Real-time self-calibration semiconductor infrared temperature measuring device and control method thereof

Abstract

The invention relates to the technical field of semiconductor infrared temperature measurement, and provides a real-time self-calibration semiconductor infrared temperature measurement device and a control method thereof. An optical front-end unit of the device comprises a rotary driving mechanism and a rotary filter wheel located in a temperature measurement light path, and a measurement light filter, a calibration attenuation piece with a known attenuation rate and a black body reference piece are arranged on the optical front-end unit; the control unit drives the rotary filter wheel to sequentially switch the three to enter a light path; the signal processing unit converts the modulated optical signals into corresponding first, second and third voltage signals, calculates the real-time total transmittance of an optical path based on the first and second voltage signals and a known attenuation rate, compensates the third signal in combination with the first signal, and outputs a corrected temperature value according to a preset voltage-temperature mapping relation. The device is used for carrying out anti-pollution interference non-contact on-line temperature measurement on a wafer or a cavity in semiconductor manufacturing, and the detection precision and the stability and the reliability in long-term use of a semiconductor industrial site are improved.

Description

Technical Field

[0001] This invention relates to the field of semiconductor infrared temperature measurement technology, and in particular to a real-time self-calibrating semiconductor infrared temperature measurement device and its control method. Background Technology

[0002] In the semiconductor manufacturing field, infrared thermometry is widely used in key areas such as wafer temperature detection and cavity thermal management due to its advantages of being non-contact, having a fast response time, and being suitable for high-temperature environments. Its basic principle is to detect the intensity of infrared radiation emitted by the target and deduce the temperature value based on the radiation-temperature mapping relationship. Its detection applications become increasingly important as process nodes continue to shrink, and advanced processes place increasingly stringent demands on the accuracy, uniformity, and repeatability of temperature control.

[0003] In practical applications, temperature measurement accuracy has long been severely affected by two types of systematic interference. First, there is background stray radiation interference. High-temperature components such as the inner walls of the process chamber, heaters, and plasma emit non-target infrared radiation. This stray light, after reflection or scattering, enters the temperature measurement optical path and superimposes on the target signal, leading to overestimation of the measurement result and fluctuations with process conditions. Second, there is transmittance attenuation in the optical transmission path. To protect internal optical components, the temperature probe usually has a protective window (e.g., a sapphire window). However, during long-term operation, the surface of this window gradually becomes dirty due to deposits, contaminants, or chemical reactions, causing a continuous decrease in the total transmittance of the entire temperature measurement optical path. This results in a continuous attenuation of the original signal, manifesting as a slow drift and systematic underestimation of the temperature reading.

[0004] To address the aforementioned issues, traditional methods primarily rely on periodic shutdowns, manual window cleaning, and offline calibration using a standard blackbody source. This approach not only disrupts the production process and reduces equipment utilization, but more importantly, it fails to reflect dynamic contamination and drift occurring during the process, making it difficult to guarantee the reliability of temperature measurements under continuous operation.

[0005] To reduce human intervention, some existing solutions attempt to introduce background subtraction mechanisms, such as acquiring and subtracting the background signal in a "targetless" state before measurement. However, such methods typically assume a stable background and constant system transmittance. Once window contamination causes changes in optical path attenuation, the background signal itself will also be distorted, leading to subtraction failure. Secondly, another solution attempts to eliminate some attenuation effects through dual-wavelength thermometry. However, this method heavily relies on the assumption that the emissivity of the measured material is similar in both wavelength bands. The emissivity of semiconductor materials (such as silicon and gallium nitride) often varies drastically with wavelength, resulting in significant errors in practical scenarios.

[0006] Therefore, there is an urgent need for a real-time self-calibrating semiconductor infrared temperature measurement device and its control method to improve the above problems. Summary of the Invention

[0007] This invention provides a real-time self-calibrating semiconductor infrared temperature measurement device and its control method. This invention is used for non-contact online temperature measurement of wafer or cavity temperature to resist contamination interference during semiconductor manufacturing.

