WO2013129559A1 - 固体撮像装置 - Google Patents

固体撮像装置 Download PDF

Info

Publication number
WO2013129559A1
WO2013129559A1 PCT/JP2013/055364 JP2013055364W WO2013129559A1 WO 2013129559 A1 WO2013129559 A1 WO 2013129559A1 JP 2013055364 W JP2013055364 W JP 2013055364W WO 2013129559 A1 WO2013129559 A1 WO 2013129559A1
Authority
WO
WIPO (PCT)
Prior art keywords
charge
signal
pixel
electrode
transfer
Prior art date
Application number
PCT/JP2013/055364
Other languages
English (en)
French (fr)
Inventor
江藤 剛治
山田 哲生
ヴ・ツルオン・ソン,ダオ
Original Assignee
Etoh Takeharu
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Etoh Takeharu filed Critical Etoh Takeharu
Priority to JP2014502360A priority Critical patent/JP6188679B2/ja
Publication of WO2013129559A1 publication Critical patent/WO2013129559A1/ja
Priority to US14/467,605 priority patent/US9503663B2/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70SSIS architectures; Circuits associated therewith
    • H04N25/76Addressed sensors, e.g. MOS or CMOS sensors
    • H04N25/77Pixel circuitry, e.g. memories, A/D converters, pixel amplifiers, shared circuits or shared components
    • H04N25/771Pixel circuitry, e.g. memories, A/D converters, pixel amplifiers, shared circuits or shared components comprising storage means other than floating diffusion
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144Devices controlled by radiation
    • H01L27/146Imager structures
    • H01L27/14601Structural or functional details thereof
    • H01L27/1464Back illuminated imager structures
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144Devices controlled by radiation
    • H01L27/146Imager structures
    • H01L27/148Charge coupled imagers
    • H01L27/14806Structural or functional details thereof
    • H01L27/14812Special geometry or disposition of pixel-elements, address lines or gate-electrodes
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144Devices controlled by radiation
    • H01L27/146Imager structures
    • H01L27/148Charge coupled imagers
    • H01L27/14831Area CCD imagers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70SSIS architectures; Circuits associated therewith
    • H04N25/71Charge-coupled device [CCD] sensors; Charge-transfer registers specially adapted for CCD sensors
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70SSIS architectures; Circuits associated therewith
    • H04N25/71Charge-coupled device [CCD] sensors; Charge-transfer registers specially adapted for CCD sensors
    • H04N25/713Transfer or readout registers; Split readout registers or multiple readout registers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70SSIS architectures; Circuits associated therewith
    • H04N25/76Addressed sensors, e.g. MOS or CMOS sensors
    • H04N25/767Horizontal readout lines, multiplexers or registers

Definitions

  • the present invention relates to a structure of a solid-state image pickup device and an operation method thereof, and more particularly to a solid-state image pickup device and an image pickup system that take a plurality of images at an ultra high speed.
  • Non-patent document 1 discloses a technique for performing high-speed imaging by connecting a plurality of CCD analog memories to one pixel.
  • Non-Patent Document 2 discloses an improvement in which the same technology is applied to the backside illumination structure so that incident light is photoelectrically converted without waste and sensitivity is dramatically improved. This technology realized an ultra-high sensitivity video camera with 16 million frames / second for 165,000 pixels.
  • FIG. 1 shows a partial plane of the high-speed imaging device disclosed in Non-Patent Document 1
  • FIG. 2 shows a cross-section of the backside illumination structure disclosed in Non-Patent Document 2.
  • Reference numeral 1 in FIG. 1 denotes a signal charge collecting unit, which collects and accumulates the signal charge and then transfers the charge to the corresponding transfer stage of the CCD analog memory unit 2. In this state, it waits until the shooting time of the next frame. Let ti be the time required for these processes.
  • the CCD analog memory unit 2 has k transfer stages, and when a signal charge is transferred from the charge collection unit, it transfers only one stage in the transfer direction indicated by the arrow. Next, after the time ti elapses, the next signal charge is transferred from the charge collecting unit to the corresponding transfer stage of the CCD analog memory unit 2, and similarly, only one stage is transferred in the direction of the arrow.
  • the signal charge finally having k + j time difference ti is vertically read out from the CCD analog memory unit 2. It is held by the CCD 3.
