JPH02223845A - Method and device for measuring particle size of extremely small particle by light scattering method - Google Patents

Abstract

PURPOSE:To measure the particle size of extremely small particles nondestructively without contacting by irradiating the surface of a sample which moves at an equal speed nearby a 1st focus of an elliptic condenser with converged laser light, detecting scattered light which is converged on a 2nd focus, and processing a detection signal. CONSTITUTION:The laser light 2 emitted by a laser 1 is passed through a chopper 3, a spatial filter 4, a collimator lens 5, a polarizing prism 6, a beam splitter (BS) 7, and a lambda/4 plate 8, and reflected by a parabolic mirror 9 to travel backward, and the light is reflected by a BS 7 and converged in a spot. The laser light which is made incident from an incidence opening 13 is converged on the 1st focus 11 of an elliptic surface mirror 10, reflected by the surface of the sample 15 in equal-speed motion, and reflected by the elliptic surface mirror 10, and the light is converged on a 2nd focus 12 and made incident on a detector E through a projection opening 14 and a parabolic light condenser 22. Its output is inputted to a computer 30 through a detector 29, whose pulse train output is integrated and the maximum value of the output voltage is compared with correlation data of the particle size and the maximum voltage to calculate the particle size of the extremely small particles sticking on a wafer 15.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention relates to a method for measuring the particle size of ultrafine particles using a light scattering method for measuring the particle size of nanometer-order ultrafine particles attached to the surface of a sample such as a silicon wafer, and Regarding the device.

[Conventional technology]

Conventionally, the method with the highest resolution for measuring the particle size of fine particles is the light scattering method, but even so, the minimum detectable particle size of a particle size measuring device using the light scattering method for a single fine particle is at most 0. However, as the degree of integration of semiconductors has increased dramatically these days, it has become necessary to detect ultrafine particles on the order of nanometers. In other words, it is known that most defects in electronic circuit patterns are caused by foreign substances on silicon wafers, and if yield and reliability in LSI manufacturing are taken into account, defects as large as 175 to 1/1O of the pattern width are known to occur. The adhesion of foreign matter on silicon wafers causes a serious problem, and as LSI patterns become more highly integrated, the pattern width also becomes submicron order.In the future, efforts will be made to reduce the amount of foreign matter adhesion on silicon wafers during the manufacturing process by cleaning the manufacturing environment. At the same time, in order to remove nanometer-order foreign matter, ie, ultrafine particles, present on the wafer, it has become necessary to detect the particle size of the ultrafine particles.

[Problem to be solved by the invention]

In view of the above-mentioned situation, the present invention aims to solve the problem of particle diameters of 2 to 10 nm attached to the surface of a sample such as a silicon wafer.
It is an object of the present invention to provide a method and apparatus for measuring the particle size of ultrafine particles using a light scattering method, which enables non-destructive and non-contact measurement of the particle size of ultrafine particles on the order of nanometers.

[Failure to solve the problem]

In order to solve the above-mentioned problem, the present invention focuses a laser beam near the first focus of an elliptical condenser, and directs the laser beam to a sample whose surface is positioned near the focus and moves at a constant speed. The extremely weak scattered light from the ultrafine particles attached to the sample surface is focused on the second focal point of the elliptical collector, and the scattered light is directed to the photocathode of the photomultiplier tube using a light guide. Then, the signal is integrated to convert the peak value into a voltage proportional to the particle size of the ultrafine particles, and the particle size of the ultrafine particles is calculated from the maximum voltage value. We have established a method for measuring the particle size of ultrafine particles using a light scattering method.

