[0023] At present, when the image sensor is working, the incident light is first filtered into three monochromatic lights of red, blue and green through a filter, and then the corresponding photodiodes collect the corresponding optical signals, output electrical signals, and restore the image through difference calculation. . Since the photodiode only has the function of photoelectric conversion, in the case of weak light, the light sensitivity is not enough, which affects the imaging effect of the image sensor.
[0024] figure 1 is a structural schematic diagram corresponding to the formation process of the image sensor. refer to figure 1 , providing a semiconductor substrate 100, in which discrete photodiodes 110 are formed, and the discrete photodiodes 110 are isolated by deep trench isolation structures 120, and the depth of the deep trench isolation structures 120 is Deeper than the photodiode 110, so as to obtain a better isolation effect and avoid the problem of photo-generated carrier diffusion between different pixel regions.
[0025] Then continue to refer to figure 1 , forming a filter layer 130 on the semiconductor substrate 100 , the filter layer 130 is divided into red, green and blue; forming a micro lens 140 on the filter layer 130 .
[0026] During use, the incident light is firstly converged by the microlens 140 and enters the filter layer 130; the filter layer 130 filters the color of the incident light, and the different filter layers 130 respectively filter out red light and green light. and blue light; the photodiode 110 in each unit area receives the corresponding color filter light for photoelectric conversion processing.
[0027] The inventors have found through research that the photodiode in the current image sensor only has the function of photoelectric conversion, which cannot meet higher requirements. In view of the above problems, an image sensor is formed, which uses a phototransistor instead of a photodiode. The phototransistor itself can not only perform photoelectric conversion, but also has a signal amplification function. Enhanced image; and as a photosensitive element, the phototransistor has higher sensitivity than the diode.
[0028] The technical solution of the present invention will be described in detail below in conjunction with the embodiments and the accompanying drawings.
[0029] Figure 2 to Figure 4 It is a schematic cross-sectional structure diagram corresponding to each step in the formation process of the photosensitive element in the image sensor of the present invention.
[0030] like figure 2 As shown, a first substrate 200 is provided, the first substrate 200 has a first surface 201 and a second surface 202, the first surface 201 and the second surface 202 are opposite surfaces; through the second The surface 202 implants first ions 211 into the first substrate 200 to form a collector 210 of a phototransistor.
[0031] In this embodiment, the method for forming the collector electrode 210 is as follows: firstly, a first photoresist layer (not shown) is formed on the second surface 202 of the first substrate 200; The photoresist layer is subjected to a photolithography process to form a collector pattern; using the first photoresist layer as a mask, implanting first ions 211 into the first substrate 200 along the collector pattern; removing all The first photoresist layer.
[0032] In this embodiment, the first ions 211 are N-type ions, such as phosphorus (P) ions or arsenic (As) ions, the implantation energy of the first ions is 160keV-240keV, and the implantation dose is 2e13/cm 2 ~5e15/cm 2.
[0033] like image 3 As shown, second ions 221 are implanted into the first substrate 200 through the second surface 202 to form the base 220 of the phototransistor.
[0034]In this embodiment, the formation method of the base 220 is as follows: firstly, a second photoresist layer (not shown) is formed on the second surface 202 of the first substrate 200; The photoresist layer is subjected to a photolithography process to form a base pattern, and the projected area of the base pattern on the second surface 202 is smaller than the projected area of the collector electrode 210 on the second surface 202; so The second photoresist layer is used as a mask, and second ions 221 are implanted into the first substrate 200 along the base pattern; and the second photoresist layer is removed.
[0035] In this embodiment, the second ions 221 are P-type ions, such as boron (B) ions. The implantation energy of the second ions is 120keV-180keV, and the implantation dose is 2e13/cm 2 ~5e15/cm 2. Since the energy of the first ion implantation is greater than the energy of the second ion implantation, the implantation depth of the collector electrode 210 (the depth relative to the ion implantation surface, that is, the second surface 202 ) is lower than the implantation depth of the base electrode 220. deep. Since the implantation depth and projection area of the collector 210 in the first substrate 200 are larger than the implantation depth and projection area of the base 220 in the first substrate 200, the collector 210 surrounding the base 220 .
