JP3172105B2 - Refractive index measuring method and refractive index sensor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring the refractive index of various liquid samples with high accuracy, and more particularly, to a method of refraction utilizing a light-dark boundary at a critical angle of total reflection of a liquid sample. The present invention relates to a refractive index measuring method for measuring a refractive index and a refractive index sensor.

[0002]

2. Description of the Related Art Identification of substances, measurement of solution concentration, measurement of the degree of mixing of a mixture, measurement of impurity concentration, monitoring of precipitates and precipitates in a solution, monitoring of reaction state in a liquid, polymerization Measurement of refractive index is used in monitoring and the like. At the manufacturing site of the petroleum industry, the concentration of other components mixed into the target petroleum product is determined by measuring the refractive index. For example, when producing octane, butane is likely to be mixed, and the refractive index of octane mixed with butane is lower in accordance with the mixed ratio of butane than the refractive index of pure octane (1.39). Therefore, by measuring the refractive index of the generated octane, the mixing ratio of butane can be determined. In addition, refractometers using such a principle are also used in fields dealing with beverages, foods, medicines, flavors, brews, surfactants and the like.

Various types of refractometers are known for measuring the refractive index. For example, Abbe's refractometer sandwiches a sample between opposing slopes of two right-angle prisms,
A liquid sample layer of about m is formed, and an emission angle corresponding to a critical angle when the sample layer is irradiated with light is measured. Since Abbe's refractometer is a transmission type, it cannot be used for, for example, a deeply colored sample such as crude oil and the like, and is not suitable for continuous monitoring at a manufacturing site because the sample needs to be injected. A reflection type refractometer is known as a refractometer for solving such a disadvantage of the transmission type refractometer. For example, JP-A-2-1141
No. 51 is a reflection type refractometer which irradiates a light beam to a measurement surface under an angle range including a limit angle with respect to total reflection of a sample, and measures a reflected light from the measurement surface as a change in light intensity with a receiver. Is disclosed. In addition to this, for example, a refractometer (SSR) manufactured by US
-72 type) and a process refractometer (PRM series) of ATAGO using a bulk prism. However, these refractometers have problems such as an increase in the size of the apparatus due to the use of a bulk prism, a lamp light source, etc., a long time for measurement, and insufficient measurement accuracy.

The applicant has filed International Publication No. WO 94/245.
No. 43 discloses a total reflection type refractive index sensor that does not use a bulk prism or a lamp light source. This sensor has a waveguide layer on a substrate, a light incident surface connected to an optical fiber that allows light to enter the waveguide layer, a detection surface that comes into contact with an object, and light reflected from the detection surface. And a light output surface for outputting the light, and light detection means such as a CCD sensor connected to the light output surface. Since the light emitted from the optical fiber has a unique spread angle, the light reaches the detection surface of the subject as a light beam centered on the central incident angle. If the critical angle of total reflection of the object is within the range of the spread of the central incident angle, reflection conditions on the detection surface differ from the critical angle, so that the light intensity on the light exit surface also changes depending on the position. This change in light intensity is detected by the CCD sensor as a light-dark boundary which is a boundary between the total reflection portion and the transmission / reflection portion. Therefore,
The correspondence between the light-dark boundary position and the refractive index is obtained in advance, and the light-dark boundary position due to the total reflection light from the subject is obtained by taking the ratio between the measurement light and the reference light, so that the refractive index of the subject can be easily determined. And it can be obtained accurately.

[0005]

As described in Japanese Patent Application Laid-Open No. HEI 8-114544 filed by the present applicant,
When the light-dark boundary is determined by the CD sensor, the change of the light-dark portion near the critical angle of total reflection is not so remarkable.
The output (light intensity ratio) of the CCD changes relatively gently with respect to the position on the CCD (corresponding to the incident angle) as shown in FIG. Conventionally, in order to determine a light-dark boundary, the measured light amount and the blank light amount (light amount measured when the subject is air) when the measured light amount is observed from the dark part to the bright part.
Were initially regarded as equal (the light intensity ratio was 100%) as a light-dark boundary. However, in such a determination method, the measurement results vary widely, and the refractive index cannot be obtained with sufficient accuracy.

JP-A-6-294737 discloses a method and an apparatus for measuring a refractive index by a total reflection method using a CCD sensor as a device for receiving light reflected from a sample boundary surface. In this publication, a total reflection critical angle is calculated by interpolating a light quantity distribution curve detected by a CCD sensor with a theoretical light quantity distribution curve. However, as disclosed in Japanese Patent Application Laid-Open No. HEI 8-114544, Fresnel diffraction occurs on the CCD sensor, and therefore the detected light amount distribution curve is accurately interpolated unless the theoretical curve is considered. It is not possible.