In order to realize the above method, a laser beam irradiation means that focuses the laser beam to a predetermined spot diameter and irradiates it onto the sample surface, and a laser beam irradiation means that focuses the laser beam to a predetermined spot diameter and irradiates it to the sample surface, and , an elliptical condenser having an entrance/exit opening for irradiating the laser beam to a vicinity including the focal point, and a sample surface located in the vicinity including the first focal point of the elliptical condenser; a moving device capable of moving at a constant speed; and a light guide having one end disposed near the second focal point of the elliptical condenser and guiding scattered light from the ultrafine particles attached to the sample surface focused at the focal point. , a photomultiplier tube is disposed at the other end of the light guide and detects the extremely weak scattered light guided by the light guide as a discrete pulse-like signal of a single photoelectron state; A light scattering method consisting of a coolable detector and a signal processing means that integrates the signal of scattered light detected by the detector, converts it into a voltage, and calculates the particle size of the ultrafine particles from the peak value voltage. A particle size measuring device for ultrafine particles was constructed.

The light guide may be a parabolic condenser having a parabolic mirror on its inner surface whose focus coincides with the second focus of the elliptical condenser, or one end of which is positioned on a hemispherical surface surrounding the second focus. Multiple optical fibers were used.

[Effect]

The method and apparatus for measuring the particle size of ultrafine particles using a light scattering method according to the present invention, which has the above-mentioned content, focuses light on the surface of a sample that is located near the first focus of an elliptical condenser and moves at a constant speed. When a laser beam is irradiated, as the ultrafine particles attached to the sample surface pass through the laser beam spot, scattered light is generated near the first focal point, with an intensity that corresponds to the particle size and an amount of light that corresponds to the movement speed. As a result, the extremely weak scattered light is focused on the second focal point of the elliptical collector, and then guided by the light guide to the photocathode of the photomultiplier tube, where a number of photoelectrons proportional to the intensity of the scattered light are emitted. generated and detected as a discrete pulse signal of a single photoelectron state, and the maximum voltage value of the voltage signal obtained by integrating the discrete signal is set to be proportional to the particle size of the ultrafine particle,
In this way, the particle size of the ultrafine particles attached to the sample surface can be measured from the maximum voltage value.

In addition, the first focal point on the elliptical mirror side is set so that the laser beam that is focused on the first focal point of the elliptical condenser and is irradiated onto the sample surface is not focused on the second point so that the strong reflected light from the surface is not focused on the second point. A laser beam exit opening is set up at an equiangular position centered on .

Furthermore, since the emission angle of the scattered light coming out from the second focal point of the elliptical condenser spans a range of about 180 degrees depending on the intensity distribution of the scattered light by the ultrafine particles, a parabolic concentrator is used as a light guide. If the optical fiber is used as a light guide, by aligning the focal point of the parabolic mirror with the second focal point, or by positioning one end of the plurality of optical fibers on a hemispherical surface surrounding the second focal point. , can be guided extremely efficiently to the photocathode of a photomultiplier tube.

〔Example〕

Next, the present invention will be further described in detail based on embodiments shown in the accompanying drawings.

Figure 1 (al, (bl) and Figure 2 show typical embodiments of the present invention, in which A is a laser beam irradiation means, B is an elliptical condenser, C is a moving device, D is a light guide, and E is a indicates a detector, and F indicates a signal processing means, respectively.

The laser beam irradiation means A has a total output (Io) of IW and a wavelength (
A laser beam 2 emitted from an argon (Ar'') laser 1 with a wavelength of 488 nm is passed through a chopper 3, and a spatial filter 4 removes distortion of the nonlinear wavefront, and a collimator lens 5 makes the spatial intensity distribution uniform. The laser beam 2 is converted into a large-diameter parallel beam, linearly polarized by a polarizing prism 6, passed through a polarizing beam splinter 7, and reflected by a parabolic mirror 9 through a λ/4 wavelength plate 8. It was constructed with an optical system that passes through the λ/4 wavelength plate 8, reflects the light in the right angle direction by the polarizing beam fritter 7, and focuses the light so that the spot diameter d becomes approximately 5 μm.
In this embodiment, the laser beam 2 is an argon laser with 48
Although a wavelength of 8 nm was used, it is possible to appropriately use a laser that can emit a long wavelength that has a high reflectance for the mirror-polished metal used in the condensing system for scattered light, which will be described later. Care must be taken because the efficiency of conversion into photoelectrons at the photocathode of the photomultiplier tube decreases, resulting in a decrease in output.

