, such as systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected, and the film measurement post-process 150, such as processes and methods after the completion of the film measurement process 140, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc. The FM historical module 116 aggregates the data from the various types of processes and stores the data in the FM network database 126. The FM network database 126 may contain the historical data from the various processes performed by the film measurement pre-process 130, film measurement process 140, and film measurement post-process 150. The FM network database 126 may contain the data parameters collected during the inspection process, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, and control parameters of the process, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc., and data parameters collected during post-processing, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. The FM historical module 116 performs the historical machine learning algorithm on the historical data stored in the FM network database 126. The historical machine learning algorithm may improve each stage of the microLED manufacturing process, such as film measurement pre-process, film measurement process, and film measurement post-process. For example, the historical machine learning algorithm may improve the pre-process by collecting historical data on relative humidity levels and film thickness measurements, analyzing the historical data to identify patterns and correlations between relative humidity levels and film thickness variations, and, based on the analysis, optimize the control of humidity levels using a predictive algorithm to ensure that relative humidity remains within the optimal ranges for accurate film thickness measurements. The historical machine learning algorithm may improve the film measurement process by continuously monitoring relative humidity levels and correlating them with film thickness measurements and measurement speed, and using real-time feedback from the ongoing measurements and historical data to dynamically adjust the measurement parameters to optimize measurement speed to improve the speed and accuracy of film thickness measurements. The historical machine learning algorithm may improve the post-process by identifying trends and patterns associated with out-of-specification measurements and their causes and based on the analysis, implement corrective actions such as recalibrations or adjustments to measurement parameters to minimize the need for rescans or retaking measurements. The FM historical module 116 sends the process adjustments to the FM inspection module 110, the FM process module 112, and the FM post-process module 114. The historical machine learning algorithm may optimize the relative humidity for accurate film thickness measurements. The historical machine learning algorithm may dynamically adjust measurement parameters to optimize measurement speed. The adjustments are sent to the FM inspection module 110 and FM process module 112 to create a more consistent final product, which is then verified by the FM post-process module 114. The FM historical module 116 returns to the base module 108. The base module 108 initiates, at step 208, the FM integration module 118. For example, the FM integration module 118 begins by being initiated by the base module 108. The FM integration module 118 performs the integration machine learning algorithm. The integration machine learning algorithm may facilitate interactions between the film measurement pre-process 130, film measurement process 140, and film measurement postprocess 150 stages of microLED manufacturing. For example, the integration machine learning algorithm may use the data from the pre-process to improve the process by collecting historical data on relative levels, temperature, air pressure, final insulator thickness, and deposition parameters, analyzing the data to establish correlations between environmental factors and film thickness variations., optimizing pre-measurement humidity control strategies based on historical insights and predictive modeling, and during film thickness measurement, continuously monitor the real-time Al-adjusted humidity levels, temperature, air pressure, and film thickness measurements, and use real-time feedback to dynamically adjust measurement parameters and optimize measurement conditions for improved accuracy and consistency. The integration machine learning algorithm may use the data from the post-process to improve the pre-process by analyzing post-measurement data, including relative humidity levels, final insulator thickness, indications of rescans, process duration, etc., identifying trends and patterns related to the effectiveness of pre- process humidity control strategies, refining pre-process humidity control strategies based on insights from data analysis, considering factors such as process duration and the occurrence of rescans, and implementing adjustments to improve the overall effectiveness of pre-process humidity control for subsequent measurements. The integration machine learning algorithm may use the data from the process to improve the pre-process by continuously monitoring real-time AL adjusted humidity levels, temperature, air pressure, film thickness measurements, and measurement speed during film thickness measurement, correlating this data to assess the impact of real-time humidity adjustments on measurement outcomes, considering the influence of temperature and air pressure, and use insights from the process analysis to refine pre-process humidity control strategies, anticipating and managing environmental variations to optimize measurement conditions. The FM integration module 118 connects to the film measurement pre- process Al module 136, the film measurement process Al module 146, and the film measurement post-process Al module 156. The FM integration module 118 connects to the film measurement pre-process 130, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process, the film measurement process 140, such as systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected, and the film measurement post-process 150, such as processes and methods after the completion of the film measurement process 140, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc. The FM integration module 118 sends the process adjustments to the film measurement pre-process Al module 136, the film measurement process Al module 146, and the film measurement post-process Al module 156. For example, the FM integration module 118 may send the optimized pre-process humidity control strategies to each of the processes to improve the accuracy and consistency during the film measurement process. In some embodiments, the FM integration module 118 may send the integration machine learning algorithm to the processes, allowing the systems to further enhance the optimization. The FM integration module 118 returns to the base module 108.

[0063]

[0062] FIG. 3 illustrates the FM inspection module 110. The process begins with the FM inspection module 110 being initiated at step 300 by the base module 108. The FM inspection module 110 connects, at step 302, to the film measurement pre-process 130. The FM inspection module 110 connects with the film measurement pre-process 130, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process. In some embodiments, the FM inspection module 110 may transmit and receive data from the film measurement pre-process 130. In some embodiments, the film measurement pre-process 130 may measure humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. In some embodiments, the FM inspection module 110 may control or send inputs to control the film measurement pre- process 130. The FM inspection module 110 collects, at step 304, the data from the film measurement pre-process 130. The FM inspection module 110 collects the pre-processing data, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, etc. The FM inspection module 110 performs, at step 306, the inspection algorithm. For example, the inspection algorithm may minimize variations in relative humidity and the related insulator thickness variations. For example, the effect of humidity on the thickness of insulator layers may vary due to the different physical and chemical properties of the materials. Polymers may absorb moisture from the environment, which may affect their physical properties, including dimensional changes or swelling, which may lead to variations in the thickness of layers or coatings made from these materials. The interaction between moisture and the polymer matrix may lead to hydrolysis reactions or plasticization, which may modify the material's structure and its physical dimensions, including thickness. Insulators generally do not absorb water in the same way polymers do. However, the presence of water molecules may still affect the growth or formation of insulator layers, primarily through mechanisms such as the facilitation of oxidation reactions or affecting the surface energy and dynamics of the insulator layer formation. In some embodiments, the impact of humidity on the thickness of insulator versus polymer layers may depend on the material's composition, the environmental conditions, and the mechanisms through which moisture interacts with the material.

[0064]

[0063] Higher relative humidity, or RH, may lead to thicker insulator layers on surfaces due to enhanced oxidation reactions or other surface chemistry effects. The impact on polymer insulators may be related to the physical absorption of moisture and its interaction with the polymer matrix. The inspection algorithm may control humidity levels and result in more consistent insulator thickness measurements by utilizing a precision environmental control system. The inspection algorithm may perform a combination of predictive modeling and real-time adjustments to optimize the environmental conditions within the film measurement tool. The inspection algorithm may be a machine learning model, such as a random forest or gradient boosting machine, that may predict the impact of various environmental factors, including humidity, on the insulator layer thickness with high accuracy. The inspection algorithm may be trained on historical data, including a wide range of environmental conditions and their corresponding effects on insulator thickness measurements. The data set may include variables such as temperature, humidity levels, substrate material properties, and previous insulator thickness measurements. The trained inspection algorithm may predict the optimal humidity levels required to achieve a target insulator thickness, taking into account the current environmental conditions and material properties. The inspection algorithm may incorporate a feedback loop using sensors to continuously monitor the actual humidity levels and the insulator thickness measurements, which may be fed back into the inspection algorithm to adjust predictions as needed. In some embodiments, actuators within the environmental control system may adjust the conditions inside the measurement tool based on the inspection algorithm’s recommendations, which may involve controlling dehumidifiers, humidifiers, or air conditioning systems to maintain the desired humidity level. In some embodiments, the control system may be adaptive, capable of adjusting its parameters in real-time based on the inspection algorithm’s output and the feedback received from the environment to ensure the system can respond to any unforeseen changes or drifts in environmental conditions, maintaining the optimal humidity level for consistent insulator thickness measurements. In some embodiments, the inspection algorithm may incorporate optimization techniques, such as genetic algorithms or simulated annealing, to find the best control strategy that minimizes variability in insulator thickness while considering operational constraints and efficiency. In some embodiments, the inspection algorithm may utilize neural networks, such as deep learning models, which may learn from historical data comprising various environmental conditions and their impact on insulator thickness to predict the exact adjustments needed in the environmental control systems to achieve the desired humidity levels that result in optimal insulator thickness, accounting for complex interactions between different factors. In some embodiments, the inspection algorithm may utilize support vector machines for regression, which may be used to predict the optimal environmental conditions necessary for achieving the target insulator thickness on substrates in the microLED process by identifying adjustments needed in the environmental control system to maintain desired conditions by being trained on historical data, including humidity levels and corresponding insulator thickness measurements.

