A method and system for measuring dissociation rate of process gas in a remote plasma source

Abstract

The application discloses a method and system for measuring dissociation rate of process gas in a remote plasma source. The method comprises: feeding a mixed gas containing inert gas and process gas to be measured into the remote plasma source; directly collecting plasma emission spectrum through an optical window at the end of a vertical exhaust pipe directly below; calculating the dissociation rate according to the spectral line intensity ratio of the inert gas and dissociated atoms; and setting a sample in a horizontal branch pipe on the side of the vertical exhaust pipe to verify the accuracy of the dissociation rate through process effect. The system comprises a remote plasma source, a vertical exhaust pipe, a horizontal branch pipe, a spectrum collection unit and a data processing unit. The application adopts an integrated structure of in-situ spectrum collection and synchronous process verification, eliminates the transmission delay and particle recombination loss of traditional FTIR detection, establishes a direct quantitative corresponding relationship between diagnostic data and process effect, and is suitable for dissociation rate measurement of various process gases.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor manufacturing equipment technology, specifically relating to a method and system for measuring the dissociation rate of process gases in a remote plasma source. Background Technology

[0002] Remote Plasma Sources (RPS) are key components in semiconductor manufacturing equipment, widely used in processes such as photoresist removal, thin film etching, and cavity cleaning. RPS generates active particles (free radicals, atoms, etc.) by ionizing process gases (such as O2, N2, NF3, etc.), enabling non-contact treatment of wafer surfaces using these highly reactive particles. The dissociation rate of the process gases directly determines the efficiency of active particle generation and the processing capacity of the RPS, and is a core indicator for evaluating RPS performance. Therefore, accurate measurement of the dissociation rate of process gases is crucial for evaluating and optimizing RPS performance. With the increasing precision requirements of semiconductor manufacturing processes, the accuracy of dissociation rate measurement directly affects the stability and consistency of key processes such as photoresist removal and thin film etching. Currently, the industry typically uses Fourier Transform Infrared Spectroscopy (FTIR) to detect the residual gas concentration in the exhaust gas to indirectly calculate the dissociation rate; simultaneously, a quadrupole mass spectrometer (QMS) is used for closed-loop verification to ensure the accuracy and reliability of the measurement results.

[0003] However, the aforementioned technologies have the following problems: With the advancement of semiconductor manufacturing technology, processes such as N2 / H2O environmentally friendly cleaning and O2 resist removal are becoming increasingly widespread. However, Fourier transform infrared spectroscopy (FTIR) technology, based on the principle of infrared absorption generated by changes in molecular dipole moments, can only detect infrared-active heteronuclear diatomic molecules (such as NF3), and cannot detect diatomic molecules (such as N2 and O2). Therefore, given the rapid growth in the application of N2 and O2, existing FTIR technology can no longer meet the measurement requirements for the dissociation rate of the aforementioned process gases.

[0004] Secondly, Chinese patent application CN 120499913A proposes to detect the amount of remaining maintenance gas in the dissociation chamber using a Fourier transform infrared spectroscopy (FTIR) instrument, and to calculate the dissociation rate (dissociation rate = 1 - M / N) based on the ratio of the input gas amount to the remaining gas amount.

[0005] However, this method has the following inherent drawbacks: This method requires waiting for the gas to flow from the RPS dissociation chamber through the transmission pipeline to the FTIR detection point, resulting in a significant physical transport delay. More seriously, highly reactive free radicals and atoms continuously undergo recombination reactions during transport (e.g., F + F → F2), causing the detected "residual gas quantity" to actually be the steady-state value after recombination reactions, rather than the transient dissociation state at the RPS outlet. This measurement lag makes this method unsuitable for real-time monitoring of RPS dynamic operating conditions.

[0006] Because this method relies on detecting the concentration of residual NF3 and other gases, and the detection point is located in the transmission pipeline downstream of the RPS, a large amount of active particles (such as F atoms) generated by dissociation are lost through gas-phase recombination or wall recombination before reaching the detection point. This not only leads to an underestimation of the dissociation rate, but more seriously, some of the dissociated active particles may regenerate feed molecules through reverse reactions (such as NF2 + F → NF3), causing the detected "residual gas quantity" to include the portion that has been regenerated after dissociation, resulting in a fundamental error in the dissociation rate calculation. Existing technologies fail to provide an effective means to distinguish between "undissociated feed gas" and "feed gas regenerated after dissociation," leading to reduced measurement reliability.

[0007] Secondly, such as Figure 1 The prior art of Chinese patent CN 120499913A, as shown, places the spectral detection device and the etching stage (process verification unit) at different locations downstream of the RPS, with a significant spatial distance between the spectral acquisition point and the process verification point. This structure leads to: The plasma state at the spectral detection point differs significantly from the state of active particles reaching the etching stage due to transmission loss, making it difficult to establish an accurate quantitative correspondence between the data from the two test units.

