Gas quenching technique for controllable crystal growth mechanisms

Abstract

A tool for producing films comprising a chuck configured to move a substrate, a slot die head configured to deposit one or more liquids onto the substrate, an air knife coupled to a gas source with a supply tube, the air knife configured to blow gas onto the substrate, a gantry configured to position the air knife, one or more controllers configured to collect one or more air knife data, one or more in-situ configured to determine an optimal position of the air knife, and a processor configured to receive position data from the in-situ monitor, receive one or more air knife data from the one or more controllers, analyze the received data with one or more machine learning / artificial intelligence (ML / AI) algorithms to predict a coating condition, compare the coating condition with a desired coating condition, and adjust one or more parameters of the air knife to meet the desired coating condition.

Description

GAS QUENCHING TECHNIQUE FOR CONTROLLABLE CRYSTAL GROWTHMECHANISMSCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of and priority to U.S. Provisional Patent Application No. 63 / 621456, filed January 16, 2024, the contents of which are incorporated by reference herein in their entirety.STATEMENT OF GOVERNMENT LICENSE RIGHTS

[0002] This invention was made with Government support under Grant Nos. 2294283 and 9528 awarded by U.S. Department of Energy. The Government has certain rights in the invention.BACKGROUND

[0003] Slot dies allow for the manufacturing of high-performance energy harvesting and storage films. However, such tools often lack desired accuracy, and do not include mechanisms for monitoring the slot die coating as it occurs, for the purpose of adjusting and improving the process. Conventional air-assisted crystallization mechanisms are limited in their modularity and capability for feedback control. Generally, only one parameter is controlled at a given time (position, outlet velocity, etc.) and requires ex-situ confirmation of the desired results. This can result in less-than-optimal films and long process development and optimization cycles.

[0004] Further, there are two great challenges in the commercialization of perovskite solar cells (PSCs): device stability and upscaling. While there has been much work done to improve the stability of perovskite devices using passivation layers and additives, the most stable perovskites to date are still limited by relatively small active areas (<1 cm2). Formulating scalable, high-performance devices require meticulous cooptimization of many parameters including precursor composition and the many inputs of various physical processing techniques (coating speed, annealing temperature, humidity, etc.). The vast majority of perovskite optimization has largely focused on material design, using regression or classification techniques to predict chemical properties of a given composition. While this has been useful at homing in on device stability and power conversion efficiency potential, optimizing PSC processing to replicate these results atlarger scales is still greatly unexplored as these scalable manufacturing techniques (slot die coating, rapid spray plasma pyrolysis, blade coating, etc.) are still largely immature.

[0005] Accordingly, tools, systems, and methods for improved in-line film manufacturing, quality control, and accuracy are needed.SUMMARY

[0006] In one aspect, disclosed herein is a tool for producing films including a chuck configured to move a substrate in a first direction, a slot die head configured to deposit one or more liquids onto the substrate, an air knife coupled to a gas source with a supply tube, the air knife configured to blow gas onto the substrate, a gantry configured to position the air knife, one or more controllers configured to collect one or more air knife data, one or more in-situ monitors configured to determine an optimal position of the air knife, and a processor configured to receive position data from the in-situ monitor, receive one or more air knife data from the one or more controllers, analyze the position data and the one or more air knife data with one or more machine learning / artificial intelligence (ML / Al) algorithms to predict a coating condition, compare the coating condition with a desired coating condition, and direct the gantry, the one or more controllers, or both to adjust one or more parameters of the air knife to meet the desired coating condition.

[0007] In some embodiments, the one or more controllers includes a temperature controller configured to adjust a temperature of gas as it flows through the supply tube. In some embodiments, the one or more controllers includes a pressure regulator configured to adjust a pressure of the gas as it flows through the supply tube.

[0008] In some embodiments, the one or more parameters of the air knife include a pressure of the gas, a mass flow rate of the gas, a temperature of the gas, the composition of the gas, the position of the air knife, an angle of gas flow, or a combination thereof.

[0009] In some embodiments, the tool further comprises an in-situ optical sensor configured to monitor the slot die head, wherein the optical sensor transmits one or more meniscus data to the processor.

[0010] In some embodiments, the processor is further configured to analyze the meniscus data with the one or more ML / Al algorithms, and adjust one or more properties of the slot die head based on analyzed meniscus data.

[0011] In some embodiments, the one or more properties of the slot die head include a flow rate of the one or more liquids, a size or lateral extents shape of a meniscusof the slot die head, an amount of the one or more liquids, an optimal coating speed of a slot die coater, a distance of the slot die head to the substrate, or a combination thereof.