For this reason, there is a demand for a method capable of accurately and accurately determining a light-dark boundary in a refractive index measurement method using a linear array type sensor such as a CCD sensor.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a refractive index measuring method and a refractive index sensor capable of accurately and accurately measuring a light-dark boundary at a critical angle of total reflection.

[0009]

According to a first aspect of the present invention, a subject is irradiated with light at various angles of incidence, the amount of light reflected from the subject is measured, and the total reflection condition of the subject is measured. In a refractive index measuring method for determining the refractive index of an object by determining a light-dark boundary based on a function indicating the characteristic of the amount of reflected light with respect to an incident angle or a reflection angle for a reference sample whose light-dark boundary is known, and an incident angle for the object Or a step of obtaining a function indicating the characteristic of the amount of reflected light with respect to the reflection angle, and a step of calculating the correlation between both functions while shifting at least one of the functions by a predetermined incident angle or reflection angle,
A method of measuring refractive index is provided, comprising the step of determining a light-dark boundary of a subject from a shift amount of an incident angle or a reflection angle at which the correlation between the two functions is greatest and a light-dark boundary of a reference sample. .

The principle of the refractive index measuring method of the present invention will be described with reference to FIG. FIG. 12 conceptually shows a profile of reflected light from the subject when a CCD sensor is used as a means for detecting a light-dark boundary based on the total reflection condition.
The reflection profile increases and decreases in the total reflection area of the reflected light, and reflects Fresnel diffraction. As a reflection profile, a reflection profile R _m (x) of a subject (a sample whose refractive index is unknown) and a reflection profile R _r (x) of a reference sample, as defined by the following equation, are calculated by the pixel position x in the CCD. Of the CCD sensor output with respect to.

[0011]

(Equation 1) (Where f _b (x) is the output profile (background) of the CCD sensor when the subject is not irradiated with light at all, and f _a (x) is the CC when the air is used as the subject.
The output profile of the D sensor, f _s (x) represents the output profile of the CCD sensor of the subject whose refractive index is unknown.)

Here, the CCD sensor output and the CCD sensor
The pixel position x in the pixel direction is the reflected light intensity and the reflected light, respectively.
Since it corresponds to the reflection angle of light, the reflection profile
The critical angle of total reflection by determining the position of the boundary between light and dark
And then determine the refractive index based on Snell's law.
Can be. In the present invention, the light-dark boundary position and the refractive index are already set.
Using a known reference sample, according to the following equation (2),
Reflection profile R_r(X) is replaced by a small unit i in the x direction
While shifting, the reflection profile R of the reference sample_r(X + n
i) (n is 0, ± 1, ± 2... ± q) and the reflection of the subject
Profile R _m Cross-correlation coefficient φ with (x)_n Each
Ask for it.

[0013]

(Equation 2)

However, the reflection profiles R _r (x) and R
_m (x) is a predetermined range (x1
.Ltoreq.x.ltoreq.x2) are functions standardized according to the following equation (3).

[0015]

(Equation 3)

Next, the x-direction shift amount ni of the reflection profile R _r (x) of the reference sample which gives the largest cross-correlation coefficient among the obtained cross-correlation coefficients φ _n is obtained. This shift amount ni corresponds to a shift of the light-dark boundary of the subject from the light-dark boundary of the reference sample. Therefore, the light-dark boundary of the subject can be obtained from the shift amount ni and the known light-dark boundary of the reference sample, and thereby the refractive index of the subject can be obtained.

In the refractive index measuring method according to the present invention, for the functions R _r (x) and R _m (x), cross points x _cr and x _{cm at} which the output of the CCD sensor is first equal to the output of the blank are obtained. , One of the functions is shifted by d1 in the pixel position direction so that their cross points coincide with each other, a correlation coefficient between the one function and the other function after the shift is obtained, and While shifting the function in the pixel direction by a small distance Δx, the correlation coefficient between the one function and the other function is obtained, and the shift amount d2 at which the correlation coefficient is the largest is obtained. It is preferable to determine the light-dark boundary of the subject from the sum (d1 + d2) and the light-dark boundary of the reference sample. In this way, the range including the light-dark boundary can be easily specified in the step of obtaining the first cross point, and the correlation coefficient is calculated over the range in the next step to shorten the light-dark boundary. It can be determined on time and accurately.

Further, after moving the one function in the pixel position direction by (d1 + d2), the phase of the one function and the other function is shifted while shifting the one function by a unit Δx ′ smaller than the minute distance Δx. The respective relation numbers are obtained, and the shift amount d3 at which the correlation coefficient becomes the largest is obtained, and the sum (d1 + d2 + d) of the respective shift amounts is obtained.
More preferably, the light-dark boundary of the subject is determined from 3) and the light-dark boundary of the reference sample. As a result, the light-dark boundary can be determined accurately and more precisely.