[0008] Therefore, there is an urgent need in this field for an RPS dissociation rate measurement method and system that can directly measure the concentration of atomic products, is applicable to diatomic molecular systems, highly integrates diagnostics and process verification, is simple in system, and has high measurement accuracy, in order to solve the many problems existing in the above-mentioned prior art. Summary of the Invention

[0009] This invention provides a method and system for measuring the dissociation rate of process gases in a remote plasma source, aiming to solve the technical problems existing in the prior art.

[0010] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0011] A method for measuring the dissociation rate of a process gas in a remote plasma source, the method comprising at least the following steps:

[0012] Step 1: A mixture of inert gas and process gas to be tested is introduced from above into the dissociation chamber of the remote plasma source. The process gas to be tested includes at least one diatomic molecule or polyatomic molecule.

[0013] Step 2: At the end of the vertical exhaust pipe below the remote plasma source, the plasma emission spectrum is directly acquired through an optical window. The emission spectrum includes the characteristic spectral lines of the inert gas and the atomic characteristic spectral lines generated by the dissociation of the process gas to be tested.

[0014] Step 3: Calculate the dissociation rate of the process gas to be tested based on the ratio of the spectral line intensities of the inert gas to the dissociated atoms in the emission spectrum;

[0015] A sample is placed inside a horizontal branch pipe on the side of the vertical exhaust pipe, and a vacuum pump is connected to the end of the horizontal branch pipe so that the plane where the sample is located is perpendicular to the airflow direction. The accuracy of the dissociation rate is verified simultaneously by the process effect of the sample.

[0016] Preferably, the inert gas is argon, the process gas to be tested is oxygen or nitrogen trifluoride, the atomic characteristic spectral line is the 844.6 nm spectral line of oxygen atoms or the 703.7 nm spectral line of fluorine atoms, and the 750.4 nm spectral line of argon atoms is selected as the reference spectral line.

[0017] Preferably, the dissociation rate is calculated based on the ratio of the inert gas to the dissociated atoms in the emission spectrum, as well as the known inert gas concentration and mixed gas flow rate parameters to obtain the dissociation rate of the process gas to be tested.

[0018] Preferably, the distance between the optical window and the fiber optic probe is 0mm to 5mm, and the fiber optic probe is set perpendicular to the surface of the optical window.

[0019] Preferably, the sample is fixed inside the horizontal branch pipe by an adjustable bracket. The horizontal branch pipe includes a horizontal adjustment module, which adjusts the total length of the horizontal branch pipe. Based on the position of the bracket, the sample is fixed at different distances to measure the effective propagation distance of the active particles. The vertical exhaust pipe includes a vertical adjustment module, which adjusts the length of the vertical exhaust pipe to reduce the vertical height when the discharge is weak.

[0020] The present invention also provides a system for measuring the dissociation rate of process gas in a remote plasma source, comprising:

[0021] A remote plasma source having a dissociation chamber and an inlet pipe disposed above the dissociation chamber;

[0022] A vertical exhaust pipe is connected directly below the remote plasma source, and an optical window is provided at the end of the vertical exhaust pipe;

[0023] A spectral acquisition unit is disposed outside the optical window and includes an optical fiber probe and a spectral analyzer disposed perpendicular to the surface of the optical window. The distance between the optical fiber probe and the optical window is 0 mm to 5 mm. The spectral acquisition unit is signal-connected to the data processing unit.

[0024] A horizontal branch pipe is provided on the side of the vertical exhaust pipe. An adjustable bracket is provided inside the horizontal branch pipe to fix the sample and make the plane where the sample is located perpendicular to the airflow direction. The horizontal branch pipe and the vertical exhaust pipe form a connected airflow path.

[0025] The data processing unit is used to receive signals from the spectral acquisition unit to calculate the dissociation rate of the process gas, and to receive process verification feedback to calibrate the measurement results.

[0026] Preferably, the lengths of the vertical exhaust pipe and the horizontal branch pipe are adjusted by a vertical adjustment module and a horizontal adjustment module, respectively.

[0027] Preferably, the end of the horizontal branch pipe is provided with a quick-release structure, which is fixed by a sealing ring.

[0028] Preferably, the adjustable bracket is a breathable fastener used to fix the position of the sample while ensuring airflow.

[0029] Preferably, the optical window is a quartz window, and the system is suitable for measuring the dissociation rate of diatomic molecular process gases.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] 1. This invention employs an integrated structure combining in-situ spectral acquisition at the end of the vertical exhaust pipe with synchronous process verification via a side horizontal branch pipe, directly capturing the plasma emission spectrum at the RPS outlet without severe recombination. Compared to traditional FTIR exhaust gas detection methods, this invention eliminates physical transmission delay and active particle recombination loss, significantly improving the timeliness and reliability of measurement results.

[0032] 2. This invention integrates the spectral diagnostic unit and the process verification unit into a single compact exhaust module. Through a vertical-horizontal orthogonal piping design, the spectral acquisition point and the sample process point are placed under the same RPS (Reference Price Segmentation) condition, thereby establishing a direct quantitative correspondence between diagnostic data and process effects. Simultaneously, by reverse-calibrating the dissociation rate calculation model using measurable parameters such as etching rate, the data correlation and measurement accuracy are effectively improved.