[0065]

[0064] Edge Al and TinyML Example 1: In some embodiments, the system may utilize Edge Al and TinyML. For example, the FM inspection module 110 may employ a TinyML model to perform real-time analysis of environmental data, such as humidity and temperature, collected from sensors integrated within the film measurement pre-process 130. This TinyML model, deployed on a microcontroller with limited computational resources, is specifically trained to predict optimal humidity levels required to achieve target insulator thickness for microLED substrates. The FM inspection module 110 begins by collecting raw data from the film measurement pre-process 130 sensors, which continuously monitor the ambient conditions. This data is fed into the TinyML model, which has been trained using historical data stored in the FM network database 126. The model performs on-device inference, leveraging its lightweight neural network architecture to process the data locally without the need for cloud connectivity, thus ensuring low latency and immediate feedback. Upon detecting a deviation from the optimal humidity range, the TinyML model triggers a control signal to the FM process module 112. The FM process module 112, which acts as the core controller, receives this control signal and adjusts the environmental control systems accordingly. This adjustment may involve activating dehumidifiers or humidifiers connected to the communication interface 104 to maintain the desired humidity level. Simultaneously, the FM process module 112 collects real-time feedback on the adjusted conditions and sends this data back to the FM inspection module 110 for further analysis. The FM inspection module 110, upon receiving the updated environmental data, re-evaluates the conditions using the TinyML model to ensure that the adjustments have brought the humidity within the optimal range. If further adjustments are necessary, the process iterates, with the FM process module 112 making incremental changes based on the continuous feedback loop. Meanwhile, the FM post-process module 114 plays a crucial role in validating the effectiveness of these adjustments. Once the film measurement process 140 is completed, the FM post-process module 114 conducts a detailed analysis of the insulator thickness measurements, comparing them against predefined specifications stored in the FM post-process database 124. The module utilizes a separate Edge Al algorithm to identify any discrepancies or deviations in the measurements that may have resulted from suboptimal environmental conditions during the measurement process. If the Edge Al algorithm detects any anomalies, it sends a notification to the FM historical module 116, which logs the event and updates the historical data set with the new findings. This updated data is then used to retrain the TinyML model periodically, enhancing its predictive accuracy for future measurements. A specific condition that would trigger this Al system is a sudden spike in ambient humidity detected by the sensors, which exceeds the predefined threshold set for optimal microLED fabrication conditions. This triggers the TinyML model to initiate the described control sequence, ensuring that the film measurement process remains within acceptable parameters. Through this detailed interaction between the FM inspection module 110, FM process module 112, and FM post-process module 114, the system achieves a robust and adaptive film measurement process that operates efficiently at the edge, minimizing the need for extensive cloud resources while maintaining high precision and reliability in microLED fabrication.

[0066]

[0065] The FM inspection module 110 sends, at step 308, the determined process data from the inspection algorithm to the FM process module 112. For example, the FM inspection module 110 may send the relative humidity and final insulator thickness to the FM process module 112 to optimize the film measurement process. In some embodiments, the FM inspection module 110 may send temperature, surface roughness, film uniformity, film composition, etc. The FM inspection module 110 stores, at step 310, the data in the FM inspection database 120. The FM inspection module 110 stores the data outputted from the inspection algorithm in the FM inspection database 120, such as the experiment ID, the relative humidity outputted by the inspection algorithm, the final insulator thickness controlled by the inspection algorithm, etc. In some embodiments, the FM inspection database 120 may include other parameter data collected or calculated during the FM inspection module 110 process, such as temperature, surface roughness, film uniformity, film composition, etc. The FM inspection module 110 returns, at step 312, to the base module 108.

[0067]

[0066] FIG. 4 illustrates the FM process module 112. The process begins with the FM process module 112 being initiated at step 400 by the base module 108. The FM process module 112 receives, at step 402, the pre-process data, such as the data on the microLED processed wafer, from the FM inspection module 110. The FM process module 112 receives the outputted data from the inspection algorithm performed in the FM inspection module 110, such as the relative humidity, allowing the film measurement process 140 to determine the optimal control parameters to optimize the film measurement process. In some embodiments, the FM process module 112 may receive temperature, surface roughness, film uniformity, film composition, etc. The FM process module 112 connects, at step 404, to the film measurement process 140. The FM process module 112 connects with the film measurement process 140, such as systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected. In some embodiments, the FM process module 112 may transmit and receive data from the film measurement process 140. The FM process module 112 collects, at step 406, the data from the film measurement process 140. The FM process module 112 collects the processing data, such as humidity, temperature, surface roughness, film uniformity, film composition, etc. The FM process module 112 performs, at step 408, the control algorithm. The control algorithm may optimize the measurement speed by controlling the humidity levels. For example, changes in humidity may affect the measurement speed as these methods are typically rapid and based on the interaction of light with the film surface. Humidity may influence the accuracy and the effective speed of the process indirectly by necessitating additional steps or corrections. Humidity may alter the refractive index of hygroscopic insulator films by changing their composition through moisture absorption. Since optical thickness measurement methods, such as ellipsometry or reflectometry, rely on changes in light polarization or reflectance that depend on the film's refractive index, variations in humidity can lead to inaccuracies in the measured film thickness if not accounted for. Some materials may physically swell or contract, with changes in humidity, affecting the actual film thickness, and identifying and correcting for these physical changes to ensure accuracy may add time to the overall measurement procedure. In environments with high humidity, condensation may form on the surfaces to be measured, scattering or absorbing the incident light and interfering with optical measurements, which may require the substrates to be dried or conditioned before measurement, indirectly slowing down the process. The control algorithm may be a linear regression model, or a decision tree algorithm utilized to control humidity levels and optimize measurement speed. For example, the control algorithm may analyze historical humidity data and corresponding measurement times to identify linear relationships between RH levels and measurement speed. By extrapolating this relationship, the control algorithm may predict the optimal RH level for minimizing measurement time in realtime.

[0068]

[0067] In some embodiments, the control algorithm may analyze the data to identify threshold values of humidity at which measurement speed tends to decrease, and based on these thresholds, the control algorithm may make decisions on when to adjust environmental conditions to maintain RH within an optimal range for efficient measurements, the control algorithm may control humidity levels by adjusting environmental parameters such as airflow and temperature. For instance, in experiments where the humidity level is higher than optimal, the control algorithm may trigger actions to increase airflow or activate dehumidifiers to reduce RH to the desired level to improve measurement speed. In experiments with lower humidity levels, the control algorithm may adjust the temperature or introduce moisture into the air to raise RH to the optimal range. These adjustments from the control algorithm ensure that the insulator thickness measurements may be performed more rapidly and consistently, contributing to increased productivity and efficiency in microLED fabrication processes. In some embodiments, the control algorithm may utilize support vector machines to optimize measurement speed, in which the control algorithm may analyze historical data of humidity levels and corresponding measurement times. The control algorithm may learn to classify or regress on the data to determine the optimal humidity range for faster measurements. In some embodiments, the control algorithm may utilize random forests to analyze historical humidity data along with other environmental variables to predict the optimal conditions for faster measurement speed. By aggregating the predictions of multiple decision trees, the control algorithm may provide more accurate and reliable recommendations for adjusting environmental parameters to maintain optimal humidity levels. The FM process module 112 adjusts, at step 410, the control parameters of the film measurement process 140. In some embodiments, the FM process module 112 may optimize the speed of the measurements by optimizing the humidity levels. In some embodiments, the FM process module may optimize the temperature. The FM process module 112 executes, at step 412, the film measurement process. The film measurement process may include collecting baseline measurements, data acquisition, data analysis, parameter calculation, interpretation and validation, etc.

