Image sensor and manufacturing method therefor

Abstract

The present disclosure relates to an image sensor and a manufacturing method therefor. The image sensor comprises at least one pixel cell, and each pixel cell comprises: a photosensitive element; and a pseudo-random array diffractive optical element (PRADOE) located in front of a light receiving side of the photosensitive element and comprising an array of unit structures oriented at a pseudo-random angle. The PRADOE is configured to cause multi-order diffraction of light to be received by the photosensitive element.

Description

Image sensor and method of manufacturing the same

[0001] Cross Reference to Related Applications

[0002] This application is based on and claims priority to CN application No. 202410891090.1, filed on July 4, 2024, the disclosure of which is incorporated herein in its entirety. TECHNICAL FIELD

[0003] The present disclosure relates generally to the field of image sensors, and more particularly, to an image sensor and a method of manufacturing the same. BACKGROUND

[0004] Complementary Metal Oxide Semiconductor (CMOS) image sensors (e.g., Silicon (Si)-based CMOS image sensors) have made significant progress in terms of process cost and performance, especially in the visible spectrum.

[0005] SUMMARY

[0006] According to an aspect of the present disclosure, there is provided an image sensor, comprising at least one pixel unit, wherein each of the at least one pixel unit comprises: a light sensing element; and a Pseudo-Random Array Diffractive Optical Element (PRADOE) located in front of a light receiving side of the light sensing element and comprising an array of cell structures oriented at pseudo-random angles, wherein the PRADOE is configured to cause multi-order diffraction of light to be received by the light sensing element.

[0007] According to another aspect of the present disclosure, there is provided a method for manufacturing an image sensor, comprising: providing a substrate having a light sensing element; and forming a Pseudo-Random Array Diffractive Optical Element (PRADOE) in front of a light receiving side of the light sensing element, the PRADOE comprising an array of cell structures oriented at pseudo-random angles, wherein the PRADOE is configured to cause multi-order diffraction of light to be received by the light sensing element.

[0008] Other features of the present disclosure, and its advantages, will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0009] The accompanying drawings, which are part of this specification, illustrate embodiments of the present disclosure and serve to explain the principles of the present disclosure.

[0010] The present disclosure can be more completely understood in consideration of the following detailed description in connection with the accompanying drawings, in which:

[0011] FIG. 1 schematically illustrates a cross-sectional side view and a top view of an image sensor, according to some embodiments of the present disclosure;

[0012] FIG. 2 schematically illustrates a top view and a perspective view of a PRAD OE included in the image sensor of FIG. 1, as well as a top view of a single cell structure of the PRAD OE;

[0013] FIG. 3 exemplarily illustrates multi-stage diffraction of light caused by a PRAD OE formed below a light-receiving surface;

[0014] FIG. 4 exemplarily illustrates multi-stage diffraction of light caused by a PRAD OE formed above a light-receiving surface;

[0015] FIGS. 5A-5B schematically depict several examples of a PRAD OE, according to some embodiments of the present disclosure, in which a pitch between centers of adjacent cell structures of the PRAD OE is different;

[0016] FIG. 6 schematically depicts several examples of a PRAD OE, according to some embodiments of the present disclosure, in which an orientation angle of cell structures of the PRAD OE is different;

[0017] FIG. 7 schematically depicts several examples of a PRAD OE, according to some embodiments of the present disclosure, in which a lateral dimension of cell structures of the PRAD OE is different;

[0018] FIG. 8 schematically illustrates a cross-sectional side view and a top view of an image sensor, according to some embodiments of the present disclosure;

[0019] FIG. 9 schematically illustrates a cross-sectional side view and a top view of an image sensor, according to some embodiments of the present disclosure;

[0020] FIG. 10 schematically illustrates a cross-sectional side view and a top view of an image sensor, according to some embodiments of the present disclosure;

[0021] FIG. 11 schematically depicts several examples of a PRAD OE, according to some embodiments of the present disclosure, in which a perimeter of the PRAD OE is shrunk relative to a perimeter of a pixel cell;

[0022] FIGS. 12-15 each schematically illustrate several examples of an image sensor, according to some embodiments of the present disclosure;

[0023] FIG. 16 schematically depicts an example of a PRAD OE applied to a 2x2 array of pixel cells, according to some embodiments of the present disclosure;

[0024] FIG. 17 schematically depicts another example of a PRAD OE applied to a 2x2 array of pixel cells, in accordance with some embodiments of the present disclosure;

[0025] FIG. 18 is a flowchart illustrating a method of manufacturing an image sensor, in accordance with some embodiments of the present disclosure;

[0026] FIGS. 19A-19L each schematically illustrate a cross-sectional side view of a device resulting from respective steps of a process in which the method of FIG. 2 is applied to manufacture the image sensor of FIG. 14.

[0027] Note that, in the embodiments described below, the same reference numbers are used in different drawings to indicate the same or similar parts or parts having the same function, and repeated description thereof will be omitted. In this specification, like reference numbers and letters are used to indicate like items, and thus once an item is defined in one drawing, that item need not be discussed further in subsequent drawings.

[0028] For ease of understanding, the positions, sizes, ranges, etc. of structures shown in the drawings, etc. are not necessarily to scale. Thus, the disclosed application is not limited to the positions, sizes, ranges, etc. disclosed in the drawings, etc. Further, the drawings are not necessarily drawn to scale and certain features can be exaggerated in order to illustrate a particular aspect. DETAILED DESCRIPTION

[0029] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. Note that the relative arrangement, numerical expressions, and numerical values of the components and steps listed in these embodiments are not limiting to the scope of the present disclosure unless otherwise specified.

[0030] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application or uses, nor is it intended to limit the application of the principles elucidated herein to the specific embodiments illustrated. The skilled person will understand that they are merely an illustration of exemplary ways in which the disclosure can be implemented, rather than an exhaustive way.

[0031] Techniques, methods, and devices known to those of ordinary skill in the art can not be discussed in detail, but should be considered part of the specification where appropriate.

[0032] Si-based CMOS image sensors exhibit much lower quantum efficiency (QE) at wavelengths longer than the visible spectrum, for example much lower QE in the near-infrared (NIR) spectrum due to the low absorption coefficient of Si for NIR wavelengths.