[0033] 3. This invention, through its adjustable vertical exhaust pipe and horizontal branch pipe structure, can flexibly adapt to the spectral acquisition requirements of different discharge intensities and support preliminary experiments on the propagation distance of active particles, providing a reliable experimental means for RPS design optimization and optimal use in customer sites. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of an existing RPS dissociation rate measurement system based on FTIR exhaust gas detection.

[0035] Figure 2 This is a schematic diagram of the overall structure of the system for measuring the dissociation rate of process gases in a remote plasma source, as described in this invention.

[0036] Figure 3 This is an argon-oxygen plasma emission spectrum acquired using the method of this invention.

[0037] Figure 4 These are comparative images of the surface morphology of samples after being processed using the system of this invention.

[0038] Figure 5 This is a comparison chart of the dissociation rate calculation results based on the spectral measurement data of this invention.

[0039] Figure 6 This is a comparison chart of the etching rates measured using the simulated etching process based on the present invention.

[0040] Explanation of reference numerals in the attached figures:

[0041] 1. Remote plasma source; 2. Vertical exhaust pipe; 21. Vertical adjustment module; 3. Horizontal branch pipe; 31. Horizontal adjustment module; 4. Optical window; 5. Fiber optic probe; 7. Inlet pipe; 8. Fixed platform; 9. Support; 10. Vacuum pump. Detailed Implementation

[0042] 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, and 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.

[0043] A method for measuring the dissociation rate of a process gas in a remote plasma source, the method comprising the following steps:

[0044] Step 1: A mixture of inert gas and process gas to be tested is introduced from above into the dissociation chamber of remote plasma source 1. The process gas to be tested includes at least one diatomic molecule or polyatomic molecule.

[0045] Step 2: At the end of the vertical exhaust pipe 2 below the remote plasma source, the plasma emission spectrum is directly acquired through the optical window 4. The emission spectrum includes characteristic spectral lines of the inert gas and atomic characteristic spectral lines generated by the dissociation of the process gas under test. This setup satisfies the in-situ acquisition method, directly capturing the plasma emission spectrum at the outlet of the remote plasma source 1. It eliminates the physical transmission delay present in traditional FTIR exhaust gas detection, avoids recombination loss of active particles during transmission, significantly improves the intensity and timeliness of the spectral signal, and ensures that the measurement results can truly reflect the dissociation state inside the remote plasma source 1.

[0046] Step 3: Calculate the dissociation rate of the process gas to be tested based on the ratio of the spectral line intensities of the inert gas to the dissociated atoms in the emission spectrum.

[0047] A sample is placed inside a horizontal branch pipe 3 on the side of the vertical exhaust pipe 2. A vacuum pump 10 is connected to the end of the horizontal branch pipe 3 so that the plane where the sample is located is perpendicular to the airflow direction. The accuracy of the dissociation rate is verified simultaneously by the process effect of the sample.

[0048] This setup enables the integration of process verification functions within the same exhaust module, placing the spectral acquisition point and the sample process point under identical remote plasma source 1 operating conditions. This establishes a direct quantitative correspondence between diagnostic data and process effects. By reverse-calibrating the dissociation rate calculation model using measurable parameters such as etching rate, the data correlation and measurement accuracy are effectively improved, solving the technical problem of the disconnect between diagnostic data and process effects in traditional methods.

[0049] Preferably, the inert gas is argon, the process gas to be measured is oxygen or nitrogen trifluoride, and the atomic characteristic spectral line is the 844.6 nm spectral line of oxygen atoms or the 703.7 nm spectral line of fluorine atoms, with the 750.4 nm spectral line of argon atoms selected as the reference spectral line. This preferred scheme, targeting gases commonly used in semiconductor processes such as oxygen and nitrogen trifluoride, uses the 750.4 nm spectral line of argon atoms as an internal standard reference, which can effectively eliminate measurement errors caused by factors such as plasma fluctuations and optical path changes, improve the stability and reliability of dissociation rate calculation, and expand the applicability of the method to various process gases.

[0050] Preferably, the dissociation rate is calculated based on the ratio of the inert gas to the dissociated atoms in the emission spectrum, as well as the known inert gas concentration and mixed gas flow rate parameters to obtain the dissociation rate of the process gas under test. This preferred scheme, by introducing inert gas concentration and mixed gas flow rate parameters, gives the dissociation rate calculation results clear physical meaning and dimensional consistency, facilitating data comparison and process optimization under different operating conditions, while improving the universality and accuracy of the calculation model.

[0051] Preferably, the distance between the optical window 4 and the fiber optic probe 5 is 0 mm to 5 mm, and the fiber optic probe 5 is arranged perpendicular to the surface of the optical window 4. This preferred embodiment, by limiting the close-range perpendicular arrangement of the fiber optic probe 5 and the optical window 4, reduces optical path loss and stray light interference, ensuring that the spectral acquisition unit obtains a high signal-to-noise ratio plasma emission spectral signal, thereby improving spectral resolution and measurement accuracy.