[0036] like Figure 4 As shown, the third ion 231 is implanted into the first substrate 200 through the second surface 202 to form the emitter 230 of the phototransistor, the collector 210, the base 220 and the emitter 230 constitute a phototransistor.
[0037] In this embodiment, the method for forming the emitter 230 is as follows: firstly, a third photoresist layer (not shown) is formed on the second surface 202 of the first substrate 200; The photoresist layer is subjected to a photolithography process to form an emitter pattern, and the projected area of the emitter pattern on the second surface 202 is smaller than the projected area of the base 220 on the second surface 202; so The third photoresist layer is used as a mask, implanting second ions 231 into the first substrate 200 along the emitter pattern; removing the third photoresist layer.
[0038] In this embodiment, the third ions 231 are N-type ions, such as phosphorus (P) ions or arsenic (As) ions; the implantation energy of the third ions is 60keV-140keV, and the implantation dose is 2e13/cm 2 ~5e15/cm 2. Since the implantation energy of the second ions is greater than the implantation energy of the third ions, the implantation depth of the base 220 is deeper than that of the emitter 230 . Since the depth and projected area of the base 220 in the first substrate 200 are larger than the depth and projected area of the emitter 230 in the first substrate 200, the base 220 surrounds The emitter 230 .
[0039] In this embodiment, the order of forming the collector, base, and emitter is in accordance with the order of implantation depth from deep to shallow. In other embodiments, the order of forming the collector, base, and emitter can be changed as required. However, the final positions of the collector, base, and emitter in the first substrate remain unchanged. Another example of the formation sequence is as follows: first, a first photoresist layer with an emitter pattern is formed on the second surface of the first substrate; using the first photoresist layer as a mask, Implanting first ions into a substrate to form an emitter; after removing the first photoresist layer, forming a second photoresist layer with a base pattern on the second surface; using the second photoresist The resist layer is a mask, implanting second ions into the first substrate to form a base; after removing the second photoresist layer, a collector electrode is formed on the second surface of the first substrate a patterned third photoresist layer; using the third photoresist layer as a mask, implanting third ions into the first substrate to form a collector.
[0040] In addition, the phototransistor formed in this embodiment is of NPN type. In other embodiments, the phototransistor can also be of PNP type, that is, the first ions implanted are P-type ions, the second ions are N-type ions, and the third ions are P-type ions.
[0041] The figure of the above-mentioned embodiment only exemplifies the situation of the phototransistors in one pixel area. In the actual process, each phototransistor is arranged discretely in the first substrate 200, and the phototransistors are isolated by a deep trench isolation structure. Moreover, the depth of the deep trench isolation structure 230 is deeper than that of the photodiode 220, so as to obtain a better isolation effect and avoid the problem of photocarrier diffusion between different pixel regions.
[0042] In this embodiment, the first substrate 200 may be a silicon substrate. In other embodiments, the material of the semiconductor substrate 200 can also be germanium, silicon germanium, silicon carbide, gallium arsenide or indium gallium, and the semiconductor substrate 200 can also be a silicon-on-insulator substrate or a silicon-on-insulator substrate. A germanium substrate on the surface, or a substrate with an epitaxial layer grown on it.