[0033] To improve the QE of an image sensor, techniques are needed that allow the light-absorbing elements to absorb more light. One approach is to increase the vertical dimension of the light-absorbing elements. However, for the most commonly used Si-based image sensors, there are limitations to increasing the thickness of Si due to factors such as difficulties in forming deep photodiodes (PDs) and providing sufficient isolation between pixels, as well as limitations in pixel design and process technology. Thus, additional approaches to increase the light absorption of the light-absorbing elements of an image sensor are desired.

[0034] The present disclosure provides an image sensor with improved QE, in which a pseudo-random array diffractive optical element (PRADOE) is provided to cause multi-order diffraction of light impinging on (or otherwise received by) the image sensor, such that light that would have been normally incident on the light-absorbing elements of the image sensor (i.e., with an incident angle of 0°) has an increased incident angle, thereby traveling a longer path before being absorbed within the light-absorbing elements. In this way, more light absorption occurs in the light-absorbing elements, thereby improving the QE of the image sensor.

[0035] Various embodiments of the image sensor of the present disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that an actual image sensor can also include other components, but in order not to obscure the gist of the present application, these other components are not discussed herein and are not shown in the accompanying drawings.

[0036] FIG. 1 illustrates an image sensor 100 according to some embodiments of the present disclosure. The image sensor 100 includes at least one pixel cell, with one pixel cell depicted in FIG. 1 as an example. Each pixel cell of the image sensor 100 includes a light-absorbing element 112. The light-absorbing element 112 can be, for example, a photodiode, or any other suitable opto-electric conversion element capable of converting received light into an electrical signal. In some embodiments, the light-absorbing element is formed within a substrate. In some other embodiments, the light-absorbing element is formed within an epitaxial layer formed on a substrate. Thus, the portion indicated by 110 in FIG. 1 can be a substrate or an epitaxial layer.

[0037] The substrate includes, but is not limited to, a semiconductor substrate, and in some embodiments, the substrate can include any suitable semiconductor material for forming an image sensor, such as a unary semiconductor material (e.g., Si, germanium (Ge), etc.), a compound semiconductor material (e.g., silicon carbide (SiC), silicon-germanium (SiGe), gallium arsenide (GaAs), gallium phosphide (GaP), indium phosphide (InP), indium arsenide (InAs), and / or indium antimonide (InSb), etc.), or a combination thereof. In some embodiments, part or all of the substrate can be doped with an impurity element at a certain concentration, for example, doped with an N-type or P-type impurity. Those skilled in the art can understand that the substrate is not particularly limited herein and can be specifically selected according to the actual application.

[0038] An epitaxial growth process can utilize the principle of two-dimensional structure similar nucleation at grain boundaries to grow single-crystalline layers with complete lattice and possibly different impurity doping and thickness along the original crystal orientation on a single-crystalline substrate. The single-crystalline layer obtained by epitaxial growth is referred to as an epitaxial layer, which can have the same or different conductivity type relative to the original single-crystalline substrate. Epitaxy includes homoepitaxy and heteroepitaxy, where the epitaxial layer obtained by homoepitaxy can be the same as the substrate material (e.g., epitaxial Si on Si substrate, etc.), and the epitaxial layer obtained by heteroepitaxy can be different from the substrate material (e.g., epitaxial GaAs on Si substrate, etc.). Epitaxial growth processes can include vapor phase epitaxy, liquid phase epitaxy, molecular beam epitaxy, etc.

[0039] As shown in FIG. 1, each of the at least one pixel cell of the image sensor 100 further includes a pseudo-random array diffractive optical element (PRADOE) 120 located in front of the light-receiving side of the photosensitive element 112. Note that the term “front” is used herein with respect to the optical path, e.g., “A” is in front of “B” means that light reaches “A” before it reaches “B”. It is known in the art that for a front-side illumination (FIS) image sensor, the light-receiving side of the image sensor is its front side, while for a back-side illumination (BIS) image sensor, the light-receiving side of the image sensor is its back side.

[0040] The PRADOE 120 includes an array of unit structures oriented at pseudo-random angles. “Oriented at pseudo-random angles” involves rotating the individual unit structures in the array differently so as to spatially modulate the diffraction angles. As a result, the angles at which the individual unit structures are rotated appear to be substantially different, but are not truly randomly distributed. Such an angle distribution is referred to herein as “pseudo-random angles”, the array of unit structures having such an angle distribution is referred to as “pseudo-random array”, and the diffractive optical element provided with such an array is referred to as “pseudo-random array diffractive optical element”.

[0041] In some embodiments, the array of unit structures is a periodic array of unit structures. For example, with reference to FIG. 2, the period of the periodic array of unit structures is denoted by P, which refers to the spacing between the centers of adjacent unit structures of the PRADOE 120. For a periodic array of unit structures, the spacing between the centers of every two adjacent unit structures of the PRADOE 120 is substantially equal to the period P. As used herein, the word “substantially” means to include any minor variations caused by design or manufacturing defects, tolerances of devices or components, environmental influences, and / or other factors. Each unit structure of the PRADOE 120 has a length L, a width W, and a thickness D. For a unit structure located at a coordinate (x i , y i ) in a two-dimensional xy-plane (e.g., parallel to a horizontal cross-section of the image sensor 100), it has an orientation angle of θ i . The unit structures of the PRADOE can be, for example, cuboids, but can also take other suitable forms that have a directionality in the xy-plane such that the orientation makes sense, such as, but not limited to, prisms and elliptic cylinders.

[0042] The PRADOE 120 is configured to cause a multi-stage diffraction of light to be received by the light sensing element 112, for example as shown in FIG. 3. The central diffraction angle is determined by the period P. In some embodiments, the period P is determined based on the wavelength of light that the pixel cell is configured to sense. This is because the same diffraction angle can be too small for short wavelengths to sufficiently increase the light incidence angle, thereby not significantly increasing the light absorption, and can be too large for long wavelengths such that the light will travel to a neighboring pixel cell, thereby undesirably increasing cross-talk. In practice, an acceptable diffraction angle can first be determined based on the target wavelength (mainly considering absorption and cross-talk), and then the period P can be determined through simulation based on the determined diffraction angle.

[0043] Periodic arrays have good simulation solvability and suppressed cross-talk. However, it is noted that the array of unit structures does not necessarily have to be a periodic array, but can also be a partially or fully aperiodic array. That is, the spacing between the centers of adjacent unit structures of the PRADOE 120 can vary across the array of unit structures. In this way, the array of unit structures can have a plurality of different spacings, and these spacings can also be determined through simulation based on the determined diffraction angle. For example, FIG. 5A shows an example of a PRADOE for a single pixel cell having two different spacings, where the spacing of the first and fourth rows of unit structures is smaller than the spacing of the second and third rows of unit structures. FIG. 5A shows a partially aperiodic array, where some unit structures have the same spacing as other unit structures. In contrast, in a fully aperiodic array, the spacing between the centers of any two adjacent unit structures can be different.

