Method and apparatus for lamp housing crack detection

Abstract

Methods and apparatus for lamp housing crack detection are provided herein. For example, a method associated with a processing chamber having a lamp housing includes sequentially processing a plurality of substrates in the processing chamber, collecting lamp housing data indicating fluid leaks in the lamp housing from a sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates, a determination is made as to whether there is a lamp housing crack based on the lamp housing data, and in response to determining that there is a lamp housing crack, at least one of triggering an alarm or stopping processing of the processing chamber.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to methods and apparatus for processing substrates, and, for example, to methods and apparatus for lamp housing crack detection. Background Art

[0002] Substrate processing using radiant heat sources is known. For example, reduced pressure chambers and atmospheric chambers may include a lamp housing that includes one or more lamps configured to heat one or more substrates to a relatively high temperature (e.g., for annealing, deoxidation, etc.). However, after extended use, the lamp housing (and / or one or more lamps) may develop cracks (see below). Figure 1 ). A cracked lamp housing may leak cooling water into the lamp tube and / or onto the quartz window through which light is radiated. As a result, the processing chamber will require relatively long periods of unscheduled downtime and corrective maintenance to replace the lamp housing, clean the leaks, and address additional failures of the lamp and / or related components due to water intrusion.

[0003] Thus, the inventors herein describe improved methods and apparatus for lamp housing crack detection. Summary of the invention

[0004] Methods and apparatus for lamp housing crack detection are provided herein. In some embodiments, a method associated with a processing chamber having a lamp housing includes sequentially processing a plurality of substrates in the processing chamber, collecting lamp housing data indicating fluid leakage in the lamp housing from a sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates, determining whether a lamp housing crack exists based on the lamp housing data, and in response to determining that a lamp housing crack exists, at least one of triggering an alarm or stopping processing in the processing chamber.

[0005] In at least some embodiments, a non-transitory computer-readable storage medium has instructions stored thereon that, when processed and executed, perform a method for processing a substrate using a lamp housing. The method includes sequentially processing a plurality of substrates in a processing chamber, collecting lamp housing data indicating fluid leakage in the lamp housing from a sensor operably connected to the lamp housing of the processing chamber during the processing of the plurality of substrates, determining whether a lamp housing crack exists based on the lamp housing data, and at least one of triggering an alarm or stopping processing in the processing chamber in response to determining that a lamp housing crack exists.

[0006] In at least some embodiments, an apparatus for processing a substrate includes: a processing chamber configured to process the substrate using radiant heat provided by a lamp housing; a sensor operably connected to the lamp housing; and a controller configured to sequentially process a plurality of substrates in the processing chamber, collect lamp housing data indicating fluid leakage in the lamp housing from the sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates, determine whether there is a crack in the lamp housing based on the lamp housing data, and at least one of triggering an alarm or stopping processing in the processing chamber in response to determining that there is a crack in the lamp housing.

[0007] Other and further embodiments of the disclosure are described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Embodiments of the present disclosure, briefly summarized above and discussed in more detail below, may be understood by reference to the illustrative embodiments of the disclosure depicted in the accompanying drawings. However, the drawings illustrate only typical embodiments of the disclosure and are therefore not to be considered limiting of the scope, as the disclosure may admit to other equally effective embodiments.

[0009] Figure 1 is a flow chart of a method associated with a processing chamber having a lamp housing in accordance with at least some embodiments of the present disclosure.

[0010] Figure 2 is configured to perform according to at least some embodiments of the present disclosure Figure 1 tools for the method.

[0011] Figure 3 is configured to communicate with Figure 2 Diagram of the lamp housing used together with the tool.

[0012] For ease of understanding, the same reference numerals are used where possible to represent the same elements common to the drawings. The drawings are not drawn to scale and may be simplified for clarity. Elements and features of one embodiment may be advantageously incorporated into other embodiments without repetition. DETAILED DESCRIPTION

