JP2017112180A - Method for producing microlens for solid-state image sensor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for producing a microlens for solid-state image sensors, capable of improving a condensing effect of a microlens.SOLUTION: In a method for producing a microlens for solid-state image sensors, a microlens 19 is formed by performing a plural times of exposure on a microlens material layer 14 of a laminate using a gray tone mask 16 while the focal point 18 of exposure light 15 is moved in a direction in parallel with the emission direction of the exposure light 15, the laminate including a planarization layer 12, color filer layers 13 of a plurality of colors, and the microlens material layer 14 laminated, in the order, on a semiconductor substrate 11 on which a plurality of light receiving elements are formed.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a method for manufacturing a microlens for a solid-state imaging device.

In the solid-state imaging device, CMOS type or CCD type photoelectric conversion elements that generate charges by absorbing light are two-dimensionally arranged, and the charges generated by the photoelectric conversion elements are transferred to the outside as electrical signals. Widely used in TV cameras, digital still cameras, etc.
The region (opening) where the photoelectric conversion element on the solid-state image sensor contributes to photoelectric conversion depends on the size and the number of pixels of the solid-state image sensor, but is limited to about 20 to 40% of the total area of the solid-state image sensor. End up. Since the small aperture leads to a decrease in sensitivity as it is, in order to compensate for this, it is a common practice to form a condensing microlens on the photoelectric conversion element.

  A thermal reflow method is known as a method for manufacturing a microlens. That is, first, a positive photosensitive resin is patterned into a substantially rectangular shape in plan view by a well-known photolithography method that performs pattern exposure and development. Next, the temperature is increased on the hot plate to melt the pattern. In this case, the lens shape is obtained by the surface tension of the resin surface. Thereafter, the resin is thermally cured (see, for example, Patent Document 1). As another method, the pattern obtained by the thermal reflow method is used as a lens matrix, and the lens matrix is used as an etching mask for the lens material located under the lens matrix, and is laid under the lens by a dry etching method. A transfer method is disclosed in which a lens shape is scraped onto a transparent resin layer (see, for example, Patent Document 2).

  Thus, the method of transferring the shape of the lens matrix to the transparent resin layer by the dry etching method is suitable for forming a relatively large microlens of 3 μm to 10 μm. Therefore, if the microlens is larger than 5 μm, the surface roughness of the lens is small, and it can be said that the level is satisfactory in practical use. However, in recent years, there has been a strong demand for downsizing of a solid-state imaging device module, and a microlens having a pixel size of 2.5 μm or less or in the vicinity of 1.5 μm is required. When a microlens having a size of 2 μm or less is formed by transfer using a dry etching method, the lens surface tends to be rough, and there is a problem that the condensing effect tends to be reduced.

In addition, in the thermal reflow method, in order to obtain a microlens array, when forming a plurality of microlenses at the same time, if the gap between adjacent lenses is narrowed, the volume of the photosensitive resin pattern expands, and the adjacent microlenses are expanded. As a result of contact between the lenses, the shape collapses at the boundary between the microlenses. When cured in this state, the lens contracts during the cooling process. In this case, since the curvature of the lens shape is different between the diagonal direction of the pixel having a substantially rectangular shape in plan view and the direction parallel to the side of the pixel, there is a problem in that the light collection efficiency is reduced due to aberration.
In addition, an exposure apparatus called a stepper (reduced projection exposure machine) using a single wavelength light as an exposure light source is used for exposure at the time of forming a microlens. When exposure is performed at a single wavelength, a phenomenon called standing wave effect occurs, and the shape of the microlens after development is caused by interference between the exposure light incident on the photosensitive resin and the reflected light from the substrate surface. Has a stepped shape, and there is a problem that a desired gently curved lens shape cannot be obtained. (For example, refer to Patent Document 3).

JP-A-11-72356 Japanese Patent Laid-Open No. 1-10666 Japanese Patent No. 4249586

  This invention solves such a problem, and provides the manufacturing method of the micro lens for solid-state image sensors which can improve the condensing effect of a micro lens.

