TW200405718A - Photoconductor on active pixel image sensor - Google Patents

Abstract

A MOS or CMOS based photoconductor on active pixel image sensor. Thin layers of semi-conductor material, doped to PIN or NIP photoconducting layers, located above MOS and/or CMOS pixel circuits produce an array of layered photodiodes. Positive and negative charges produced in the layered photodiodes are collected and stored as electrical charges in the MOS and/or CMOS pixel circuits. The present invention also provides additional MOS or CMOS circuits for reading out the charges and for converting the charges into images. With the layered photodiode of each pixel fabricated as continuous layers of charge generating material on top of the MOS and/or CMOS pixel circuits, extremely small pixels are possible with almost 100 percent packing factors. MOS and CMOS fabrication techniques permit sensor fabrication at very low costs. In preferred embodiments all of the sensor circuits are incorporated on or in a single crystalline substrate along with the sensor pixel circuits. Techniques are disclosed for tailoring the spectral response of the sensor for particular applications.

Description

200405718 (1) Description of the invention: 1. Technical field to which the invention belongs The present invention is US Patent Application No. 10 / 〇72,637 (filed on 2/5/2002), 10 / 229,953 (filed on 8/27/2002), 10 / 229,954 (filed on 8/27/2002), 10 / 229,955 (filed on 8/27/2002), 10 / 229,956 (filed on 8/27/2002) and 10 / 371,618 (filed on 2/22/2003) Continuation In Part (CIP). The present invention relates to an image sensor, and more particularly to an image sensor based on MOS and CMOS. 2. Prior Technology A typical electronic image sensor is composed of a large number of tiny photodetector arrays, collectively referred to as a "pixel array". The sensor usually generates a signal representing each pixel (a signal, the amplitude of which is proportional to the intensity of the light received by each detector in the array. The image element in the electronic camera generates a light image of the scene and projects it Until the book wins ^ 丨 丨 u ^ ^ ^ ^. ^ Torikyo descends, and then the electronic image sensor converts the light image into a $ 10,000 丨 | 1_-and lists the signal of the son. The electronic camera is generally Package number of components 'to convert the image to digital ^ k |' The image can be processed by a digital processor for storage, digital transfer or surface; Flying 717. A variety of semiconductor components can be used to obtain the image 'It includes Charge Coupled Device (CCD), Photodiode Array, and ^ 卞 ~ 和 私 何> King-in-the-wood type. The most common electronic image sensor uses Γγγι K ^ CCD detector array converts light into electronic signals. CCD detectors have existed for many years, and their technology has matured and developed well. One of the biggest lack of CCDs is love for production, people *, Other integrated circuit technologies Metal

H: \ HU \ TYS \ ii □ case \ 87767 \ 87767.DOC 200405718 Metal Oxide Semiconductor (MOS) or Complementary Metal Oxide Semiconductor (CMOS) technologies are not compatible. The processing circuit must be placed in a wafer outside the CCD element. Another existing image sensor is based on MOS or CMOS technology. The sensor typically has multiple transistors in each pixel. The most common CMOS sensor design is to have photo-sensing circuitry and active circuitry in each pixel cell. This is called an active pixel sensor. Pixel Sensor, APS). The active circuit is composed of a plurality of transistors interconnected by metal lines. In this way, the area occupied by circuits that are not transparent to visible light cannot be used for light sensing. Therefore, each pixel cell generally contains light-sensing and non-light-sensing circuits. In addition to the circuits combined with each pixel cell, the CMOS sensor also has other digital signal and analog signal processing circuits, such as sample-and-hold amplifiers, analog digital converters, and digital signal processing logic. The circuits are all integrated into a monolithic element. Pixel arrays and other digital and analog circuits are manufactured using the same basic fabrication process. Small cameras using CCD sensors generally consume a large amount of energy (compared to cameras with CMOS sensors) and require high rail-to-rail voltage swings to operate the CCD. This can cause problems with current mobile electronic devices, such as mobile phones and Personal Digital Assistants (PDAs). On the other hand, small cameras using CMOS sensors can provide solutions for energy consumption, but due to the shallow integration of silicon substrates

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC 200405718 (shallow junction) depth and its active transistor circuit occupy the area needed for precious light sensing. Traditional small cameras using CMOS are limited to the above-mentioned CMOS active type The inherent characteristics of the pixel sensor make its light sensing poor.

U.S. Patent Nos. 5,528,043, 5,886,353, 5,998,794, and 6,1 63,030 illustrate conventional techniques for imaging using CMOS circuits. The patent has been granted to the applicant's employer. Patent No. 5,528,043 describes an X-ray detector with a read circuit using a CMOS array on a single wafer. In this example, a separate processor is used to process the images. Patent No. 5,886,353 describes a general pixel structure using a hydrogenated amorphous silicon layer, such as p-i-n or p-n or other derivatives in combination with a CMOS circuit for the pixel array. Patent Nos. 5,998,794 and 6,163,030 describe various methods for making electrical connections to CMOS circuits at the bottom of a pixel. All of the aforementioned U.S. patents are incorporated herein by reference.

Integrating a CMOS or MOS sensor with an external processor will increase complexity and increase production costs. Therefore, an improved image sensing technology is needed, which can provide low cost and high quality compared to previous sensors. , High performance and smaller size image sensor. 3. Summary of the Invention The present invention provides an active pixel image sensor provided with a MOS or CMOS photoconductor. A thin semiconductor material layer doped into a PIN or NIP photoconductive layer and located on a MOS or CMOS pixel circuit creates a layered photodiode array. The positive and negative charges generated in the layered photodiode are collected and stored in a MOS or CMOS pixel circuit. The invention also provides additional

H: \ HU \ TYS · mouth case \ 87767 \ 87767.DOC 200405718

MOS or CMOS circuit to read the charge and convert it into an image. By processing the layered photodiodes of each pixel as a continuous layer of charge generating material on top of MOS and / or CMOS pixel circuits, extremely small pixels may have a packing factor of almost 100%. Using MOS or CMOS manufacturing technology can make the manufacturing cost of the sensor very low. In a preferred embodiment, all the sensor circuits are integrated into or on a single crystalline substrate together with the sensor pixel circuits. The invention also discloses a spectrum response design technique of a sensor suitable for a particular application. For example, in the preferred embodiment, the spectral range of the sensor can be adjusted to cover ultraviolet and near-infrared and visible light ranges, or any portion of that wide range. Some preferred embodiments of the present invention include additional features that minimize cross talk between pixels. This additional feature includes a gate bias transistor in each pixel point to maintain the electrical potential of the charge collection unit of each pixel to be approximately the same. In another embodiment, in order to increase its resistance, carbon is doped in the semiconductor material surrounding the charge collection unit. This embodiment provides a quantum efficiency higher than 50% and an extremely fast response time equal to or even a few hundred microseconds. In a particularly preferred embodiment, the sensor is a 300,000 pixel array (3.2 mm x 2.4 mm, 640 x 480) with 5 micron square pixels, which is compatible with lenses in 1 / 4.5 inch optical format. In the preferred embodiment, the sensor and focusing optics are integrated into a mobile phone camera to allow simultaneous transmission of video images during voice communications. As a result, low-cost cameras are produced in large numbers and can be made very small (for example, smaller than the human eye). The cost of mass production of cameras above 300,000 pixels is planned to be less than $ 10 per unit. Compare

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC 200405718 The preferred embodiment also includes ultraviolet and infrared cameras with high quantum efficiency, fast response time, and = pixel array, such as the high quality of 2 pixels TV (high definition television) format sensor. Fourth, the implementation of the first preferred embodiment of a single-chip camera circuit with a photoconductor on the active pixel sensor, a sensor is ", this: Ming-the preferred embodiment is a photodiode The second component of the array, the! J early wafer camera, the photodiode array is composed of a photoconductive layer on the top of the king-type array. (Applicant calls the P0AP sensor, and the P0AP is ph〇t〇c. 〇nduct〇r 〇n Active

Plxel's dimples, which are photoconductors on active pixels). In this particular sense, 307,200 pixel dots are arranged in a 64 × 480 pixel array, and there is a transparent electrode on the photoconductive layer. The pixel size is 5 micrometers and micrometers, and the filling rate is almost 100%. The size of the active part of the sensor is 3.2mm x 2.4mm. The preferred lens unit is a t-shaped lens with 1/4 · 5 inch optics. One of the preferred applications of this camera is as a mobile phone component in Figs. 1A and B. In Fig.a, the camera is a whole limb of a mobile phone 2a, and the lens is shown in Fig. 4A. In FIG. 1B, the camera 6 is separated from the mobile phone 2B and connected through a 3-pin connector 10. The lens of the camera is shown in Fig. 4B, and reference numeral 8 in the figure shows the cover of the camera. FIG. 1 c is a block diagram, which includes a lens 4, a lens holder 12, an image chip 14, a sensor pixel array 100, a circuit board 16, and a pin connector 10 for displaying the main functions of the camera 4B in the figure . CMOS sensor

