Method for determining electron temperature in plasma process
By generating and analyzing hydrogen plasma emissions, the method determines electron temperature without plasma disruption, addressing measurement challenges in high-pressure environments and film deposition processes.
Patent Information
- Application Number
- JP2024073925
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-04-30
- Publication Date
- 2025-11-12
AI Technical Summary
Existing methods for determining electron temperature in plasma processes using a probe disrupt the plasma, are unreliable under high gas pressure, and face challenges in film deposition due to film adhesion, making accurate measurement difficult.
Generate hydrogen plasma, measure Fulcher band emission and dissociative continuum light, normalize their intensities, and use a pre-calculated relationship to determine electron temperature without affecting the plasma, applicable even under high gas pressure.
Enables accurate electron temperature determination without disrupting the plasma, suitable for film-forming processes under high gas pressure, ensuring reliable measurements.
Smart Images
Figure 2025168993000002 
Figure 2025168993000003 
Figure 2025168993000004
Abstract
Description
[Technical Field]
[0001] The present invention relates to a method for determining electron temperature in a plasma process. [Background technology]
[0002] Plasma processes are widely used in the fabrication of semiconductor devices and other devices. In plasma processes, electron temperature is widely recognized as an important parameter that determines the success or failure of the process. Electron temperature is an index that indicates the magnitude of the thermal kinetic energy of electrons. Generally, electron temperature is determined by inserting a probe into the plasma and measuring the current-voltage characteristics.
[0003] As a method for determining the electron temperature using a general probe, for example, the method described in Non-Patent Document 1 is known. [Prior art documents] [Non-patent literature]
[0004] [Non-Patent Document 1] Michael A. Lieberman, Alan J. Lichtenberg, Principles of Plasma Discharges and Materials Processing, Wiley-Interscience; 2nd edition (April 14, 2005). Summary of the Invention [Problem to be solved by the invention]
[0005] There are three problems with the method of determining the electron temperature using a probe: (1) Inserting a probe into the plasma disrupts the plasma, making it difficult to determine the electron temperature using a probe in an actual process; (2) Under high gas pressure process conditions, collisions between the probe and gas molecules reduce the reliability of the electron temperature; and (3) In film deposition plasmas used for insulating film deposition, film adhesion to the probe makes it difficult to measure the electron temperature.
[0006] The present invention has been made in view of the above circumstances, and aims to provide a method for determining electron temperature in a plasma process, which can determine the electron temperature without adversely affecting the plasma, and which is also applicable to film-forming plasma under high gas pressure. [Means for solving the problem]
[0007] The present invention has the following aspects. [1] generating hydrogen plasma by introducing hydrogen molecules into the plasma; measuring Fulcher band emission and dissociation continuum light originating from the hydrogen molecules in the emission spectrum of the hydrogen plasma; normalizing the emission intensity of the Fulcher band emission with the emission intensity of the dissociative continuum; a step of determining the electron temperature of the hydrogen plasma from the relationship between the emission intensity of the Fulcher band emission normalized by the emission intensity of the dissociative continuum light obtained by pre-calculation and the electron temperature of the hydrogen plasma, and the value of the emission intensity of the Fulcher band emission normalized by the emission intensity of the dissociative continuum light. [2] The method for determining the electron temperature in a plasma process according to [1], wherein the emission intensity of the Fulcher band emission is the average value of the emission intensity at wavelengths of 585 nm or more and 645 nm or less. [3] The method for determining the electron temperature in a plasma process according to [1], wherein the wavelength of the dissociative continuous light is 200 nm or more and 500 nm or less. [Effects of the Invention]
[0008] According to the present invention, it is possible to provide a method for determining the electron temperature in a plasma process that can determine the electron temperature without adversely affecting the plasma and that is also applicable to film-forming plasma under high gas pressure. [Brief explanation of the drawings]
[0009] [Figure 1] FIG. 1 is a diagram showing an example of a measurement of an emission spectrum of hydrogen plasma (an example actually measured by optical emission spectroscopy (OES)). [Figure 2] FIG. 10 is a diagram showing an example of calculation of an emission spectrum of hydrogen plasma. [Figure 3] FIG. 10 shows the calculation results of normalizing the emission intensity of Fulcher band luminescence with the emission intensity of dissociative continuum light at a wavelength of 200 nm when the electron temperature is 1.0 eV and 5.0 eV. [Figure 4] FIG. 10 shows the relationship between the calculated Fulcher band luminescence intensity normalized by the calculated dissociative continuum luminescence intensity and the electron temperature of hydrogen plasma, and the calculated Fulcher band luminescence intensity normalized by the calculated dissociative continuum luminescence intensity. DETAILED DESCRIPTION OF THE INVENTION
[0010] An embodiment of the method for determining the electron temperature in a plasma process according to the present invention will be described. It should be noted that the present embodiment is specifically described to allow a better understanding of the gist of the invention, and does not limit the present invention unless otherwise specified.