[0069]

[0068] For example, some measurement techniques, such as ellipsometry or reflectometry, may require a baseline measurement of the bare substrate without the thin film and this is used as a reference for comparing measurements with the thin film present. The measurement process begins, and data is acquired according to the selected technique, which may involve scanning the sample surface, illuminating it with light of specific wavelengths, applying electrical probes, etc., depending on the measurement method. The acquired data is analyzed to extract relevant information about the thin film properties, which may involve mathematical modeling, fitting experimental data to theoretical models, performing calculations based on measurement results, etc. Using the analyzed data, various parameters of interest are calculated. For example, in ellipsometry, parameters such as film thickness, refractive index, and extinction coefficient are determined from the measured ellipsometry angles. The calculated parameters are interpreted in the context of the thin film's intended application and manufacturing process. Any deviations from expected values are investigated to ensure measurement accuracy and reliability. The FM process module 112 stores, at step 414, the data in the FM process database 122. The FM process module 112 stores the data in the FM process database 122, such as the data collected from the film measurement process 140, the output of the inspection algorithm, and the output of the control algorithm. The FM process database 122 may contain the experiment ID, the relative humidity, the final insulator thickness, the optimized speed of the measurement, etc. The FM process module 112 returns, at step 416, to the base module 108.

[0070]

[0069] FIG. 5 illustrates the FM post-process module 114. The process begins with the FM postprocess module 114 being initiated at step 500 by the base module 108. The FM post-process module 114 connects, at step 502, to the film measurement post-process 150. The FM post-process module 114 may connect to the film measurement post-process 150, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc.

[0070] Synthetic Data and AI-Generated Digital Twins Example 1: In some embodiments, the system may utilize Synthetic Data and AI-Generated Digital Twins. For example, the system can simulate the microLED fabrication environment by creating a digital twin of the cleanroom and film measurement equipment. This digital twin is generated using synthetic data derived from historical measurements, environmental conditions, and operational parameters stored in the FM network database 126. The digital twin serves as a virtual replica of the physical system, allowing for predictive modeling and real-time process optimization. The FM inspection module 110, FM process module 112, and FM post-process module 114 interact with this digital twin to enhance measurement accuracy and efficiency. The FM inspection module 110 uses synthetic data to simulate various humidity levels and their impact on insulator thickness, enabling the inspection algorithm to predict optimal humidity settings for each batch of microLED substrates. This predictive capability is achieved by inputting historical data on humidity variations and corresponding thickness measurements into a machine learning model, such as a gradient boosting machine, which outputs the ideal humidity conditions to achieve target thickness specifications. The FM process module 112 receives these predicted conditions as control signals, adjusting realtime environmental parameters like airflow and temperature in the physical cleanroom to match the optimal settings determined by the digital twin.

[0071]

[0071] This adjustment is facilitated through a communication interface 104 that transmits control signals to environmental control systems, ensuring that the physical conditions align with the virtual model's predictions. Meanwhile, the FM post-process module 114 uses the digital twin to conduct virtual inspections of the resultant film measurements. By comparing real-time data from the film measurement process 140 with the digital twin's predictions, the post-process algorithm identifies deviations and triggers corrective actions, such as rescanning or parameter recalibration. For instance, if the measured insulator thickness deviates from the predicted value by more than 0.02 microns, the FM post-process module 114 sends a signal to the film measurement process 140 to initiate a rescan. This feedback loop is continuously refined through the FM historical module 116, which aggregates data from both the digital twin and physical processes to update the predictive models. The historical machine learning algorithm within the FM historical module 116 analyzes patterns and trends, informing the digital twin's future simulations and ensuring that the system adapts to evolving process conditions. The integration of synthetic data and ALgenerated digital twins thus enables a dynamic and responsive film measurement system, where real-time adjustments are grounded in robust virtual modeling, enhancing both the precision and throughput of microLED fabrication.

[0072]

[0072] The FM post-process module 114 collects, at step 504, the post-process data. The FM postprocess module 114 may collect the parameter data, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. The FM post-process module 114 extracts, at step 506, the data from the FM inspection database 120 and FM process database 122. The FM post-process module 114 extracts the data, such as material data, parameter data, etc., of the pre-process and film measurement process. The FM post-process module 114 performs, at step 508, the post-process algorithm. The post-process algorithm may optimize the environmental conditions to minimize the need for rescanning or retaking measurements. For example, in film measurement systems, there may be a mechanism for automated rescanning or retaking measurements if the initial measurement falls outside the specified tolerances or specifications to ensure that measurements are accurate and reliable. The insulator may be rescanned by determining if a measurement is outside a specification threshold. Once the threshold has been exceeded, the system automatically triggers the need for a rescan, then the system performs another measurement, and the new measurement is compared against the specific tolerances or specifications. Once a valid measurement is obtained, the results are documented and reported. The measured data may change through sample variation, environmental changes, instrument drift, surface contamination, etc. The post-process algorithm may analyze historical humidity data and corresponding measurement times to identify linear relationships between RH levels and measurement speed. By extrapolating this relationship, the post-process algorithm may predict the optimal RH level for minimizing measurement time in real-time. In some embodiments, the post-process algorithm may utilize decision tree algorithms, which may analyze the data to identify threshold values of humidity at which measurement speed tends to decrease, and based on these thresholds, the post-process algorithm may make decisions on when to adjust environmental conditions to maintain RH within an optimal range for efficient measurements. The post-process algorithm may control humidity levels by adjusting environmental parameters such as airflow and temperature.

[0073]

[0073] For example, when the humidity level is higher than optimal, the post-process algorithm may trigger actions to increase airflow or activate dehumidifiers to reduce RH to the desired level to improve measurement speed. When the humidity level is lower than optimal, the post-process algorithm may adjust the temperature or introduce moisture into the air to raise RH to the optimal range. The adjustments by the post-process algorithm may ensure that the insulator thickness measurements can be performed more rapidly and consistently, which may contribute to increased productivity and efficiency in microLED fabrication processes. In some embodiments, the postprocess algorithm may utilize support vector machines, which may be used to analyze historical data of humidity levels and corresponding measurement times and may learn to classify or regress on the data to determine the optimal humidity range for faster measurements. In some embodiments, the post-process algorithm may utilize random forests, which may analyze historical humidity data along with other environmental variables to predict the optimal conditions for faster measurement speed, and by aggregating the predictions of multiple decision trees, the post-process algorithm may provide more accurate and reliable recommendations for adjusting environmental parameters to maintain optimal humidity levels.