[0012] In some embodiments, the gas is nitrogen gas.

[0013] In some embodiments, the gantry comprises a first bar coupled to the air knife, wherein the first bar is configured to move the air knife in a second direction perpendicular to the first direction, a second bar coupled to the first bar, wherein the second bar is configured to move the air knife vertically, and a first support disposed on a first side of the gantry, and a second support disposed on a second side of the gantry, opposite the first, wherein the first and the second support are configured to move the air knife in a third direction.

[0014] In some embodiments, the one or more ML / Al algorithms comprises a Bayesian algorithm. In some embodiments, the in-situ monitor is configured to measure photoluminescence (PL) of the substrate. In some embodiments, the gas source is controlled independently from the slot die head.

[0015] In some embodiments, the gantry is configured to move the air knife continuously as the slot die head deposits the one or more substances.

[0016] In another aspect, disclosed herein is a method of producing an in-line processed film, including depositing one or more liquid materials with a slot die head onto a substrate, as the substrate moves underneath the slot die head, positioning an air knife to a target area with a gantry, blowing gas onto the substrate, collecting one or more data of the air knifed substrate with one or more in-situ monitors, analyzing the one or more data with one or more machine leaming / artificial intelligence (ML / Al) algorithm to predict a coating condition, comparing the coating condition with a desired coating condition, and adjusting one or more parameters of the air knife based to meet a desired coating condition.

[0017] In some embodiments, the one or more parameters of the air knife include a pressure of the gas, a mass flow rate of the gas, a temperature of the gas, the composition of the gas, a position of the air knife, an angle of gas flow, or a combination thereof.

[0018] In some embodiments, the method further includes collecting one or more meniscus data of the slot die head with an in-situ optical sensor, analyzing the one or more meniscus data with one or more machine learning / artificial intelligence (ML / Al) algorithm to predict the coating condition, and adjusting one or more parameters of the slot die head based on a difference between the coating condition and the desired coating condition.

[0019] In some embodiments, the one or more parameters of the slot die head include a flow rate of the one or more liquids, a size or shape of a meniscus of the slot die head, an amount of the one or more liquids, an optimal coating speed of a slot die coater, a distance of the slot die head to the substrate, or a combination thereof.

[0020] In some embodiments, the in-line processed film is a semiconductor film. In some embodiments, the in-line processed film is a perovskite film. In some embodiments, the in-line processed film is a battery. In some embodiments, the in-line processed film is a battery material such as an energy storage anode, cathode, separator, adhesion layer, interfacial resistance reduction layer or electrolyte layer.

[0021] In some embodiments, the in-situ monitor is configured to measure photoluminescence (PL) of the substrate. In some embodiments, the in-situ monitor is configured to measure time-resolved photoluminescence (TRPL) of the substrate. In some embodiments, the in-situ monitor is configured to measure X-ray fluorescence (XRF) of the substrate. In some embodiments, the in-situ monitor is configured to measure Raman spectroscopy of the substrate. In some embodiments, the in-situ monitor is configured to measure confocal thickness microscopy of the substrate.

[0022] In some embodiments, the one or more ML / Al algorithms comprises a Bayesian algorithm.

[0023] This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This summary is not intended to identify key features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.DESCRIPTION OF THE DRAWINGS

[0024] The foregoing aspects and many of the attendant advantages of this disclosure will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein:

[0025] FIGURE 1 shows an example tool having an independent, motion- controlled air knife that is connected to a nitrogen air supply, in accordance with the present technology;

[0026] FIGURE 2 is an example tool having an air knife gantry and meniscus camera mounted on FOM slot die coater, in accordance with the present technology;

[0027] FIGURES 3A-3B show fully crystalized 100 mm perovskite on glass manufactured using a tool as described herein, in accordance with the present technology; and

[0028] FIGURE 4 is an example method of using a tool, in accordance with the present technology.DETAILED DESCRIPTION

[0029] Described herein is a machine-learning optimized, next-generation scalable energy film coating tool for advanced manufacturing of energy harvesting and storage films that can produce higher performance devices at higher manufacturing throughputs. Combined hardware and software developments work together to produce in-situ feedback and process controls that can produce training data sets and in-situ feedback for machine learning optimization and process control. Producing controlled, high quality energy device films such as perovskites, battery electrodes or electrolyte layers requires careful control of process variables, such as solution (or liquid) flow rates, linear translation speed, drying gas flow rates, treatment periods and a low humidity process environment. In addition to process control, rapid feedback, quicker response to process excursions and reduced process scale-up time, the ability to capture actionable data streams for machine-learning training are useful to optimize coating processes. In one aspect, disclosed herein is an improvement to conventional slot die machines, based on the Alpha coater system (FOM Technologies™, “FOM”) that includes gas flow impingement and optical feedback imaging and sensors and a machine-learning optimized control system. In some embodiments, the tool is configured to achieve an improved manufacturing capability that will increase the adoption of advanced manufacturing techniques in energy devices manufacturing and accelerate the production and adoption of renewable energy generation and storage.

