JPH0547041B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to an electronic counting device that counts the number of electrons emitted from a sample upon irradiation with light energy.

(Prior art) Conventionally, as a method for measuring the thickness of an oxide film formed on the surface of a sample such as a semiconductor, the sample surface is irradiated with light, and the light is emitted to the outside through the thin film on the sample. A method is known in which the film thickness is measured by counting the electrons generated by an electron detector (Japanese Patent Application Laid-open No.
60-262005, etc.).

(Problems to be Solved by the Invention) However, in such conventional electronic counting devices, it is necessary to optimize the amount of light irradiated onto the sample, the anode voltage and grid voltage of the electron detection section, etc. before starting measurement. However, if the temperature, atmospheric pressure, and humidity change during measurement, the optimal measurement conditions will deviate from the initial settings, and if you are not aware of this and take measurements, you may not get good measurement results. I had a problem with not being able to get it.

(Means for Solving the Problems) The present invention has been made in view of the above-mentioned conventional problems, and it is possible to always perform measurement under optimal measurement conditions without being affected by changes in the environmental conditions of the measurement location. The purpose of the present invention is to provide an electronic counting device that can obtain highly reliable measurement results.

In order to achieve this object, the present invention provides an electronic counting device that irradiates a sample with light and introduces electrons emitted from the sample into an electron detection section to count the number of electrons. a detection means for detecting the temperature, atmospheric pressure and humidity of the detection means; an environment determination means for determining and outputting that the detection value of the detection means has changed beyond a predetermined threshold value; and an environment determination means for measuring the elapsed time from the start of measurement. , a time measuring means that outputs an output when a predetermined set time is reached; and a display means that displays a request for initial settings of the light amount, anode voltage, and grid voltage when receiving the output of the environment determining means or the time measuring means. ; is provided.

(Function) According to the electronic counting device of the present invention having such a configuration, for example, the temperature is ±5° C. or more,
±10mmHg for atmospheric pressure, ±10 for humidity
If a change of more than % occurs, the light intensity Lp, anode voltage
A request for initial setting of Va and grid voltages Vg ₁ and Vg ₂ is displayed, informing that optimal measurement results cannot be obtained under the current setting conditions.

If such an initial setting request display is displayed, the initial values can be reset manually or automatically by the operator so that the light intensity, anode voltage, and grid voltage are optimized under the current environmental conditions. As a result, measurement work can always be performed under optimal measurement conditions.

Furthermore, the elapsed time from the start of measurement is monitored. For example, when the elapsed time reaches 4 hours, a request for initial settings is displayed to prevent the reliability of the measurement data from decreasing due to the measurement time being too long. can be prevented.

(Example) FIG. 1 is an explanatory diagram showing an example of the present invention.

In FIG. 1, 1 is an electron detection unit, which has a metal case 3 with a detection window 2 opened at the bottom.
Case 3 is connected to ground. An anode ring 4 is disposed inside the case 3, and an anode voltage Va is applied to the anode ring 4 from a high voltage source 23.
Va can be varied in the range of Va=3.0KV to 4.0KV as necessary.

A first grid electrode 5 is provided on the detection window 2 side of the anode ring 4.
and second grid electrode 6 are sequentially arranged.

The output of the first pulse generator 20 is connected to the first grid electrode 5, and the first pulse generator 20 supplies the first grid electrode 5 with a first grid voltage G1, for example, G1=
100V is applied. Further, the output of the second pulse generator 22 is connected to the second grid electrode 6.
The pulse generator 22 applies, for example, G2=80V as the second grid voltage G2.

First pulse generator 20 and second pulse generator 2
2 is connected to the output of an amplifier 18 which extracts the voltage change of the anode ring 4 via a capacitor C.