[0044] The disclosure is mainly exemplified with a periodic array of unit structures for the purpose of illustration and not limitation, but it can be understood that the "period P" described herein means "the spacing between the centers of adjacent unit structures" for aperiodic array, which can not be the same.

[0045] In some embodiments, the size of a pixel unit is a multiple of the period P. For example, FIG. 5B shows several examples of PRADOEs for a single pixel unit with different periods P, in which the period P of the left PRADOE is one third of the size of the pixel unit, the period P of the middle PRADOE is one fourth of the size of the pixel unit, and the period P of the right PRADOE is one fifth of the size of the pixel unit. This facilitates providing the period continuity of PRADOEs between pixel units.

[0046] Many applications can require image sensors to be able to detect multiple wavelengths or wavelength ranges. Therefore, optimizing an image sensor for a single wavelength can not enable the image sensor to be suitable for a wide range of wavelengths at the same time, because different wavelengths of light have different diffraction angles after passing through the same diffractive optical element. In other words, for example, if a single diffraction angle is set by optimization for a center wavelength of a wavelength range in the NIR range, the light of the wavelengths in this wavelength range that are longer than the center wavelength (e.g., near the upper boundary of the wavelength range) will cause more crosstalk than the light of the center wavelength when both are diffracted at the single diffraction angle, while the light of the wavelengths in this wavelength range that are shorter than the center wavelength (e.g., near the lower boundary of the wavelength range) will harvest less increase in absorption than the light of the center wavelength when both are diffracted at the single diffraction angle, thereby reducing the imaging quality of the image sensor. For this reason, in some embodiments, the pseudo-random angles of the unit structures of the PRADOE 120 can be configured so that different levels of multi-level diffraction have different diffraction angles. That is, although the center diffraction angle has been determined by the period P, the variation of the diffraction angles of the levels can still be created by the pseudo-random angles of the unit structures of the PRADOE 120. In this way, a broadband diffraction angle can be achieved for all wavelengths instead of a single diffraction angle, which improves the design freedom of optimizing the image sensor over a wide range of light sources or multiple wavelengths. In practice, the acceptable diffraction angle range can be first determined according to the target wavelength series (mainly considering absorption and crosstalk), then the period P is determined by simulation according to the center diffraction angle of the determined diffraction angle range, and further the pseudo-random angles of the unit structures are determined by simulation according to the diffraction angle variation of the determined diffraction angle range. For example, FIG. 6 shows several examples of PRADOEs for a single pixel unit with different pseudo-random angle distributions.

[0047] Once the period P and the pseudo-random angle are determined, simulations can be performed on the PRADOE with other design parameters (including the length L, the width W, the thickness D, and / or the refractive index n of the PRADOE material) varied while the period P and the pseudo-random angle are kept unchanged at the determined values. The design parameters of the PRADOE with the highest QE are finally selected. Typically, the material selection of the PRADOE is limited by the CMOS process, so in practice, the refractive index n can not be adjusted so freely. Therefore, the refractive index n is typically predetermined, and simulations can be performed on the PRADOE with the length L, the width W, the thickness D varied while the refractive index n, the period P, and the pseudo-random angle are kept unchanged at the determined values.

[0048] In some embodiments, the unit structures of the PRADOE 120 can have the same length and width as each other, as shown in the left part of FIG. 7. This can help reduce the computational burden of the simulation. In some embodiments, at least some of the unit structures of the PRADOE 120 differ from each other in at least one of the length and the width, as shown in the right part of FIG. 7. If the size of each unit structure is allowed to vary individually, it can help the simulation to produce the best results. In addition, for the consideration of process difficulty, the unit structures of the PRADOE 120 can have the same thickness as each other. However, it is still feasible to make the thickness of at least some of the unit structures of the PRADOE 120 different from each other, for example, by multiple etching.

[0049] There are various methods to fabricate the PRADOE 120. In some embodiments, for example, as shown in FIG. 1 and FIG. 3, the PRADOE 120 is formed below the light-receiving surface of the substrate or epitaxial layer 110. For example, the substrate or epitaxial layer 110 can be etched from the light-receiving surface of the substrate or epitaxial layer 110 to form an array of trenches corresponding to the array of unit structures, and then the array of trenches is filled with a PRADOE material. The PRADOE material can be any suitable material with a different refractive index from the substrate or epitaxial layer 110.

[0050] In other embodiments, for example, as shown in FIG. 4, the PRADOE 120 is formed above the light-receiving surface of the substrate or epitaxial layer 110. For example, a layer of material other than the PRADOE material (e.g., used as a sacrificial layer) can be formed on top of the light-receiving surface of the substrate or epitaxial layer, and the layer of material is etched to form an array of trenches corresponding to the array of unit structures, and then the array of trenches is filled with the PRADOE material. The PRADOE material can be any suitable material with a different refractive index from the sacrificial layer. For another example, a layer of PRADOE material can be formed on top of the light-receiving surface of the substrate or epitaxial layer, and then the layer of PRADOE material is etched to form the array of unit structures.

[0051] The PRADOE can include one or more materials. In some embodiments, the PRADOE includes a single layer or multiple layers of anti-reflection (AR) material. As an example, the AR material can include, but is not limited to, silicon oxide (e.g., Si02), silicon nitride (e.g., Si3N4), silicon oxynitride (e.g., Si3N4O), or a combination thereof. In some embodiments, the PRADOE includes a first layer of passivation material and a second layer of AR material over the first layer. The passivation material can help to passivate defects (e.g., dangling bonds) generated by etching used to form the PRADOE, which can otherwise become dark sources leading to severe noise in the image sensor. For example, the passivation material can be a negatively charged material such as hafnium oxide (e.g., Hf02) or aluminum oxide (e.g., AI2O3). These negatively charged species can induce positive charges in the defect regions covered by them, which trap the electrons generated in the defect regions, thereby preventing dark sources from entering the light sensing elements.

[0052] There are other cases where two or more materials are used in the PRADOE. For example, it can sometimes be difficult to fill the etched portion with a single material depending on the depth and / or width of the etched portion. In particular, when using techniques such as atomic layer deposition (ALD) to fill the etched portion, the deposition rate can be very low, leading to longer processing times. In such cases, the etched portion can be partially filled with one material using ALD, and then the remaining portion can be filled with another material using a deposition process with a higher deposition rate.

