JPH07117484B2 - Correction method of particle size distribution in fine particle measurement - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention measures the number of fine particles and their particle size distribution by irradiating fine particles contained in a fluid with a laser beam and detecting scattered light from the fine particles. And a method for correcting the particle size distribution in such an apparatus.

[Conventional technology]

In the manufacturing process of semiconductors and pharmaceuticals, in order to inspect the cleanliness of ambient air, ultrapure water, the quality of chemicals, etc., and in the fields of medicine, biology, etc., to inspect the state of cells,
A fine particle detector for detecting fine particles such as dust and cells is used, and such a fine particle detector generally employs a light scattering method.

FIG. 9 is a block diagram showing a method of receiving 90 ° side scattered light among the laser light scattering methods. In the figure, reference numeral 1 denotes a light source unit, and various lasers are used. The laser beam 2 emitted from this light source 1 is usually focused by a converging lens 3 and irradiated onto a measurement fluid 4. A transparent flow cell 5 holds the measurement fluid 4. Reference numeral 6 denotes a beam block that blocks the transmitted beam after being irradiated with the measurement fluid. The fluid is assumed to flow in the Z-axis direction perpendicular to the paper surface, that is, assuming the paper surface to be the X and Y planes.

When the measurement fluid 4 contains fine particles, the fine particles pass through the laser beam 2, and at that time, scattered light 7 corresponding to the particle diameter is generated.

The scattered light 7 is condensed by the light receiving lens 8, passes through the diaphragm 9, and then is guided to the photodetector 10, where it is converted into an electric signal pulse. This signal pulse is amplified by the amplifier 11 and enters the measuring circuit 12. The measuring circuit 12 calculates the number of particles from the number of pulses and the particle size from the peak value of the pulse.
When the scattered light contains a fluorescent component, an optical filter or a spectroscope is inserted in front of the photodetector 10 to detect only the fluorescent component, so that the properties of the particles can be examined.

[Problems to be Solved by the Invention]

Especially important as the performance required for such a fine particle detector is (a) high detection sensitivity of the particle size, and (b) accurate particle size analysis. It is very difficult to use the device that adopts the above, and there are the following two main reasons.

When a laser beam is used as the irradiation light, the beam has a power distribution (mostly a Gaussian distribution) in its cross section, so the irradiation light intensity changes depending on the passage position of the particles in the beam, and the scattered light intensity also changes. It is difficult to convert the scattered light intensity into a particle size because it depends on the passage position of the particle.

In order to remove stray light or to limit the light receiving area to the beam waist, insert a diaphragm 9 such as a pinhole or slit in front of the detector 10, and use A and B in Fig. 10 (a) and (b). Although the visual field shown is formed, the light receiving efficiency in the visual field is not constant. That is, if particles pass through the visual field area A in FIG. 10, all the scattered light collected by the condenser lens enters the detector, and the light receiving efficiency becomes constant at 100%. If the particles pass through, only a part of the scattered light will enter the detector and the light receiving efficiency will be less than 100%.

FIG. 11 shows this from another point of view.

This figure shows an image (becomes circular) on the pinhole 9 of scattered light emitted from particles at arbitrary positions x and y in the light-receiving system visual field region 13 as shown in FIG. In the case of FIG. 11 (a) in which the image 14 is entirely contained in the pinhole 9 (radius D), the light receiving efficiency is 100%, but in the case of simultaneous (c) where it does not enter the pinhole 9 at all. It becomes 0%. Also, in the case of the same figure (b), 0 depending on the value of x, y
And a value between 100%. That is, as shown in FIG. 10 (b), the particles passing through the visual field region B are measured to be smaller than they actually are, which causes an error in the particle size distribution measurement. It should be noted that Y in the visual field area 13 of the light receiving system shown in FIG.
There is a relation between 1/2 (= R) of the maximum width in the direction and the pinhole radius D, where R is D / M, where M is the magnification of the condenser lens.

Of the above two reasons, as for the problem due to the power distribution of the laser beam, as a countermeasure, for example,
As shown in the figure, the light receiving area is limited to the central part of the beam, or the laser beam is scanned as shown in FIG. 14 to apparently make the beam power distribution in the light receiving area uniform. The method described above has been adopted (Japanese Patent Laid-Open No. Sho 61-61).
-288139 gazette).