[0013] The methods and apparatus described herein are configured for use with a processing chamber configured for processing a substrate. For example, a processing chamber may include one or more lamp housing sensors that may be configured to detect lamp housing pressure, such as when used in conjunction with a reduced pressure processing chamber. In at least some embodiments, the methods and apparatus described herein use one or more humidity sensors that may be configured to detect relative humidity in the lamp housing, such as when used in conjunction with an atmospheric pressure processing chamber. For example, in any embodiment, data collected from one or more lamp housing sensors or humidity sensors may be used to predict when a crack will appear / form on the lamp housing. For example, for one or more lamp housing sensors, the collected data allows a user to determine a change in lamp housing pressure that may be associated with an early crack. Similarly, for a humidity sensor, the collected data allows a user to determine a change in lamp housing humidity that may be associated with an early crack. Automatic time series modeling using a window based on at least one of a moving average (maximum minus minimum) of a range of collected data or a standard deviation of a range of collected data may also be used to set limits on the collected data to alert a user to early crack detection (e.g., faulty vs. normal). The methods and apparatus described herein can reduce, if not eliminate, the need for unplanned downtime and corrective maintenance to replace lamp housings, clean leaks, and address additional failures of lamp housings (and / or lamp heads) and / or related components due to water intrusion.

[0014] Figure 1 is a flow chart of a method 100 for processing a substrate according to at least some embodiments of the present disclosure, and Figure 2 A tool 200 (or apparatus) may be used to perform the method 100 .

[0015] Method 100 may be performed in tool 200, which may include any suitable processing chamber configured for one or more of physical vapor deposition (PVD), chemical vapor deposition (CVD), and / or atomic layer deposition (ALD), such as plasma enhanced ALD or thermal ALD (e.g., without plasma formation). Examples of processing systems that may be used to perform the inventive methods disclosed herein may include, but are not limited to, one or more processing chambers commercially available from Applied Materials, Inc. of Santa Clara, California. Other processing chambers, including those available from other manufacturers, may also be used as appropriate in conjunction with the teachings provided herein.

[0016] The tool 200 may be embodied in individual processing chambers that may be provided in a standalone configuration or as part of a cluster tool, such as described below with reference to Figure 2An integrated tool (tool 200) is described. Examples of integrated tools may include, but are not limited to, one or more processing chambers commercially available from Applied Materials, Inc. of Santa Clara, California. The methods described herein may be practiced using other cluster tools having suitable processing chambers coupled thereto or in other suitable processing chambers. For example, in some embodiments, the inventive methods discussed above may be performed in an integrated tool such that there are limited or no vacuum breaks between processing steps. For example, reducing vacuum breaks may limit or prevent contamination (e.g., oxidation) of one or more metal layers or other portions of a substrate.

[0017] The integrated tool includes a processing platform 201 (vacuum sealed processing platform), a factory interface 204, and a controller 202. The processing platform 201 includes a plurality of processing chambers, such as 214A, 214B, 214C, and 214D operably coupled to a transfer chamber 203 (vacuum substrate transfer chamber) and processing chambers 214E and 214F operably coupled to a buffer chamber 208 (vacuum substrate buffer chamber).

[0018] The factory interface 204 is connected through one or more load lock chambers (two load lock chambers, such as Figure 2 206A and 206B) are operably coupled to the buffer chamber 208. In at least some embodiments, one of the buffer chamber 208 or the transfer chamber 203 of the tool 200 can be omitted. One or more modules or channels can be disposed between the buffer chamber 208 and the transfer chamber 203 and can be configured to receive one or more substrates from the buffer chamber 208 and / or the transfer chamber 203. In at least some embodiments, the module 218A and the module 218B are disposed between the buffer chamber 208 and the transfer chamber 203 and are configured to receive one or more substrates from the buffer chamber 208 and / or the transfer chamber 203 during operation, as will be described in more detail below. As described above, the module 218A and the module 218B can have a transparent cover.

[0019] In some embodiments, the factory interface 204 includes a docking station 207, a factory interface robot 238 to facilitate the transfer of one or more semiconductor substrates (wafers). The docking station 207 is configured to receive one or more front opening pods (FOUPs). Figure 2In the embodiment of the present invention, four FOUPS are shown, such as 205A, 205B, 205C, and 205D. The factory interface robot 238 is configured to transfer substrates from the factory interface 204 to the processing platform 201 through load lock chambers such as 206A and 206B. Each load lock chamber 206A and 206B has a first port coupled to the factory interface 204 and a second port coupled to the transfer chamber 203. The load lock chambers 206A and 206B are coupled to a pressure control system (not shown) that evacuates and exhausts the load lock chambers 206A and 206B to facilitate the passage of substrates between the vacuum environment of the buffer chamber 208 and the substantially atmospheric environment (e.g., atmosphere) of the factory interface 204. In at least some embodiments, the buffer chamber 208 can be maintained at a substantially atmospheric environment. In embodiments where the buffer chamber 208 is not used, the load lock chambers 206A and 206B facilitate the transfer of substrates between the transfer chamber 203 and the factory interface 204. The buffer chamber 208 and the transfer chamber 203 each have a vacuum robot 242 positioned to transfer / receive one or more substrates. For example, the vacuum robot 242 of the buffer chamber 208 can receive / transfer substrates 221 between the load lock chambers 206A and 206B, the processing chambers 214E and 214F, and the modules 218A and 218F. Similarly, the vacuum robot 242 of the transfer chamber 203 can receive / transfer substrates 221 between the processing chambers 214A, 214B, 214C, and 214D and the modules 218A and 218F.