[0053] FIG. 8 illustrates an image sensor 200 including a substrate or epitaxial layer 210 having light sensing elements 212 and a PRADOE 220, in accordance with some embodiments of the present disclosure. The image sensor 200 differs from the image sensor 100 only in that the PRADOE 220 includes two layers of material. Specifically, the PRADOE 220 includes a first layer 222 and a second layer 224 over the first layer 222. As described above, the first layer 222 can include a passivation material, while the second layer 224 can include an AR material, or alternatively, the first layer 222 can be formed using a first deposition process with one material, while the second layer 224 can be formed using a second deposition process with another material having a higher deposition rate than the first deposition process. It can also be understood that the PRADOE of the present disclosure is not limited to a single layer or a two-layer structure, but can have more layers.

[0054] To avoid crosstalk between pixel cells, in some embodiments, the pixel cell further includes an isolation structure at least partially surrounding the photosensitive element. The isolation structure can extend along part or all of the depth of the photosensitive element, thereby suppressing crosstalk as well as blooming. The PRADOE can include the same material as the isolation structure or a different material from the isolation structure, and the same or different etching and filling processes can be employed for the PRADOE and the isolation structure. For example, the isolation structure can be formed using conventional techniques in the art for shallow trench isolation (STI) or deep trench isolation (DTI), and thus can use conventional materials for STI or DTI, such as silicon oxide (e.g., SiO2), silicon nitride (e.g., Si3N4), silicon oxynitride (e.g., Si3N4O), or a combination thereof. In some embodiments, a layer of passivation material as described above can also be formed between the isolation trench and the isolation structure to prevent defect generation in the etched portion from dark sources.

[0055] In some embodiments, the spacing between the center of the isolation structure and the center of the unit structure of the PRADOE closest to the isolation structure can be the same as the period P. This can improve the periodicity of the diffraction pattern to some extent. The worse the periodicity, the wider the range of diffraction angles of light. Too wide a range of diffraction angles of light can cause an increase in crosstalk between adjacent pixel cells.

[0056] FIG. 9 shows an image sensor 300 including a substrate or epitaxial layer 310 having photosensitive elements 312 and a PRADOE 320, according to some embodiments of the present disclosure. The image sensor 300 differs from the image sensor 100 only in that the pixel cell of the image sensor 300 further includes an isolation structure 330 surrounding the photosensitive elements 312 and extending along part of the depth of the photosensitive elements 312. Specifically, the isolation structure 330 can be formed by etching the substrate or epitaxial layer 310 to form an isolation trench surrounding the photosensitive elements 312, and then filling the isolation trench with an isolation material. For example, the etching can be performed from the light-receiving side, e.g., for a BSI image sensor, the etching can be backside etching. In some cases, the etching for the isolation structure 330 and the etching for the PRADOE 320 can both be performed from the light-receiving side. Furthermore, the isolation trench for the isolation structure 330 and the array of trenches for the PRADOE 320 can be filled with the same material in the same deposition process.

[0057] FIG. 10 illustrates an image sensor 300’ that includes a substrate or epitaxial layer 310 with photosensitive elements 312, PRADOEs 320, and isolation structures 330’ that surround the photosensitive elements 312, in accordance with some embodiments of the present disclosure. The image sensor 300’ differs from the image sensor 300 only in that the isolation structures 330’ extend along the entire depth of the photosensitive elements 312. In particular, the isolation structures 330’ can also be formed by etching the substrate or epitaxial layer 310 to form isolation trenches that surround the photosensitive elements 312, and then filling the isolation trenches with an isolation material. For example, the etching can be performed from the side opposite the light-receiving side, e.g., for a BSI image sensor, the etching can be front-side etching. Further, the etching can be performed partially from the light-receiving side and partially from the opposite side.

[0058] It can be appreciated that in some examples of the above cases where the PRADOEs are formed above the light-receiving surface of the substrate or epitaxial layer, the etching for the isolation structures can include etching of a sacrificial layer, such that the upper surface of the isolation structures is aligned with the upper surface of the PRADOEs.

[0059] In some embodiments, the PRADOEs are shrunk relative to the perimeter of the pixel cells. For example, as shown in FIG. 11, where both pixel cells with isolation structures and pixel cells without isolation structures are illustrated, the area of the PRADOEs does not fill the entire area of the pixel cells, but is applied in a shrunk form. This can be process-friendly, as some people can want the PRADOEs to be separated from the isolation structures for ease of manufacturing. Typically, it is desirable for the PRADOEs to be laterally aligned with the photosensitive elements. For example, the center of the PRADOEs can be laterally aligned with the center of the photosensitive elements. In some cases, it can be desirable for the area of the PRADOEs to be substantially the same as or larger than the area of the photosensitive elements. However, this is not always necessary, as the pixel cells typically include microlenses that can converge the received light, such that the beam size can be shrunk within the area of the PRADOEs.

[0060] FIGS. 12-15 each schematically illustrate several examples of image sensors, in accordance with some embodiments of the present disclosure. Although these figures illustrate BSI image sensors as examples, this is not meant to be any limitation on the present disclosure. It can be appreciated that the teachings of the present disclosure are also applicable to FIS image sensors. For example, the PRADOEs can be located in front of the light-receiving side of the photosensitive elements of a FIS image sensor, but typically behind the metal wiring layers of a FIS image sensor.

[0061] Figure 12 illustrates an image sensor 400, where a top view shows a 2x2 array of pixel cells, and a cross-sectional side view shows two adjacent pixel cells in the 2x2 array, in accordance with some embodiments of the present disclosure. As shown in Figure 12, a pixel cell of the image sensor 400 includes a microlens 401, a color filter 402, a light shield 403 (e.g., in the form of a metal mesh), an AR layer 404 (e.g., in the form of a composite layer having multiple sub-layers), a PRADOE 405, a photosensitive element 406 in a substrate 407, an active element 408 (e.g., a transistor), and a metal interconnect structure 409. Note that the bottom-most sub-layer of the AR layer 404 can be formed in the same deposition process with the same material as the PRADOE 405.

