JPH0513194A - Method for measuring negative ion in plasma - Google Patents

Abstract

PURPOSE:To provide a method for measuring a negative ion in a plasma, where density of the negative ion in the plasma is locally measured with accuracy. CONSTITUTION:A probe P1 is disposed in a desired measuring portion in a plasma. A laser 5 is irradiated in the vicinity of the probe P1 of the plasma. An electron, which is photodesorbed from a negative ion in the plasma by radiation of the laser 5, is collected by the probe P1 so that a photodesorbed electron current is measured. Consequently, density of the negative ion is locally measured.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a process plasma used for etching and film formation, a negative ion source, etc.
The present invention relates to a method for measuring negative ions in plasma for measuring the state of various negative ions including radicals.

[0002]

2. Description of the Related Art The inventors of the present invention have been conducting research to measure the state of process plasma, and, for example,
Plasma generation method including negative ions and probe measurement method in this plasma (H. Amemiya: J. Phys. D, 23 (1990)
999.) or the results of spectroscopic measurement of negative ions (O ⁻ , O ₂ ⁻ ) in RF discharge such as oxygen using laser light galvano method (T.Suzuki and T.Kasuya: Phys.Rev., A36).
(1987) 2129.) and so on.

As a conventional method for measuring negative ions by the optical galvanometer method, halogen and its derivatives having a large electron affinity have been measured. In these cases, short-wavelength lasers are required, and pulsed lasers that enable this have been mainly used. For example ^I -, Cl ^- for the N ₂ excitation pulse dye ^lasers, Cl ^-, BCl ₃ - For N ₂ pulsed laser, Cl ^- include excimer laser pumped pulsed dye laser for. In addition, as a research on the negative ion source for neutral beam irradiation for heating of fusion plasma, the research on the H ⁻ source has been performed conventionally.
^- it is pulsed measured using a ruby laser has been carried out. Further, measurement in plasma containing SF ₆ ⁻ using N ₂ laser is also performed.

Incidentally, in the conventional negative ion measurement by the optical galvano method, photo-desorbed electrons are collected by using an anode or a cathode, and a change in impedance is observed, not a local density measurement. .

Among the above-mentioned lasers, the N ₂ laser has a short pulse width and an unstable pulse height, so that it is difficult to absorb plasma noise and obtain data with a good signal-to-noise ratio (S / N). . In general, in order to increase the reliability of measurement, continuous (cw) oscillation light is often preferable to pulsed light because noise suppression means such as various filters and lock-in methods can be applied. The above measurement for O ⁻ is
Using dye CW laser, output 200mW, S / N ratio about 300
Has been obtained.

[0006]

However, in the above-mentioned conventional method, since the anode or the cathode is used to collect the electrons photo-desorbed from the negative ions, the electrons reach these electrodes. Since it is consumed by the collision effect up to that point, the S / N ratio is poor, and there is a problem that it is not possible to obtain locality without knowing which position in the plasma is being measured.

Further, although negative ions in plasma may have many components, it is unclear which negative ion species is being measured by the conventional method, which may cause a large error in determining the negative ion density. There was a problem that there is a property.

The present invention has been made in consideration of such conventional circumstances, and an object of the present invention is to provide a method for measuring negative ions in plasma capable of accurately measuring the density of negative ions in plasma. is there.

[0009]

That is, according to the invention of claim 1, a probe is arranged at a desired measuring portion in plasma, and a laser is irradiated in the vicinity of the probe of the plasma, and the laser is irradiated. Electrons photo-desorbed from the negative ions in the plasma are collected by the probe and the photo-desorption electron current is measured to measure the density of the negative ions.

According to the second aspect of the present invention, plasma is sequentially irradiated with lasers of a plurality of types of wavelengths, and photodesorption electron currents corresponding to the irradiations of these lasers are measured. It is characterized in that the negative ion species in the plasma are identified and the density is determined from the photodetachment cross-section of the above.