The voltage applied to the anode ring 4 falls in a pulsed manner as shown in FIG.
As shown in FIG. 2b, the first pulse generator 20 generates a rectangular wave that becomes a quenching voltage Vq by increasing the first grid voltage G1 by a predetermined voltage ΔV to prevent gas discharge. A pulse is output for a quenching time τ. Here, the voltage ΔV that provides the quenching voltage Vq is set to, for example, ΔV=300V, and therefore, when receiving a voltage pulse, the voltage G1 of the first grid electrode 5 changes from the previous 100V to 400V, which is an increase of 300V. .

On the other hand, when the second pulse generator 22 receives a voltage pulse from the amplifier 18, it generates a square wave pulse that changes the second grid voltage G2 to -30V over a quenching time τ, as shown in FIG. occurs. For example, if the second grid voltage G2 is 80V, a rectangular wave pulse lowered by 110V will be generated.

The action of the rectangular wave pulse over the quenching time τ by the first pulse generator 20 and the second pulse generator 22 is such that when the electrons emitted from the sample 12 approach the anode ring 4, a high electric field near the anode ring 4 is generated. , the electrons are accelerated to cause a gas discharge phenomenon, but the first pulse generator 20 increases the quenching voltage Vq over the quenching time τ.
By increasing the voltage, the potential difference between the anode ring 4 and the first grid electrode 5 is lowered by, for example, ΔV=300V,
This prevents light generated by the gas discharge and secondary electrons from positive ions from reaching the discharge voltage, thereby preventing an avalanche of discharge.

In addition, the output of the rectangular wave pulse lowered to -30V over the quenching time τ of the second pulse generator 22 captures and neutralizes the positive ions generated by the gas discharge accompanied by the amplification effect with the second grid electrode 6. This prevents positive ions from reaching the sample 12 and affecting the photoelectron emission function, and at the same time blocks the introduction of electrons from the outside.

Below the detection window 2 of the electron detection section 1 is a sample stage 13.
sample 12 set on the sample stage 13.
A predetermined single wavelength light is irradiated from the light source device 7. The light source device 7 includes a light source 8 such as a deuterium lamp, a monochromator 9 that shortens the wavelength of light from the light source 8, and further includes slits 10 and 11 before and after the monochromator 9 for adjusting the light intensity. There is. The monochromator 9 has a function of scanning short wavelength light in a predetermined wavelength range, for example, 200 nm to 350 nm, and by this wavelength scanning, measurement results for determining the work relationship of the sample 12, for example, can be obtained.

A light receiving element 15 is installed at the position where the light from the light source device 7 is irradiated on the sample stage 13 when the sample 12 is not set. It is possible to detect the light amount from the light source device 7 before measurement, and based on the measurement results of the light amount measuring means 16, the light amount setting (correction coefficient setting) at the time of initial setting, which will be explained later. It is done.

On the other hand, a counting means 24 is provided as a measuring circuit for counting the number of electrons based on the voltage pulse output from the amplifier 18 of the electron detection section 1.
outputs, for example, the electron counting rate per unit time (cps). The output of the counting means 24 is given to the calculating means 26, and the calculating means 26 calculates the anode voltage Va of the anode ring 4, the grid voltage G1 of the first grid electrode 5, and the It has an arithmetic processing function that suppresses the grid voltage G2 of the second grid electrode 6 to an optimal value at the time of initial setting. Furthermore, since the calculation means 26 is supplied with the measurement output of the light amount measuring means 16, the calculation means 26 corrects and outputs the electronic counting rate N based on the amount of light detected by the light amount measurement means.

Following the calculation means 26, the display means 28 is provided, and the display means 28 is formed on the surface of the sample 12 provided in the operation method 26, which is displayed as it is, or the number of electronic counts obtained by the Countermade means 24. The calculation result of the film thickness T of the oxide film will be displayed.

On the other hand, an environment monitor means 32 is provided to determine changes in the environmental conditions at the measurement location, and a temperature sensor 34, an air pressure sensor 36, and a humidity sensor 38 are connected to the environment monitor means 32.