The elliptical condenser 5B is made by cutting a metal lump into a substantially hemispherical shape centered on the major axis of a spheroid, and mirror-finishing it to form an ellipsoidal mirror 10 on the inner surface. The focal point 11 and the second focal point 12 are set to be located slightly outward from the plane connecting the surrounding ridge lines, and the laser beam is placed equiangularly around the first focal point 11 on the ellipsoidal mirror 10 side. The laser light 2 incident from the entrance and exit ports 13 and 14 is reflected from the surface of a sample 15 such as a silicon wafer placed near the first focal point 11, and is reflected by the ellipsoidal mirror 10.
All the reflected light from the sample 15 is made to exit from the exit port I4 so as not to be reflected and focused on the second focal point 12.

When mounting a plate-shaped sample 15 having parallel surfaces, the moving device C can use an X-Y table 16 that moves in a plane, fixes the sample 15 on the upper surface, and moves the sample 15. The surface is always positioned near the first focal point 11 of the elliptical condenser B and is moved at a constant speed in the X direction and the Y direction, and the driving stepping motor 1717 connected to each table is a motor drive. It is rotated by device 18.

In this way, each part of the surface of the sample 15 is moved in the vicinity of the first focal point 11, and as a result, the surface of the sample 15 is scanned with the spot of the laser beam 2. In addition, the sample 15
Although it is not impossible to fix the optical system and move the optical system including the laser beam irradiation means A, it is not practical. In the case of the sample 15 having a curved surface, it should be possible to move it in the Z direction as well, or if the curvature is constant, the upper table should be made to roll relative to the lower table so as to match the curvature. It is necessary to do something.

In this embodiment, the light guide D has a parabolic mirror 19 made of a mirror-polished metal surface on its inner surface, and an opening 21 near its focal point 20.
A parabolic condenser 22 is used, the focal point 20 of which is aligned with the second focal point 12 of the elliptical condenser B, and the other open end is connected to a detector E, which will be described later. The reason why this parabolic condenser 22 is used in this embodiment is that the emission angle of the scattered light 2' emerging from the second focal point 12 of the elliptical condenser B is determined by the scattered light intensity distribution due to the ultrafine particles P. This is to efficiently condense such wide-angle scattered light 2', which spans a range of approximately 180 degrees. Although not shown, it is also possible to arrange one end of a plurality of optical fibers on a hemispherical surface surrounding the second focal point 12 and bundle the other ends to guide the optical fibers to the detector E.

The detector E is placed in a cooling container 24 having a double structure so that a refrigerant can be sealed around the outer periphery while leaving a part of the window 23.
A photomultiplier tube 2 installed with its light-receiving surface facing the window portion 23
5 is installed inside, and the window portion 23 is provided with a double window cell 26 to prevent dew condensation from forming on the light receiving surface. A predetermined voltage VC is applied to the photomultiplier tube 25 by a high voltage type [27] so that the extremely weak scattered light 2' can be detected as a discrete pulse-like signal of a single photoelectron state.