  • Each pixel 9 includes a signal charge collecting unit 1 and an associated CCD analog memory, and charge collection and transfer are simultaneously performed in parallel in all pixels. Accordingly, the video signal charge group (one frame signal charge) of all the pixels 9 is recorded at the time interval ti, and is stored in the imaging device for k + j frames having the time difference ti.
  • the vertical readout CCD 3 is operated to transfer the signal charge in the direction of the arrow in the figure and sequentially to the horizontal transfer CCD 6.
  • horizontal read transfer transfer from right to left on the paper surface
  • a video signal of k + j frames (frames) having a time interval ti can be output to the outside.
  • FIG. 1 shows a substantially horizontal cut surface of the sheet of FIG. 1 from the signal charge collecting section to the CCD memory.
  • Reference numeral 11 denotes a signal charge collection unit
  • reference numeral 12 denotes a collection control electrode for controlling charge collection (hereinafter referred to as “charge collection electrode” or “collection electrode”)
  • reference numeral 13 denotes a CCD analog memory unit
  • reference numeral 14 denotes a CCD.
  • the transfer electrode, reference numeral 15 denotes an n-type transfer channel layer of the CCD, and reference numeral 16 denotes an element isolation layer, all of which are formed on the surface side of the semiconductor (Si).
  • a high-concentration p-layer 20 is formed on the back surface of Si, and the bulk forms a junction structure of a low-concentration p-layer 19 and a low-concentration n-layer, and is maintained in a strong reverse bias state. For this reason, almost all the carriers in the bulk are depleted to the outside, and an electric field is formed from the front side to the back side.
  • Light (electromagnetic wave) 21 incident from the back surface of Si is converted into a pair of electrons and holes in Si. At this time, the signal electrons are attracted to the front surface side by the electric field, and the holes move to the back surface and are discharged to the high concentration p layer 20.
  • a p-well barrier layer 17 is formed to prevent electrons drawn to the surface side from entering the CCD analog memory section. According to this structure, it is possible to collect almost all the photoelectrically converted electrons in the signal charge collecting unit 11. The time required for charge collection in this case is several nanoseconds or less, and a p-well barrier layer 17 is formed to prevent electrons attracted to the surface side from entering the CCD analog memory section. According to this structure, it is possible to collect almost all electrons photoelectrically converted by the charge collecting unit. The time required for charge collection in this case is several nanoseconds or less, and can be shortened to about 1 nanosecond.
  • the speed (frame / second or frame / second) of the conventional high-speed imaging apparatus described with reference to FIG. 1 is the reciprocal of the frame time interval ti. Therefore, how to shorten the frame time interval ti is a technical point of the ultra high-speed imaging device.
  • the frame time interval ti is expressed by the following formula (1).
  • tc is the charge collection time until the signal charge generated by the incident light reaches the charge collection unit
  • tt is the charge transfer time required for transfer from the charge collection unit to the memory
  • tw is until the next frame is imaged. It is a waiting time.
  • the time tc for the signal charge generated by the incident light to reach the charge collection unit can be shortened to several nanoseconds or less.
  • the transfer time to the CCD or the transfer time tt between the CCD transfer stages requires 10 nanoseconds or more, and generally several tens of nanoseconds.
  • Non-Patent Document 3 discloses an ultrahigh-speed image sensor using a CMOS circuit and a CMOS memory. An imaging apparatus equipped with this element achieves an imaging speed of 10 million frames / second for 100,000 pixels.
  • the characteristics of this image sensor are as follows. (1) 100 analog memories are formed for each pixel in the periphery of the image sensor chip. (2) On each pixel column, 1 ⁇ 4 of the signal readout lines of the number of pixels on the pixel column is arranged, and the image signal of all pixels of one image is transmitted around the element by only four signal readout operations. Transferred to the memory (outside the light receiving surface) and recorded. (3) The signal storage capacity is increased by configuring each analog memory with a combination of a normal CMOS capacitor and a capacitor made of two layers of polysilicon. (4) In addition, the introduction of a power supply circuit for each pixel and a device for noise cancellation have been introduced.
  • the charge transfer time tt is much longer than the charge collection time tc. Therefore, shortening the charge transfer time tt or eliminating the influence thereof is a technique for achieving further speedup.
  • Non-Patent Document 2 invented a special back element structure that can use the technique of Non-Patent Document 1 as it is, and the aperture ratio (ratio of the light receiving area of incident light to the pixel area) was set to 100%.