[0020] In some embodiments, the processing chambers 214A, 214B, 214C, 214D, 214E, and 214F may include at least an ALD chamber, a CVD chamber, a PVD chamber, an electron beam deposition chamber, and / or an electroplating, electroless (EEP) deposition chamber. Likewise, in some embodiments, one or more optional service chambers (shown as 216A and 216B) may be coupled to the buffer chamber 208. The service chambers 216A and 216B may be configured to perform other substrate processing, such as degassing, bonding, chemical mechanical polishing (CMP), annealing, substrate cleaning (e.g., pre-cleaning to remove oxidation), wafer dicing, etching, plasma dicing, orientation, substrate metrology, cooling, etc.

[0021] The controller 202 controls the operation of the tool 200 using direct control of the process chambers 214A, 214B, 214C, 214D, 214E, and 214F and the device 212, or alternatively, by controlling a computer (or controller) associated with the process chambers 214A, 214B, 214C, 214D, 214E, and 214F, the device 212, and the tool 200. In operation, the controller 202 enables data and feedback to be collected from the various chambers and systems to optimize the performance of the tool 200. For example, the controller 202 may receive data from one or more sensors operably coupled to one or more of the process chambers 214A, 214B, 214C, 214D, 214E, and 214F, the service chambers 216A and 216B, and / or the modules 218A and 218F. The controller 202 generally includes a central processing unit 230, a memory 234, and support circuits 232. The central processing unit 230 may be any form of general purpose computer processor that may be used in an industrial environment. The support circuits 232 are conventionally coupled to the central processing unit 230 and may include caches, clock circuits, input / output subsystems, power supplies, etc. Software routines such as the processing methods described above may be stored in the memory 234 (e.g., a non-transitory computer readable storage medium) and, when executed by the central processing unit 230, transform the central processing unit 230 into a special purpose computer (e.g., the controller 202). The software routines may also be stored and / or executed by a second controller (not shown) located remotely from the tool 200.

[0022] Continue to refer Figure 1 , one or more substrates, thermal module assemblies, etc. may be initially loaded into one or more of four FOUPS, such as 205A, 205B, 205C, and 205D ( Figure 2 ). For example, in at least some embodiments, a substrate 221 (wafer) may be loaded into FOUP 205B. The diameter of substrate 221 may be 150 mm, 200 mm, 300 mm, etc. Substrate 221 may be formed of germanium, silicon, silicon carbide, silicon oxide, etc. In at least some embodiments, substrate 221 may have a diameter of 300 mm and may be formed of silicon. In at least some embodiments, one or more metal layers may be deposited on substrate 221. For example, one or more metal layers may include aluminum, cobalt, copper, nitride, titanium, tantalum, etc. In at least some embodiments, substrate 221 may include a metal layer containing cobalt and tungsten.

[0023] Once loaded, the factory interface robot 238 may transfer the substrate 221 from the factory interface 204 to the processing platform 201 through, for example, the load lock chamber 206A. The vacuum robot 242 of the buffer chamber 208 may transfer the substrate 221 from the load lock chamber 206A to one or more of the processing chambers 214A, 214B, 214C, 214D, 214E, and 214F, the service chambers 216A and 216B, and / or the modules 218A and 218B.

[0024] For example, in at least some embodiments, the vacuum robot 242 of the buffer chamber 208 can transfer the substrate 221 from the load lock chamber 206A to the service chambers 216A and 216B and / or modules 218A and 218B, wherein one or more of degassing, bonding, chemical mechanical polishing (CMP), annealing, substrate cleaning (oxidation removal processing), wafer cutting, etching, plasma cutting, orientation, substrate metrology, and / or cooling can be performed.