HAHUVrYS \ import case \ 87767 \ 87767.DOC -10- 200405718 The sensor pixel array is based on the photoconductive device, reading circuit, reading timing / control circuit, and sensor timing on the active pixel array. / Control electronics and analog digital conversion circuits. The sensor includes: 1. a CMOS-based daylight array, including a 64 × 480 × charge receiver and a 640 × 480 CMOS pixel circuit; and 2. a CMOS reading circuit, including analog digital (At〇 D) Conversion and timing control circuits. This sensor array is described in the US Patent No. 5,886,353 in the technical background (visible light sensor arrays are similar (especially in texts 19 to 24 and Figure 27), which will be incorporated herein by reference. Various sensing A detailed description of the sensor array is also described in the patents listed on page 6 of this specification, which will also be incorporated herein by reference. Heads 2, 3A, 3B, and 3C describe preferred sensors for the mobile phone camera. Characteristics of the array. The general layout of the sensor is shown as 100 in Figure 2. The sensor includes a pixel array 102 and a read and timing / control circuit 104. Figure 3A shows the pixel array One of the layered structures of the five-day element. The sensor strip array is covered with a color filter (collor filter), each day element is coated with only one color filter to limit only the color spectrum One component can be transmitted through. The preferred color filter kit is composed of three types of broadband color filters, with transmission peaks at 450 nm (blue), 550 nm (green), and 630. Nano (red). For blue and green filters, The full width at half maximum of Fenhai color filter is about 50 nanometers. The red light filter generally allows transmission of all light rays in the near infrared. For the application of visible images, infrared cut-off is required ( cut-0ff) filter is used to adjust H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -11-200405718 The peak value of the red light response is 630 nm, and its full width at half maximum is 50 nm. The Kallas are used for visible light sensing applications. As shown in Figure 3C, four pixels form a four-piece set. Two of the four pixels cover a transmission peak at 55 nm < color filter Film, which is called "green light pixels". One pixel covers a color filter with a transmission peak at 450 nm (blue light pixels), and the other pixel covers a transmission peak at 630 nm (red light pixels). Color filter. The two green pixels are placed on the upper right and lower left of the four-piece set. The red pixels are placed on the upper left of the four-piece set. The blue light picture The four-piece set is placed at the bottom right of the four-piece set. The four-piece set covered with color filters is repeatedly placed over the entire 640x48 0 array. Figure 3A shows a top filter layer 106, in which green and blue filters are alternately interspersed in a row of pixels. Below the filter layer is a transparent surface electrode layer 108, which consists of about 0. 〇6 micron thick Indium Tin Oxide (ITO) layer, which is conductive, and visible light can be transmitted in this layer. The photoconductive layer is shown in Figure 3A, under the conductive surface electrode layer A photoconductive layer composed of three sub-layers. The uppermost sub-layer is an n-type doped lice-crystallized amorphous bite layer 110 'with a thickness of about 0.05 micrometers, and a non-doped gasification layer with a thickness of about 0.5 micrometers below it. Amorphous silicon layer 112. The 112 layers are called "intrinsic" layers. Except for photon irradiation, this intrinsic layer appears to have high resistivity. Under the non-doped layer ιΐ2 is about 0.01 micrometers thick and has a high resistivity? Type doped hydrogenated amorphous stone layer 114. In the case of N-type doping, | 'a preferred scale doping concentration is in the range of about 1 atom per cubic meter. For p-type doping, the preferred hip impurity concentration is in the range of atoms / cm3. The three-layer hydrogenated amorphous stone layer

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -12- 200405718 produces a diode effect on each pixel circuit. Applicants claim that this layer 4 is called -p or P-I-N photoconductive layer. PIN and NIP layers

A very important element of the present invention is the layout of the intrinsic layer and the doped layer of the light guide material on an active MOS or CMOS circuit. The reader should understand that the order in which the layers are placed can be reversed. That is, those adjacent to the transparent upper electrode layer 108 may be a P-type doped layer 'and in this case, those adjacent to the lower pixel electrode may be an n-type layer. -The intrinsic layer is located between the ρ-type layer and the "layer. Most charge pairs (⑽ pairs) are generated in the intrinsic layer. The applicant in this specification and the scope of the patent application will be cited in the order of power: The p-type layer is close to the transparent upper electrode layer and the η-type layer is close to the pixel electrode. The M (ph0toelectric) layer may be a piN photovoltaic layer (or a PIN layer); in the opposite case, the configuration will be a Νρ photovoltaic layer (Or the private NIP layer.) For example, the optoelectronic layer shown in the embodiment of FIG.

Layer in the embodiment shown in FIG. 4 is a PIN layer. In the following examples: In the detailed description, the applicant added a barrier between the p-type layer and the intrinsic layer ^ (similar to the intrinsic layer, but thinner and slightly doped with carbon). The photovoltaic layer system It is the PBIN layer or the NIBP layer. The technique of arranging this layer will be described in detail below. In the preferred embodiment described in detail in the instruction manual, all optoelectronic layers are reverse biased. Therefore, for the typical configuration of the embodiment shown in FIG. 3A, before each storage (charge accumulation) stage, During the reset operation, a positive bias charge of + 3.3V is applied to the transparent upper conductive layer, and the pixel capacitor is grounded by resetting. During the charge accumulation process, the positive charge generated by the photoconductive layer capacitor 246 increases its potential. The fast response transistor reads the potential of this increase. After charging or reading the capacitor, the capacitor passes-reset transistor

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -13- 200405718 Grounding. This reset operation reduces the voltage on the pixel capacitor (246 shown in Figure 3B) to a transistor threshold voltage of about 0.7V. For the PIN configuration as described in Figures 5 and 6, during each reset operation, the transparent electrode 355 is grounded and the pixel capacitor is charged forward to approximately 2.7V. In the accumulation process in this case, the negative charge current formed by the charge generated by the PIN photodiode layer will flow to the pixel capacitor for partial discharge. The positive charge flows to the transparent electrode 355 and is grounded. In this case, the decrease in voltage across the pixel capacitor is the measured value of the light detected by the pixel during the accumulation process. When the reduced voltage of the charge is read, the pixel capacitor is charged again to about 2.7 volts in the reset step. Add about 1022 atoms / cm3 of carbon atoms or molecules to the layer 114 in Fig. 3A to increase its resistance. This minimizes horizontal crosstalk between the pixel points and their corresponding spatial resolution loss. The PIN photoconductive layer is not patterned using a lithography process, but has a uniform thin film structure (in the horizontal plane), which simplifies the manufacturing process. Within this sublayer 114 are 307,200 4.6 x 4.6 micron electrodes 116, which define 3,07,200 pixels in a better sensor array. The electrode 116 is made of titanium nitride (TiN). Directly below the electrode 116 is a CMOS pixel circuit 118. In this document, the applicant refers to the electrode 116 as a "pixel electrode" or a "collecting electrode." The pixel collecting electrode is used in combination with the upper transparent electrode to supply a voltage drop in the photoconductive layer, so that the charge generated by the light flows to the storage capacitor. FIG. 3B illustrates the elements in the pixel circuit 118. FIG. The CMOS pixel circuit 118 uses three transistors 250, 248, and 260. U.S. Patent 5,886,353 describes the operation of a tri-transistor pixel circuit in detail. The circuit used in this embodiment is to reduce the wafer surface support as much as possible \ 111; \ 丁 丫 5 \ 进 □ 案 \ 87767 \ 87767.DOC -14- 200405718. The patent application mentioned on the sixth page of the specification describes other finer reading circuits in detail, and other embodiments are explained in detail in the later part of this specification. As shown in FIG. 3A, a pixel collection electrode 116, which is also composed of titanium nitride, is connected to the charge collection node 120 shown in FIG. 3B. The pixel circuit 118 includes a collecting electrode 116, a collecting capacitor 246, a source follower buffer transistor 248, a selection transistor 260, and a reset transistor 250. The pixel circuit 118 uses a p-type channel transistor as the reset transistor 250, and an n-type channel transistor as a source to follow the transistor 248 and the selection transistor 260. The voltage on COL (output) 256 is proportional to the charge Q (input) stored by the collection capacitor 246. By reading the node twice, the voltage difference is proportional to the amount of light detected by the photosensitive structure 122 after the light is irradiated once and the reset is performed again. The node 262 of the pixel circuit 118 is a positive voltage Vcc (typically 2.5 to 5V). The '353 patent describes the pixel circuit of the array in detail. The '353 patent and other embodiments in subsequent sections of this specification describe in detail various integrated circuit lithography techniques that may be used to make the components in the sensor of Figure 3B. Other functions of the camera are shown in this preferred embodiment shown in FIG. 2. In the preferred embodiment, an additional MOS or CMOS circuit is provided on the same crystalline substrate used to collect the charge, which is used to convert the charge into an electrical signal, amplify the signal, and analog signals. Converted into digital signals and digital signal processing. The data output by the sensor section 100 is in digital form and has a continuous pixel stream. The sensor chip area includes a standard clock generation function (not shown here, but described in detail in the '353 patent mentioned in the background section). It can be seen that the signal represents the beginning of the frame, the beginning of the line, the end of the frame, and the end of the line, and the pixels are distributed in all parts of the image chip. H: \ HU \ TYS \ Import Case \ 87767 \ 87767.DOC -15- 200405718 minutes to synchronize the data stream. Environmental analyzer circuit: The data output by the sensor is registered in a data analysis circuit 14o, where the statistics of the images are performed. It is better to divide the sensor area into isolated sub-areas. The average signal in the cut area is compared with the individual signals in the area to distinguish the characteristics of the image data. For example, the following characteristics of the lighting environment are measured: 1 ·% of light source brightness in the image plane 2 · Spectrum composition of the light source required for white balance 3 · Reflectance of the imaging target 4 · Reflection spectrum of the imaging target 5 · The reflection uniformity of the imaging target will provide only the image features to the decision and control circuit (decision and less circmt) I44 ', and the image passing through the data analysis circuit r: will be changed to better. In this embodiment, the average value of the first basic color signal, the average value of the second basic color signal q / color signal, and the average value of the brightness signal among the measured image features. The ^ ^ ^ ,, ^ ^ ^, but the statistical value of the data will be calculated and the original data will be passed to the image processing. These values are very useful for determining the target's reflection table shovel mesh center, the binary P ^ -TL > ^ & and lighting conditions. As for the color A, the statistics are based on the statistics of the entire image model in each sub-image area. 乂 = Che Fujia, but the light weighted average is _some—selected child = read. Implementation of the importance of allowing it to make nj regions, such as the center