The present invention also provides a refractive index sensor suitable for performing the above-described refractive index measuring method. That is, according to the second aspect of the present invention, it has a contact surface with the subject whose refractive index is to be measured, and emits a light incident path on which light enters the contact surface and reflected light from the contact surface. A waveguide layer having an exit path formed therein, a light supply source connected to the waveguide layer for supplying light to the incident path, and detecting the reflected light emitted from the waveguide layer through the exit path. An array-type sensor for the, refractive index sensor to determine the refractive index of the subject based on a light-dark boundary corresponding to the total reflection critical angle observed from the array-type sensor, wherein the light-dark boundary is a known reference sample and From the result of detection of the unknown subject by the array type sensor, a function R _r (x) indicating the characteristic of the amount of reflected light with respect to the array position x of the array type sensor for the reference sample, and the array of the array type sensor for the subject Reflected light for position x The cross-correlation coefficient with determined over a predetermined range of the function showing the characteristics _{R m (x), R m} (x + ni) and R
_r (x) or R _m (x) and R _r
(X + ni) (where i is a value equal to or smaller than the array unit of the array type sensor, and n is ± 1, ± 2,...) Over a predetermined range. Means, and second calculation means for obtaining a light-dark boundary of the subject from ni indicating the largest cross-correlation coefficient among the obtained cross-correlation coefficients and a known light-dark boundary position of the reference sample. Is provided.

In the refractive index sensor according to the present invention, the array type sensor is a linear array type sensor arranged in the x direction, and the characteristics of the amount of reflected light with respect to the arrangement position x of the linear array type sensor with respect to the reference sample and the subject; Function R
_r (x) and R _m (x), and R _m (x + ni) and R _r
Preferably, the apparatus further comprises means for storing at least one of (x) or a cross-correlation coefficient between R _m (x) and R _r (x + ni).

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments and examples of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view of a main part of a specific example of a total reflection type refractive index sensor according to the present invention. The total reflection type refractive index sensor 50 has a waveguide structure composed of the cladding 2 / core 3 / cladding 4 between the substrate 1 and the substrate 6. Such a structure is formed by sequentially laminating a clad glass, a core glass, and a clad glass on a substrate 1, and further attaching a substrate 6 via an adhesive 5. The lower substrate 1 and the upper substrate 6 can be, for example, a Si substrate or a metal substrate. As the material of the core 3 and the claddings 2 and 4, materials generally used as materials for optical fibers can be used. For example, the material of the core / cladding is SiO ₂ / SiO ₂ + GeO ₂ , SiO ₂ / SiO ₂ +
TiO ₂ , SiO ₂ + SiF ₄ / SiO ₂ or the like can be used. The claddings 2, 4 and the core 3 can be deposited on the lower substrate 1 by a conventional film forming technique such as CVD or sputtering. As the adhesive 5, for example, an epoxy resin is used.

The structure of the total reflection type optical sensor 50 is not limited to the above-described laminated structure. Instead of the waveguide layer, a waveguide structure of clad / core / clad is formed by a waveguide glass having a thickness of 0.5 mm to 1 mm. A structure may be used in which the core is formed and adhered so that the core is sandwiched between cladding materials, and further, if necessary, sandwiched between substrates. For the core waveguide glass, glass such as quartz glass or optical glass, or for optical crystals such as sapphire, zirconia or diamond, and for the cladding material, silica glass or optical glass with a lower refractive index than the core waveguide glass Glass, sapphire, optical crystals such as zirconia and diamond, and substrates having good thermal conductivity such as Si and metal are used. For the adhesive between the clad and the core, for example, epoxy resin for optical use is used,
As an adhesive between the substrate and the clad, for example, an epoxy resin or the like is used.

This laminate has a light incident surface 7 for allowing light to enter the waveguide layer formed by the cladding 2 / core 3 / cladding 4;
And a light emitting surface 9 for outputting reflected light. The light incident surface 7 is connected to an optical fiber coupling section 11 in which a single mode optical fiber 10 is embedded. The light emitting surface 9 is connected to the CCD sensor 32. The optical fiber 10 is, for example, GaAs-Al
It is connected to a light source (not shown) such as a semiconductor laser such as GaAs, a He—Ne laser, and a light emitting diode (LED).