[0043] In this embodiment, the phototransistor is used as a photosensitive device to convert the received optical signal into an electrical signal, and at the same time, gain and amplify the received optical signal. It should be noted that, in order to be more conducive to the smooth absorption of light of different colors (light of different wavelengths), optionally, phototransistors of different light absorption color gamuts can be made at different depths, for example, the photoelectric transistors in the red light absorption region The absorption depth of the triode (relative to the depth of the light incident surface, that is, the first surface 201) is the deepest, and the absorption depth of the phototransistor in the green light absorption region is shallower (less than the absorption depth of the phototransistor in the red light absorption region), and the blue light absorption region. The absorption depth of the phototransistor in the red light absorption region is the shallowest (less than the absorption depth of the phototransistor in the green light absorption region); correspondingly, the implantation depth of the phototransistor in the red light absorption region (relative to the ion implantation surface, that is, the depth of the second surface 202) is the lowest Shallow, the injection depth of the phototransistor in the green light absorption region is deeper (greater than the injection depth of the phototransistor in the red light absorption region), and the injection depth of the phototransistor in the blue light absorption region is the deepest (greater than the injection depth of the phototransistor in the green light absorption region depth).
[0044] Subsequently, the first surface 201 of the first substrate 200 can be thinned first, and then a filter layer is formed on the thinned first surface 201; then a microlens is formed on the filter layer .
[0045] The image sensor with a phototransistor formed in the above embodiment includes: a first substrate 200, the first substrate 200 has a first surface 201 and a second surface 202, the first surface 201 and the second surface 202 is the opposite surface; the phototransistors are discretely arranged in the first substrate 200; the filter layer is located on the first substrate 200, and each of the filter layers 200 is respectively arranged correspondingly to the phototransistors .
[0046] Wherein, each phototransistor includes a collector, a base and an emitter, and the projected area of the base 220 on the second surface 202 is smaller than the projected area of the collector 210 on the second surface 202, And the implantation depth of the base 220 in the first substrate 200 is smaller than the implantation depth of the collector 210 in the first substrate 200; the emitter 230 is on the second surface 202 The projection area of the base 220 is smaller than the projection area of the base 220 on the second surface 202, and the implantation depth of the emitter 230 in the first substrate 200 is smaller than that of the base 220 in the first substrate 202. The implantation depth in the substrate 200. Thus, the collector surrounds the base, which surrounds the emitter.
[0047] Furthermore, before the image sensor of the embodiment of the present invention forms the filter layer and the micro-lens, it is necessary to form a metal interconnection structure and bond with the substrate including the logic device.
[0048] Figure 5 to Figure 9 It is a schematic cross-sectional structure diagram corresponding to each step in the formation process of the image sensor using phototransistor in the present invention. The steps on how to form the phototransistor in the first substrate are described in the above-mentioned Figure 2 to Figure 4 It is introduced in the description of , and will not be repeated here. Wherein, the first substrate is used as a substrate of a pixel wafer and is mainly used to form optoelectronic devices, and the second substrate is used as a substrate of a logic wafer and is mainly used to form logic devices.
[0049] like Figure 5 As shown, an insulating stack 260 is formed on the second surface 202; metal plugs and metal wiring are sequentially formed in each layer of the insulating stack 260, wherein the topmost layer of metal connected to the emitter 230 The surface of the wiring 270 is flush with the top of the insulating stack 260 .
[0050] In this embodiment, the formation process of the insulating stack 260 and the metal plugs and metal wiring in each layer is as follows: firstly, a first insulating layer is formed on the second surface 202; then the first insulating layer is etched. layer, forming a dual damascene structure exposing the emitter 230, the base 220, and the collector 210 respectively; filling the dual damascene structure with conductive substances; forming a second insulating layer on the first insulating layer. Two insulating layers; etching the second insulating layer to expose the first insulating layer to form openings corresponding to each dual damascene structure; filling the openings with metal substances to form metal interconnection lines; continue to press The above steps form an insulating layer, form metal plugs in the insulating layer, and then form an insulating layer, and form metal interconnection lines corresponding to the corresponding metal plugs in the insulating layer until the requirements of the process are met.