On the other hand, with respect to the problem due to the nonuniformity of the light receiving efficiency, there is a method of flowing a clean sheath flow around the sample flow and flowing the sample only in the central part of the visual field region. However, this method has a problem that the mechanism for forming the sheath flow (the shape of the flow cell, etc.) is complicated and the cost becomes high. Further, when the laser beam is narrowed down and the light receiving region is made small to several tens to several hundreds of μm, it becomes difficult to set the sample flow in the central part, which is a problem in practicality.

Therefore, an object of the present invention is to eliminate the particle size distribution measurement error caused by the unevenness of the light receiving efficiency in the light receiving region among the above-mentioned problems in a scattered light type fine particle detecting device by a simple and inexpensive method. is there.

[Means for Solving the Problems]

In order to solve the above problem, in the present invention, a measurement fluid containing fine particles is irradiated with a light beam, and scattered light emitted from the fine particles by the light beam is received by limiting the field area of the light receiving system. When performing the measurement with a device for measuring the number of fine particles and the particle size distribution from the scattered light, the light receiving efficiency distribution in the light receiving system visual field region caused by the limitation of the light receiving system visual field region is calculated, and this light receiving efficiency distribution is calculated. Is corrected according to the power distribution of the light beam, and based on the corrected light receiving efficiency distribution, for each particle size category specified by the device, when the particle group corresponding to the representative particle size is measured, it is output from the device. Calculate the particle size distribution of each expected
The difference between each predicted particle size distribution output and the true particle size distribution is obtained as each error distribution, and the particle size distribution of the measurement result is corrected to the true particle size distribution based on each error distribution.

[Action]

After obtaining the light receiving efficiency distribution of the particles by simulation (simulation), the particle size distribution is obtained by correcting the light receiving efficiency distribution according to the power distribution of the light beam, and from the true particle size distribution of the particle size distribution. The deviation ratio of is calculated as a correction coefficient, and the particle size distribution of the measurement result is corrected based on this correction coefficient, so that the particle size distribution error can be easily and inexpensively removed.

〔Example〕

FIG. 1 is a schematic flowchart showing an embodiment of the present invention.

That is, first, the particle size distribution of fine particles is predicted by calculation simulation (see), then the correction coefficient is obtained (see), and the measurement result is corrected based on this correction coefficient (
reference). If step and are executed in advance to obtain a correction coefficient and the correction coefficient is stored in a memory or the like, the measurement result obtained by the particle detection device is simply corrected based on the stored correction coefficient. Can be processed only.

Hereinafter, the processing contents of the steps will be described in more detail.

In the device shown in FIG. 9 above, it can be seen that the scattered light intensity changes depending on the position (x, y) in the visual field region of the light receiving particle, that is, the light receiving efficiency η and the laser beam distribution. . Therefore, conversely, if the scattered light intensity is known, the particle size of the particles can be known. That is, assuming that the particle diameter is D ₀ and the scattered light intensity when the light receiving efficiency is 100% is W ₀ = f (D ₀ ), the particle having the particle diameter D _{0 at} the position (x, y) in the light receiving system visual field region The scattered light intensity W (x, y, D ₀ ) when is transmitted is represented by W = (x, y, D ₀ ) = η (x, y) · W ₀ (D ₀ ), and D ₀ = Since it is f ⁻¹ (W ₀ ), the scattered light intensity W can be converted into the particle size D from the relation of D = f ⁻¹ (W). However, f ⁻¹ represents the inverse function of f.

Therefore, first, the light receiving efficiency η of the light receiving system visual field region is obtained as follows.

For this purpose, consider a relational diagram as shown in FIG. 2 in which the pinhole image and the scattered light image are converted to the light receiving system visual field region (measurement region) side via the condenser lens. The figure shows the relationship between the pinhole image and the scattered light image in Y and Z coordinates.

From this relationship diagram, the light receiving efficiency η is the area ratio of the scattered light image entering the pinhole, that is, η = (area of the scattered light image included in the pinhole) ÷
(Area of scattered light image) It can be seen that it can be obtained by the following formula. Here, considering only the case of x> 0, y> 0, i) y + r ₁ ≦ R η = 1 ii) y−r ₁ ≧ R η = 0 iii) yr ₁ <R When <y + r ₁ , 0 <η <1. Therefore, η in the case of iii) is obtained in detail as follows.