[0011]

According to the first aspect of the present invention having the above-mentioned structure, a laser is irradiated in the vicinity of the probe arranged at a desired measuring portion in the plasma, and the electrons photo-desorbed from the negative ions in the plasma are collected by the probe. The density of negative ions is measured by measuring the photodetached electron current. Therefore, not only locality can be obtained, but also by appropriately changing the probe voltage and applying an optimum positive bias to the plasma potential, the sensitivity can be increased, and the S / N ratio can be improved to increase the density of negative ions. Can be locally and accurately measured.

According to the second aspect of the invention, plasma is irradiated with two or more types of cw or pulse lasers by wavelength sweeping, and the photoresponse causes electrons to be desorbed from each negative ion species in the plasma. Measure the current based on. Then, the negative ion species in the plasma are identified and the density is determined from the photodetachment cross section of each negative ion species. Therefore, even when a plurality of negative ion species are present in the plasma, the negative ion species can be identified and the density thereof can be accurately obtained.

[0013]

The details of the method for measuring negative ions in plasma according to the present invention will be described below with reference to the drawings.

FIG. 1 shows the structure of a measuring system according to an embodiment of the present invention. In the figure, 1 is a laser light source. In this embodiment, the laser light source 1 is a cw argon ion laser and a cw dye. A laser was used.

In the figure, reference numeral 2 denotes a discharge tube formed in a cylindrical shape. Inside the discharge tube 2, a cathode K and an anode A are provided along the tube axis direction, and a voltage is applied from a power supply Va. By these, these cathode K and anode A
Is configured to generate a plasma between and. Further, in the discharge tube 2, as shown by an arrow in the figure, a probe P for electron collection which is configured to be movable in the radial direction of the discharge tube 2 is inserted. In this embodiment, the two probes P1 and P2 are provided with a space in the tube axis direction of the discharge tube 2.
Is provided.

At both ends of the discharge tube 2, light transmitting windows 3, 4 are provided.
Is provided, and the laser 5 from the laser light source 1 is provided.
Is configured to enter the discharge tube 2 through the light transmission window 3 through the chopper 6, the mirror 7, and the condenser lens 8 and to exit to the outside through the light transmission window 4 on the opposite side. The mirror 7 is configured so that its angle can be adjusted.
The laser 5 can be focused on a desired site in the discharge tube 2 by adjusting the angle of the laser beam 5 to direct the laser 5 in a desired direction in the discharge tube 2 and moving the focusing position by the lens 8 in the tube axis direction. Is configured.

In the present embodiment, the probe P1 (or the probe P2) is arranged at a desired measurement site by using the measuring system having the above-mentioned structure, and the laser 5 is focused by the mirror 7 and the lens 8 in the vicinity of the probe P1. Then, the laser 5 is applied to the plasma in the vicinity of the probe P1. Then, the probe P is supplied by the probe power source Vp.
A predetermined bias voltage is applied to 1, and the electrons photo-desorbed from the negative ions by the irradiation of the laser 5 are collected by this probe P1. A change in current (light-exiting electron current) ΔIp due to irradiation of the laser 5 is detected by the lock-in amplifier 9 as a change in voltage drop across the resistor Rp. That is, the lock-in amplifier 9 inputs .DELTA.Ip.multidot.Rp through the noise filter 10 and extracts the difference in the presence or absence of laser light by phase sensitive detection using the signal from the chopper 6 as a reference signal.

In the system of FIG. 1, the input of the lock-in amplifier 9 is switched to ΔI · Ra from the anode A, and for comparison, the electrons photo-desorbed from the negative ions are collected by the anode A. It is configured so that it can also be measured. Further, the laser 5 transmitted through the discharge tube 2 is reflected by the mirror 11 and is incident on the power monitor 12, so that the power of the transmitted light can be monitored.

Here, when the laser 5 with power P _k and frequency ν _k is focused on a certain point in the plasma, the negative ions in that portion cause photodesorption to generate electrons, and the probability is σ. It is given by _i (hν _k ). Here, h is the Planck's constant, and σ _i is the photodetachment cross section of the i-type negative ion. When each negative ion species density in the plasma and n _i, the (k
The generation rate Γ _{k of} electrons generated by photodetachment by the laser 5 (seed) is given by the following equation.

[0020] Here, b is the radius of the beam of the laser 5.
The physical meaning of the equation (1) is that photons per unit area per second (P
_k / _πb ² _{hν k)} is that the elimination light electrons from the negative ions of the probability sigma _i in unit volume per n _i.