[0052] Preferably, the sample is fixed within the horizontal branch pipe 3 by a bracket 9. The horizontal branch pipe 3 includes a horizontal adjustment module 31, which adjusts the total length of the horizontal branch pipe 3. Based on the position of the bracket 9, the sample is fixed at different distances to measure the effective propagation distance of the active particles. The vertical exhaust pipe 2 includes a vertical adjustment module 21, which adjusts the length of the vertical exhaust pipe 2 to reduce its vertical height when the discharge is weak. This preferred solution, through an adjustable pipeline structure, achieves the experimental function of determining the propagation distance of active particles and the possibility of adjusting the spectral acquisition position. It provides a reliable experimental means for the design optimization of the remote plasma source 1 and the optimal use in the customer's field. Simultaneously, it can shorten the vertical height to enhance the spectral signal intensity when the discharge is weak, expanding the applicability and experimental function of the method.

[0053] The present invention also provides a system for measuring the dissociation rate of process gas in a remote plasma source, comprising:

[0054] The remote plasma source 1 has a dissociation chamber and an air inlet pipe 7 disposed above the dissociation chamber;

[0055] A vertical exhaust pipe 2 is connected directly below the remote plasma source 1, and an optical window 4 is provided at the end of the vertical exhaust pipe 2;

[0056] A spectral acquisition unit, located outside the optical window 4, includes a fiber optic probe 5 and a spectrometer positioned perpendicular to the surface of the optical window 4. The distance between the fiber optic probe 5 and the optical window 4 is 0 mm to 5 mm. The spectral acquisition unit is signal-connected to the data processing unit. It provides raw data input for dissociation rate calculation, and the signal connection with the data processing unit enables real-time data transmission and automated processing.

[0057] A horizontal branch pipe 3 is disposed on the side of the vertical exhaust pipe 2. A bracket 9 is disposed inside the horizontal branch pipe 3 to fix the sample and make the plane where the sample is located perpendicular to the airflow direction. The horizontal branch pipe 3 and the vertical exhaust pipe 2 form a connected airflow path.

[0058] The data processing unit is used to receive signals from the spectral acquisition unit to calculate the dissociation rate of the process gas, and to receive process verification feedback to calibrate the measurement results.

[0059] Preferably, the lengths of the vertical exhaust pipe 2 and the horizontal branch pipe 3 are adjusted by the vertical adjustment module 21 and the horizontal adjustment module 31, respectively. This preferred structure enables the system to flexibly adapt to the spectral acquisition requirements of different discharge intensities, supports preliminary experiments on the propagation distance of active particles, provides reliable experimental means for the design optimization of the remote plasma source 1 and the optimal use at the customer's site, and enhances the adaptability and functionality of the system.

[0060] Preferably, the end of the horizontal branch pipe 3 is provided with a quick-release structure, which is fixed by a sealing ring. This preferred structure facilitates quick replacement of the sample and system maintenance, while the sealing ring ensures the air circuit is airtight, and reduces system maintenance time and operational complexity.

[0061] Preferably, the support 9 is a breathable fixing component, used to fix the sample position while ensuring airflow. This preferred structure ensures that the active particles can act uniformly on the sample surface while maintaining smooth airflow.

[0062] Preferably, the optical window 4 is a quartz window, and the system is suitable for measuring the dissociation rate of diatomic or polyatomic process gases. This preferred material has good ultraviolet-visible light transmittance, high temperature resistance, and plasma etching resistance, and is suitable for detecting characteristic spectral lines such as oxygen atoms at 844.6 nm and fluorine atoms at 703.7 nm.

[0063] Example 1

[0064] This invention provides a method for measuring the dissociation rate of process gases in a remote plasma source. This method achieves accurate quantitative measurement of the dissociation rate of process gases by combining optical emission spectroscopy with the principle of inert gas calibration.

[0065] Step 1: Gas Mixing and Introduction

[0066] A mixture of an inert gas and the analyte gas is introduced from above into the dissociation chamber of a remote plasma source. The analyte gas contains at least one diatomic or polyatomic molecule, such as oxygen or nitrogen trifluoride. Argon is selected as the inert gas, serving as an internal standard gas that does not participate in the chemical reaction; its ground-state atomic number density remains constant, providing a stable reference for subsequent dissociation rate calculations. Under the influence of plasma, the analyte gas undergoes a dissociation reaction, producing corresponding atomic products.

[0067] Step 2: Spectral Acquisition

[0068] Plasma emission spectra are directly acquired through an optical window at the end of a vertical exhaust pipe directly below the remote plasma source. The distance between the optical window and the fiber optic probe is controlled within the range of 0 mm to 5 mm, and the fiber optic probe is set perpendicular to the surface of the optical window to ensure the accuracy and reproducibility of the spectral acquisition. The acquired emission spectra include characteristic spectral lines of inert gases and atomic characteristic spectral lines generated by the dissociation of the process gas under test.

[0069] For the argon-oxygen mixed gas system, the characteristic emission line for oxygen atoms is chosen at 844.6 nm, with an excitation threshold energy of approximately 10.99 eV. Compared to the 777.4 nm line, this avoids interference from cascade transitions and self-absorption effects. For argon atoms, the 750.4 nm line is selected as the reference line, with an excitation threshold energy of approximately 13.48 eV, similar to the 844.6 nm line for oxygen atoms. Furthermore, this line is almost entirely generated by electron-bearing collisions of ground-state argon, and the cascade transition pathway is negligible.

