Optical filter and imaging device

Abstract

An optical filter (1) comprises a first dielectric multilayer film (10), a near-infrared absorbing glass substrate (20), a resin film (30), and a second dielectric multilayer film (40) in this order, the fracture toughness of the near-infrared absorbing glass substrate (20) being less than 0.3 MPa·m1 / 2, the coefficient of thermal expansion of the near-infrared absorbing glass substrate (20) being 135 × 10-7 / K or more, a crack region extending in a direction along the main surface of the near-infrared absorbing glass substrate (20) and a pair of mirror surface layers with the crack region therebetween being observed on an end surface of the near-infrared absorbing glass substrate (20), the surface roughness Ra of the crack region being 0.5-1.5 µm, the surface roughness Ra of the mirror surface layers being less than 0.5 um, and the width W of the crack region in the plate thickness direction of the near-infrared absorbing glass substrate (20) being 15-40% of the plate thickness T of the near-infrared absorbing glass substrate (20).

Description

Optical filters and imaging devices

[0001] This disclosure relates to optical filters and imaging devices.

[0002] In imaging devices using solid-state image sensors, optical filters are used to transmit visible light (hereinafter also referred to as "visible light") and block near-infrared wavelength light (hereinafter also referred to as "near-infrared light") in order to reproduce colors well and obtain sharp images. Various types of such optical filters can be used, such as reflective filters that use light interference to reflect the light to be blocked by alternately stacking dielectric thin films with different refractive indices on one or both sides of a transparent substrate to form a dielectric multilayer film. In addition, a near-infrared absorbing glass substrate may be used to impart near-infrared absorption characteristics to the transparent substrate, or a resin layer containing a coloring agent may be placed on one or both sides of the near-infrared absorbing glass substrate to complement its near-infrared absorption characteristics.

[0003] In recent years, the demand for thinner solid-state image sensor modules used in mobile devices such as mobile phones and smartphones has led to a need for thinner optical glass sheets that make up optical filters. However, when optical glass sheets become thinner, if bending stress is applied to the optical glass, the likelihood of cracks propagating from chips or minute cracks present at the edges of the glass (the boundary between the main surface and the edge of the glass) increases, potentially leading to breakage and a deterioration in the bending strength of the optical filter. Therefore, higher bending strength is required for the optical glass that makes up optical filters.

[0004] An optical glass is disclosed that has higher bending strength and dimensional accuracy, and has a cut surface cut along a modified region composed of a plurality of modified parts formed by light irradiated to focus inward, wherein the optical glass has a crack originating from the modified part of the modified region, the depth of the modified region tip from the cut surface to the tip of the crack is 3 to 20% of the thickness of the optical glass, and the width of the modified region in the thickness direction is 13 to 50% of the thickness of the optical glass (see, for example, Patent Document 1).

[0005] Patent Document 1: Patent No. 6551404

[0006] However, even when using the optical glass described in Patent Document 1, it is sometimes not possible to obtain an optical filter with sufficient bending strength and dimensional accuracy. Furthermore, when the optical filter has a resin layer, burrs may form on the resin layer containing the coloring agent at the end face of the optical filter, which can lead to a deterioration in the dimensional accuracy of the optical filter. This disclosure has been made in view of the above-mentioned conventional circumstances, and aims to provide an optical filter with excellent bending strength and dimensional accuracy, and an imaging device using the same.

[0007] The specific means for achieving the above objectives are as follows: <1> A first dielectric multilayer film, a near-infrared absorbing glass substrate, a resin film, and a second dielectric multilayer film are provided in this order, wherein the fracture toughness of the near-infrared absorbing glass substrate is 0.3 MPa·m １／２ The coefficient of thermal expansion of the near-infrared absorbing glass substrate is less than 135 × 10 －７An optical filter wherein the temperature is 1 / K or higher, and on the end face of the near-infrared absorbing glass substrate, a crack region extending in a direction along the main surface of the near-infrared absorbing glass substrate and a pair of mirrored layers sandwiching the crack region are observed, the surface roughness Ra of the crack region on the end face is 0.5 to 1.5 μm, the surface roughness Ra of the mirrored layers on the end face is less than 0.5 μm, and the width W of the near-infrared absorbing glass substrate in the thickness direction with respect to the crack region is 15 to 40% of the thickness T of the near-infrared absorbing glass substrate. <2> The optical filter according to <1>, which satisfies the following spectral characteristics (i) to (iii) when the first dielectric multilayer film side is the incident direction side. (i) When the average transmittance in the spectral transmittance curve at an incident angle of 0 degrees is T[450-600(0deg)AVE] for wavelengths of 450-600 nm, and the average transmittance in the spectral transmittance curve at an incident angle of 60 degrees is T[450-600(60deg)AVE] for wavelengths of 450-600 nm, the absolute value of the difference between T[450-600(0deg)AVE] and T[450-600(60deg)AVE] is 15% or less. (ii) In the spectral transmittance curve at an incident angle of 0 degrees, the wavelength IR30(0deg) at which the transmittance is 30% is in the range of 600-700 nm, and in the spectral transmittance curve at an incident angle of 60 degrees, the wavelength IR30(60deg) at which the transmittance is 30% is in the range of 600-700 nm. The optical filter according to <1> or <2>, wherein the absolute value of the difference between IR30 (0 deg) and IR30 (60 deg) is 20 nm or less (iii) In the spectral transmittance curve at an incident angle of 0 degrees, the average transmittance T [750 - 1000 (0 deg) AVE] at wavelengths of 750 to 1000 nm is 1% or less <3> When at least one of the first dielectric multilayer film side and the second dielectric multilayer film side is the incident direction side, in the spectral reflectance curve at an incident angle of 5 degrees, the average reflectance RI [1000 - 1100 (5 deg) AVE] at wavelengths of 1000 to 1100 nm is 15% or less.<4> An optical filter according to any one of <1> to <3>, wherein, when the first dielectric multilayer film side is the incident direction side, the transmittance at a wavelength of 1030 nm in the spectral transmittance curve at an incident angle of 5 degrees is T[1030(5deg)], and the reflectance at a wavelength of 1030 nm in the spectral reflectance curve at an incident angle of 5 degrees is R[1030(5deg)], and the absorption rate at 1030 nm calculated by the following formula (A) is 80% or more. Absorption rate at 1030 nm (%) = 100 - T[1030(5deg)] - R[1030(5deg)] (A) <5> An optical filter according to any one of <1> to <4>, wherein, in the spectral transmittance curve of the near-infrared absorbing glass substrate at an incident angle of 0 degrees, the transmittance at 1030 nm is 10% or less. <6> An optical filter according to any one of <1> to <5>, wherein the crack region is composed of one crack layer, and the surface roughness Ra(1) at the center of the width W in the thickness direction of the crack region at the end face, the surface roughness Ra(2) at a position 1 / 3 × W away from the center of the width W in the thickness direction of the crack region at the end face toward one main surface side of the near-infrared absorbing glass substrate, and the surface roughness Ra(3) at a position 1 / 3 × W away from the center of the width W in the thickness direction of the crack region at the end face toward the other main surface side of the near-infrared absorbing glass substrate, satisfy the following formulas (B) to (C). (B) Ra(1) > Ra(2) (C) Ra(1) > Ra(3) (C) <7> The optical filter according to any one of <1> to <5>, wherein the crack region is composed of two layers: a first crack layer and a second crack layer having a thinner width in the thickness direction of the crack region than the first crack layer, and when the first crack layer is divided into two in the thickness direction of the crack region to form a first layer facing the second crack layer and a second layer not facing the second crack layer, the surface roughness at the center of the width in the thickness direction of the first layer is defined as surface roughness Ra(4), the surface roughness at the center of the width in the thickness direction of the second layer is defined as surface roughness Ra(5), and the surface roughness at the center of the width in the thickness direction of the second crack layer is defined as surface roughness Ra(6), wherein surface roughness Ra(4), surface roughness Ra(5), and surface roughness Ra(6) satisfy the following formulas (D) to (E).Ra(4) > Ra(5) (D) Ra(4) > Ra(6) (E) <8> The optical filter according to any one of <1> to <5>, wherein the crack region is composed of at least three crack layers, and from among the at least three crack layers, crack layer A located closest to the center of the width W in the thickness direction of the crack region is selected, and the surface roughness at the center of the width of crack layer A in the thickness direction is set to surface roughness Ra(7), and the surface roughness at the center of the width of the two crack layers B and crack layer C located on both sides of crack layer A in the thickness direction is set to surface roughness Ra(8) and surface roughness Ra(9), respectively, wherein surface roughness Ra(7), surface roughness Ra(8), and surface roughness Ra(9) satisfy the following formulas (F) to (G). Ra(7)>Ra(8) (F) Ra(7)>Ra(9) (G) <9> The optical filter according to any one of <1> to <8>, wherein a hackle is observed in the mirror layer, extending from the crack region toward one main surface side or the other main surface side of the near-infrared absorbing glass substrate, and the length of the hackle is defined as the distance between a line segment L1 passing through one end of the hackle and parallel to the main surface of the near-infrared absorbing glass substrate and a line segment L2 passing through the other end of the hackle and parallel to the main surface of the near-infrared absorbing glass substrate, and the length of the hackle is 30 μm or less. <10> The near-infrared absorbing glass substrate contains the components P, Al, K, Cu, F and R (R is at least one selected from Li, Na, Rb and Cs), and Al. ３＋ An optical filter according to any one of <1> to <9>, comprising phthalic acid glass having a content of 2 to 20% by mass, and the expected value of the ionic radius of the alkali metal components consisting of K and R being 80 pm or more and less than 133 pm. <11> An imaging device having an optical filter according to any one of <1> to <10>.

[0008] According to this disclosure, an optical filter with excellent bending strength and dimensional accuracy, and an imaging device using the same are provided.

[0009] This is a side view of the optical filter 1. This figure illustrates a method for calculating the width W of the crack region 22. This figure illustrates a method for calculating the width W of the crack region 22 when the crack region 22 is composed of three crack layers 22A, 22B, and 22C. This figure illustrates a method for measuring surface roughness Ra. This figure illustrates a method for measuring surface roughness Ra. This figure illustrates the distribution of surface roughness Ra when the crack region 22 is composed of one crack layer. This figure illustrates the distribution of surface roughness Ra when the crack region 22 is composed of two layers, a first crack layer 22D and a crack layer 22E whose width in the plate thickness direction of the crack region 22 is thinner than that of the first crack layer 22D. This figure illustrates the distribution of surface roughness Ra when the crack region 22 is composed of at least three crack layers. This figure shows the end face of the near-infrared absorbing glass substrate 20 used to explain the length of individual hackles 29 formed on the mirrored layer 24. This figure shows the measurement results of the spectral transmittance curve at an incident angle of 0 degrees and at an incident angle of 60 degrees for the optical filter in Example 1, using a UV-Vis spectrophotometer with the dielectric multilayer film 1 on the side facing the incident direction.

[0010] In this disclosure, numerical ranges indicated using "~" mean a range that includes the numbers before and after "~" as the minimum and maximum values, respectively. In numerical ranges described in stages in this disclosure, the upper or lower limit stated in one numerical range may be replaced with the upper or lower limit of another numerical range described in stages. Also, in numerical ranges described in this disclosure, the upper or lower limit stated in one numerical range may be replaced with the values ​​shown in the examples. In this disclosure, a combination of two or more preferred embodiments is a more preferred embodiment. In this disclosure, the amount of each component means the total amount of multiple substances if there are multiple substances corresponding to each component, unless otherwise specified.

[0011] In this disclosure, "transmittance" means measured transmittance. In this disclosure, for a specific wavelength range, a transmittance of, for example, 90% or more means that the transmittance does not fall below 90% across the entire wavelength range, i.e., the minimum transmittance in that wavelength range is 90% or more. Similarly, for a specific wavelength range, a transmittance of, for example, 1% or less means that the transmittance does not exceed 1% across the entire wavelength range, i.e., the maximum transmittance in that wavelength range is 1% or less. The average transmittance in a specific wavelength range is the arithmetic mean of the transmittances for every 1 nm in that wavelength range. In this disclosure, the average reflectance in a specific wavelength range is the arithmetic mean of the reflectances for every 1 nm in that wavelength range. In this disclosure, optical properties mean values ​​measured using an ultraviolet-visible spectrophotometer. In this disclosure, the angle of incidence means the angle between the incident light and a straight line (normal) perpendicular to the surface of the object being measured.