  • Non-Patent Document 4 discloses an image sensor technology that effectively photoelectrically converts most of light incident on a pixel by using a combination of an on-chip microlens and a laminated waveguide by front surface irradiation without using back surface irradiation. ing.
  • This image sensor is for a normal digital camera, but this technology can also be applied to a high-speed image sensor.
  • Eto et al. 1 million image / second image pickup device with oblique linear CCD type pixel peripheral recording area, Journal of the Institute of Image Information and Television Engineers, Vol. 56, no. 3, pp. 483-486, 2002.
  • Eto et al. 16 million images / second back-illuminated image sensor, Journal of the Institute of Image Information and Television Engineers, Vol. 65, no. 3, pp. 349-353, 2011.
  • Tochigi et al. Global shutter high-speed image sensor with a shooting speed of burst 1 Tpixel / s and continuous 780 Mpixel / s, Video Information Media Society Technical Report, Vol. 36, no. 18, pp. 9-12, 2012.
  • Each pixel has a single photoelectric conversion means, m ( ⁇ 3) charge collection means, and signal transfer for transferring the collected image signal (signal charge or a signal obtained by converting it into another signal such as a voltage).
  • An image pickup apparatus including M rows ⁇ N columns of pixels including means, wherein the photoelectric conversion means is not individually associated with m charge collection means, but is common to m charge collection means It is means for photoelectrically converting incident light incident on a large area corresponding to the majority (51% or more) of the area, and comprises means for sequentially collecting generated signal charges by m charge collecting means.
  • High-speed imaging is possible by sequentially making incident light incident on two solid-state image pickup means with a half mirror. According to this method, if the number of shots is two, incident light of 50% of the total pixel area can be photoelectrically converted, but if the number of shots is three or more, only incident light of about 33% or less of the total pixel area can be obtained. I can't do photoelectric conversion. The loss of incident light in high-speed imaging is fatal, and a conversion ratio of about 33% or less is unacceptable. In the present invention, by sequentially collecting signals from a single photoelectric conversion means for each pixel to a plurality of charge collection means for each pixel, the number of shots is 3 or more while ensuring a conversion efficiency of 51% or more. High-speed shooting can be realized.
  • Control of charge collecting means is independent and signal transfer means is common to each group
  • m unit pixel circuits are Charge collection means belonging to another group is divided into L groups of m / L (L ⁇ 2) unit pixel circuits, and the signal charges sequentially collected by the charge collection means belonging to one of these groups Means for simultaneously transferring signal charges while sequentially collecting them.
  • the signal transfer means operates at the same time for each group and transfers signals within the group all at once, so it must be shared within the group.
  • the impurity concentration of the barrier layer is low in the portion where the charge collecting means is localized.
  • the barrier between the photoelectric conversion layer and the circuit layer is an oxide film
  • the barrier layer may be an insulating material.
  • an oxide film is used as an insulating layer. A hole is formed in the oxide film layer at the position of the charge collecting means, and signal charges are collected through the hole.
  • the pixel shape can be densely spread as a regular hexagon.
  • the pixel shape is square, and one small square space remains for each pixel.
  • a control circuit common to the unit pixel circuit in each pixel is formed.
  • the photoelectric conversion means includes at least a waveguide and a photodiode.
  • Signal charges generated by light collected over the entire pixel area are collected by the charge collecting means A: Signal charges are transferred from the charge collecting means A while the remaining (m ⁇ 1) collecting means sequentially collect charges)
  • the signal charge collection time from the photoelectric conversion means to the charge collection means is tc
  • the transfer time for transferring the image signal from the collection means to the outside is tt
  • the frame time interval of continuous images is ti
  • the minimum frame time interval is Is tm.
  • the minimum frame time interval that can be theoretically realized is assumed to be tu.
  • signal charges may be collected by sequentially switching m charge collecting means for each collection time tc.
  • photographing can be performed with a minimum frame time interval that can be theoretically realized.
  • the number of continuous shots is limited to m.
  • the charge collecting means A After collecting the signal charge by one of the m charge collecting means, the charge collecting means A, and then collecting the signal charge sequentially while collecting the signal charge by the remaining (m ⁇ 1) charge collecting means. If a means for transferring to the outside of the means A and emptying the charge collecting means A is provided, the (m + 1) th image signal can also be collected by A. By repeating this, it is possible to capture m ⁇ (k + 1) frames of continuous images.