[0008] According to a first aspect of the present invention, a real-time self-calibrating semiconductor infrared temperature measuring device is provided. The device includes an optical front-end unit, a signal processing unit, and a control unit. The optical front-end unit includes a rotary drive mechanism and a rotary filter wheel, the rotary filter wheel being located in the temperature measuring optical path. The rotary drive mechanism is connected to the rotary filter wheel. The rotary filter wheel is provided with a measuring filter, a calibration attenuator with a known attenuation rate, and a blackbody reference plate. The control unit is connected to the rotary drive mechanism. During a measurement cycle, the control unit controls the rotary drive mechanism to drive the rotary filter wheel to rotate, so that the temperature measuring optical path passes through the blackbody reference plate, the... The system includes a calibration attenuator and a measurement filter. A signal processing unit receives the optical signal modulated by the rotating filter wheel and converts it into a voltage signal. The voltage signal includes a first signal corresponding to the blackbody reference, a second signal corresponding to the calibration attenuator, and a third signal corresponding to the measurement filter. Based on the first signal, the second signal, and the known attenuation rate of the calibration attenuator, the unit calculates the real-time total transmittance of the temperature measurement optical path. It also compensates the third signal based on the real-time total transmittance and the first signal to obtain a compensated voltage signal. Finally, based on the compensated voltage signal and a preset voltage-temperature mapping relationship, the unit calculates the corrected temperature value.

[0009] In one embodiment, the rotating filter wheel has a rotating shaft connected to a rotating drive mechanism; a measurement filter, a calibration attenuator, and a blackbody reference plate are circumferentially distributed around the rotating shaft; the rotating drive mechanism is used to drive the rotating shaft, the measurement filter, the calibration attenuator, and the blackbody reference plate to rotate together.

[0010] In one implementation, the measurement filter is a narrowband interference filter that allows only a specific band of the target radiation to pass through.

[0011] In one embodiment, the calibration attenuator is a neutral density filter with a constant transmittance ranging from 0 to 1.

[0012] In one embodiment, the surface of the blackbody reference sheet is a blackbody coating; the optical density of the blackbody coating in the operating wavelength band of the temperature measuring device is greater than 4.

[0013] In one embodiment, the signal processing unit includes a photodetector, a transimpedance amplifier, a programmable gain amplifier, and an analog-to-digital converter; the photodetector is used to convert an optical signal modulated by a rotating filter wheel into a current signal; the transimpedance amplifier is used to convert the current signal into a voltage signal; the programmable gain amplifier is used to amplify the voltage signal; and the analog-to-digital converter is used to convert the amplified analog voltage signal into a digital signal, providing a basis for subsequent temperature value calculation.

[0014] In one embodiment, the device further includes a window and a lens group located in the temperature measurement optical path; the window is located between the rotating filter wheel and the target being measured to isolate the temperature from the gas environment; the lens group is located between the window and the rotating filter wheel to focus the thermal radiation from the target being measured onto the photodetector.

[0015] According to a second aspect of the present invention, a control method for a real-time self-calibrating semiconductor infrared temperature measuring device is provided, for use in any of the infrared temperature measuring devices of the first aspect. The control method includes: controlling a rotary drive mechanism to drive a rotary filter wheel to rotate in each measurement cycle, so that a blackbody reference plate, a calibration attenuator, and a measurement filter are respectively placed in the temperature measuring optical path; acquiring a first signal corresponding to the blackbody reference plate, a second signal corresponding to the calibration attenuator, and a third signal corresponding to the measurement filter; calculating the real-time total transmittance of the temperature measuring optical path based on the first signal, the second signal, and the known attenuation rate of the calibration attenuator; compensating the third signal based on the real-time total transmittance and the first signal to obtain a compensated voltage signal, and calculating a corrected temperature value based on the compensated voltage signal and a preset voltage-temperature mapping relationship.

[0016] In one implementation, calculating the real-time total transmittance of the temperature-measuring optical path includes: according to the formula T_sys=(V_cal-V_bg) / (V_prev) k) Calculate the real-time total transmittance T_sys; where V_bg is the first signal corresponding to the blackbody reference, V_cal is the second signal corresponding to the calibration attenuator, k is the known attenuation rate of the calibration attenuator, and V_prev is the third signal corresponding to the measured filter in the previous measurement cycle.

[0017] In one embodiment, the third signal is compensated by: calculating the compensated voltage signal V_corrected according to the formula V_corrected=(V_raw-V_bg) / T_sys; where V_raw is the third signal corresponding to the measuring filter in the current measurement cycle, V_bg is the first signal corresponding to the blackbody reference, and T_sys is the real-time total transmittance of the temperature measuring optical path.