As shown in FIG. 1 (al) and FIG. Integrate the discrete pulse-like signal S in the photoelectronic state (shown in Figure 3 (C1)) and convert the maximum voltage value Vm into a voltage Vp (shown in Figure 3 (dl)) proportional to the intensity of the scattered light 2'. A detection circuit 29 and a computer 30 that reads the peak value Vm of the voltage vp signal and converts it into the particle size Dp of the ultrafine particles P.
Therefore, the detection circuit 29 is constituted by a CR integration circuit (integration constant τ=CR) that integrates the pulse train output of the photomultiplier tube 25 in an analog manner, and the combinator 3
0 is the scattered light 2' corresponding to the particle size Dρ of the spherical ultrafine particles P.
is guided to the photomultiplier tube 25 by the condensing system, and the maximum voltage value Vm of the output voltage ■p obtained by the detection circuit 29 is theoretically predicted in advance, taking into consideration the reflection and conversion efficiency of each part. Correlation data between the particle ('V D p and the maximum voltage value ■m) is stored, and the maximum value of the output voltage vp from the actual detection circuit 29 is read and compared with the above data to determine the particle size of the ultrafine particle P. The particle diameter Dp is calculated by comparison.Also, when the output of the photomultiplier tube 25 is small, C
It is also possible to insert a high-speed pulse amplifier before the R integration circuit, and it is also possible to binarize the output of the photomultiplier tube 25 and digitally integrate it. The computer 30 stores information such as the output P of the laser 1, the spot (winter d), the scanning speed vs. It is practical to manually create the correlation data between the particle diameter Dp and the maximum voltage value Vm using these input values.

In this embodiment, a projection circuit 31 is provided which displays input parameters and measurement results in conjunction with the computer 30.
has been established.

Next, to describe the details of the particle size measurement principle of the present invention, as shown in FIG. Assuming a Gaussian distribution (shown in FIG. 3 (bl)), the intensity in the direction of the radius r is given by a point where the intensity is l/e2.

Here, Io is the total output (IW) of the laser beam 2. The spot diameter d of this laser beam 2 on the surface of the sample 15 is set to 5 μm in this example, which is very large compared to the particle diameter Dp (1 to lonm) of the ultrafine particles P to be measured. , the scattered light 2' from the cough ultrafine particles P is generated in accordance with the spot light intensity 1 (r) during the time that the spot diameter d passes through the particles. The total scattered light intensity Is of the scattered light 2' is expressed as being valid when the particle size Dp is sufficiently small compared to the wavelength (488 nm) of the laser light 2. Here, α-Dpπ/λ is a dimensionless parameter, and when α is sufficiently smaller than 1, formula (2) holds true, and the crystal is a complex refractive index for ultrafine particles P, which varies depending on the type of ultrafine particles P. They are different.

Of the total scattering intensity Is, the photocathode 2 of the photomultiplier tube 25
The scattered light intensity It that can be received by 8 is Ii-βIs
It is expressed as (3). Here, β is the light collection efficiency that takes into account the loss of the light collection system, and the input and output loss at each focal point of the elliptical collector B is 9%, and the ellipsoid! JIIO (aluminum)
The reflectance of the parabolic mirror 19 of the parabolic condenser 22 is 90%.
If the reflectance of (aluminum) is 90%, approximately β=0
．． 67, but the loss due to the entrance/exit port 13.14 of the elliptical condenser B, the fact that the entrance/exit port 13.14 is tilted, and the scattering within an angle of 2π steradian out of the total scattered light intensity +s Taking into consideration that light 2' is focused,
In this embodiment, β=0.575. And the seventh
The figure shows the relationship between particle diameter Dp (nm) and received scattered light intensity It (W) for silver (Ag), gold (Au), and silicon carbide (SiC).
, aluminum oxide (^j! 203) and silicon dioxide (Sin2) are illustrated.

In addition, as shown in FIG.
If h is the blank constant and C is the speed of light, then Nρh= It λ/ h c (
4), and the number of photoelectrons Npk, which is the number of photons Nρh multiplied by the quantum efficiency η at the photocathode 28, is transferred to the photocathode 28.
The electrons of the number Nk of photoelectrons, which is the number Npk of photoelectrons multiplied by the dynode focusing efficiency δ (δ = 0.7), are amplified by the amplification factor μ and become a discrete pulse-like signal S of a single photoelectron/electronic state. , is output from the photomultiplier tube 25. That is, the number of photoelectrons Nk is Nk(r) = I t(λ
/hc) ηδ−βIs(λ/hc) ηδ (pcs/5
ee) (5) However, since the spot light intensity distribution is a Gaussian distribution as shown in FIG. 3 (C1), this discrete pulse-like signal S has a densely generated distribution within the spot diameter d.