[0074]

[0074] Large Language Models (LLMs) Example 1: In some embodiments, the system may utilize Large Language Models (LLMs). For example, the FM inspection module 110, FM process module 112, and FM post-process module 114 may interact to enhance the film measurement process for microLED substrates by leveraging LLMs to interpret complex data patterns and provide actionable insights. The system begins with the FM inspection module 110 collecting data from the film measurement pre-process 130, including environmental parameters such as humidity and temperature, as well as material properties like insulator thickness and surface roughness. This data is then fed into the LLM, which has been pre-trained on extensive datasets encompassing various environmental conditions and their effects on film measurements. The LLM analyzes this input to predict potential deviations in insulator thickness that may arise during the film measurement process. The LLM generates a control signal based on its analysis, which is transmitted to the FM process module 112. This control signal may include recommended adjustments to the environmental conditions, such as modifying the airflow rate or adjusting the temperature to maintain optimal humidity levels, ensuring consistency in insulator thickness measurements. The FM process module 112 receives this control signal and interfaces with the film measurement process 140, where it adjusts the operational parameters of the measurement equipment, such as an ellipsometer or spectrophotometer, to align with the LLM's recommendations. This adjustment is crucial for maintaining the precision of optical measurements, as changes in environmental conditions can affect the refractive index of the materials being measured. As the measurement process progresses, real-time data, including the speed of measurements and optical properties like refractive index and transmission spectra, are continuously fed back into the LLM. This feedback loop allows the LLM to refine its predictions and control signals dynamically, enhancing the accuracy and efficiency of the measurement process. Upon completion of the film measurement process, the FM post-process module 114 utilizes the LLM to conduct a comprehensive analysis of the collected data. The LLM compares the measured parameters against predefined specifications and historical data stored in the FM network database 126 to identify any deviations or anomalies. If the LLM detects that the measurement speed exceeded the optimal threshold due to environmental fluctuations, it generates a recommendation for a rescan, which is communicated back to the FM process module 112. Additionally, the LLM provides insights into potential root causes of any deviations, such as unexpected humidity spikes, and suggests corrective actions to prevent recurrence in future measurements. This information is stored in the FM post-process database 124 for further analysis and continuous improvement. A specific condition that would trigger the LLM system is a sudden increase in relative humidity detected by the sensors during the film measurement process, which could potentially lead to inaccuracies in insulator thickness measurements. In response, the LLM would analyze the situation, predict the impact on measurement accuracy, and issue a control signal to adjust environmental parameters accordingly. This detailed interaction between the FM inspection module 110, FM process module 112, and FM post-process module 114, facilitated by the LLM, ensures a robust and adaptive film measurement system capable of maintaining high precision and reliability in microLED fabrication.

[0075]

[0075] Generative Video Al Example 1: In some embodiments, the system may utilize Generative Video Al. For example, the FM inspection module 110, FM process module 112, and FM postprocess module 114 may interact to generate real-time visualizations of the film measurement process, enhancing the accuracy and efficiency of microLED substrate analysis. The input data for the Generative Video Al may include high-resolution images and video feeds captured by cameras integrated within the film measurement process 140. These cameras continuously monitor the microLED substrates, capturing detailed visual data of the epitaxial layer and its optical properties. The Generative Video Al processes this input data using convolutional neural networks (CNNs) to identify patterns and anomalies in the film's thickness and uniformity. The decision logic of the Al involves comparing the real-time visual data against a database of historical measurements stored in the FM network database 126. This comparison allows the Al to detect deviations from expected film properties and predict potential defects or inconsistencies. Upon identifying such deviations, the Al generates a control signal that is sent to the FM process module 112. This control signal may instruct the module to adjust specific parameters of the film measurement process, such as altering the light wavelength used in spectroscopic reflectometry to enhance measurement precision. Additionally, the Al may trigger the FM post-process module 114 to initiate a rescan or retake of measurements if the detected deviations exceed predefined thresholds. The FM postprocess module 114 then employs the post-process algorithm to reassess the measurement data, ensuring compliance with quality standards.

[0076]

[0076] A specific example of a condition that would trigger the Generative Video Al system is the detection of a sudden change in the refractive index of the film, which may indicate an environmental fluctuation affecting measurement accuracy. In such cases, the Al not only alerts the FM process module 112 to recalibrate the measurement equipment but also updates the FM inspection database 120 with new parameters, ensuring future measurements account for similar environmental conditions. The interaction between these modules is facilitated by the communication interface 104, which ensures seamless data exchange and control signal transmission. The Generative Video Al also incorporates feedback loops, where the outcomes of process adjustments are fed back into the Al model to refine its predictive capabilities. This continuous learning mechanism allows the system to adapt to new patterns and improve its decision-making accuracy over time. By integrating Generative Video Al, the system provides a robust framework for real-time monitoring and dynamic adjustment of the film measurement process, ensuring high precision and reliability in microLED fabrication.

[0077]

[0077] The FM post-process module 114 performs, at step 510, the post-process inspection. For example, the film measurement post-process may include data analysis and interpretation, feedback loops for process adjustments, quality control and assurance processes, integration with device performance testing, etc. Once the film thickness and optical properties data have been collected, software tools may compare the measured values against predefined standards and specifications to identify any deviations. The data obtained from the film measurements may be fed back into the film measurement process to make real-time adjustments to reduce defects. The film measurement data is post-processed by being analyzed to identify trends and patterns that may indicate underlying process issues, which may require corrective measures to be implemented. Then, the film measurement data is correlated with the performance characteristics of the microLED to understand how variations in film properties affect the performance of the device. The FM postprocess module 114 stores, at step 512, the data in the FM post-process database 124. The FM postprocess module 114 stores the data in the FM post-process database 124, such as the data collected from the film measurement post-process 150, the output of the post-process algorithm, etc. The FM post-process database 124 may contain the experiment ID, the relative humidity, the final insulator thickness, the speed of the measurement, if a rescan is necessary, etc. The FM process database 122 may contain the data for each experiment that is being processed by the film measurement post-process 150. The FM post-process module 114 returns, at step 514, to the base module 108.

[0078]

[0078] FIG. 6 illustrates the FM historical module 116. The process begins with the FM historical module 116 being initiated at step 600 by the base module 108. The FM historical module 116 connects, at step 602, to the film measurement pre-process 130, the film measurement process 140, and the film measurement post-process 150. The historical module 116 connects to the film measurement pre-process 130, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process, the film measurement process 140, such as systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected, and the film measurement post-process 150, such as processes and methods after the completion of the film measurement process 140, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc. The FM historical module 116 aggregates, at step 604, the data from the various types of processes and stores the data in the FM network database 126. The FM network database 126 may contain the historical data from the various processes performed by the film measurement pre-process 130, film measurement process 140, and film measurement post-process 150. The FM network database 126 may contain the data parameters collected during the inspection process, such as humidity, temperature, insulator thickness measurements, surface roughness, film uniformity, film composition, and control parameters of the process, such as speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states, molecular structure, etc., and data parameters collected during post-processing, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc. The FM historical module 116 performs, at step 606, the historical machine learning algorithm on the historical data stored in the FM network database 126. The historical machine learning algorithm may improve each stage of the microLED manufacturing process, such as film measurement pre-process, film measurement process, and film measurement post-process. For example, the historical machine learning algorithm may improve the pre-process by collecting historical data on relative humidity levels and film thickness measurements, analyzing the historical data to identify patterns and correlations between relative humidity levels and film thickness variations, and, based on the analysis optimize the control of humidity levels using a predictive algorithm to ensure that relative humidity remains within the optimal ranges for accurate film thickness measurements. The historical machine learning algorithm may improve the film measurement process by continuously monitoring relative humidity levels and correlating them with film thickness measurements and measurement speed, and using real-time feedback from the ongoing measurements and historical data to dynamically adjust the measurement parameters to optimize measurement speed to improve the speed and accuracy of film thickness measurements. At step 608, the historical machine learning algorithm may improve the post-process by identifying trends and patterns associated with out-of-specification measurements and their causes and, based on the analysis, implement corrective actions such as recalibrations or adjustments to measurement parameters to minimize the need for re-scans or retaking measurements.