[0025] For example, as described above, the substrate 221 can be processed in one or more processing chambers that are configured to process the substrate using radiant heat provided by the lamp housing. For example, the service chambers 216A and 216B (e.g., radiation chambers) can be configured to perform an annealing process (annealing chambers) or an oxidation removal process (oxidation chambers), each of which uses relatively high heat. When performing such processes, the service chambers 216A and 216B can be equipped with one or more sensors 300 (e.g., lamp housing sensors and / or relative humidity sensors) that are operably coupled to the lamp housing via one or more suitable coupling devices (e.g., bolts, clamps, etc.). For example, when the service chambers 216A and 216B are configured to be used in a reduced pressure environment (e.g., annealing process), the one or more sensors 300 can be configured to detect lamp housing pressure (e.g., pressure sensors). Similarly, when the service chambers 216A and 216B are configured for use in an atmospheric pressure environment (e.g., an oxidation removal process), the one or more sensors 300 can be configured to detect relative humidity in the lamp housing (e.g., a relative humidity sensor). In either embodiment, the data collected from the one or more sensors 300 can be used to predict when a lamp housing crack will appear / form. The lamp housing crack can cause coolant 304 (e.g., cooled or uncooled water from a cooling water line, not shown) to leak from the lamp housing 302 to one or more lamp modules 306, which are configured to emit / radiate high heat (e.g., as shown by arrow 310) via the lamp window 308 to process the substrate 221. The leaking coolant 304 can damage, if not destroy, one or more lamp modules 306, which can result in unplanned downtime, work hours, and possible failure of other related components.

[0026] Thus, the inventors have discovered that by using empirical data related to pressure and / or relative humidity obtained during one or more annealing processes or oxidation removal processes, respectively, lamp housing cracks can be determined / predicted and damage caused thereby can be reduced, if not eliminated. The empirical data can be obtained over a longer period of time (e.g., one week, two weeks, one month, two months, one year, two years, etc.). For example, with respect to reduced pressure environments, the inventors have discovered that the collected data allows a user to determine changes in lamp housing pressure that may be associated with early lamp housing cracks. Similarly, with respect to atmospheric pressure environments, the inventors have discovered that the collected data allows a user to determine changes in lamp housing humidity that may be associated with early housing cracks. In at least some embodiments, automatic time series modeling using a window based on at least one of a moving average (maximum value minus minimum value) of a range of collected data or a standard deviation of a range of collected data can also be used to set limits on the collected data to alert a user of early crack detection (e.g., faulty vs. normal).

[0027] Thus, at 102, method 100 includes sequentially processing a plurality of substrates in a processing chamber, and at 104, method 100 includes collecting lamp housing data indicating fluid leakage in the lamp housing (which may cause cracks in the lamp housing) from a sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates. For example, automated time series modeling may include one or more of the following: a) collecting data for a plurality of substrate (wafer) runs (annealing and / or oxidation removal processes), b) for each substrate run, calculating a range (e.g., a range may be defined as a maximum sensor value minus a minimum sensor value (e.g., obtained from one or more sensors 300 for the substrate run)), c) calculating a moving average of the range obtained from b), e.g., the moving average may be based on the last X runs, d) calculating a standard deviation of the range obtained from the second step, e.g., the standard deviation may be based on the last X runs, e) calculating a threshold limit by adding the moving average obtained from the third step to the standard deviation obtained from the fourth step, and f) for the next substrate run, calculating a range and comparing the calculated range to the threshold limit obtained from e).

[0028] Next, at 106, the method 100 includes determining whether there is a lamp housing crack based on the lamp housing data. For example, as described above, the substrate 221 can be transferred to one or both of the service chambers 216A and 216B under the control of the controller 202. For example, in at least some embodiments, the substrate 221 can be transferred to the service chamber 216A, where an oxidation removal process can be performed on the substrate 221. Alternatively or additionally, the substrate 221 can be transferred to the service chamber 216B, where an annealing process can be performed on the substrate 221.

[0029] During method 106, method 100 may include comparing a first subset of lamp housing data collected when processing a first substrate of the plurality of substrates (e.g., empirical data collected during 102 and 104 at a) through e) with a second subset of lamp housing data collected when processing a second substrate of the plurality of substrates (e.g., 104 at f). In at least some embodiments, during 106, method 100 may include determining a threshold limit based on the first subset of lamp housing data (e.g., obtained during 104 at e), generating a metric based on the second subset of lamp housing data, and including comparing the metric to the threshold limit (e.g., obtained during 104 at f).