  In order to solve the above problems, in one embodiment of the present invention, a planarization layer, a plurality of color filter layers, and a microlens material layer are stacked in this order on a semiconductor substrate on which a plurality of light receiving elements are formed. Exposure that forms a microlens by performing multiple exposures by moving the focus of exposure light in a direction parallel to the emission direction of exposure light using a gray-tone mask on the microlens material layer of the laminate A process is performed.

  According to one embodiment of the present invention, since a gray-tone mask is used, a smooth hemispherical microlens can be formed. Moreover, the standing wave effect can be suppressed by moving the focus of the exposure light in a direction parallel to the emission direction of the exposure light and performing the exposure a plurality of times. Thereby, the manufacturing method of the micro lens for solid-state image sensors which can improve the condensing effect of a micro lens can be provided.

It is a figure which shows the manufacturing method of the micro lens for solid-state image sensors which concerns on embodiment. It is a top view of the micro lens for solid-state image sensors. It is a graph showing the cross-sectional shape of a micro lens.

Embodiments according to the present invention will be described below with reference to the drawings.
1A to 1F are views showing a method for manufacturing a microlens for a solid-state imaging device according to an embodiment of the present invention. First, as shown in FIG. 1A, a planarizing layer 12 is formed on a semiconductor substrate 11 (a light receiving element is not shown) on which a plurality of light receiving elements are formed to reduce unevenness and improve smoothness. Then, a color resist is formed on the planarizing layer 12, and a color filter layer 13 in which color filters of each color are spread at predetermined positions based on the Bayer arrangement is laminated by a plurality of photolithography processes. The color filter layer 13 can be a primary color filter composed of green, red and blue, or a complementary color filter composed of yellow, cyan and magenta. The vertical and horizontal dimensions of each color curler filter of the color filter layer 13 are, for example, 1.0 μm or more and 10.0 μm or less, and preferably 1.5 μm or more and 2.5 μm or less.

Next, as shown in FIG. 1B, a microlens material layer 14 is laminated on the color filter layer 13. The microlens material layer 14 is formed by applying an acrylic resin by a spin coat method and thermosetting on a hot plate. The film thickness of the microlens material layer 14 is, for example, not less than 0.5 μm and not more than 5.0 μm, preferably not less than 0.5 μm and not more than 2.0 μm.
As the material of the microlens material layer 14, a resin material having heat flow resistance can be adopted. For example, there is a thermoplastic resin material that has a high glass transition temperature and does not lose its shape before being cured by heat treatment under conditions of 100 ° C. or higher and 220 ° C. or lower. Specifically, it is preferable to contain a base resin having a mass average molecular weight (Mw: measured value in terms of styrene by gel permeation chromatography (GPC)) of 10,000 to 30,000. More preferably, the base resin has a mass average molecular weight of 20000 or more and 30000 or less. When the mass average molecular weight of the base resin is 10,000 or more, heat resistance and heat flow resistance can be improved. Moreover, generation | occurrence | production of the residue at the time of image development can be suppressed because the mass mean molecular weight of base resin is 30000 or less. By using such a resin, it is possible to avoid the expansion of the volume of the microlens pattern and the contact between the adjacent microlenses 19, and as a result, the occurrence of shape collapse at the boundary portion between the adjacent microlenses 19 can be prevented. it can.

As the material of the microlens material layer 14, an acrylic resin, a fluorine-based acrylic resin, an epoxy resin, a polyester resin, a urethane resin, a melamine resin, a urea resin, a styrene resin, a phenol resin, or a copolymer thereof can be used. . In particular, an acrylic resin having high heat resistance is preferable. An acrylic resin, a styrene resin, etc. are common, The refractive index shall be the range of about 1.5-1.7. These resins may be used alone or in combination of two kinds. If the material used for the microlens material layer 14 has a high refractive index, the incident light can be refracted more greatly and can be incident on the light receiving element, and the condensing effect of the microlens 19 can be further enhanced, and thus sensitivity can be increased. Can be improved. In order to increase the refractive index of the microlens 19, a metal oxide may be included. Examples of the metal oxide include titanium oxide, zirconium oxide, and silicon oxide. In particular, titanium oxide is preferable.
Further, the material of the microlens material layer 14 can contain other additives as required so as not to impair the characteristics of the present invention. As other additives, there are an adhesion assistant for improving adhesion to the substrate, a surfactant, a leveling agent, a dispersing agent, and a curing agent for improving coating property.

