JPH0580614B2 - - Google Patents

Description

[Detailed description of the invention]

Color finishing equipment rarely, if ever, achieves a satisfactory color match to a color standard without performing an adjustment operation known as "shading." Color recoloring usually requires relatively minor but critical handling of the formulated pigment composition to modify the cumulative effects that manufacturing variables have on the pigment dispersion. Traditionally, recoloring operations are performed by highly skilled and experienced individuals who require extensive work experience to become proficient in their craft. After all, visual color correction is a kind of special technology, and effective process control is difficult. More recently, such visual recoloring has been supplemented by the use of equipment to instrumentally characterize the composition of paints or pigments.
Colorimeters and spectrophotometers are well known in the art and are used to measure certain optical properties of various coatings coated on test panels. A typical spectrophotometer measures the amount of light reflected at various wavelengths in the visible spectrum by a painted panel held at a fixed angle relative to the direction of the incident light source. The paint reflection coefficient allows paint chemists to calculate color values that characterize the color of various paints. For paints that do not contain light reflective flakes or fragments (ie, non-metallic paints), the reflection coefficient does not vary with the angle of the panel relative to the direction of the incident light except at the gloss (specular) angle. A single spectrophotometer reading at any particular angle therefore produces a reflectance value that accurately characterizes the paint. However, in order to obtain a pleasant aesthetic effect, light-reflecting flakes are often used in paints in the paint industry (i.e. metallic paints). Paints containing light-reflecting flakes, such as aluminum, bronze, coated mica, and other materials, are characterized by a "two-tone" or "flip-flop" effect in which the apparent color of the paint changes at different viewing angles. . This effect is due to the orientation of the flakes in the coating. Since the color of such metallic paints obviously varies as a function of the illumination and viewing angles, a single spectrophotometer reading is insufficient to accurately characterize the paint. Although measurement studies have shown that visual color differences between two metallic paints are detectable at an infinite number of angles, it is clear that practical reasons preclude the collection of reflection coefficients for an infinite number of viewing angles. It is. However, previous research has also shown that measuring the optical properties of metallic paints at just two specific angles can provide useful properties. See, eg, US Pat. No. 3,690,771, which is incorporated herein by reference. The present invention relates to the finding that unexpectedly improved optical properties of metallic paints are obtained when measurements are carried out at three specific angles. According to the invention, the optical properties of a paint containing metallic particles or flakes can be determined instrumentally using multi-angle spectrophotometric or colorimetric measurements to derive the color constants of the paint. An improved method is provided that consists of using three multi-angle measurements to characterize the It is recognized that directional reflection must be taken into account in the optical characterization of surfaces containing metallic particles, such as metallic paints and films. Metallic paints contain light reflective flakes or fragments such as aluminum, bronze, coated mica, or other materials. These flakes or fragments function much like small mirrors, reflecting light directionally rather than diffusely. The directional reflective properties of metallic coatings give rise to a phenomenon known as goniochromatism, which is defined as a variation in the color of the coating as a function of the direction of illumination and observation. This phenomenon is also sometimes described as "two-tone,""flop,""flip-flop,""flash,""sidetone," etc. In short, the color of metallic paint appears differently at different viewing angles. To explain this directional or angular reflection, or goniochromatism, the reflection coefficient measured by a spectrophotometer must be employed at multiple angles. The reflection coefficient of a coating is the ratio of the luminous flux reflected from a coating sample to the luminous flux reflected from a perfectly reflective diffuser when both the sample and the perfect diffuser are illuminated identically. A completely white reflector has a value of 1. A completely black non-reflector has a value of 0. Reflection coefficients are used to calculate color descriptor values used to define colors and color differences. The color tristimulus values (X, Y, Z) are the reflection coefficient data (R) and the sensitivity of the human eye (,
y, ) and the illumination of the light source (E), all of which are functions of the wavelength (λ) of the visible spectrum. The equation for determining tristimulus values is as follows. X = ₃₆₀ ∫ ⁸³⁰ R (λ) E (λ) (λ) dλ Y = ₃₆₀ ∫ ⁸³⁰ R (λ) E (λ) (λ) dλ Z = ₃₆₀ ∫ ⁸³⁰ R (λ) E (λ) (λ) dλ This tristimulus value can be used to calculate color description values related to the visual perception of colors and color differences.
One of the many descriptive value combinations that can be used is
CIELAB Perceptual Color Scale (CIELAB
perceptual color scale), which is recommended by the International Commission on Illumination. [Recommendations on Uniform Color Spaces, Color Difference Equations, Psychometric Color Terms]
Spaces，Color Difference Equations，
Psychometric Color Terms)” Supplement No.
2-C Publication (CIE)
Publication) No.15 (E1.3.1) 1971/CT (1.3)
1978, Bureau Central De La CIE] Tristimulus transformations were used to calculate lightness (L ^* ), redness/greenness (a ^* ), yellowness/blueness ( b ^* ), perceptual color values describing saturation (C) or hue (H) can be calculated. Colors are completely described by combinations of L, a, b or L, C, H values.
The following equation, specified by the International Commission on Illumination, relates the tristimulus values to L ^* , a ^* , and b ^* . L ^* = 116 (Y/Yo) ^1/3 -16 a ^* = 500 [(X/Xo) ^1/3 - (Y/Yo) ^1/3 ] b ^* = 200 [(Y/Yo) ^1/3 −(Z/Zo) ^1/3 ] where Xo, Yo and Zo are the tristimulus values of a perfectly white body for the given illumination, and X, Y and Z are the tristimulus values for the color. The saturation (C) and hue (H) descriptive values are related to the a ^* and b ^* values as follows. C = (a ^*2 + b ^*2 ) ^1/2 H = tan ^-1 (b ^* / a ^* ) Often, a sample batch of a paint, for example, is compared to a standard color, the difference is measured, and then Adjustment of the sample with appropriate additives is required to bring the sample within acceptable standard values. The difference in color between the color standard and the batch sample is described as follows. △L ^* = L ^* (Batch) - L ^* (Standard) △a ^* = a ^* (Batch) - a ^* (Standard) △b ^* = b ^* (Batch) - b ^* (Standard) The obtained value is Conforms to visual assessment of lightness (ΔL ^* ), redness/greenness (Δa ^* ) and yellowness/blueness (Δb ^* ). Further discussion will focus on illumination and observation conditions for goniochromatic colorimetry using tristimulus values (X, Y, Z) and perceptual color values (L ^* , a ^* , b ^* , C, H). Quantify the impact of changes in
The particular color description value used is only one of many possible choices of tristimulus value transformations that can be used for this task. Tristimulus values for metallic paints and therefore
The L ^* , a ^* , and b ^* values change regularly due to regular variations in the observation angle of the coating. FIG. 1 shows the directional color behavior of a solution lacquer medium red metallic color. Samples were prepared by spraying a conventional air mist onto an aluminum substrate and then baking at 155° C. for 30 minutes. Standard white (BaSO ₄ )
Reflection coefficient measurements were carried out in six sets of illumination and observation directions using a reflection spectrophotometer specifically designed to measure reflection properties in various measurement configurations. This device is essentially a standard spectrophotometer consisting of a light source, a monochromator, a variable measurement arrangement module, a photodetector and associated control and readout electronics. Tristimulus values are calculated as described above using the reflection coefficients for each measured feature array. Figure 1 shows the tristimulus value X,
The angular dependence of Y and Z is explained. The angle is measured from the specular angle (or "specular" angle). The values used in FIG. 1 are as follows.