Next, FIG. 2 is a diagram for explaining the detection system and operation principle of the total reflection type optical sensor 50 of FIG. FIG.
Shows a cross-sectional structure of the total reflection type optical sensor 50 including the core 3 of the waveguide layer of FIG. The optical fiber 10 is a light incident surface 7 of the waveguide layer.
And the light from the light source is
At the light incident position 7 ′, the light is emitted from the end of the optical fiber 10 into the waveguide layer at a spread angle of ± Δ, usually about 6 to 8 degrees, from which the waveguide layer ( Through the incident path) and reaches the detection surface 8 in contact with the test liquid M with a spread angle (α ± Δ) centered on the central incident angle α. The center of the arrival point is shown as B and both ends are shown as A and C. At the point A, the incident angle is (α−Δ), and at the point C, the incident angle is (α + Δ). If the light is reflected from the interface between the detection surface 8 and the test liquid M, the reflected light from the points A, B and C
At points D, E, and F at the light emission position 9 ′. Points D to E
The light that has reached the interval between F and F reaches the CCD sensor 32. CCD
The sensor 32 includes a linear array type CCD sensor in which a plurality of pixels are linearly arranged in the direction from D to F (for example,
A pixel having 2048 bits (2048 bits) in the direction from D to F (14 μm / pixel) can be used.

In order to accurately determine the light / dark boundary of the emitted light, C
The output of the CD sensor 14 is output to a calculation unit 34 via a signal line 33.
Send to The calculation unit 34 calculates a function, a correlation coefficient, and the like defined by an expression described later. Both the first and second calculation units in this specification can be the calculation unit 34. The calculation unit 34 can further include a memory (not shown) for storing the calculation result.

As shown in FIG. 2, the incident light from the optical fiber 10 has a spread angle α ± Δ, and the reflected light corresponding thereto is detected by the CCD sensor 32. The reflection angle at which total reflection from M occurs is determined by the CCD sensor 32.
Can be detected as a light-dark boundary. By measuring the light-dark boundary using several test liquids having different refractive indexes in advance, the refractive index of the test liquid and the critical angle for total reflection, C, are measured.
The relationship with the pixel position of the CD sensor 32 can be obtained in advance.

Next, the subject M detected by the CCD sensor
A method for determining a light / dark boundary in light reflected from a camera will be described. In the present invention, using the output pattern of the reference sample whose refractive index is known in advance by the refractive index sensor 50 (the output pattern of the CDD sensor), the light-dark boundary is determined from the output pattern of the subject according to the following first to fourth steps. decide.

[First Step—Measurement of Reflection Profile of Subject] First, in the following three cases, the output profile of the CCD sensor with respect to the pixel position x of the CCD sensor is measured, and these are defined as follows. . 1) Output profile of the CCD sensor when the subject is not irradiated with any light (optical black): f _b (x) 2) Output profile of the CCD sensor when using air as the subject (blank): f _a ( x) 3) Output profile of the CCD sensor when using an object whose refractive index is unknown: f _s (x)

A ratio R _s (x) as defined in the following equation (1) is calculated using the function indicating each output profile.

(Equation 4) _{Here, (f s (x) -} f b (x)) represents the change in the amount of light reflected from the subject for the pixel position x, (f _a
(x) −f _b (x)) indicates a change in the amount of reflected light from air with respect to the pixel position x. These ratios R _s (x) indicate the ratio of the amount of reflected light of the subject to the total amount of reflected light from air. In theory, when total reflection from the subject occurs, R _s (x) = 1. Become. Therefore, R _s (x) is represented by a graph with respect to x, and a cross point x _cm where R _s (x) = 1 is first obtained when viewed from the dark side (a pixel position corresponding to a higher incident angle). .

[Second Step: Deviation d1 between Cross Points]
Measurement] Next, a reference sample whose refractive index is known in advance is prepared, and the cross point _xcs is determined for the reference sample as well. Here, R _s (x) of the reference sample is represented by R _r (x), and R _s (x) of the subject is represented by R _m (x). However,
R _r (x) and R _m (x), respectively, the range for obtaining the cross-correlation coefficient to be described later _{R s (x) ((x} cm -b) ~ (x cm +
The function is standardized according to the following equation (3) ′ over a)).

(Equation 5)

In order to match the cross point x _cs of the reference sample with the cross point x _cm of the subject, R _r
The profile representing (x) is moved in the x direction. This movement amount is defined as d1. That is, d1 = x _cm −x _cs .

[Third Step—Measurement of Pixel Deviation d2 by Cross-Correlation Coefficient] Next, the mutual phase relationship between R _r (x) after moving as described above and R _m (x) of the subject. The number is obtained based on the following equation (2) ′. The range for obtaining the cross-correlation coefficient is a range of a pixel in the + direction and b pixel in the-direction near the cross point x _cm . This range (x _cm
It is preferable to select a range from -b) to (x _cm + a) up to a position where a bright and dark portion is generated by Fresnel diffraction, and in this example, a range including up to the third-order Fresnel diffraction light is set.

[0033]

(Equation 6) Next, a profile R _r (x + d1 + n) in which R _r (x) is moved one pixel at a time (n is a moving pixel amount, and n = ±
1, ± 2...) And R _m (x) are obtained in accordance with equation (2) ". After obtaining the cross-correlation coefficients for each moving pixel amount, the cross-correlation coefficients are obtained. Is obtained, and the amount of movement n is set to d2.