[0030] While the tool may be used to improve gas quenching of semiconductor films, it should be noted that the techniques to be used during this research are model agnostic. In some embodiments, the initial machine learning framework utilizes Bayesian optimization. This technique is able to be tuned to any given target variable and its versatility with sparsely collected data. In the case of perovskite, performance can be measured in-situ using time resolved photoluminescence to predict future coating conditions that maximize this target. However, this approach is still widely applicable tobattery materials, using laser profilometry to monitor edge effects and surface topography of electrodes that are vital to cell assembly further down the processing line.

[0031] In some embodiments, these techniques are used to provide “near real time” data for adjusting parameters of an air knife, a slot die head, or both. As seen in FIGS. 2 and 3 A-3B, collecting meniscus imaging data allows for true real time correction of slot die coating parameters. This method allows for improved accuracy of meniscus imaging that involves digital twinning of images.

[0032] FIG. 1 shows an example tool 100 having an independent, motion- controlled air knife 115 that is connected to a nitrogen air supply, in accordance with the present technology. The tool 100 is capable of feedback control of air pressure, flow rate, and temperature through in-situ monitoring of the drying front accompanied with slot-die feedback control through meniscus imaging. In some embodiments, the rate of solvent evaporation is controlled by the temperature and motion control of the tool. In some embodiments, the pressure and flow regulators (controllers 132 and 133, respectively) aid in optimizing film uniformity and thickness.

[0033] In some embodiments, disclosed herein is a tool 100 for producing films including a chuck 105 configured to move a substrate S in a first direction DI, a slot die head 110 configured to deposit one or more liquids onto the substrate, an air knife 115 coupled to a gas source with a supply tube 120, the air knife 115 configured to blow gas onto the substrate (as shown by the dashed vertical arrows), a gantry 125 configured to position the air knife 115, one or more controllers 130 configured to collect one or more air knife data, an in-situ monitor 135 configured to determine an optimal position of the air knife 115, and a processor 140 configured to receive position data from the in-situ monitor, receive one or more air knife data from the one or more controllers 130 analyze the position data and the one or more air knife data with one or more machine leaming / artificial intelligence (ML / AI) algorithms to predict a coating condition, compare the coating condition with a desired coating condition, and direct the gantry 125, the one or more controllers 130 or both to adjust one or more parameters of the air knife 115 to meet the desired coating condition.

[0034] In some embodiments, the one or more controllers 130 includes a temperature controller 131 configured to adjust a temperature of gas as it flows through the supply tube. In some embodiments, the one or more controllers 130 include a mass flow controller 132, configured to adjust mass flow rate of the gas as it flows through the supplytube 120. In some embodiments, the one or more controllers 130 includes a pressure regulator 133 configured to adjust a pressure of the gas as it flows through the supply tube 120.

[0035] In some embodiments, the one or more parameters of the air knife 115 include a pressure of the gas, the mass flow of the gas, the composition of the gas, a temperature of the gas, a position of the air knife, an angle of the air flow, or a combination thereof.

[0036] In some embodiments, the tool further comprises an in-situ optical sensor 145 configured to monitor the slot die head 110, wherein the optical sensor 145 transmits one or more meniscus data to the processor 140. In some embodiments, in-situ monitoring occurs in two places during the slot-die coating: before and after the substrate has been contacted by the air knife 115. In some embodiments, both wet and dried substrate may be optimized by adjusting the position of the air knife 115, and / or slot die data. Both wet and dried substrate may be affected by the air knife 115 stream and slot die coating conditions. Because of this, the tool 100 may include a meniscus camera (optical sensor 145) and an in-situ monitor (in-situ monitor 135). It should be understood that while the in-situ monitor 135 is explicitly called “in-situ,” other monitors and / or sensors may also be considered “in- situ.” This includes, but is not limited to optical sensor 145, mass flow controller 135, and pressure regulator 133.

