Bosch process etching endpoint detection method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of semiconductor detection, and discloses a Bosch process etching endpoint detection method and device, electronic equipment and a storage medium, the method comprising the following steps: obtaining light intensity time sequence data of each Bosch process cycle based on characteristic light intensity signals corresponding to the Bosch process cycle; determining a target frequency value according to the Bosch process cycle; extracting a frequency component with the target frequency value and a period phase corresponding to the target frequency component in the light intensity time sequence data, wherein the period phase represents the relative position of the target frequency component in the Bosch process cycle; determining the modulation depth of each Bosch process cycle based on the light intensity time sequence data, wherein the modulation depth represents the response amplitude of the characteristic light intensity signal to the gas switching signal; and detecting the etching endpoint based on the modulation depth and the period phase change amount between adjacent Bosch process cycles. The application can improve the recognition ability of the endpoint, and further improve the precision of etching endpoint detection.

Description

Technical Field

[0001] This invention relates to the field of semiconductor inspection technology, specifically to a method, apparatus, electronic device, and storage medium for detecting the etching endpoint of Bosch processes. Background Technology

[0002] Endpoint detection is a crucial step in semiconductor etching processes, requiring precise determination of whether the etching has reached the target interface, such as penetrating the stop layer beneath the silicon wafer. Currently, the etching endpoint is primarily determined by monitoring changes in the light intensity of specific characteristic spectral lines (such as SiF characteristic spectral lines).

[0003] Existing endpoint detection techniques typically rely on the premise that the plasma is in a quasi-steady state or a slowly changing state. The endpoint characteristics generally manifest as monotonic or trend-like changes in the intensity of specific characteristic spectral lines over a long timescale. However, in the Bosch process used for TSV (Through-Silicon Via) silicon etching, the etching gas and passivation gas are alternately introduced in a high-frequency periodic manner, and the process aperture ratio is extremely low (often less than 0.5%). This causes the light intensity change signal to switch rapidly and periodically with the Bosch process cycle. Furthermore, the light intensity change signal is extremely weak, resulting in the weak trend change corresponding to the etching endpoint being masked by the periodic modulation components, making accurate detection of the etching endpoint difficult.

[0004] Therefore, when using existing endpoint detection technology for etching endpoint detection, the weak trend change corresponding to the etching endpoint is masked by the periodic modulation component. This can easily lead to technical problems such as endpoint detection failure, misjudgment, or poor detection reliability due to the mismatch between the endpoint criterion and the actual physical process, thereby reducing the reliability and accuracy of Bosch process etching endpoint detection. Summary of the Invention

[0005] This invention provides a method, apparatus, electronic device, and storage medium for detecting the etching endpoint in Bosch processes, in order to solve the problems of endpoint detection failure, misjudgment, or poor detection reliability caused by insufficient time resolution and mismatch between endpoint criteria and actual physical processes.

[0006] In a first aspect, the present invention provides a method for detecting the etching endpoint in a Bosch process, the method comprising: The characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle are collected to obtain the light intensity time sequence data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by the gas switching signal triggered by the external process. Determine the target frequency value based on the Bosch process cycle; Within a continuously preset number of Bosch process cycles, determine the light intensity timing data corresponding to each Bosch process cycle; Extract the periodic phase corresponding to the target frequency component from each light intensity time series data to obtain the periodic phase corresponding to each Bosch process cycle; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase characterizes the relative position of the target frequency component in the corresponding Bosch process cycle. Based on the time-series data of each light intensity, the modulation depth of each Bosch process cycle is determined. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal. Determine the corresponding periodic phase change between each adjacent Bosch process cycle; When the phase change in each cycle is greater than the preset phase stability threshold, and the modulation depth is less than the preset modulation depth threshold, the Bosch process etching endpoint is determined to have been reached.

[0007] In one optional implementation, the Bosch process cycle includes an etching window and a passivation window; based on light intensity timing data, the modulation depth of each Bosch process cycle is determined, including: For each Bosch process cycle, determine the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window within the Bosch process cycle; The modulation depth of the Bosch process cycle is determined based on the degree of change between the first and second light intensity time series data.

[0008] In one optional implementation, the modulation depth of the Bosch process cycle is determined based on the degree of change between the first light intensity timing data and the second light intensity timing data, including: Determine the first light intensity average value corresponding to the first light intensity time series data, and determine the second light intensity average value corresponding to the second light intensity time series data; The average values ​​of the first and second light intensities are normalized to obtain the modulation depth of the Bosch process cycle.

[0009] In one optional implementation, corresponding to the light intensity timing data for each Bosch process cycle, the periodic phase corresponding to the target frequency component in the light intensity timing data is extracted, including: For each Bosch process cycle, frequency analysis is performed on the light intensity time series data corresponding to the Bosch process cycle to obtain the target frequency component in the light intensity time series data whose frequency is the target frequency value. The phase of the target frequency component at the midpoint of the Bosch process cycle is determined to obtain the periodic phase.

[0010] In one optional implementation, corresponding to the light intensity timing data for each Bosch process cycle, the periodic phase corresponding to the target frequency component in the light intensity timing data is extracted, including: Corresponding to the light intensity timing data of each Bosch process cycle, a digital phase-locked loop is used with the target frequency value as the reference frequency to perform phase-locked tracking on the light intensity timing data, so as to obtain the continuous phase corresponding to the target frequency component in the light intensity timing data. The phase value corresponding to the midpoint of the Bosch process cycle is determined as the cycle phase.

[0011] In one optional implementation, the process of extracting the periodic phase corresponding to the target frequency component from the light intensity timing data corresponding to each Bosch process cycle further includes: For each Bosch process cycle, Fourier transform is performed on the light intensity time series data corresponding to the Bosch process cycle to obtain the target frequency component in the light intensity time series data whose frequency is the target frequency value. The periodic phase is determined based on the phase value of the target frequency component.

[0012] In one alternative implementation, the modulation depth threshold is set to 0.05-0.1.

[0013] In one alternative implementation, the phase stabilization threshold is set to 20° to 45°.

[0014] In one alternative implementation, the preset number of Bosch process cycles is 3 to 5.

[0015] Secondly, the present invention provides a Bosch process etching endpoint detection device, the device comprising: The acquisition module is used to acquire characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle, and obtain light intensity time series data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by the gas switching signal triggered by the external process. The frequency determination module is used to determine the target frequency value based on the Bosch process cycle. The timing data determination module is used to determine the light intensity timing data corresponding to each Bosch process cycle within a continuously preset number of Bosch process cycles. The extraction module is used to extract the periodic phase corresponding to the target frequency component in each light intensity time series data to obtain the periodic phase corresponding to each Bosch process cycle; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase represents the relative position of the target frequency component in the corresponding Bosch process cycle. The modulation depth determination module is used to determine the modulation depth of each Bosch process cycle based on the time-series data of each light intensity. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal. The detection data determination module is used to determine the corresponding periodic phase change between each adjacent Bosch process cycle; The detection module is used to determine the end point of the Bosch process etching when the phase change in each cycle is greater than a preset phase stability threshold and the modulation depth is less than a preset modulation depth threshold.

[0016] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the Bosch process etching endpoint detection method described in the first aspect or any corresponding embodiment thereof.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the Bosch process etching endpoint detection method described in the first aspect or any corresponding embodiment thereof.

[0018] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the Bosch process etching endpoint detection method described in the first aspect or any corresponding embodiment thereof.

[0019] Compared with existing endpoint detection technologies, the Bosch process etching endpoint detection method, apparatus, electronic equipment, and storage medium proposed in this application have the following advantages: By constructing light intensity time-series data based on the characteristic light intensity signals corresponding to each Bosch process cycle, and extracting the modulation depth corresponding to each Bosch process cycle and the periodic phase change between adjacent Bosch process cycles from the light intensity time-series data, the actual etching process under the Bosch process can be characterized from two dimensions: response amplitude and response time-series stability. This makes the detection basis of the etching endpoint match the actual etching physical process, reduces the interference of periodic gas switching on the identification of etching endpoint features, improves the ability to identify and distinguish the state of the etching endpoint, and thus improves the accuracy of Bosch process etching endpoint detection.