H: VHUVFYS \ import case \ 87767 \ 87767.DOC -16- 200405718 area. Decision and control circuit: The decision and control circuit 144 uses the image parameter signal received from the data analyzer 140 to perform automatic exposure and auto_white_baiance (feature control) and evaluates the quality of the detected image. Based on this evaluation, the control element provides a feedback signal to the sensor 1G0 to modify some adjustable aspects of the image data provided by the sensor series. And the control element provides control signals and parameters for the image processing circuit 142. This change can be based on the sub-image or the entire image. The feedback signal from the self-control circuit U4 to the sensor_ provides active control for the sensor elements (substrate, image absorption layer, and reading circuit) to optimize the characteristics of the image data. In particular, the feedback control provides the ability to program the sensor to modify the operation (or control parameters) of the sensing element. The output signal from the camera, the control signals and parameters of the image processing circuit 42 may include specific corrective changes to the image data. Image processing circuit: The image processing circuit 142 receives the image data from the data analyzer and considers the control signal received from the control module to provide an output image data signal. The signal data is based on a control algorithm to optimize its parameters. Optimization. In this circuit, the image data is processed pixel by pixel according to the algorithm, so that each pixel is represented by three color elements. Color saturation, color hue, contrast, brightness can be adjusted to obtain the desired image quality. The image processing circuit provides color interpolation for each pixel and adjacent pixels of the same color filter, so that each pixel can be represented by three color components H: \ HU \ TYS \ iin ^ \ 87767 \ 87767.DOC -17- 200405718. This material has provided enough information for each pixel, so the sensor can generate a color that people can perceive for each pixel. At the same time, it is also capable of color adjustment, so that any expected difference between the color response of the sensor and the human perspective can be optimized. Communication protocol circuit: The communication protocol circuit 146 reorganizes the image data received from the image processing circuit so that its down-stream equipment conforms to the communication protocol, such as an industrial standard or a dedicated standard. The protocol can be in bit_serial or bit-parallel format. Preferably, the communication protocol circuit 146 converts the image processing data into luminance and chrominance components, as described in the ITU_RBT · 601_4 standard. The output of the image chip can be easily used on other components on the market by cooperating with the materials. Other agreements can also be used in specific applications. Input and output interface circuit: The input and output interface circuit 148 receives the data from the communication protocol circuit 46 and converts it into electrical signals that can be detected and recognized by downstream equipment. In the preferred embodiment, the input and output interface circuit 148 provides circuitry so that external devices can obtain data from the image chip and read nest information from the programmable parameter portion of the image chip. Chip package: The image chip is packaged in a 8mm x 8mm plastic chip carrier (chip carrie) with a glass cover. Depending on economy and application requirements, other types and sizes of wafer carriers can also be used. The glass cover can be replaced by other types of transparent objects. The broken glass cover can be coated with an anti-reflectance coating or an infrared cut filter.

H: \ HU \ TYS \ MLM \ 87767 \ 87767.DOC -18-200405718 (off filter *). In another embodiment, if the module is enclosed by the substrate of the installation image, and is combined with the lens holder in a high-cleanness clean room as a cover, the glass cover may not be needed. Camera The lens 4 shown in Fig. 1C is based on the optical format of i / 4 5, f 8, and has a fixed focal length of 3 to 5 meters. Because the size of the chip is small, the entire camera module can be less than 1G mm (length) x 10 mm (width) Xίο halo (height), which is actually smaller than human eyes! Such a compact module is very suitable for portable electronic devices such as mobile phones. The lens mount P is made of black plastic to prevent light leakage and internal reflection. The image chip is inserted into a lens holder with unidirectional indentations on all sides, and is provided as an independent unit when the image chip is inserted and securely fastened. The module has metal leads on an 8mm x 8mm wafer carrier, which can be soldered to a typical circuit board. Feedback and control examples Camera exposure control: The sensor 100 can be used as a light detector to determine lighting conditions. Because the sensor signal is directly proportional to the light detected by each pixel, the camera can perform directing to generate a "nominal" signal at a preset lighting level. When the signal is lower than the “nominal” value, it means that the surrounding “light level” is lower than expected. In order to restore the telecommunication signal to the "nominal" level, the exposure time of the pixels and / or the sensor generation or; the signal amplification factor in the image processing module can be automatically adjusted. The camera can be programmed to divide the entire image into several sub-regions to ensure that changes in operation can be based on a sub-region or to give greater weight to the region of interest. H: \ HU \ TYS \ iiPS \ 87767 \ 87767.DOC -19- 200405718 Camera White Balance Control: The camera can be used under any "light source". Each light source may have a different spectral distribution. As a result, the signal output of the sensor may be different under different light sources. However, users may wish to display the same image when displayed on different display devices, such as when printed on paper or displayed on a Cathode Ray Tube (CRT). This means that different < light sources (daylight, flash, tungsten bulb, etc.) should be treated as a white object. Because the sensor has pixels covered by the main color filters, the camera can be programmed to measure the relative intensity of the light source from the image data. The environmental analyzer is used to receive the statistical data of the artifacts and determine the spectral components, and perform the necessary parameter adjustments during the operation of the sensor or the image processing to generate a signal that can display the same image as people feel under white light. Other POAP sensor pixel cell arrays POAP array with reduced crosstalk Figure 4 is a cross-sectional view of a preferred embodiment of a poAp sensor 300 with reduced crosstalk. As shown in the figure, 'the sensor 300 includes a substrate 310 and PIN layers 350, 345, and 3 40' is similar to the sensor shown in FIG. 3A (except that the layer order of the sensors in FIG. 3A is reversed to provide a NIP layer outer). The sensor in FIG. 4 further includes interconnection structures 315, 32, and 325 of the bottom layer, the middle layer, and the upper layer which are continuously formed on the substrate 31 and below the piN layer. In this embodiment, a metal via 335 is used as a pixel electrode, which is connected to a pixel circuit below through a multiple interconnection structure. Each pixel of the sensor pixel array includes a via 335 provided on the substrate 310, interconnect structures 315, 32, and metal lines 336. As will be described in detail here, the substrate 31 includes a pixel cell circuit.

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -20- 200405718 It is only the circuit in the example that includes the additional function of reducing crosstalk. The I n-type layer 34 is formed at a position adjacent to the upper interconnection structure π. The soil layer 345 is formed adjacent to the n-type layer 34o, and the p-type layer 35o is formed in a square and adjacent to the 1-type layer 345. The p-type layer 350, i-type layer 345, and n-layer 340 form a contact layer, which, together with other elements described below, form an IN light sensor array, which is also referred to herein as pixel sensing. Device. The layer region is electrically connected to the pixel private circuit through the conductive via 335 and the metallized region 336 (shown in the figure again). The semi-transparent conductive layer is disposed adjacent to the p-type layer 350. The bottom and middle layer interconnect structures 315 and 320 are standard cm0s interconnects. It is composed of a non-thin thin sill material (such as SiO2) and has a conductive structure passing through the insulating material. The CMQS mutual The connection structure includes the via 335 and the metallized region 336 shown in FIG. 4 and includes a conductive structure (not shown in the figure). It provides electrical connection with the pixel transistor “under the interconnection structure shown by reference numeral 337 in FIG. 5. The upper layer interconnect structure 325 is similar to 315 layers and 32 layers. ^ Some difficulties are difficult to read, and the structure can be read higher. The reader should understand that the upper layers are made by well-known integrated circuit manufacturing technology. These diagrams (see Figure 4) are not intended to illustrate detailed processes, such as metallized areas and via manufacturing techniques (such as vias-they are usually not placed on top of other vias ^. However, the diagram does show sensing The various components of the device are generally connected to each other.