[0079]

[0079] Large Language Models (LLMs) Example 2: In some embodiments, the system may utilize Large Language Models (LLMs). For example, the FM historical module 116 may employ LLMs to analyze the vast repository of historical data stored in the FM network database 126, which includes parameters such as humidity, temperature, insulator thickness measurements, and optical properties. The LLMs are trained on this extensive dataset to understand the intricate relationships between these variables. When a new batch of microLED substrates is introduced, the FM inspection module 110 collects initial environmental data, such as current humidity and temperature, from the film measurement pre-process 130. This data is then fed into the LLMs within the FM historical module 116. The LLMs use their learned patterns to predict potential deviations in film thickness and optical properties that might arise under the current conditions. Based on these predictions, the FM historical module 116 generates a set of recommended adjustments to the film measurement process 140, such as altering the humidity control parameters. These recommendations are transmitted as control signals to the FM process module 112, which adjusts the environmental control systems, such as dehumidifiers and temperature regulators, to maintain optimal conditions during the film measurement process. The FM process module 112 then executes the film measurement process, collecting real-time data on film thickness and optical properties. This data is continuously fed back into the FM historical module 116, where the LLMs refine their predictions and adjustments in a feedback loop. Additionally, the FM post-process module 114 performs a final inspection of the processed substrates, comparing the measured film properties against predefined specifications. If deviations are detected, the FM post-process module 114 sends this information back to the FM historical module 116, where the LLMs analyze the discrepancies and update their predictive models accordingly. This iterative process ensures that the system continuously learns from each batch, improving its predictive accuracy and control precision over time. An example condition that would trigger the LLM system is a detected increase in ambient humidity by 15% above the baseline, which could potentially affect film thickness measurements. Upon detecting such a condition, the FM inspection module 110 would immediately relay this data to the LLMs in the FM historical module 116, prompting a recalibration of the environmental controls to preemptively mitigate any adverse effects. Through this detailed interplay of modules and data flows, the system ensures precise and consistent film measurements for microLED substrates, leveraging the predictive power of LLMs to adapt to dynamic environmental conditions.

[0080]

[0080] Al Agents Example 1: In some embodiments, the system may utilize Al Agents. For example, the FM historical module 116, FM process module 112, and FM post-process module 114 may collaborate through Al Agents to optimize the film measurement process for microLED substrates. The Al Agents receive input data from the FM network database 126, which contains historical data on environmental conditions such as humidity and temperature, as well as film thickness measurements and process parameters. The Al Agents are programmed with decision logic that employs a combination of machine learning algorithms, including random forests and support vector machines, to analyze this historical data and identify patterns or trends that correlate specific environmental conditions with optimal measurement outcomes. Upon detecting a deviation from optimal conditions, the Al Agents generate control outputs to adjust the environmental parameters in real-time. For instance, if the FM historical module 116 identifies that a relative humidity level above 55% RH consistently leads to increased measurement times, the Al Agents will send a control signal to the FM process module 112 to activate dehumidifiers or increase airflow, thereby reducing the humidity to the optimal range of 45-50% RH. This control signal is transmitted through the communication interface 104, ensuring seamless integration with the environmental control systems. Concurrently, the FM post-process module 114 uses Al Agents to evaluate the measurement outcomes stored in the FM post-process database 124. If the postprocess algorithm detects that the insulator thickness measurements are outside the specified tolerances, the Al Agents will trigger a re-scan by sending a command to the film measurement process 140. This command includes specific parameters such as the adjusted humidity level and the recalibrated measurement speed, which are calculated based on the Al Agents' analysis of the historical data. The Al Agents also facilitate a feedback loop by updating the FM historical module 116 with new data from the rescans, allowing the system to learn and refine its predictive models continuously. An example condition that would trigger the Al system is a sudden spike in humidity detected by sensors during the film measurement process, which the Al Agents recognize as a potential risk for measurement inaccuracies. In response, the Al Agents execute a pre-defined protocol to adjust the environmental conditions and initiate a re-scan if necessary, ensuring that the measurements remain within the desired specifications. This interaction between the FM historical module 116, FM process module 112, and FM post-process module 114, orchestrated by Al Agents, enables the system to maintain high precision and efficiency in the automated film measurement of microLED substrates.

[0081]

[0081] Al Agents Example 2: In some embodiments, the system may employ Al Agents to enhance the film measurement process for microLED substrates by focusing on predictive maintenance and anomaly detection. The FM inspection module 110, FM integration module 118, and FM historical module 116 collaborate through Al Agents to ensure system reliability and performance optimization. The Al Agents receive input data from the FM inspection database 120, which includes real-time sensor data on equipment health indicators such as vibration levels, power consumption, and operational temperature. The Al Agents are equipped with decision logic that utilizes machine learning algorithms, including neural networks and anomaly detection models, to analyze these indicators and predict potential equipment failures or deviations from normal operating conditions. Upon identifying an anomaly, such as an unexpected increase in vibration levels that could indicate mechanical wear, the Al Agents generate a control output to initiate a maintenance protocol. This control output is transmitted to the FM integration module 118, which coordinates with the FM process module 112 to schedule a maintenance window, ensuring minimal disruption to ongoing operations. The FM integration module 118 then sends a control signal through the communication interface 104 to adjust the film measurement schedule, temporarily pausing operations to allow for necessary maintenance activities. Concurrently, the FM historical module 116 updates its database with the new maintenance and anomaly data, enabling continuous improvement of the predictive models. An example condition triggering the Al system is a detected rise in operational temperature beyond the predefined threshold, which the Al Agents identify as a precursor to equipment overheating. In response, the Al Agents execute a pre-defined maintenance protocol, ensuring that the system operates within safe parameters. This interaction between the FM inspection module 110, FM integration module 118, and FM historical module 116, orchestrated by Al Agents, facilitates proactive maintenance and enhances the reliability of the automated film measurement system for microLED substrates.

[0082]

[0082] Multimodal Al Example 1: In some embodiments, the system may utilize Multimodal Al. For example, the FM process module 112, FM post-process module 114, and FM historical module 116 may interact to enhance the accuracy and efficiency of film measurements in microLED fabrication. The Multimodal Al system receives input data from various sensors and databases, including humidity levels, temperature, insulator thickness measurements, and optical properties such as refractive index and absorption coefficient. The FM process module 112 begins by receiving pre-process data from the FM inspection module 110, which includes environmental parameters such as humidity and temperature. This data is then used to calibrate the film measurement process 140. The Multimodal Al utilizes a decision-making algorithm that combines data from the FM inspection database 120 and the FM process database 122 to predict optimal measurement conditions. For instance, if the humidity level deviates from the optimal range of 50% RH, the Al system may issue a control signal to adjust the environmental parameters by activating dehumidifiers or humidifiers to stabilize the humidity level. The Al system also analyzes optical properties data, such as changes in refractive index, to adjust the measurement speed and ensure accurate thickness readings. This is achieved by dynamically altering the measurement protocol within the film measurement process 140, based on real-time feedback from the sensors. The FM post-process module 114 then takes the processed data and performs a post-process algorithm that identifies any deviations from expected values. If the insulator thickness measurements exceed a specified threshold, a control signal is sent to initiate a re-scan of the affected substrates. The post-process data, including any rescans and adjustments made, is stored in the FM post-process database 124. The FM historical module 116 aggregates this data along with historical records from the FM network database 126 to perform a historical machine learning algorithm. This algorithm identifies long-term trends and patterns, such as recurring environmental conditions that lead to measurement inaccuracies. The historical insights are then used to update the predictive models in the FM process module 112 and FM post-process module 114, allowing for continuous improvement of the measurement process. For example, if the historical data indicates that a specific range of temperature fluctuations consistently affects measurement accuracy, the Al system may adjust the baseline temperature control parameters to mitigate these effects. A specific condition that would trigger the Al system is a sudden spike in humidity levels detected by the sensors, which could potentially affect the film thickness measurements. In response, the Al system would immediately assess the impact of this environmental change using its predictive model and issue control signals to adjust the measurement parameters accordingly. This interaction between the FM process module 112, FM post-process module 114, and FM historical module 116, facilitated by the Multimodal Al, ensures that the film measurement process is both adaptive and precise, allowing for real-time adjustments and long-term process optimization.