[0020] Under the continuous constraint of N consecutive Bosch process cycles, the etching endpoint of the Bosch process can be detected by a joint determination mechanism of modulation depth and cycle phase. This can avoid misjudgment by a single cycle phase criterion or a single modulation depth criterion, thereby improving the reliability of Bosch process etching endpoint detection.

[0021] The modulation depth corresponding to the Bosch process cycle is determined by the degree of change between the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window within the Bosch process cycle. This modulation depth can quantify the response amplitude of the characteristic light intensity signal to the gas switching signal, providing a reliable basis for subsequent etching endpoint detection, thereby improving the accuracy and stability of Bosch process etching endpoint detection.

[0022] By extracting the periodic phase corresponding to each Bosch process cycle, the synchronization state between the characteristic light intensity signal and the Bosch process cycle can be obtained, providing a basis for etching endpoint detection based on periodic phase changes. At the same time, by determining the phase corresponding to the target frequency component at the midpoint of each Bosch process cycle as the periodic phase, the phase calculation deviation caused by unstable process cycle boundaries or asynchronous sampling frequencies can be avoided, improving the accuracy of periodic phase determination and thus enhancing the precision of etching endpoint detection. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first process of the Bosch process etching endpoint detection method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the second process of the Bosch process etching endpoint detection method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of modulation depth and periodic phase according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the third process of the Bosch process etching endpoint detection method according to an embodiment of the present invention; Figure 6 This is a schematic diagram of light intensity timing data under an ideal state according to an embodiment of the present invention; Figure 7 This is a schematic diagram of light intensity timing data during the Bosch etching process according to an embodiment of the present invention; Figure 8 This is a schematic diagram of the Bosch process cycle modulation depth variation curve according to an embodiment of the present invention; Figure 9 This is a schematic diagram of the periodic phase change curve of the Bosch process cycle according to an embodiment of the present invention. Figure 10 This is a structural block diagram of a Bosch process etching endpoint detection device according to an embodiment of the present invention; Figure 11 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] As an optional application scenario of this invention, such as Figure 1 As shown, the external process equipment 110 is communicatively connected to the server 120 to achieve Bosch process etching endpoint detection. The external process equipment 110 can be an etching device used to output a gas switching signal triggered by it. The server 120 receives the gas switching signal and determines multiple consecutive Bosch process cycles based on the signal. Simultaneously, it acquires the light intensity timing data corresponding to each Bosch process cycle and then detects the etching endpoint of the Bosch process based on the light intensity timing data. The server 120 can be any type of computing system or server capable of providing computing power, including but not limited to mainframes, edge computing nodes, and computing devices in cloud environments.

[0028] It should be noted that, Figure 1 This is merely an example of an application scenario and does not limit the scope of protection of this invention.

[0029] In conventional etching processes, etching is achieved by introducing etching gas (such as SF6) and passivation gas (such as C4F8). Specifically, under the action of a high-energy electric field, the etching gas and passivation gas are excited into plasma, and the active particles in the plasma are responsible for etching or passivation. Among them, the plasma excited by the etching gas SF6 generates specific characteristic spectral lines (such as SiF characteristic spectral lines) with the silicon material.

[0030] When etching reaches its endpoint, the light intensity of the specific characteristic spectral line usually shows a significant decreasing trend. Correlation endpoint detection technology determines the etching endpoint by monitoring the light intensity change of the specific characteristic spectral line (such as the SiF characteristic spectral line). However, correlation endpoint detection technology is usually based on the premise that the plasma is in a quasi-steady state or a slowly changing state, and its endpoint characteristics generally manifest as a monotonic change or trend change in the light intensity of the specific characteristic spectral line over a long time scale.

[0031] This application proposes a method for detecting the end point of Bosch etching process. The core idea is to treat the gas switching signal from external process equipment (such as etching equipment) as a periodic excitation u(t) of the system, where u(t) is a square wave with a fixed frequency. The characteristic spectral intensity signal I(t) corresponding to the target reactive species generated by the reaction of plasma excited by the etching gas with silicon material, obtained through spectral monitoring, is considered the system output. During etching, due to the presence of silicon material, the system has a response gain K>0 for the etching step time window, and the output characteristic intensity signal exhibits modulation synchronized with the gas switching excitation signal. When etching approaches or reaches its end point, the silicon material is gradually consumed, the system response gain K approaches zero, and the output intensity signal degenerates into random fluctuations mainly composed of background noise, unrelated to the gas switching excitation signal. Based on this physical mechanism, this application further extracts the modulation depth and periodic phase change characteristics of the characteristic light intensity signal to characterize the response amplitude and response timing stability of the characteristic light intensity signal to the gas switching signal, respectively. It describes the real etching process under the Bosch process from two dimensions: response intensity and response synchronization, thereby ensuring that the endpoint detection basis is consistent with the actual physical process and effectively improving the accuracy, stability and reliability of Bosch process etching endpoint detection.

[0032] According to an embodiment of the present invention, a method for detecting the etching endpoint of a Bosch process is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] This embodiment provides a method for detecting the etching endpoint of a Bosch process, which can be used in the aforementioned mobile server. Figure 2This is a flowchart of the Bosch process etching endpoint detection method according to an embodiment of the present invention, as follows: Figure 2 As shown, the process includes the following steps: Step S201: Collect characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle to obtain light intensity time sequence data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by the gas switching signal triggered by the external process.

[0034] Optionally, the gas switching signal triggered by the external process refers to the process control signal output by the controller of the external process equipment (such as etching equipment), which is used to control the alternating switching of the etching gas in the etching stage and the passivation gas in the passivation stage according to a preset timing sequence. This gas switching signal is used to characterize the switching timing sequence of each step in the Bosch process cycle and can be used to determine the cycle boundary and target frequency corresponding to each Bosch process cycle.

[0035] A Bosch process cycle refers to the execution cycle in the Bosch deep silicon etching process corresponding to one alternation between the etching and passivation stages. Its cycle boundary can be determined by the gas switching control signal. Characteristic light intensity signals refer to the light intensity information corresponding to the characteristic spectral lines of the target reactive species generated during the etching stage when the plasma excited by the etching gas reacts with the silicon material, such as the light intensity information corresponding to the characteristic spectral lines of SiF. Light intensity time-series data refers to the time-series data obtained by continuously acquiring characteristic light intensity signals at a preset sampling frequency in each Bosch process cycle of the Bosch deep silicon etching process. It consists of light intensity values ​​corresponding to multiple sampling times arranged in chronological order, and can characterize the dynamic process of characteristic light intensity signals changing with the Bosch process cycle.

[0036] For example, by acquiring the gas switching signal output by the etching equipment controller, the switching time between the etching stage and the passivation stage is identified, thereby determining the start time, end time, and cycle length corresponding to each Bosch process cycle. For each Bosch process cycle of the Bosch deep silicon etching process, the target characteristic spectral lines are detected using a spectral acquisition device, and the light intensity value at a preset wavelength is extracted to obtain the characteristic light intensity signals corresponding to multiple preset times within the Bosch process cycle; by aggregating the characteristic light intensity signals in chronological order, the light intensity time-series data of the Bosch process cycle can be obtained.

[0037] Step S202: Determine the target frequency value based on the Bosch process cycle.

[0038] Optionally, in this embodiment, the target frequency value refers to the periodic frequency corresponding to the Bosch process cycle, used to characterize the process switching frequency formed by the alternating switching of etching gas and passivation gas. Since the characteristic light intensity signal undergoes periodic changes driven by the Bosch process cycle, the light intensity time series data contains frequency components corresponding to the Bosch process cycle.

[0039] For example, suppose the Bosch process cycle is... Then the target frequency value F is For example, when the Bosch process cycle is 50ms, the corresponding target frequency value is 20Hz; when the Bosch process cycle is 40ms, the corresponding target frequency value is 25Hz.

[0040] Step S203: Within a continuously preset number of Bosch process cycles, determine the light intensity timing data corresponding to each Bosch process cycle.

[0041] For multiple consecutive Bosch process cycles, determine the light intensity timing data corresponding to each Bosch process cycle.