• Γ 丄 丄 NN lamp table is better, such as hydrogen> (| crystalline silicon. Other materials that may be used to make the PIN layer include u-amorphous carbon, amorphous petrochemicals, amorphous silicon, and amorphous silicon. Yu Ling彳 去 余, > In the 1st best case, the transparent conductive layer -21-

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC 43! 200405718 It can also be made of indium tin oxide. However, other conductive materials can also be used to form the transparent conductive layer 355, including tin oxide, nitride, thin film and the like. A protective layer (not shown) can also be formed on the transparent conductive layer 355. Sensor manufacturing The sensor of the present invention can be manufactured using any conventional CMQS (or MQs) manufacturing technology. For example, the well-known 0.25 micron lithography technique can be used. With this 0.25 micron technology, it is possible to manufacture pixel cells with a size of 4 micrometers and micrometers, and have a fill factor of almost 100%. The standard lithography technology can be used to fabricate the CMos pixel circuit shown in Figure 3A, 118 and Figure 3B, 118 on a silicon wafer. Interconnects also prefer to use standard integrated circuit technology. When the private day shown in FIG. 5 is also made, a layer having a thickness of about 0.25 micrometers is formed on the silicon wafer to form the above structure. For example, the electrode 116 can be fabricated by spimering a sublayer with a thickness of about 0.2 micrometers, followed by a yellow photoresistor and an etching process. After the excess metal is removed by an etching step, the silicon dioxide layer 315 is formed by a CVD technique. By using photoresist and etching techniques, the circular holes required for the vias 335 can be created, and then the circular holes are filled with a conductive material such as tungsten using a sputtering technique. The surface of the first layer 315 is then planarized using a chemical mechanical polishing technique. Then, another interconnection layer 32 is formed on the first layer 3 5 using the same or the same technique. Similarly, second layers 325 are formed on the second layer, each layer being approximately 0.5 micron thick. Then, a photoconductive layer is added by a PECVD technique and a top conductive layer is added by a sputtering method. H: \ HU \ TYS \ ii □ Case \ 87767 \ 87767.DOC -22- 200405718 The POAP sensor of the present invention may include a pixel cell array with various individual pixel cells. For example, the rows and columns of the pixel cell array can be expanded or compressed to obtain the expected array size and configuration (for example: 120x160, 256x256, 512x512, 1024x1024, 2048x2048, 4096x4096, etc.). However, the present invention is not limited to a specific size or geometry, and any array configuration can be applied using a CMOS process to achieve and include the maximum pixel density. The invention further considers modifying the size of individual pixel cells, as well as modifying the size and geometry of the pixel cell array to accommodate any possible cost constraints. Four-transistor POAP sensor circuit Figures 5 and 6 show the function of a POAP image sensor using four-transistor pixels. Specifically, each pixel cell contains three transistors described in the first preferred embodiment: a reset transistor 250, a source follower transistor 248, and a column selection transistor 260. In this embodiment, an additional transistor is provided, that is, a gate bias transistor 360. The gate of transistor 360 is biased by a gate voltage Vcgb. The transistor 360 is provided between the pixel electrode 116 and the charge detection node 120. In FIG. 6, incident light from the photodiode 370 generates positive and negative charges in the photoconductive layer, which generates an effective current Ipd. This effective current Ipd partially discharges the capacitor 346 without significantly changing the potential of the pixel electrode 116. Therefore, the gate transistor 360 effectively isolates the pixel electrode 330 from the charge detection node CS during the accumulation process (ie, the charge accumulation stage). However, the current flowing in any pixel circuit discharges the capacitor 246 to the same extent as the circuit of FIG. 3B. As shown in FIG. 3B, the sensor uses a traditional three-transistor pixel cell configuration, and does not combine a gate transistor 360, which will sense and receive light energy from each of them. □ Case \ 87767 \ 87767.DOC -23-200405718 volume-dependent potential difference. As a result, the neighboring electrode 116 of the pixel cell with the restricted triode configuration may sense a potential difference of about 100 microvolts to 1000 microvolts (assuming a supply voltage of about 3 volts). The presence of this potential difference may cause unwanted pixel crosstalk and may increase pixel blur. Therefore, if the gate transistor is not provided, other techniques may be considered, such as increasing the resistance of layer 114 (shown in FIG. 3A) or layer 340 (shown in FIG. 4). As explained previously, this can be achieved by adding carbon to this layer. As shown in Figs. 5 and 6, the transistor 250 functions as a reset transistor, and its gate is connected to the column reset RST (Fig. 5), and its source is connected to the gate transistor 360. A source follower transistor 248 includes a gate connected to the transistor Mcgb. The selection transistor 260 includes a gate connected to the column selection RSL (FIG. 5), and a drain connected to the source of the source follower transistor 248. During the charge accumulation process, the incident light effectively strikes the photodiode 370 and generates a photocurrent flowing from the capacitor 246 to the transistor 360. Therefore, the photocurrent discharges the capacitor 246 to a degree related to the intensity of incident light on the photodiode. However, as mentioned above, because of the constant gate voltage across transistor 360, the potential on electrode 116 does not change significantly due to the discharge of capacitor 246. As mentioned above, the present invention can make both a PIN diode and a NIP diode structure. Therefore, anyone with ordinary skill will realize that individual pixel cell circuits using a PIN diode structure shown in FIG. 6 can be adjusted to accommodate a NIP diode structure. An example of this structure is shown in FIG. 7. Fig. 7 illustrates a circuit diagram which can be used for individual pixel cells having the NIP diode structure shown. The various electronic components shown in the circuit are substantially similar to the PIN diode structure shown in Fig. 5, but they are adjusted to accommodate the NIP diode structure. For example, as shown in Figure H: \ HU \ TYS \ il □ case \ 87767 \ 87767.DOC-24- 200405718: Shown: Electric Diode 37. versus! The opposite of 6. At the same time, the reset transistor "1-pole receiving-represented as a reverse reset signal of chat", which resets the capacitor 246 to ground by resetting the transistor 250. As a result, in each reset operation, the voltage of the capacitor drops to the threshold voltage of the transistor 25 (about 0.7 volts): In the circuit of FIG. 7, the voltage on the translucent electrode layer 电极 is maintained-a constant supply Voltage, such as 3.3 volts. The present invention is described as having a -planar, or substantially planar photoconductive layer (e.g., n-type layer 340). However, the present invention is also possible with other configurations, and is considered in the present invention < For example, the sensor may be modified to include a non-planar photoconductive layer. Other embodiments of the present invention using a discontinuous, channeled: patterned photoconductive layer configuration will be described graphically below. Non-continuous photoconductive layer Refer to FIG. 8, which is a cross-sectional view of an embodiment of a POA sensor of the present invention, which is denoted by 400. The sensor 400 is similar to the sensors 100 and 300 in many respects, and may include a substrate 310 and a multilayer interconnection structure 352 formed on the substrate 310 (having its private circuit). In addition, the sensor 400 further includes a conductive layer 355, a p-type layer 350, an i-type layer 345, and an n-type layer 340. In addition, the substrate 3 10 may include a pixel cell circuit for reducing or eliminating day-to-day crosstalk. There is a significant difference between the photoconductive layer at the bottom of the sensor 300 and the sensor 400 (such as the n-type layer in a ρΐN diode or the p-type layer in a NIP diode), which is related to the conduction via 335 Electrical connections. In particular, the sensor 400 is shown as having a discontinuous n-type layer 340 formed by a channel 342 between adjacent pixel electrodes 335. Typically, because the material of the intrinsic layer 345 has a higher resistivity than the n-type layer 340, this channel configuration provides an extra distance between adjacent pixels: H: \ HU \ TYS ^ □ case \ 87767 \ 87767.DOC- 25- 200405718 ^ and electrically isolated. However, because the non-continuous n-type layer 34 of the sensor 400 can be used in conjunction with the pixel cell circuit described above, the sensor can reduce pixel crosstalk to "without reducing the fill factor." Lower. Fig. 8α shows a variation of the embodiment of Fig. 8. In contrast to making the channel ditched, the person who requested the towel used a raised, SiO2 wall to create a non-connected η-type layer. Deposited and etched like other internal metal dielectric layers. Channel layer Referring to FIG. 9, a cross-sectional view of another embodiment of the present day and month POAP sensor is shown, which is labeled 500. The sensor 500 is The aspect is similar to the sensor structure. Like the channel design used in the sensor, the sensor 500 also includes a channel 344 between adjacent pixel electrodes 335. However, the channel 3 is private. It is different from the channel 342 (in the sensor 400). The layer 340 in the sensor 500 is continuous, which is different from the non-continuous n-type layer used in the sensor 400. As shown in FIG. 9 The channel configuration typically provides a certain degree of isolation between pixel cells by increasing the pen resistance between adjacent pixels associated with the n-type layer. Figure 10 'shows a cross-sectional view of another embodiment of the POAP sensor of the present invention, which is labeled 600. The sensor 600 is similar to the sensor 500 in many respects, such as including a substrate 310 and formed on The multilayer interconnect structure 352 on the substrate 31. Similar to the other sensors described above, the sensor 600 also includes a conductive layer 355, a p-type layer 350, and an i-type layer 345. However, compared to other sensors, The sensor 600 also includes a patterned n-type layer 34, but it is different from the channel design used by the sensors 400 and 500. The pattern configuration shown in FIG. 10 is typically between adjacent pixel cells. Provide a certain degree of electrical isolation. However, in some applications such as H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -26- 200405718, the electrical isolation provided by the n-type layer 340 patterning alone may not be sufficient. If In this way, the sensor 600 may be configured with the pixel cell circuit including the gate bias transistor (transistor 360 shown in FIG. 5) to obtain the desired pixel cell isolation. Optional structure The individual pixel cells used in the present invention are described as a pixel cell junction using a tri- or quad-transistor. However, other alternative structures can also be used in the consideration of the present invention. For example, the pixel cell circuit used in the sensor of the present invention may include two to six transistors, or more. The following is illustrated by illustration Method description: Other pixel cell structures can be used to maintain higher potential uniformity on the pixel electrode array. Hexamorphic pixel cell Figure 11 illustrates a circuit diagram of a six-transistor pixel cell, which can be used with PIN diode Individual pixel cells of the structure. The circuit shown in Figure II is similar in many respects to that shown in Figure 6. However, compared to the design in Figure 6, the circuit in Figure 11 contains two additional transistors (ie, Mpg and Mtf transistors). In FIG. 11, the transistor Mcgb is biased by a constant gate voltage Vcgb, so each electrode 116 of the pixel electrode array can be maintained at approximately the same potential. The reset transistor Mrst, the source follower transistor Msf, the column selection transistor Mrsl, and the gate bias piezoelectric crystal Mcgb have functions similar to those of the four-transistor pixel cell (Fig. 6). A fifth transistor Mpg 410, a photogate transistor, may be used as a MOS capacitor to store charge carriers (such as electrons) generated by photons. A sixth transistor Mtf 4 1 1, that is, the transmission transistor, can be used to transfer the charge stored in the transistor Mpg to the charge detection node CS. Relative to a small number of electricity H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -27- 200405718 crystal configuration (such as four-cell crystal cell), using one-six-cell pixel cell—advantages The pixel structure of the 7T crystal allows an associated double sampling, which significantly reduces the pixel reading noise. In order to reduce the fixed pattern noise between pixel points in the CMOS sensor reading circuit, a general technique called double sampling is used. In this technology, the reset reference voltage of each pixel can be read and subtracted from the optical signal voltage to eliminate the offset voltage caused by equipment mismatch. Idle may be different. However, each reset operation often introduces a transient noise, commonly referred to as KTC noise. For example, in a four-transistor pixel cell, the charge storage node (the node that stores the charge carriers generated by photons) and the charge detection node (the node that converts charge into a voltage output) are usually the same node. In this way, the reset reference voltage on the charge detection node of each pixel cell can only be read after the optical signal voltage is read (otherwise the stored photo charge will be cleared). Therefore, the 'reset reference voltage is independent of the optical signal voltage. Therefore, the KTC noise associated with the signal voltage cannot be eliminated in a four-transistor design. The double sampling performed in this way is generally called non-associative double sampling. Although uncorrelated double sampling may reduce fixed pattern noise between pixel points, generally it cannot reduce KTC noise, and in many cases it will increase KTC noise instead. On the other hand, in the case of a six-transistor pixel cell, the transmission transistor Mtf effectively isolates the charge storage node (the MOS capacitor under the gate of Mpg) from the charge detection node CS. During the charge accumulation process, the photogate transistor Mpg may generate a bias in the deep depletion mode (for NMOS, Vpg = Vdd) and disconnect the transmission transistor. At the end of charge accumulation, a reset pulse from transistor 250 resets the charge detection node CS. Then, first use the transistor H. \ HU \ TYS \ import case \ 87767 \ 87767.DOC -28- 200405718 to read the reset reference voltage on the charge detection node, and then turn on the transmission transistor 411 and The gate voltage pulse of Mpg 410 enters the accumulation mode (for NMOS, Vpg = 0) to dump the stored photocharge to the charge detection node, so that the optical signal voltage can be read. Because the reset reference voltage contains the same KTC noise as in the optical signal voltage, subtracting the reset reference voltage from the optical signal voltage externally can generally completely eliminate the KTC noise. This operation is called correlated double sampling. Due to the additional two transistors, the six-crystal pixel cell configuration generally requires a larger pixel cell area than the four-crystal pixel cell configuration. However, even if a larger pixel cell is required, the six-transistor pixel cell configuration is very useful in applications where it is necessary to reduce fixed pattern noise between pixels and at the same time to substantially get rid of KTC noise. FIG. 12 illustrates a circuit diagram of a six-cell pixel cell, which can be used for individual pixel cells having a NIP diode structure. The various electronic components shown in this circuit are basically the same as the PIN diode structure shown in FIG. 11, and the variation is to accommodate a NIP diode structure. For example, FIG. 11 shows a six-transistor pixel cell that can be used with a PIN POAP sensor structure. The capacitor 246 is discharged by the photodiode 370 during the accumulation process, and the decrease in charge is measured as a read Voltage. In contrast, FIG. 12 shows a six-transistor day cell that can be used with a NIP POAP sensor structure. The capacitor 246 is charged by the photodiode 370 during the accumulation process, and the increase in charge is measured as a read Take the voltage. In addition, the six-transistor pixel cell shown in FIG. 12 includes reverse PG, TF, and RST signals, which is opposite to the non-reverse PG, TF, and RST signals used in the PIN diode structure shown in FIG. 11. Four-transistor NIP structure H: \ HU \ TYS \ il Case \ 87767 \ 87767.DOC -29- 200405718 FIG. 13 is a cross-sectional view of an embodiment of the present invention, which is labeled 700. The sensor 700 includes a structure similar to that used in, for example, the sensor 300 (FIG. 5) in many respects. Specifically, the sensor 700 includes a multilayer interconnection structure including layers 315, 320, and 325 formed on a substrate 310. However, the sensing sensor 700 includes a NIP diode structure, as opposed to the PIN diode configuration of the sensor 300 in FIG. 5. Sensor 700 and any other sensor embodiments may be equipped with additional transistors to meet the special design needs described above. Barrier Layer and Interlayer Dielectric (ILD) FIG. 14 is a cross-sectional view of a sensor 800. The sensor 800 includes a barrier layer 836 in the PIN photodiode structure. A pBiN photodiode structure is generated. The sensor also includes a special inner dielectric layer 820 under the interconnect structure. As shown in the figure, the sensor 800 includes a substrate 810 (including the pixel circuit described above) and an inner interposer formed on the substrate 810 (including the pixel circuit described above) and located under the interconnect structure 815 Electrical layer 82. This embodiment also includes a diffusion region 830 formed in the substrate, which provides an electrical contact location for the via 825. The diffusion region is preferably diffused to the silicon substrate region with a P or N-type doped material to make the region conductive. An additional barrier layer 836 between the p-type layer and the 丨 -type layer improves the current conductivity through the photoconductive circuit. Its thickness is preferably less than 50 angstroms, and it may be composed of materials similar to the 丨 type layer, but with a proportion of carbon. Applicant's experiments show that adding a very small amount of p-type doping to the barrier layer may improve the performance of the photoconductive circuit. The additional inner dielectric layer 82 is a very thin insulating layer 1 similar to the layer 在. In a preferred embodiment, a thermal process is applied to the silicon wafer (to produce a very thin High quality Si02 layer). The material of layer 815-2 is formed by CVD technology ^ A A. i.t-H: \ HIATYS ^ □ case \ 87767 \ 87767.DOC -30- 200405718. The reader is reminded that the production systems of layers 820, 815, 825, and 830 follow the industry standard integrated circuit manufacturing process. Figure 15 shows a more detailed view of the pixel cell structure that can be applied to any sensor embodiment of the present invention. In particular, Fig. 15 shows the upper portion of individual pixel cells having a transparent conductive layer 845. The transparent conductive layer ⑷ is formed adjacent to the PBIN radiation absorbing structure 835, and the other side of the pBiN radiation absorbing structure 8 is close to the upper surface of the multilayer interconnection structure 815. The multilayer interconnect structure 815 is shown to have three different layers 860, 865, and 870. Each layer of the interconnect structure 815 may include a dielectric material (such as silicon oxide, silicon nitride, or other similar materials) as an insulating material to isolate metal conductors in various layers, metallized regions 873, vias 840, 825 And other metal transistor connections (not shown). This via and the metallized region provide electrical connection 'to the diffusion region 83 and it is in contact with the photoconductive layer for a pixel electrode. Metal pad FIG. 16A is a schematic cross-sectional view of a sensor using a pixel cell structure, which is denoted by reference numeral 900. Compared to the sensor in FIG. 5, the sensor 900 includes a metal pad 876 electrically connected to the via 840. The metal pad 876 may be combined with the via 840 to provide a radiation-absorbing structure 835 (eg, a bottom n-type layer in a pBIN configuration or a bottom p-type layer in a NIBP layered design) and a diffusion region 830 Good electrical connection. The metal pad 8 76 may be formed of a suitable conductive material such as titanium nitride or tungsten. Optional metal pad design Figure 16B is a schematic cross-sectional view of a sensor using a pixel cell structure, which is denoted by reference numeral 950. The sensor 950 does not use an interconnect structure 815, but also H: \ HU \ TYS \ il case \ 8T767 \ 87767.DOC • 31-200405718 does not use an inner layer dielectric (ILD) layer 82, which is It is used in many other sensor designs of the present invention. In the illustrated PBIN configuration, a charge-collecting pixel electrode array can be defined by a diffusion region 830 array and -metal 塾 876 array. Additional Circuit FIG. PA shows the relationship between the pixel cell array 802 and the circuit area 804 according to some preferred embodiments of the present invention. In the figure, the sensor_ displays a pixel cell array (N × M) with -individual pixels. In particular, the circuit area 804 is isolated to the extent that it only occupies the sensor 800 for one side. The circuit area 804 contains a pixel cell array I for fetching and controlling circuits for supporting 800 sensors. For example, the circuit area 804 can also be made up of an analog-to-digital converter (ADC), a digital signal processor (Dsp), a clock and control circuit, and a circuit that provides image processing support. The circuit area 804 may also include an RF circuit to meet the image data transmission and reception requirements in the wireless imager. The POAP sensor of the present invention may include an array of pixel cells having a large range of individual pixel cells. For example, the (N × M) pixel cell array columns and rows can be individually expanded or compressed to achieve the expected array size and configuration (for example, 120x160 256x256, 512x512, 1024x1024, 2048x2048, 4096x4096, etc.). However, it must be understood that the present invention is not limited to the size of the array or the shape of the array, and any array configuration can be applied using CMOS equipment to achieve and include the maximum pixel density. And in January, it can consider changing the size of individual pixel cells, as well as the size and geometry of the pixel cell array, to accommodate any lens restrictions that may exist.