  Next, in order to control the convex shape of the microlens 19 by an exposure method, a special exposure mask called a gray tone mask 16 is used. The gray tone mask 16 is formed by forming a light-shielding film having a high light transmittance on a quartz substrate at a portion corresponding to a thin film portion of a lens element to be manufactured. That is, it is a mask having a light-shielding film with a gradation of gradation. The gradation of the gradation of the gray tone mask 16 is achieved by a partial difference in the number (roughness) per unit area of small diameter dots that are not resolved by the exposure light 15. By forming the microlens 19 by an exposure method using the gray tone mask 16, it is possible to suppress the roughness of the lens surface, reduce light scattering on the lens surface, and obtain a smooth microlens 19.

  Further, when exposing the microlens material layer 14, an exposure apparatus called a general stepper is used. As the exposure wavelength, ultraviolet light is used, and 365 nm is preferable. In general, the reflectance of the semiconductor substrate 11 is high, and a standing wave is generated by interference between incident light on the microlens material layer 14 and reflected light reflected by the semiconductor substrate 11. Therefore, it is difficult to obtain a microlens 19 having a good shape. Therefore, in the present embodiment, the microlens material layer 14 is exposed a plurality of times by using the gray tone mask 16 and moving the focal point 18 of the exposure light 15 in a direction parallel to the emission direction of the exposure light 15. Thus, the microlens 19 was formed. As a result, it was found that the standing wave effect can be suppressed and a preferable lens shape can be obtained.

  First, as shown in FIG. 1C, the exposure light 15 is irradiated onto a gray-tone mask 16 having a predetermined pattern, and the transmitted light beam is reduced by a condensing optical system 17 so as to reduce the microlens material layer 14. Is exposed to a reduced transmitted light beam (hereinafter also referred to as “first light”). Note that the distance from the surface position of the microlens material layer 14 is ΔZ in a direction parallel to the irradiation direction of the exposure light (hereinafter also referred to as “Z-axis direction”). ΔZ is “−” on the irradiation direction side of the surface position of the microlens material layer 14 and “+” on the opposite side to the irradiation direction. When the focal point 18 of the condensing optical system 17 coincides with the position of the surface of the microlens material layer 14, the defocus amount ΔZ is “0”. In the present embodiment, the amount of defocus during exposure with the first light is ΔZ = 0.

Next, as shown in FIGS. 1D and 1E, the exposure light 15 is irradiated onto the gray tone mask 16 having a predetermined pattern, and the transmitted light beam is reduced by the condensing optical system 17, The microlens material layer 14 is exposed with a reduced transmitted light beam (hereinafter also referred to as “second light”). At this time, the defocus amount is set so as to be in a so-called defocus (focal shift) state so that the focal point 18 is deviated from the focal position of the first light. As the defocus amount, a desired focus offset amount is set using the autofocus function of the exposure machine. In the present embodiment, the defocus amount during exposure with the second light is ΔZ = + Δ. Thereby, in this embodiment, the focus 18 of the exposure light 15 is moved in a direction parallel to the emission direction of the exposure light 15 to perform the second and subsequent exposures.
Thus, in the present embodiment, the second exposure is performed on the microlens material layer 14 by defocusing the second light with respect to the first light. Therefore, the phase of the reflected light from the semiconductor substrate 11 can be shifted according to the defocus amount, and the incident light can be canceled by the reflected light. Thereby, if the defocus amount is optimized, the standing wave effect can be effectively suppressed.