[0042] Step S204: Extract the periodic phase corresponding to the target frequency component in each light intensity time series data to obtain the periodic phase corresponding to each Bosch process cycle; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase characterizes the relative position of the target frequency component in the corresponding Bosch process cycle.

[0043] The periodic phase is used to characterize the time offset or relative position of the target frequency component in the light intensity timing data within each Bosch process cycle. Essentially, it reflects the time offset of the periodic change in the characteristic light intensity signal relative to the Bosch process cycle switching sequence defined by the gas switching signal. Since the Bosch process cycle is determined by the periodic alternation of etching and passivation gases, it has a stable time reference. Therefore, the periodic component in the light intensity timing data synchronized with this process switching sequence can be represented as a fixed-frequency periodic signal. Its phase parameter characterizes the time offset of this periodic signal relative to the start of the Bosch process cycle. This time offset corresponds to the relative position of the target frequency component within a Bosch process cycle, thus characterizing the timing of the response of the characteristic light intensity signal during the gas switching cycle (Bosch process cycle). During the normal etching stage, as the etching reaction continues and remains stable, the characteristic light intensity signal can stably follow the changes in the gas switching signal, so that the periodic phase corresponding to each Bosch process cycle remains basically consistent. However, as the etching end point approaches, the amount of material that can participate in the etching reaction gradually decreases, the response of the characteristic light intensity signal to the gas switching weakens and a timing drift occurs, resulting in changes or fluctuations in the periodic phase.

[0044] In this embodiment, the periodic phase corresponding to the target frequency component is extracted for multiple consecutive Bosch process cycles. This periodic phase is used to characterize the relative position of the target frequency component in the Bosch process cycle, that is, to characterize the time offset relationship between the periodic change of the characteristic light intensity signal and the switching sequence of the Bosch process cycle. By extracting the periodic phase corresponding to each Bosch process cycle, the synchronization state between the characteristic light intensity signal and the Bosch process cycle can be obtained, providing a basis for subsequent etching endpoint detection based on periodic phase changes. Furthermore, joint analysis of multiple consecutive Bosch process cycles can more stably reflect the dynamic characteristics of the characteristic light intensity signal changing with the etching process, providing a reliable data foundation for subsequent Bosch process etching endpoint detection based on the amount of periodic phase change and modulation depth.

[0045] Step S205: Based on the time-series data of each light intensity, determine the modulation depth of each Bosch process cycle. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal.

[0046] In the Bosch process, etching gases (such as SF6) and passivation gases (such as C4F8) are alternately switched at a fixed period, causing the plasma state to change periodically with the process cycle. This results in periodic fluctuations in the intensity of the target characteristic spectral lines (i.e., the characteristic light intensity signal). For example, during the normal etching stage, due to the active etching reaction, the characteristic light intensity signal responds significantly to the gas switching. The difference in characteristic light intensity signals between the etching and passivation stages is significant, resulting in large periodic fluctuations and a high modulation depth. As the etching process approaches its end, the amount of silicon material available for the etching reaction decreases, the etching reaction weakens, and the difference in characteristic light intensity signals between the etching and passivation stages gradually decreases. Consequently, the response amplitude of the characteristic light intensity signal to the gas switching decreases, and the modulation depth gradually decreases.

[0047] Therefore, modulation depth can be understood as the degree of periodic fluctuation of the characteristic light intensity signal caused by the alternating switching of etching gas and passivation gas within the Bosch process cycle. It can reflect the strength of the response of the characteristic light intensity signal to the gas switching signal and the activity level of the current etching reaction, and can be used to assist in the detection of the etching endpoint of the Bosch deep silicon etching process.

[0048] Step S206: Determine the corresponding periodic phase change between each adjacent Bosch process cycle.

[0049] During the normal etching phase, the etching reaction is continuous and stable, and the characteristic light intensity signal can stably follow the gas switching signal, ensuring that the periodic phase of each Bosch process cycle remains basically consistent. However, near the end of the etching process, as the amount of material available for the etching reaction gradually decreases, the response of the characteristic light intensity signal to the gas switching weakens and timing drift occurs, leading to changes or fluctuations in the periodic phase. In other words, the amount of periodic phase change characterizes the synchronization stability between the characteristic light intensity signal and the Bosch process cycle. Therefore, the amount of periodic phase change between adjacent Bosch process cycles determined in this embodiment can characterize the degree of change in the timing position of the characteristic light intensity signal response between adjacent Bosch process cycles, that is, reflect the stability change of the synchronization relationship between the characteristic light intensity signal and the gas switching signal. Simultaneously, the modulation depth can be understood as the degree of periodic fluctuation of the characteristic light intensity signal caused by the alternating switching of etching and passivation gases within the Bosch process cycle, reflecting the strength of the characteristic light intensity signal's response to the gas switching signal and the activity level of the current etching reaction.

[0050] By simultaneously analyzing the periodic phase change between adjacent Bosch process cycles and the modulation depth of each Bosch process cycle, the dynamic changes of the plasma state during the Bosch etching process can be characterized from two dimensions: response timing stability (characterized by the periodic phase change) and response amplitude (characterized by the modulation depth). This allows for the depiction of the actual etching process under the Bosch process, ensuring that the endpoint detection criteria are consistent with the actual physical process, thereby improving the accuracy, stability, and reliability of Bosch process etching endpoint detection.

[0051] Step S207: When the phase change of each cycle is greater than the preset phase stability threshold and the modulation depth is less than the preset modulation depth threshold, it is determined that the Bosch process etching endpoint has been reached.

[0052] The etching process is considered complete when, within N consecutive Bosch process cycles, the modulation depth is simultaneously less than a preset modulation depth threshold and the periodic phase change is greater than a preset phase stability threshold. Each modulation depth can be denoted as... The periodic phase change between adjacent Bosch process cycles can be expressed as: .

[0053] in, This represents the periodic phase corresponding to the i-th Bosch process cycle. The period phase is the period phase corresponding to the (i-1)th Bosch process cycle adjacent to the i-th Bosch process cycle.

[0054] For example, within N consecutive Bosch process cycles, where N can be set to 3 to 5, the Bosch etching process is determined to have reached its etching endpoint if the following conditions are met simultaneously: (1) Modulation depth of each detection cycle All are less than the preset modulation depth threshold , where the modulation depth threshold The preferred value is 0.05 to 0.1.

[0055] (2) Phase change in each period All are greater than the preset phase stabilization threshold. Or, the phase of each Bosch process cycle exhibits irregular fluctuations; among which, the phase stability threshold... It can be set from 20° to 45°, preferably 30°.

[0056] Modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal and is highly sensitive to changes in etching reaction intensity. However, in actual detection, it may be affected by factors such as background light intensity drift (e.g., baseline drift) and spectral line intensity fluctuations, leading to misjudgments based on a single modulation depth criterion. The periodic phase corresponding to the Bosch process cycle characterizes the synchronization relationship between the characteristic light intensity signal and the Bosch process cycle. Its changes are highly sensitive to the time offset between the characteristic light intensity signal and the gas switching signal. However, under the influence of plasma transient disturbances or signal acquisition noise, short-term phase fluctuations may occur, affecting the stability of the single periodic phase criterion. Specifically, if only the modulation depth decreases while the periodic phase remains stable, it may correspond to non-etching endpoint states such as changes in optical measurement conditions or fluctuations in the source intensity of the etching equipment; if only the periodic phase fluctuates while the modulation depth does not change significantly, it may correspond to short-term noise disturbances or unstable signal acquisition.

[0057] In some alternative implementations, the phase stabilization threshold is set to 20° to 45°.

[0058] Optionally, the phase stabilization threshold in this embodiment is set to 20° to 45°. In practical applications, the phase stabilization threshold can be adaptively adjusted according to the Bosch process cycle length, the signal-to-noise ratio (SNR) of the spectral signal, and the current process state. The adjustment range can be 15° to 50°. For example, for a typical Bosch deep silicon etching process with a process cycle length of 100 ms and an SNR of approximately 10 dB, the phase stabilization threshold can be set to 30°. When the spectral signal SNR is low and the phase fluctuation increases, the phase stabilization threshold can be appropriately increased to 35° to 45° to enhance noise immunity. When the spectral signal SNR is high, or when the endpoint characteristics become more apparent as the etching process progresses, the phase stabilization threshold can be appropriately decreased to 15° to 25° to improve the sensitivity of etching endpoint detection.