Assuming that the length of the probe P that collects electrons is L, the collection volume is πb ² L, and the increased current ΔIp flowing through the probe P is such that e is the electron charge. Given in. This equation (2), a negative ion species density from △ Ip can be calculated n _i. In addition,
In the equation (2), the beam diameter b disappears.

Next, with reference to FIGS. 2 to 5, the results of various measurements performed in the measurement system of FIG. 1 will be described.

FIG. 2 shows a probe current (ip) when a cw argon ion laser is used as the laser light source 1.
) -Voltage (Vp) characteristics and optical galvano signal ΔIp. Laser input power is converted to 100mW and plotted. As you can see from this graph,
ΔIp increases with Vp. In particular, the increase above the space potential Vs is because the speed of photodetached electrons is increased due to the acceleration, and this amplification action improves the S / N ratio. If Vp is further increased, the probe current enters the dielectric breakdown region and the effect of photodesorption electrons disappears due to the electron multiplication effect,
ΔIp drops to zero. Therefore, the efficiency of electron collection can be improved by selecting an appropriate Vp.

The graph of FIG. 3 shows changes in the optical galvano signal ΔIp at a constant Vp when the laser spot is swung in the vertical and horizontal directions with respect to the probe. Laser input power is converted to 100mW. As can be seen from this graph, the signal can be obtained most efficiently when the laser spot is focused and irradiated on the sheath end around the probe P. This is because the electrons photo-desorbed from the negative ions existing at the end of the sheath immediately enter the sheath, are accelerated by the sheath voltage, and reach the probe P. As described above, in this embodiment, the sensitivity can be increased as compared with the case where the galvano signal is picked up from the anode A or the cathode K. Also, the probe P and the laser 5
By moving the, local measurements can be performed.

In the graph of FIG. 4, the laser light source 1 is c
2 shows changes of the optical galvano signal ΔIp from the probe P and the optical galvano signal ΔI from the anode A with respect to the discharge current I when the w argon ion laser is used.
The laser input power is converted to 100mW. As can be seen from this graph, the optical galvano signal ΔIp from the probe P is larger than the optical galvano signal ΔI from the anode A by almost two digits. In addition, the difference between ΔIp and ΔI seen in the current dependence is that ΔIp represents the average negative ion density of the entire discharge, and ΔIp is the local negative ion density near the probe. This is because it represents the ion density.

In the graph of FIG. 5, the laser light source 1 is c
It shows changes in the optical galvano signal output ΔIp at a constant discharge current and a constant probe voltage when the wavelength of the laser is swept using a w dye laser. In the case of a dye laser, since the output differs depending on the wavelength, the optical galvano signal ΔIp of each wavelength is converted into a constant power and plotted. In this case, the laser power at the _jth frequency ν _j ,
An optical galvanometer signal respectively P _j, △ expressed and _{I j, △ I j = (} P j / hν j) {σ 1 (ν j) n 1 + σ 2 (ν j) n 2} (3) becomes equation To get Here, it is considered that the main negative ions in the oxygen plasma are O ⁻ and O ₂ ⁻ , n ₁ and n ₂ are their respective densities, and σ ₁ (ν _j ) and σ ₂ (ν _j ) are at ν _j . It is the photodetachment cross section of each ion.

From equation (3), ΔI at a plurality of ν _{j
From j} , n ₁ and n ₂ can be determined by the method of least squares.
In principle, it can be determined from two equations, but the accuracy is improved by using the output of many wavelengths.

Next, a second embodiment will be described with reference to FIG.

In FIG. 6, 20 is a diode laser, and 21 is a laser drive circuit for driving the diode laser 20. This diode laser 2
0 is, for example, wavelength 0.9 μm, 0.83 μm, 0.78 μm
, 0.67 μm is used. Further, the diode laser 20 is provided with a temperature control mechanism using a Peltier element, and the laser drive circuit 21 is provided with a phase sensitive detection function by an optical modulation mechanism. As a result, the laser output is stabilized and the sensitivity is improved. The temperature of the diode laser 20 is variably controlled at 20 ° C to 40 ° C.