Figure 3 is for explaining this. Since η is represented by the ratio of the areas S ₁ and S _{5 in} the figure, S ₁ = πr ₁ ² S ₂ = πR ² θ ₁ / π = Since R ² θ ₁ S ₃ = πr ₁ ² θ ₂ / π = r ₁ ² θ ₂ S ₄ = yRsinθ ₁ and S ₅ = S ₂ + S ₃ −S ₄ , η = S ₅ / S ₁ = ( R ² θ ₁ + α ² x ² θ ₂ −yRsin θ ₁ ) ÷ (πα ² x ² ). However, r ₁ = αx, α = tanω (ω; converging angle), and θ ₁ and θ ₂ are θ ₁ = cos ⁻¹ {(R ² + y ² −r ₁ ² ) / 2Ry from the cosine law, respectively. } Θ ₂ = cos ⁻¹ {(r ₁₂ + y ² −R ² ) / 2r ₁ y}.

The above is the case where the power distribution of the laser beam is not taken into consideration. Considering this with the scanning beam system as shown in FIG. 14, it becomes as follows.

Now, let's change the convergent beam radius at beam waist 2A in FIG.
If r _a , then r _a = 4λf / 3πr ₀ , and the beam radius r _b at a distance y from the beam waist 2A is r _b = r _a {1+ (λy / πr _a ² ) ² } ^{1 ]} since the ^2, power density ratio of the laser becomes _{P b / P a = (r} a / r b) 2 = 1 / {1+ (9πyr 0 2 / 16λf 2) 2}. Therefore, the light receiving efficiency η'in consideration of the power distribution of the laser beam is expressed by η '= η / {1+ (9πy ₀ ² / 16λf ² ) ² }. Here, r ₀ is the laser beam diameter, λ is the laser wavelength, and f is the focal length of the converging lens.

From the above, the scattered light intensity W when a particle having a particle diameter D ₀ passes through an arbitrary position (x, y) in the visual field region of the light receiving system is W (x, y, D ₀ ) = η ′ (x, y) · W ₀ (D ₀ ).

Therefore, assuming that the particle diameter D ₀ and the scattered light intensity W ₀ when the light receiving efficiency is 100% are known, the range 0 <x <L 0 <y <R + Ltanω in the coordinate region shown in FIG. Both are divided into 100 equal parts, for example, and the scattered light intensity W is calculated for 10,000 pairs of (x, y), and from this, the particle size D
Ask for.

The 10,000 sets of particle size D (x, y) thus obtained are counted as A _{1 to} A
FIG. 4A shows the relationship with _n . Then, if such an operation is performed for the representative values D _{01 to} D _0n of each particle size channel as shown in FIG. 8B, n sets of distributions will be obtained as shown in FIG. And (d) in the same figure.

As described above, it is considered that the reason why the result as if the particles are large dispersed particles is obtained in the case of the monodisperse particles is mainly due to the device error. Therefore, the present invention makes the following corrections.

First, the difference (error distribution) between the calculated particle size distribution and the true particle size distribution is calculated. This relationship is illustrated in FIG. 5, and the error distribution B is shown in correspondence with A _ij as follows.

B ₁₁ B ₂₁ , B ₂₂ B ₃₁ , B ₃₂ , B ₃₃ ... _Bn1 , B _n2 , B _n3 ……… B _nn where B _ij = A _ij (i ≠ j) B _ii ＝ A _ii − T _i (i = j). The true distribution is considered to be the number of particles passing through the shaded area in FIG. 12, and is represented here as a known constant value.

Next, the error distribution B _ij is divided by the number of particles T _i in the true distribution to obtain a correction coefficient P _ij as described below.

P ₁₁ P ₂₁ ,, P ₂₂ P ₃₁ ,, P ₃₂ , P ₃₃ ... P _n1 , P _n2 , P _n3 ……… P _nn where P _ij = B _ij / T _i (i ≠ j) P _ii ＝ B _ii / T _i + B _ii ) (i = j).

Therefore, the error distribution is subtracted from the particle size distribution (N _{1 to} N _n ) measured by the device as shown in FIG. 9 to obtain the true distribution (N ₁ ′ to
The calculation formula (correction calculation formula) for obtaining N _n ′) is as follows.

N _n ′ ＝ N _n −N _n P _nn N _n-1 ′ ＝ N _n-1 −N _n ′ P _n , _n-1 −N _n-1 × P _n-1 , _n-1 N _n-2 ′ ＝ N _n-2 −N _n ′ P _n , _n-2 −N _n-1 ′ × P _n-1 , _n-2 −N _n-2 P _n-2 , _n-2 N ₁ ′ ＝ N ₁ − N _n ′ P _n1 −N _n-1 ′ P _n-1 , ₁ −N _n-2 ′ P _n-2 , ₁ …… N
₁ P ₁₁ Figure 6 shows a concrete example.