[0018] Compared with existing technologies, the advantages of this invention are as follows: By introducing a blackbody reference plate, a calibration attenuator, and a measurement filter through a rotating filter wheel in each measurement cycle, the signal processing unit can simultaneously acquire the corresponding first, second, and third signals. Based on the first signal, the influence of background noise can be accurately removed. Combined with the second signal and the known attenuation rate of the calibration attenuator, the real-time total transmittance of the current temperature measurement optical path can be deduced. This transmittance comprehensively reflects system attenuation changes caused by factors including protective window contamination and optical component aging. Furthermore, by using this real-time total transmittance and the composite compensation of the third signal with the first signal, the dual influence of background interference and system attenuation on the original temperature measurement signal is effectively eliminated, thus obtaining a compensated voltage signal that only reflects the true radiation intensity of the measured target. Based on this, a high-precision corrected temperature value can be directly output through a preset voltage-temperature mapping relationship. Therefore, this device does not require external calibration or manual intervention and can achieve dynamic sensing and automatic correction of optical path state changes during continuous operation, improving detection accuracy and the stability and reliability of infrared thermometry in long-term use in the semiconductor industry. Attached Figure Description

[0019] Figure 1 This is a structural block diagram of a real-time self-calibrating semiconductor infrared temperature measurement device according to an exemplary embodiment.

[0020] Figure 2 This is a flowchart illustrating a control method for a real-time self-calibrating semiconductor infrared temperature measuring device according to another exemplary embodiment.

[0021] Explanation of the reference numerals in the figure:

[0022] 100. Real-time self-calibrating semiconductor infrared temperature measurement device; 1. Optical front-end unit; 2. Signal processing unit; 3. Control unit; 11. Rotary drive mechanism; 12. Rotary filter wheel; 121. Measuring filter; 122. Calibration attenuator; 123. Blackbody reference plate; 21. Photodetector; 22. Transimpedance amplifier; 23. Programmable gain amplifier; 24. Analog-to-digital converter. Detailed Implementation

[0023] Unless otherwise defined, the technical or scientific terms used in this specification should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. Specific embodiments of the invention will be described below with reference to the accompanying drawings. It should be noted that, in order to provide a concise description, this specification cannot provide a detailed description of all features of the actual embodiments. Without departing from the spirit and scope of the invention, those skilled in the art can make modifications and substitutions to the embodiments of the invention, and the resulting embodiments are also within the protection scope of the invention.

[0024] like Figure 1 As shown, the first embodiment of the present invention provides a real-time self-calibrating semiconductor infrared temperature measuring device. The device includes an optical front-end unit 1, a signal processing unit 2, and a control unit 3. The optical front-end unit 1 includes a rotary drive mechanism 11 and a rotary filter wheel 12, with the rotary filter wheel 12 located in the temperature measuring optical path. The rotary drive mechanism 11 is connected to the rotary filter wheel 12. The rotary filter wheel 12 is provided with a measuring filter 121, a calibration attenuator 122 with a known attenuation rate, and a blackbody reference 123. The control unit 3 is connected to the rotary drive mechanism 11. During the measurement cycle, the control unit 3 controls the rotary drive mechanism 11 to drive the rotary filter wheel 12 to rotate, so that the temperature measuring optical path passes through the blackbody reference 123. The system includes a reference plate 123, a calibration attenuator 122, and a measurement filter 121. The signal processing unit 2 receives the optical signal modulated by the rotating filter wheel 12 and converts it into a voltage signal. The voltage signal includes a first signal corresponding to the blackbody reference plate 123, a second signal corresponding to the calibration attenuator 122, and a third signal corresponding to the measurement filter 121. Based on the first signal, the second signal, and the known attenuation rate of the calibration attenuator 122, the system calculates the real-time total transmittance of the temperature measurement optical path. Based on the real-time total transmittance and the first signal, the system compensates the third signal to obtain a compensated voltage signal. Based on the compensated voltage signal and a preset voltage-temperature mapping relationship, the system calculates the corrected temperature value.

[0025] In one embodiment, the rotating filter wheel 12 has a rotating shaft connected to a rotating drive mechanism 11; the measuring filter 121, the calibration attenuator 122, and the blackbody reference plate 123 are circumferentially distributed around the rotating shaft; the rotating drive mechanism 11 is used to drive the rotating shaft, the measuring filter 121, the calibration attenuator 122, and the blackbody reference plate 123 to rotate together.

[0026] In some specific embodiments, the rotary drive mechanism 11 includes a miniature stepper motor, the output shaft of which is coaxially connected to the shaft of the rotating filter wheel 12. Under the command of the control unit 3, the stepper motor rotates precisely at a preset angle, thereby driving the rotating filter wheel 12 to rotate synchronously as a whole, so that the measuring filter 121, the calibration attenuator 122, and the blackbody reference 123 are sequentially and stably switched into the temperature measurement optical path. This drive mechanism has a compact structure and high positioning repeatability, and can achieve rapid and reliable switching between three optical states without introducing additional vibration or optical axis misalignment, providing precise timing assurance for the self-calibration process within each measurement cycle.