The laser drive circuit 21 includes an oscillator 22.
The diode laser 20 can be turned on and off at the same frequency by introducing a sine wave signal of several kHz from the oscillator 22. Therefore, the signal of the oscillator 22 is input to the lock-in amplifier 9 as a reference signal. Further, when the output of the lock-in amplifier 9 is drawn on a recording paper, fluctuations occur in the zero level when there is no laser irradiation due to plasma noise. Therefore, in this embodiment, for the purpose of clarifying the output due to laser irradiation, a shutter 23 is inserted between the mirror 7 and the lens 8 and the irradiation of the laser 24 is switched about every two minutes.

Thus, in this embodiment, the pressure is 0.6T.
Oxygen plasma is generated with orr and discharge current of 1 mA, and this oxygen plasma is irradiated with a laser 24 having a wavelength of 0.78 μm, so that the average output signal from the anode A is ΔI =
We were able to obtain 4nA. At this time, in order to improve the sensitivity, a mirror 25 is provided on the outlet side of the discharge tube 2 to reflect the laser 24 emitted through the discharge tube 2 and introduce it again into the discharge tube 2. Incident laser light output was increased almost twice.

Next, the same measurement was carried out by replacing the diode laser 20 with a wavelength of 0.83 μm with the result that the wavelength was 0.78 μm.
The output signal of 1/3 was obtained compared with the case of m.

The main negative ions in the oxygen plasma are O
^- and O ₂ ^- is. Here, the density of O ⁻ is n ₁ , O ₂ ⁻
The density of n ₂ , the power of the laser of wavelength 0.78 μm is P ₁ , the frequency is ν ₁ , the power of the laser of wavelength 0.83 μm is P ₂ , the frequency is ν ₂ , and O for each laser is
^- an optical de-transection area sigma ₁ of _{(ν 1), σ 1 (} ν 2), O 2 for each laser ^- of the optical de-transection area σ ₂ (ν _1),
σ ₂ (ν ₂ ), the output for each laser is ΔI ₁ , ΔI
_Then , the following equation holds.

ΔI ₁ = (P ₁ / hν ₁ ) {σ ₁ (ν ₁ ) n ₁ + σ ₂ (ν ₁ ) n ₂ } (4a) ΔI ₂ = (P ₂ / hν ₂ ) {σ ₁ ( _{ν 2) n 1 + σ 2} (ν 2) n 2} (4b) this (4a), in expression _{(4b), P 1 = 40mW P 2 =} 60mW hν 1 = 1.59eV hν 2 = 1.49eV σ 1 (ν ₁ ) = 4.2 × 10 _-18 cm ² σ ₂ (ν ₁ ) = 1.0 × 10 _-18 cm ² σ ₁ (ν ₂ ) = 2.2 × 10 _-18 cm ² σ ₂ (ν ₂ ) = 0.85 × 10 _-18 obtaining _{^{cm 2 △ I 1 = 4 nA}} △ I 2 = (△ I 1/3) substituting ^{_{nA n 1 = 2.1 × 10 9}} cm -3 n 2 = 3.3 × 10 9 cm -3. By this, each density of O ⁻ and O ₂ ⁻ could be determined.

Although two types of diode lasers 20 are used in the above embodiment, ΔI may be measured by using one or more lasers having other wavelengths. In such a case, n ₁ and n ₂ can be determined by the least squares method from two or more equations for two unknowns. Although a commercially available near-infrared diode is used in the above-mentioned embodiment, the range of application to negative ion species will be increased if the wavelength shortening by the double wave of the diode laser is realized. The method is economical because it can be realized with relatively low equipment cost.

In this embodiment, the photodetachment electron current was measured by the anode A, but the probes P1 and P2 were used.
The same measurement can be performed by using, and in this case, the same effect as that of the above-described embodiment can be obtained.

FIG. 7 shows the structure of a measuring system in the third embodiment using a pulse laser. In FIG. 7, reference numeral 30 denotes a YAG laser.
A laser 31 of 1.06 μm, 0.53 μm, 0.353 μm, etc. is emitted. Reference numeral 32 is a dye laser, which has a wavelength ranging from near ultraviolet to near infrared. This dye laser 32 is a YAG
From the laser 30, a mirror 33, a condenser lens 34,
It is excited by the laser 31 which is incident through the mirror 35. In the figure, A1 and A2 are dampers.