[0011] [Method for determining electron temperature in plasma processes] A method for determining electron temperature in a plasma process according to one embodiment of the present invention includes the steps of: generating hydrogen plasma by introducing hydrogen molecules into the plasma (hereinafter referred to as the "first step"); measuring Fulcher band emission and dissociative continuum emission originating from the hydrogen molecules in the emission spectrum of the hydrogen plasma (hereinafter referred to as the "second step"); normalizing the intensity of the Fulcher band emission with the intensity of the dissociative continuum emission (hereinafter referred to as the "third step"); and determining the electron temperature of the hydrogen plasma from a relationship between the Fulcher band emission intensity normalized with the emission intensity of the dissociative continuum emission, which has been calculated in advance, and the electron temperature of the hydrogen plasma, and the value of the Fulcher band emission intensity normalized with the emission intensity of the dissociative continuum emission (hereinafter referred to as the "fourth step").
[0012] "First step" In the first step, hydrogen plasma is generated by introducing hydrogen molecules into the plasma. For example, the hydrogen plasma used to determine the electron temperature is generated by a parallel-plate capacitively coupled high-frequency discharge. The discharge conditions are controlled by adjusting the pressure of the hydrogen gas introduced into the plasma, the high frequency power, the excitation frequency, the electrode spacing, and the like. The gas pressure of hydrogen introduced into the plasma is not particularly limited, but is preferably, for example, 0.03 Torr or more and 30 Torr or less. When the gas pressure is equal to or more than the lower limit, sufficient plasma can be generated. When the gas pressure is equal to or less than the upper limit, sufficient plasma can be generated.
[0013] The high frequency power is not particularly limited, but is preferably, for example, 10 W or more and 1 kW or less. When the high frequency power is equal to or more than the lower limit, sufficient plasma can be generated. When the high frequency power is equal to or less than the upper limit, sufficient plasma can be generated.
[0014] The excitation frequency is not particularly limited, but is preferably, for example, 13.56 MHz or more and 100 MHz or less. When the excitation frequency is equal to or more than the lower limit, sufficient plasma can be generated. When the excitation frequency is equal to or less than the upper limit, sufficient plasma can be generated.
[0015] The electrode spacing is not particularly limited, but is preferably, for example, 5 mm or more and 50 mm or less. When the electrode spacing is equal to or more than the lower limit, sufficient plasma can be generated. When the electrode spacing is equal to or less than the upper limit, sufficient plasma can be generated.
[0016] "Second step" In the second step, the Fulcher band emission and dissociation continuum emission originating from hydrogen molecules are measured from the emission spectrum of the hydrogen plasma. The emission spectrum of hydrogen plasma consists of continuum emission, molecular band emission, and atomic line emission. Among these emission spectra, this embodiment focuses on the Fulcher band emission and radiative dissociation continuum originating from hydrogen molecules. This is because the Fulcher band emission and radiative dissociation continuum (1) have strong emission intensities, (2) show good agreement with the calculated spectrum described below, and (3) have clear dependence on the electron temperature.