[0062] Figure 13 illustrates an image sensor 500, where a top view shows a 2x2 array of pixel cells, and a cross-sectional side view shows two adjacent pixel cells in the 2x2 array, in accordance with some embodiments of the present disclosure. As shown in Figure 13, a pixel cell of the image sensor 500 includes a microlens 501, a color filter 502, a light shield 503 (e.g., in the form of a metal mesh), an AR layer 504 (e.g., in the form of a composite layer having multiple sub-layers), a PRADOE 505, a photosensitive element 506 in a substrate 507, an active element 508 (e.g., a transistor), and a metal interconnect structure 509. The image sensor 500 also includes an isolation structure 510 that separates the pixel cells from each other. The isolation structure 510 extends along a partial depth of the photosensitive element 506. Note that the bottom-most sub-layer of the AR layer 504 can be formed in the same deposition process with the same material as the PRADOE 505 and the isolation structure 510.

[0063] Figure 14 illustrates an image sensor 600, where a top view shows a 2x2 array of pixel cells, and a cross-sectional side view shows two adjacent pixel cells in the 2x2 array, in accordance with some embodiments of the present disclosure. As shown in Figure 14, a pixel cell of the image sensor 600 includes a microlens 601, a color filter 602, a light shield 603 (e.g., in the form of a metal mesh), an AR layer 604 (e.g., in the form of a composite layer having multiple sub-layers), a PRADOE 605, a photosensitive element 606 in a substrate 607, an active element 608 (e.g., a transistor), a metal interconnect structure 609, and an isolation structure 610. The isolation structure 610 extends along a partial depth of the photosensitive element 606. In particular, the PRADOE 605 and the isolation structure 610 include a first layer of passivation material and a second layer of AR material thereon. Note that the bottom-most sub-layer of the AR layer 604 can be formed with the same material as the PRADOE 605 and the isolation structure 610 in the same deposition process as the PRADOE 605 and the isolation structure 610.

[0064] Figure 15 illustrates an image sensor 700, where a top view shows a 2x2 array of pixel cells, and a cross-sectional side view shows two adjacent pixel cells in the 2x2 array, in accordance with some embodiments of the present disclosure. As shown in Figure 15, a pixel cell of the image sensor 700 includes a microlens 701, a color filter 702, a light shield 703 (e.g., in the form of a metal mesh), an AR layer 704 (e.g., in the form of a composite layer having multiple sub-layers), a PRADOE 705, a photosensitive element 706 in a substrate 707, an active element 708 (e.g., a transistor), a metal interconnect structure 709, and an isolation structure 710. The isolation structure 710 extends along an entire depth of the photosensitive element 706. In particular, the PRADOE 705 and the isolation structure 710 include a first layer of passivation material and a second layer of AR material thereon. Note that the bottom-most sub-layer of the AR layer 704 can be formed with the same material as the PRADOE 705 and the isolation structure 710 in the same deposition process as the PRADOE 705 and the isolation structure 710. More particularly, the formation of the PRADOE 705 involves backside etching, while the formation of the isolation structure 710 involves frontside etching.

[0065] In the present disclosure, an image sensor can be a near-infrared (NIR) image sensor. For example, an image sensor can be a NIR BIS image sensor, such as a Si-based NIR BIS image sensor. As mentioned above, conventional Si-based image sensors have poor performance in the NIR spectrum. In contrast, a NIR image sensor according to the present disclosure has significantly improved QE due to the enhanced light absorption brought by the PRADOE.

[0066] Additionally, the image sensor can be a visible light image sensor. For example, the image sensor can be a visible light BIS image sensor, such as a Si-based visible light BIS image sensor. While conventional Si-based image sensors already have good performance in the visible light spectrum, the QE of the visible light image sensor according to the present disclosure is further improved due to the enhanced light absorption by the PRADOE.

[0067] While the above figures mainly depict one pixel cell or a 2x2 array of pixel cells, the image sensor can have two, three, four, five, or more pixel cells. In some embodiments, the image sensor includes a first pixel cell configured to sense light of a first wavelength and including a first PRADOE, and a second pixel cell configured to sense light of a second wavelength different from the first wavelength and including a second PRADOE. The first and second PRADOEs can differ from each other in at least one of the following: length of the cell structure, width of the cell structure, orientation angle of the cell structure, pitch (e.g., period P) between centers of adjacent cell structures. Note that the first and second PRADOEs can also have different thicknesses or refractive indices, but changing both parameters can make the process more complicated. As mentioned above, the parameters of the PRADOE for a pixel cell can be optimized for the target wavelength(s) of the pixel cell.

[0068] For example, FIG. 16 schematically depicts an example of PRADOEs applied to a 2x2 array of pixel cells similar to a Bayer array (two green pixel cells G, one red pixel cell R, and one blue pixel cell B) according to some embodiments of the present disclosure. The PRADOEs for the green pixel cells G, blue pixel cells B, and red pixel cells R have different periods (P), different lateral dimensions (L, W), and different pseudo-random angle distributions. For various types of pixel cell arrays, the optimization of the PRADOE for each pixel cell can be performed. Specifically, without PRADOEs, pixel cells designed for different wavelengths or wavelength ranges can have different levels of absorption and cross-talk. When PRADOEs are applied, they enhance the absorption, thereby improving the signal-to-noise ratio, while the increase in cross-talk is controlled to be small and acceptable, thus not substantially degrading the image contrast and color accuracy. Therefore, different types of PRADOEs can be applied to each pixel cell to suit its specific characteristics.

[0069] In some embodiments, the image sensor includes: a first pixel cell configured to sense light of a first wavelength; and a second pixel cell configured to sense light of a second wavelength different from the first wavelength, wherein the first pixel cell includes a PRADOE but the second pixel cell does not include a PRADOE. In some examples, the first pixel cell has a smaller absorption of the light of the first wavelength than the second pixel cell has of the light of the second wavelength. In some examples, when the light of the second wavelength has a same angle of incidence as the light of the first wavelength, a crosstalk caused by the light of the second wavelength between the second pixel cell and its neighboring pixel cell is greater than a crosstalk caused by the light of the first wavelength between the first pixel cell and its neighboring pixel cell.

[0070] For example, FIG. 17 schematically depicts another example of PRADOEs applied to a 2x2 array of pixel cells (two green pixel cells G, one red pixel cell R, and one blue pixel cell B) similar to a Bayer array, according to some embodiments of the present disclosure. Only the green pixel cells G and the red pixel cell R are provided with PRADOEs. No PRADOE is provided for the blue pixel cell B because, for example, the absorption coefficient of Si for blue light is already high, and thus there is no need to further enhance the absorption of blue light. In fact, if the absorption of blue light is too much better than the absorption of other colors of light, it can cause color cast, thereby reducing the imaging quality. In fact, PRADOEs can be selectively applied to certain types of pixel cells, for example, where an increase in QE is needed. Generally, PRADOEs can be applied when the increase in QE is significant while the increase in crosstalk is small and acceptable.