In such a system, when the laser 36 having energy W _k and frequency ν _k is focused on a certain point in the plasma, the negative ions in that portion cause photodetachment to generate electrons, The cross-sectional area is given by σ _i (hν _k ). Here, the density of each negative ion species in the plasma is defined as n _i
Then, the amount of current increase based on the number of electrons generated by photodetachment by the laser light (k type) is given by the following equation.

ΔI _i = I _i [1−exp {−σ _i (ν _k ) W _k / (hν _k πb ² )}] (5) where I _i is the maximum level of ΔI _i , and b is The beam radius. The physical meaning of the above formula is photons W _k / h per unit area.
ν _k πb ² is that light leaving the electrons from the unit volume per n _i.

As the energy of light increases, ΔI _i
Becomes saturated towards I _i . Here, a linear region where saturation does not occur is applied. Let L be the length by which the probe and electrode collect electrons. The collection volume is πb.
² L, which flows to the probe and electrode, but the surrounding charge Q Given in. Here, Q _i is the charge component due to the i-type negative ion, and where e is the electron charge, Q _i = n _i πb ² Le. In the system of FIG. 7, assuming that the laser 36 and the negative ion density are constant over the length of the cathode K of the discharge tube 2, the photo-desorbed electrons are collected by the anode A. Va is a voltage for applying a bias to the anode A, and a change amount ΔI of the current due to laser irradiation is taken out as a change amount of the voltage drop generated across the resistor Ra, and ΔI · R is input to the digital oscilloscope 37. A signal from the YAG laser 30 is input to the digital oscilloscope 37, and the output signal waveform S is extracted using this signal as a trigger signal.

It should be noted that the same measurement can be performed when signals are picked up from the probes P1 and P2, but with a pulsed laser, when the light hits the probes P1 and P2 directly, ablation occurs, so care must be taken to apply it only near the sheath end. There is.

FIG. 8 shows the output of the digital oscilloscope 37 when the measurement is performed in the system of FIG. The laser wavelength is 0.53 μm, the energy is 4 mJ, and the laser pulse width is 10 ns, but the pulse waveform is extended to 40 μs due to the influence of the sheath in front of the electrode placed in the plasma.

Similar to the above-described embodiment, the main negative ions in oxygen plasma are O ⁻ and O ₂ ⁻ , the density of O ⁻ is n ₁ and the density of O ₂ ⁻ is n ₂ , and the wavelength is 1.06 μm. , 0.53μ
The energies of the respective lasers of m are W ₁ and W ₂ , the frequencies thereof are ν ₁ and ν ₂ , the photodetachment cross sections of O ⁻ with respect to the respective laser beams are σ ₁ (ν ₁ ), σ ₁ (ν ₂ ), The photodetachment cross section of O ₂ ^{− to} the laser light is σ ₂ (ν ₁ ), σ
₂ (ν ₂ ), output for each laser irradiation is Q ₁ , Q ₂
Then, the following equation holds.

Q ₁ = (W ₁₁ / hν ₁ ) {σ ₁ (ν ₁ ) n ₁ + σ ₂ (ν ₁ ) n ₂ } + (W ₁₂ / hν ₂ ) {σ ₁ (ν ₂ ) n ₁ + σ ₂ (Ν ₂ ) n ₂ } (7) The same equation holds for Q ₂ . Here, W _ii
Is the energy of the inner wavelength ν _i component of the i-th laser light. In equation (7), W ₁₁ = 20mJ W ₂₁ = 4mJ W ₁₂ = W ₂₁ = 0 hν ₁ = 1.16eV hν ₂ = 2.34eV σ ₁ (ν ₁ ) = 0 σ ₂ (ν ₁ ) = 0.8 × 10 ₋ By substituting ₁₈ cm ² σ ₁ (ν ₂ ) = 6.5 × 10 _-18 cm ² σ ₂ (ν ₂ ) = 1.5 × 10 _-18 cm ² , n ₁ and n ₂ can be calculated. Each density of O ⁻ and O ₂ ⁻ can be determined.