(Equation 7)

[Fourth Step—Measurement of Pixel Deviation d3 by Cross-Correlation Coefficient] After the third step, the function R _{r of the} reference sample is obtained.
The profile of (x) is set in the pixel direction (x direction) by (d1
After moving by + d2), each time the profile of R _r (x) is moved in the range of ± 0.9 pixel in 0.1 pixel units, a cross-correlation coefficient is calculated according to the following equation (4). Then, the movement amount h that gives the largest value among the cross-correlation coefficients is obtained, and this is set as d3.
Note that the R _r (x) of 0.1 pixels, R _r
(n) and the interpolated value of R _r (n + 1) or R _r (n) and R _r (n-1),
_{r (n) + i (R} r (n + 1) -R r (n)) ( where, i = 0.1, 0.2,
..., 0.8, 0.9) or _{R r (n) + i (} R r (n) -R r
(n-1)) (where i = -0.1, -0.2, ..., -0.8, -0.9)
Was used. Also, interpolation values from adjacent ± 1 pixels,
For _{example, (R r (n-1} ) + 2R r (n) + R r (n + 1)) / 4 + i
(R _r (n + 1) −R _r (n−1)) / 2 (where i = −0.9, −0.8,
・ ・, 0, ・ ・, 0.8, 0.9).

(Equation 8)

The sum of the pixel movement amounts Δd = (d1 + d2 +) at which the cross-correlation coefficient is the largest in the second to fourth steps.
d3) is calculated. Δd indicates a shift amount of the position of the light-dark boundary of the subject with respect to the reference sample whose light-dark boundary is known. Therefore, the light-dark boundary position of the subject can be specified from the light-dark boundary position of the reference sample and Δd. Examples in which the method of the present invention is more specifically implemented will be described below.

[0036]

[Example] Hexane (n-hexane), octane (n-octane),
C10 (decane (n-decane)), C12 (dodecane (n-dod)
ecane)), and using cyclohexane and carbon tetrachloride as the test subjects, the profile of the reflected light from each test subject was obtained using the total reflection optical sensor 50 shown in FIGS. 1 and 2 (the first step described above). .

The materials and dimensions of the total reflection type refractive index sensor 50 are as follows. Optical glass having a thickness of 1 mm as a core layer (manufactured by Schott, trade name: BK7 (refractive index: 1.516
Using 3)), a structure sandwiched between silicon substrates having a thickness of 0.5 mm was adopted. The waveguide glass and the silicon substrate were bonded to each other with a thermosetting resin. A multi-mode fiber was used as the optical fiber 10 for supplying incident light to the total reflection type refractive index sensor, and its end face was polished so that the outgoing light had a spread of 6 °. As incident light, monochromatic light of 589 nm obtained by dispersing light from a white light source with an interference filter was incident on the optical fiber 10. The incident angle was 71 °, and the optical path length in the waveguide layer of the sensor was 100 mm. CC
The accumulation time of the D sensor was set to 40 ms. The sensitivity of the CCD sensor is 59.4v / Lxs. The temperature of the subject was adjusted to 20 ° C. using circulating water in a thermostat. In order to measure the baseline (background) of the CCD sensor, stray light was prevented by covering the sensor section with a dark curtain.

FIG. 3 shows the profile of the reflected light for each of the above objects. FIG. 3 shows the output of the CCD with respect to the pixel position (pixel No.) of the CCD sensor. The measurement was performed using air as the subject (displayed as ref) and the baseline of the CCD sensor (displayed as dark)
At the same time. Note that the baseline of the CCD sensor corresponds to the above-described optical black f _b (x). FIG.
It can be seen that the light-dark boundary of hexane is located outside the half-value width of the total reflection light from the air surface.

[0039] was converted into R _s according to the equation from the curve shown in FIG. 3 above for each subject (1). Therefore, represent R _s (x) is a graph for the pixel location x, the first R _s as viewed from the dark portion side (pixels corresponding to the higher incidence angle position)
A cross point x _{cm where} (x) = 1 was obtained.

The maximum point of the reflected light profile shown in FIG. 3 of the subject is located at the center of the pixel at C12.
Is used as a reference sample. Normalized in accordance with the second step described above of Rs (x) for each analyte and reference sample, respectively and converted to R _m (x) and _{R r} (x). FIG. 4 shows R _m (x) when octane is used as a subject. FIG.
Shows the results of 10 measurements, superimposed, and shows the vicinity of the primary Fresnel diffraction pattern in an enlarged manner. In the conventional method for determining a light-dark boundary, a light-dark boundary is determined from a cross point where R _m (x) = 1. However, in this method, there is a variation of about one pixel in the light-dark boundary between ten measurements. Understand.