[0027] In other specific embodiments, the measuring filter 121, the calibration attenuator 122, and the blackbody reference 123 are circumferentially distributed on the rotating filter wheel 12. Their arrangement order can be flexibly set according to actual application requirements and does not need to be fixed to a specific order. The control unit 3 only needs to ensure that all three are sequentially switched to the temperature measurement optical path and that the corresponding signals are acquired in each measurement cycle to complete the self-calibration calculation. This design avoids rigid constraints on the switching sequence, improves the flexibility of system scheduling, and is compatible with various operating strategies (such as priority temperature measurement, priority background acquisition, etc.) without affecting the accurate inversion of real-time total transmittance and the temperature compensation effect.

[0028] In one embodiment, the measuring filter 121 is a narrowband interference filter that allows only a specific band of the target radiation to pass through.

[0029] In some specific embodiments, the optical parameters of the narrowband interference filter are optimized according to the temperature range and material properties of the target object. For example, for high-temperature measurement scenarios (such as 600–1200℃) commonly used in semiconductor processes, a narrowband interference filter with a center wavelength of 900nm and a bandwidth of 10nm can be selected. This filter constitutes the measurement area on the rotating filter wheel 12. Its function is to allow only the thermal radiation of the target object within this specific wavelength band to pass through during normal temperature measurement, effectively suppressing the interference of stray radiation of other wavelengths in the cavity (such as plasma emission, broadband radiation of the heater, etc.) on the detection signal, thereby improving the signal-to-noise ratio and temperature measurement selectivity.

[0030] In other specific embodiments, the calibration attenuator 122 is a neutral density filter, forming the calibration area on the rotating filter wheel 12. This neutral density filter has a flat spectral response across the entire temperature measurement wavelength range, and its transmittance k is a known constant ranging from (0, 1). For example, a high-precision neutral density filter with a transmittance of 1% (i.e., k = 0.01) can be selected. Before leaving the factory, the filter undergoes rigorous calibration using a standard blackbody source, and the calibrated k value is stored as a system constant in the non-volatile memory of the control unit 3 or the signal processing unit 2. Its core function is to provide a known, stable, and wavelength-independent light intensity attenuation reference: when the calibration attenuator 122 is engaged in the temperature measurement optical path, the signal received by the detector reflects the known input ratio under the current total attenuation state of the optical path, enabling the system to deduce the real-time total transmittance, including factors such as protective window contamination and optical element aging, based on this reference, for subsequent accurate compensation of the original temperature measurement signal.

[0031] In one embodiment, the surface of the blackbody reference sheet 123 is a blackbody coating; the optical density of the blackbody coating in the operating wavelength band of the temperature measuring device is greater than 4.

[0032] In some specific embodiments, a blackbody reference plate 123 constitutes a completely black background area on the rotating filter wheel 12. Its substrate is a metal or ceramic sheet coated with a high-absorption-rate blackbody coating, such as Acktar Fractal Black, Nextel Velvet, or other commercially available ultra-black coatings. This coating has extremely high absorptivity within the operating wavelength range of the temperature measuring device (e.g., 0.8–1.1 μm or 8–14 μm, depending on the specific process detection application), corresponding to an optical density (OD) greater than 4, blocking more than 99.99% of incident infrared radiation. When the rotating filter wheel 12 switches the completely black background area to the temperature measuring optical path, external thermal radiation from the direction of the target being measured is almost completely absorbed. The signal received by the detector mainly originates from internal electronic dark noise and trace amounts of background stray light leaking from non-main optical path directions. This state provides the system with a near-zero input reference that includes current background interference components, enabling precise subtraction of background offsets and ensuring that subsequent compensation of the original temperature measurement signal reflects only the true radiation intensity of the target.

[0033] In some examples, the blackbody reference sheet 123 is a mechanical carrier with high stability, high thermal conductivity, and easy processing. The high-absorption coating applied to it is the core of achieving the "completely black background" function. The combination of the two achieves the function of "physically and completely blocking the target optical path," enabling accurate measurement of the background.

[0034] In one embodiment, the signal processing unit 2 includes a photodetector 21, a transimpedance amplifier 22, a programmable gain amplifier 23, and an analog-to-digital converter 24; the photodetector 21 is used to convert the optical signal modulated by the rotating filter wheel 12 into a current signal; the transimpedance amplifier 22 is used to convert the current signal into a voltage signal; the programmable gain amplifier 23 is used to amplify the voltage signal; and the analog-to-digital converter 24 is used to convert the amplified analog voltage signal into a digital signal, providing a basis for subsequent temperature value calculation.