Further, the defocus amount, that is, the movement amount ΔZ of the focal point 18 may be set according to the following equation (2), for example. Thereby, the standing wave effect can be effectively suppressed. Even if the defocus amount ΔZ does not completely match the following equation (2), the effect of suppressing the standing wave effect can be obtained.
ΔZ = m · (λ / 2n · N) (2)
Where λ is the exposure wavelength, n is the refractive index of the microlens material, m is an integer, and N is the number of exposures. For example, when the wavelength of the exposure light 15 is 365 nm, the refractive index n of the microlens material layer 14 is about 1.6, and the number of exposures is 2, ΔZ in the above equation (2) is about 228 nm (= λ / 2n). To do. By using this ΔZ as a defocus amount and moving the focal position of the exposure light 15 in a direction parallel to the emission direction, the phase of the second light is shifted from the first light by this defocus amount. Become. Thereby, the influence of a standing wave effect can be suppressed and a favorable lens shape can be obtained.

  Next, the exposed laminated body (laminated body in which the semiconductor substrate 11, the flat layer 12, the color filter layer 13, and the microlens layer 14 are laminated in this order) is developed and heat-treated, and FIG. As shown, a hemispherical microlens 19 is formed on the color filter layer 13. FIG.1 (f) is sectional drawing at the time of fracture | rupture by the aa line | wire of FIG. FIG. 2 is a plan view of a microlens for a solid-state imaging device manufactured by the manufacturing method of the present embodiment. Thereby, a smooth microlens 19 having a lens surface roughness (Ra) of 50 nm or less can be obtained.

(Effect of this embodiment)
The invention according to this embodiment has the following effects.
(1) The method for manufacturing a microlens for a solid-state imaging device according to this embodiment includes a planarizing layer 12, a plurality of color filter layers 13, and a microlens material on a semiconductor substrate 11 on which a plurality of light receiving elements are formed. Using the gray-tone mask 16 having a predetermined pattern, the focal point 18 of the exposure light 15 is parallel to the emission direction of the exposure light 15 with respect to the microlens material layer 14 of the laminate in which the layers 14 are laminated in this order. The microlens 19 is formed by moving in the direction and performing a plurality of exposures.
According to such a configuration, since the gray tone mask 16 is used, a smooth hemispherical microlens 19 can be formed. Further, the standing wave effect can be suppressed by moving the focal point 18 of the exposure light 15 in a direction parallel to the emission direction of the exposure light 15 and performing the exposure a plurality of times. Thereby, the manufacturing method of the micro lens for solid-state image sensors which can improve the condensing effect of the micro lens 19 can be provided.

(2) In the method for manufacturing the microlens for a solid-state imaging device according to the present embodiment, the defocus amount ΔZ (the movement amount ΔZ of the focal point 18) is set according to the following equation (3).
ΔZ = m · (λ / 2n · N) (3)
According to such a configuration, the influence of the standing wave effect can be effectively suppressed.

(3) In the method for manufacturing a microlens for a solid-state imaging device according to this embodiment, the mass average molecular weight of the base resin of the microlens material layer 14 is 10,000 or more and 30,000 or less.
According to such a configuration, the heat flow resistance of the microlens 19 can be improved. Therefore, it can be avoided that the volume of the microlenses 19 expands and the adjacent microlenses 19 come into contact with each other. Thereby, shape collapse can be prevented at the boundary portion of the microlens 19.

Embodiments of a method for manufacturing a microlens for a solid-state imaging device according to the present invention will be described below.
Example 1
As shown in FIG. 1 (f), the microlens for a solid-state imaging device of Example 1 is obtained by laminating a planarization layer 32, a color filter layer 33, and a microlens 39 in this order on a semiconductor substrate 31. is there. FIG.1 (f) is sectional drawing at the time of fracture | rupture by the aa line | wire of FIG.
As shown in FIG. 1A, first, a thermosetting acrylic resin solution is applied onto the semiconductor substrate 31 by spin coating, and heat treatment is performed on a hot plate at 200 ° C. for 5 minutes. A planarizing layer 32 having a thickness of 1 μm was formed. Next, an acrylic photosensitive green coloring resist was applied on the formed planarizing layer 32 by spin coating, and prebaking treatment was performed at 80 ° C. for 1 minute on a hot pouch. Next, the photosensitive green colored resist was subjected to pattern exposure using an i-line stepper (NSR-TFHi12 manufactured by Nikon Corporation, exposure wavelength 365 nm), and then with an organic alkali developer (TMAH concentration 0.05%) for 1 minute. The film was developed, rinsed thoroughly with pure water, and drained and dried. Thereafter, post baking was performed at 220 ° C. for 6 minutes to form a G (green) pixel pattern (color filter pattern). Similarly to the G (green) pixel pattern (color filter pattern), the R (red) pixel pattern and the B (blue) pixel pattern are formed at predetermined positions based on the Bayer array using a photolithography method, and the pixel size is 2 A 0.0 μm color filter layer 33 was obtained. When the R (red), G (green), and B (blue) color filter patterns were measured, the film thicknesses were each 1.0 μm.