[0037] In some embodiments, the processor 140 is further configured to analyze the meniscus data with the one or more ML / Al algorithms and adjust one or more properties of the slot die head based on analyzed meniscus data. In some embodiments, the one or more properties of the slot die head 110 include a flow rate of the liquid, a size or shape of a meniscus of the slot die head, an amount of liquid, an optimal coating speed of the slot die coater, a distance of the slot die head to the substrate, or a combination thereof. In some embodiments, the processor 140 is coupled with the tool 100. In some embodiments, the processor 140 may be separate from the tool 100, such as on a remote computer or device.

[0038] In some embodiments, the gas is nitrogen gas. In some embodiments, the gantry 125 includes a first bar 126 coupled to the air knife 115. The first bar 126 may be configured to move the air knife 115 in a second direction D2 perpendicular to the first direction DI. As shown in FIG. 1, the second direction D2 may be one of two directions along an axis perpendicular to the first direction DI . In some embodiments, the gantry 125 further includes a second bar 127 coupled to the first bar 126, where the second bar 127 isconfigured to move the air knife 115 vertically V. In some embodiments, the gantry 125 further includes a first support 128 disposed on a first side of the gantry 125, and a second support 129 disposed on a second side of the gantry 125, opposite the first side. In operation, the first support 128 and the second support 129 are configured to move the air knife 115 in a third direction D3. In some embodiments, the third direction D3 is movement along a same axis as the first direction DI. In some embodiments, the third direction D3 is the same as the first direction DI. In some embodiments, the third direction D3 is a direction opposite the first direction DI. One skilled in the art will recognize that the first support 128 and the second support 129 may move the knife in either direction.

[0039] In some embodiments, the gas source is controlled independently from the slot die head 110. In some embodiments, the gantry 125 is configured to move the air knife 115 continuously as the slot die head 110 deposits the one or more substances.

[0040] In some embodiments, the one or more ML / Al algorithms comprises a Bayesian algorithm. As used here, a Bayesian algorithm in machine learning is a statistical approach that uses Bayes’ Theorem to calculate the probability of an event based on prior knowledge and new evidence, allowing models to update their beliefs as they encounter new data, making it particularly useful in situations with limited data or when incorporating prior information is valuable. In some embodiments, the Bayesian algorithm is the “Naive Bayes” classifier, which is often used for classification tasks by assuming independence between features within a class. The core principle behind Bayesian algorithms is Bayes’ Theorem, which allows you to calculate the posterior probability (probability of a hypothesis given observed data) based on the likelihood (probability of observing the data given the hypothesis), prior probability (initial belief about the hypothesis), and marginal likelihood (normalizing constant). An advantage of Bayesian algorithms is the ability to incorporate prior knowledge about a problem into the model, which can be helpful when dealing with limited data. Unlike some other machine learning algorithms that provide deterministic predictions, Bayesian models output probability distributions, which can provide valuable insight into the uncertainty of the prediction. In some embodiments, a classification algorithm that is not a Bayesian algorithm, such as a decision tree, logistic regression algorithm, support vector machine, random forest, K-Nearest Neighbors (KNN), an artificial neural network (ANN), or gradient boosting algorithm is used.

[0041] In some embodiments, the in-situ monitor 135 is configured to measure time-resolved photoluminescence (TRPL) of the substrate S. In some embodiments, thein-situ monitor 135 may measure photoluminescence (PL) of the substrate S. In some embodiments, the in-situ monitor 135 is configured to measure X-ray fluorescence (XRF) of the substrate S. In some embodiments, the in-situ monitor 135 is configured to measure Raman spectroscopy of the substrate S. In some embodiments, the in-situ monitor 135 is configured to measure confocal thickness microscopy of the substrate S.

[0042] In some embodiments, the purpose of the tool 100 disclosed herein is to redesign the gas quenching process of aqueously processed thin film materials, in order to achieve highly tunable and reproducible crystalline structures of large area semiconductors. This level of control is highly desirable, as the rate and temperature at which thin films are annealed has a direct impact on their physio-electronic performance as electron carriers. Although aspects of this disclosure focus on a case study in slot-die coated perovskites, this technology may be incorporated into multiple technologies into a single tool capable of being mounted on any in-line processing unit. In some embodiments, the tool 100 decouples the gas quencher from the coating process. The described tool 100 is configured to dry and coat at different rates. In conventional systems, the drying rate is fixed at whatever the coating speed is. The gantry also allows for continuous motion during the coating process (if desired).

[0043] FIG. 2 is another example tool 100 having an air knife gantry 125 and meniscus camera 135 mounted on FOM slot die coater 110, in accordance with the present technology. In some embodiments, the processor 140 is coupled to the tool 140. In some embodiments, the processor 140 includes a screen or monitor for providing and / or analyzing real-time data from the tool 100.