Furthermore, the number of single photoelectron pulses Nks per spot diameter d generated when the ultrafine particles P are detected by a scanning laser spot is Nb5 = Nk - d/vg (number of particles), where the spot scanning speed V S (m/5ee) is ) (6), and the relationship between the particle diameter Dp and the number of single photoelectron pulses NkS is shown in FIG. 8 using the scanning speed V9 as a parameter. In the figure, a region where the number of single photoelectron pulses NkB is 10 or less is indicated by a broken line as an unstable detection region. The present invention integrates the number of single photoelectron pulses Nks emitted during the time when the scanning laser spot is irradiating the ultrafine particles P through the detection circuit 29, and calculates the output voltage vp as shown in FIG. 3 (dl). The measurement principle is to detect the peak value and find out the particle size Dp from the maximum voltage value Vs.

Further, FIG. 5 shows an equivalent circuit including the photomultiplier tube 25 and the detection circuit 29. When the amplification factor of the photomultiplier tube 25 is μ, the photoelectrons generated at the photocathode 28 are the total charge Q.−eμ (e is the charge element) and the width is t,
(usually 2 to Ions) is output as a pulse. When the capacitor capacity including the load resistance Rt and stray capacitance connected to the output part of the photomultiplier tube 25 is C, the CR circuit output voltage is ■. is expressed as follows, that is, the charging voltage VOC at Q<t≦t has a time constant of τ (for example, R=
50Ω, C=τ. These relationships are shown in FIG. 6, and the width t1
The charging voltage due to the pulse of ■0, and [. When the discharge voltage after seconds is 2, 1w and t. By appropriately selecting the time constant τ, that is, the value of CR, it continues to rise without attenuating, and the integrated output voltage vp of the pulse train shown in Fig. 3 (dl) can be obtained. The maximum voltage value Vm is proportional to the number of single photoelectron pulses Nks, that is, proportional to the intensity It of scattered light received from the ultrafine particles P, and thus the particle size Dp of the ultrafine particles P can be measured. .

Finally, the measurement sensitivity S/N ratio of the present invention will be described. m
The noise that greatly affects the S/N ratio is the dark current of the photomultiplier tube 25, and the number of single photoelectron pulses due to the dark current is
Then, the S/N ratio is defined by the following equation.

S / N - N k / N d (10) Therefore, at room temperature (20°C), the dark current causes about 30
00 single photoelectron pulses are obtained and the photomultiplier tube 2
By cooling 5 (-30°C), the single photoelectron pulse per unit time due to dark current can be suppressed to about 75 flll, and from the relationship between S/N ratio and particle size Dp shown in Figure 9, S If the /N ratio is 10 or more as the measurement limit, -
By cooling to about 30℃, the particle size Dp becomes 2.5μ.
It is possible to detect the particle size of ultrafine particles P of m or more, and the particle size D
To detect ultrafine particles P with p of 10 μm, a photomultiplier tube 2 is used.
Since the S/N ratio can be maintained at around 103 without cooling 5,
Detection is possible if noise such as electromagnetic induction is taken into account.

Furthermore, as mentioned above, the intensity of the scattered light 2' is extremely weak, so the photomultiplier tube 25 is used at its sensitivity limit.
By increasing the output of the laser 1 as much as possible, the detection sensitivity can be slightly increased, but there is a risk that the sample 15 and the ultrafine particles P will be thermally destroyed by the irradiation of the laser beam 2. Care must be taken in selection. For example, if the IW laser l is focused on a 5 μm spot and a silicon wafer is selected as the sample 15, the surface temperature of the silicon wafer in a steady state is approximately 45 μm according to thermal conduction theory.
5°C, and it was found that the surface does not melt. Furthermore, when gold fine particles with a particle size Dp of 10° nm are selected as the ultrafine particles P, the surface temperature in a steady state is approximately 550°C, and it is found that the surface does not melt. It turns out there isn't.