[0083]

[0083] Generative Video Al Example 2: In some embodiments, the system may implement Generative Video Al to enhance predictive maintenance and operational efficiency within the microLED fabrication process. Specifically, the FM historical module 116, FM integration module 118, and film measurement process Al module 146 collaborate to utilize video-based predictive analytics for equipment health monitoring. The input data for the Generative Video Al consists of high-resolution video streams captured by cameras strategically positioned to observe the mechanical components of the film measurement equipment, such as motors and alignment systems. These video streams are processed using recurrent neural networks (RNNs) to analyze temporal patterns and detect early signs of mechanical wear or misalignment. The decision logic of the Al involves comparing the real-time video data against a repository of historical maintenance records and operational benchmarks stored in the FM network database 126. By identifying discrepancies or deviations from normal operational patterns, the Al can infer potential equipment failures or inefficiencies. Upon detecting such conditions, a control signal is generated and dispatched to the FM integration module 118, which coordinates with the film measurement process Al module 146 to adjust operational parameters, such as modifying the speed of the measurement process to prevent further mechanical stress. Additionally, the FM historical module 116 updates the FM network database 126 with new insights, enabling the system to refine its predictive models for future maintenance scheduling. A specific scenario that would activate the Generative Video Al system is the identification of abnormal vibration patterns in the video data, indicative of potential motor misalignment. In response, the Al not only prompts the FM integration module 118 to initiate corrective actions but also logs the event in the FM network database 126 for subsequent analysis and model training. The communication interface 104 facilitates seamless data exchange and control signal dissemination among the modules, ensuring real-time responsiveness. This integration of Generative Video Al fosters a proactive maintenance strategy, reducing downtime and enhancing the reliability of the microLED fabrication process through continuous learning and adaptation to evolving operational conditions.

[0084]

[0084] Edge Al and TinyML Example 2: In an alternative embodiment, the system leverages Edge Al and TinyML to enhance the efficiency of the film measurement post-process 150. The FM historical module 116 employs a TinyML model to analyze historical data patterns stored in the FM network database 126, focusing on deviations in measurement accuracy and speed. This TinyML model, optimized for deployment on edge devices with constrained resources, processes historical data to identify optimal environmental conditions that minimize measurement discrepancies. Upon identifying a pattern or trend indicating potential inaccuracies, the TinyML model generates an inference indicating necessary adjustments to the measurement parameters. This inference is communicated to the FM integration module 118, which acts as a conduit for implementing these adjustments across the system. The FM integration module 118, upon receiving the inference, dispatches a control signal to the FM process module 112 to adjust parameters such as measurement speed or optical alignment. Concurrently, the FM integration module 118 updates the FM inspection module 110 with refined environmental control strategies, ensuring that future measurements commence under optimized conditions. This iterative feedback loop is further refined by the FM post-process module 114, which conducts a comprehensive analysis of the final measurement data, comparing it against the specifications stored in the FM post-process database 124. Should the FM post-process module 114 detect any anomalies or deviations from expected values, it triggers a notification back to the FM historical module 116, prompting a reevaluation of the TinyML model's parameters and retraining as necessary. A specific condition that activates this Al system is the detection of a consistent pattern of measurement inaccuracies in the historical data, prompting the TinyML model to infer corrective actions. Through the coordinated interaction of the FM historical module 116, FM integration module 118, FM process module 112, and FM post-process module 114, the system achieves an adaptive, edgebased optimization of the film measurement process, enhancing precision and reliability in microLED fabrication without the dependency on extensive cloud resources.

[0085]

[0085] Synthetic Data and AI-Generated Digital Twins Example 2: In some embodiments, the system may utilize Synthetic Data and AI-Generated Digital Twins. For example, the system may generate synthetic data to simulate various environmental conditions, such as humidity and temperature, that affect the film measurement process of microLED substrates. The FM inspection module 110 may employ this synthetic data to create ALgenerated digital twins of the microLED substrates, allowing for a virtual representation of the substrates under varying conditions. The digital twin serves as a dynamic model that mirrors the physical substrate, enabling the system to predict potential deviations in insulator thickness due to environmental changes. This predictive modeling is achieved by inputting the synthetic data into the inspection algorithm, which is a machine learning model trained on historical data, including variables such as temperature, humidity levels, substrate material properties, and previous insulator thickness measurements. The inspection algorithm processes this data to predict the optimal environmental conditions required to achieve the target insulator thickness. The FM process module 112 receives these predictions and adjusts the control parameters of the film measurement process 140 accordingly. For instance, if the digital twin indicates a potential increase in insulator thickness due to higher humidity levels, the FM process module 112 may adjust the airflow or activate dehumidifiers to maintain the desired humidity level. This adjustment is executed through a control signal sent from the FM process module 112 to the environmental control system, ensuring that the physical measurement process aligns with the optimal conditions predicted by the digital twin. Concurrently, the FM historical module 116 aggregates data from both the synthetic simulations and actual measurements, storing this information in the FM network database 126. The historical machine learning algorithm within the FM historical module 116 analyzes this aggregated data to identify patterns and correlations between environmental conditions and film thickness variations. These insights are then used to refine the predictive models in the FM inspection module 110, enhancing the accuracy of future digital twin simulations. The FM integration module 118 plays a crucial role in this process by facilitating the continuous update of the predictive models and control strategies. It receives process adjustments from the FM historical module 116 and disseminates these updates to the FM inspection module 110, FM process module 112, and FM post-process module 114. This ensures that all modules operate with the most current and accurate data, maintaining a closed-loop system that adapts to both simulated and real-world conditions. A specific condition that may trigger this Al system is a detected deviation in the real-time humidity levels from the optimal range during the film measurement process. Upon detection, the system initiates the generation of synthetic data to simulate potential outcomes, updates the digital twin, and adjusts the physical process parameters to mitigate any adverse effects, thereby ensuring consistent and accurate film measurements. This detailed interaction between the modules, driven by synthetic data and digital twins, provides a robust framework for optimizing the film measurement process in microLED fabrication.

[0086]

[0086] The FM historical module 116 sends, at step 610, the process adjustments to the FM inspection module 110, the FM process module 112, and the FM post-process module 114. The historical machine learning algorithm may optimize the relative humidity for accurate film thickness measurements. The historical machine learning algorithm may dynamically adjust measurement parameters to optimize measurement speed. The adjustments are sent to the FM inspection module 110 and FM process module 112 to create a more consistent final product, which is then verified by the FM post-process module 114. The FM historical module 116 returns, at step 612, to the base module 108.

[0087]

[0087] FIG. 7 illustrates the FM integration module 118. The process begins with the FM integration module 118 being initiated at step 700 by the base module 108. The FM integration module 118 performs, at step 702, the integration machine learning algorithm. The integration machine learning algorithm may facilitate interactions between the film measurement pre-process 130, film measurement process 140, and film measurement post-process 150 stages of microLED manufacturing. For example, the integration machine learning algorithm may use the data from the pre-process to improve the process by collecting historical data on relative levels, temperature, air pressure, final insulator thickness, and deposition parameters, analyzing the data to establish correlations between environmental factors and film thickness variations., optimizing premeasurement humidity control strategies based on historical insights and predictive modeling, and during film thickness measurement, continuously monitor the real-time Al-adjusted humidity levels, temperature, air pressure, and film thickness measurements, and use real-time feedback to dynamically adjust measurement parameters and optimize measurement conditions for improved accuracy and consistency. The integration machine learning algorithm may use the data from the post-process to improve the pre-process by analyzing post-measurement data, including relative humidity levels, final insulator thickness, indications of rescans, process duration, etc., identifying trends and patterns related to the effectiveness of pre-process humidity control strategies, refining pre-process humidity control strategies based on insights from data analysis, considering factors such as process duration and the occurrence of rescans, and implementing adjustments to improve the overall effectiveness of pre-process humidity control for subsequent measurements. The integration machine learning algorithm may use the data from the process to improve the pre- process by continuously monitoring real-time Al-adjusted humidity levels, temperature, air pressure, film thickness measurements, and measurement speed during film thickness measurement, correlating this data to assess the impact of real-time humidity adjustments on measurement outcomes, considering the influence of temperature and air pressure, and use insights from the process analysis to refine pre-process humidity control strategies, anticipating and managing environmental variations to optimize measurement conditions.