[0012] <Optical Filter> The optical filter of this disclosure comprises, in this order, a first dielectric multilayer film, a near-infrared absorbing glass substrate, a resin film, and a second dielectric multilayer film, wherein the fracture toughness of the near-infrared absorbing glass substrate is 0.3 MPa·m. １／２ The coefficient of thermal expansion of the near-infrared absorbing glass substrate is less than 135 × 10 －７The temperature is above / K, and at the end face of the near-infrared absorbing glass substrate, a crack region extending in a direction along the main surface of the near-infrared absorbing glass substrate and a pair of mirror-finished layers sandwiching the crack region are observed, the surface roughness Ra of the crack region at the end face is 0.5 to 1.5 μm, the surface roughness Ra of the mirror-finished layers at the end face is less than 0.5 μm, and the width W in the thickness direction of the near-infrared absorbing glass substrate with respect to the crack region is 15 to 40% of the thickness T of the near-infrared absorbing glass substrate. The optical filter of this disclosure has excellent bending strength and dimensional accuracy. The reason for this is not clear, but it is presumed to be as follows: Glass substrates with fracture toughness and thermal expansion coefficient within a predetermined range have the characteristic of being easily fractured by cutting with laser light. On the other hand, the cut end face of a glass substrate that is easily fractured tends to have minute irregularities and cracks called hackles. Cracks that occur on the cut end face of a glass substrate are likely to become the starting point for further cracking of the glass substrate, which can reduce the bending strength of the glass substrate. In this disclosure, it is presumed that by setting the width of the crack region on the end face of the near-infrared absorbing glass substrate to 40% or less of the thickness T of the near-infrared absorbing glass substrate, cracks at the glass end face become less likely to occur, and the bending strength is improved. On the other hand, it is presumed that by setting the width of the crack region on the end face of the near-infrared absorbing glass substrate to 15% or more of the thickness T of the near-infrared absorbing glass substrate, the amount of meandering at the glass end face is reduced, improving the dimensional accuracy of the optical filter, and also making it easier to suppress the generation of resin burrs at the cut portion of the resin layer due to meandering. Here, resin burrs refer to the resin layer protruding from the outer shape of the optical filter when viewed from above.

[0013] The configuration of the optical filter described herein will be explained below with reference to the drawings. Note that the sizes of the components in each drawing are conceptual, and the relative sizes of the components are not limited thereto. Furthermore, components having substantially the same function are given the same reference numerals throughout all drawings, and redundant explanations may be omitted.

[0014] Figure 1 is a side view of an optical filter 1 according to one embodiment of the optical filter of the present disclosure. The optical filter 1 comprises, in this order, a first dielectric multilayer film 10, a near-infrared absorbing glass substrate 20, a resin film 30, and a second dielectric multilayer film 40. On the end face of the near-infrared absorbing glass substrate 20, a crack region 22 extending in a direction along the main surface of the near-infrared absorbing glass substrate 20 and a pair of mirror-finish layers 24 sandwiching the crack region 22 are observed. The crack region 22 is provided spaced apart from both the one main surface 26 and the other main surface 28 of the near-infrared absorbing glass substrate 20, and the mirror-finish layers 24 are observed between the crack region 22 and the one main surface 26, and between the crack region 22 and the other main surface 28.

[0015] The crack region 22 observed on the end face of the near-infrared absorbing glass substrate 20 is a region that occurs when the optical filter is fragmented using laser light during optical filter manufacturing. The end face of the optical filter 1 corresponds to the cut surface formed by irradiating laser light through an optical system equipped with an optical lens adjusted so that the focal point aligns with the interior of the near-infrared absorbing glass substrate 20 constituting the optical filter 1, forming a crack region 22 that serves as the starting point for cutting in the thickness direction of the near-infrared absorbing glass substrate 20 along the planned cutting line, and extending the crack region 22. Laser light irradiation conditions include the wavelength of the laser light, pulse width, repetition frequency, irradiation time, energy intensity, and number of scans. These irradiation conditions are set appropriately in consideration of the thickness of the near-infrared absorbing glass substrate 20, the type of glass constituting the near-infrared absorbing glass substrate 20, etc. The planned cutting line is typically a grid-like scanning line such that the planar shape of the optical filter 1 becomes square or rectangular.

[0016] The separation between the crack region 22 and the mirror layer 24 can be determined based on an image of the end face of the near-infrared absorbing glass substrate 20. The image of the end face of the near-infrared absorbing glass substrate 20 can be taken by an optical microscope. FIG. 2 is a diagram for explaining a method of calculating the width W of the crack region 22, and is a diagram showing the near-infrared absorbing glass substrate 20 constituting the optical filter 1 in FIG. 1 extracted. At the boundary between the crack region 22 and the mirror layer 24 (mirror layer 24a) on one main surface 26 side, as shown in FIG. 2, a plurality of convex portions M (convex portion M １ , convex portion M ２ ... convex portion M ｎ－１ , convex portion M ｎ ) protruding from the crack region 22 side toward one main surface 26 side of the near-infrared absorbing glass substrate 20 are observed. From one main surface 26 to each vertex of the individual convex portions M １ , convex portion M ２ ... convex portion M ｎ－１ , convex portion M ｎ , a perpendicular line orthogonal to one main surface 26 is drawn, and the distance a １ , distance a ２ ... distance a ｎ－１ , distance a ｎ between one main surface 26 and each vertex of the individual convex portions M １ , distance a ２ ... distance a ｎ－１ , distance a ｎ is obtained. The average value (distance a) of the obtained distances a １ , distance a ２ ... distance a ｎ－１ , distance a ｎ is taken as the width a of the mirror layer 24a. Also, a straight line parallel to one main surface 26 and separated from one main surface 26 by a distance a is taken as the boundary line A between the crack region 22 and the mirror layer 24a. Further, at the boundary between the crack region 22 and the mirror layer 24 (mirror layer 24b) on the other main surface 28 side, as shown in FIG. 2, a plurality of convex portions N (convex portion N １ , convex portion N ２ ... convex portion N ｎ－１ , convex portion N ｎ ) protruding from the crack region 22 side toward the other main surface 28 side of the near-infrared absorbing glass substrate 20 are observed. For the plurality of convex portions N, in the same manner as the convex portion M, the other main surface 28 and the individual convex portions N １ , convex portion N ２ ... convex portion Nｎ－１ , protruding part N ｎ distance b from each vertex １ , distance b ２ ...distance b ｎ－１ , distance b ｎ We find the distance b obtained １ , distance b ２ ...distance b ｎ－１ , distance b ｎ The average value (distance b) is taken as the width b of the mirrored layer 24b. Furthermore, a straight line parallel to the other main surface 28 and at a distance b from the other main surface 28 is taken as the boundary line B between the crack region 22 and the mirrored layer 24b. The width W of the crack region 22 is calculated by subtracting the sum of widths a and b from the plate thickness T of the near-infrared absorbing glass substrate 20.

[0017] Figure 2 shows the case where the crack region 22 is composed of one crack layer, but the crack region 22 may be composed of two or more crack layers. Figure 3 is a diagram illustrating the method for calculating the width W of the crack region 22 when the crack region 22 is composed of three crack layers 22A, 22B, and 22C. When the crack region 22 exists as multiple crack layers on the end face of the near-infrared absorbing glass substrate 20 as shown in Figure 3, the distance a is calculated at the boundary between the crack layer 22A closest to one main surface 26 of the near-infrared absorbing glass substrate 20 and the mirrored layer 24a using the method described above. １ distance a ２ ...distance a ｎ－１ distance a ｎ The average value (distance a) is calculated and set as the width a of the mirrored layer 24a. Furthermore, a straight line parallel to one main surface 26 and at a distance a from that main surface 26 is defined as the boundary line A between the crack region 22 and the mirrored layer 24a. Similarly, at the boundary between the cracked layer 22C closest to the other main surface 28 of the near-infrared absorbing glass substrate 20 and the mirrored layer 24b, the distance b is calculated. １ , distance b ２ ...distance b ｎ－１ , distance b ｎThe average value (distance b) is calculated and set as the width b of the mirrored layer 24b. A straight line parallel to the other main surface 28 and at a distance b from the other main surface 28 is set as the boundary line B between the crack region 22 and the mirrored layer 24b. The width W of the crack region 22 is calculated by subtracting the sum of widths a and b from the plate thickness T of the near-infrared absorbing glass substrate 20.

[0018] In the optical filter 1, the width W of the crack region 22 in the thickness direction is 15 to 40% of the thickness T of the near-infrared absorbing glass substrate 20, preferably 15 to 35%, and more preferably 20 to 30%. The width of the mirror layer 24 (i.e., width a and width b) is not particularly limited. Width a and width b may be the same or different. The sum of width a and width b is 60 to 85% of the thickness T.

[0019] The surface roughness Ra of the crack region 22 at the end face of the near-infrared absorbing glass substrate 20 is set to 0.5 to 1.5 μm. Preferably, the surface roughness Ra of the crack region 22 is 0.55 to 1 μm, and more preferably 0.6 to 0.8 μm. Furthermore, the surface roughness Ra of the mirror layer 24 at the end face of the near-infrared absorbing glass substrate 20 is set to less than 0.5 μm. Preferably, the surface roughness Ra of the mirror layer 24 is 0.4 μm or less, and more preferably 0.35 μm or less. The surface roughness Ra of the mirror layer 24 may be 0 μm or more. The optical filter 1 comprising the near-infrared absorbing glass substrate 20 in which the surface roughness Ra of the crack region 22 and the surface roughness Ra of the mirror layer 24 are within the above ranges suggests that it is fragmented by irradiation with laser light.

[0020] The surface roughness Ra of the crack region 22 and the mirror layer 24 at the end face of the near-infrared absorbing glass substrate 20 refers to the value measured using a laser microscope by the method described below. For example, a Keyence VKX-3000 laser microscope can be used as the laser microscope.

[0021] When the crack region 22 is composed of a single crack layer, the surface roughness Ra of the crack region 22 is measured using a laser microscope as follows. Figure 4 is a diagram illustrating the method for measuring the surface roughness Ra, and shows the near-infrared absorbing glass substrate 20 that constitutes the optical filter 1 in Figure 1. On the end face of the near-infrared absorbing glass substrate 20, a measurement area P is set with a scan width in the range of 10 to 200 μm and in the direction along the main surface of the near-infrared absorbing glass substrate 20, so that the entire crack region 22 is covered in the thickness direction of the near-infrared absorbing glass substrate 20. The pitch width in the height direction of the end face of the near-infrared absorbing glass substrate 20 is set to 0.1 nm. The pitch width in the height direction is the same for the following. Height information in the measurement area P is acquired using a laser microscope. Based on the acquired height information, the center W of the width W in the crack region 22 is determined. Ｃ A straight line Q is set that passes through the measurement region P and has a length of 10 to 100% of the length in the direction along the main surface of the near-infrared absorbing glass substrate 20 in the measurement region P where the high and low information has been acquired. The average line roughness on the straight line Q is acquired, and the obtained value is taken as the surface roughness Ra in the crack region 22. In Figure 4, the straight line Q is 100% of the length in the direction along the main surface of the near-infrared absorbing glass substrate 20 in the measurement region P.

[0022] When the crack region 22 is composed of two or more crack layers, the surface roughness Ra of the crack region 22 is measured using a laser microscope as follows. Figure 5 is a diagram illustrating the method for measuring the surface roughness Ra, and shows the near-infrared absorbing glass substrate 20 that constitutes the optical filter 1 in Figure 1. In Figure 5, the crack region 22 is composed of three crack layers 22A to 22C. The width W of the crack region 22 is determined by the method described above, and the center W of the width W is determined. Ｃ To identify the widths WA, WB, and WC of each crack layer 22A to 22C, the same method as the calculation method for the width W of the crack region 22 based on Figure 2 is used, and the center WA of the widths WA, WB, and WC of each crack layer 22A to 22C is determined. Ｃ WB Ｃ and WC ＣEach of these is identified. The center W of the width W of the crack region 22 among the crack layers 22A to 22C. Ｃ Identify a crack layer having the center closest to the center W of the crack region 22. In Figure 5, the center W of the width W of the crack region 22 is identified. Ｃ The crack layer with the center closest to center WB Ｃ This is a crack layer 22B having the following characteristics. The scan width is set to a range of 10 to 200 μm and the measurement area P is set to a range of 1 mm or more in the direction along the main surface of the near-infrared absorbing glass substrate 20 so that the crack layer (crack layer 22B) identified in the thickness direction of the near-infrared absorbing glass substrate 20 is completely covered, and the same procedure is followed for the center WB Ｃ A straight line Q passing through the crack region is set, the average line roughness along the straight line Q is obtained, and the obtained value is defined as the surface roughness Ra in the crack region 22.