[0070] For the argon-nitrogen-fluorine mixed gas system, the 703.7 nm characteristic spectral line of fluorine atoms is selected, corresponding to the 3p→3s transition of F atoms, and the 750.4 nm spectral line of argon atoms is used as the reference spectral line.

[0071] Step 3: Calculation of dissociation rate

[0072] The dissociation rate of the process gas under test is calculated based on the intensity ratio of the inert gas to the dissociated atoms in the emission spectrum. The dissociation rate calculation is based on the intensity ratio of the inert gas to the dissociated atoms in the emission spectrum, combined with known inert gas concentration and mixed gas flow parameters, and obtained by establishing a dissociation rate calculation model using the principle of element conservation.

[0073] The intensity of spectral lines in an emission spectrum is proportional to the number density of the corresponding excited-state particles and the probability of radiative transitions. Under quasi-steady-state equilibrium, the generation and loss of excited states are balanced, allowing us to deduce the relationship between spectral line intensity and electron density, as well as the ground-state atomic number density. Since argon is an inert gas and does not participate in the reaction, its ground-state atomic number density is constant. Therefore, a relationship can be established between the argon and oxygen spectral line intensities, which includes instrument and atomic constants, as well as the ratio of excitation rate coefficients.

[0074] For oxygen dissociation rate, calculations are performed based on the principle of element conservation and the general relationship between argon-oxygen flow rate and spectral line intensity. For nitrogen trifluoride, the dissociation rate of NF3 is determined based on F atom concentration, and a proportional relationship between F atom concentration and argon atom concentration is established. This proportional relationship includes factors such as excitation cross-section ratio, wavelength, and transition probability.

[0075] Example: Measurement of oxygen dissociation rate in argon-oxygen plasma

[0076] This embodiment takes an argon-oxygen mixture as an example to illustrate in detail the specific implementation process and calculation results of measuring the oxygen dissociation rate in a remote plasma source using the method and system described in this invention.

[0077] Argon and oxygen are mixed at a volumetric flow rate ratio of 1:9, with the total flow rate controlled at 500 sccm. The mixture is then introduced from above into the dissociation chamber of the remote plasma source 1 through inlet pipe 7. The discharge power of the remote plasma source 1 is set to 8000W, and the working pressure is maintained at 3 torr10 Pa. The length of the vertical exhaust pipe 2 is adjusted to 150 mm using vertical adjustment module 21, and the length of the horizontal branch pipe 3 is adjusted to 200 mm using horizontal adjustment module 31.

[0078] Outside the optical window 4 at the end of the vertical exhaust pipe 2, the fiber optic probe 5 is positioned perpendicular to the surface of the optical window 4, with the distance between them adjusted to 2mm. The plasma emission spectrum at the outlet of the remote plasma source 1 is directly acquired through the spectral acquisition unit, with a spectral acquisition range of 200nm-900nm and an integration time set to 100ms.

[0079] like Figure 3 As shown, the acquired emission spectra clearly display the characteristic spectral lines of argon atoms at 750.4 nm, and oxygen atoms at 777.4 nm and 844.6 nm. Among them, the 777.4 nm oxygen atom spectral line has the highest intensity, reaching approximately 21,000 counts; the 750.4 nm argon atom spectral line has an intensity of approximately 800 counts; and the 844.6 nm oxygen atom spectral line has an intensity of approximately 5,000 counts.

[0080] Based on the spectral characteristics analysis, although the 777.4 nm spectral line of the oxygen atom has the highest intensity, this excited-state particle has contributions from cascade transitions in addition to electron collision excitation, and also exhibits a significant self-absorption effect. Therefore, it is not suitable for the accurate calculation of the dissociation rate.

[0081] The excitation threshold energy of the 844.6 nm spectral line of oxygen atoms is approximately 10.99 eV. This spectral line is almost entirely generated by ground-state oxygen atoms through electron collisions. The cascade conversion path is negligible, and the self-absorption effect is weak, making it suitable for dissociation rate calculations.

[0082] The excitation threshold energy of the 750.4 nm argon atom spectral line is approximately 13.48 eV, which is similar to that of the 844.6 nm argon atom spectral line. Furthermore, this spectral line is entirely generated by ground-state argon atoms through electron collisions, and the cascade conversion path is negligible. Simultaneously, the wavelength difference between 750.4 nm and 844.6 nm is approximately 94 nm, effectively avoiding spectral overlap interference. Therefore, the 750.4 nm argon atom spectral line was chosen as the internal standard reference line.

[0083] Based on the above spectral line selection, a dissociation rate calculation model is established. Let the argon flow rate in the argon-oxygen mixture be... Oxygen flow rate is The ratio of the initial argon atom number density to the oxygen molecule number density is:

[0084]

[0085] Under quasi-steady-state equilibrium conditions, the relationship between the intensity of the emission spectral lines and the ground-state particle number density is as follows:

[0086]

[0087]

[0088] in, and For instrument and atomic constants, [O] and [Ar] represent the electron density, respectively, and the number densities of ground-state oxygen and argon atoms. and These are the electron collision excitation rate coefficients for the corresponding spectral lines.