  • k is the number of image signal memories that are inside or outside each unit pixel circuit and belong to the unit pixel circuit. The reason why the value is (+1) is that image signals are also stored in the m charge collecting means.
  • the charge collecting means A can collect a new signal charge. That is, if 7tc ⁇ tt, since 7tc, the signal charge has been transferred from the charge collecting means A over tt time, so that a new signal charge can be collected at A.
  • a normal image sensor is used, and four pixel groups adjacent vertically and horizontally are regarded as one group pixel, and one pixel A out of the four pixels is used. If the signal charge is collected for one frame time interval, the charge generated during the next three frame time intervals is discharged from the drain, and this is repeated in succession by four pixels, it is possible to shoot at the frame time interval tc. It is. It is also possible to continuously transfer the image signal from the pixel A to the memory in the pixel or to the memory outside the pixel, with the signal charges being concentrated in the three pixels.
  • the light receiving area of the group pixel when shooting one frame is one pixel of the original pixel. Therefore, the aperture ratio is 1 ⁇ 4 or less of the original aperture ratio, and the sensitivity is 1 ⁇ 4 or less of the original sensitivity. In ultra-high-speed shooting, the decrease in sensitivity is fatal.
  • the aperture ratio can be increased to 80% or more, and the sensitivity is extremely high. High-speed shooting is possible.
  • Control of charge collecting means is independent and signal transfer means is common to each group
  • the charge collecting means / transfer means group is divided into two groups of X and Y, and signal charges (or image signals) sequentially collected by the charge collecting means belonging to group X are sequentially collected by the charge collecting means belonging to group Y. Transfer all at once.
  • the number of wirings for controlling the charge collecting means that is, m sets of charge collecting means is necessary.
  • the number of wirings required for charge transfer is only the number of groups, that is, only two types. Therefore, the number of wirings required for charge transfer and temporary storage is reduced from m sets to two sets. As a result, it is possible to effectively reduce the number of metal wirings on one pixel column and increase the number of pixels by reducing the pixel size while enjoying the high-speed shooting by transferring signals while sequentially collecting signal charges. be able to.
  • the time allocatable for transferring the group X signal charge is m / 2 times the charge collection time tc, that is, tc ⁇ (m / 2).
  • the time that can be allocated for transferring the signal charge when the image signal is transferred while collecting the charge by the other (m ⁇ 1) after collecting the charge by one charge collecting means is as described above.
  • Non-Patent Document 2 Charge is collected by selectively punching through the potential barrier between the photoelectric conversion layer and the charge collecting means.
  • the charge collecting means group is localized on the center side of each pixel, the following three effects can be obtained by increasing the concentration of the barrier layer around the pixel and decreasing it at the center portion.
  • a strong potential gradient is generated at the boundary between the photoelectric conversion layer and the barrier layer from the periphery of the pixel toward the center, and thereby the signal charge generated at the periphery of the pixel is directed toward the center of the pixel at high speed.
  • the barrier layer becomes stronger toward the periphery of the pixel, the probability of stray signal charges to the memory or the like built on the surface side decreases.
  • the center of the pixel means the center or centroid when the pixel is circular or regular polygonal. Further, when the charge collecting means is arranged along a horizontal or vertical center line with a rectangular pixel, it means a band-like region along the center line.
  • the barrier between the photoelectric conversion layer and the circuit layer is an oxide film
  • SOI Silicon on Insulator
  • a hole is formed in the oxide film layer at the position of the charge collecting means, and signal charges are collected through the hole. Charges are selectively collected by sequentially increasing the voltage of the electrode of the charge collection gate on the surface side of the hole. Also in this case, the transfer path can be depleted.
  • the m unit pixel circuits are laid out so as to form a regular polygon. For example, if the apex angle is 60 °, a set of six charge collecting means, transfer means, and other circuits (memory) can be spread to form a regular hexagonal pixel. In this case, hexagonal pixels can be spread without a free space.
  • the six charge collecting means are arranged at equal intervals on the circumference around the pixel center. Since the distance from the pixel periphery to the center is substantially equal, the charge collection time for each charge collection means is equal.
  • signal charges can be collected effectively by the mortar-shaped barrier layer.