[0023] The surface roughness Ra of the mirrored layer 24 is measured using a laser microscope as follows. In Figure 4, a measurement area R is set on the end face of the near-infrared absorbing glass substrate 20, with a scan width in the range of 10 to 200 μm, and in the direction along the main surface of the near-infrared absorbing glass substrate 20, so that the entire mirrored layer 24a is covered in the thickness direction of the near-infrared absorbing glass substrate 20. Height information is acquired in the measurement area R using a laser microscope. Based on the acquired height information, the average line roughness is obtained on a straight line S that passes through the center of the mirrored layer 24a in the thickness direction and has a length of 10 to 100% of the length in the direction along the main surface of the near-infrared absorbing glass substrate 20 in the measurement area R from which the height information was acquired. Similarly, the average line roughness on the straight line S of the mirrored layer 24b is obtained. The average of the average line roughness on the straight line S for the pair of mirrored layers 24a and 24b is taken as the surface roughness Ra of the mirrored layer 24.

[0024] The surface roughness Ra in the crack region 22 may be uniform in the thickness direction or may have a distribution in the thickness direction. When the surface roughness Ra in the crack region 22 has a distribution in the thickness direction, it is preferable from the viewpoint of suppressing the length of the hackle that the surface roughness Ra at the center of the width W in the thickness direction of the crack region is higher than the surface roughness Ra at other locations.

[0025] The following describes, based on the drawings, the preferred configurations for cases where the crack region 22 is composed of one crack layer, two crack layers, and three or more crack layers, in which the surface roughness Ra has a distribution in the thickness direction of the plate.

[0026] Figure 6 illustrates the distribution of surface roughness Ra when the crack region 22 is composed of a single crack layer. Note that in Figure 6, the protrusions M and N at the boundary between the crack region 22 and the mirror-finish layer 24 are omitted. When the crack region 22 is composed of a single crack layer, the center W of the width W in the thickness direction of the crack region 22 at the end face of the near-infrared absorbing glass substrate 20 Ｃ The surface roughness Ra(1) and the center W of the width W in the thickness direction of the crack region 22 at the end face of the near-infrared absorbing glass substrate 20. Ｃ The surface roughness Ra(2) at a position 1 / 3 × W away in the thickness direction toward one main surface 26 of the near-infrared absorbing glass substrate 20, and the center W of the width W in the thickness direction of the crack region 22 at the end face of the near-infrared absorbing glass substrate 20. Ｃ It is preferable that the surface roughness Ra(3) at a position 1 / 3 × W away from the other main surface 28 of the near-infrared absorbing glass substrate 20 in the thickness direction satisfies the following equations (B) to (C). The relationship between surface roughness Ra(2) and surface roughness Ra(3) is not particularly limited; surface roughness Ra(2) and surface roughness Ra(3) may be the same, or surface roughness Ra(2) and surface roughness Ra(3) may be different. Ra(1) > Ra(2) (B) Ra(1) > Ra(3) (C)

[0027] Here, the surface roughness Ra(1) is center W Ｃ The straight line Q shown in Figure 6 passes through and is parallel to the direction along the main surface of the near-infrared absorbing glass substrate 20. １ This refers to the average line roughness shown above. Also, the surface roughness Ra(2) is calculated using the center W. Ｃ A straight line Q shown in Figure 6 passes through a position 1 / 3 × W away in the thickness direction toward one main surface 26 side of the near-infrared absorbing glass substrate 20, and is parallel to the direction along the main surface of the near-infrared absorbing glass substrate 20. ２The above refers to the average line roughness. Also, the surface roughness Ra(3) is centered W Ｃ A straight line Q shown in Figure 6 passes through a position 1 / 3 × W away in the thickness direction toward the other main surface 28 side of the near-infrared absorbing glass substrate 20, and is parallel to the direction along the main surface of the near-infrared absorbing glass substrate 20. ３ This refers to the roughness of the average line shown above. (Line Q) １ Top, straight line Q ２ Upper and straight line Q ３ The average line roughness can be determined by the same method as the measurement of the surface roughness Ra of the crack region 22. The surface roughness Ra(1) is preferably 0.5 to 1.5 μm, more preferably 0.55 to 1 μm, and even more preferably 0.6 to 0.8 μm.

[0028] Figure 7 is a diagram illustrating the distribution of surface roughness Ra when the crack region 22 is composed of two layers: a first crack layer 22D and a second crack layer 22E which is thinner in the thickness direction of the crack region 22 than the first crack layer 22D. Note that in Figure 7, the convex portions M and N present at the boundary between the crack region 22 and the mirror-finish layer 24 are omitted. When the first crack layer 22D, which has a width WD, is divided into two in the thickness direction of the crack region 22 to form a first layer 22D1 facing the second crack layer 22E and a second layer 22D2 not facing the second crack layer 22E, the center WD1 of the width WD1 in the thickness direction of the first layer 22D1 is shown. Ｃ The surface roughness in is defined as surface roughness Ra(4), and the center WD2 of the width WD2 in the thickness direction of the second layer 22D2 is defined as WD2 Ｃ The surface roughness in is defined as surface roughness Ra(5), and the center WE of the width WE in the thickness direction of the second crack layer 22E is Ｃ When the surface roughness in is defined as surface roughness Ra(6), it is preferable that surface roughness Ra(4), surface roughness Ra(5), and surface roughness Ra(6) satisfy the following equations (D) to (E). The relationship between surface roughness Ra(5) and surface roughness Ra(6) is not particularly limited; surface roughness Ra(5) and surface roughness Ra(6) may be the same, or they may be different. Ra(4) > Ra(5) (D) Ra(4) > Ra(6) (E)

[0029] Here, the surface roughness Ra(4) is center WD1Ｃ The straight line Q shown in Figure 7 passes through and is parallel to the direction along the main surface of the near-infrared absorbing glass substrate 20. ４ This refers to the average line roughness shown above. Also, the surface roughness Ra(5) is based on the central WD2 Ｃ The straight line Q shown in Figure 7 passes through and is parallel to the direction along the main surface of the near-infrared absorbing glass substrate 20. ５ This refers to the average line roughness shown above. Also, surface roughness Ra(6) is measured at the center WE. Ｃ The straight line Q shown in Figure 7 passes through and is parallel to the direction along the main surface of the near-infrared absorbing glass substrate 20. ６ This refers to the roughness of the average line shown above. (Line Q) ４ Top, straight line Q ５ Upper and straight line Q ６ The average line roughness can be determined by the same method as the measurement of the surface roughness Ra of the crack region 22. The surface roughness Ra (4) is preferably 0.5 to 1.5 μm, more preferably 0.55 to 1 μm, and even more preferably 0.6 to 0.8 μm.

[0030] Figure 8 illustrates the distribution of surface roughness Ra when the crack region 22 is composed of at least three crack layers. Figure 8 explains the case where the crack region 22 is composed of three crack layers (crack layer 22F, crack layer 22G, and crack layer 22H) as an example. Note that in Figure 8, the convex portions M and N at the boundary between the crack region 22 and the mirror-finish layer 24 are omitted. Of the three crack layers (crack layer 22F, crack layer 22G, and crack layer 22H), the center W of the width W in the plate thickness direction of the crack region 22 is... Ｃ The center WF of the width WF in the plate thickness direction of the crack layer 22F, which corresponds to the crack layer A located closest to it. Ｃ The surface roughness in this region is defined as surface roughness Ra(7), and for the crack layer 22G corresponding to crack layer B and the crack layer 22H corresponding to crack layer C located on both sides of the crack layer 22F in the thickness direction, the width WG and the center WG of the width WH in the thickness direction are defined as follows: Ｃ and center WH ＣWhen the surface roughnesses are surface roughness Ra(8) and surface roughness Ra(9) respectively, it is preferable that surface roughness Ra(7), surface roughness Ra(8), and surface roughness Ra(9) satisfy the following formula (F) to the following formula (G). The relationship between surface roughness Ra(8) and surface roughness Ra(9) is not particularly limited, and surface roughness Ra(8) and surface roughness Ra(9) may be the same, or surface roughness Ra(8) and surface roughness Ra(9) may be different. Ra(7) > Ra(8) (F) Ra(7) > Ra(9) (G)

[0031] Here, surface roughness Ra(7) is the average line roughness on a straight line Q shown in FIG. 8 that passes through the center WF Ｃ and is parallel to the direction along the main surface of the near-infrared absorption glass substrate 20. ７ Also, surface roughness Ra(8) is the average line roughness on a straight line Q shown in FIG. 8 that passes through the center WG Ｃ and is parallel to the direction along the main surface of the near-infrared absorption glass substrate 20. ８ Also, surface roughness Ra(9) is the average line roughness on a straight line Q shown in FIG. 8 that passes through the center WH Ｃ and is parallel to the direction along the main surface of the near-infrared absorption glass substrate 20. ９ The average line roughness on the straight line Q ７ on the straight line Q ８ and on the straight line Q ９ can be obtained by the same method as the measurement of the surface roughness Ra of the crack region 22. Surface roughness Ra(7) is preferably 0.5 to 1.5 μm, more preferably 0.55 to 1 μm, and even more preferably 0.6 to 0.8 μm.

[0032] When separating the optical filter 1 into individual pieces, the laser light irradiation conditions may involve one or more laser light scans. If the number of scans is two or more, the focal point of the laser light may be shifted in the thickness direction of the near-infrared absorbing glass substrate 20. When the focal point is shifted in the thickness direction, multiple crack layers may be formed individually along the direction of the main surface of the near-infrared absorbing glass substrate 20 with each laser light scan, and a crack region 22 may be formed by multiple crack layers, or the crack layers may be integrated to form a crack region 22. For example, by shifting the focal point in the thickness direction and scanning the laser light three times, and sequentially changing the intensity of the laser light from one main surface 26 side of the near-infrared absorbing glass substrate 20 in the order of weak-strong-weak with each scan, the surface roughness Ra(1) to Ra(9) are more likely to satisfy equations (B) to (G).

[0033] In the mirror-finish layer 24, hackles (streaky patterns occurring in the mirror-finish layer 24) may be observed, extending from the crack region 22 toward one main surface 26 or the other main surface 28 of the near-infrared absorbing glass substrate 20. Since cracks in the optical filter are more likely to occur starting from the hackles, it is preferable that the hackles be short. Figure 9 shows the end face of the near-infrared absorbing glass substrate 20 used to explain the length of individual hackles 29 that have formed in the mirror-finish layer 24. The length H of the hackle 29 is defined as the distance between a line segment L1 passing through one end 29a of the hackle 29 and parallel to the main surface of the near-infrared absorbing glass substrate 20, and a line segment L2 passing through the other end 29b of the hackle 29 and parallel to the main surface of the near-infrared absorbing glass substrate 20. The length of each individual hackle is determined based on an optical microscope photograph of the end face of the optical filter 1. The arithmetic mean of the lengths of 20 individual hackles is defined as the length of the hackle in this disclosure. The length of the hackle is preferably 30 μm or less, more preferably 20 μm or less, and even more preferably 10 μm or less. Shorter hackle is preferable.

[0034] From the perspective of thinning the optical filter 1, the thickness T of the near-infrared absorption glass substrate 20 is preferably 0.5 mm or less, more preferably 0.3 mm or less. From the perspective of the strength of the optical filter 1, the thickness T of the near-infrared absorption glass substrate 20 is preferably 0.10 mm or more, more preferably 0.13 mm or more, and particularly preferably 0.15 mm or more.