Next, after calculating the cross point x _{cr of} C12 and the shift amount d1 for matching the cross point x _cm of each subject according to the second step, R _m (x) and R _r are determined according to the third step. The cross-correlation coefficient of (x + d1) was determined. The integration range in equation (3) ′ is a = 30, b = −1
0 was set. For octane and C12, d1 (octane)
= -316. The cross-correlation coefficient between each subject and C12 was determined for the case where the waveform (R _r (x)) of C12 was moved by one pixel from the cross point x _cm of each subject over five pixels before and after. FIG. 5 shows the relationship between the pixel movement amount of octane and C12 and the cross-correlation coefficient. FIG.
It can be seen that the correlation between the two functions R _m (x) and R _r (x) at a position shifted by +1 pixel from the octane cross point x _cm is higher. That is, d2 (octane) = 1.

Then, according to the fourth step described above, C12
After shifting R _r a (x) (d1 + d2) , i.e., R _r (x) is after the _{R r (x + d1 + d2} ), R r (x + d1 + d2) portionwise 0.1 pixels over more ± 1 pixel When moving, R _m (x) and R _r (x +
d1 + d2 + h) (h is ± 0.1, ± 0.2... ± 0.9). Note that 0.
For R _r (x) in units of one pixel, R _r (n) and R _r
(n + 1) or the interpolated value of R _r (n) and R _r (n-1), R _r (n) +
i (R _r (n + 1) −R _r (n)) or R _r (n) + i (R _r (n) −
R _r (n-1)) (where i = 0.1, 0.2, ..., 0.8, 0.9)
Was used. FIG. 6 shows the results. Figure 6 than C12 waveform (function Rr (x + d1 + d2) ) when the moved by -0.2 pixels both function _{R r (x + d1 + d2} ) and R _m (x)
(Octane) has the highest correlation. That is, d3 = −0.2.

From the above results, the light-dark boundary of octane is
D1 + d2 + d3 = −316 + 1 from the light / dark boundary of C12
−0.2 = −315.2 pixels were found to be present at the position shifted.

The same procedure as that for octane was carried out for the other specimens other than octane, and the above-described fourth step was carried out.
The amount of pixel movement for the light-dark boundary of No. 2 was determined. The results are shown in the table below.

[0045]

[Table 1] Object Pixel movement amount (d1 + d2 + d3) Hexane -568.4 pixels Octane -315.2 pixels C10 -133.1 pixels Cyclohexane + 62.5 pixels Carbon tetrachloride +594.0 pixels

The light-dark boundary obtained as described above and the refractive index obtained therefrom were verified by the following method. It has been found that the refractive index of a liquid generally changes with a slope of about 4.3 × 10 ⁻⁴ / ° C. Therefore, first, the light-dark boundary and the refractive index of the subject were measured at various temperatures, and the refractive index was determined from the light-dark boundary of the subject. 15 ° C ~
The measurement was performed at several temperatures in a temperature range of 25 ° C., and a relational expression between the temperature and the refractive index was obtained. The temperature was measured using a Pt-100 ohm resistance thermometer. Next, in the case of using octane as a subject, measurement was performed 10 to 20 times, and a refractive index at a temperature of 20 ° C. was obtained based on a relational expression. FIG. 7A shows the relationship between the refractive index at a temperature of 20 ° C. and its frequency. For comparison, the same test was performed for the case where the light-dark boundary and the refractive index were determined only from the cross points corresponding to the conventional method. The results obtained by the conventional method are also shown in FIG.

In FIG. 7, it can be seen that the results of the present invention obtained in the first to fourth steps have smaller variations in the obtained refractive index than in the case of the prior art. Standard deviation σ
_{When n-1} was obtained, the result of this method was as large as 1.4 × 10 ^-5 and in the case of the prior art, as large as 4.3 × 10 ^-5 . Therefore, it can be said that the refractive index obtained by the method for determining a light-dark boundary of the present invention has higher reliability than the conventional method.

In the same manner as above, the relational expression between the temperature and the refractive index is obtained for the test object other than octane.
Was determined. FIG. 8 shows a case where hexane is used as a subject, FIG. 9 shows a case where C10 is used as a subject, FIG. 10 shows a case where cyclohexane is used as a subject, and FIG. 11 shows a case where carbon tetrachloride is used as a subject. Each is shown. The following table shows the results of this method and the standard deviation σ _n-1 obtained from the results of the conventional method.

[0049]

Table 2 Standard deviation σn-1 Subject Conventional method Method of the present invention Octane 4.3 × 10 ^-5 1.4 × 10 ^-5 Hexane 1.0 × 10 ^-4 1.6 × 10 ^-5 C10 2.0 × 10 ^-5 5.7 × 10 ^{− 6} Cyclohexane 2.4 × 10 ^-5 7.9 × 10 ^-6 Carbon tetrachloride 2.5 × 10 ^-5 1.7 × 10 ^{-5 From the} above results of hexane, C10, cyclohexane and carbon tetrachloride, the method of the present invention is compared with the conventional method. It can be seen that good results have been obtained.