[0017] The emission spectrum of the hydrogen plasma is measured by optical emission spectroscopy, which can use, for example, a spectroscope, an optical fiber, and a light-receiving lens.
[0018] "Third step" In the third step, the emission intensity of the Fulcher band emission measured in the second step is normalized by the emission intensity of the dissociative continuum emission measured in the second step.
[0019] The emission intensity of the Fulcher band emission is preferably the average value of the emission intensity at a wavelength of 585 nm or more and 645 nm or less, and more preferably the average value of the emission intensity at a wavelength of 595 nm or more and 625 nm or less. When the wavelength of the Fulcher band emission is equal to or greater than the lower limit, the emission intensity can be measured sufficiently. When the wavelength of the Fulcher band emission is equal to or less than the upper limit, the emission intensity can be measured sufficiently.
[0020] The wavelength of the dissociation continuum is preferably 200 nm or more and 500 nm or less, and more preferably 200 nm or more and 350 nm or less. If the wavelength of the dissociation continuum is below the lower limit, the sensitivity of the spectrometer used for measurement decreases, making measurement difficult. If the wavelength of the dissociation continuum exceeds the upper limit, overlap with other molecular band emissions makes measurement difficult.
[0021] "Fourth step" In the fourth step, the electron temperature of the hydrogen plasma is determined from the relationship between the Fulcher band emission intensity normalized by the dissociative continuum emission intensity obtained in advance and the electron temperature of the hydrogen plasma, and the value of the Fulcher band emission intensity normalized by the dissociative continuum emission intensity obtained in the third step.
[0022] Specifically, the emission spectrum of hydrogen molecules is calculated using the collisional-radiative (CR) mode, and a graph is created showing the relationship between the Fulcher band emission intensity normalized by the dissociative continuum emission intensity and the electron temperature of the hydrogen plasma. The Fulcher band emission intensity normalized by the dissociative continuum emission intensity obtained in the third step is applied to this graph to determine the electron temperature of the hydrogen plasma.
[0023] According to the method for determining the electron temperature in a plasma process of this embodiment, the electron temperature can be determined without adversely affecting the plasma, and it can also be applied to film-forming plasma under high gas pressure. [Example]
[0024] The present invention will be explained in more detail below with reference to examples, but the present invention is not limited to the following examples.
[0025] [Example] The hydrogen plasma used to determine the electron temperature was generated by a parallel-plate capacitively coupled radio-frequency discharge. The discharge conditions were as follows: hydrogen gas pressure introduced into the plasma was 0.03 Torr to 30 Torr, radio-frequency power was 10 W to 1 kW, excitation frequency was 13.56 MHz to 100 MHz, and electrode spacing was 5 mm to 50 mm.
[0026] The emission spectrum of the generated hydrogen plasma was measured by optical emission spectroscopy. Specifically, the emission spectrum was measured using a Hamamatsu Photonics spectrometer (PMA-12 C10027). The measurement wavelength range was 200 nm to 950 nm, the wavelength resolution was 2.0 nm, and sensitivity calibration was performed. Figure 1 shows an example of a measurement of the emission spectrum of hydrogen plasma (measured by optical emission spectroscopy). In this example, we focused on the Fulcher band emission and the radiative dissociation continuum originating from hydrogen molecules.
[0027] Figure 2 shows an example of a calculation of the emission spectrum of hydrogen plasma. A collisional-radiative (CR) model was used to calculate the emission spectrum. Here, we assumed an electron temperature of 2.3 eV, a molecular vibrational temperature of 4200 K, a rotational temperature of 300 K, and a wavelength resolution of 2.0 nm. To compare with the measurement results shown in Figure 1, the intensity of the Fulcher band emission was normalized by the intensity of the dissociative continuum at a wavelength of 200 nm. The calculation results shown in Figure 2 were compared with the measurement results shown in Figure 1, and it was confirmed that the shape and intensity of the emission spectrum were well reproduced. Specifically, the measurement results shown in Figure 1 and the calculation results shown in Figure 2 were found to be nearly identical in the spectral shape of the Fulcher band emission (around 595 nm to 625 nm) and the rate of change (slope) of the intensity with wavelength of the dissociative continuum (around 200 nm to 500 nm).