[0035] In some specific embodiments, the signal processing unit 2 includes a high-sensitivity photodetector 21 and a signal conditioning circuit board. The high-sensitivity photodetector 21 is located on the temperature-measuring optical path behind the rotating filter wheel 12, and is used to receive infrared radiation modulated by the filter wheel and convert it into a weak current signal. Depending on the operating wavelength, the detector can be a mercury cadmium telluride (HgCdTe, abbreviated as MCT) detector (suitable for near-infrared to mid-infrared, e.g., 1–5 μm) or a lithium tantalate pyroelectric detector (suitable for broadband far-infrared, e.g., 8–14 μm) to match the peak radiation wavelength corresponding to the typical temperature range in semiconductor processes.

[0036] In some examples, the signal conditioning circuit board is located immediately after the output of the photodetector 21 and integrates a transimpedance amplifier 22, a programmable gain amplifier 23, and an analog-to-digital converter 24 (ADC). The transimpedance amplifier 22 linearly converts the weak current signal output by the detector into a voltage signal; the programmable gain amplifier 23 dynamically adjusts the gain and performs low-pass filtering on the voltage signal to adapt to signal amplitudes under different temperature conditions and suppress high-frequency noise; the high-precision ADC 24 converts the amplified and filtered analog voltage signal into a digital signal, which is then used by the control unit 3 or embedded processor to execute subsequent self-calibration algorithms and temperature inversion calculations. This signal conditioning chain ensures high-fidelity, low-noise conversion from picoampere-level photocurrent to digital sampled values, providing a reliable electrical signal foundation for real-time compensation and high-precision temperature measurement.

[0037] In one embodiment, the device further includes a window and a lens group located in the temperature measuring optical path; the window is located between the rotating filter wheel 12 and the target being measured, and is used to isolate the temperature from the gas environment; the lens group is located between the window and the rotating filter wheel 12, and focuses the thermal radiation from the target being measured onto the photodetector 21.

[0038] In some specific embodiments, the protective window is located at the very front of the probe, between the target being measured and subsequent optical components. It is typically made of sapphire, a material with high infrared transmittance, high hardness, and chemical inertness. Its main function is to physically isolate the internal precision optical system from the external high-temperature, corrosive process gas, or particulate environment, ensuring long-term stable operation of the device in the harsh semiconductor manufacturing chamber. However, during continuous use, process byproducts, contaminants, or thin films inevitably accumulate on the outer surface of the window, leading to a gradual decrease in its transmittance, which in turn causes attenuation of the temperature measurement signal and systematic drift.

[0039] In other specific embodiments, the collecting lens group is located between the protective window and the rotating filter wheel 12, and consists of one or more focusing lenses made of infrared transmitting materials (such as zinc selenide (ZnSe), zinc sulfide (ZnS), or silicon (Si). Its function is to efficiently converge the heat radiation emitted from the target being measured and image it onto the sensitive surface of the photodetector 21, thereby improving the system's luminous flux, spatial resolution, and signal-to-noise ratio. This lens group, together with the protective window, constitutes the front-end optical interface of the temperature measurement optical path. Its overall transmittance performance is directly affected by the contamination state of the window, thus becoming a key link in the self-calibration mechanism of this invention that requires real-time compensation.

[0040] In some embodiments, the infrared temperature measurement device employs a highly integrated embedded architecture, merging the functions of the control unit 3 and the signal processing unit 2 onto a single chip platform to form a compact, efficient, and low-latency control and processing core. This core is implemented by a microcontroller unit (MCU) or digital signal processor (DSP), which integrates a high-precision analog-to-digital converter (ADC) 24, a motor control timer, a digital filtering module, and an industrial communication interface.

[0041] Specifically, the MCU / DSP is directly connected to the stepper motor driver in the rotary drive mechanism 11. By precisely generating pulse and direction signals, it controls the temperature measurement optical path of the rotating filter wheel 12 to switch as needed between the normal temperature measurement zone (measuring filter 121), the calibration zone (calibration attenuator 122), and the completely black background zone (blackbody reference 123). A built-in position feedback mechanism ensures the repeatability and reliability of the optical alignment. Simultaneously, the analog signal from the photodetector 21, after being processed by the front-end conditioning circuit, is sampled by a high-resolution ADC either internally or externally to the MCU / DSP. The resulting digital signal immediately executes a complete self-calibration algorithm within the chip, including background noise extraction (V_bg), real-time total transmittance calculation (T_sys), original signal compensation (V_corrected), and temperature inversion based on a pre-stored voltage-temperature mapping relationship.

[0042] In addition, the control and processing core also integrates a standard industrial communication module, which supports real-time uploading of calibrated temperature values ​​and system health status (such as current T_sys, pollution warning signs, and filter wheel position status) to a host computer or factory automation system via RS-485 bus (Recommended Standard 485, RS-485), compatible with Modbus communication protocol or Industrial Ethernet (Industrial Ethernet) interface, to achieve seamless integration and remote operation and maintenance.