[0089] Since the excitation threshold energies of argon atoms (750.4 nm) and oxygen atoms (844.6 nm) are similar (13.48 eV vs 10.99 eV), under the same electron energy distribution function, their excitation rate coefficients are... It can be considered a constant.

[0090] Dividing the two equations, we get:

[0091]

[0092] make ,but:

[0093]

[0094] Since argon is an inert gas and does not participate in chemical reactions, its ground-state atomic number density remains constant. .

[0095] According to the law of conservation of elements, the initial oxygen molecule number density The number density of oxygen molecules remaining after dissociation [O2] and the number density of oxygen atoms generated [O] satisfy the following:

[0096]

[0097] The oxygen dissociation rate η is defined as:

[0098]

[0099] Based on the above relationships, the general formula for calculating the dissociation rate and spectral line intensity is finally obtained:

[0100]

[0101] Where C' is a comprehensive calibration coefficient that includes the flow ratio and the instrument constant, which can be determined through standard samples or theoretical calculations.

[0102] according to Figure 3 The spectral data shown are measured. , The spectral line intensity ratio is 6.25. Combining this with the pre-calibrated coefficient C' = 8.5, the oxygen dissociation rate under the current operating conditions is calculated as follows:

[0103] η = 6.25 / (8.5 + 6.25) = 42.4%

[0104] A silicon wafer sample was placed 100 mm from the bifurcation point inside the horizontal branch pipe 3, and fixed by the support 9 with the sample plane perpendicular to the airflow direction. Under the same operating conditions of the remote plasma source 1, the oxidation etching rate of the silicon wafer was measured to be 360 ​​Å / s. By changing the discharge power of the remote plasma source 1, a quantitative correlation between the dissociation rate and the etching rate was established, and the accuracy of the dissociation rate calculation model was verified in reverse, with the relative error controlled within 5%.

[0105] The total length of the horizontal branch pipe 3 was adjusted by the horizontal adjustment module 31, and the sample was fixed at distances of 50mm, 100mm, 150mm, and 200mm from the bifurcation point, respectively. The etching rate changes at different positions were measured. The results showed that the etching rate decreased exponentially with increasing distance, and the effective propagation distance was approximately 180mm, providing experimental basis for the process optimization and client application of the remote plasma source 1.

[0106] Example: Measurement of Nitrogen Trifluoride Plasma Dissociation Rate

[0107] Using the same system structure as in the previous embodiment, the process gas was replaced with nitrogen trifluoride (NF3), and the flow ratio of argon to nitrogen trifluoride was set to 2:8. The remote plasma source 1 had a discharge power of 10 kW and an operating pressure of 3 torr.

[0108] The main dissociation product of nitrogen trifluoride is fluorine atoms, which exhibit a strong emission line at 703.7 nm (corresponding to the 3p→3s transition). Using the 750.4 nm spectral line of argon atoms as an actinometer, the ratio of fluorine atom concentration to argon atom concentration was established:

[0109]

[0110] in It is a proportionality constant that includes the excitation cross-section ratio, wavelength factor, transition probability, etc., and is determined through pre-calibration.

[0111] Measured Calculated Combining the flow ratio and the element conservation principle, the dissociation rate of nitrogen trifluoride was calculated to be 38.6%. Verification through silicon wafer etching showed that this dissociation rate was in good agreement with the measured etching rate, demonstrating the applicability of the method of this invention to polyatomic molecular process gases.

[0112] Step 4: Verification of Process Effect

[0113] A sample is placed inside a horizontal branch pipe on the side of the vertical exhaust pipe, with the plane of the sample perpendicular to the airflow direction. The accuracy of the dissociation rate is simultaneously verified by observing the process effect of the sample. The sample is fixed inside the horizontal branch pipe using an adjustable bracket. By adjusting the length of the horizontal branch pipe, the sample is fixed at different distances to measure the effective propagation distance of the active particles. By adjusting the length of the vertical exhaust pipe, the vertical height needs to be reduced when measuring the spectrum under conditions of weak discharge, thus optimizing the measurement conditions.

[0114] like Figure 4 As shown, after approximately 20 seconds of simulated process treatment, the sample before treatment showed a significant change in surface morphology compared to the sample after treatment, demonstrating the convenience and efficiency of the simulated process system. Furthermore, by measuring the material removal rate of the sample under the action of active particles, the correlation between the dissociation rate and the process effect was verified. The verification methods may include, but are not limited to, the following methods:

[0115] Film thickness measurement method: Use an ellipsometry or profilometer to measure the film thickness change of the sample before and after processing, and calculate the etching rate.

[0116] Gravimetric method: The amount of etching is estimated by measuring the mass loss of the sample before and after processing using a precision balance.