The environment monitoring means 32 reads and stores the detected values of the temperature sensor 34, atmospheric pressure sensor 36, and humidity sensor 38 as initial values when initializing the light intensity, anode voltage, and grid voltage prior to the start of measurement, and then performs the measurement. In this case, each detected value is compared with an initial value at every predetermined detection cycle, and when a change from the initial value is greater than a predetermined threshold value, an environmental change determination output is generated. Here, the threshold values for determining changes in temperature, atmospheric pressure, and humidity in the environment monitoring means 32 are set to, for example, ±5° C. for temperature, ±10 mmHg for atmospheric pressure, and ±10% for humidity.

According to the present invention, a time measuring means 30 is provided in conjunction with such an environment monitoring means 32, and the time measuring means 30 measures the elapsed time from the start of measurement, and when a predetermined set time, for example 4 hours, is reached. produces a time measurement output when

The outputs of the environment monitor means 32 and the time measurement means 30 are given to the calculation means 26, and when receiving the outputs of the time measurement means 30 and/or the environment monitor means 32, the calculation means 26 requests the display means 28 for initial settings. The display is performed by lighting a lamp or sounding a buzzer. In addition, in this example,
Since the calculation means 26 has an automatic setting function of light intensity, anode voltage, and grid voltage, the time measurement means 30
Alternatively, upon receiving the output from the environment monitoring means 32, the initial settings of the light amount, anode voltage, and grid voltage are automatically performed again.

FIG. 3 is a flowchart showing initial setting request processing based on environmental changes or elapsed time in the embodiment of FIG. 1.

In FIG. 3, when the measurement operation is started after completing the initial settings of the light intensity, anode voltage, and grid voltage prior to measurement, the temperature sensor 34 is
The detected values of the atmospheric pressure sensor 36 and the humidity sensor 38 are stored as initial values T ₀ , P ₀ , W ₀ , and at the same time the timer of the time measuring means 30 is started.

Next, the process advances to step S2, and when a predetermined detection cycle is reached, the temperature sensor 34 and the atmospheric pressure sensor 3 are activated.
6 and the humidity sensor 38, and in steps S3 to S5, the absolute value of the difference between the detected value and the initial value is compared with predetermined threshold values ΔT, ΔP, and ΔW. Furthermore, in step S6, it is determined whether the measurement time of the measuring means 30 has reached the set time.
If any of the sensor detection values exceeds the threshold in steps S3 to S5, the process proceeds to step S7 and the calculation means 2
6, the initial setting request is displayed on the display means 28, and the process proceeds to step S7, where the operator manually operates the initial setting request display or the calculation means 2
Initial settings of light intensity, anode voltage, and grid voltage are performed by automatic control by 6. Of course, step
The same applies when the measurement time reaches the set time in S6.

Next, the setting control of the light amount, anode voltage, and grid voltage performed by the calculation means 26 during initial setting will be explained.

First, the adjustment of the amount of light from the light source device 7 will be explained.
FIG. 4 is a graph showing changes in the amount of light when the monochromator 9 of the light source device 7 scans the sample 12 with single wavelength light in a predetermined wavelength range.

As shown by the solid line A in this graph, in the initial state, the light source 8 and lenses of the light source device 7 are free of dirt, so the amount of light when wavelength scanning is performed by the monochromator 9 is Obtained as the maximum amount of light.

Incidentally, the spectral characteristics of the light source 8 are not constant at each wavelength, and exhibit intensity changes, for example, as shown by curve A in FIG. 4.

Therefore, the calculation means 26 corrects the electron counting rate N so that the variation in the spectral intensity of the light source device 7 becomes a constant reference light amount W ₀ given by the light amount W ₀ at the wavelength λ ₀ , for example.

That is, the wavelength range λ ₀ to λn is measured by the monochromator 9.
Each wavelength λ ₀ , λ ₁ , λ ₂ ,
The amount of light W ₀ at each of ...λi, ...λn,
Find W ₁ , W ₂ , ... Wi, ... Wn and use the correction coefficient K ₀ ~ to correct the measured light amount to the reference light amount W ₀
Find Kn for each wavelength using the following formula.