[Table] The analysis of the angle dependence plot in Figure 1 is as follows.
Clarify the point. (1) Tristimulus values are not constant with respect to angular variation.
Values from multi-angle measurements are therefore required to accurately describe the color behavior of a sample. (2) The plot is a function that decreases monotonically as the angle from the specular angle increases, so a simple mathematical model should describe this curve. (3) The plot is a curve, and the mathematical model is probably higher than that of a linear equation (straight line). Similar angle dependence plots were obtained for a wide variety of metallic colors. And all show similar results. The significance of these results is that they define the problem of metallic color characterization and definition. Since all metallic colors exhibit monotonically decreasing tristimulus values with a similar curve in the measurement direction from the specular angle,
There is a systematic angular color behavior that allows a simple measurement scheme to be developed. Multiple measurements are required to fully characterize this behavior. Since L ^* , a ^* and b ^* are color values normally used for characterizing the color of coatings, the determination of the number of measurements required to satisfactorily characterize the angular dependence of these values is the most important. Since the plots of all three variables are fairly similar, L ^*
The mathematical property that fits the plot of the angular dependence of is
It is reasonable to assume that a plot of the angular dependence of a ^* and b ^* will also fit. The best fit of various mathematical models to the lightness (L ^* ) angular dependence curve is of a quadratic type requiring three measurements. This can be shown by considering the average residual error of the L ^* values for the six measurement directions as predicted by various prediction models using subsets of the six measurement directions. The reflection coefficients of six measurement configurations for 37 solution lacquer metallic colors are measured and the lightness values L ^* for these configurations are calculated. Samples were prepared and measured as described above. The objective is to create a system for metallic color characterization that provides optimal information with minimal effort. This is done by considering whether a subset of the six measured configuration data is sufficient to predict color brightness behavior in all six configurations. A linear metallic color characterization model based on linear equations is initiated. Such an equation for the angular dependence of metallic brightness has the form: L ^* = a ₁ + a ₂ 〓 where L ^* is the lightness, φ is the angle from the specular angle, and a ₁ and a ₂ are fitted from measurements using at least two different measurement directions. is a constant for each color characteristic). Similarly, a quadratic equation of the form is used. ^* L=a ₁ +a ₂ φ+a ₃ φ ² (variables in the formula are the same as in the linear example, but here another constant a ₃ is added)
A minimum of three measurement directions are required here. Table 1 shows the average sum of squared residuals for 6 measurement configurations for 37 metallic colors. If the average sum of squared residuals is low, a model is formed that describes the color dependence of the metallic color in the measurement direction.