[0043] This integrated design not only significantly reduces the number of external components and signal transmission paths, and reduces noise interference and system latency, but also improves overall reliability and anti-interference capabilities, making it particularly suitable for installation scenarios inside semiconductor manufacturing equipment with limited space and complex electromagnetic environments.

[0044] like Figure 2 As shown, the second embodiment of the present invention provides a control method for a real-time self-calibrating semiconductor infrared temperature measuring device, used in any of the infrared temperature measuring devices described in the above embodiments. The control method includes the following steps S1-S4: S1, in each measurement cycle, control the rotary drive mechanism to drive the rotary filter wheel to rotate so that the blackbody reference plate, calibration attenuator and measurement filter are respectively placed in the temperature measurement optical path.

[0045] S2, acquire the first signal corresponding to the blackbody reference, the second signal corresponding to the calibration attenuator, and the third signal corresponding to the measurement filter.

[0046] S3. Calculate the real-time total transmittance of the temperature measurement optical path based on the first signal, the second signal, and the known attenuation rate of the calibration attenuator.

[0047] S4. Based on the real-time total transmittance and the first signal, the third signal is compensated to obtain the compensated voltage signal. Based on the compensated voltage signal and the preset voltage-temperature mapping relationship, the corrected temperature value is calculated.

[0048] In some specific embodiments, during the background reference measurement stage, the control unit (such as an MCU) first drives a stepper motor to precisely rotate the rotating filter wheel to the completely black background area, i.e., aligning the completely black background sheet with the temperature measurement optical path. This position is confirmed in real time by an integrated position sensor (such as a Hall sensor or photoelectric encoder), ensuring the reliability and repeatability of the optical switching. Subsequently, the control unit triggers the analog-to-digital converter (ADC) to sample the detector output multiple times and calculate its average value, denoted as V_bg. Since the completely black background area uses a high-absorption blackbody coating with an optical density greater than 4, infrared radiation from the direction of the measured target is almost completely blocked. Therefore, the signal V_bg acquired at this time mainly includes detector dark current, electronic circuit thermal noise, and a very small amount of stray background radiation leaking from non-main optical path paths. It can be regarded as a "pure background" signal that does not contain effective radiation from the target under the current operating state. This background value will be used for background subtraction of the original temperature measurement signal in subsequent steps, which is a prerequisite for achieving high-precision compensation.

[0049] In other specific embodiments, during the online calibration phase of the system attenuation, the control unit drives the rotating filter wheel to switch to the calibration area, aligning the calibration attenuator with the temperature measurement optical path. This attenuator is a neutral density filter with a known and stable transmittance k within the operating wavelength range; its optical characteristics have been precisely calibrated at the factory using a standard blackbody source and are stored in the system memory. At this time, the thermal radiation from the target object passes sequentially through the contaminated protective window, the collecting lens group, and the calibration attenuator before reaching the detector. The signal processing unit acquires the output voltage in this state, denoted as V_cal.

[0050] In the core calculation step, the algorithm calls upon the original temperature measurement signal V_prev recorded in the measurement area (measurement filter) during the previous complete measurement cycle, and combines it with the previously obtained background benchmark V_bg, substituting it into the formula: T_sys=(V_cal-V_bg) / (V_prev) k); The real-time total transmittance T_sys of the current temperature measurement optical path is calculated. This value comprehensively reflects the light intensity attenuation effect caused by all factors, including protective window contamination, lens aging, and optical path alignment misalignment. When the window is clean, T_sys is close to 1; however, as contamination accumulates, T_sys will gradually decrease to less than 1. Therefore, the system not only achieves real-time quantitative diagnosis of the optical path attenuation state but also provides key parameters for subsequent dynamic compensation of the current temperature measurement signal, thus transforming traditional passive drift into an active and correctable system variable.

[0051] In some specific embodiments, during the compensation temperature measurement and output stage, the control unit drives the rotating filter wheel back to the normal temperature measurement area, so that the measuring filter is re-aligned with the temperature measurement optical path. At this time, the thermal radiation from the target being measured passes through the protective window, the collecting lens group, and the narrowband interference filter before entering the photodetector. The signal processing unit collects the raw output voltage in this state, which is denoted as V_raw.

[0052] Subsequently, the system performs two key signal compensation steps: First, background subtraction is performed, calculating V_target = V_raw - V_bg, effectively eliminating common-mode shift introduced by stray radiation from the chamber, electron dark current, and circuit noise, retaining the net signal contributed only by target radiation; then, attenuation compensation is performed, normalizing the signal using the formula V_corrected = V_target / T_sys, "restoring" the light intensity attenuation caused by window contamination or optical component aging to the equivalent level under an ideal clean optical path. This step essentially uses the real-time total transmittance T_sys calibrated in the previous cycle to perform dynamic gain correction on the current signal, thereby offsetting the effects of long-term drift.