[0088]

[0088] Multimodal Al Example 2. In some embodiments, the system may employ Multimodal Al to optimize the integration of film measurement data with production processes. The FM integration module 118, FM inspection module 110, and FM historical module 116 collaborate to enhance the real-time adaptability of the microLED fabrication process. The Multimodal Al system receives input data from the FM inspection database 120, including parameters such as surface roughness, film uniformity, and chemical composition. The FM inspection module 110 initiates by analyzing pre-process data to determine the initial quality of the substrates. This data is then communicated to the FM integration module 118, which utilizes a machine learning algorithm to assess the compatibility of the current batch with historical production data stored in the FM network database 126. The Al system infers potential adjustments in the production parameters by comparing current substrate characteristics with historical patterns of successful fabrication outcomes. Upon detecting discrepancies, such as deviations in surface roughness beyond acceptable limits, the Al system generates a control output that modifies the deposition parameters in the film measurement process 140. This control signal may involve adjusting the deposition rate or altering the chemical vapor composition to achieve the desired uniformity. Concurrently, the FM historical module 116 performs a comprehensive analysis of past production cycles to identify recurring conditions that have historically led to suboptimal outcomes. The insights gained are used to refine the predictive models within the FM integration module 118, ensuring that future substrate batches are processed with improved precision. For instance, if historical data reveals that a specific range of chemical composition variations consistently results in enhanced device performance, the Al system may adjust the baseline chemical parameters to align with these findings. A specific condition that would trigger the Al system is the detection of an unexpected increase in surface roughness, which could compromise the microLED performance. In response, the Al system would immediately evaluate the impact using its inference model and issue control signals to adjust the deposition parameters accordingly. This interaction among the FM integration module 118, FM inspection module 110, and FM historical module 116, facilitated by the Multimodal Al, ensures that the film measurement process is seamlessly integrated with production adjustments, allowing for both immediate corrective actions and strategic long-term improvements.

[0089]

[0089] The FM integration module 118 connects, at step 704, to the film measurement pre-process Al module 136, the film measurement process Al module 146, and the film measurement postprocess Al module 156. The FM integration module 118 connects to the film measurement pre- process 130, such as preparing and positioning the film to be measured, including cleaning the surface, ensuring proper alignment, etc., utilized before the film measurement process 140, designed to prepare the microLED for optimal performance during the film measurement process, the film measurement process 140, such as systems or equipment for automated film measurement of microLED chips in which precision measurements of thin film thickness and optical properties are collected, and the film measurement post-process 150, such as processes and methods after the completion of the film measurement process 140, such as data analysis and interpretation, integrating measurement data into the production process, quality control, etc. The FM integration module 118 sends, at step 706, the process adjustments to the film measurement pre-process Al module 136, the film measurement process Al module 146, and the film measurement post-process Al module 156. For example, the FM integration module 118 may send the optimized pre-process humidity control strategies to each of the processes to improve the accuracy and consistency during the film measurement process. In some embodiments, the FM integration module 118 may send the integration machine learning algorithm to the processes, allowing the systems to further enhance the optimization. The FM integration module 118 returns, at step 708, to the base module 108.

[0090]

[0090] FIG. 8 illustrates the FM inspection database 120. The FM inspection database 120 provides an example of the results of the inspection algorithm performed in the FM inspection module 110. The FM inspection database 120 may contain the experiment ID, the relative humidity outputted by the inspection algorithm, the final insulator thickness controlled by the inspection algorithm, etc. The first example illustrates the results of 10 experiments conducted to measure the insulator thickness on microLED process substrates using a film measurement device in an automated manner. Each row represents a distinct experiment, detailing the experiment ID, the humidity level in relative humidity or RH during the measurement, and the final insulator thickness measured on a control sample from a known specification in a reference area. The humidity levels show a normal variation with current controls ±10% range of a baseline of 50% RH, and the resulting insulator thickness are example measurements. For example, in experiment 1, with a humidity level of 60 RH, the final insulator thickness measured was 1.05 microns, indicating an increase in thickness correlated with higher humidity levels. Experiment 3, conducted at a lower humidity level of 44 RH, resulted in a thinner insulator layer of 0.97 microns. These variations illustrate the sensitivity of the final insulator thickness measurements to changes in environmental humidity, illustrating the importance of maintaining controlled humidity levels for precise and accurate measurements in the microLED fabrication process. The second example illustrates the results of 10 experiments aimed at measuring insulator thickness in microLED processes utilizing the inspection algorithm and enhanced controller to minimize variations in relative humidity and the related insulator thickness variations. Each row represents an individual experiment, detailing the experiment ID, the more consistently controlled humidity level through the inspection algorithm, and the final insulator thickness measured on a control sample. For example, experiments 1 and 2, with humidity levels precisely controlled at 50 RH, resulted in a consistent insulator thickness of 1.000 microns, demonstrating the effectiveness of the inspection algorithm in maintaining target environmental conditions. Experiments 3 and 5, where the humidity slightly increased to 52 RH, saw a slight increase in insulator thickness to 1.005 microns. Experiments 6, 7, and 8, with a reduced humidity level of 48 RH, showed a minor decrease in thickness to 0.995 microns. The second example illustrates the inspection algorithm’s ability to reduce variability in environmental conditions, leading to more consistent and predictable insulator thickness measurements in microLED fabrication processes. In some embodiments, the FM inspection database 120 may contain parameters, optimized parameters, parameter adjustments from the inspection algorithm, etc.

[0091]

[0091] FIG. 9 illustrates the FM process database 122. The FM process database 122 provides an example of the results of the control algorithm performed in the FM process module 112. The FM process database 122 may contain the experiment ID, the relative humidity, the final insulator thickness, the optimized speed of the measurement, etc. The first example illustrates the speed of measurement in seconds, which may be based on the variability in RH from the baseline of 50%. The deviation from the baseline shows measurement time changing, starting from a setpoint speed of 5 seconds under optimal conditions to the actual time it takes to get the measurement through the optical measurement tool. In the context of optical measurement devices, settling refers to the process by which the instrument stabilizes and reaches a consistent state after being exposed to changes in its environment or operating conditions. When a measurement device settles, it has adjusted to any fluctuations in temperature, humidity, or other environmental factors and is ready to provide accurate and reliable measurements. The settling time of an optical instrument is the duration it takes for the device to achieve this stable state. During this time, the instrument may undergo internal automatic adjustments or calibration procedures to compensate for changes in its surroundings. Settling time is crucial for ensuring the accuracy and repeatability of measurements. For example, in optical thickness measurements of insulator films for microLED fabrication, the optical tool may require a settling period to adapt to variations in humidity levels. This settling time allows the instrument to compensate for any changes in the optical properties of the insulator material due to moisture absorption, ensuring that subsequent measurements are performed under consistent conditions. The first example illustrates that as the humidity levels deviate further from the optimal 50% RH, the time required for measurements increases, reflecting the additional complexity in ensuring accurate film thickness readings. For example, experiment 1, with a humidity level of 60% RH, displays a higher measurement time of 7 seconds, which indicates the potential need for additional calibration or adjustments due to the higher humidity. Measurements closer to the optimal humidity condition, such as experiments 4, 8, and 10, are quicker, at 5.4 seconds, demonstrating the efficiency of maintaining controlled environmental conditions during the film thickness measurement process. Focusing solely on optical film thickness measurements for insulators in the context of microLED fabrication, the justification for the speed of measurements being affected by relative humidity (RH) changes is nuanced. Optical measurement techniques, such as spectroscopic reflectometry or ellipsometry, primarily depend on the interaction of light with the material's surface to gauge thickness. These methods are inherently fast, but the overall process efficiency and speed may be indirectly influenced by RH through hygroscopic effects, condensation on the surface of the insulator, calibration due to environmental factors, requiring data analysis techniques to isolate and remove the effects of the environmental conditions, etc. In the second example, the columns are labeled to indicate the Al adjustments by the control algorithm to humidity measurements and the resulting enhancement in the speed of measurements for insulator thickness in microLED fabrication. In this example, the control algorithm may be a linear regression model or a decision tree algorithm to control humidity levels and optimize measurement speed. In some embodiments, the FM process database 122 may store other parameter data collected or calculated from the FM process module 112, such as humidity, temperature, surface roughness, film uniformity, film composition, etc.