  Here, in the color filter layer 33, red pigment: CI Pigment Red 254 (manufactured by BASF Japan Ltd., “Ilgar Four Red B” is used to form R (red), G (green), and B (blue) color filters. -CF ") and CIPigment Red 177 (" chromoftal red A2B "manufactured by BASF Japan KK), green pigment: CI Pigment Green 36 (" Lionol Green 6YK "manufactured by Toyocolor Co., Ltd.) and CI Pigment Yellow 150 (Bayer's “Fungchon First Yellow Y-5688”), Blue pigment: CIPigment Blue 15 (Toyocolor Co., Ltd. “Lionol Blue ES”) CIPigment Violet 23 (BASF Japan Co., Ltd. “Paliogen Violet” 5890 ") together with an acrylic resin and a cyclohexanone solvent. A color resist was used. The amount of the color material added was about 50% in terms of the solid content ratio in the resist.

Next, as shown in FIG. 1B, a positive photoresist (refractive index 1.6) containing a base resin having a mass average molecular weight of 30000 is applied on the color filter layer 33 by a spin coating method. The mold photoresist was pre-baked at 90 ° C. for 1 minute on a hot plate. As a result, a microlens material layer 34 having a thickness of 0.8 μm was formed.
Next, as shown in FIG.1 (c), the gray-tone mask 36 which has the predetermined pattern for the exposure light 35 using i line | wire stepper (NSR-TFHi12 by Nikon Corporation, exposure wavelength 365nm). Then, the transmitted light flux was reduced to ¼ by the condensing optical system 37 and exposed to the microlens material layer 34 for a predetermined time. This is the first exposure. The defocus amount ΔZ at the time of exposure was set to “0”, and the focus offset amount of the focal point 38 was set to “0”.

  Next, as shown in FIG. 1 (d), a gray tone mask 36 having a predetermined pattern of exposure light 35 using an i-line stepper (NSR-TFHi12 manufactured by Nikon Corporation, exposure wavelength 365 nm). Then, the transmitted light flux was reduced to ¼ by the condensing optical system 37 and exposed to the microlens material layer 34 for a predetermined time. This is the second exposure. The defocus amount ΔZ at the time of exposure was set to “+0.228”, and the focus offset amount of the focal point 38 was set to “+0.2”. The focus offset amount is a significant figure up to the first decimal place. ΔZ was calculated according to the above equation (2). Here, λ = 365 nm, n = 1.6, m = 1, and N = 2.

  Next, the development processing for 1 minute was performed with the organic alkali developing solution (TMAH density | concentration 1.19%). Next, it was thoroughly rinsed with pure water and drained and dried. Thereafter, post-baking was performed at 200 ° C. for 6 minutes. Next, bleaching was performed using an exposure apparatus under the condition of 300 mJ, and the microlens material layer 34 was used as the microlens 39 and the microlens 39 was formed on the color filter layer 33 as shown in FIG. . The microlens 39 is a smooth hemispherical lens having a lens height of 0.8 μm, and no influence of standing waves was observed. FIG. 3 shows the shape of the surface layer of the microlens 39 obtained using an AFM (atomic force microscope).