[0035] The fracture toughness of the near-infrared absorption glass substrate 20 is less than 0.3 MPa·m １／２ and the coefficient of thermal expansion is 135×10 －７ / K or more, which is preferable from the perspectives of bending strength and dimensional accuracy. The fracture toughness is 0.27 MPa·m １／２ or less, and the coefficient of thermal expansion is 140×10 －７ / K or more, which is more preferable. The fracture toughness is 0.25 MPa·m １／２ or less, and the coefficient of thermal expansion is 150×10 －７ / K or more, which is even more preferable. The fracture toughness of the near-infrared absorption glass substrate 20 may be 0.1 MPa·m １／２ or more, and the coefficient of thermal expansion may be 200×10 －７ / K or less.

[0036] The fracture toughness of the glass substrate is the value (K1c) calculated by the following formula in the fracture toughness measurement method (IF method) defined in JIS R1607:2015. The measurement of the fracture toughness of the glass substrate is performed using a Vickers hardness tester (for example, ARS9000F manufactured by Future Tech Co., Ltd., and analysis software: FT-026) under environmental conditions where the room temperature is 23°C and the humidity is approximately 30%. In this measurement, cracks extend from the indentation formed by the indenter and grow over time. Therefore, within 30 seconds after separating the indenter from the glass substrate, the crack length is measured.

[0037] K1c = 0.026·E １／２ ·P １／２ ·a·C －３／２ [[ID=三十一]] In the above formula, E is the Young's modulus, P is the indentation load, a is 1 / 2 of the average of the indentation diagonal length, and C is 1 / 2 of the average of the crack length.

[0038] The thermal expansion coefficient of the glass substrate was measured using the differential formula specified in JIS R3102:1995, and is the average value of measurements taken between 50 and 300°C.

[0039] In the spectral transmittance curve of the near-infrared absorbing glass substrate 20 at an incident angle of 0 degrees, the transmittance at 1030 nm is preferably 10% or less from the viewpoint of improving the light-shielding performance of the optical filter in the infrared region.

[0040] The composition of the near-infrared absorbing glass substrate 20 is not particularly limited as long as it has excellent absorption characteristics in the near-infrared region, and examples include phthalate glass and phosphate glass containing copper ions. Among these, phthalate glass containing copper ions is preferred from the viewpoint of obtaining good near-infrared absorption characteristics while maintaining the weather resistance of the optical filter.

[0041] The phthalic acid glass constituting the near-infrared absorbing glass substrate 20 (hereinafter also referred to as the phthalic acid glass of this disclosure) contains the following components: P, Al, K, Cu, F, and R (R is at least one selected from Li, Na, Rb, and Cs), and Al ３＋ Preferably, the content is 2 to 20% by mass, and the expected value of the ionic radius of the alkali metal component consisting of K and R is 80 pm (picometers) or more and less than 133 pm (picometers).

[0042] The phthalic acid glass of this disclosure is a copper phthalic acid glass containing P, Al, K, Cu, F, and R (where R is at least one selected from Li, Na, Rb, and Cs). Glass with P as the main component has the effect of enhancing absorption capacity and sharp cut properties in the near-infrared region. Furthermore, the inclusion of F and Al in the glass can improve weather resistance. In addition, by adjusting the expected value of the ionic radius of the alkali metal components consisting of K and R to 80 pm or more and less than 133 pm, the absorption capacity and sharp cut properties in the near-infrared region can be improved, and the inclusion of K and R can improve weather resistance.

[0043] Alkali metal components are, K ＋ Li ＋ Na ＋ , Rb＋ and Cs ＋ The expected values ​​of the ionic radii of each component are defined as follows. The ionic radii of each alkali metal component are as follows: Li ＋ Ionic radius r Ｌｉ It is 60 pm, Na ＋ Ionic radius r Ｎａ It was 95 p.m., K ＋ Ionic radius r Ｋ 133 pm, Rb ＋ Ionic radius r Ｒｂ It is 148 p.m., Cs ＋ Ionic radius r Ｃｓ It is 169 pm. These ionic radii are based on the literature of L. Pauling (1931-1933), THE NATURE OF THE CHEMICAL BOND (translated by Masao Koizumi in 1963, Kyoritsu Shuppan). "Cation %" is a unit that expresses the content of each cation component in moles, when the total content of all cation components contained in the glass is set to 100 mol%. The expected value of the ionic radius of alkali metal components is calculated by the following formula.

[0044] The expected value of the ionic radius = [Ionic radius of each alkali metal component × sum of cation amounts of each component] / [sum of cation amounts of all alkali metal components]. Specifically, it is as follows:

[0045] [Ionic radius of each alkali metal component × sum of cation amounts of each component] = Ionic radius r Ｌｉ × Li cation amount + ionic radius r Ｎａ × Amount of Na cation + Ionic radius r Ｋ × Amount of cations in K + Ionic radius r Ｒｂ × Amount of cation in Rb + Ionic radius r Ｃｓ × Amount of Cs cations, [Total amount of cations of all alkali metal components] = Amount of Li cations + Amount of Na cations + Amount of K cations + Amount of Rb cations + Amount of Cs cations

[0046] The phthalic acid glass disclosed herein contains alkali metal components (Li ＋ Na ＋ _K ＋ , Rb＋ , Cs ＋ By setting the expected value of the ionic radius of ) to 80 pm or more, it is possible to maintain a sharp absorption shape with enhanced absorption in the near-infrared region while keeping the transmittance in the red region high. The reason for this is presumed to be as follows: Cu in glass ２＋ Non-bridged oxygen molecules coordinate to form a regular octahedron. This Cu ２＋ If the symmetry of the non-crosslinked oxygen coordinating to it is high, it will have a sharp absorption peak in the near-infrared region, but for reasons described later, if the symmetry of these non-crosslinked oxygen atoms decreases, Cu ２＋ The absorption peak shifts, and the shape of the glass transmittance curve changes from a sharp absorption shape to a broad absorption shape. It has been reported in "Kohei Kadono (2009), 'Optical Properties of Glass II.' NEW GLASS Vol.24 No.2" that the absorption spectrum of transition metals containing Cu is easily altered by changes in the coordination environment in glass. ２＋ The non-crosslinked oxygen that coordinates to the element is attracted to the surrounding electronegative components, causing a decrease in symmetry. Electronegativity is a property that represents the strength of the force with which the nucleus of an atom attracts surrounding electrons. Ionic radius is a value that indicates the distance from the nucleus to the outermost electron shell of an atom. Among atoms of the same group, the greater the distance between the nucleus and the bonding electron pair, the lower the electronegativity, so it can be said that components with a large ionic radius have low electronegativity. Therefore, by including components with a large ionic radius among alkali metal components in the glass, Cu ２＋ The symmetry of the non-crosslinked oxygen that coordinates to it is not reduced, and high absorption capacity and high sharpness in the near-infrared region can be achieved.

[0047] On the other hand, alkali metal components (Li ＋ Na ＋ _K ＋ , Rb ＋ , Cs ＋Setting the expected value of the ionic radius of ) to 133 pm or higher raises concerns about a decrease in weather resistance. The reason for this is presumed to be as follows: Weather resistance is evaluated by the degree of deterioration of the glass surface caused when the glass is left standing for a long time under high temperature and high humidity conditions. Under high temperature and high humidity conditions, H present on the glass surface ＋ The hydrogen penetrates the interior of the glass, attacks the -O-P-O- structure, and causes hydrolysis. As a result, H is released from the glass surface. ３ PO ４ The H2O2 remains in a liquid state and reacts with the glass, causing foreign matter to precipitate and degrade the glass surface. When a large amount of alkali metal components with a large ionic radius are included, the force that attracts non-crosslinked oxygen in the glass weakens, reducing the strength of the glass structure. Therefore, when glass is left standing for a long time under high temperature and high humidity conditions, the H2O2 present on the glass surface deteriorates. ＋ If the alkali metal component penetrates more easily into the glass, that is, the hydrolysis reaction is accelerated, and the weather resistance of the glass decreases. For the reasons above, it is desirable that the expected value of the ionic radius of the alkali metal component be 80 pm or more and less than 133 pm. If it is 80 pm or more, sufficient effects of high absorption capacity in the near-infrared region and improved sharpness can be obtained, and if it is less than 133 pm, problems such as a decrease in weather resistance are less likely to occur. For this reason, it is more preferably 85 pm or more, even more preferably 90 pm or more, even more preferably 95 pm or more, most preferably 100 pm or more, and more preferably 130 pm or less, even more preferably 128 pm or less, even more preferably 124 pm or less, and most preferably 120 pm or less.

[0048] The phthalic acid glass of this disclosure contains alkali metal components consisting of K and Rb. By including two or more alkali metal components in the glass, weather resistance can be improved. The reason for this is presumed to be as follows: Glass with low weather resistance deteriorates on the glass surface under high temperature and high humidity, and precipitates and liquid leaching are observed. The cause of this is thought to be alkali metal components and H ＋ One example is the ion exchange reaction. When glass is left standing for a long time under high temperature and high humidity conditions, H present on the glass surface... ＋An ion exchange reaction occurs between the alkali metal ions on the glass surface and H. Specifically, the alkali metal components dissolve to the glass surface through the ion exchange reaction, and H ＋ H can easily penetrate inside the glass. ＋ The effect of this on glass is as described above. Because alkali metal components have a larger ion diffusion coefficient compared to other components, their ion mobility is high, and H ＋ This facilitates ion exchange reactions, reducing the weather resistance of the glass.

[0049] It is known that by combining two or more alkali metal components and incorporating them into glass, the ion mobility of each alkali metal component is reduced due to the mixed alkali effect. This effect reduces the H present on the glass surface. ＋ This suppresses the ion exchange reaction with alkali metal components, thereby preventing a decrease in weather resistance.

[0050] The components that can constitute the phthalic acid glass of this disclosure and their preferred content are described below.

[0051] In the phthalic acid glass of this disclosure, P (phosphorus) is P ５＋ It is contained as P. ５＋ P is the most abundant cationic component in phthalic acid glass and is also a component that enhances sharpness in the near-infrared region. ５＋ The content is preferably 20 to 70% by mass. ５＋ If the content is 20% by mass or more, the effect can be sufficiently obtained, and if it is 70% by mass or less, problems such as glass instability or reduced weather resistance are less likely to occur. For this reason, it is more preferably 25% by mass or more, even more preferably 30% by mass or more, even more preferably 33% by mass or more, even more preferably 60% by mass or less, even more preferably 55% by mass or less, even more preferably 50% by mass or less, and most preferably 45% by mass or less. Note that P ５＋ From the viewpoint of suppressing crucible erosion and inhibiting the volatilization of components, the use of phosphoric acid or a salt thereof is preferred as the raw material.

[0052] In the fluorine glass of this disclosure, F (fluorine) is F － It is contained as F. － This is a component used to stabilize the glass and improve its weather resistance. In this disclosure, F is contained in the glass. － When all other constituent elements are considered to be 100% by mass, the amount of F contained in the glass － The content of F is shown as an external percentage. － The content is preferably 3 to 60% by mass when calculated externally. － If the content is 3% by mass or more when divided by external weight, sufficient weather resistance can be obtained, and if it is 60% by mass or less when divided by external weight, problems such as a decrease in the transmittance of light in the visible region, a decrease in the absorption capacity of light in the near-infrared region, sharp cut performance, a decrease in mechanical properties such as strength, hardness, and elastic modulus, and an increase in ultraviolet transmittance are less likely to occur. More preferably, it is 4% by mass or more when divided by external weight, even more preferably 6% by mass or more when divided by external weight, even more preferably 8% by mass or more when divided by external weight, most preferably 10% by mass or more when divided by external weight, and more preferably 50% by mass or less when divided by external weight, even more preferably 40% by mass or less when divided by external weight, even more preferably 30% by mass or less when divided by external weight, and most preferably 20% by mass or less when divided by external weight.

[0053] In the phthalic acid glass of this disclosure, Cu (copper) is Cu ＋ or Cu ２＋ Although it is contained as, in this disclosure all Cu ２＋ The amount present is listed if it were to exist as such. Cu ２＋ This is a component for improving absorption in the near-infrared region. Also, Cu ２＋ Because it has the property of attracting phosphate chains in glass to form a cross-linked structure, the glass structure is strengthened, improving weather resistance and glass strength. ２＋The content is preferably 1 to 20% by mass. If it is 1% by mass or more, the near-infrared absorption capacity of the glass is ensured. Preferably it is 2% by mass or more, more preferably 3% by mass or more, even more preferably 4% by mass or more, and even more preferably 5% by mass or more. Also, if it is 20% by mass or less, the glass is less likely to become unstable and the risk of devitrification is reduced. Preferably it is 18% by mass or less, more preferably 16% by mass or less, even more preferably 14% by mass or less, and even more preferably 12% by mass or less.