[0092]

[0092] FIG. 10 illustrates the FM post-process database 124. The FM post-process database 124 provides an example of the results of the post-process algorithm performed in the FM post-process module 114. The example data illustrates the lot number or ID, the initial moisture percentage, temperature, time, final moisture percentage, moisture reduction percentage, etc. The first example illustrates that rows 1 and 2 have speeds of 7.0 and 6.6 seconds, respectively, which exceed the specified threshold of 6.0 seconds. Therefore, these rows would require a res-can or retake to ensure that the insulator thickness measurements meet the required specifications for speed. Rows 3 to 10 have speeds within the acceptable range and would not require a re-scan based on the specified threshold. The second example illustrates improved RH control by the post-process algorithm, and the film thickness measurements are more consistent, and only one row, experiment ID 11, would require a re-scan. The post-process algorithm improved humidity control leads to more reliable and precise thickness measurements and reduces the need for rescanning or retaking measurements. Therefore, most of the rows do not require a re-scan, indicating the effectiveness of the post-process algorithm in improving measurement accuracy and ensuring that film thickness remains within specification limits. In the second example, the post-process algorithm controls humidity levels by adjusting environmental parameters such as airflow and temperature. For example, in experiments where the humidity level is higher than optimal, such as experiment 11 with 60% RH, the postprocess algorithm may trigger actions to increase airflow or activate dehumidifiers to reduce RH to the desired level, thereby improving measurement speed. In experiments with lower humidity levels, such as experiment 13 with 44% RH, the post-process algorithm may adjust the temperature or introduce moisture into the air to raise RH to the optimal range. These proactive adjustments ensure that the insulator thickness measurements can be performed more rapidly and consistently, which contributes to increased productivity and efficiency in microLED fabrication processes. In some embodiments, the FM post-process database 124 may store other parameter data collected or calculated from the FM post-process module 114, such as measurement accuracy, comparison against specifications, deviations from expected values, statistical analysis of measurement data, etc.

[0093]

[0093] The functions performed in the processes and methods may be implemented in differing orders. Furthermore, the outlined steps and operations are only provided as examples, and some of the steps and operations may be optional, combined into fewer steps and operations, or expanded into additional steps and operations without detracting from the essence of the disclosed embodiments.

Claims

CLAIMSWhat is claimed is:

1. An automated film measurement (AFM) method for microLED chips system, comprising: an automated film measurement inspection module that measures at least one parameter from a first group of parameters and calculates an output of at least one parameter from a second group of parameters to help control the automated film measurement processing module; an automated film measurement processing module which inputs the at least one parameter from said second group to control the maskless aligner using at least one parameter from a third group of parameters; an automated film measurement post-processing module which measures at least one parameter from a fourth group of parameters that determines how well the parameters of said first and second group were running; an automated film measurement historical module which calculates a plurality of group parameters based upon parameters of the first group, second group, third group, and fourth group; and an automated film measurement integration module to update the automated maskless aligner processing module based upon the results of the automated maskless aligner historical module.

2. The method of claim 1, wherein the AFM process module acts as the core controller, utilizing inputs from the AFM inspection module to govern the AFM process to ensure optimal and accurate speed of the measurements, epitaxial layer measurements, including thickness and uniformity, optical properties, including refractive index, absorption coefficient, transmission / reflection spectra, electrical properties, including resistivity, conductivity, carrier mobility, mechanical properties, including film stress, strain, elasticity, chemical properties, including surface composition, bonding states and molecular structure.

3. The method of claim 1, wherein the AFM post-process module conducts inspections to assess the execution of parameters and identify potential deviations, and the AFM historical module analyzes data collected across various stages, allowing for continuous learning and patternrecognition and subsequently an AFM integration module then employs historical insights to update and optimize the process, enabling adaptive improvements over time.

4. The method of claim 1, wherein a communication interface, which is a hardware or a software component that enables communication between an AFM Al network, an AFM pre-process, an AFM process, and an AFM post-process.

5. The method of claim 4, wherein the communication interface comprises a set of protocols, rules, and standards that define how information is transmitted and received between devices is a physical connector, wireless network, or software application and include components such as drivers, software libraries, and firmware that are used to control and manage the communication process.

6. The method of claim 4, wherein the communication interface is compatible with a USB, aBluetooth, or a Wi-Fi and the communication interface communicates with a network.

7. The method of claim 1, wherein there is a memory that comprises suitable logic, circuitry, and / or interfaces that are configured to store a machine code and / or a computer program with at least one code section executable by a processor.

8. The method of claim 1, wherein there is a base module, which initiates the AFM inspection module, the AFM process module, the AFM post-process module, the AFM historical module, and the AFM integration module.

9. The method of claim 8, wherein the AFM inspection module begins by being initiated by the base module wherein the AFM inspection module connects to the AFM pre-process wherein the AFM inspection module collects data from the AFM pre-process.

10. The method of claim 8, wherein the AFM inspection module performs an inspection algorithm, and the AFM inspection module sends the determined process data from the inspection algorithmto the AFM process module wherein the AFM inspection module stores the data in the AFM inspection database and AFM inspection module returns to the base module.

11. The method of claim 8, wherein the AFM process module begins by being initiated by the base module wherein the AFM process module receives the process data, such as the data on the microLED processed wafer, from the AFM inspection module and the AFM process module connects to the AFM process.

12. The method of claim 11, wherein the AFM process module collects the data from the AFM process and the AFM process module performs a control algorithm and the AFM process module adjusts the control parameters of the AFM process and executes the AFM process wherein AFM process module stores the data in the AFM process database and the AFM process module returns to the base module.

13. The method of claim 8, wherein the AFM post-process module begins by being initiated by the base module and the AFM post-process module connects to the AFM post-process wherein the AFM post-process module collects the post-process data.

14. The method of claim 13, wherein the AFM post-process module extracts the data from the AFM inspection database and AFM process database and performs the post-process algorithm subsequently the AFM post-process module performs the post-process inspection wherein the AFM post-process module stores the data in the AFM post-process database and the AFM postprocess module returns to the base module.

15. The method of claim 8, wherein the AFM historical module begins by being initiated by the base module and the AFM historical module connects to the AFM pre-process, the AFM process, and the AFM post-process wherein the AFM historical module aggregates data from the various types of processes and stores the data in the AFM network database.

16. The method of claim 15, wherein the AFM historical module performs a historical machine learning algorithm on the historical data stored in the AFM network database and sends the processadjustments to the AFM inspection module, the AFM process module, and the AFM post-process module and subsequently the AFM historical module returns to the base module.

17. The method of claim 8, wherein the AFM integration module begins by being initiated by the base module and the AFM integration module performs the integration machine learning algorithm and connects to the AFM pre-process Al module, the AFM process Al module, and the AFM post-process Al module.

18. The method of claim 17, wherein the AFM integration module sends the process adjustments to the AFM pre-process Al module , the AFM process Al module , and the AFM post-process Al module and the AFM integration module returns to the base module.

19. An automated film measurement of microLED substrates system, comprising: an automated film measurement inspection module; an automated film measurement process module; an automated film measurement post-process module; an automated film measurement historical module; and an automated film measurement integration module; wherein, an automated film measurement inspection module is used to determine the quality of incoming processing material and, an automated film measurement process module is used to process the processing material, an automated film measurement post-process module is used to inspect the resultant process material for acceptability, and an automated film measurement historical module is used to analyze the post process results of the most recent process, against all historical data to find enhanced process changes in the automated film measurement process module and an automated film measurement integration module to update the automated film measurement process module based upon the results of the automated film measurement historical module.

Images