H: \ HU \ TYS \ iiP ^ \ 87767 \ 87767.DOC 442 -32- 200405718 Figures 17B to C show additional configurations that can be used in the POAp sensor of the present invention. FIG. 17B shows a sensor 800 having a (NxM) pixel cell array arranged between two circuit regions 850, and FIG. 17C shows a sensor 800 having a (NxM) pixel cell array. The circuit area § 〇4 is fabricated around the entire leptin cell array. The arrangement of pixel cell arrays and circuit areas shown in Figures 17A to C provides sensors that can be used in a variety of different applications. For example, a sensor module may be manufactured by scaling a plurality of individual sensors using one or more sensor configurations shown in the figure. It will be appreciated that the circuit area configuration shown in Figs. 17A to C can be used in any sensor design described herein. NIBP POAP Sensor FIG. 18A is a cross-sectional view of a sensor according to an alternative embodiment of the present invention, which is denoted by reference numeral 1000. The sensor 1000 includes a structure similar to a sensor that may be used in many aspects, for example, the sensor 100 (FIG. 14). Except for this example, the radiation absorbing layer 885 has a NIBP structure instead of a PBIN structure. . 18B and 18C are schematic cross-sectional views of a sensor using a pixel cell structure with a metal pad. This sensor is similar in many respects to the sensor and pixel cell arrays shown in Figures 16A and 16B, except that the radiation absorbing layer is PBIN. Other Sensor Structures Other preferred embodiments of the present invention are shown in Figs. FIG. 19 is a cross-sectional view of an alternative embodiment of a POAP sensor using a discontinuous photoconductive layer design according to the present invention. FIG. 20 is a cross-sectional view of another alternative embodiment of poAp sensing using a trench-type photoconductive layer design according to the present invention. FIG. 21 is a cross-sectional view of another alternative embodiment of a POAP sensor using a patterned photoconductive layer design according to the present invention. FIG. 22 is a cross-sectional view of an embodiment of the PBIN sensor of the present invention, and some available

H: \ HU \ TYS \ igp ^ \ 87767 \ 87767.DOC -33- 200405718 Supports related pixel cell circuits of PBIN diode structure. FIG. 23 is a cross-sectional view of an NIBP sensor according to the present invention, and some related pixel cell circuits that can be used to support the NIBP diode structure. FIG. 24 illustrates a circuit diagram of a six-transistor structure, which can be used for individual pixel cells having a PBIN diode structure. FIG. 25 illustrates a circuit diagram of a six-transistor structure, which can be used for individual pixel cells having a NIBP diode structure. FIG. 26 illustrates a circuit diagram that can be used for individual pixel cells having a PBIN diode structure without using a gate bias transistor. Fig. 27 illustrates a circuit diagram which can be used for individual pixel cells having a NIBP diode structure without using a gate bias transistor. FIG. 28 illustrates a circuit diagram of a five-transistor structure, which can be used for individual pixel cells having a PBIN diode structure. FIG. 29 illustrates a circuit diagram of a five-transistor structure, which can be used for individual pixel cells having a NIBP diode structure. POAP Sensor Manufacturing FIG. 30 illustrates an operation flowchart for manufacturing a POAP sensor. For the formation of the POAP sensor according to the embodiment of the present invention, please refer to the operation shown in FIG. 30 and the manufacturing stages of the sensor shown in FIGS. 3A to F. In FIG. 30, the first operation 605 is to provide a substrate 8 1 0 having an array of electrically conductive diffusion regions 830 (FIG. 31A). In the preferred embodiment, the pixel circuit unit includes at least two transistors placed on the substrate and connected to the diffusion region 830. Next, in operation 610, an inner dielectric layer 820 (ILD) is formed, which has an array formed by contact channels 825. The contact channel 825 is connected to one of the diffusion regions 830 in an array formed by the diffusion regions 830 (Fig. 31B). In operation 615, a multilayer interconnection structure 815 is formed on the ILD layer 820 (FIG. 31C). The conductive vias 840 in the multilayer interconnect structure 815 can be formed from a combination of H: \ HU \ TYS \ iinil \ 87767 \ 87767.DOC -34- 200405718 suitable conductive materials, such as tungsten, copper, aluminum, or other well-known Similar materials in integrated circuit manufacturing technology to generate vias. Generally, the conductive vias 840 are formed by a CVD process, but other techniques (such as sputtering) can also be used. Metal conductors that connect pixel circuit elements may also be included in the multilayer interconnect structure. In an optional operation, a metal pad 876 may be patterned adjacent to the interconnect structure 815. In this way, the electrical connection between the via 840 and the photoconductive layer 835 (FIG. 31D) will be improved. In the next operation, individual PMN layers are successively deposited on the interconnect structure 815 to form the radiation absorbing structure 835 (FIG. 31E). Alternatively, in the case where an NIBp diode structure is required (sensor 1000 as shown in FIG. 18A), each individual NiBp photodiode layer is continuously deposited on the interconnect structure 815 to form radiation absorption structure. In general, the layers of the radiation absorbing structure 835 are deposited using PECVD. Furthermore, it can be understood that, according to one embodiment, because one or more layers need not be patterned (such as the bottom N layer in a PBin diode structure), the radiation absorbing structure 835 can be used in a continuous deposition process. Middle formation. A transparent conductive layer 845 may be deposited on the top layer of the radiation absorbing structure 835 (FIG. 31F). The transparent conductive layer 845 is a top layer of the radiation absorbing structure 835 (such as a p-type layer of a PBIN diode structure or an n-type layer of a NIBp diode structure) to provide ground or an electrical connection to a voltage source. The transparent conductive layer 8 4 5 can be deposited by a reactive plating method. The transparent conductive layer can be formed by vapor deposition. However, if the transparent conductive layer 845 is formed of titanium nitride, a CVD or sputtering technique is generally used at this time. Applications, other features, and variations shown here are considered to be the preferred embodiments of the present invention. However, it is clear that H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -35- 200405718 Make various modifications and changes without departing from the scope and essence. Λ Application-A sensor unit with proper configuration using the POA sensor structure can be used in various applications, including cameras, machine vision systems, car navigation systems, video phones, computer input devices, surveillance systems, autofocus systems , Galaxy tracking, motion detection systems, image stabilization systems, scanners and other similar devices. People who can benefit from the use of children's equipment include: law enforcement, medical treatment, burned out door, emergency services and search and rescue operations, and military and intelligence groups. Other applications include non-destructive inspection, preventive maintenance operations, business safety applications, and the automotive industry, for example, providing drivers with low-brightness visual enhancement devices. Doping The doping of the layers of the sensor of the present invention can be performed by well-known semiconductor processes. This process can be used to form a layer with a doping gradient. For example, the doping concentration will change with the depth of the layer, or the doping concentration may change suddenly. The doping gradient or doping profile can be optimized for a particular sensor design. Although a number of examples of suitable materials for manufacturing the radiation absorbing structure 835 have been described, the present invention is not limited thereto. In a preferred embodiment, the applicant uses Plasma Enhanced Chemical Vapor Deposition (PECVD) technology to generate a NIP photodiode layer, which is well known in the industry. We first grow a p-type layer. This p-type layer is in contact with the pixel pad. The film uses the "P-type impurity system side". In order to increase the resistivity, we can also change the impurity in the p-type layer On the order of one meter. Carbon atoms / molecules are on the order of 1G21, 22 atoms / cubic meter. : Sign: It is best to contain only silicon and thorium atoms, and the concentration of hydrogen atoms is preferably in the order of 2 10 _1022 atoms / cm 3. Of course, some undesired impurities may be contained in the film. The n-type layer in contact with IT0 should contain only ^ -type impurities and hydrogen atoms. In this example, the n-type impurity f has a concentration of 1 () 2. _1 () 21 Atoms / Zambeam top grade Phosphorus. In the p-type layer and the n-type layer, the order of hydrogen atoms is 軚; the degree of order is 10 Lio22 atoms / cm3. Protective layer The transparent conductive layer 845 (shown in FIG. 31F) may be made of indium tin oxide (IT0), tin oxide, titanium nitride, thin silicide, or other similar materials. A protective layer (not shown in the figure) may be formed on the transparent conductive layer 845. This protective layer provides mechanical protection, electrical isolation, and anti-reflective properties. Defect Correction Circuit Defect correction circuit can be used. The purpose of this circuit is to eliminate the "single pixel defect" or "small set defect" in the video stream without determining the defect location in advance. The working principle of this circuit is to program the sensor to check a pixel and its nearest pixels, and compare the value of the pixel point with the standard value of the area. If the value of the pixel point is higher or lower than the standard threshold value determined by experience, the pixel point will be considered "defective" and the value of the pixel point will be replaced by the standard value. This technology can eliminate defects. However, if the information is only one pixel wide (although this situation is very unlikely to occur in normal camera applications), the real information will also be eliminated. This circuit can potentially substantially improve the yield of the sensor (yiel sentence, because of the present and foreseeable future system H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -37- 200405718 Cheng & Yield The loss is caused by the "single pixel defect". In the preferred embodiment, the microcrystals are broken. The basic materials used to make the radiation absorbing layer (ie, PIN, NIP, PBIN or NIBP layer) are microcrystals (microamps) and microcrystals. (Germanium) instead of amorphous fragments. As a result, the spectral response width was increased from 35nm to 750nm in amorphous silicon to 300nm to 1000nm in microcrystalline silicon and microcrystalline silicon ( Germanium) 300nm to 1200nm. Another advantage of using microcrystalline silicon or microcrystalline silicon (germanium) is to improve the response time by at least 10 times, which can lead the sensor in the KHz frame rate. . High-definition television (HDTV) sensor For a high-definition television sensor of 2 million pixels, the applicant integrated two unique circuits, a series of "column ADC" ", The other is a" deita-double sampling (DDS) "circuit. In a better sensor, each row (the The preferred sensor has 920 rows (column X 1280 rows) to provide an analog-to-digital converter. The purpose of the column analog-to-digital converter is to ease the sampling rate of the analog-to-digital converter.-2 · 丨Arrays of megapixels are executed at 30 Hz and 10 bits, which means that a single analog digital converter needs to execute 60M 10-bit samples per second. This puts the analog digital converter at the most cutting-edge design and manufacturing technology. By using a "column-to-analog digital converter", the applicant can reduce the sampling rate by the number of columns, which is 1920 in this example. Therefore, the applicant can ease the sampling rate requirements of the analog-to-digital converter and use the same on other circuits. CMOS technology. The DDS circuit continuously converts the two signals, one is the real signal and the other is the reference output of each pixel. The applicant's circuit converts the two signals into digital values and uses the difference between the two signals to replace