As described above, the refractive index measuring method and the refractive index sensor of the present invention have been specifically described with reference to the embodiments and the examples. However, the present invention is not limited to these, and the modified examples of the disclosed specific examples can be used. It goes without saying that it includes improvements. For example, in the embodiment, the waveform of the reference sample is shifted by fixing the waveform of the subject, but the waveform of the reference sample may be fixed by shifting the waveform of the subject. Further, both waveforms may be moved. In addition, the integration range for obtaining the shift amount and the correlation coefficient may take any value according to the type of the subject or the reference sample.
In addition, although the waveform of C12 was used as the waveform of the reference sample, it is also possible to use a waveform of another sample having a similar refractive index, a waveform obtained by deforming this waveform into a similar shape, or a waveform obtained from a theoretical formula of Fresnel diffraction. good. Further, as the material and the structure of the refractive index sensor, any material can be adopted as long as the principle of the present invention can be implemented. For example, using a single mode optical fiber as the optical fiber,
By providing a lens on the incident surface, it is possible to change the shape of the reference light and improve the accuracy of the measurement of the refractive index of an object having a light-dark boundary at a portion where the light intensity of the reference light is weak.

[0051]

According to the refractive index measuring method of the present invention, while the reflection profile of the reference sample having a known light-dark boundary is shifted with respect to the reflection profile of the object, the shift that maximizes the cross-correlation coefficient between the two. Since the light-dark boundary of the subject is determined by obtaining the amount, the light-dark boundary can be determined extremely accurately and with high accuracy. Further, even if Fresnel diffraction appears in the reflection profile, the light-dark boundary can be accurately determined without being affected by the Fresnel diffraction. Further, after first specifying the range where the light-dark boundary appears by the cross point, the shift amount is determined using the cross-correlation coefficient, so that the light-dark boundary can be determined efficiently and accurately. Further, by calculating the correlation coefficient while shifting the reflection profile of the reference sample by dividing the shift amount into several steps, it is possible to perform measurement in a shorter time and with higher accuracy.

The refractometer of the present invention used on the basis of such a refractive index measuring method is extremely suitable for highly accurate measurement of other physical quantities measured using the refractive index as a parameter, for example, temperature, precipitation point, concentration and the like. Becomes effective.

[Brief description of the drawings]

FIG. 1 is a perspective view of a main part of a specific example of a total reflection type refractive index sensor according to the present invention.

FIG. 2 is a sectional view of a sensor including a core layer of the sensor shown in FIG. 1, illustrating a detection system and an operation principle of the sensor.

FIG. 3 is a graph showing reflection profiles of various subjects.

FIG. 4 shows R _m when octane is used as a subject.
It is a graph which shows (x).

FIG. 5 is a graph showing the relationship between the cross-correlation coefficient of the reflection profile of octane and C12 and the amount of pixel movement (1 pixel unit).

FIG. 6 is a graph showing a relationship between a cross-correlation coefficient between R _s (x + d1 + d2) and an octane reflection profile R _m (x) and a pixel movement amount (0.1 pixel unit).

7 (a) shows the refractive index of octane obtained by the method of the present invention at a temperature of 20 ° C. and its measurement frequency, and FIG. 7 (b) shows the refraction of octane obtained by a conventional method at a temperature of 20 ° C. It is a graph which shows the relationship between a rate and the measurement frequency.

8 (a) shows the refractive index of hexane obtained at the temperature of 20 ° C. obtained by the method of the present invention and its measurement frequency, and FIG. 8 (b) shows the refractive index of hexane obtained at 20 ° C. obtained by the conventional method. It is a graph which shows the relationship between a rate and the measurement frequency.

FIG. 9 (a) shows the temperature of C10 obtained by the method of the present invention at 20 ° C.
9 (b) is a graph showing the relationship between the refractive index of C10 at a temperature of 20 ° C. and the measurement frequency obtained by the conventional method.

FIG. 10 (a) shows the refractive index of cyclohexane obtained at the temperature of 20 ° C. and the measurement frequency thereof,
0 (b) is a graph showing the relationship between the refractive index of cyclohexane obtained by the conventional method at a temperature of 20 ° C. and the measurement frequency.

FIG. 11 (a) is a graph showing the refractive index of carbon tetrachloride obtained by the method of the present invention at a temperature of 20 ° C. and its measurement frequency.
(b) is the temperature of carbon tetrachloride obtained by the conventional method at 20.
5 is a graph showing the relationship between the refractive index at ° C. and the measurement frequency.

FIG. 12 is a graph illustrating the principle of the refractive index measuring method of the present invention.