In addition, in the particle size measuring method of the present invention, the type of ultrafine particles P cannot be determined, and the complex refractive index 5 differs depending on the type. Therefore, as shown in FIG. Therefore, calculation of the particle size Dp includes an inherent error. The ultrafine particles P that are likely to adhere to the surface of the silicon wafer sample 15 are
As shown in FIG. 7, assuming that only the ultrafine particles P exist, for example, if the scattered light intensity is 10-1'W, the particle size Dp will be within 3.5 to 5.5 nm. When measuring ultrafine particles P of approximately 5 nm, the error is ±1.5 nm, and when the scattered light intensity is 10-''W, the particle size Dp is within the range of 5 to 10 nm. It is predicted that the error when measuring ultrafine particles P of 7 to 3 nm is ±2.5 nm.In this way, the error caused by the type of ultrafine particles P on the measured value of particle diameter Dp is relatively small. As can be seen from equation (2) above, the contribution of grains + Dp is the sixth power, and the contribution to the complex refractive index is about the second power,
The contribution of the particle size Dp is overwhelmingly greater, and therefore, as mentioned above, the type of ultrafine particles P has little influence on the measured value.

In addition, in the present invention, the shape of the ultrafine particles P is assumed to be spherical, but the actual shape varies and spherical shapes are rather rare. Since we have made it possible to receive light from all objects, even if the scattered light 2' is biased due to distorted ultrafine particles P,
Since the total scattered light intensity depends almost only on the outer shape of the ultrafine particles P as seen from the direction of incidence of the laser beam 2, the essence of the present invention is not lost.

Note that if the type of ultrafine particles P is detected in advance using another method, it is possible to measure the particle size with even higher accuracy.

〔Effect of the invention〕

According to the method and apparatus for measuring the particle size of ultrafine particles using the light scattering method of the present invention, the particles attached to the surface of the sample have a diameter of 2 to 200 nm by simply scanning the sample surface with a laser spot.
The particle size of ultrafine particles of about 1 Onnm can be measured non-destructively and non-contact with high accuracy and in real time.

The optical system also includes a laser beam irradiation means that focuses the laser beam to a predetermined spot diameter and irradiates it onto the sample surface;
At an equiangular position centered on the first focal point on one side of the ellipsoidal mirror,
an elliptical condenser having an exit opening for irradiating the laser beam to a vicinity including the focal point; and an elliptical condenser having one end disposed near a second focal point so that the laser beam is condensed at the focal point. Since it is composed of a light guide that guides the scattered light from the ultrafine particles attached to the sample surface, it is possible to extremely efficiently collect the scattered light within 2π steradians out of the total scattered light from the B particles attached to the sample surface of the laser beam. To guide the strong reflected light that is directly reflected from the sample surface to the outside of the elliptical condenser through an exit opening provided in the elliptical condenser so as not to affect the extremely weak scattered light. was completed.

Furthermore, regarding the extremely weak scattered light detection system, a photoelectron intensifier is disposed at the other end of the light guide and detects the extremely weak scattered light guided by the light guide as a discrete pulse-like signal in a single photoelectron state. A detector equipped with a multiplier and capable of cooling the photomultiplier; the signal of the scattered light detected by the detector is integrated and converted into a voltage, and the particle size of the ultrafine particles is determined from the peak value voltage. By using a signal processing means that calculates It can be detected as a discrete pulse-like signal in the photoelectronic state, and by integrating the discrete pulse-like signal and converting its maximum value into a voltage signal proportional to the particle size, the particle size can be determined from the voltage signal by simple processing. It became possible to calculate.

and a parabolic condenser having a parabolic mirror on its inner surface whose focal point coincides with the second focus of the elliptical condenser as a light guide;
Alternatively, by using a plurality of optical fibers having one end positioned on a hemispherical surface surrounding the second focal point, scattered light emitted from the second focal point of the elliptical condenser at an exit angle spanning 180 degrees can be efficiently collected. It emits light and can guide it to the photocathode of a photomultiplier tube.