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -38- 200405718 represents the real video signal. This circuit is designed to minimize any offset (due to the circuit or process < non-uniformity) between the charge storage capacitance in the pixel and the input of the edge-to-digital converter. This DDS circuit can improve uniformity and alleviate the requirement for uniformity of manufacturing technology. Applications of small cameras Embodiments of the present invention provide a potential camera with a very small size, very low manufacturing cost, and very high quality. Generally speaking, there will be some restraint between size, quality and cost, but with the large I production in the cost range of a few dollars, the size of the millimeter size and the image measured in millions of pixels or 100,000 pixels Quality, the scope of application of the present invention is very wide. In addition to mobile phones, other potential applications are listed below: An analog video camera-digital video camera-security camera-digital camera-a personal computer camera-toys-endoscopes ~ military drones, bombs and missiles-sports equipment-high definition Since the quality television sensor can be made smaller than a human eyeball, one embodiment of the present invention is a camera shaped as a human eye. Due to its low cost, the eyeball camera can be incorporated into many toys and new products. A cable can be attached as an optical nerve to monitor from a

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -39- 200405718 (such as a personal computer monitor) to obtain image data. The eyeball camera can be incorporated into a doll or a mannequin, and even equipped with a rotating element and a feedback circuit so that the eyeball can move within its visual range. If no cable is used, the image data can also be transmitted wirelessly using cell phone technology. This small-sized camera allows Parker-Mobile Phone Transmitter to be mounted in the negative headboard of professional football. 1: Fans can watch the movements of professional athletes like sports. In fact, the camera with a transmitter can even be mounted on a football, so the moon dagger is enough to provide a very interesting action. These are just examples of the thousands of potential applications for this tiny, cheap, high-quality camera. This small camera can be used to monitor the light intensity profile without a lens, and to output changes in its intensity and profile, which is crucial in optical communication applications. Because this requires detecting the beam profile to obtain the highest transmission efficiency. When other light sensing materials are used instead of amorphous silicon, the camera can be used to extend lightsensing beyond visible light. For example, one can use microcrystalline silicon (germanium) to extend light sensing to the near-infrared light range, and its cameras are ideal for night vision. We use such a package in a preferred embodiment. The sensor is mounted on a chip carrier and is snapped onto a lens housing. One can also change the assembly sequence by first soldering the sensor to a sensor base, then covering the sensor with a lens holder with a lens and mechanically fixing it to a PCB to produce a camera. To those skilled in the art, it is only a natural modification of the present invention. Other applications of the present invention In addition to the applications described above, a sensor unit using a POAP sensor structure and a reasonable configuration can also be used in many applications, such as machine vision systems, vehicle navigation systems, video phones, computer input devices , Surveillance system, H: \ HU \ TYS \ Entry case \ 87767 \ 87767.DOC • 40- 200405718 focus system, galaxy tracker, motion detection system, image stabilization system, field = and other types of equipment. Benefits from the use of this device include law enforcement: secondary treatment, fire services, emergency services and search and rescue operations, as well as military and intelligence ^ In addition, "applications include non-destructive testing, preventive maintenance operations, strong For example, to provide drivers with low brightness vision and other changes. The transparent layer can be replaced by a grid of very thin conductors. As shown in Figure 2: The fetch circuit and the camera circuit 148m148 can be partially or entirely located under the CMOS pixel array to make a very tiny camera. The c-deleted circuit can be divided or replaced entirely by M0S circuits as shown in 12. Some of the circuits L40 to 148 can be located on one or more wafers, rather than on a chip with a sensing array (eg. Circuits 144 and 146 are divided into chips manufactured on an early-morning wafer. For each g, 卩, the last one, the early-morning alone < the number of pixels with cost advantages in the processor can be increased or decreased from 0.3 megapixels. It can be almost limited. Therefore, the scope of protection of the present invention should be determined by the scope of the following patent applications and their legal equivalents. 5. Brief description of the drawings Figure 1AΗB shows a mobile phone equipped with a camera, and the camera uses the invention CMOS sensor array; Λ Figure 1C shows some details of the camera; Figure 2 shows some details of a CMOS integrated circuit using the principles of the present invention;

H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -41 · 200405718 Figure 3 A is a partial cross-sectional view of a pixel cell structure with five pixels in a sensor array using one of the principles of the present invention; Figure 3B shows a single pixel CMOS pixel circuit; Figure 3C shows a grid pattern of a color filter; Figure 4 is a cross-sectional view of a sensor with a photodiode array, the photodiode The array consists of a photoconductive layer on an active pixel array of CMOS circuit components; FIG. 5 is a cross-sectional view of a first embodiment of the POAP sensor of the present invention, and some related pixel cell circuits, which can be used for Supports a PIN photodiode structure; Figure 6 illustrates a circuit diagram that can be used for individual pixel cells with a PIN photodiode structure; Figure 7 illustrates a circuit diagram that can be used for individual pixels with a NIP diode structure Figure 8 and Figure 8A are wearing views of an optional embodiment of the POAP sensor of the present invention, which uses a discontinuous photoconductive layer design; Figure 9 is another alternative of the POAP sensor of the present invention A cross-sectional view of the embodiment, which uses a trench-type photoconductive layer design; FIG. 1 0 is a cross-sectional view of another alternative embodiment of the POAP sensor of the present invention, which uses a patterned photoconductive layer design; FIG. 11 illustrates a circuit diagram of a six-transistor structure, which can be used for a PIN diode structure Individual pixel cells; Figure 12 illustrates a circuit diagram of a six-transistor structure, which can be used for individual pixel cells with a NIP diode structure; H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -42- 200405718 FIG. 13 is a cross-sectional view of an alternative embodiment of the POAP sensor of the present invention, and some related pixel cell circuits that can be used to support the NIP diode structure; FIG. 14 is a diagram of a sensor with a barrier layer A schematic cross-sectional view of the first embodiment; FIG. 15 illustrates a partial cross-sectional view of a pixel cell structure, which can be used in any embodiment of the POAP sensor of the present invention; FIG. 16A is a PBIN sensor using a metal pad 16B is a schematic cross-sectional view of an alternative embodiment of a PBIN sensor; FIG. 17A is a top view of a POAP sensor of the present invention, showing a pixel cell array and a single sensor circuit area of an embodiment The relationship between them; FIG. 17B is the POAP sense of the present invention The top view of the device shows the correlation between the pixel cell array and the multi-sensor circuit area of another embodiment; FIG. 17C is a top view of the POAP sensor of the present invention, showing the pixel cell array and The relationship between the four-sided sensor circuit areas; Figure 18A is a cross-sectional view of an embodiment of the NIBP sensor of the present invention; Figure 18B is an optional embodiment of an NIBP sensor using a metal pad 18C is a cross-sectional view of another alternative embodiment of a NIBP sensor; FIG. 19 is a cross-sectional view of an alternative embodiment of a POAP sensor using a discontinuous photoconductive layer design according to the present invention; FIG. 20 is a cross-sectional view of another alternative embodiment of a POAP sensor using a trench-type photoconductive layer design according to the present invention; FIG. 21 is another alternative embodiment of a POAP sensor using a patterned photoconductive layer design according to the present invention; Sectional view of the selected embodiment; H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC -43-200405718 Figure 22 is a sectional view of one embodiment of the PBIN sensor of the present invention, and some can be used to support PBIN Related pixel cell circuit of diode structure; Figure 23 is a NI of the present invention A cross-sectional view of the BP sensor and some related pixel cell circuits that can be used to support the NIBP diode structure. Figure 24 illustrates a circuit diagram of a six-transistor structure that can be used for individual pixel cells with a PBIN diode structure. ; Figure 25 illustrates a circuit diagram of a six-transistor structure, which can be used for individual pixel cells with a NIBP diode structure; Figure 26 illustrates a circuit diagram, which can be used for individual pixel cells with a PBIN diode structure, without A gate bias transistor is required; Figure 27 illustrates a circuit diagram that can be used for individual pixel cells with a NIBP diode structure without the need for a gate bias transistor; Figure 28 illustrates a five-transistor structure Circuit diagram, which can be used for individual pixel cells with PBIN diode structure; Figure 29 illustrates a circuit diagram of a five-electrode structure, which can be used for individual pixel cells with NIBP diode structure; Figure 30 illustrates an operation flow Figures, which can be used to manufacture a POAP sensor according to one embodiment of the present invention; and Figures 31A to 31F are partial cross-sectional views showing manufacturing steps, which can be used to manufacture a POAP sensor according to one embodiment of the present invention. Device . Six, component symbol description 2A, 2B mobile phone 4 lens 8 camera cover 4B, 6 camera 10 connector H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC-44-200405718 800, 900, 950, 1000 100 , 300, 400, 500, 600, 700, sensor pixel array 102, 802 pixel array 1 0 4, 8 0 4 reading and timing / control circuit 106 filter layer 110, 112, 114, 340, 118 Pixel circuit 12 lens holder 122 light sensing structure 14 image chip 142 image processing circuit 146 communication protocol circuit 16 circuit board 250, 248, 260, 360, 410 256 COL (output) 310, 810 sensor substrate 108 electrode layer, 345, 860, 865, 870 layers 116 electrodes 120 charge collection nodes 335 vias 140 data analysis circuits 144 decision and control circuits 148 input and output interface circuits 246 capacitors, 411 transistors 262 nodes 315, 320, 325, 337, 352, 815 interconnect structure 336 metal line 355 transparent conductive layer 605, 610, 615, 620, 804, 850 circuit area 825, 840 via 835 radiation absorption structure 845 conductive layer 876 metal pad 342, 344 channel 3 70 photodiode 625, 630 operation 820 Barrier layer 836 a dielectric layer 830 diffusion region 875 region 885 of the radiation absorbing metal layer