Table: Adding one more measurement taken near the specular angle reduces the sum of squared errors of the prediction from 529.3 to 12.5 in L ^* units. This shows that the metallic color lightness behavior in any direction can be well predicted from measurements in three selected directions. Higher accuracy can be achieved by adding a larger number of measurement angles or by using a higher order equation using a larger number of measurement angles. However, none of such measures significantly and unexpectedly increases the accuracy that can be achieved by using a quadratic model incorporating one more angular measurement than a two-angle system. That is, three suitably selected measurement directions are the optimal choice for providing maximum information about the metallic color for minimum measurement effort. The example data shows optimal results for brightness values. Similar results are obtained for tristimulus values (X, Y, Z), perceptual color values (a ^* ,
b ^* , C, H), color difference values (ΔL ^* , Δa ^* , Δb ^* ) or other tristimulus values. Various measurement techniques can be used in collecting data regarding the optical characterization of coatings. The first technique is object modulated reflection (OMR), which fixes the source and observer or detector reference points and varies the position of the object. This technique is illustrated in US Pat. No. 3,712,745, which is incorporated herein by reference. Two other techniques are Detector Modified Reflection (DMR) and Irradiance Modified Reflection (IMR).
In DMR, the detector is varied while the light source and object are fixed. And in IMR, the irradiation or light source is varied, while the detector and object are fixed. As discussed above, analysis of the data (collected using DMR) shows that the optimal measurement combination for characterizing goniochromatic effects in metallic coatings is measured at three angles, namely (1 ) An angle close to the specular reflection angle, (2) Approximately 45° from the specular reflection angle
and (3) measurements made at angles far from the specular reflection angle. In general, for typical metallic colors, the angular dependence of L ^* , C, and H values is independent of whether it is measured by OMR, DMR, IMR, or some combination of these.
It has a quadratic expression for any set of angles that are regularly varied from near the specular angle to far from the specular angle. Therefore, whether OMR, DMR or IMR is used, the three measurements are
The angular dependence curves of L ^* , C, H can be optimally characterized. FIG. 2 represents a preferred embodiment of the invention in which a DMR is used. Incident light source against the paint film
Position it at a 45° angle. Three detectors are positioned at three different angles (measured from the specular angle) for the measurement of optical properties. (1) Detector No. 1 15° (near the specular reflection angle) (2) Detector No. 2 45° (perpendicular to the coating surface) (3) Detector No. 3 110° (far from the specular reflection angle) ) It may not be possible to choose the same angle for a system using IMR or OMR, but one skilled in the art can easily determine an appropriate angle. The improved method of the present invention can be used to characterize not only metallic coatings, but also any metallic particle-containing surface, such as plastics, including reflective metallic flakes. This improvement method is L ^* , a ^*
It is particularly useful when recoloring paints where the and b ^* values have been determined against a standard. in this case,
Manufacturing paint batches according to a certain recipe. I made this batch of painted panels, and the L ^* ,
Measure the a ^* and b ^* values. Often, even when carefully manufactured, paint batches do not meet standards because of pigment variations and pigment dispersion color drift. △L ^* , △a ^* and △b ^* of the batch
Calculate the value. If they are outside the acceptable tolerances, make calculations for adding pigments in the form of millbase and add this millbase to the batch and fabricate the second panel as described above. and measure its value. This process is repeated until an acceptable color match between the standard and the paint batch is achieved.

[Brief explanation of the drawing]

FIG. 1 is a graph showing the relationship between X, Y, Z tristimulus values and angle from specular reflection, and FIG. 2 is a schematic representation of a preferred spectrophotometric system for practicing the method of the invention.

Claims (1)

[Scope of Claims] 1. A method for instrumentally characterizing the optical properties of coatings containing metallic flakes using multi-angle measurements with a spectrophotometer to derive color constants of the coating, comprising: A three-point multi-angle test using a detector calibrated reflectance technique performed at angles of 15°, 45° and 110° measured from the specular angle at an illumination angle of 45° for a metallic coating characterized by The X, Y _, Z ^tristimulus values _of the coating ^are determined using the following formula: (λ) dλ and Z = ₃₆₀ ∫ ⁸³⁰ R (λ) E (λ) (λ) dλ (In the formula, R is reflection coefficient data,
The data is based on the sensitivity of the human eye, and E
is the illumination from the light source, and λ is 360 ~
830nm, which is the function of wavelength in the visible light spectrum), and brightness (L ^* ), redness/greenness (a ^* ), yellowness/
The perceptual color value of a paint film in terms of blueness (b ^* ) saturation (C) and hue (H) is calculated using the following formula: L ^* = 116 (Y/Yo) ^1/3 -16 a ^* = 500 [(X/Yo) Xo) ^1/3 - (Y/Yo) ^1/3 ] b ^* = 200 [(Y/Yo) ^1/3 - (Z/Zo) ^1/3 ] C = (a ^*2 + b ^*2 ) ^{1/ 2} and H=tan ^-1 (b ^* /a ^* ) (where Xo, Yo and Zo are the tristimulus values of a perfectly white body for constant irradiation, and X, Y and Z
is the tristimulus value for color). 2. 1 in the method of recoloring metallic paint
2. The method of claim 1, wherein the optical properties of the metallic coating are characterized instrumentally in two steps.