[0051] like Image 6 As shown, a second substrate 300 is provided; the second substrate 300 includes a source follower (Source Follower) and a row selection (Row Select) region I, a reset (Reset) region II, and a transmission (Transfer) region III; An isolation region 320 is formed in the second substrate 300 for separation between each region; N is implanted in the second substrate 300 in the source follower and row selection region I and in the reset region II. Type ions are formed to form source follower and row selection region N well 330a and reset region N well 330b; P type ions are implanted into the second substrate in the transfer region III to form transfer region P well 330c.
[0052] In this embodiment, the material of the second substrate 300 is consistent with that of the first substrate.
[0053] In this embodiment, when forming the N well 330a in the source follower and row selection region and the N well 330b in the reset region, the transfer region III is covered with photoresist or other mask layers, and then N-type ions are implanted into the source follower and row selection region. Within the second substrate 300 are the selection region I and the reset region II. When forming the P well 330c in the transfer region, cover the source follower and row selection region I and the reset region II with photoresist or other mask layers, and then implant P-type ions into the second substrate 300 in the transfer region III Inside.
[0054] In other embodiments, the transfer region P well 330c may also be formed first, and then the source follower and row selection region N well 330a and the reset region N well 330b are formed.
[0055] In this embodiment, the N-type ions may be phosphorus ions or arsenic ions, the implantation energy may be 80keV-220keV, and the implantation dose may be 2e13/cm 2 ~3e15/cm 2. The P-type ions may be boron ions, the implantation energy may be 80keV-220keV, and the implantation dose may be 2e13/cm 2 ~3e15/cm 2.
[0056] like Figure 7As shown, a source follower gate (SF) 340a' and a row select gate (RS) 340a are respectively formed on the second substrate 300 in the source follower and row select region I, and in the reset region II A reset gate (RST) 340b is formed on the second substrate 300, and a transfer gate (TG, Transfer Gate) 340c is formed on the second substrate in the transfer region III; between the source follower gate 340a' and the The second substrate 300 on both sides of the row selection gate 340a forms a first deeply doped region 350a, and the source follower gate 340a' shares one of the first deeply doped regions 350a with the row selection gate 340a; The second substrate 300 on both sides of the reset gate 340b forms a second deeply doped region 350b; the second substrate 300 on both sides of the transfer gate 340c forms a third deeply doped region 350c .
[0057] In this embodiment, the source follower gate 340a', the row selection gate 340a, the reset gate 340b and the transmission gate 340c are formed simultaneously. The general process is as follows: a gate layer is formed on the second substrate 300; a photoresist layer is formed on the gate layer; the photoresist layer is exposed and developed to define a gate pattern; the photoresist layer is used as the mask, etch the gate layer to expose the second substrate 300, and form corresponding gates in each region.
[0058] In other embodiments, before forming the gate layer, a gate dielectric layer is formed on the second substrate 300 . When etching the gate layer, the gate dielectric layer is also etched to expose the second substrate 300 .
[0059] In other embodiments, spacers may also be formed on both sides of the gate after forming the gate.
[0060] In this embodiment, the first deeply doped region 350a and the second deeply doped region 350b are formed at the same time, and P-type ions are implanted, and the P-type ions may be boron ions; the third deeply doped region The region 350c can be formed before or after the first deeply doped region 350a and the second deeply doped region 350b are formed, and N-type ions are implanted, and the N-type ions can be phosphorus ions or arsenic ions.
[0061] In this embodiment, the first deeply doped region 350a, the second deeply doped region 350b, and the third deeply doped region 350c are distinguished as source or drain after voltage is applied. Wherein, the third deeply doped region 350c can be respectively used as a floating diffusion region (FD, Floating Difusion) and a temporary storage area (TSA, Temporary Storage Area), and the phototransistor can temporarily store the received optical signal into an electrical signal. Describe the temporary storage area.
[0062] like Figure 8 As shown, a dielectric stack is formed on the second substrate 300; metal plugs connected to each gate and each deeply doped region are formed in the dielectric stack; metal plugs connected to each of the metal plugs are formed. As for the metal interconnection, the surface of the topmost metal interconnection 370 connected to one of the third deeply doped regions 350 c (temporary storage region) of the transmission region III is flush with the top of the dielectric stack 360 .