[0044] FIGS. 3A-3B show fully crystalized 100 mm perovskite on glass manufactured using a tool as described herein, in accordance with the present technology. As shown in FIGS. 3A-3B, the perovskite has been fully crystallized.

[0045] In another aspect, disclosed herein is a method of producing an in-line processed film, including providing a substrate S, depositing one or more liquid materials with a slot die head 110 onto the substrate S, as the substrate S moves underneath the slot die head 110, positioning an air knife 115 to a target area with a gantry 125, blowing compressed air onto the substrate S, collecting one or more data of the air knife 115 with one or more in-situ monitor, analyzing the one or more data with one or more machine learning / artificial intelligence (ML / Al) algorithm to predict a coating condition, comparing the coating condition with a desired coating condition, and adjusting one or moreparameters of the air knife 115 based to meet a desired coating condition, as described in FIG. 4.

[0046] In some embodiments, the one or more parameters of the air knife 115 include a pressure of gas, a temperature of gas, a mass glow rate of gas, the composition of the gas, a position of the air knife 115, an angle of the gas flow, or a combination thereof. In some embodiments, the method further includes collecting one or more meniscus data of the slot die head 110 with an in-situ optical sensor (such as optical sensor 145), analyzing the one or more meniscus data with one or more machine leaming / artificial intelligence (ML / Al) algorithm to predict the coating condition, and adjusting one or more parameters of the slot die head 110 based on a difference between the coating condition and the desired coating condition. In some embodiments, the one or more parameters of the slot die head 110 include a flow rate of the liquid, a size or shape of a meniscus of the slot die head 110, an amount of liquid, an optimal coating speed of the slot die coater, the distance of the slot die head 110 to the substrate S, or a combination thereof

[0047] In some embodiments, the in-line processed film is a semiconductor film. In some embodiments, the in-line processed film is a perovskite film. In some embodiments, the in-line processed film is a battery. In some embodiments, the in-line processed film is a battery material such as an energy storage anode, cathode, separator, adhesion layer, interfacial resistance reduction layer or electrolyte layer. In some embodiments, the in-situ monitor is configured to measure time-resolved photoluminescence (TRPL) of the substrate. In some embodiments, the one or more ML / Al algorithms comprises a Bayesian algorithm. In some embodiments, the in-situ monitor 135 may measure photoluminescence (PL) of the substrate S. In some embodiments, the in-situ monitor 135 is configured to measure X-ray fluorescence (XRF) of the substrate S. In some embodiments, the in-situ monitor 135 is configured to measure Raman spectroscopy of the substrate S. In some embodiments, the in-situ monitor 135 is configured to measure confocal thickness microscopy of the substrate S.

[0048] FIG. 4 is an example method 400 of using the tool described herein, in accordance with the present technology. In some embodiments, the tool (such as tool 100) includes a chuck (such as chuck 105), a slot die head (such as slot die head 110), an air knife (such as air knife 115), a supply tube (such as supply tube 120), a gantry (such as gantry 125), one or more controllers (such as one or more controllers 130), an in-situ monitor (such as in-situ monitor 140), and an optical sensor (such as optical sensor 145).In some embodiments, the gantry includes a first bar (such as first bar 126), a second bar (such as second bar 127), a first support (such as first support 128), and a second support (such as second support 129). In some embodiments, the one or more controllers include a temperature controller (such as temperature controller 131), a mass flow controller (such as mass flow controller 132), and / or a pressure regulator (such as pressure regulator 133).

[0049] In block 405, one or more liquid materials are deposited onto a substrate (such as substrate S) with the slot die head. In some embodiments, the one or more liquid materials are deposited throughout the method 400, not just as a discrete step in block 405.

[0050] In block 410, the air knife is positioned to a target area with the gantry. In some embodiments, the air knife can be positioned with the first bar, the second bar, the first support, and the second support. This allows the air knife to be positioned in three dimensions (i.e., along three axes). In some embodiments, the air knife can be positioned and moved continuously.

[0051] In block 415, gas is blown from the air knife onto the substrate. In some embodiments, the gas is nitrogen gas. In some embodiments, the gas is supplied to the air knife with the supply tube.

[0052] In block 420, one or more air knife data is collected. In some embodiments, one or more slot die head data is also collected. The air knife data may include any number of air knife parameters, such as a pressure of the gas, a composition of the gas, the mass flow of the gas, a temperature of the gas, a position of the air knife, an angle of the gas flow, or a combination thereof. Further, slot die head data may include any number of slot die head parameters, including a flow rate of the liquid, a size of a meniscus of the slot die head, an amount of one or more liquid materials, an optimal coating speed of the slot die coater, the distance of the slot die head to the substrate, or a combination thereof.