[0059] Furthermore, this embodiment can also employ an adaptive threshold strategy to determine the phase stability threshold. Specifically, a predetermined number (e.g., 20) of consecutive Bosch process cycles can be selected as statistical samples during the normal etching phase, and the periodic phase change corresponding to each adjacent Bosch process cycle can be calculated. mean and standard deviation and in accordance with A phase stability threshold is determined, where the mean and standard deviation are used to characterize the statistical distribution range of periodic phase changes under normal etching conditions. Subsequently, as the deep silicon etching process continues, the above statistical results (i.e., the mean) can be further analyzed. and standard deviation The system updates the phase stability threshold and adjusts it dynamically to adapt to different process conditions, signal-to-noise ratios, and phase fluctuation characteristics under different etching states, thereby improving the accuracy of Bosch process etching endpoint detection and environmental adaptability.

[0060] In this embodiment, the modulation depth is used to characterize the response amplitude of the characteristic light intensity signal to the gas switching signal, reflecting the activity level of the current etching reaction; the periodic phase change is used to characterize the stability of the characteristic light intensity signal's response to the Bosch process cycle switching timing between adjacent Bosch process cycles, reflecting the change in the synchronization relationship between the characteristic light intensity signal and the Bosch process cycle. During the normal etching stage, the characteristic light intensity signal has a strong response to the gas switching signal, a high modulation depth, and the periodic phase corresponding to adjacent Bosch process cycles remains relatively stable. As the etching process gradually approaches its end, the response of the characteristic light intensity signal to the gas switching gradually weakens, the modulation depth decreases, and the synchronization relationship between the characteristic light intensity signal and the Bosch process cycle changes, leading to a decrease in periodic phase stability. Therefore, by jointly analyzing the modulation depth and the periodic phase change between adjacent Bosch process cycles, the etching state can be characterized from two dimensions: response amplitude and response stability, improving the accuracy and reliability of Bosch process etching end detection.

[0061] Based on this, the present invention adopts a joint determination mechanism based on modulation depth and periodic phase, and introduces a continuous constraint of N consecutive Bosch process cycles. Only when the modulation depth is less than the modulation depth threshold and the instability of the synchronization relationship represented by the periodic phase change continues to exist within N consecutive cycles, the Bosch process etching process is determined to have reached the etching endpoint. This can reduce the risk of misjudgment in the etching endpoint detection during the Bosch process etching process and improve the accuracy and reliability of the etching endpoint detection.

[0062] Compared to other endpoint detection technologies that rely solely on the intensity variation trend of characteristic spectral lines, which can lead to the weak trend changes corresponding to the etching endpoint being masked by periodic modulation components, the Bosch process etching endpoint detection method provided in this embodiment can effectively extract the dynamic response features during the periodic modulation process of the Bosch process, reduce the interference of periodic gas switching on endpoint feature recognition, and improve the ability to identify and distinguish the etching endpoint state, thereby improving the accuracy and reliability of Bosch process etching endpoint detection.

[0063] The Bosch process etching endpoint detection method provided in this embodiment constructs light intensity time-series data based on the characteristic light intensity signals corresponding to each Bosch process cycle, and extracts the modulation depth corresponding to each Bosch process cycle and the periodic phase change between adjacent Bosch process cycles from the light intensity time-series data. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal, and the periodic phase change characterizes the stability of the synchronization relationship between the characteristic light intensity signal and the Bosch process cycle. This approach characterizes the actual etching process under the Bosch process from two dimensions: response intensity and response synchronization relationship. It ensures that the detection basis for the etching endpoint matches the actual etching physical process, reduces the interference of periodic gas switching on endpoint feature recognition, improves the ability to identify and distinguish the etching endpoint state, and thus improves the accuracy of Bosch process etching endpoint detection.

[0064] This embodiment provides a method for detecting the etching endpoint of a Bosch process, which can be used in the aforementioned mobile server. Figure 3 This is a flowchart of the Bosch process etching endpoint detection method according to an embodiment of the present invention, as follows: Figure 3 As shown, the process includes the following steps: Step S301: Collect characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle to obtain light intensity time-series data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by a gas switching signal triggered by an external process. For details, please refer to... Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0065] Step S302: Determine the target frequency value based on the Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0066] Step S303: Within a predetermined number of consecutive Bosch process cycles, determine the light intensity timing data corresponding to each Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0067] Step S304: Extract the periodic phase corresponding to the target frequency component from each light intensity time series data to obtain the periodic phase corresponding to each Bosch process cycle; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase characterizes the relative position of the target frequency component in the corresponding Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S204 of the illustrated embodiment will not be described again here.

[0068] Step S305: Based on the time-series data of each light intensity, determine the modulation depth of each Bosch process cycle. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal.

[0069] Specifically, the Bosch process cycle includes an etching window and a passivation window; step S304 above includes: Step S3051: For each Bosch process cycle, determine the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window within the Bosch process cycle.

[0070] The Bosch process cycle includes an etching window and a passivation window, where the etching window corresponds to the etching gas introduction stage, and the passivation window corresponds to the passivation gas introduction stage. The Bosch process cycle is determined based on an externally triggered gas switching signal; specifically, the start time of each Bosch process cycle is precisely defined according to this gas switching signal. Etching window (SF6 etching gas introduction period), passivation window (The time period during which the passivating gas C4F8 is introduced) and the end time. For example, consider one Bosch process cycle. The etching step takes [time]. Then etch the window for: .

[0071] Passivation window for: .

[0072] Based on the gas switching signal, the start and end times of each Bosch process cycle are precisely determined, and the etching and passivation windows within that Bosch process cycle are further divided, thus forming a time reference that strictly corresponds to the gas switching signal of the external process. By establishing this time reference, subsequent analysis of characteristic light intensity signals is performed within the process cycle framework synchronized with the gas switching signal. This ensures that the extraction of modulation depth and cycle phase is based on a clear gas switching signal excitation sequence, achieving precise alignment of light intensity timing data with the Bosch process cycle. This provides a reliable data foundation for subsequent etching endpoint detection based on modulation depth and cycle phase.

[0073] First light intensity time-series data are acquired within the etching window, and second light intensity time-series data are acquired within the passivation window. Both the first and second light intensity time-series data are time-series data obtained by continuously sampling the characteristic light intensity signal based on a preset sampling frequency. Optionally, a spectroscopic detection device, such as an AOTF-OES (Acousto-Optic Tunable Filter - Optical Emission Spectrometer), is configured to continuously acquire the characteristic light intensity signal corresponding to the target characteristic spectral line (e.g., SiF 440nm). Within a single Bosch process cycle, the spectroscopic detection device acquires the characteristic light intensity signal at a sampling frequency higher than the frequency corresponding to the Bosch process cycle, thereby forming the light intensity time-series data I(t). The sampling process is synchronized with the etching and passivation windows determined by the gas switching signal to ensure that the acquired light intensity time-series data can accurately characterize the dynamic change process of the characteristic light intensity signal within the Bosch process cycle. For example, multiple sampling times are evenly distributed or set according to a preset strategy within each Bosch process cycle to ensure a sufficient number of sampling points are obtained within both the etching and passivation windows. For instance, at least 5 sampling points are set within each window to improve the stability of the light intensity statistics within the window and to provide a reliable data basis for subsequent modulation depth calculation and periodic phase extraction.

[0074] By using the above method, segmented sampling of the same characteristic light intensity signal is achieved in different process windows (i.e., etching window and passivation window) to characterize the light intensity change characteristics corresponding to the etching stage and passivation stage respectively, providing basic data support for subsequent modulation depth calculation and periodic phase analysis.

[0075] Step S3052: Based on the degree of change between the first light intensity timing data and the second light intensity timing data, determine the modulation depth of the Bosch process cycle.

[0076] In some optional implementations, step S3052 above includes: Step a1: Determine the first light intensity average value corresponding to the first light intensity time series data, and determine the second light intensity average value corresponding to the second light intensity time series data.