K ₀ =W ₀ /W ₀ K ₁ =W ₀ /W ₁ K ₂ =W ₀ /W ₂ 〓 Ki = W ₀ /Wi 〓 Kn = W ₀ /Wn ...... (1) On the other hand, in the counting means 24 The measured number of electrons _N0 is the number of electrons in the time excluding the quenching time τ, which is the dead time, as shown in Figure 2, so for example, N= _N0 /(1- _N1・τ) ………(2) However, N: Number of emitted electrons N ₀ : Number of measured electrons τ: Quenching time τ as dead time
The number N of emitted electrons is calculated by correcting the number of electrons.

Then, the calculating means 26 uses the correction coefficient of the corresponding wavelength given by the above equation (1) for the number of emitted electrons N obtained by the above equation (2) obtained at each scanning wavelength, Nt=N・Ki ...(3) Find the true number of emitted electrons Nt obtained by irradiation with the reference light amount W ₀ .

Next, as shown by the broken line B in FIG. When a request for initial setting is displayed based on the output, the monochromator 9 scans a single wavelength light in the wavelength range λ ₀ to λn (automatically or manually) prior to measuring the sample 12, and the light amount measuring means 16 The light quantity characteristics with respect to the scanning wavelength as shown by the broken line B in FIG. 4 are determined by the following.

If the light amount of the broken line B, where the light amount has decreased in this way, is determined in the scanning wavelength range, the denominator W ₀₁ to Wn ₁ in the above equation (1) is replaced with the correction coefficient K ₀ to Kn.
After that, the sample 1 obtained by the counting means 24
Calculate the number of emitted electrons N from the measured number of electrons N ₀ from Eq. is calculated from equation (3) above.

As a result, even if the amount of light irradiated onto the sample 12 decreases due to deterioration of the light source 8 in the light source device 7 or dirt on the optical member, etc., the amount of light is always the same as the amount of light in the initial state (however, the spectral characteristics are corrected to a constant value). The true number of emitted electrons Nt is the same as when irradiated with
can be found.

In addition, in the case of the light amount measuring means 16 shown in FIG. 1, when the amount of light detected by the light receiving element 15 falls below a predetermined threshold level, an alarm signal is output and the optical system is cleaned or the light source 8 is replaced. You may also encourage them.

Next, the setting control of the anode voltage Va and the grid voltages G1, G2 at the time of initial setting in the calculation means 26 will be explained.

FIG. 5 is a flowchart showing the electrode voltage setting process in the calculation means 26.

First, in steps S1 to S4 shown in FIG. 5, the high voltage source 23 is controlled to optimally set the anode voltage Va to be applied to the anode ring 4.

This optimum anode voltage setting process is performed by setting a sample 12 serving as an appropriate reference on a sample stage 13, and emitting electrons by irradiating single wavelength light of a predetermined wavelength from the light source device 7, as shown in FIG. , electrons emitted from the sample 12 are introduced into the electron detection section 1, and the number N of electrons is measured by the counting means 24.

That is, in the initial state, the calculation means 26 adjusts the anode voltage Va from the high voltage source 23 to a voltage variable range of 3.0 to 3.0.
The lowest value Va of 4.0KV is set to 3.0KV, and the anode voltage Va is increased from this initial state as shown in step S1. In response to the increase in anode voltage in step S1, the counting means 24 is applied in the next step S2.
The electron counting rate (number of electrons per unit time) obtained from N is the threshold determined by the background noise.
When it is determined that the electron counting rate N is equal to or higher than _the threshold value _N0 , the anode voltage at this time is detected as the discharge starting voltage Vs, and the next step S3
Proceed to. In step S3, the discharge starting voltage Vs
The voltage obtained by adding half the voltage of the predetermined plateau voltage width Pw is calculated as the optimum anode voltage Va ₀ , and in the next step S4, the high voltage source 23 is controlled so that the optimum anode voltage Va ₀ is obtained. .