[0054] Furthermore, the total amount of Cu is expressed as the sum of Cu in mass percent, including monovalent, divalent, and other existing valencies, and represents the total content of all components of the glass disclosed herein (however, F － When the content of (excluding) is taken as 100% by mass, the range of total Cu content in the glass is preferably 1 to 20% by mass. If the total Cu content is 1% by mass or more, the effect of absorption capacity in the near-infrared region can be sufficiently obtained, and if it is 20% by mass or less, the decrease in visible region transmittance can be suppressed. ＋ The percentage of the content is (Cu ＋ A range of 0.01 to 4.0 mass% is preferred for (total Cu content) × 100 [mass%].

[0055] In the phthalic acid glass of this disclosure, Al (aluminum) is Al ３＋ It is contained as Al ３＋ Al is a component that forms glass and is used to increase the strength and weather resistance of glass. ３＋ If the content is 2% by mass or more, the effect can be obtained sufficiently, and if it is 20% by mass or less, problems such as glass instability or a decrease in near-infrared absorption ability and sharp cut performance are less likely to occur. ３＋ The content is preferably 2 to 20% by mass. ３＋ The content of is more preferably 3% by mass or more, even more preferably 3.5% by mass or more, even more preferably 4% by mass or more, most preferably 5% by mass or more, and more preferably 19% by mass or less, even more preferably 18% by mass or less, even more preferably 15% by mass or less, and most preferably 13% by mass or less.

[0056] Al ３＋ As a raw material, AlF ３ Al ２ O ３ Al(OH) ３ These can be used, and among them, the increase in dissolution temperature and the generation of unmelted material, and F － AlF ３ It is preferable to use [this].

[0057] Li (lithium) is a component used to lower the melting temperature of glass, lower the liquidus temperature of glass, improve the weather resistance of glass, and stabilize glass. ＋ The content is preferably 0 to 30% by mass. Li ＋ If the content is 30% by mass or less, the glass is less likely to become unstable. Since the presence of Li reduces the absorption capacity in the near-infrared region and sharp cut properties, it is more preferably 28% by mass or less, even more preferably 25% by mass or less, even more preferably 20% by mass or less, and most preferably 10% by mass or less. ＋ If only Li is used, weather resistance will improve, but absorption capacity and sharpness in the near-infrared region will decrease. ＋ It is necessary to include one or more alkali metal components with a larger ionic radius.

[0058] Sodium (Na) is a component that lowers the melting temperature of glass, lowers the liquidus temperature of glass, and stabilizes glass. ＋ The content is preferably 0 to 40% by mass. ＋ If the content is 40% by mass or less, the glass is less likely to become unstable. More preferably 30% by mass or less, even more preferably 25% by mass or less, even more preferably 20% by mass or less, and most preferably 10% by mass or less. The alkali metal component is Na ＋If only Na is used, either improved weather resistance or improved absorption capacity and sharpness in the near-infrared region can be obtained, and the properties improved will differ depending on the composition system. However, it is difficult to improve both properties at the same time. Therefore, to improve weather resistance, Na is used. ＋ It must contain one or more alkali metal components other than Na, and to improve absorption and sharpness in the near-infrared region. ＋ It is necessary to include alkali metal components with larger ionic radii.

[0059] In the phthalic acid glass of this disclosure, K (potassium) is K ＋ It is contained as K. ＋ This component has effects such as lowering the melting temperature of glass, lowering the liquidus temperature of glass, and improving the absorption capacity and sharpness in the near-infrared region. ＋ The content is preferably 1 to 40% by mass. ＋ A content of 40% by mass or less is preferable because it makes the glass less likely to become unstable. More preferably 2% by mass or more, even more preferably 5% by mass or more, even more preferably 8% by mass or more, and most preferably 10% by mass or more, and also K ＋ Since the presence of reduces weather resistance, it is preferably 30% by mass or less, more preferably 25% by mass or less, even more preferably 20% by mass or less, and most preferably 14% by mass or less. ＋ Using only this improves absorption and sharpness in the near-infrared region, but reduces weather resistance. Therefore, to improve weather resistance through the alkali mixing effect, K ＋ It is necessary to include one or more alkali metal components other than those listed above.

[0060] Rb (rubidium) is a component that has effects such as lowering the melting temperature of glass, lowering the liquidus temperature of glass, and improving the absorption capacity and sharp-cut properties in the near-infrared region. ＋ The content of Rb is preferably 0 to 20% by mass. ＋ A content of 20% by mass or less is preferable because it makes the glass less likely to become unstable. ＋Since the presence of reduces weather resistance, it is more preferably 15% by mass or less, even more preferably 10% by mass or less, and even more preferably 5% by mass or less. The alkali metal component is Rb ＋ Using only this improves absorption and sharpness in the near-infrared region, but reduces weather resistance. Therefore, Rb is added to improve weather resistance through the alkali mixing effect. ＋ It is necessary to include one or more alkali metal components other than those listed above.

[0061] Cs (cesium) is a component that has effects such as lowering the melting temperature of glass, lowering the liquidus temperature of glass, and improving the high absorption capacity and sharp cut properties in the near-infrared region. ＋ The content is preferably 0 to 20% by mass. ＋ A content of 20% by mass or less is preferable because it makes the glass less likely to become unstable. ＋ Since the presence of reduces weather resistance, it is more preferably 15% by mass or less, even more preferably 10% by mass or less, and even more preferably 5% by mass or less. The alkali metal component is Cs ＋ Using only Cs improves absorption and sharpness in the near-infrared region, but reduces weather resistance. Therefore, Cs is used to improve weather resistance through the alkali mixing effect. ＋ It is necessary to include one or more alkali metal components other than those listed above.

[0062] K ＋ and R ＋ (R ＋ Li ＋ Na ＋ , Rb ＋ and Cs ＋ At least one of the following is a component that lowers the melting temperature of the glass, lowers the liquidus temperature of the glass, or stabilizes the glass. ＋ and K ＋ The total amount, i.e., Li ＋ Na ＋ _K ＋ , Rb ＋ and Cs ＋ Total amount (ΣR ＋ +K ＋If the amount is 14% by mass or more, the effect is sufficiently obtained, and if it is 42% by mass or less, the glass is less likely to become unstable, which is preferable. ＋ and K ＋ The total amount is preferably 14.5 to 40% by mass. More preferably 16% by mass or more, even more preferably 18% by mass or more, even more preferably 35% by mass or less, even more preferably 33% by mass or less, even more preferably 30% by mass or less, and most preferably 28% by mass or less.

[0063] Magnesium (Mg) is a component used to lower the melting temperature of glass, lower the liquidus temperature of glass, stabilize glass, increase the strength of glass, and improve the weather resistance of glass. ２＋ The content is preferably 0 to 20% by mass. Mg ２＋ If the content is 20% by mass or less, problems such as glass instability or a decrease in near-infrared ray blocking properties are less likely to occur. More preferably, it is 15% by mass or less, even more preferably 10% by mass or less, and even more preferably 5% by mass or less.

[0064] Calcium (Ca) is a component that lowers the melting temperature of glass, lowers the liquidus temperature of glass, stabilizes glass, increases the strength of glass, and improves the weather resistance of glass. ２＋ The content is preferably 0 to 20% by mass. Ca ２＋ If the content is 20% by mass or less, problems such as glass instability or a decrease in near-infrared ray blocking properties are less likely to occur. More preferably it is 1% by mass or more, even more preferably 2% by mass or more, even more preferably 18% by mass or less, even more preferably 15% by mass or less, even more preferably 10% by mass or less, and most preferably 7% by mass or less.

[0065] Sr (strontium) is a component used to lower the melting temperature of glass, lower the liquidus temperature of glass, stabilize glass, increase the strength of glass, and improve the weather resistance of glass. ２＋ The content is preferably 0 to 30% by mass.２＋ If the content is 30% by mass or less, problems such as glass instability or reduced near-infrared cutting performance are less likely to occur. More preferably 1% by mass or more, even more preferably 2% by mass or more, even more preferably 4% by mass or more, most preferably 5% by mass or more, and more preferably 25% by mass or less, even more preferably 20% by mass or less, even more preferably 16% by mass or less, and most preferably 14% by mass or less.

[0066] Barium (Ba) is a component used to lower the melting temperature of glass, lower the liquidus temperature of glass, stabilize glass, enhance the absorption capacity of near-infrared light, and improve the sharpness of near-infrared light. ２＋ The content is preferably 0 to 40% by mass. Ba ２＋ If the content is 40% by mass or less, problems such as glass instability are less likely to occur. More preferably it is 1% by mass or more, even more preferably 5% by mass or more, even more preferably 10% by mass or more, most preferably 13% by mass or more, and more preferably 35% by mass or less, even more preferably 30% by mass or less, even more preferably 20% by mass or less, and most preferably 17% by mass or less.

[0067] R" ２＋ (R" ２＋ Mg ２＋ Ca ２＋ , Sr ２＋ , and Ba ２＋ At least one of the following is a component that lowers the melting temperature of the glass, lowers the liquidus temperature of the glass, stabilizes the glass, etc. ２＋ The total amount, i.e., Mg ２＋ Ca ２＋ , Sr ２＋ , and Ba ２＋ Total amount (ΣR") ２＋ If the amount is 14% by mass or more, the effect is sufficiently obtained, and if it is 40% by mass or less, the glass is less likely to become unstable. Therefore, ΣR" ２＋The content is preferably 14 to 40% by mass. More preferably 14.5% by mass or more, even more preferably 16% by mass or more, even more preferably 20% by mass or more, most preferably 22% by mass or more, and more preferably 38% by mass or less, even more preferably 35% by mass or less, even more preferably 33% by mass or less, and most preferably 30% by mass or less.

[0068] Zinc (Zn) has effects such as lowering the melting temperature of glass and lowering the liquidus temperature of glass. ２＋ The content is preferably 0 to 20% by mass. Zn ２＋ If the content is 20% by mass or less, problems such as glass instability, deterioration of glass solubility, and reduction of near-infrared ray blocking properties are less likely to occur. More preferably, it is 15% by mass or less, even more preferably 10% by mass or less, and even more preferably 5% by mass or less.

[0069] P ５＋ Content of / ΣR' (R' is Al ３＋ Mg ２＋ and Li ＋ At least one selected from the above (ΣR' is the total amount of R') is preferably 3.0 to 7.5.

[0070] P ５＋ This component enhances sharpness in the near-infrared region, but it also has the effect of reducing weather resistance. Furthermore, Al ３＋ Li ＋ Mg ２＋ These are components that each have the effect of improving weather resistance.

[0071] Therefore, P for ΣR' ５＋ By reducing the content ratio of to 7.5 or less, the weather resistance of the glass can be improved. Also, P relative to ΣR' ５＋ By setting the content ratio of to 3.0 or higher, the sharp cut performance of the glass in the near-infrared region can be maintained at a high level. P relative to ΣR' ５＋The proportion of the content is more preferably 3.5 or more, even more preferably 4.0 or more, even more preferably 4.5 or more, and more preferably 7.0 or less, even more preferably 6.5 or less, even more preferably 6.0 or less, and most preferably 5.5 or less.

[0072] Boron (B) may be included in a range of 20% by mass or less to stabilize the glass. ３＋ If the content is 20% by mass or less, problems such as deterioration of the glass's weather resistance or reduction in near-infrared ray blocking properties are less likely to occur. More preferably, it is 15% by mass or less, even more preferably 10% by mass or less, even more preferably 8% by mass or less, and most preferably 5% by mass or less.

[0073] In the phthalic acid glass of this disclosure, SiO ２ , GeO ２ , ZrO ２ , SnO ２ , TiO ２ , CEO ２ WO ３ , Y ２ O ３ La ２ O ３ , Gd ２ O ３ Yb ２ O ３ Nb ２ O ５ These components may be included in a total amount of 10% by mass or less to improve the weather resistance of the glass. If the content of these components is 10% by mass or less, problems such as the formation of devitrified foreign matter in the glass or a decrease in near-infrared ray blocking properties are less likely to occur. Preferably, it is 4% by mass or less, more preferably 3% by mass or less, even more preferably 2% by mass or less, and even more preferably 1% by mass or less.