[0077] Step a2: Normalize the first average light intensity and the second average light intensity to obtain the modulation depth of the Bosch process cycle.

[0078] Statistical processing is performed on the light intensity values ​​corresponding to the characteristic light intensity signals at each preset time (each sampling point) in the first light intensity time series data to obtain the average first light intensity value corresponding to the etching window. It can be represented as ,in, The duration of the etching window, The time interval of the etching window within the i-th Bosch process cycle; simultaneously, the light intensity values ​​corresponding to the characteristic light intensity signals at each preset time (each sampling point) in the second light intensity timing data are statistically processed to obtain the average second light intensity corresponding to the passivation window. It can be represented as , To passivate the window duration, Let be the time interval of the passivation window within the i-th Bosch process cycle. The first average light intensity and the second average light intensity can characterize the light intensity level of the corresponding process window.

[0079] Optionally, the modulation depth of the Bosch process cycle is obtained by normalizing the first average light intensity and the second average light intensity. This allows for the quantification of the relative light intensity difference between the etched and passivated windows, and the modulation depth. It can be determined by the following formula: .

[0080] Modulation depth The response amplitude of the characteristic light intensity signal to the gas switching signal within the Bosch process cycle was quantified, thus characterizing the activity level of the etching reaction within the Bosch process cycle. As the etching process nears its end, the response of the characteristic light intensity signal to the gas switching weakens, the light intensity difference between the etching window and the passivation window gradually decreases, and correspondingly, the modulation depth gradually decreases. For example, the modulation depth... A value approaching 1 indicates a strong response of the characteristic light intensity signal to gas switching, suggesting that the etching reaction is still active, typically corresponding to a non-terminal state; modulation depth A value approaching 0 indicates a weakening or absence of response from the characteristic light intensity signal to gas switching, suggesting that the etching reaction is gradually decaying and the etching process is nearing its end. Modulation depth The difference in characteristic light intensity signals between the etching window and the passivation window is represented by a normalized form. Compared with the direct use of absolute light intensity values, it is not affected by factors such as spectrometer drift, optical attenuation, or changes in detector sensitivity. It can stably reflect the strength of the response of the characteristic light intensity signal to the gas switching signal, thus more reliably characterizing the activity of the etching reaction and ensuring the accuracy of the etching endpoint determination in the Bosch process.

[0081] In this embodiment, the first average light intensity corresponding to the etching window and the second average light intensity corresponding to the passivation window are used to characterize the overall light intensity level of the etching window and the passivation window, thereby reducing the impact of random noise, transient fluctuations and single-point outliers on the detection results and improving the stability of light intensity feature extraction. Furthermore, by normalizing the first average light intensity and the second average light intensity, the modulation depth is obtained. The relative light intensity difference between the etching window and the passivation window can quantify the response amplitude of the feature light intensity signal to the gas switching signal. Even if there is absolute light intensity drift caused by factors such as optical window contamination, detector sensitivity changes, and fluctuations in the overall plasma luminescence intensity, the modulation depth can still maintain sensitivity to changes in the etching state, thereby more accurately characterizing the activity level of the current etching reaction.

[0082] The Bosch process etching endpoint detection method provided in this embodiment determines the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window for each Bosch process cycle. This allows for the separation and characterization of characteristic light intensity signals corresponding to different process stages within the same Bosch process cycle, thereby accurately reflecting the dynamic changes in characteristic light intensity signals under the alternating action of etching and passivation gases. Furthermore, by determining the modulation depth corresponding to the Bosch process cycle based on the degree of change between the first and second light intensity timing data, the response strength of the characteristic light intensity signal to the gas switching signal can be quantified. Therefore, the modulation depth can characterize the activity level of the current etching reaction and transform the dynamic response characteristics during the Bosch process cycle switching process into a quantifiable modulation depth parameter, providing a reliable basis for subsequent etching endpoint detection, thereby improving the accuracy and stability of Bosch process etching endpoint detection.

[0083] Step S306: Determine the corresponding cycle phase change between each adjacent Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S306 of the illustrated embodiment will not be described again here.

[0084] Step S307: When the phase change in each cycle is greater than a preset phase stability threshold, and the modulation depth is less than a preset modulation depth threshold, the Bosch process etching endpoint is determined to have been reached. For details, please refer to [link to relevant documentation]. Figure 2 Step S207 of the illustrated embodiment will not be described again here.

[0085] In some alternative implementations, such as Figure 4The diagram illustrates the modulation depth and periodic phase. The horizontal axis represents the Bosch process cycle number, the left vertical axis represents the modulation depth, and the right vertical axis represents the periodic phase. Near the end of the etching process, the modulation depth gradually decreases and remains low below the modulation depth threshold, indicating a significant weakening of the response amplitude of the characteristic light intensity signal to the gas switching signal. Simultaneously, during the normal etching phase, the periodic phase remains relatively stable with only minor fluctuations, indicating a relatively stable synchronization between the characteristic light intensity signal and the Bosch process cycle. However, near the end of the etching process, the periodic phase exhibits significant irregular jumps or increased fluctuations, indicating instability in the synchronization between the characteristic light intensity signal and the gas switching timing. By jointly observing the modulation depth and periodic phase, the evolution of the etching process can be characterized from two dimensions: response amplitude and response timing stability, thereby enabling reliable identification of the Bosch process etching endpoint.

[0086] In this embodiment, a joint determination mechanism based on modulation depth and periodic phase is introduced, and a continuous constraint of N consecutive Bosch process cycles is introduced. Specifically, the Bosch process etching process is determined to have reached the etching endpoint only when the phase change of each cycle is greater than a preset phase stability threshold, and the phenomenon that the modulation depth is less than a preset modulation depth threshold persists for N consecutive cycles. This can avoid short-term phase fluctuations that could affect the stability of the single-cycle phase criterion or cause misjudgment of the single modulation depth criterion due to factors such as background light intensity drift (e.g., baseline drift). This improves the reliability of endpoint detection and reduces the risk of misjudgment.

[0087] The Bosch process etching endpoint detection method provided in this embodiment determines the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window for each Bosch process cycle. It can separate and characterize the characteristic light intensity signals corresponding to different process stages within the same Bosch process cycle, thereby accurately reflecting the dynamic change process of the characteristic light intensity signals under the alternating action of etching gas and passivation gas. Furthermore, by determining the modulation depth corresponding to the Bosch process cycle based on the degree of change between the first and second light intensity time-series data, the response strength of the characteristic light intensity signal to the gas switching signal can be quantified, providing a reliable basis for subsequent etching endpoint detection, thereby improving the accuracy and stability of Bosch process etching endpoint detection. The periodic phase change between adjacent Bosch process cycles can characterize the degree of change in the timing position of the characteristic light intensity signal response between adjacent Bosch process cycles, that is, reflect the stability change of the synchronization relationship between the characteristic light intensity signal and the gas switching signal. Based on this, this application adopts a joint determination mechanism based on modulation depth and periodic phase, and introduces a continuous constraint of N consecutive Bosch process cycles. Only when the phenomenon that the modulation depth is less than the modulation depth threshold and the periodic phase change is greater than the modulation depth threshold persists for N consecutive cycles is the Bosch process etching process determined to have reached the etching endpoint. This can reduce the risk of misjudgment in etching endpoint detection during the Bosch process etching process and improve the accuracy and reliability of etching endpoint detection.

[0088] This embodiment provides a method for detecting the etching endpoint of a Bosch process, which can be used in the aforementioned mobile server. Figure 5 This is a flowchart of the Bosch process etching endpoint detection method according to an embodiment of the present invention, as follows: Figure 5 As shown, the process includes the following steps: Step S501: Collect characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle to obtain light intensity time-series data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by a gas switching signal triggered by an external process. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0089] In some alternative embodiments, such as Figure 6 The diagram shown is a schematic representation of the light intensity timing data under ideal conditions in this embodiment, illustrating the correspondence between the Bosch process cycle, the gas switching control signal, and the characteristic light intensity signal, as well as their variation characteristics before and after the etching endpoint. Figure 6The horizontal axis represents time, and the upper curve shows the gas switching signal of the Bosch process and its corresponding periodic structure. A high level corresponds to the introduction of SF6 etching gas, and a low level corresponds to the introduction of C4F8 passivation gas, thus forming a periodic square wave control signal. For example, the etching window (SF6 stage) is from 0ms to 50ms, and the passivation window (C4F8 stage) is from 50ms to 100ms, thus forming a repeating Bosch process cycle. At the same time, the upper curve also shows the trigger pulse (gas switching signal) output by the external process controller, which is used to identify the start or switching time of each Bosch process cycle.