[0053] After compensation is completed, the system inputs V_corrected to the voltage-temperature mapping relationship pre-stored in the MCU. This mapping is based on the calibration obtained at the factory using a standard blackbody source under clean optical path conditions, and can be a high-precision lookup table or a fitted polynomial function. From this, the final corrected temperature value T is derived and output, ensuring that the temperature measurement results are not affected by contamination or aging.

[0054] Finally, the raw signal V_raw acquired in this measurement is updated and stored as V_prev, which serves as the reference for calculating the system transmittance T_sys in the next measurement cycle, thus forming a closed-loop, continuous, and self-sustaining online self-calibration process.

[0055] It is worth noting that the above process consists of three key steps forming a complete self-calibration cycle: the first step is the background reference measurement in the completely black background area, the second step is the online calibration of the system attenuation in the calibration area, and the third step is the compensation temperature measurement and result output in the normal temperature measurement area.

[0056] This self-calibration cycle can be triggered periodically at preset time intervals (e.g., every minute), or it can be actively initiated when the system detects a sudden drop in the real-time total transmittance T_sys (indicating sudden contamination of the indicator window or optical path anomaly), thus achieving event-driven dynamic calibration. During normal operation, the device remains in the normal temperature measurement zone for continuous temperature detection most of the time, only briefly switching to the completely black background zone and calibration zone during each calibration cycle to complete background acquisition and attenuation calibration. The switching process is rapid and has minimal interference with the main temperature measurement task, thereby ensuring high-precision self-calibration while maintaining high continuity and real-time performance of temperature measurement.

[0057] This invention, through the hardware innovation of integrating a "three-state filter wheel" into a single optical path, constructs three physically meaningful and functionally coordinated optical states: a completely black background area, a known attenuation calibration area, and a normal temperature measurement area, along with a matching three-step self-calibration algorithm. This method utilizes the completely black background area to accurately acquire the background noise reference V_bg, which is free from target radiation. Through the calibration area, combined with historical temperature measurement data and a known attenuation rate, the real-time total transmittance T_sys, reflecting window contamination and system aging, is derived. Then, the original signal acquired in the normal temperature measurement area undergoes a composite compensation process of "first subtracting background, then removing attenuation," thereby simultaneously eliminating the two core error sources—background stray light interference and optical path attenuation drift—within a single measurement cycle. The entire process requires no external calibration, does not rely on material emissivity models, and does not require production interruption. It achieves closed-loop real-time calibration by switching from hardware to algorithm compensation, fundamentally solving the long-standing problem of traditional infrared thermometry's inability to simultaneously address background suppression and contamination compensation in semiconductor industrial settings. This provides a new technical path for high-precision, high-reliability temperature detection scenarios in semiconductor manufacturing.

[0058] This embodiment constructs an in-situ observation system capable of solving interference parameters by integrating three regions on a rotating filter wheel: a completely black background region, a known attenuation calibration region, and a narrowband temperature measurement region. With this hardware, the algorithm employs a lightweight strategy of "three-step loop, two-form compensation": within each measurement cycle, background reference acquisition, system attenuation calibration, and compensated temperature measurement output are completed sequentially. This collaborative mechanism achieves simultaneous perception, in-situ decoupling, and real-time compensation of the two long-standing challenges of background interference and window contamination with minimal hardware cost. It not only avoids the complex design of multiple optical paths, multiple detectors, or external reference sources, but also achieves high-precision self-calibration without interrupting continuous temperature measurement, demonstrating a high degree of unity between hardware simplicity and algorithmic intelligence.

[0059] In this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The term "multiple" refers to two or more unless otherwise expressly defined.

[0060] The above description of the embodiments is intended to enable those skilled in the art to understand and apply the present invention. It will be apparent to those skilled in the art that various modifications can be made to these embodiments, and the general principles described herein can be applied to other embodiments without creative effort. Therefore, the present invention is not limited to the embodiments described herein, and any improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope and spirit of the invention are within the scope of the present invention.