[0117] Profilometry: Use scanning electron microscopy (SEM) or atomic force microscopy (AFM) to observe the morphology and depth of the etching.

[0118] By comparing spectral measurement data from different models of equipment with actual etching rate results, the high accuracy of the spectroscopic method for measuring dissociation rate was verified. Under the same gas conditions, the equipment with a higher dissociation rate corresponds to a higher etching rate. The results of the two testing methods are consistent, fully demonstrating the reliability and practicality of this method.

[0119] This method enables real-time and accurate measurement of the dissociation rate of process gases in remote plasma sources using optical emission spectroscopy, providing an effective detection method for the optimization and control of plasma processes.

[0120] To verify the accuracy of the measurement method described in this invention, comparative tests were conducted using two types of remote plasma sources, SS10 and SS10-H, under the same operating conditions. With a fixed oxygen-argon flow ratio of 2:1, the total flow rate was adjusted from 10% to 100%, and the oxygen dissociation rate of both sources was measured using the method of this invention. Figure 5 As shown, the dissociation rate of the SS10-H model is consistently higher than that of the SS10 model. At 100% flow rate, the dissociation rate of the SS10-H reaches 42.06%, while that of the SS10 is only 24.81%.

[0121] The above measurement results were further verified by simulating an etching process, measuring the etching rates of two different types of silicon-based thin films under the same process conditions. Figure 6 As shown, the etching rate of the SS10-H model was significantly higher than that of the SS10 model at all time points, reaching 368 Å / s at 60 seconds, while that of the SS10 was only 213 Å / s. The results of the two testing methods showed a high degree of consistency, that is, the SS10-H model with a higher dissociation rate exhibited a higher etching rate, proving that the dissociation rate measurement method based on in-situ spectral acquisition described in this invention can accurately reflect the actual process performance of the remote plasma source and has good accuracy and reliability.

[0122] Example 2

[0123] This invention provides a system for measuring the dissociation rate of process gases in a remote plasma source. This system can achieve accurate measurement and real-time monitoring of the dissociation rate of process gases.

[0124] The system includes a remote plasma source with a dissociation chamber and an inlet pipe positioned above the dissociation chamber. The dissociation chamber generates plasma and dissociates the process gas, while the inlet pipe supplies the process gas to be dissociated into the dissociation chamber. Within the dissociation chamber, the process gas undergoes a dissociation reaction under the influence of the plasma, producing corresponding atomic products.

[0125] A vertical exhaust pipe is connected directly below the remote plasma source to guide the dissociated gas downstream. An optical window, made of quartz, is located at the end of the vertical exhaust pipe. This quartz window offers excellent optical transmittance and corrosion resistance, enabling it to withstand the high temperatures and chemical corrosion of the plasma environment. The quartz window allows the emission spectrum generated by dissociation to pass through, providing an optical path for subsequent spectroscopic detection.

[0126] The spectral acquisition unit is located outside the optical window and includes a fiber optic probe and a spectrometer positioned perpendicular to the optical window surface. The distance between the fiber optic probe and the optical window is 0mm to 5mm, a distance setting that ensures effective acquisition of spectral signals while avoiding direct interference from the plasma environment. The fiber optic probe collects the emission spectral signals passing through the quartz window, and the spectrometer analyzes and processes the collected spectra to obtain spectral intensity data at different wavelengths. The spectral acquisition unit is connected to the data processing unit to achieve real-time transmission of spectral data.

[0127] The horizontal branch pipe is located on the side of the vertical exhaust pipe, forming a connected airflow path. An adjustable bracket of adjustable length is installed inside the horizontal branch pipe to fix the sample and ensure that the plane of the sample is perpendicular to the airflow direction. The adjustable bracket is a breathable fixing component, which secures the sample position while allowing airflow, enabling the dissociated active gas to fully contact the sample for etching or other processing. The breathable design avoids airflow blockage, ensuring the continuity and uniformity of the airflow.

[0128] The data processing unit receives signals from the spectral acquisition unit to calculate the dissociation rate of the process gas and receives process verification feedback to calibrate the measurement results. The data processing unit calculates the dissociation rate of the process gas by analyzing the spectral intensity at a specific wavelength and using an inert gas as an internal standard. This system is suitable for measuring the dissociation rate of diatomic process gases, such as oxygen and nitrogen.

[0129] In a preferred embodiment, the lengths of the vertical exhaust pipe and the horizontal branch pipe are adjustable via telescopic connecting fittings. These fittings allow for adjustments to pipe lengths to suit different process requirements and equipment configurations, improving the system's adaptability and flexibility. Adjusting the pipe lengths optimizes the gas flow path, ensuring that dissociated products maintain appropriate concentrations and activity upon reaching the detection location.

[0130] In another preferred embodiment, the end of the horizontal branch pipe is provided with a quick-release structure, which is secured by a sealing ring. The quick-release structure facilitates sample replacement and maintenance, while the sealing ring ensures a tight seal, preventing gas leakage from affecting measurement accuracy. This design significantly improves the system's operational convenience and maintenance efficiency.