[0071] In certain cases, PRADOEs can be selectively not applied to certain types of pixel cells, for example, where a reduction in crosstalk is needed. The reduction in crosstalk can advantageously improve image contrast. In some examples, PRADOEs can be provided only for green pixel cells G, but not for red pixel cells R and blue pixel cells B.

[0072] Accordingly, in another aspect, the present disclosure provides a method for manufacturing an image sensor. As shown in FIG. 18, the method 800 for manufacturing an image sensor includes: at step S802, providing a substrate having a photosensitive element; and at step S804, forming a pseudo-random array diffractive optical element (PRADOE) in front of a light-receiving side of the photosensitive element, the PRADOE including an array of unit structures oriented at pseudo-random angles. The PRADOE is configured to cause a multi-order diffraction of light to be received by the photosensitive element. In some embodiments, the array of unit structures can be a periodic array of unit structures. In some other embodiments, the array of unit structures can be a (partially or fully) aperiodic array of unit structures.

[0073] In some embodiments, the light sensing elements are formed within the substrate, and forming the PRAD OE includes one of: etching the substrate from a light receiving surface of the substrate to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with the first material; or forming a second material layer over the light receiving surface of the substrate, etching the second material layer to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with the first material; or forming a first material layer over the light receiving surface of the substrate, and etching the first material layer to form the array of unit structures.

[0074] In some embodiments, the light sensing elements are formed within an epitaxial layer formed on the substrate, and forming the PRAD OE includes one of: etching the epitaxial layer from a light receiving surface of the epitaxial layer to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with the first material; or forming a second material layer over the light receiving surface of the epitaxial layer, etching the second material layer to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with the first material; or forming a first material layer over the light receiving surface of the epitaxial layer, and etching the first material layer to form the array of unit structures.

[0075] For example, the first material can be any of the above-described materials for the PRAD OE, and the second material can be any suitable material having a different refractive index than the PRAD OE. In some examples, the second material layer can be formed by epitaxial growth or deposition.

[0076] In some embodiments, filling the array of trenches with the first material includes first filling the array of trenches with a passivation material, and then filling the array of trenches with an anti-reflective material. For example, the passivation material is a negatively charged material.

[0077] In some embodiments, filling the array of trenches with the first material includes filling the array of trenches with multiple materials at different deposition rates.

[0078] In some embodiments, the method 800 further includes etching the substrate or the epitaxial layer to form an isolation trench at least partially surrounding the light sensing element, and filling the isolation trench with a third material to form an isolation structure. For example, the isolation structure extends along part or all of the depth of the light sensing element. The third material can be any suitable material capable of providing isolation between pixel units, such as those used for STI or DTI. The third material can or can not be the same as the first material.

[0079] In some embodiments, the etching for the isolation trench and the etching for the array of trenches are both performed from the light receiving side. For example, the isolation trench and the array of trenches are filled with the same material in the same deposition process. In some embodiments, the etching for the isolation trench is performed from a side opposite the light receiving side.

[0080] In some embodiments, the spacing between the center of the isolation structure and the center of the unit structure of the PRADOE closest to the isolation structure is the same as the spacing between the centers of adjacent unit structures of the PRADOE.

[0081] In some embodiments, the method 800 further comprises forming an anti-reflective layer over the PRADOE, wherein the anti-reflective layer or a bottommost sub-layer of the anti-reflective layer is formed in the same deposition process with the same material as the PRADOE.

[0082] In some embodiments, the unit structures of the PRADOE are cuboids. In some embodiments, at least some of the unit structures of the PRADOE differ from each other in at least one of length and width. In some embodiments, the unit structures of the PRADOE have the same length and width. In some embodiments, the spacing between the centers of adjacent unit structures of the PRADOE is determined based on the wavelength of light that the pixel unit is configured to sense.

[0083] In some embodiments, the light-sensing element comprises a first light-sensing element configured to sense light of a first wavelength and a second light-sensing element configured to sense light of a second wavelength different from the first wavelength, wherein the PRADOE comprises a first PRADOE for the first light-sensing element and a second PRADOE for the second light-sensing element, wherein the first PRADOE and the second PRADOE differ from each other in at least one of: length of unit structures, width of unit structures, orientation angle of unit structures, spacing between centers of adjacent unit structures.

[0084] In some embodiments, the light-sensing element comprises a first light-sensing element configured to sense light of a first wavelength and a second light-sensing element configured to sense light of a second wavelength different from the first wavelength, wherein the PRADOE is formed in front of the first light-sensing element but not in front of the second light-sensing element.

[0085] In some embodiments, the image sensor is a near-infrared back-illuminated image sensor.

[0086] Embodiments of the method 800 are similar to those described above for the image sensor, and reference can be made to the various embodiments of the image sensor of the present disclosure, and thus repeated description is omitted here.

[0087] For non-limiting illustration purposes, FIGS. 19A-19L each schematically illustrate a cross-sectional side view of a device resulting from respective steps of a process in which the method of FIG. 2 is applied to manufacture the image sensor of FIG. 14. Two adjacent pixel units are exemplarily illustrated.

[0088] As shown in FIG. 19A, a substrate 901 is provided. As shown in FIG. 19B, a light sensing element 902 (such as a PD) is formed within the substrate 901. As shown in FIG. 19C, peripheral circuitry 903 such as transistors and metal wiring is formed over the light sensing element 902. As shown in FIG. 19D, the substrate 901 having the light sensing element 902 and peripheral circuitry 903 formed thereon is flipped over and a thinning process is performed. As shown in FIG. 19E, the backside of the substrate 901 (now looking like upside down) is etched to form a periodic array of trenches 904 corresponding to the periodic array of cell structures of the PRAD OE designed for the two pixel cells. As shown in FIG. 19F, the backside of the substrate 901 is further etched to form an isolation trench 905 around each pixel cell. As shown in FIG. 19G, the trenches are partially filled with a passivation layer 906. As shown in FIG. 19H, an AR layer 907 fills the rest of the trenches. The portion of the AR layer 907 within the periodic array of trenches forms the PRAD OE, the portion of the AR layer 907 within the isolation trench forms the isolation structure, and the rest of the AR layer 907 forms the bottom layer in the bilayer AR structure. As shown in FIG. 191, an additional AR layer 908 is deposited as the top layer in the bilayer AR structure. As shown in FIG. 19J, a metal mesh 909 is formed over the bilayer AR structure to serve as a light shield to avoid cross talk. As shown in FIG. 19K, a color filter 910 is installed for each pixel cell, which passes light of wavelengths within the design wavelength range of the respective pixel cell and blocks light of wavelengths outside the design wavelength range. As shown in FIG. 19L, a microlens 911 is installed for each pixel cell for converging the received light rays to avoid cross talk.