[0053] In block 425, the one or more air knife data and / or slot die head data are analyzed with an ML / Al algorithm to predict a coating condition (i.e., a too thick coating, a too thin coating, an uneven coating, a desired coating, etc.). In some embodiments, the ML / Al algorithm is a Bayesian algorithm as described herein. In other embodiments, the ML / Al algorithm may be another classification algorithm as described herein.

[0054] In block 430, the coating condition is compared with a desired coating condition. In some embodiments, the desired coating condition is a desired thickness ofcoating, a desired consistency or evenness of coating, or the like. In some embodiments, the desired coating condition is predetermined by the application of the substrate.

[0055] In block 435, one or more parameters of the air knife and / or the slot die head are adjusted to meet the desired coating condition. One skilled in the art will recognize that adjusting the one or more parameters of the air knife and / or slot die head includes maintaining a same parameter of the air knife and / or slot die head to continue meeting the desired coating condition. In some embodiments, the air knife may be positioned to a new location by the gantry to meet the desired coating condition, the gas flow rate may be increased or decreased, the amount of liquid material dispensed may be increased or reduced, or the like.

[0056] The tool may continuously monitor the slot die head and / or the air knife and adjust the parameters of the air knife and / or the slot die head in real-time. That is, the tool may compare the coating condition of both wet and dried coating and adjust the air knife and / or the slot die head accordingly. This results in a feedback loop, as shown by the dashed arrow.EXAMPLES

[0057] 1. A tool for producing films comprising a chuck configured to move a substrate in a first direction, a slot die head configured to deposit one or more liquids onto the substrate, an air knife coupled to a gas source with a supply tube, the air knife configured to blow gas onto the substrate, a gantry configured to position the air knife, one or more controllers configured to collect one or more air knife data, one or more in-situ monitors configured to determine an optimal position of the air knife, and a processor configured to receive position data from the in-situ monitor, receive one or more air knife data from the one or more controllers, analyze the position data and the one or more air knife data with one or more machine learning / artificial intelligence (ML / Al) algorithms to predict a coating condition, compare the coating condition with a desired coating condition, and direct the gantry, the one or more controllers, or both to adjust one or more parameters of the air knife to meet the desired coating condition.

[0058] 2. The tool of Example 1, wherein the one or more controllers includes a temperature controller configured to adjust a temperature of gas as it flows through the supply tube.

[0059] 3. The tool of Example 1 or Example 2, wherein the one or more controllers includes a pressure regulator configured to adjust a pressure of the gas as it flows through the supply tube.

[0060] 4. The tool of any one of Examples 1-3, wherein the one or more parameters of the air knife include a pressure of the gas, the composition of the gas, a mass flow rate of the gas, a temperature of the gas, the position of the air knife, an angle of air flow, or a combination thereof.

[0061] 5. The tool of any one of Examples 1-4, wherein the tool further comprises an in-situ optical sensor configured to monitor the slot die head, wherein the optical sensor transmits one or more meniscus data to the processor.

[0062] 6. The tool of any one of Examples 1-5, wherein the processor is further configured to analyze the meniscus data with the one or more ML / Al algorithms, and adjust one or more properties of the slot die head based on analyzed meniscus data.

[0063] 7. The tool of Example 6, wherein the one or more properties of the slot die head include a flow rate of the one or more liquids, a size or shape of a meniscus of the slot die head, an amount of the one or more liquids, an optimal coating speed of a slot die coater, a distance of the slot die head to the substrate, or a combination thereof.

[0064] 8. The tool of any one of Examples 1-7, wherein the gas is nitrogen gas.

[0065] 9. The tool of any one of Examples 1-8, wherein the gantry comprises a first bar coupled to the air knife, wherein the first bar is configured to move the air knife in a second direction perpendicular to the first direction, a second bar coupled to the first bar, wherein the second bar is configured to move the air knife vertically, and a first support disposed on a first side of the gantry, and a second support disposed on a second side of the gantry, opposite the first, wherein the first and the second support are configured to move the air knife in a third direction, opposite the first direction.

[0066] 10. The tool of any one of Examples 1-9, wherein the one or more ML / Al algorithms comprises a Bayesian algorithm.

[0067] 11. The tool of any one or Examples 1-10, wherein the in-situ monitor is configured to measure time-resolved photoluminescence (TRPL) of the substrate.

[0068] 12. The tool of any one or Examples 1-11, wherein the in-situ monitor is configured to measure photoluminescence (PL) of the substrate.