H: \ HU \ T YS \ import case \ 87767 \ 87767.DOC

Claims (1)

200405718 Patent application scope: 1. An active metal-oxide-semiconductor (MOS) or complementary metal-oxide-semiconductor (CMOS) sensor, including: A) — a crystalline substrate; B) — a charge-generating photoconductive layer, Including at least two charge generating material layers for converting light into charges; C) a plurality of MOS or CMOS pixel circuits, which define a plurality of pixels, and are manufactured in a crystalline substrate under the charge generating photoconductive layer, Each pixel circuit defines a pixel and includes: a charge collecting electrode, a capacitor, and at least two transistors, the pixel circuit is used to collect and read out the charges generated by the charge generating material layer; D) a surface electrode, It is a thin transparent layer or grid type and is located on the charge generating material layer; and E) a power source for providing a voltage drop across the charge generating photoconductive layer. 2. The sensor according to item 丨 of the patent application, wherein the plurality of Mos or CMOS pixel circuits are a plurality of CMOS pixel circuits. 3. For the sensor in the scope of patent application, a metal pad is used as a charge collection electrode for each pixel. 4. The sensor according to item 丨 of the patent application scope, wherein the plurality of pixels are at least 100,000 pixels. 5. The sensor as described in item 丨 of the patent application scope, further comprising an additional MOS or CMOS circuit in or on the same crystalline substrate as the active sensor for conversion to The image charge read by the element circuit. H: \ HU \ TYS \ import case \ 87767 \ 87767.DOC λ r Γ · 200405718 6 · For example, the scope of the patent application is "Sense, where the charge generating material includes a p-type doped layer, an intrinsic Layer and — n-type layer. 7. The sensor according to item 6 of the patent claim, wherein the surface electrode is composed of indium tin oxide (ITO). 8. The sensor according to item 6 of the patent application, wherein the n-type layer is adjacent to the surface pen and the P-type layer is adjacent to the charge collecting electrode. 9. The sensor according to item 6 of the patent application scope, further comprising a barrier layer between the p-type layer and the intrinsic layer. Not the sensor described in item 8 of the patent scope, where? The mold layer contains carbon atoms or molecules. The scope of patent application! The sensor according to the above item, wherein the plurality of pixel circuits and each pixel circuit includes a gate bias transistor for separating the charge collecting electrode from the capacitor. 12. The sensor according to item 6 of the patent application, wherein each pixel circuit of the plurality of pixel circuits includes a gate bias transistor which separates the charge collecting electrode from the capacitor. 13. The sensor as described in item 12 of the Shen-Jin patent scope, wherein the gate-biased electric crystal maintains a constant voltage. 14. The sensor according to item 1 of the patent application scope, further comprising an interconnect structure 'formed on the crystalline substrate and under the charge generating photoconductive layer. 1 5. The sensor according to item 14 of the scope of patent application, wherein the interconnect structure includes at least two sub-layers, and each sub-layer includes conductive vias for electrically connecting the plurality of day-to-day circuits with the charge to generate photoelectricity导 层。 Guide layer. 16. The sensor described in item 1 of the scope of the patent application, which also includes a data analysis circuit manufactured on the H: \ HU \ TYS \ ^ L ': J ^ \ 87767 \ 87767.DOC 200405718 crystal substrate. 17. = Apply for the job description described in item 16, which also includes an image processing circuit manufactured on the substrate. 18.2 The sensor described in item 17 of the patent scope also includes a decision and control circuit manufactured on the substrate. 19. = The sensor described in item 18 of the scope of patent application, which also includes a communication circuit manufactured on the crystal substrate. 20. The sensor as described above in the scope of the patent application, wherein the sensor is an integral part of a camera, and the camera is connected to a mobile phone by a cable. 21 · If the patent scope, please! The sensor according to the item, wherein the surface electrode is composed of an indium tin oxide layer. 22. The sensor according to item 丨 of the patent application scope, wherein the at least two charge generating layers mainly include a hydrogenated amorphous silicon layer. 23. The sensor as described in page "Scope of the patent application", wherein the at least two charge generating material layers are composed of microcrystalline silicon or microcrystalline silicon (germanium). 24. As described in item 1 of the patent application scope The sensor includes a protective layer. 2 5 The sensor according to item 1 of the patent application scope further includes a circuit for locating a defective pixel and replacing normal data for the defective pixel. 26. The sensor according to item 1 of the scope of patent application, wherein the sensor is located in an integral part of a camera in a mobile phone. 27. The sensor according to item 1 of the scope of patent application, which An array of color filters located on top of the surface electrode is also included. 28. The sensor according to item 27 of the scope of patent application, wherein the color phosphor is composed of red, green and blue filters The filter is configured according to two green and one red H: \ HU \ TYS \ into the case \ 87767 \ 87767.DOC 200405718 and the blue four-piece set. 9 · As the first patent application scope The sensor described above, wherein the sensor is a camera made in the shape of a human eyeball. 〇_ The sensor as described in item 18 of the scope of patent application, wherein the decision and control pen circuit includes a processor, which is programmed with a control sub-algorithm for analyzing pixel data, and is based on the data. Controls the signal output from the sensor. The sensor as described in item 30 of the patent application of the μ μ patent, wherein the processor controls the signal output by the illumination time of the whole pixel. The sensor described in the above item, wherein the processor controls the signal output by adjusting the signal amplification. For example, the sensor described in item 1 of the patent application, wherein the sensor is-combined with a device As part of the camera, the device is selected from the following groups:-analog camcorders;-digital camcorders;-security cameras;-digital cameras;-personal computer cameras;-toys;-endoscopes;-military drones , Bombs, missiles;-sports equipment; and-high-definition television sensors. 34 ·-Known with active MOS or CMOS sensor array ", Qi machine, use charge to generate light to generate electrons Image The machine contains: A) —crystalline substrate; H: \ HU \ TYS \ ^ t □ case \ 87767 \ 87767.DOC -4- 200405718 B) — charge generation photoconductive layer, including at least two charge generation material layers for Light is converted into charge; C) A plurality of MOS or CMOS pixel circuits are manufactured in a crystalline substrate under the charge generating photoconductive layer. Each pixel circuit includes: a charge collection electrode, a capacitor, and at least three transistors. The pixel circuit is used to collect and read out the charges generated by the charge generating layer; D) — surface electrode, which is a thin transparent layer or grid type, and is located on the electric generating layer, E) — power supply, For providing an inverse bias across the charge to generate a photoconductive layer; F) additional MOS or CMOS circuits with active sensing in and / or on the same crystalline substrate to convert the charge into an image G) additional MOS or CMOS circuits with active sensor arrays for timing and signal synchronization in and / or on the same crystalline substrate; and H) focusing optics for The light generated by the electron hole is focused to the Movable sensor array. H: \ HU \ TYS \ il □ case \ 87767 \ 87767.DOC