(Example 2)
In Example 2, the mass average molecular weight of the base resin of the positive photoresist used for the microlens material layer 34 was changed to 10,000. Otherwise, the configuration was the same as in Example 1.
(Example 3)
In Example 3, the mass average molecular weight of the base resin of the positive photoresist used for the microlens material layer 34 was changed to 18000. Otherwise, the configuration was the same as in Example 1.
Example 4
In Example 4, the defocus amount when performing exposure with the second light is changed to Δ = + 0.5 (m = 3). Other than that, it was set as the structure similar to Example 1 (refer FIG.1 (d)).

(Example 5)
In Example 5, the defocus amount when performing exposure with the second light was changed to Δ = + 0.9 (m = 4). Other than that, it was set as the structure similar to Example 1 (refer FIG.1 (d)).
(Example 6)
In Example 6, the defocus amount when performing exposure with the second light was changed to Δ = −0.5 (m = −3). Other than that, it was set as the structure similar to Example 1 (refer FIG.1 (e)).
(Example 7)
In Example 7, the defocus amount when performing exposure with the second light was changed to Δ = −0.9 (m = −4). Other than that, it was set as the structure similar to Example 1 (refer FIG.1 (e)).
(Example 8)
In Example 8, the refractive index of the base resin of the positive photoresist used for the microlens material layer 34 was changed to 1.5. Otherwise, the configuration was the same as in Example 1.

(Comparative Example 1)
In Comparative Example 1, the microlens material layer 34 was changed to be exposed only by the first light without being divided into a plurality of times. Otherwise, the configuration was the same as in Example 1.
(Comparative Example 2)
In Comparative Example 2, the mass average molecular weight of the base resin of the positive photoresist used for the microlens material layer 34 was changed to 800. Otherwise, the configuration was the same as in Example 1.
(Comparative Example 3)
In Comparative Example 3, the mass average molecular weight of the base resin of the positive photoresist used for the microlens material layer 34 was changed to 40000. Otherwise, the configuration was the same as in Example 1.

Next, the shape, lens height, and surface roughness of the microlens 39 formed by the manufacturing methods of Examples 1 to 8 and Comparative Examples 1 to 3 were evaluated.
[Shape of micro lens 39]
The shape of the microlens 39 was observed with a length measurement SEM (eCD2-XP manufactured by KLA-Tencor Corporation). The case where the shape of the microlens 39 was hemispherical was evaluated as a pass “◯”, and the case where the microlens 39 was not hemispherical was evaluated as a disapproval “x”.
[Lens height and surface roughness of the microlens 39]
The lens height and surface roughness of the microlens 39 were measured with an AFM (atomic force microscope) manufactured by Toyo Technica Co., Ltd. (i-nano).
The evaluation results are shown in Table 1.

As is clear from Table 1, in the manufacturing methods of Examples 1 to 8, hemispherical microlenses 39 having no influence of standing waves and having a smooth surface were obtained.
Moreover, in the manufacturing methods of Comparative Examples 1 to 3, a desired lens shape could not be formed.

11, 31 ... Semiconductor substrate 12, 32 ... Flattening layer 13, 33 ... Color filter layer 14, 34 ... Microlens material layer 15, 35 ... Exposure light 16, 36 ... Gray tone masks 17, 37 ... Condensing optical system 18, 38 ... Focus 19, 39 ... Micro lens

Claims (3)

  Predetermined for the microlens material layer of the laminate in which a planarizing layer, a plurality of color filter layers, and a microlens material layer are laminated in this order on a semiconductor substrate on which a plurality of light receiving elements are formed. Solid-state imaging characterized in that a microlens is formed by performing a plurality of exposures by moving a focal point of exposure light in a direction parallel to an emission direction of the exposure light using a gray tone mask having a pattern Manufacturing method of microlens for element. 2. The method of manufacturing a microlens for a solid-state imaging device according to claim 1, wherein the movement amount ΔZ of the focal point is set according to the following formula (1).
ΔZ = m · (λ / 2n · N) (1)
Where λ is the exposure wavelength, n is the refractive index of the microlens material, m is an integer, and N is the number of exposures.   3. The method of manufacturing a microlens for a solid-state imaging device according to claim 1, wherein a mass average molecular weight of the base resin of the microlens material layer is 10,000 or more and 30,000 or less.

Images