[Brief explanation of the drawing]

FIG. 1 (al is a simplified layout diagram showing a typical embodiment of the present invention, FIG. Figure ta+ is a simplified cross-sectional view showing the state in which ultrafine particles on the sample surface are irradiated with laser light, Figure 3 (bl is a graph showing the intensity distribution of the laser beam, and Figure 3 (C1 is the output section of the photomultiplier tube). Figure 3 is a graph showing the discrete pulse-like signal of a single photoelectron state in (dl is a graph of the voltage waveform corresponding to the particle size at the output of the detection circuit, Figure 4 is the circuit of the photomultiplier tube and the detection circuit. 5 is an equivalent circuit diagram, FIG. 6 is an enlarged view of the voltage waveform of dl in FIG. Figure 9 is a graph of the S/N ratio versus particle size. A: Laser beam irradiation means, B: Elliptical condenser, C: Moving device, D: Light guide, E : Detector, F: Signal processing means. 1: Laser, 2: Laser light, 3: Chopper, 4: Spatial filter, 5: Collimator lens, 6: Polarizing prism, 7: Polarizing beam splitter, 8: λ/4 Wave plate, 9: parabolic mirror, 10: ellipsoidal mirror, 11: first focal point, 1
2: second focal point, 13: entrance port, 14: exit port, 15: sample, o: x-y table, 17: stamping motor, 18: motor drive device, 19: parabolic mirror, 20: focal point , 21: opening, 22: parabolic concentrator, 23: window, 2
4: cooling container, 25: photomultiplier tube, 26: double window cell, 27: high voltage power supply, 28: photocathode, 29: detection circuit,
30: Computer, 31: Projection circuit.

Claims (1)

[Scope of Claims] 1) Focusing a laser beam near the first focal point of an elliptical condenser, and irradiating the laser beam onto a sample whose surface is positioned near the focal point and is moving at a constant speed; The extremely weak scattered light from the ultrafine particles attached to the sample surface is focused on the second focal point of the elliptical condenser, and the scattered light is guided to the photocathode of the photomultiplier tube by a light guide to generate a single photoelectron. A light scattering method that detects the state as a discrete pulse-like signal, integrates the signal, converts the peak value into a voltage proportional to the particle size of the ultrafine particles, and calculates the particle size of the ultrafine particles from the maximum voltage value. A method for measuring the particle size of ultrafine particles. 2) A laser beam irradiation means that focuses the laser beam to a predetermined spot diameter and irradiates it onto the sample surface;
an elliptical condenser having an entrance and exit opening for irradiating the laser beam to a vicinity including the focal point, and a sample surface located in the vicinity including the first focus of the elliptical condenser, a moving device capable of moving at high speed; a light guide having one end disposed near the second focal point of the elliptical condenser and guiding scattered light from the ultrafine particles adhering to the sample surface focused at the focal point; A photomultiplier tube is disposed at the other end of the light guide and detects extremely weak scattered light guided by the light guide as a discrete pulse-like signal of a single photoelectron state, and the photomultiplier tube is cooled. A light scattering method comprising: a detector capable of detecting scattered light; a signal processing means for integrating the signal of the scattered light detected by the detector, converting it into a voltage, and calculating the particle size of the ultrafine particles from the peak value voltage; Ultrafine particle size measuring device. 3) The light guide according to claim 2, wherein the light guide is a parabolic condenser formed by forming a parabolic mirror on the inner surface whose focus is aligned with the second focus of the elliptical condenser. Ultrafine particle size measuring device using light scattering method. 4) Ultrafine particles produced by the light scattering method according to claim 2, which uses a plurality of optical fibers whose end faces are located on a hemispherical surface surrounding the second focal point of the elliptical condenser as the light guide. particle size measuring device.