[0069] 13. The tool of any one or Examples 1-12, wherein the in-situ monitor is configured to measure X-ray fluorescence (XRF) of the substrate.

[0070] 14. The tool of any one or Examples 1-13, wherein the in-situ monitor is configured to measure Raman spectroscopy of the substrate.

[0071] 15. The tool of any one or Examples 1-14, wherein the in-situ monitor is configured to measure confocal thickness microscopy of the substrate.

[0072] 16. The tool of any one of Examples 1-15, wherein the gas source is controlled independently from the slot die head.

[0073] 17. The tool of any one of Examples 1-16, wherein the gantry is configured to move the air knife continuously as the slot die head deposits the one or more substances.

[0074] 18. A method of producing an in-line processed film, comprising depositing one or more liquid materials with a slot die head onto a substrate, as the substrate moves underneath the slot die head, positioning an air knife to a target area with a gantry, blowing gas onto the substrate, collecting one or more data of the air knife with one or more in-situ monitors, analyzing the one or more data with one or more machine learning / artificial intelligence (ML / Al) algorithm to predict a coating condition, comparing the coating condition with a desired coating condition, and adjusting one or more parameters of the air knife based to meet a desired coating condition.

[0075] 19. The method of Example 18, wherein the one or more parameters of the air knife include a pressure of the gas, a mass flow rate of the gas, a composition of the gas a temperature of the gas, a position of the air knife, an angle of gas flow, or a combination thereof.

[0076] 20. The method of Example 18 or Example 19, wherein the method further comprises collecting one or more meniscus data of the slot die head with an in-situ optical sensor, analyzing the one or more meniscus data with one or more machine learning / artificial intelligence (ML / AI) algorithm to predict the coating condition, and adjusting one or more parameters of the slot die head based on a difference between the coating condition and the desired coating condition.

[0077] 21. The method of any one of Examples 18-20, wherein the one or more parameters of the slot die head include a flow rate of the one or more liquids, a size or shape of a meniscus of the slot die head, an amount of the one or more liquids, an optimal coating speed of a slot die coater, a distance of the slot die head to the substrate, or a combination thereof.

[0078] 22. The method of any one of Examples 18-21 wherein the in-line processed film is a semiconductor film.

[0079] 23. The method of any one of Examples 18-22, wherein the in-line processed film is a perovskite film.

[0080] 24. The method of any one of Examples 18-23, wherein the in-line processed film is a battery material such as an energy storage anode, cathode, separator, adhesion layer, interfacial resistance reduction layer or electrolyte layer.

[0081] 25. The method of any one of Examples 18-24, wherein the in-situ monitor is configured to measure (PL) of the substrate.

[0082] 26. The method of any one of Examples 18-25, wherein the in-situ monitor is configured to measure time-resolved photoluminescence (TRPL) of the substrate.

[0083] 27. The tool of any one or Examples 18-26, wherein the in-situ monitor is configured to measure X-ray fluorescence (XRF) of the substrate.

[0084] 28. The tool of any one or Examples 18-27, wherein the in-situ monitor is configured to measure Raman spectroscopy of the substrate.

[0085] 29. The tool of any one or Examples 18-28, wherein the in-situ monitor is configured to measure confocal thickness microscopy of the substrate.

[0086] 30. The method of any one of Examples 18-29, wherein the one or more ML / Al algorithms comprises a Bayesian algorithm.

[0087] While illustrative embodiments have been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the disclosure.

Claims

CLAIMSWe claim:

1. A tool for producing films comprising: a chuck configured to move a substrate in a first direction; a slot die head configured to deposit one or more liquids onto the substrate; an air knife coupled to a gas source with a supply tube, the air knife configured to blow gas onto the substrate; a gantry configured to position the air knife; one or more controllers configured to collect one or more air knife data; one or more in-situ monitors configured to determine an optimal position of the air knife; and a processor configured to: receive position data from the in-situ monitor, receive one or more air knife data from the one or more controllers, analyze the position data and the one or more air knife data with one or more machine learning / artificial intelligence (ML / Al) algorithms to predict a coating condition, compare the coating condition with a desired coating condition, and direct the gantry, the one or more controllers, or both to adjust one or more parameters of the air knife to meet the desired coating condition.

2. The tool of Claim 1, wherein the one or more controllers includes a temperature controller configured to adjust a temperature of gas as it flows through the supply tube.

3. The tool of Claim 1 , wherein the one or more controllers includes a pressure regulator configured to adjust a pressure of the gas as it flows through the supply tube.