[0089] The words "left," "right," "front," "back," "top," "bottom," "over," "under," "upper," "lower," and the like in the description and the claims, if any, are used for descriptive purposes and not necessarily for describing permanent relative positions. It is to be understood that the words so used are interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in other orientations than those illustrated or otherwise described herein. For example, if the device is inverted, then the features originally described as above other features can be described as below the other features. The device can also be oriented in other ways (rotated 90 degrees or in other orientations) and the relative spatial relationships would be correspondingly interpreted.

[0090] In the description and claims, when an element is referred to as being "on", "attached", "connected", "coupled", or "in contact" with another element, it can be directly on, attached, connected, coupled, or in contact with the other element, or one or more intervening elements can also be present. In contrast, when an element is referred to as being "directly on", "directly attached", "directly connected", "directly coupled", or "directly in contact" with another element, there are no intervening elements present. In the description and claims, a feature that is arranged "adjacent" to another feature can mean that the feature has a portion that overlaps the adjacent feature or a portion that is above or below the adjacent feature.

[0091] As used herein, the word "exemplary" is used to mean "serving as an example, instance, or illustration." Any implementation of the described implementations described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other implementations. Furthermore, the disclosure is not intended to be limited to the implementations shown in the text, figures, or examples described herein. Moreover, the disclosure is not intended to be limited to the aspects presented and / or implied by the text, figures, or examples described herein. As used herein, the word "substantially" means including any minor variations as a result of design, manufacturing, and / or material tolerances, environmental impacts, and / or other factors. The word "substantially" also allows for differences that are within normal manufacturing and / or material tolerances, environmental impacts, and / or other factors that can exist in an actual implementation.

[0092] Additionally, the terms "first", "second", and / or like terms are used herein only to facilitate discussion, and thus, are not intended to limit the scope of the disclosure. For example, unless otherwise specifically stated, the terms "first", "second", and other such numerical terms do not imply a sequence or order to steps, but rather are used for purposes of distinction only. It is also to be understood that the term "comprising" as used herein is intended to indicate that the presence of stated features, integers, steps, components, and / or elements, but not to exclude the presence or addition of one or more other features, integers, steps, components, elements, and / or groups thereof. In the disclosure, the term "provide" is used in a broad sense to encompass all ways of obtaining an object, and thus "providing an object" includes, but is not limited to, "purchasing", "preparing / manufacturing", "arranging / setting", "installing / fitting", and / or "ordering" the object, etc.

[0093] As used herein, the term "and / or," includes any and all combinations of one or more of the associated listed items. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0094] Those skilled in the art will realize that the boundaries between the above described operations merely illustrative. The multiple operations can be combined into a single operation, a single operation can be distributed in additional operations and operations can be executed at least partially overlapping in time. Moreover, alternative embodiments can include a number of instances of a particular operation, and the order of the operations can be altered in other various embodiments. However, other modifications, variations, and alternatives are also possible. The aspects and elements of all such embodiments can be combined in any manner and / or with other embodiments, aspects and elements without departing from the scope of the present disclosure. It will therefore be appreciated that the description and drawings set forth herein do not restrict the scope of the disclosure. The disclosure is susceptible to numerous modifications, adaptations, and variations. Specific embodiments are not intended to limit the scope of the disclosure but rather to provide examples of the disclosure.

[0095] In addition, as used in the specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Further, as used in the specification and the appended claims, the term "or" means "and / or" unless the context clearly dictates otherwise. Furthermore, the words "comprise," "comprising," "include," "including," and the like, when used in the present specification and in the following claims, are intended to mean "including but not limited to," unless otherwise explicitly limited by the context.

[0096] While certain embodiments of the disclosure have been described herein, the present disclosure is not limited to these embodiments. Various modifications, alterations, and permutations of the disclosed embodiments can be made and equivalents thereof employed without departing from the scope of the disclosure. Accordingly, the foregoing description is intended to be illustrative only and not limiting of the disclosure.

Claims

1. An image sensor comprising at least one pixel cell, wherein, Each of the at least one pixel cell comprises: a light sensing element; and a pseudo-random array diffractive optical element (PRADOE) located in front of a light receiving side of the light sensing element and comprising an array of unit structures oriented at pseudo-random angles, wherein the PRADOE is configured to cause multi-order diffraction of light to be received by the light sensing element.

2. The image sensor of claim 1, wherein, The pseudo-random angles are configured such that different orders of the multi-order diffraction have different diffraction angles.

3. The image sensor of claim 1, wherein: the light sensing element is formed within a substrate, and the PRADOE is formed over a light receiving surface of the substrate; or the light sensing element is formed within an epitaxial layer formed on a substrate, and the PRADOE is formed over a light receiving surface of the epitaxial layer.

4. The image sensor of claim 1, wherein: the light sensing element is formed within a substrate, and the PRADOE is formed under a light receiving surface of the substrate; or the light sensing element is formed within an epitaxial layer formed on a substrate, and the PRADOE is formed under a light receiving surface of the epitaxial layer.

5. The image sensor of claim 1, wherein, The unit structures of the PRADOE are cuboids.

6. The image sensor of claim 1, wherein, At least some of the unit structures of the PRADOE differ from each other in at least one of length and width.

7. The image sensor of claim 1, wherein, The unit structures of the PRADOE have the same length and width as each other.

8. The image sensor of claim 1, wherein, A pitch between centers of adjacent unit structures of the PRADOE is determined based on a wavelength of light the pixel cell is configured to sense.

9. The image sensor of claim 1, wherein, The PRADOE comprises a first layer of passivation material and a second layer of anti-reflective material over the first layer.

10. The image sensor of claim 9, wherein, The passivation material is a negatively charged material.

11. The image sensor of claim 1, wherein, The array of unit structures is a periodic array of unit structures.

12. The image sensor of claim 11, wherein, A size of the pixel cell is a multiple of a pitch between centers of adjacent unit structures of the PRADOE.