[0131] The system operates based on emission spectroscopy to measure the dissociation rate. Taking an argon-oxygen mixture as an example, argon, as an inert gas, does not participate in chemical reactions, and its ground-state atomic number density remains constant, serving as an internal standard reference. Under plasma conditions, oxygen partially dissociates into oxygen atoms, which produce a characteristic emission line at 844.6 nm, while argon atoms produce a characteristic emission line at 750.4 nm. By measuring the ratio of the spectral intensities at these two wavelengths and combining this with the argon-oxygen flow rate ratio, the dissociation rate of oxygen can be calculated.

[0132] The system monitors the emission spectra of dissociation products in real time through a spectral acquisition unit. The data processing unit calculates the dissociation rate based on the spectral intensity data and calibrates the measurement results using feedback from sample process verification, ensuring accuracy and reliability. This system achieves an organic integration of dissociation rate measurement and process verification, providing an effective technical means for the optimization and control of plasma processes.

[0133] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for measuring the dissociation rate of process gas in a remote plasma source, characterized in that, The method includes the following steps: Step 1: A mixture of inert gas and process gas to be tested is introduced from above into the dissociation chamber of the remote plasma source (1), wherein the process gas to be tested includes at least one diatomic molecule; Step 2: At the end of the vertical exhaust pipe (2) below the remote plasma source, the plasma emission spectrum is directly collected through the optical window (4). The emission spectrum includes the characteristic spectral lines of the inert gas and the atomic characteristic spectral lines generated by the dissociation of the process gas to be tested. Step 3: Calculate the dissociation rate of the process gas to be tested based on the ratio of the spectral line intensities of the inert gas to the dissociated atoms in the emission spectrum.

2. The method according to claim 1, characterized in that, The method further includes: Step 4: Set a sample in the horizontal branch pipe (3) on the side of the vertical exhaust pipe (2), and connect a vacuum pump (10) to the end of the horizontal branch pipe so that the plane where the sample is located is perpendicular to the airflow direction. Simultaneously verify the accuracy of the dissociation rate through the process effect of the sample.

3. The method according to claim 1, characterized in that, The inert gas is argon, the process gas to be tested is oxygen or nitrogen trifluoride, the atomic characteristic spectral lines are the 844.6 nm spectral line of oxygen atoms or the 703.7 nm spectral line of fluorine atoms, and the 750.4 nm spectral line of argon atoms is selected as the reference spectral line.

4. The method according to claim 1, characterized in that, The dissociation rate is calculated based on the ratio of the inert gas to the dissociated atoms in the emission spectrum, as well as the known inert gas concentration and mixed gas flow rate parameters, to obtain the dissociation rate of the process gas to be tested.

5. The method according to claim 1, characterized in that, The distance between the optical window (4) and the fiber optic probe (5) is 0 mm to 5 mm, and the fiber optic probe (5) is set perpendicular to the surface of the optical window (4).

6. The method according to claim 1, characterized in that, The sample is fixed inside the horizontal branch pipe (3) by an adjustable bracket (9). The horizontal branch pipe (3) includes a horizontal adjustment module (31). The total length of the horizontal branch pipe (3) is adjusted by the horizontal adjustment module (31). The sample is fixed at different distances by the position of the bracket (9) to measure the effective propagation distance of the active particles. The vertical exhaust pipe (2) includes a vertical adjustment module (21). The length of the vertical exhaust pipe (2) is adjusted by the vertical adjustment module (21) to reduce the vertical height when the discharge is weak.

7. A system for measuring the dissociation rate of process gases in a remote plasma source, characterized in that, include: The remote plasma source (1) has a dissociation chamber and an inlet pipe (7) disposed above the dissociation chamber. A vertical exhaust pipe (2) is connected directly below the remote plasma source (1), and an optical window (4) is provided at the end of the vertical exhaust pipe (2). The spectral acquisition unit is located outside the optical window (4) and includes an optical fiber probe (5) and a spectral analyzer arranged perpendicular to the surface of the optical window (4). The distance between the optical fiber probe and the optical window is 0 mm to 5 mm. The spectral acquisition unit is connected to the data processing unit. A horizontal branch pipe (3) is set on the side of the vertical exhaust pipe (2). An adjustable bracket (9) is set inside the horizontal branch pipe (3) to fix the sample and make the plane where the sample is located perpendicular to the airflow direction. The horizontal branch pipe (3) and the vertical exhaust pipe (2) form a connected airflow path. The data processing unit is used to receive signals from the spectral acquisition unit to calculate the dissociation rate of the process gas, and to receive process verification feedback to calibrate the measurement results.

8. The system according to claim 6, characterized in that, The lengths of the vertical exhaust pipe (2) and the horizontal branch pipe (3) are adjusted by the vertical adjustment module (21) and the horizontal adjustment module (31), respectively.

9. The system according to claim 6, characterized in that, The end of the horizontal branch pipe is provided with a quick-release structure, which is fixed by a sealing ring.

10. The system according to claim 6, characterized in that, The adjustable bracket (9) is a breathable fixing component used to fix the sample position while ensuring airflow; the optical window is a quartz window, and the system is suitable for measuring the dissociation rate of various process gases.

Images