[0030] For example, when performing the oxidation removal process, one or more sensors 300 (e.g., relative humidity sensors) collect relative humidity data of the lamp housing 302 and transmit the relative humidity data to the controller 202. Similarly, when performing the annealing process, one or more sensors 300 (e.g., pressure sensors) collect pressure data of the lamp housing 302 (e.g., pressure changes along the surface of the lamp housing 302) and transmit the lamp housing pressure data to the controller 202.

[0031] For example, the controller 202 may compare the collected lamp housing data with previously stored lamp housing data (e.g., empirical data obtained from annealing processes and / or oxidation removal processes previously performed during a) through e) of 102 and 104). The empirical data may be obtained using time series modeling over a fixed time period of about 1 week to about 2 weeks, or in some embodiments, less than a week or longer than 2 weeks (e.g., 1 day to about 2 years). In at least some embodiments, the time period may not be fixed and continuous time series modeling may be used to collect data. In at least some embodiments, the controller 202 may be configured to perform the comparison at predetermined time intervals (e.g., a predetermined number of seconds, minutes, hours, etc.) or continuously during the process.

[0032] Next, at 108, the method 100 includes at least one of triggering an alarm or stopping processing in the processing chamber in response to determining that a lamp housing crack exists. For example, when performing an oxidation removal process, if the relative humidity of the lamp housing 302 is above a predetermined threshold (e.g., about 50% humidity to about 60% humidity), the controller 202 may trigger an alarm and / or automatically stop processing of the substrate. In at least some embodiments, the controller 202 may trigger an alarm and a user may manually stop processing of the substrate. In at least some embodiments, the controller 202 may also be configured to detect a time frame within which the relative humidity (e.g., for processing of at least one substrate) remains at a predetermined threshold. Similarly, when performing an annealing process, if the pressure of the lamp housing 302 exceeds a predetermined threshold (e.g., about 20 Torr to about 30 Torr), the controller 202 may trigger an alarm and / or stop processing of the substrate. In at least some embodiments, the controller 202 may also be configured to detect a time frame within which the pressure (e.g., for at least one substrate) remains at a predetermined threshold. In at least some embodiments, for a particular substrate, the predetermined thresholds for relative humidity and pressure can be based on at least one of: a moving average (maximum value minus minimum value) of the range of collected data, or a standard deviation of the range of collected data obtained during the corresponding time series modeling within the time period (e.g., a time period of about 1 week to about 2 weeks, or in some embodiments, less than a week or longer than 2 weeks).

[0033] The method 100 including operations 102-108 may also be used in conjunction with the processing chambers 214A, 214B, 214C, 214D, 214E, and 214F if the processing chambers 214A, 214B, 214C, 214D, 214E, and 214F are configured to perform one or more of an annealing process or an oxidation removal process.

[0034] Additionally, in some embodiments, the method 100 may be based on predictions from a machine learning process / artificial intelligence process that may predict when a lamp housing crack may occur based on trends in previous and current pressure / humidity measurements. For example, the collected lamp data may be input to a trained machine learning model (e.g., which may be stored in memory 234) that was previously trained based on old data, and the output of the trained machine learning model may be used to trigger an alarm, as described above. For example, in at least some embodiments, during 106, the method 100 may include extracting one or more features from the time series lamp housing data and providing the one or more features to a trained neural network that is configured to provide an output to indicate whether a lamp housing crack exists. In at least some embodiments, the one or more features may include a maximum minus a minimum range of sensor values ​​obtained during substrate processing, as described above in a) to f) during 102 and 104. For example, the collected data obtained may be input to a trained neural network, and by using a moving average and / or a standard deviation, the trained neural network may determine a predetermined threshold (e.g., pressure / humidity) and trigger an alarm when the predetermined threshold is met.

[0035] While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the disclosure may be devised without departing from the basic scope thereof.

Claims

1. A method associated with a processing chamber having a lamp housing, the method comprising: sequentially processing a plurality of substrates in the processing chamber; collecting lamp housing data indicative of a fluid leak in the lamp housing from a sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates; Determining whether there is a crack in the lamp housing according to the lamp housing data; as well as In response to determining that the lamp housing crack exists, at least one of triggering an alarm or stopping processing in the processing chamber is performed.