4. The tool of Claim 1, wherein the one or more parameters of the air knife include a pressure of the gas, a composition of the gas, a mass flow rate of the gas, a temperature of the gas, the position of the air knife, an angle of gas flow, or a combination thereof.

5. The tool of Claim 1, wherein the one or more in-situ monitors comprises an in-situ optical sensor configured to monitor the slot die head, wherein the optical sensor transmits one or more meniscus data to the processor.

6. The tool of Claim 1, wherein the processor is further configured to: analyze the meniscus data with the one or more ML / Al algorithms; and adjust one or more properties of the slot die head based on analyzed meniscus data.

7. The tool of Claim 6, wherein the one or more properties of the slot die head include a flow rate of the one or more liquids, a size or shape of a meniscus of the slot die head, an amount of the one or more liquids, an optimal coating speed of a slot die coater, a distance of the slot die head to the substrate, or a combination thereof.

8. The tool of Claim 1, wherein the gas is nitrogen gas.

9. The tool of Claim 1, wherein the gantry comprises: a first bar coupled to the air knife, wherein the first bar is configured to move the air knife in a second direction perpendicular to the first direction; a second bar coupled to the first bar, wherein the second bar is configured to move the air knife vertically; and a first support disposed on a first side of the gantry, and a second support disposed on a second side of the gantry, opposite the first, wherein the first and the second support are configured to move the air knife in a third direction, opposite the first direction.

10. The tool of Claim 1, wherein the one or more ML / Al algorithms comprises a Bayesian algorithm.

11. The tool of Claim 1 , wherein the one or more in-situ monitors are configured to measure photoluminescence (PL) of the substrate.

12. The tool of Claim 1, wherein the one or more in-situ monitors are configured to measure X-ray fluorescence (XRF) of the substrate.

13. The tool of Claim 1, wherein the one or more in-situ monitors is configured to measure Raman spectroscopy of the substrate.

14. The tool of Claim 1, wherein the one or more in-situ monitors is configured to measure confocal thickness microscopy of the substrate.

15. The tool of Claim 1, wherein the gas source is controlled independently from the slot die head.

16. The tool of Claim 1, wherein the gantry is configured to move the air knife continuously as the slot die head deposits the one or more substances.

17. A method of producing an in-line processed film with the tool of Claim 1, comprising: depositing one or more liquid materials with a slot die head onto a substrate, as the substrate moves underneath the slot die head; positioning an air knife to a target area with a gantry; blowing gas onto the substrate; collecting one or more data of the air knife with one or more in-situ monitors; analyzing the one or more data with one or more machine learning / artificial intelligence (ML / Al) algorithm to predict a coating condition; comparing the coating condition with a desired coating condition; and adjusting one or more parameters of the air knife based to meet a desired coating condition.

18. The method of Claim 17, wherein the one or more parameters of the air knife include a pressure of the gas, a composition of the gas a mass flow rate of the gas, a temperature of the gas, a position of the air knife, an angle of gas flow, or a combination thereof.

19. The method of Claim 17, wherein the method further comprises: collecting one or more meniscus data of the slot die head with an in-situ optical sensor; analyzing the one or more meniscus data with one or more machine learning / artificial intelligence (ML / Al) algorithm to predict the coating condition; and adjusting one or more parameters of the slot die head based on a difference between the coating condition and the desired coating condition.

20. The method of Claim 17, wherein the one or more parameters of the slot die head include a flow rate of the one or more liquids, a size or shape of a meniscus of the slot die head, an amount of the one or more liquids, an optimal coating speed of a slot die coater, a distance of the slot die head to the substrate, or a combination thereof.

21. The method of Claim 17, wherein the in-line processed film is a semiconductor film.

22. The method of Claim 17, wherein the in-line processed film is a perovskite film.

23. The method of Claim 17, wherein the in-line processed film is a battery material such as an energy storage anode, cathode, separator, adhesion layer, interfacial resistance reduction layer or electrolyte layer.

24. The method of Claim 17, wherein the one or more in-situ monitors are configured to measure photoluminescence (PL) of the substrate.

25. The method of Claim 17, wherein the one or more in-situ monitors are configured to measure time-resolved photoluminescence (TRPL) of the substrate.

26. The method of Claim 17, wherein the one or more in-situ monitors are configured to measure X-ray fluorescence (XRF) of the substrate.

27. The method of Claim 17, wherein the one or more in-situ monitors is configured to measure Raman spectroscopy of the substrate.

28. The method of Claim 17, wherein the one or more in-situ monitors is configured to measure confocal thickness microscopy of the substrate.

29. The method of Claim 17, wherein the one or more ML / Al algorithms comprises a Bayesian algorithm.

Images