[0074] Fe ２ O ３ , Cr ２ O ３ , Bi ２ O ３ NiO, V ２ O ５ MnO ２Both CoO and CoO are components that, when present in glass, reduce the transmittance of visible light. Therefore, it is preferable that these components are substantially absent from the glass. Here, substantially absent from the glass means that they are not included except for unavoidable impurities, and that these components are not actively added. Specifically, this means that the content of each of these components in the glass is approximately 100 ppm by mass or less.

[0075] The resin film 30 preferably contains a dye that absorbs infrared rays. Such dyes may be selected from, for example, squarylium dyes, phthalocyanine dyes, merocyanine dyes, and cyanine dyes.

[0076] Furthermore, the resin constituting the resin film 30 is not particularly limited as long as it is transparent. The resin may be selected from, for example, polyester resin, acrylic resin, epoxy resin, ene-thiol resin, polycarbonate resin, polyether resin, polyarylate resin, polysulfone resin, polyethersulfone resin, poly-paraphenylene resin, polyarylene ether phosphine oxide resin, polyamide resin, polyimide resin, polyamide-imide resin, polyolefin resin, cyclic olefin resin, polyurethane resin, and polystyrene resin.

[0077] From the viewpoint of the spectral characteristics, glass transition temperature (Tg), and adhesion of the resin film 30, it is preferable that the resin be selected from polyimide resin, polycarbonate resin, polyester resin, and acrylic resin.

[0078] Furthermore, from the viewpoint of heat resistance, the glass transition temperature (Tg) of the resin is preferably 200°C or higher.

[0079] The resin film 30 may contain only one type of resin, or two or more types may be used in combination.

[0080] In the optical filter 1, as shown in Figure 1, the resin film 30 is provided on one side of the near-infrared absorbing glass substrate 20, but the resin film 30 may be provided on both sides of the near-infrared absorbing glass substrate 20.

[0081] The method for forming the resin film is not particularly limited. The resin film can be formed by preparing a coating solution by dissolving or dispersing a dye, a resin or raw material component of the resin, and any additional components as needed in a solvent, coating this solution onto a support, drying it, and further curing it as needed. The support in this case may be a near-infrared absorbing glass substrate that constitutes an optical filter, or a releaseable support used only when forming the resin film. The solvent may be any dispersion medium or solvent that can stably disperse or dissolve the components.

[0082] Furthermore, the coating solution may contain a surfactant to improve voids caused by minute bubbles, indentations caused by the adhesion of foreign matter, and repulsion during the drying process. In addition, methods such as immersion coating, cast coating, or spin coating can be used for applying the coating solution. After applying the above coating solution to the support, a resin film is formed by drying. Furthermore, if the coating solution contains raw material components of a transparent resin, a curing treatment such as thermosetting or photocuring is performed.

[0083] Furthermore, the resin film can also be manufactured in film form by extrusion molding. The obtained film-like resin film can be laminated onto a near-infrared absorbing glass substrate and integrated by thermocompression bonding or the like to produce a substrate. The resin film may be present as one layer or as two or more layers within the optical filter. If there are two or more layers, each layer may have the same configuration or different configurations.

[0084] The thickness of the resin film is preferably 10 μm or less, more preferably 5 μm or less, from the viewpoint of in-plane film thickness distribution within the substrate after coating and appearance quality, and preferably 0.5 μm or more from the viewpoint of exhibiting desired spectral characteristics with an appropriate dye concentration. If the optical filter has two or more layers of resin film, it is preferable that the total thickness of each resin film is within the above range.

[0085] In the optical filter 1, it is preferable that at least one of the first dielectric multilayer film 10 and the second dielectric multilayer film 40 is designed as a near-infrared reflective layer (hereinafter also referred to as the NIR reflective layer). The other dielectric multilayer film is preferably designed as an NIR reflective layer, a reflective layer having a reflection region other than the near-infrared region, or an anti-reflective layer. In the optical filter 1, the first dielectric multilayer film 10 may be an NIR reflective layer, and the second dielectric multilayer film 40 may be designed as an anti-reflective layer. The optical filter of this disclosure may also include other dielectric multilayer films other than the first dielectric multilayer film and the second dielectric multilayer film. If the optical filter of this disclosure includes other dielectric multilayer films, the other dielectric multilayer films may be placed between the near-infrared absorbing glass substrate and the resin film.

[0086] The NIR reflective layer is a dielectric multilayer film designed to block light in the near-infrared region. For example, the NIR reflective layer has wavelength selectivity, transmitting visible light and primarily reflecting light in the near-infrared region. The reflective region of the NIR reflective layer may also include the light-blocking region of the resin film in the near-infrared region. The NIR reflective layer may be designed to further block light in wavelengths other than the near-infrared region, such as light in the near-ultraviolet region, not limited to its NIR reflection characteristics.

[0087] The NIR reflective layer is composed of, for example, a dielectric multilayer film in which dielectric films with different refractive indices are alternately stacked. Examples of dielectric films include a low refractive index dielectric film, a medium refractive index dielectric film, and a high refractive index dielectric film, and it is preferable to stack two or more of these dielectric films in combination. In particular, from the viewpoint of improving light shielding in the infrared region to the extent that ripple does not occur in the visible region, it is preferable to have at least a medium refractive index film.

[0088] The high refractive index film preferably has a refractive index of 2.2 or higher, and more preferably 2.2 to 2.5. Examples of materials for the high refractive index film include Ta ２ O ５ , TiO ２ ,TiO,Ti ２ O ３ Nb ２ O ５Other commercially available products include the OS50 (Ti) manufactured by Canon Optron Corporation. ３ O ５ ), OS10 (Ti ４ O ７ ), OA500 (Ta ２ O ５ and ZrO ２ A mixture of OA600 (Ta ２ O ５ and TiO ２ Examples include mixtures of TiO. ２ It is preferable.

[0089] The medium refractive index film preferably has a refractive index of 1.6 to 2.1. Examples of materials for the medium refractive index film include ZrO ２ Nb ２ O ５ Al ２ O ３ , HfO ２ Also, the OM-4 and OM-6 (Al) sold by Canon Optron Corporation. ２ O ３ and ZrO ２ Examples include a mixture of ), OA-100, H4 and M2 (alumina antania) sold by Merck, etc. Of these, Al is chosen based on its film-forming properties, reproducibility in refractive index, stability, etc. ２ O ３ Compounds of the system and Al ２ O ３ and ZrO ２ A mixture of these is preferred.

[0090] The low refractive index film preferably has a refractive index of 1.5 or less, and more preferably 1.4 to 1.5. Examples of materials for the low refractive index film include SiO ２ SiO ｘ N ｙ MgF ２ These are some examples. Other commercially available products include Canon Optron's S4F and S5F (SiO ２ and Al ２ O ３ A mixture of these can be cited. Of these, in terms of reproducibility, stability, and economy in film formation, SiO ２ It is preferable.

[0091] To create a dielectric multilayer film that satisfies both gentle light-shielding characteristics in the near-infrared region and low ripple characteristics in the visible light region, one approach is to combine several dielectric films with different spectral characteristics when transmitting and selecting the desired wavelength band. Specifically, by reducing the refractive index difference between the medium-refractive-index and low-refractive-index layers, reflection ripple in the visible light region can be suppressed. Conventional infrared cut filters use SiO ２ / TiO ２ While low-refractive-index and high-refractive-index materials are used to enhance light shielding in the near-infrared region, this configuration tends to generate reflection ripple in the visible light region. In the optical filter configuration of this disclosure, the transmittance in the near-infrared region can be reduced by the absorption characteristics of the near-infrared absorbing glass substrate and resin film, so it is not always necessary to use a low-refractive-index layer / high-refractive-index layer combination that tends to generate ripple in the visible light region.

[0092] The NIR reflective layer has a total number of layers of dielectric multilayer films constituting the reflective layer, preferably 20 or more, more preferably 30 or more, and even more preferably 35 or more. From the viewpoint of suppressing the occurrence of warping and an increase in layer thickness, the total number of layers is preferably 100 or less, more preferably 75 or less, and even more preferably 60 or less. Furthermore, the overall thickness of the NIR reflective layer is preferably 2 to 10 μm.

[0093] Furthermore, for the formation of dielectric multilayer films, vacuum deposition processes such as CVD, sputtering, and vacuum evaporation, as well as wet deposition processes such as spraying and dipping, can be used.

[0094] The NIR reflective layer may provide predetermined optical properties with a single layer (a group of dielectric multilayer films) or with two or more layers. When there are two or more layers, each reflective layer may have the same or different configuration. When there are two or more reflective layers, they are usually composed of multiple reflective layers with different reflection bands. When two reflective layers are provided, one may be a near-infrared reflective layer that shields light in the short-wavelength band of the near-infrared region, and the other may be a near-infrared / near-ultraviolet reflective layer that shields light in both the long-wavelength band and the near-ultraviolet region of the near-infrared region.

[0095] Examples of anti-reflective layers include dielectric multilayer films, intermediate refractive index media, and moth-eye structures with gradually changing refractive indices. Among these, dielectric multilayer films are preferred from the viewpoint of optical efficiency and productivity. The anti-reflective layer is obtained by alternately stacking dielectric multilayer films, similar to the reflective layer.

[0096] The optical filter 1 may include layers other than the first dielectric multilayer film 10, the near-infrared absorbing glass substrate 20, the resin film 30, and the second dielectric multilayer film 40. These other layers may include layers that provide absorption by inorganic nanoparticles, controlling the transmission and absorption of light in a specific wavelength range. Specific examples of inorganic nanoparticles include ITO (Indium Tin Oxides), ATO (Antimony-doped Tin Oxides), cesium tungstate, and lanthanum boride. ITO nanoparticles and cesium tungstate nanoparticles have high transmittance of visible light and light absorption over a wide range in the infrared wavelength region exceeding 1200 nm, making them suitable for use when shielding against such infrared light is required.

[0097] In the optical filter of this disclosure, when the first dielectric multilayer film side (the dielectric multilayer film 10 side in Figure 1) is the incident direction side, it is preferable that the following spectral characteristics (i) to (iii) are satisfied, from the viewpoint of suppressing changes in spectral characteristics during oblique incidence and obtaining good infrared shielding performance. (i) When the average transmittance in the spectral transmittance curve at an incident angle of 0 degrees is T[450-600(0deg)AVE] for wavelengths of 450-600 nm, and the average transmittance in the spectral transmittance curve at an incident angle of 60 degrees is T[450-600(60deg)AVE] for wavelengths of 450-600 nm, the absolute value of the difference between T[450-600(0deg)AVE] and T[450-600(60deg)AVE] is 15% or less. (ii) In the spectral transmittance curve at an incident angle of 0 degrees, the wavelength IR30(0deg) at which the transmittance is 30% is in the range of 600-700 nm, and in the spectral transmittance curve at an incident angle of 60 degrees, the wavelength IR30(60deg) at which the transmittance is 30% is in the range of 600-700 nm. (iii) The absolute value of the difference between IR30 (0 deg) and IR30 (60 deg) is 20 nm or less. (iii) In the spectral transmittance curve at an incident angle of 0 degrees, the average transmittance T [750 - 1000 (0 deg) AVE] for wavelengths of 750 to 1000 nm is 1% or less.

[0098] Satisfying spectral characteristic (i) means that the visible light transmittance in the 450-600 nm range does not change significantly even at a high incidence angle of 60 degrees, i.e., ripple is suppressed. The absolute value of the difference in spectral characteristic (i) is more preferably 13% or less, and even more preferably 11% or less. T[450-600 (0 deg) AVE] is preferably 85% or more, more preferably 87% or more, and even more preferably 88% or more, from the viewpoint of improving visible light transmittance. T[450-600 (60 deg) AVE] is preferably 75% or more, more preferably 77% or more, and even more preferably 78% or more, from the viewpoint of suppressing changes in visible light transmittance at high incidence angles.