Here, when the anode voltage Va is changed, the count value N has the characteristics shown in FIG. That is, the anode voltage
As Va increases, the counting rate N exceeds the threshold N ₀ determined by background noise at a certain voltage.
is obtained, and the voltage exceeding this threshold value N ₀ is detected as the discharge starting voltage Vs. Furthermore, when the anode voltage is increased, the counting rate remains approximately constant within a certain range with respect to changes in the anode voltage, and the range in which the counting rate N remains constant with respect to changes in the anode voltage is called a plateau. Define the voltage width Pw.

Further, when the anode voltage is increased beyond the plateau voltage width Pw, the counting rate N rapidly increases and becomes uncountable.

The plateau voltage width Pw that provides the range of anode voltage in which the counting rate is approximately constant as shown in FIG. 6 is a voltage equal to the change ΔV in the first grid voltage G1 that provides the quenching voltage Vq shown in FIG. given as width.

Therefore, the optimum anode voltage Va ₀ determined in step S3 of FIG _. 5 is set to the voltage at the center of the plateau voltage width Pw in the characteristic curve shown in FIG. 6.

In this way, the anode optimum voltage Va ₀ is the plateau width Pw
When set at the center of Va, the characteristic curve shown in Figure 6 shifts as shown by the broken line due to changes in atmospheric pressure and temperature, but as long as this variation is within ±Pw/2, the optimum anode voltage Va ₀ is plateau width Pw
The counting rate N hardly changes even if the atmospheric pressure or temperature changes, and it is possible to obtain stable counting results even when the atmospheric pressure or temperature changes.

Referring again to FIG. 5, after the setting of the optimum anode voltage Va ₀ is completed through the processing up to step S4, the first grid voltage is adjusted in the next step S5.
Set G ₁ to the initial voltage V ₁ a. For example, since the first grid voltage G1 can be varied within the range of G1=80 to 120V, it is set to _V1a =80V.

With the first grid voltage G1 set to the initial voltage G1= _V1a in this way, the second grid voltage G2 is varied in the range of, for example, 40 to 110V in the next step S6. Then, in the next step S7, the counting rate N when the second grid voltage G2 is varied in step S6 is monitored, and G2=V ₂ a is detected as the second grid voltage G2 at which the counting rate N becomes maximum.
Next, in step S8, it is checked whether the final setting of the first grid voltage G1 has been completed, and if the final setting has not been completed, the process returns to step S9, and a predetermined voltage ΔV ₁ is added to the first grid voltage G1 up to that point. Set the newly added first grid voltage and repeat the steps in the same way.
Repeat steps S6 to S8.

FIG. 7 shows the second grid voltage G2 obtained in steps S6 to S9 in the flowchart of FIG.
It is a characteristic graph showing the relationship between the counting rate N and the count rate N.

The characteristic graph in FIG. 7 shows that the first lattice voltage G1
is varied in seven steps: V ₁ a, V ₁ b, V ₁ c, ...V ₁ g,
It shows the change in the counting rate N when the second grid voltage G2 is varied at each first grid voltage, and as is clear from this characteristic graph, the first grid voltage G1
The second grid voltage G2 that gives the peak values Na, Nb, ... Ne of the counting rate N obtained by varying the second grid voltage G2 when is constant is G2 = V ₂ a, V ₂ b, ...
…Can be detected as V ₂ g.

Referring to Figure 5 again, in step S8
Once the final setting of grid voltage G1 has been determined,
Proceeding to the next step S10, for example, as shown in FIG. 7, a second grid voltage G2, which gives the maximum counting rate Nmax, is selected from among the second grid voltages V ₂ a, V ₂ b, . . . V ₂ g, as shown in FIG.
For example, detect G2=V ₂ e.