FIG. 13 is a graph illustrating a method for determining a light-dark boundary according to the related art.

[Explanation of symbols]

 Reference Signs List 1 lower substrate 2, 4 clad 3 core 5 adhesive layer 6 upper substrate 7 light incidence surface 8 detection surface 9 light emission surface 10 optical fiber 32 CCD sensor 34 operation unit 50 total reflection type refractive index sensor M test liquid

────────────────────────────────────────────────── ─── Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G01N 21/00-21/61 ECLA

Claims (8)

(57) [Claims] An object is irradiated with light at various angles of incidence, the amount of light reflected from the object is measured, and a light-dark boundary is determined based on the condition of total reflection of the object, whereby the refractive index of the object is determined. In the method for measuring the refractive index to be determined, a step of obtaining a function indicating the characteristic of the amount of reflected light with respect to the incident angle or the angle of reflection for the reference sample having a known light-dark boundary and a function of indicating the characteristic of the amount of reflected light with respect to the angle of incidence or the reflection angle of the subject And calculating a correlation between at least one of the functions while shifting at least one of the functions by a predetermined incident angle or a reflection angle; and a shift amount of the incident angle or the reflection angle and a reference sample at which the correlation between the two functions is maximized. Determining a light-dark boundary of the subject from the light-dark boundary of the object. 2. A light-dark boundary based on the total reflection condition is defined as CC
Observed using a D sensor, a function indicating the characteristic of the amount of reflected light with respect to the incident angle or the reflected angle of the reference sample and the subject is represented by the output waveform functions R _r (x) and R _m of the CCD sensor with respect to the pixel position x of the CCD sensor. The refractive index measuring method according to claim 1, wherein the refractive index is determined as (x). 3. For the functions R _r (x) and R _m (x), cross points x _cr , x _{cm at} which the output of the CCD sensor is first equal to the output of the blank are obtained, and the cross points match. One of the functions is shifted by d1 in the pixel position direction so as to obtain a correlation coefficient between the one function and the other function after the shift, and further, the one function is shifted by a small distance Δx for each pixel. While shifting in the direction, the correlation coefficient between the one function and the other function is obtained, and the shift amount d2 at which the correlation coefficient becomes the largest is obtained. The sum of the shift amounts (d1 + d2) and the reference sample are obtained. 3. The method according to claim 2, wherein a light-dark boundary of the subject is determined from the light-dark boundary. 4. After moving the one function in the pixel position direction by (d1 + d2), the one function and the other function are shifted by a unit Δx ′ smaller than the minute distance Δx in the pixel position direction. The correlation coefficient with the function is obtained, and the shift amount d3 at which the correlation coefficient is the largest is obtained. The light-dark boundary of the object is obtained from the sum (d1 + d2 + d3) of the shift amounts and the light-dark boundary of the reference sample. The refractive index measuring method according to claim 3, wherein 5. The function according to claim 2, wherein the functions R _r (x) and R _m (x) are standardized over a range in which the correlation coefficient is calculated. The described refractive index measuring method. 6. A light guide having a contact surface with an object whose refractive index is to be measured, wherein a light incident path for entering light to the contact surface and an exit path for emitting reflected light from the contact surface are formed. A wave layer, a light source connected to the waveguide layer and supplying light to the incident path, and an array-type sensor for detecting the reflected light emitted from the waveguide layer through the emission path, A refractive index sensor for determining a refractive index of the subject based on a light-dark boundary corresponding to a critical angle of total reflection observed from the array-type sensor, wherein the light-dark boundary is a known reference sample and an unknown object. From the result detected by the sensor, a function R _r (x) indicating the characteristic of the amount of reflected light with respect to the array position x of the array type sensor for the reference sample and the characteristic of the amount of reflected light with respect to the array position x of the array type sensor for the subject are shown. shows the function R _m ( ) And a cross-correlation coefficient with determined over a predetermined range of, the cross-correlation coefficient between R _m (x + ni) and R _r (x) or, R _m and _{(x) R r (x +} ni) ( where , I are values equal to or less than the array unit of the array type sensor,
n is ± 1, ± 2,...) over a predetermined range, a first calculating means, and a cross-correlation coefficient having the largest value among the obtained cross-correlation coefficients And a second calculating means for obtaining a light-dark boundary of the subject from ni and a known light-dark boundary position of the reference sample. 7. A linear array sensor in which the array sensors are arranged in the x direction, and a characteristic of a reflected light amount with respect to an arrangement position x of the linear array sensor for the reference sample and the subject, a function R _r (x). And R _m (x), and R _m (x + ni) and R _r (x) or R _m (x) and R _r
7. The refractive index sensor according to claim 6, further comprising means for storing at least one of the cross-correlation coefficients with (x + ni). 8. The refractive index sensor according to claim 6, wherein said array type sensor is a CCD sensor.