[0090] The central curve represents the characteristic light intensity signal I(t) acquired by the spectral acquisition device. Before the etching endpoint, due to the periodic alternation of etching and passivation gases, the plasma state changes synchronously with the Bosch process cycle, causing the characteristic light intensity signal to exhibit periodic modulation characteristics consistent with the Bosch process cycle, i.e., a square wave modulated signal that fluctuates with the time period. However, after etching approaches or reaches its endpoint, as the amount of material available to participate in the etching reaction decreases, the response of the etching reaction to gas switching weakens, and the periodic modulation characteristics of the characteristic light intensity signal gradually disappear, transforming into a low-amplitude, approximately flat noise-like fluctuation signal.

[0091] The lower schematic diagram shows the sampling time when the characteristic light intensity signal is continuously sampled at a preset sampling frequency. This is used to construct the light intensity time series data corresponding to each Bosch process cycle and to provide a basic data source for subsequent modulation depth calculation and phase extraction of the target frequency component.

[0092] Step S502: Determine the target frequency value based on the Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0093] Step S503: Within a predetermined number of consecutive Bosch process cycles, determine the light intensity timing data corresponding to each Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0094] Step S504: Extract the periodic phase corresponding to the target frequency component in each light intensity time series data to obtain the periodic phase corresponding to each Bosch process cycle; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase characterizes the relative position of the target frequency component in the corresponding Bosch process cycle.

[0095] Specifically, step S504 includes: Step S5041: Corresponding to each Bosch process cycle, perform frequency analysis on the light intensity time series data corresponding to the Bosch process cycle to obtain the target frequency component in the light intensity time series data whose frequency is the target frequency value.

[0096] For each Bosch process cycle, frequency analysis is performed on the light intensity time-series data acquired within that Bosch process cycle to extract periodic response components corresponding to the Bosch process cycle, thus obtaining the target frequency component. The target frequency component corresponds to the periodic signal component in the light intensity time-series data that is synchronized with the process cycle defined by the gas switching signal, and its frequency is the reciprocal of the Bosch process cycle.

[0097] Step S5042: Determine the phase of the target frequency component at the midpoint of the Bosch process cycle to obtain the periodic phase.

[0098] The phase value of the target frequency component at the midpoint of the corresponding Bosch process cycle is determined and used as the periodic phase of that cycle. Optionally, to unify the phase sampling reference between different Bosch process cycles, achieve comparability of phase characteristics between different Bosch process cycles, and eliminate phase deviations introduced by incomplete alignment of cycle start boundaries, this embodiment can select the midpoint of each Bosch process cycle as the phase sampling reference and use the phase value corresponding to that moment as the periodic phase of that cycle.

[0099] For example, the Fourier transform method can be used to process the light intensity time series data, extract the target frequency component, and determine its phase value according to the complex representation of the target frequency component; or the phase-locked tracking method can be used to perform phase synchronization tracking of the light intensity time series data with the target frequency as a reference to obtain the corresponding real-time phase information.

[0100] The Bosch process etching endpoint detection method provided in this embodiment uses periodic phase to characterize the relative position of the target frequency component within the Bosch process cycle, i.e., to characterize the time offset relationship between the periodic change of the characteristic light intensity signal and the Bosch process cycle switching sequence. By extracting the periodic phase corresponding to each Bosch process cycle, the synchronization state between the characteristic light intensity signal and the Bosch process cycle can be obtained, providing a basis for etching endpoint detection based on periodic phase changes. By determining the phase corresponding to the midpoint of each Bosch process cycle as the periodic phase, cross-cycle comparisons of the periodic phases of each Bosch process cycle can be made, thereby effectively characterizing the change in the synchronization relationship between the characteristic light intensity signal and the gas switching signal. At the same time, it can avoid phase calculation deviations caused by unstable process cycle boundaries or asynchronous sampling frequencies, improving the accuracy of periodic phase determination and thus enhancing the precision of etching endpoint detection.

[0101] In some optional implementations, step S504 may further include: Step b1 involves using a digital phase-locked loop (PLL) with the target frequency value as the reference frequency to track the light intensity timing data for each Bosch process cycle, thereby obtaining the continuous phase corresponding to the target frequency component in the light intensity timing data.

[0102] Step b2: Determine the phase value corresponding to the midpoint of the Bosch process cycle as the cycle phase.

[0103] For example, for the light intensity timing data corresponding to each Bosch process cycle, a digital phase-locked loop (PLL) is used to perform phase-locked tracking with the target frequency value as the reference frequency to obtain the continuous phase corresponding to the target frequency component in the light intensity timing data. PLL tracking can extract the temporal position changes of the characteristic light intensity signal within a continuous cycle, reflecting the synchronous stability of the etching reaction to the gas switching timing. Subsequently, the phase value corresponding to the midpoint of each Bosch process cycle is selected as the periodic phase of that cycle. This periodic phase quantifies the response time position of the light intensity signal to the process cycle switching timing and can be used for cross-cycle comparison to characterize the stability and synchronicity of the etching process.

[0104] The Bosch process etching endpoint detection method provided in this embodiment uses a digital phase-locked loop to perform phase-locked tracking of light intensity timing data to obtain the continuous phase corresponding to the target frequency component. The phase value corresponding to the midpoint of the Bosch process cycle is determined as the periodic phase, thereby realizing the stable expression of the periodic phase of the Bosch process cycle and improving the comparability of cross-cycle phases and the robustness of etching endpoint detection.

[0105] In some optional implementations, step S504 may further include: Step c1, corresponding to each Bosch process cycle, performs Fourier transform processing on the light intensity time series data corresponding to the Bosch process cycle to obtain the target frequency component in the light intensity time series data with the target frequency value.

[0106] Step c2: Determine the periodic phase based on the phase value of the target frequency component.

[0107] For each Bosch process cycle, the light intensity time-series data within that Bosch process cycle is subjected to Fourier transform processing to extract the target frequency component corresponding to the Bosch process cycle from the frequency domain. The frequency of the target frequency component is the reciprocal of the Bosch process cycle, and the phase is determined based on the complex representation of the target frequency component. Optionally, to obtain a periodic phase that can characterize the state of a single Bosch process cycle, for the i-th Bosch process cycle, the midpoint of that Bosch process cycle is used as a representative time, and the phase value of the target frequency component corresponding to that midpoint time is extracted and defined as the periodic phase corresponding to the i-th Bosch process cycle. Using the phase value corresponding to the midpoint of the cycle as the periodic phase can reduce the influence of transient fluctuations at the Bosch process cycle switching boundary on the phase measurement results and improve the stability and consistency of phase comparison between adjacent cycles.

[0108] The Bosch process etching endpoint detection method provided in this embodiment effectively filters out non-periodic noise interference and improves the stability of periodic phase extraction by using Fourier transform to extract the target frequency component and determine its phase.

[0109] Step S505: Based on the time-series data of each light intensity, determine the modulation depth of each Bosch process cycle. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal. For details, please refer to [link to relevant documentation]. Figure 3 Step S305 of the illustrated embodiment will not be described again here.

[0110] Step S506: Determine the corresponding cycle phase change between each adjacent Bosch process cycle. For details, please refer to [link to relevant documentation]. Figure 2 Step S306 of the illustrated embodiment will not be described again here.

[0111] Step S507: When the phase change in each cycle is greater than a preset phase stability threshold, and the modulation depth is less than a preset modulation depth threshold, the Bosch process etching endpoint is determined to have been reached. For details, please refer to [link to relevant documentation]. Figure 2 Step S207 of the illustrated embodiment will not be described again here.