13. The image sensor of claim 1, wherein, A perimeter of the PRADOE is shrunk relative to a perimeter of the pixel cell.

14. The image sensor of any one of claims 1-13, wherein, The pixel cell further comprises an isolation structure at least partially surrounding the light sensing element.

15. The image sensor of claim 14, wherein, The isolation structure extends along a portion or all of a depth of the light sensing element.

16. The image sensor of claim 14, wherein, The PRADOE comprises the same material as the isolation structure.

17. The image sensor of claim 14, wherein, The array of unit structures is a periodic array of unit structures, and wherein a pitch between a center of the isolation structure and a center of a unit structure of the PRADOE closest to the isolation structure is the same as a pitch between centers of adjacent unit structures of the PRADOE.

18. The image sensor of claim 1, wherein, Each of the at least one pixel cell comprises: a first pixel cell configured to sense light of a first wavelength and comprising a first PRADOE; and a second pixel cell configured to sense light of a second wavelength different from the first wavelength and comprising a second PRADOE, wherein the first PRADOE and the second PRADOE differ from each other in at least one of: length of unit structures, width of unit structures, orientation angle of unit structures, pitch between centers of adjacent unit structures.

19. The image sensor of claim 1, wherein, The at least one pixel cell includes a first pixel cell configured to sense light of a first wavelength, and the image sensor further includes a second pixel cell configured to sense light of a second wavelength different from the first wavelength, wherein the first pixel cell includes a PRADOE but the second pixel cell does not include a PRADOE.

20. The image sensor of claim 19, wherein, Absorption of light of the first wavelength by the first pixel cell is less than absorption of light of the second wavelength by the second pixel cell.

21. The image sensor of claim 19, wherein, When light of the first wavelength and light of the second wavelength have the same angle of incidence, crosstalk caused by light of the second wavelength between the second pixel cell and its neighboring pixel cells is greater than crosstalk caused by light of the first wavelength between the first pixel cell and its neighboring pixel cells.

22. The image sensor of claim 1, wherein, The image sensor is a near-infrared backside illuminated image sensor.

23. A method for manufacturing an image sensor, comprising: providing a substrate having photosensitive elements; and forming a pseudo-random array diffractive optical element (PRADOE) in front of a light-receiving side of the photosensitive elements, the PRADOE including an array of unit structures oriented at pseudo-random angles, wherein the PRADOE is configured to cause multi-order diffraction of light to be received by the photosensitive elements.

24. The method of claim 23, wherein, The photosensitive elements are formed within the substrate, and forming the PRADOE includes: etching the substrate from a light-receiving surface of the substrate to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with a first material; or forming a second material layer over the light-receiving surface of the substrate, etching the second material layer to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with a first material; or forming a first material layer over the light-receiving surface of the substrate, and etching the first material layer to form the array of unit structures.

25. The method of claim 23, wherein, The photosensitive elements are formed within an epitaxial layer formed on the substrate, and forming the PRADOE includes: etching the epitaxial layer from a light-receiving surface of the epitaxial layer to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with a first material; or forming a second material layer over the light-receiving surface of the epitaxial layer, etching the second material layer to form an array of trenches corresponding to the array of unit structures, and filling the array of trenches with a first material; or forming a first material layer over the light-receiving surface of the epitaxial layer, and etching the first material layer to form the array of unit structures.

26. The method of claim 24 or 25, wherein, Filling the array of trenches with the first material includes filling the array of trenches with a passivation material first, and filling the array of trenches with an anti-reflective material second.

27. The method of claim 26, wherein, The passivation material is a negatively-charged material.

28. The method of claim 24 or 25, wherein, Filling the array of trenches with the first material includes filling the array of trenches with a plurality of materials at different deposition rates.

29. The method of claim 24, further comprising: etching the substrate to form an isolation trench at least partially surrounding the photosensitive elements; filling the isolation trench with a third material to form an isolation structure.

30. The method of claim 25, further comprising: etching the epitaxial layer to form an isolation trench at least partially surrounding the light sensing element; filling the isolation trench with a third material to form an isolation structure.

31. The method of claim 29 or 30, wherein, The etching for the isolation trench and the etching for the array of trenches are both from the light-receiving side.

32. The method of claim 31, wherein, The isolation trench and the array of trenches are filled with the same material in the same deposition process.

33. The method of claim 29 or 30, wherein, The etching for the isolation trench is from the side opposite the light-receiving side.

34. The method of claim 29 or 30, wherein, The isolation structure extends along part or all of the depth of the light sensing element.

35. The method of claim 29 or 30, wherein, The array of unit structures is a periodic array of unit structures, and wherein a pitch between a center of a unit structure of the PRADOE closest to the isolation structure and a center of the isolation structure is the same as a pitch between centers of adjacent unit structures of the PRADOE.

36. The method of claim 23, further comprising forming an anti-reflective layer on the PRAD OE, wherein, The anti-reflective layer or a bottommost sub-layer of the anti-reflective layer is formed with the same material in the same deposition process as the PRADOE.

37. The method of claim 23, wherein, The unit structures of the PRADOE are cuboids.

38. The method of claim 23, wherein, At least some of the unit structures of the PRADOE differ from each other in at least one of length and width.

39. The method of claim 23, wherein, The unit structures of the PRADOE have the same length and width.

40. The method of claim 23, wherein, A pitch between centers of adjacent unit structures of the PRADOE is determined based on a wavelength of light that a pixel unit is configured to sense.

41. The method of claim 23, wherein, The light sensing element includes a first light sensing element configured to sense light of a first wavelength and a second light sensing element configured to sense light of a second wavelength different from the first wavelength, and wherein the PRADOE includes a first PRADOE for the first light sensing element and a second PRADOE for the second light sensing element, wherein the first PRADOE and the second PRADOE differ from each other in at least one of: length of unit structures, width of unit structures, orientation angle of unit structures, pitch between centers of adjacent unit structures.

42. The method of claim 23, wherein, The light sensing element includes a first light sensing element configured to sense light of a first wavelength and a second light sensing element configured to sense light of a second wavelength different from the first wavelength, wherein a PRADOE is formed in front of the first light sensing element but no PRADOE is formed in front of the second light sensing element.

43. The method of claim 23, wherein, The image sensor is a near-infrared backside illuminated image sensor.

44. The method of claim 23, wherein, The array of unit structures is a periodic array of unit structures.

Images