[0099] Satisfying spectral characteristic (ii) means that the spectral curve in the 600-700 nm region is less likely to shift even at a high incidence angle of 60 degrees. The absolute value of the difference in spectral characteristic (ii) is more preferably 17 nm or less, and even more preferably 15 nm or less. IR30 (0 deg) is preferably 600-700 nm, more preferably 620-660 nm, and even more preferably 630-650 nm, from the viewpoint of improving the transmittance of red light.

[0100] Satisfying spectral characteristic (iii) means that the near-infrared region can be sufficiently shielded. The average transmittance T is more preferably 0.5% or less, and even more preferably 0.2% or less.

[0101] In the optical filter of this disclosure, when at least one of the first dielectric multilayer film side and the second dielectric multilayer film side is the incident direction side, the average reflectance RI [1000-1100 (5 deg) AVE] at wavelengths of 1000 to 1100 nm in the spectral reflectance curve at an incident angle of 5 degrees is preferably 15% or less, more preferably 12% or less, and even more preferably 10% or less, from the viewpoint of improving light shielding in the infrared region.

[0102] In the optical filter of this disclosure, when the first dielectric multilayer film side (dielectric multilayer film 10 side in Figure 1) is the incident direction side, and the transmittance at a wavelength of 1030 nm in the spectral transmittance curve at an incident angle of 5 degrees is T[1030(5deg)], and the reflectance at a wavelength of 1030 nm in the spectral reflectance curve at an incident angle of 5 degrees is R[1030(5deg)], the absorption rate at 1030 nm calculated by the following formula (A) is preferably 80% or more, more preferably 85% or more, and even more preferably 90% or more, from the viewpoint of improving light shielding in the infrared region. Absorption rate at 1030 nm (%) = 100 - T[1030(5deg)] - R[1030(5deg)] (A)

[0103] <Imaging Device> The imaging device of the present disclosure has an optical filter of the present disclosure. Preferably, the imaging device of the present disclosure further has a solid-state image sensor and an imaging lens. The optical filter of the present disclosure can be used, for example, by being placed between the imaging lens and the solid-state image sensor, or by being directly attached to the solid-state image sensor, imaging lens, etc. of the imaging device via an adhesive layer.

[0104] The embodiments of this disclosure are described below. A UV-Vis spectrophotometer (Hitachi High-Technologies Corporation, UH-4150 model) was used to measure each spectral characteristic. Unless otherwise specified, the spectral characteristics are measured at an incident angle of 0° (perpendicular to the main surface of the optical filter).

[0105] <Preparation of Glass Substrates> Glass substrate 1 and glass substrate 2 were prepared as near-infrared absorbing glass substrates. The raw materials were weighed and mixed so that the glass components after melting had the composition (mass%) shown in Table 1. The raw materials were placed in a 1 L crucible and heated and melted in an electric furnace at a predetermined melting temperature for 2 hours. After that, the mixture was clarified and stirred, and then cast into a rectangular mold measuring 100 mm (length) x 80 mm (width) x 20 mm (height) preheated to 50-500°C. After being held at 300-500°C, it was slowly cooled at approximately 1°C / min to obtain plate-like glass substrates 1 and 2 measuring 40 mm (length) x 40 mm (width) x 0.2 mm (thickness) with both sides optically polished. The following raw materials were used for each glass: P ５＋ In the case of H ３ PO ４ I used Al ３＋ In this case, AlF ３ I used Li. ＋ In this case, LiF, LiNO ３ Na was used. ＋ In this case, NaF was used. ＋ In this case, KF was used. Mg ２＋ In this case, MgO was used. Ca ２＋ In this case, CaF ２ Sr was used. ２＋ In the case of SrF ２ I used Ba. ２＋ In that case, BaF ２ It was used. Cu ２＋ In this case, CuO was used. F － In this case, the fluoride raw materials of the above components were used. In addition to the components listed in Table 1, O was added to the glass as an anion. ２－ Includes. O ２－ The content of highly volatile F －Although not specified because it varies depending on the content, both glass substrate 1 and glass substrate 2 contain O ２－ It contains [this]. Furthermore, known glass raw materials can be used. Note that the amounts of each glass component listed in Table 1 are rounded to one decimal place, with the second decimal place being rounded off. Meanwhile, in Table 1, "Total" and "ΣR" are [the relevant values]. ＋ +K ＋ " and "ΣR ’’２＋ The values ​​listed in the "Total" column are the sum of the amounts of each glass component, including the second decimal place and beyond, rounded to the first decimal place. Therefore, the values ​​calculated based on the amounts of each glass component listed in Table 1 are the values ​​listed in "Total" and "ΣR" in Table 1. ＋ +K ＋ " or "ΣR ’’２＋ The values ​​listed in the " " column may not match.

[0106]

[0107] <Fabrication of Optical Filters> Optical filters were fabricated through the following process. Examples 1 to 4 and Example 10 are examples, and Examples 5 to 9 are comparative examples. An optical filter 1 before fragmentation having the layer structure shown in Table 2 was fabricated as follows: Silica (SiO₂) was deposited onto one surface of the glass substrate 1 by vapor deposition. ２ ) / Titania (TiO ２ A dielectric multilayer film 1 was formed by alternately stacking a total of 48 layers of silica (SiO₂). Subsequently, silica (SiO₂) was deposited on the other surface of the glass substrate 1 by vapor deposition. ２ ) / Titania (TiO ２ A dielectric multilayer film 2 was formed by stacking layers of ) one by one. A resin film 1 with a thickness of 1 μm was formed on the dielectric multilayer film 2 by spin coating using the resin liquid 1 described below. Next, silica (SiO) was deposited on the resin film 1 by vapor deposition. ２ ) / Titania (TiO ２ A pre-fragmentation optical filter 1 was fabricated by depositing a dielectric multilayer film 3, which consists of nine layers of alternating layers of ).

[0108] On the other hand, the glass substrate 1 was changed to glass substrate 2, the resin liquid 1 was changed to resin liquid 2, and the dielectric multilayer film configuration was changed as shown in Table 2 to fabricate the pre-fragmentation optical filter 2.

[0109]

[0110] Resin solution 1 was prepared as follows. To polyimide resin (Mitsubishi Gas Chemical Company, Inc., "C3G30G" (product name), refractive index 1.59), dyes 1, 2, and 4 were added to the resin solids at the concentrations (mass%) listed in Table 3. As a resin diluent, γ-butyrolactone and cyclohexanone were added in a mass ratio of 1:1, and the mixture was heated to 50°C and stirred for 2 hours to prepare resin solution 1. On the other hand, resin solution 2 was prepared in the same manner as resin solution 1, except that dyes 3 and 4 were added to the resin solids at the concentrations (mass%) listed in Table 3. Details of dyes 1 to 4 are as follows.

[0111]

[0112] Dye 1 (squallium compound, maximum absorption wavelength (752 nm)): Synthesized according to International Publication No. 2017 / 135359. Dye 2 (squallium compound, maximum absorption wavelength (722 nm)): Synthesized according to International Publication No. 2014 / 088063 and International Publication No. 2016 / 133099. Dye 3 (squallium compound, maximum absorption wavelength (708 nm)): Synthesized according to U.S. Patent No. 5543086. Dye 4 (merocyanine compound, maximum absorption wavelength (399 nm)): Synthesized according to German Patent Publication No. 10109243.

[0113]

[0114] A polyimide resin solution with a resin concentration of 8.5% by mass was prepared by dissolving a polyimide resin (C3G30G, product name, manufactured by Mitsubishi Gas Chemical Company, Inc., refractive index 1.59) in a γ-butyrolactone (GBL) to cyclohexanone ratio of 1:1 (by mass). Each of the above dyes 1 to 4 was added to the resin solution at a concentration of 7.5 parts by mass per 100 parts by mass of resin, and the mixture was stirred and dissolved at 50°C for 2 hours to obtain a coating solution. The obtained coating solution was applied to alkali glass (D263 glass, manufactured by Schott, Inc., thickness 0.2 mm) by spin coating, forming coating films with a film thickness of approximately 1.0 μm. The spectral transmittance curves of the obtained coating films were measured in the wavelength range of 350 to 1200 nm using a UV-Vis spectrophotometer. The maximum absorption wavelength in the resin was determined for each of the above dyes 1 to 4. The results obtained are as described above.

[0115] <Formation of End Faces> A Yb:KGW laser (center wavelength 1030 nm) was used as the laser light source, and the laser beam was incident on the glass substrate 1 constituting the pre-pieced optical filter 1 or the glass substrate 2 constituting the pre-pieced optical filter 2 to obtain pieced optical filters. For Examples 1 to 3 and 5 to 8, the number of laser beam scans was set to 1. For Example 4, the number of laser beam scans was set to 3. For Example 10, the number of laser beam scans was set to 2. In addition, the laser output for each laser beam scan was selected appropriately so that the crack region did not reach the main surface of the glass substrate, and the average laser energy per pulse was set to 5 to 30 μJ, and was adjusted so that the positions of the crack regions generated in each scan overlapped. For Example 9, the pre-pieced optical filter 1 was cut using a dicing device under the conditions of a #600 blade and a processing speed of 3 mm / sec to obtain pieced optical filters.

[0116] [Evaluation] The surface roughness Ra of the crack region and the mirrored layer observed on the end face of each optical filter in Examples 1 to 10 was determined using the method described above. The results are shown in Table 4. Crack regions and mirrored layers were observed on the end faces of each optical filter in Examples 1 to 8 and Example 10. On the other hand, no crack regions or mirrored layers were observed on the end face of the optical filter in Example 9, and a uniformly rough end face was observed. In the column for the crack region in Example 9 in Table 4, the surface roughness Ra of the end face of Example 9 is recorded.

[0117] Furthermore, the surface roughness Ra(1), Ra(2), and Ra(3) of each optical filter in Examples 1 to 10 were determined using the method described above. Note that the crack regions observed on the end faces of the optical filters in Examples 1 to 8 and Example 10 consisted of a single crack layer. In Example 9, the entire end face of the optical filter was considered a crack region consisting of a single crack layer, and the surface roughness Ra(1), Ra(2), and Ra(3) were determined using the method described above. The results are shown in Table 4.

[0118] The width W in the thickness direction of the crack region observed on the end face of each optical filter in Examples 1-8 and Example 10 was calculated using the method described above. Based on the obtained width W value and the thickness of the glass substrate (0.2 mm), the width W (%) of the crack region in the thickness direction relative to the thickness of the glass substrate was determined. The results are shown in Table 4. Note that no mirror layer was observed on the end face of the optical filter in Example 9.

[0119] The fracture toughness and thermal expansion coefficient of glass substrates 1 and 2 used in Examples 1 to 10 were measured using the measurement method described above. The results are shown in Table 4. Glass substrate 1 was used for the optical filters in Examples 1 to 4 and Examples 6 to 10, while glass substrate 2 was used for the optical filter in Example 5.

[0120] For each optical filter in Examples 1 to 10, the spectral reflectance curve was measured at an incident angle of 5 degrees using a UV-Vis spectrophotometer, with the side containing the dielectric multilayer film 1 facing the incident direction. Based on the obtained spectral reflectance curves, the average reflectance RI [1000-1100 (5 deg) AVE] of each optical filter at wavelengths of 1000 to 1100 nm was determined. The results are shown in Table 4.

[0121] For each optical filter in Examples 1 to 10, the spectral transmittance curve and spectral reflectance curve at an incident angle of 5 degrees were measured using a UV-Vis spectrophotometer, with the side containing the dielectric multilayer film 1 as the incident direction. Based on the obtained spectral transmittance curve and spectral reflectance curve, the absorption rate at 1030 nm was calculated for each optical filter using the following formula (A), where T[1030(5deg)] is the transmittance at 1030 nm in the spectral transmittance curve at an incident angle of 5 degrees, and R[1030(5deg)] is the reflectance at 1030 nm in the spectral reflectance curve at an incident angle of 5 degrees. The results are shown in Table 4. Absorption rate at 1030 nm (%) = 100 - T[1030(5deg)] - R[1030(5deg)] (A)

[0122] For glass substrates 1 and 2 used in Examples 1 to 10, the spectral transmittance curves at an incident angle of 0 degrees were measured using a UV-Vis spectrophotometer. Based on the obtained spectral transmittance curves, the transmittance of the glass substrates at 1030 nm was determined. The results are shown in Table 4.