[0112] The Bosch process etching endpoint detection method provided in this embodiment uses periodic phase to characterize the relative position of the target frequency component within the Bosch process cycle, i.e., to characterize the time offset relationship between the periodic change of the characteristic light intensity signal and the Bosch process cycle switching sequence. By extracting the periodic phase corresponding to each Bosch process cycle, the synchronization state between the characteristic light intensity signal and the Bosch process cycle can be obtained, providing a basis for etching endpoint detection based on periodic phase changes. By determining the phase corresponding to the midpoint of each Bosch process cycle as the periodic phase, cross-cycle comparisons of the periodic phases of each Bosch process cycle can be made, thereby effectively characterizing the change in the synchronization relationship between the characteristic light intensity signal and the gas switching signal. At the same time, it can avoid phase calculation deviations caused by unstable process cycle boundaries or asynchronous sampling frequencies, improving the accuracy of periodic phase determination and thus enhancing the precision of etching endpoint detection.

[0113] The modulation depth of each Bosch process cycle and the determination process of the corresponding periodic phase change between adjacent Bosch process cycles in this embodiment have low computational complexity, and can be implemented in real time on a low-cost processor (such as an FPGA), meeting the real-time requirements (millisecond-level response) of etching equipment for the endpoint signal. Furthermore, the quantitative comparison results between the Bosch process etching endpoint detection method proposed in this application and existing endpoint detection technologies (i.e., methods that determine the etching endpoint by monitoring the light intensity changes of specific characteristic spectral lines) are shown in the table below.

[0114]

[0115] Compared with existing endpoint detection technologies, the Bosch process etching endpoint detection method proposed in this application shows significant improvements in several key technical dimensions. First, the effective time resolution of this application can reach 100Hz or even 1kHz, which is approximately 5-50 times higher than the existing technology's resolution of less than 20Hz. This allows for the capture of high-frequency periodic modulation characteristics of the Bosch process, enabling real-time monitoring of rapid etching dynamics. Second, the minimum detectable aperture ratio of this application is as low as 0.3%, significantly better than the existing technology's detection limit of ≥2%, a reduction of approximately 7 times, thus accurately determining the endpoint even at the micro-etching feature stage. Furthermore, the interference of factors such as window contamination or equipment power drift on endpoint determination is significantly reduced in this application, improving robustness by more than 10 times; in contrast, existing methods directly rely on the light intensity threshold, making them susceptible to drift and prone to misjudgment. Regarding the misjudgment rate, this application achieves less than 0.5% under typical operating conditions, while existing technologies are typically 5-10%, representing a reduction of approximately 1-2 orders of magnitude. Furthermore, this application calculates the modulation depth and periodic phase in real time, with an algorithm latency of less than 1 ms, while existing technologies relying on software post-processing typically have a latency exceeding 100 ms, representing a two-order-of-magnitude improvement in response speed. Finally, the criterion of this application directly corresponds to the response gain K, possessing clear physical interpretability. Compared to existing methods that rely on empirically derived light intensity thresholds, this significantly improves the reliability and interpretability of the determination results. In summary, the Bosch process etching endpoint detection method proposed in this application outperforms existing technologies in terms of resolution, sensitivity, robustness, false positive rate, real-time performance, and physical interpretability.

[0116] As an exemplary embodiment, a typical Bosch deep silicon etching process is used for endpoint detection and verification. Specifically, the Bosch process cycle... The etching process takes 100 ms, with SF6 gas introduced for 50 ms during the etching stage and C4F8 gas introduced for 50 ms during the passivation stage. Ar is used as the auxiliary gas. The characteristic spectral line of SiF (440 nm) is selected as the monitoring object to characterize the characteristic light intensity signal during the etching reaction process. In this exemplary embodiment, the light intensity time series data corresponding to the characteristic light intensity signal before the etching endpoint can be expressed as follows: .in, The background light intensity represents the stable light intensity reference value outside the etching and passivation processes; S(t) represents the periodic excitation function composed of the gas switching signal, used to characterize the process switching timing of the Bosch process cycle. It represents the square wave modulation signal synchronized with the Bosch process cycle, taking a value of 1 in the SF6 etching stage and a value of 0 in the C4F8 passivation stage; A represents the periodic response amplitude generated by the etching reaction, which is a physical parameter of the signal; n(t) represents Gaussian white noise, used to simulate plasma fluctuations and spectral detection noise. After the etching endpoint is reached, due to the significant weakening or disappearance of the etching reaction, the characteristic light intensity signal mainly consists of the background light intensity and noise. The light intensity timing data can be expressed as follows: For example, setting A=0.2, used to simulate weak endpoint signal scenarios under low aperture ratio conditions; noise standard deviation The sampling frequency of the spectral acquisition device is 1 kHz.

[0117] like Figure 7 The diagram shows the timing data of light intensity during the Bosch etching process. Before the etching endpoint is reached, the characteristic light intensity signal in the timing data exhibits a clear periodic modulation characteristic synchronized with the Bosch process cycle. As the etching process approaches and reaches its endpoint, the periodic modulation component weakens significantly, and the characteristic light intensity signal gradually becomes a random noise signal fluctuating around the background light intensity. Furthermore, the modulation depth M is calculated for each Bosch process cycle. Figure 8 The diagram shows the variation curve of the periodic modulation depth in the Bosch process. Before reaching the etching endpoint, the modulation depth M remains at a relatively high and stable level (approximately 0.18). After the etching reaches its endpoint, due to the weakening of the etching reaction, the light intensity difference between the etching and passivation stages significantly decreases, and the modulation depth M rapidly decreases and remains in a low-value region close to the noise level. Simultaneously, a sliding window Fourier transform is used to extract the periodic phase corresponding to the target frequency component. Figure 9 The figure shows a schematic diagram of the periodic phase change curve of the Bosch process cycle. During the normal etching stage, the periodic phase remains relatively stable, indicating that there is a stable synchronization relationship between the characteristic light intensity signal and the Bosch process cycle. However, after the etching end point is reached, the periodic phase fluctuates significantly and loses stability due to the significant weakening of the periodic response component, indicating that the synchronization relationship between the characteristic light intensity signal and the gas switching timing (Bosch process cycle) degrades.

[0118] In the endpoint determination process, the modulation depth threshold is set to 0.05, the phase stability threshold to 45°, and the number of consecutive preset Bosch process cycles to 3. When three consecutive Bosch detection process cycles simultaneously satisfy the conditions that the modulation depth is lower than the modulation depth threshold and the phase change of the cycle exceeds the phase stability threshold, the etching is determined to have reached the endpoint. Experimental results show that this embodiment can accurately trigger the endpoint arrival signal shortly after the endpoint occurs, approximately at 15.3 seconds (i.e., about three cycles after the etching endpoint occurs).

[0119] The above embodiments demonstrate that the Bosch process etching endpoint detection method of this application can effectively extract weak time modulation feature signals in a strong noise background. (Approximately 20%), indicating its good adaptability to low signal-to-noise ratio conditions. Based on the joint criterion of modulation depth and periodic phase, it can effectively identify the endpoint state at the etching endpoint, demonstrating that this application has good distinguishing ability in both response amplitude and response timing stability dimensions. Furthermore, since the modulation depth and periodic phase are obtained based on normalization processing and frequency domain / phase feature extraction, and do not depend on the light intensity determination threshold, this method is effective for background light intensity. The drift has low sensitivity, which improves the adaptability to changes in the detection environment to a certain extent.

[0120] This embodiment also provides a Bosch process etching endpoint detection device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0121] This embodiment provides a device for detecting the etching endpoint in a Bosch process, such as... Figure 10 As shown, it includes: The acquisition module 1001 is used to acquire characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle, and obtain light intensity time sequence data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by the gas switching signal triggered by the external process.

[0122] The frequency determination module 1002 is used to determine the target frequency value based on the Bosch process cycle.

[0123] The timing data determination module 1003 is used to determine the light intensity timing data corresponding to each Bosch process cycle within a continuously preset number of Bosch process cycles.

[0124] The extraction module 1004 is used to extract the periodic phase corresponding to the target frequency component in each light intensity time series data to obtain the periodic phase corresponding to each Bosch process cycle; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase represents the relative position of the target frequency component in the corresponding Bosch process cycle.