2. The method of claim 1, wherein determining whether the lamp housing has cracks comprises: A first subset of the lamp housing data collected while processing a first substrate of the plurality of substrates is compared to a second subset of the lamp housing data collected while processing a second substrate of the plurality of substrates.

3. The method according to any one of claims 1 or 2, wherein the comparison comprises: determining a threshold limit based on said first subset of said lamp housing data; as well as generating a metric based on said second subset of said lamp housing data; as well as The metric is compared to the threshold limit.

4. The method of claim 1 , wherein determining whether the lamp housing crack exists comprises: extracting one or more features from the lamp housing data; as well as The one or more features are provided to a trained neural network, which provides an output indicating whether a crack in the lamp housing is present. 5 . The method of claim 4 , wherein the one or more characteristics include a maximum minus a minimum range of sensor values ​​obtained during processing of the substrate.

6. The method of any one of claims 1, 2, 4 or 5, wherein the sensor is a pressure sensor configured to detect pressure along a surface of the lamp housing, or a humidity sensor configured to detect relative humidity associated with the lamp housing. The method of claim 1 , wherein the processing chamber is a reduced pressure chamber. The method of claim 1 , wherein the processing chamber is an atmospheric chamber.

9. The method of any one of claims 1, 2, 4, 5, 7 or 8, wherein the processing chamber is at least one of an annealing chamber or an oxidation chamber.

10. A non-transitory computer readable storage medium having stored thereon instructions which, when executed by a processor, result in a method associated with a processing chamber having a lamp housing, the method comprising: sequentially processing a plurality of substrates in the processing chamber; collecting lamp housing data indicative of a fluid leak in the lamp housing from a sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates; Determine whether there is a crack in the lamp housing according to the lamp housing data; as well as In response to determining that the lamp housing crack exists, at least one of triggering an alarm or stopping processing in the processing chamber is performed.

11. The non-transitory computer readable storage medium of claim 10, wherein determining whether a lamp housing crack exists comprises: A first subset of the lamp housing data collected while processing a first substrate of the plurality of substrates is compared to a second subset of the lamp housing data collected while processing a second substrate of the plurality of substrates.

12. The non-transitory computer-readable storage medium of any one of claims 10 or 11, determining a threshold limit based on the first subset of the lamp housing data; generating a metric based on said second subset of said lamp housing data; as well as The metric is compared to the threshold limit.

13. The non-transitory computer readable storage medium of claim 10, wherein determining whether a lamp housing crack exists comprises: extracting one or more features from the time series lamp housing data; as well as The one or more features are provided to a trained neural network that provides an output indicating whether a lamp housing crack is present.

14. The non-transitory computer readable storage medium of any one of claims 10, 11, or 13, wherein the one or more characteristics include a maximum minus a minimum range of sensor values ​​obtained during processing of the substrate.

15. The non-transitory computer-readable storage medium of claim 10, wherein the sensor is a pressure sensor configured to detect pressure along a surface of the lamp housing or a humidity sensor configured to detect relative humidity associated with the lamp housing.

16. The non-transitory computer readable storage medium of claim 10, wherein the processing chamber is a reduced pressure chamber.

17. The non-transitory computer readable storage medium of claim 10, wherein the processing chamber is an atmospheric chamber.

18. The non-transitory computer readable storage medium of any one of claims 10, 11, 13, 15, 16, or 17, wherein the processing chamber is at least one of an annealing chamber or an oxidation chamber.

19. An apparatus for processing a substrate, comprising: a processing chamber configured to process a substrate using radiant heat provided by the lamp housing; a sensor operably connected to the lamp housing; as well as A controller, the controller being configured to: sequentially processing a plurality of substrates in the processing chamber; collecting lamp housing data indicative of a fluid leak in the lamp housing from the sensor operably connected to the lamp housing of the processing chamber during processing of the plurality of substrates; Determine whether there is a crack in the lamp housing according to the lamp housing data; as well as In response to determining that a lamp housing crack is present, at least one of triggering an alarm or stopping processing in the processing chamber is performed.

20. The processing chamber of claim 19, wherein the controller is further configured to compare a first subset of the lamp housing data collected while processing a first substrate of the plurality of substrates with a second subset of the lamp housing data collected while processing a second substrate of the plurality of substrates.

Images