[0123] For the optical filters of Example 1 and Example 5, the spectral transmittance curves at an incident angle of 0 degrees and at an incident angle of 60 degrees were measured using a UV-Vis spectrophotometer, with the side on which the dielectric multilayer film 1 is provided as the incident direction. The measurement results for the optical filter of Example 1 are shown in Figure 10. In Figure 10, T0 is indicated on the spectral transmittance curve at an incident angle of 0 degrees (solid line), and T60 is indicated on the spectral transmittance curve at an incident angle of 60 degrees (dashed line).

[0124] Based on the spectral transmittance curves, the average transmittance of the optical filters in Example 1 and Example 5 at wavelengths of 450-600 nm in the spectral transmittance curve at an incident angle of 0 degrees (T[450-600(0deg)AVE]) and at wavelengths of 450-600 nm in the spectral transmittance curve at an incident angle of 60 degrees (T[450-600(60deg)AVE]) were determined. The results are shown in Table 5. Furthermore, the absolute value (absolute value A) of the difference between T[450-600(0deg)AVE] and T[450-600(60deg)AVE] was calculated. The results are shown in Table 5.

[0125] Furthermore, based on the spectral transmittance curves, it was confirmed that for the optical filters in Example 1 and Example 5, the wavelength IR30(0deg) at which the transmittance in the spectral transmittance curve at an incident angle of 0 degrees is 30% is in the range of 600 to 700 nm, and the wavelength IR30(60deg) at which the transmittance in the spectral transmittance curve at an incident angle of 60 degrees is 30% is in the range of 600 to 700 nm. The specific values ​​of IR30(0deg) and IR30(60deg) are shown in Table 5. Based on the values ​​of IR30(0deg) and IR30(60deg), the absolute value of the difference between IR30(0deg) and IR30(60deg) (absolute value B) was calculated. The obtained results are shown in Table 5.

[0126] Furthermore, based on the spectral transmittance curves at an incident angle of 0 degrees, the average transmittance T [750-1000(0deg)AVE] for the optical filters of Example 1 and Example 5 at an incident angle of 0 degrees for wavelengths of 750 to 1000 nm was determined. The obtained results are shown in Table 5.

[0127] For each optical filter in Examples 1 to 10, resin burr evaluation was performed using the method described below. The resin burr evaluation was then ranked according to the following criteria. The results are shown in Table 4. Resin burrs were measured using the optical microscope mode of a Keyence VKX-3000 laser microscope. For resin burr measurement, any optical microscope with a length measurement function may be used. Specifically, the optical filters were observed after fragmentation using the above microscope, and if the resin protruded from the outer shape of the optical filter, it was considered that resin burrs had occurred. The evaluation was based on the number of optical filters with resin burrs among 100 optical filters produced using the same procedure. A: Fewer than 7 optical filters with resin burrs out of 100. B: 7 or more but less than 15 optical filters with resin burrs out of 100. C: 15 or more optical filters with resin burrs out of 100.

[0128] Four-point bending strength tests were conducted on each optical filter from Examples 1 to 10. The four-point bending strength was measured in accordance with the "Four-Point Bending Strength Test" specified in JIS R1601:2008. Here, the test specimens were 5 mm x 5 mm square in size, with a support point pitch of 3 mm, a load point pitch of 1 mm, and a radius of curvature of 0.25 mm at the tip of the support that served as both a support and a load point. Bending strength was measured for 16 specimens under each condition, and the average value was calculated. A Shimadzu AGS-J measuring instrument was used. The four-point bending strength was then ranked according to the following criteria. The results are shown in Table 4. A: 350 MPa or more B: 300 MPa or more and less than 350 MPa C: Less than 300 MPa

[0129] Optical microscope images of the end faces of each optical filter in Examples 1 to 10 were obtained. Based on the obtained images, the hackle length was determined using the method described above. The hackle lengths were ranked according to the following criteria. The results are shown in Table 4. Note that in Example 9, although the end face of the optical filter was rough overall, no hackle was observed. A: 0 μm or more and less than 10 μm B: 10 μm or more and less than 30 μm C: 30 μm or more

[0130]

[0131]

[0132] From the evaluation results shown in Table 4, the following can be concluded: The fracture toughness of the glass substrate is 0.3 MPa·m １／２ The above is true, and the coefficient of thermal expansion is 135 × 10 －７The optical filter in Example 5, which was less than / K, had a large hackle width and poor dimensional accuracy. The optical filter in Example 8, in which the surface roughness Ra of the crack region at the end face exceeded 1.5 μm, had low four-point bending strength and a large hackle width, resulting in poor bending strength and dimensional accuracy. The optical filter in Example 9, in which no crack region or mirror layer was observed and the surface roughness Ra of the end face exceeded 1.5 μm, had low four-point bending strength and poor bending strength. The optical filter in Example 6, in which the width W in the thickness direction of the crack region was less than 15% of the thickness T of the near-infrared absorbing glass substrate, had severe resin burrs and poor dimensional accuracy. The optical filters in Examples 7 and 8, in which the width W in the thickness direction of the crack region exceeded 40% of the thickness T of the near-infrared absorbing glass substrate, had low four-point bending strength and a large hackle width, resulting in poor bending strength and dimensional accuracy. On the other hand, the optical filters of Examples 1 to 4 and 10, in which the fracture toughness and thermal expansion coefficient of the glass substrate are within a predetermined range, the crack region observed on the end face of the glass substrate and the surface roughness Ra of the mirror layer are within a predetermined range, and the width W of the crack region in the thickness direction is within a predetermined range with respect to the thickness T of the near-infrared absorbing glass substrate, had fewer resin burrs, higher four-point bending strength, a smaller hackle width, and sufficient bending strength and dimensional accuracy compared to the optical filters of Examples 5 to 9.

[0133] 1 Optical filter 10 First dielectric multilayer film 20 Near-infrared absorbing glass substrate 22 Cracked region 24 Mirror layer 26 One main surface 28 Other main surface 29 Huckle 30 Resin film 40 Second dielectric multilayer film

[0134] The disclosure of Japanese Patent Application No. 2025-025462, filed on 19 February 2025, is incorporated herein by reference in its entirety. All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

Claims

1. The device comprises a first dielectric multilayer film, a near-infrared absorbing glass substrate, a resin film, and a second dielectric multilayer film in this order, wherein the fracture toughness of the near-infrared absorbing glass substrate is 0.3 MPa·m. １／２ The coefficient of thermal expansion of the near-infrared absorbing glass substrate is less than 135 × 10 －７ An optical filter having a temperature of 1 / K or higher, wherein the end face of the near-infrared absorbing glass substrate has a crack region extending in a direction along the main surface of the near-infrared absorbing glass substrate and a pair of mirror-finished layers sandwiching the crack region, the surface roughness Ra of the crack region at the end face is 0.5 to 1.5 μm, the surface roughness Ra of the mirror-finished layers at the end face is less than 0.5 μm, and the width W of the near-infrared absorbing glass substrate in the thickness direction with respect to the crack region is 15 to 40% of the thickness T of the near-infrared absorbing glass substrate.

2. The optical filter according to claim 1, which satisfies the following spectral characteristics (i) to (iii) when the first dielectric multilayer film side is the incident direction side. (i) When the average transmittance in the spectral transmittance curve at an incident angle of 0 degrees is T[450-600(0deg)AVE] for wavelengths of 450-600 nm, and the average transmittance in the spectral transmittance curve at an incident angle of 60 degrees is T[450-600(60deg)AVE] for wavelengths of 450-600 nm, the absolute value of the difference between T[450-600(0deg)AVE] and T[450-600(60deg)AVE] is 15% or less. (ii) In the spectral transmittance curve at an incident angle of 0 degrees, the wavelength IR30(0deg) at which the transmittance is 30% is in the range of 600-700 nm, and in the spectral transmittance curve at an incident angle of 60 degrees, the wavelength IR30(60deg) at which the transmittance is 30% is in the range of 600-700 nm. (iii) The absolute value of the difference between IR30 (0 deg) and IR30 (60 deg) is 20 nm or less. (iii) In the spectral transmittance curve at an incident angle of 0 degrees, the average transmittance T [750 - 1000 (0 deg) AVE] for wavelengths of 750 to 1000 nm is 1% or less.

3. The optical filter according to claim 1, wherein, when at least one of the first dielectric multilayer film side and the second dielectric multilayer film side is the incident direction side, the average reflectance RI [1000-1100 (5 deg) AVE] at wavelengths of 1000 to 1100 nm in the spectral reflectance curve at an incident angle of 5 degrees is 15% or less.

4. When the first dielectric multilayer film side is the incident direction side, and the transmittance at a wavelength of 1030 nm in the spectral transmittance curve at an incident angle of 5 degrees is T[1030(5deg)], and the reflectance at a wavelength of 1030 nm in the spectral reflectance curve at an incident angle of 5 degrees is R[1030(5deg)], the optical filter according to claim 1, wherein the absorption rate at 1030 nm calculated by the following formula (A) is 80% or more. Absorption rate at 1030 nm (%) = 100 - T[1030(5deg)] - R[1030(5deg)] (A) 5. The optical filter according to claim 1, wherein the transmittance at 1030 nm in the spectral transmittance curve of the near-infrared absorbing glass substrate at an incident angle of 0 degrees is 10% or less.

6. The optical filter according to claim 1, wherein the crack region is composed of a single crack layer, and the surface roughness Ra(1) at the center of the width W in the thickness direction of the crack region on the end face, the surface roughness Ra(2) at a position 1 / 3 × W away from the center of the width W in the thickness direction of the crack region on the end face toward one main surface side of the near-infrared absorbing glass substrate, and the surface roughness Ra(3) at a position 1 / 3 × W away from the center of the width W in the thickness direction of the crack region on the end face toward the other main surface side of the near-infrared absorbing glass substrate satisfy the following formulas (B) to (C): Ra(1) > Ra(2) (B) Ra(1) > Ra(3) (C) 7. The optical filter according to claim 1, wherein the crack region is composed of two layers: a first crack layer and a second crack layer having a thinner width in the thickness direction of the crack region than the first crack layer, and when the first crack layer is divided into two in the thickness direction of the crack region to form a first layer facing the second crack layer and a second layer not facing the second crack layer, the surface roughness at the center of the width of the first layer in the thickness direction is defined as surface roughness Ra(4), the surface roughness at the center of the width of the second layer in the thickness direction is defined as surface roughness Ra(5), and the surface roughness at the center of the width of the second crack layer in the thickness direction is defined as surface roughness Ra(6), such that surface roughness Ra(4), surface roughness Ra(5), and surface roughness Ra(6) satisfy the following formulas (D) to (E). Ra(4)>Ra(5) (D) Ra(4)>Ra(6) (E) 8. The optical filter according to claim 1, wherein the crack region is composed of at least three crack layers, and from among the at least three crack layers, crack layer A located closest to the center of the width W in the thickness direction of the crack region is selected, and the surface roughness at the center of the width of crack layer A in the thickness direction is defined as surface roughness Ra(7), and the surface roughness at the center of the width of the two crack layers B and C located on both sides of crack layer A in the thickness direction is defined as surface roughness Ra(8) and surface roughness Ra(9), respectively, such that surface roughness Ra(7), surface roughness Ra(8), and surface roughness Ra(9) satisfy the following formulas (F) to (G). Ra(7)>Ra(8) (F) Ra(7)>Ra(9) (G) 9. The optical filter according to claim 1, wherein a hackle is observed in the mirror-finish layer, extending from the crack region toward one main surface side or the other main surface side of the near-infrared absorbing glass substrate, and the length of the hackle is defined as the distance between a line segment L1 passing through one end of the hackle and parallel to the main surface of the near-infrared absorbing glass substrate and a line segment L2 passing through the other end of the hackle and parallel to the main surface of the near-infrared absorbing glass substrate, and the length of the hackle is 30 μm or less.

10. The near-infrared absorbing glass substrate contains the components P, Al, K, Cu, F, and R (where R is at least one selected from Li, Na, Rb, and Cs), and Al ３＋ The optical filter according to claim 1, comprising phthalic acid glass having a content of 2 to 20% by mass, and an expected value of the ionic radius of the alkali metal components consisting of K and R being 80 pm or more and less than 133 pm.

11. An imaging apparatus having an optical filter according to any one of claims 1 to 10.

Images