Although the YAG laser 30 is used in this embodiment, a laser of another wavelength, a dye laser or the like may be installed and pumped by a YAG or N ₂ laser or the like, and in that case. From each equation for two or more wavelengths for two unknowns, the least squares method is applied to n ₁ , n ₂
Can be determined.

The major feature of the pulse method as described above is that instantaneous measurement is possible, and local measurement by combination with a probe is also possible. Also, a commercially available YAG laser,
It is particularly suitable for measurement of oxygen negative ions and the like by using a dye laser, but if shorter wavelengths become available, the range of application will increase to negative ion species with higher attachment energy.

In the above embodiment, if the oxygen plasma contains O ₃ ⁻ , O ₄ ^−, etc., by using a laser with four or more wavelengths, four or more formulas (4) or (7) can be obtained. The density of each negative ion species can be determined from a similar equation.

As described above, according to these embodiments,
It is possible to measure the local density of negative ions in the plasma and, if multiple negative ion species are present, to identify the ion species and measure the density, improving the precision and reproducibility of the plasma. Can be planned.

The method of the present invention can be applied to the measurement of negative ions including radicals in plasmas for processes such as etching, negative ion sources, electronless plasmas (plasma containing only ions), and the measurements are also compared. It is simple and can be done by a small system. Further, by applying the hollow cathode galvatron as a standard tube for measurement, a portable standardized discharge tube is provided and can be used for signal comparison in various plasma measurements.

In the method of the present invention, argon ions,
Various lasers such as continuous lasers such as dyes and laser diodes, or pulsed lasers such as excimer, YAG, dyes and ruby can be used. When continuous lasers are used, high precision measurement is performed using lock-in technique. When a pulsed laser is used, it is possible to perform instantaneous measurement with a digital oscilloscope, averaging measurement, and high precision measurement with a Boxker integrator.

[0051]

As described above, according to the method of the present invention, the local density of negative ions in plasma is measured, and if there are a plurality of negative ion species, the ion species are identified. The density can be measured, and the precision and reproducibility of plasma can be improved.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a measurement system according to an embodiment of the present invention.

FIG. 2 is a graph showing a relationship between an optical galvanometer signal and probe characteristics.

FIG. 3 is a graph showing laser light irradiation position dependency of an optical galvano signal.

FIG. 4 is a graph showing a relationship between an optical galvano signal from an anode and an optical galvano signal from a probe.

FIG. 5 is a graph showing an optical galvano signal in the case of wavelength sweep of a dye laser.

FIG. 6 is a diagram showing a configuration of a measurement system of an example using a cw diode laser.

FIG. 7 is a diagram showing a configuration of a measurement system of an example using a pulse laser.

FIG. 8 is a graph showing an optical galvano signal by a pulse laser.

[Explanation of symbols]

1 laser light source 2 discharge tubes 3,4 light transmission window 5 laser 6 chopper 7,11 Mirror 8 lenses 9 Lock-in amplifier 10 noise filter 12 Power monitor K cathode A anode Va, Vp power supply P1 and P2 probes Ra, Rp resistance

Claims (4)

[Claims] 1. A probe is arranged at a desired measuring portion in plasma, and a laser is irradiated to the vicinity of the probe of the plasma, and electrons emitted from the negative ions in the plasma are desorbed by the laser irradiation. A method for measuring negative ions in plasma, characterized in that the density of negative ions is measured by collecting with the probe and measuring a photodetached electron current. 2. A plasma is sequentially irradiated with lasers of a plurality of types of wavelengths, and photodesorption electron currents corresponding to the irradiations of these lasers are measured. A method for measuring negative ions in plasma, which comprises identifying negative ion species in plasma and determining density. 3. The laser is a cw laser, a dye laser, an argon ion laser, or a semiconductor laser.
Alternatively, the method for measuring negative ions in plasma according to claim 2. 4. The laser is a pulse laser, which is a YAG laser, a ruby laser, an excimer laser having a wavelength range from near ultraviolet to infrared, or a dye laser. Item 2. A method for measuring negative ions in plasma according to Item 2.