Claims

1. A real-time self-calibrating semiconductor infrared temperature measurement device, characterized in that, The device comprises an optical front-end unit, a signal processing unit and a control unit; The optical front-end unit comprises a rotary drive mechanism and a rotary filter wheel, and the rotary filter wheel is located in a temperature measurement light path; the rotary drive mechanism is connected to the rotary filter wheel; The rotary filter wheel is provided with a measurement filter, a calibration attenuation piece with a known attenuation rate and a blackbody reference piece; The control unit is connected to the rotary drive mechanism, and when a measurement cycle is performed, the control unit is used to control the rotary drive mechanism to drive the rotary filter wheel to rotate, so that the temperature measurement light path passes through the blackbody reference piece, the calibration attenuation piece and the measurement filter; The signal processing unit is used to receive the light signal modulated by the rotary filter wheel and convert it into a voltage signal, the voltage signal comprising a first signal corresponding to the blackbody reference piece, a second signal corresponding to the calibration attenuation piece and a third signal corresponding to the measurement filter, based on the first signal, the second signal and the known attenuation rate of the calibration attenuation piece, to calculate the real-time total transmittance of the temperature measurement light path, and based on the real-time total transmittance and the first signal, to compensate the third signal to obtain a compensated voltage signal, and based on the compensated voltage signal and a preset voltage-temperature mapping relationship, to calculate a corrected temperature value.

2. The infrared temperature measurement device of claim 1, wherein, The rotary filter wheel has a rotating shaft, the rotating shaft is connected to the rotary drive mechanism; the measurement filter, the calibration attenuation piece and the blackbody reference piece are distributed circumferentially around the rotating shaft; and the rotary drive mechanism is used to drive the rotating shaft, the measurement filter, the calibration attenuation piece and the blackbody reference piece to rotate together.

3. The infrared temperature measurement device of claim 1, wherein, The measurement filter is a narrowband interference filter that only allows a specific waveband of target radiation to pass through.

4. The infrared temperature measurement device of claim 1, wherein, The calibration attenuation piece is a neutral density filter, and the transmittance of the neutral density filter is a constant, with a value range of (0, 1).

5. The infrared temperature measurement device of claim 1, wherein, The surface of the blackbody reference piece is a blackbody coating; and the optical density of the blackbody coating in the working waveband of the temperature measurement device is greater than 4.

6. The infrared temperature measurement device of claim 1, wherein, The signal processing unit comprises a photodetector, a transimpedance amplifier, a programmable gain amplifier and an analog-to-digital converter; the photodetector is used to convert the light signal modulated by the rotary filter wheel into a current signal; the transimpedance amplifier is used to convert the current signal into a voltage signal; the programmable gain amplifier is used to amplify the voltage signal; and the analog-to-digital converter is used to convert the analog voltage signal after amplification into a digital signal, providing a basis for subsequent temperature value calculation.

7. The infrared temperature measurement device of claim 6, wherein, The device further comprises a viewing window and a lens group located in the temperature measurement light path; The viewing window is located between the rotary filter wheel and the measured target, and is used to isolate temperature and gas environment; The lens group is located between the viewing window and the rotary filter wheel, and focuses thermal radiation from the measured target onto the photodetector.

8. A control method of a real-time self-calibrating semiconductor infrared temperature measuring device, for the infrared temperature measuring device according to any one of claims 1 to 7, characterized in that, The control method comprises: In each measurement cycle, the rotary drive mechanism is controlled to drive the rotary filter wheel to rotate, so that the blackbody reference piece, the calibration attenuation piece and the measurement filter are placed in the temperature measurement light path, respectively; acquire a first signal corresponding to the blackbody reference plate, a second signal corresponding to the calibration attenuation plate, and a third signal corresponding to the measurement filter plate; calculate a real-time total transmittance of the temperature measurement optical path according to the first signal, the second signal, and a known attenuation rate of the calibration attenuation plate; compensate the third signal according to the real-time total transmittance and the first signal to obtain a compensated voltage signal, and calculate a corrected temperature value based on the compensated voltage signal and a preset voltage-temperature mapping relationship.

9. The control method according to claim 8, characterized by, The real-time total transmittance of the temperature measurement light path is calculated, including: calculating the real-time total transmittance T_sys according to the formula T_sys=(V_cal-V_bg) / (V_prev k); wherein V_bg is the first signal corresponding to the black body reference piece, V_cal is the second signal corresponding to the calibration attenuation piece, k is the known attenuation rate of the calibration attenuation piece, and V_prev is the third signal corresponding to the measurement filter piece in the previous measurement period.

10. The control method according to claim 8 or 9, characterized by, The compensation of the third signal includes: calculating a compensated voltage signal V_corrected according to the formula V_corrected=(V_raw-V_bg) / T_sys; wherein V_raw is the third signal corresponding to the measurement filter plate in the current measurement period, V_bg is the first signal corresponding to the blackbody reference plate, and T_sys is the real-time total transmittance of the temperature measurement optical path.

Images