The figure (b) shows the measured distribution.
When correction is performed by taking into account the correction coefficient that represents the ratio of the error distribution to the true distribution shown in ( ₃ ), the true distribution N ₃ ′ as shown in FIG.
~ N ₆ ′ is required.

In the above description, the correction is performed by sequentially performing calculation from the larger particle size to the smaller particle size. However, the matrix [A] of the simulation result Ai of the particle size distribution is considered, and the inverse matrix is calculated and corrected. Is also possible.

Further, the measurement results of polystyrene latex (PSL) particles, which are said to have a uniform particle size, are compared with those in the case where they are not corrected as in the present invention, and those in FIGS.
Shown in the figure.

FIG. 7 shows an example in which the particle size is 0.410 [μm].
The figure is an example when the particle size is 0.208 and 0.506 [μm]. In any case, it can be seen that the particle size distribution is close to the true distribution, and the device error is removed.

〔The invention's effect〕

According to the present invention, the measurement error of the particle size distribution due to the non-uniformity of the light receiving efficiency in the field area of the light receiving system is corrected based on the calculation simulation result without depending on the experiment. Without the need to remove it easily and inexpensively.

[Brief description of drawings]

FIG. 1 is a schematic flow chart showing an embodiment of the present invention, FIG. 2 is an explanatory view for explaining the relationship between a scattered light image and a pinhole, and FIG. 3 is an explanation for explaining a calculation method of light receiving efficiency. 4 and 5 are explanatory diagrams for explaining a simulation method of particle size distribution, and FIG. 5 is an explanatory diagram for explaining an error distribution of a particle size distribution obtained by simulation calculation with respect to a true distribution. FIG. 6 is an explanatory diagram for specifically explaining the correction calculation method according to the present invention, FIGS. 7 and 8.
Both figures show the case where the correction method according to the present invention is applied,
FIG. 9 is an explanatory view for explaining the result in comparison with the case where it is not, FIG. 9 is a configuration diagram showing a general example of a particle measuring device, and FIG. 10 is a view for explaining a relationship between a light receiving system visual field region and light receiving efficiency. FIG. 11 is an explanatory view for explaining the light receiving efficiency of the scattered light image, FIG. 12 is an explanatory view for explaining the light receiving system visual field region, FIG. 13 and FIG. 14 are both It is explanatory drawing for demonstrating the relationship between a laser beam and its power distribution. Reference numeral 1 ... Laser light source, 2 ... Laser beam, 2A ... Beam waist, 3 ... Converging lens, 4 ... Measuring fluid, 5
...... Flow cell, 6 ... Beam block, 7 ... Scattered light, 8 ... Condenser lens, 9 ... Aperture (pinhole), 10
...... Photo detector, 11 ...... Amplifier, 12 …… Measurement circuit, 13 ……
Light-receiving system field of view, 14 ... Scattered light image.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Takeo Tanaka 1-1 Tanabe Nitta, Kawasaki-ku, Kawasaki-shi, Kanagawa Fuji Electric Co., Ltd. (72) Yuji Oki 1 Nitta Tanabe, Kawasaki-ku, Kawasaki-shi, Kanagawa No. 1 within Fuji Electric Co., Ltd. (56) Reference JP-A-61-29738 (JP, A)

Claims (1)

[Claims] 1. A measurement fluid containing fine particles is irradiated with a light beam, and the scattered light emitted from the fine particles by the light beam is received by limiting the visual field of the light-receiving system, and the scattered light is received from the received light. When performing the measurement with a device for measuring the number of fine particles and the particle size distribution, the light receiving efficiency distribution in the light receiving system visual field region caused by the limitation of the light receiving system visual field region is calculated, and the power of the light beam is calculated in this light receiving efficiency distribution. It is expected to be output from the device when the particle size corresponding to the representative particle size is measured for each particle size category defined by the device, based on the corrected light receiving efficiency distribution that is corrected according to the distribution. Calculate the particle size distribution of each,
The feature is that the difference between the expected output of each particle size distribution and the true particle size distribution is obtained as each error distribution, and the particle size distribution of the measurement result is corrected to the true particle size distribution based on each error distribution. A method for correcting the particle size distribution in the measurement of fine particles.