[0063] In this embodiment, the process of forming the dielectric stack 360 and the metal plugs and metal wiring in each layer is similar to the process of forming the insulating stack 260 and the metal plugs and metal wiring in each layer. This will not be repeated here.
[0064] like Figure 9 As shown, the insulating stack 260 on the first substrate 200 is bonded to the dielectric stack 360 on the second substrate 300 to form an image sensor.
[0065] After bonding, the metal wires 270 in the insulating stack 260 are bonded to the metal wires 370 in the dielectric stack 360 . In this example, please refer to Figure 11 The metal wiring 270 and the metal wiring 370 function as a bonding pad (Bonding Pad), and its size may be greater than or equal to 0.2 μm, which is more conducive to bonding and interconnection. It should be noted that, by overlooking Figure 11 It can be seen that the source follower and the row selection region N-well 330a, the reset region N-well 330b, and the transfer region P-well 330c are not all on the same cross section. Here, the cross-sectional structure schematic diagram is misplaced for the convenience of representing the internal structure. deal with.
[0066] In this embodiment, the bonding process may specifically be a bonding method such as a plasma activated bonding method.
[0067] like Figure 10 As shown, the first surface 201 is thinned; a filter layer 280 is formed on the thinned first surface 201; a microlens 290 is formed on the surface of the filter layer 280, and the filter layer 280 and the microlens 290 correspond to the phototransistor, and external light enters the phototransistor through the microlens 290 and the filter layer 280 .
[0068] In this embodiment, the filter layer 280 is a red filter layer, a green filter layer or a blue filter layer, and the filter layer 280 of one color is formed above one phototransistor. The incident light entering the filter layer 280 is filtered by the filter layer to become monochromatic light (red light, green light or blue light), and then illuminates the phototransistor to excite electrons from the phototransistor. The microlens structure 290 is used to focus the incident light and converge the incident light onto the phototransistor.
[0069] During specific implementation, the red filter layer corresponds to the phototransistor in the red light absorption region, the green filter layer corresponds to the phototransistor in the green light absorption region, the blue filter layer corresponds to the phototransistor in the blue light absorption region, and the photoelectric transistor in the red light absorption region corresponds to the phototransistor in the red light absorption region. The absorption depth of the triode (relative to the depth of the light incident surface, that is, the first surface 201) is the deepest, and the absorption depth of the phototransistor in the green light absorption region is shallower (less than the absorption depth of the phototransistor in the red light absorption region), and the blue light absorption region. The absorption depth of the phototransistor is the shallowest (less than the absorption depth of the phototransistor in the green light absorption region).
[0070] Figure 12 It is the schematic diagram of the image sensor circuit of the present invention. like Figure 12 As shown, a certain voltage is added to the collector and base of the phototransistor to make it in an amplified state, and the external light passes through the R, G, and B filter layers to filter out corresponding optical signals; In the region, the electron-hole pairs generated by excitation increase the concentration of minority carriers, which greatly increases the reverse saturation current of the collector, completes photoelectric conversion, and generates photogenerated current, which is injected into the emitter for amplification and transmission To the temporary storage area (TSA); when the transfer gate (TG) is turned on, photoelectrons flow into the floating diffusion area (FD), and then perform subsequent image processing such as digital-to-analog conversion.
[0071] It can be seen from the above principles that the phototransistor can amplify the signal, so the image sensor using the phototransistor has higher sensitivity.
[0072] Although the present invention has been disclosed as above with a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can use the methods and technical contents disclosed above to analyze the present invention without departing from the spirit and scope of the present invention. Therefore, any simple modification, equivalent change and modification made to the above implementation methods according to the technical essence of the present invention, which do not depart from the content of the technical solution of the present invention, all belong to the technical solution of the present invention. protected range.