[0125] The modulation depth determination module 1005 is used to determine the modulation depth of each Bosch process cycle based on the time-series data of each light intensity. The modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal.

[0126] The detection data determination module 1006 is used to determine the corresponding periodic phase change between each adjacent Bosch process cycle.

[0127] The detection module 1007 is used to determine the end point of the Bosch process etching when the phase change of each cycle is greater than the preset phase stability threshold and the modulation depth is less than the preset modulation depth threshold.

[0128] In some optional implementations, the Bosch process cycle includes an etching window and a passivation window; the modulation depth determination module 1005 is further configured to determine, for each Bosch process cycle, the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window; and to determine the modulation depth of the Bosch process cycle based on the degree of change between the first light intensity timing data and the second light intensity timing data.

[0129] In some optional implementations, the modulation depth determination module 1005 is further configured to determine the first light intensity average value corresponding to the first light intensity timing data, and to determine the second light intensity average value corresponding to the second light intensity timing data; and to normalize the first light intensity average value and the second light intensity average value to obtain the modulation depth of the Bosch process cycle.

[0130] In some optional implementations, the extraction module 1004 is further configured to perform frequency analysis on the light intensity time series data corresponding to each Bosch process cycle, to obtain the target frequency component in the light intensity time series data whose frequency is the target frequency value; and to determine the phase corresponding to the target frequency component at the midpoint of the Bosch process cycle, thereby obtaining the periodic phase.

[0131] In some optional implementations, the extraction module 1004 is also used to perform phase-locked tracking on the light intensity time series data corresponding to each Bosch process cycle using a digital phase-locked loop with the target frequency value as the reference frequency, to obtain the continuous phase corresponding to the target frequency component in the light intensity time series data; and to determine the phase value corresponding to the midpoint of the Bosch process cycle as the periodic phase.

[0132] In some optional implementations, the extraction module 1004 is further configured to perform Fourier transform processing on the light intensity time series data corresponding to each Bosch process cycle to obtain the target frequency component in the light intensity time series data with the target frequency value; and determine the period phase based on the phase value of the target frequency component.

[0133] In some optional implementations, the modulation depth threshold is set to 0.05-0.1.

[0134] In some alternative implementations, the phase stabilization threshold is set to 20° to 45°.

[0135] In some alternative implementations, the preset number of Bosch process cycles is 3 to 5.

[0136] The Bosch process etching endpoint detection device provided in this embodiment of the invention can execute the Bosch process etching endpoint detection method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0137] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0138] The following is a detailed reference. Figure 11 This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processing unit, etc.) 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory 1102 or a program loaded from a memory 1108 into a random access memory 1103. The read-only memory may be a ROM, and the random access memory may be RAM. The random access memory 1103 also stores various programs and data required for the operation of the electronic device. The processor 1101, the read-only memory 1102, and the random access memory 1103 are interconnected via a bus 1104. An input / output interface 1105 is also connected to the bus 1104.

[0139] Typically, the following devices can be connected to the input / output interface 1105: input devices 1106 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 1107 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; memory devices 1108 including, for example, magnetic tape, hard disk, etc.; and communication devices 1109. Communication device 1109 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 11 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0140] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 1109, or installed from a memory 1108 (the memory being magnetic tape, hard disk, etc.), or installed from a read-only memory 1102. When the computer program is executed by the processor 1101, it performs the functions defined in the Bosch process etching endpoint detection method of the embodiments of the present invention.

[0141] Figure 11 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0142] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the Bosch process etching endpoint detection method shown in the above embodiments is implemented.

[0143] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0144] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for detecting the etching endpoint in a Bosch process, characterized in that, The method includes: The characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle are collected to obtain light intensity time sequence data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by a gas switching signal triggered by an external process. Determine the target frequency value based on the Bosch process cycle; Within a continuously preset number of Bosch process cycles, determine the light intensity timing data corresponding to each Bosch process cycle; Extract the periodic phase corresponding to the target frequency component in each of the light intensity time series data to obtain the periodic phase corresponding to each of the Bosch process cycles; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase represents the relative position of the target frequency component in the corresponding Bosch process cycle. Based on the light intensity time series data, the modulation depth of each Bosch process cycle is determined, and the modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal. Determine the corresponding periodic phase change between each adjacent Bosch process cycle; When the phase change of each period is greater than a preset phase stability threshold, and the modulation depth of each period is less than a preset modulation depth threshold, the end point of the Bosch process etching is determined to have been reached.

2. The method according to claim 1, characterized in that, The Bosch process cycle includes an etching window and a passivation window; determining the modulation depth of each Bosch process cycle based on the light intensity timing data includes: For each Bosch process cycle, determine the first light intensity timing data corresponding to the etching window and the second light intensity timing data corresponding to the passivation window within the Bosch process cycle; The modulation depth of the Bosch process cycle is determined based on the degree of change between the first light intensity time series data and the second light intensity time series data.

3. The method according to claim 2, characterized in that, Determining the modulation depth of the Bosch process cycle based on the degree of change between the first light intensity time-series data and the second light intensity time-series data includes: Determine the first average light intensity corresponding to the first light intensity time series data, and determine the second average light intensity corresponding to the second light intensity time series data; The first average light intensity and the second average light intensity are normalized to obtain the modulation depth of the Bosch process cycle.

4. The method according to claim 1, characterized in that, The step of extracting the periodic phase corresponding to the target frequency component in each of the light intensity time series data includes: Corresponding to each Bosch process cycle, frequency analysis is performed on the light intensity time series data corresponding to the Bosch process cycle to obtain the target frequency component in the light intensity time series data with the target frequency value; wherein, Fourier transform processing is used to perform frequency analysis on the light intensity time series data corresponding to the Bosch process cycle. The phase of the target frequency component corresponding to the midpoint of the Bosch process cycle is determined to obtain the cycle phase.

5. The method according to claim 1, characterized in that, The step of extracting the periodic phase corresponding to the target frequency component in each of the light intensity time series data includes: Corresponding to the light intensity timing data of each Bosch process cycle, a digital phase-locked loop is used with the target frequency value as the reference frequency to perform phase-locked tracking on the light intensity timing data, so as to obtain the continuous phase corresponding to the target frequency component in the light intensity timing data. The phase value corresponding to the midpoint of the Bosch process cycle is determined as the cycle phase.

6. The method according to claim 1, characterized in that, The modulation depth threshold is set to 0.05 to 0.

1.

7. The method according to claim 1, characterized in that, The phase stability threshold is set to 20° to 45°.

8. The method according to claim 1, characterized in that, The preset number of Bosch process cycles is 3 to 5.

9. A device for detecting the end point of etching in a Bosch process, characterized in that, The device includes: The acquisition module is used to acquire characteristic light intensity signals corresponding to multiple preset times in each Bosch process cycle, and obtain light intensity time sequence data corresponding to multiple consecutive Bosch process cycles. The Bosch process cycle is determined by a gas switching signal triggered by an external process. The frequency determination module is used to determine the target frequency value based on the Bosch process cycle. The timing data determination module is used to determine the light intensity timing data corresponding to each Bosch process cycle within a continuously preset number of Bosch process cycles. The extraction module is used to extract the periodic phase corresponding to the target frequency component in each of the light intensity time series data to obtain the periodic phase corresponding to each of the Bosch process cycles; the target frequency component is the frequency component in the light intensity time series data whose frequency value is the target frequency value, and the periodic phase represents the relative position of the target frequency component in the corresponding Bosch process cycle. The modulation depth determination module is used to determine the modulation depth of each Bosch process cycle based on the light intensity time series data, wherein the modulation depth characterizes the response amplitude of the characteristic light intensity signal to the gas switching signal. The detection data determination module is used to determine the corresponding periodic phase change between each adjacent Bosch process cycle; The detection module is used to determine that the Bosch process etching endpoint has been reached when the phase change of each period is greater than a preset phase stability threshold and the modulation depth of each period is less than a preset modulation depth threshold.

10. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the Bosch process etching endpoint detection method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the Bosch process etching endpoint detection method according to any one of claims 1 to 8.

12. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the Bosch process etching endpoint detection method as described in any one of claims 1 to 8.

Images