[0028] As mentioned above, it was confirmed that the measurement results shown in Figure 1 and the calculation results shown in Figure 2 are almost identical. Therefore, the dependence of the emission spectrum on the electron temperature was investigated by calculation. When the electron temperature was 1.0 eV and 5.0 eV, the emission intensity of the Fulcher band emission was normalized by the emission intensity of the dissociative continuum at a wavelength of 200 nm. The results are shown in Figure 3. From the results shown in Figure 3, it was found that the emission intensity of the Fulcher band emission changes depending on the electron temperature. On the other hand, it was confirmed that the emission intensity of the dissociative continuum remains almost unchanged. Specifically, it was found that the emission intensity of the Fulcher band emission increases with an increase in the electron temperature.
[0029] Here, we focused on the fact that the Fulcher band emission intensity changes with the electron temperature. Figure 4 shows the relationship between the calculated Fulcher band emission intensity normalized by the emission intensity of the dissociative continuum and the electron temperature of hydrogen plasma, as well as the calculated Fulcher band emission intensity normalized by the emission intensity of the dissociative continuum. In Figure 4, the Fulcher band emission intensity is calculated by normalizing the average emission intensity at wavelengths of 595 nm to 625 nm with the dissociative continuum at a wavelength of 200 nm or 350 nm when the electron temperature is changed. The results shown in Figure 4 indicate that the electron temperature (Te) can be determined from the Fulcher band emission intensity normalized by the emission intensity of the dissociative continuum. Specifically, the emission spectrum of hydrogen plasma was measured, and the average intensity of the Fulcher band emission (I_Fulcher) in the wavelength range of 595nm to 625nm was calculated. Then, the average intensity of the Fulcher band emission (I_Fulcher) was normalized by the intensity of the dissociative continuum emission at wavelengths of 200nm or 350nm (I_200nm, I_350nm) to derive the values (I_Fulcher / I_200nm, I_Fulcher / I_350nm). The electron temperature Te can be determined from the point where this value intersects with the curve in the figure.
[0030] Table 1 shows examples of determining the electron temperature (Te) for various hydrogen plasmas. The hydrogen plasmas for which the electron temperature was determined were generated under different gas pressure conditions. In this example, the gas pressures were 0.2 Torr and 5.0 Torr.
[0031] [Table 1]
[0032] From the results shown in Table 1, the value of I_Fulcher / I_200nm = 0.0718 was measured at a gas pressure of 0.2 Torr, and from this value, the electron temperature (Te) was determined to be 3.9 eV. Also, at a gas pressure of 0.2 Torr, the value of I_Fulcher / I_350nm = 0.655 was measured for the intensity of dissociative continuum light at different wavelengths, and from this value, the electron temperature (Te) was determined to be 4.0 eV.
Claims
1. generating a hydrogen plasma by introducing hydrogen molecules into the plasma; measuring Fulcher band luminescence and dissociation continuum luminescence originating from the hydrogen molecules in the emission spectrum of the hydrogen plasma; normalizing the emission intensity of the Fulcher band emission with the emission intensity of the dissociative continuum; a step of determining the electron temperature of the hydrogen plasma from the relationship between the emission intensity of the Fulcher band emission normalized by the emission intensity of the dissociative continuum light obtained by pre-calculation and the electron temperature of the hydrogen plasma, and the value of the emission intensity of the Fulcher band emission normalized by the emission intensity of the dissociative continuum light.
2. 2. The method for determining an electron temperature in a plasma process according to claim 1, wherein the emission intensity of the Fulcher band emission is an average value of emission intensity in the wavelength range of 585 nm to 645 nm.
3. 2. The method for determining an electron temperature in a plasma process according to claim 1, wherein the wavelength of the dissociative continuous light is 200 nm or more and 500 nm or less.