Subsequently, in step S11, the first pulse generator _{20 and} the second set in the pulse generator 22.

The process for setting the optimum grid voltage is completed through the above process, and further, in the flowchart of FIG. 5, the electron counting rate is determined by setting the electron detection section 1 to the background counting mode in step S12.

In this background counting mode, as shown in the signal waveform diagram of Fig. 8, the voltage applied to the second grid electrode 6 by the second pulse generator 22 is fixed at -30V to block the introduction of electrons from the outside. Therefore, the mode is set in which only the electrons generated as noise inside the electron detection section 1 are counted.

After counting the counting rate that gives background noise by setting the background counting mode in step S12, in the next step S13, it is compared with the background noise threshold value _N0 ,
If the counting rate N is smaller than the threshold value N ₀ , it is assumed that the grid voltage optimum setting process has been performed normally, and the setting process ends. On the other hand, if the counting rate N in the background counting mode is greater than the threshold value _N0 , the process proceeds to step S14, where an alarm is issued to urge adjustment of the electronic detection section 1 itself.

In the embodiment shown in FIG. 1, when the initial setting request is displayed in response to the output from the environment monitor means 32 or the time measurement means 30, the calculation means 26
In this example, the light intensity, anode, and grid voltage are automatically initialized, but it is of course possible for the operator to manually perform the initial settings upon receiving an initial setting request display. It is.

Further, the threshold values for determining changes in temperature, atmospheric pressure, and humidity in the environment monitoring means 32 in the embodiment shown in FIG. If you want to increase it, you can set a smaller threshold.

(Effects of the Invention) As explained above, according to the present invention, even if there is a change in environmental conditions such as temperature, pressure, humidity, etc. during measurement, if a change in environmental conditions exceeding a set threshold is detected, the initial settings are set. This will immediately indicate that the optimum measurement conditions have been deviated from during the measurement, and the light intensity and each electrode voltage of the electronic detection section will be reset to the optimum conditions. Highly reliable measurement data can always be obtained under optimal measurement conditions.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing one embodiment of the present invention, FIG. 2 is a signal waveform diagram showing the voltage applied to each electrode provided in the electron detection section of FIG. 1, and FIG. A flowchart showing the initial setting process based on changes in environmental conditions and measurement time in the example, FIG. 4 is a graph showing the relationship between the light intensity and the light source wavelength, and FIG. 5 shows the relationship between the anode voltage and the grid voltage in the present invention. Flowchart showing the optimal setting process, No. 6
The figure is a graph showing the relationship between the electron count rate and the anode voltage, FIG. 7 is a characteristic graph showing the relationship between the count rate N and the second grid voltage G2 using the first grid voltage G1 as a parameter, and FIG. FIG. 3 is a signal waveform diagram showing electrode voltage in background counting mode. 1: Electronic detection section, 2: Detection window, 3: Case,
4: anode ring, 5: first grid electrode, 6: second grid electrode, 7: light source device, 8: light source, 9: monochromator, 10, 11: slit, 12: sample, 1
3: sample stage, 15: light receiving element, 16: light amount measuring means, 18: amplifier, 20: first pulse generator, 2
2: Second pulse generator, 23: High voltage power supply, 24:
Counting means, 26: Calculation means, 28: Display means, 3
0: Time measurement means, 32: Environment monitoring means, 3
4: temperature sensor, 36: atmospheric pressure sensor, 38: humidity sensor.

Claims (1)

[Claims] 1. In an electronic counting device that irradiates a sample with light and introduces electrons emitted from the sample into an electron detection section to count the number of electrons, the temperature, atmospheric pressure, and humidity of the measurement location. a detection means for detecting a change in the detection value of the detection means; an environment determination means for determining and outputting a change in the detection value of the detection means exceeding a predetermined threshold value; and a display means for displaying a request for initial settings of light intensity, anode voltage, and grid voltage when receiving the output from the environment determining means or the time measuring means. Characteristic electronic counting device.