  • the set of 8 circuit in the unit pixel is a regular octagon. If regular octagons are spread in a plane, the same number of small square surfaces as regular octagons remain. Normally, a circuit common to the eight unit pixel circuits is required.
  • CMOS readout circuit may be formed in a square area (see an example described later).
  • the actual pixel pitch is 1 / ⁇ 2 of the square arrangement. That is, there is an advantage that the real space resolving power is increased.
  • FIG. 3 is a detailed plan configuration diagram of a pixel according to the first embodiment.
  • FIG. 3 is a plan view showing first charge collection of each pixel in Example 1.
  • 7 is a plan view showing second charge collection of each pixel in Embodiment 1.
  • FIG. 6 is a plan view showing an eighth charge collection of each pixel in Example 1.
  • FIG. 10 is a plan view showing the sixteenth charge collection of each pixel in Example 1.
  • FIG. 3 is a partial plan view showing an electrode configuration and a control voltage feeder in Example 1.
  • FIG. 3 is a potential distribution diagram illustrating a charge collection principle and a charge transfer principle by a three-phase drive CCD in Example 1.
  • 3 is a drive timing chart for controlling charge collection and operation of a three-phase drive CCD in Embodiment 1.
  • FIG. 3 is a cross-sectional structure diagram in the horizontal direction centered on one pixel in Embodiment 1.
  • FIG. 6 is a cross-sectional structure diagram illustrating a thirteenth charge collection state of one pixel in Embodiment 1.
  • FIG. 6 is a cross-sectional structure diagram illustrating a 14th charge collection state of one pixel in Embodiment 1.
  • FIG. 3 is a cross-sectional structure diagram illustrating a state where charges of all pixels in Example 1 are discharged to the drain all at once.
  • FIG. 3 is a potential distribution diagram along the depth direction of the semiconductor showing the charge collection principle in Example 1;
  • FIG. 3 is a potential distribution diagram along the depth direction of the semiconductor showing the charge discharging principle in Example 1;
  • FIG. 3 is a drive timing diagram illustrating the entire operation sequence of the first embodiment. It is a fragmentary top view which shows the electrode structure in Example 2 of this invention, and the feeder of a control voltage. It is sectional structure figure of the horizontal direction centering on 1 pixel in Example 3 of this invention.
  • 1 is a system block diagram when the present invention is applied to an imaging system. It is a whole structure of Example 4 of this invention.
  • FIG. 3 is a cross-sectional structure diagram illustrating a state where charges of all pixels in Example 1 are discharged to the drain all at once.
  • FIG. 3 is a potential distribution diagram along the depth direction of the semiconductor showing
  • FIG. 6 is a plan structure diagram of a pixel in Example 4;
  • FIG. 10 is a diagram illustrating an operation principle of a pixel in Example 4. It is a plane structure fragmentary diagram of Example 5 of the present invention. The movement of signal charges in one pixel in the operation principle of the fifth embodiment will be described. The detailed planar structure of the pixel of Example 5 is shown. The transfer process is shown in detail among the operating principles of the fifth embodiment. It is a plane structure fragmentary diagram of Example 6 of the present invention.
  • FIG. 10 is a plan layout view of electrodes of Example 6. It is a figure which shows the mask for implants for producing
  • FIG. 10 is a plan layout view of electrodes of Example 6. It is a figure which shows the mask for implants for producing
  • FIG. 10 is a plan view of a potential distribution when a high voltage VH is applied to one of the charge collection electrodes of Example 6.
  • FIG. 10 is a cross-sectional view of a potential distribution when a high voltage VH is applied to one of the charge collection electrodes of Example 6. It is a plane layout view of Example 7 of the present invention. It is sectional drawing of Example 8 of this invention.
  • FIG. 10 is a cross-sectional view of a modified example of Example 8. It is sectional drawing of Example 9 of this invention. It is sectional drawing of Example 10 of this invention.
  • FIG. 3 shows a planar structure of the first embodiment of the present invention.
  • Reference numeral 31 indicates an imaging region in which M ⁇ N pixels are arranged, 32 surrounded by a small black thick frame indicates one pixel, and a so-called CCD charge including m ⁇ n charge collecting portions for each pixel 32.
  • a transfer stage is formed.
  • Reference numeral 33 denotes a horizontal transfer CCD that receives the signal charges transferred in parallel in the vertical direction of the arrow 35 from the imaging unit one by one and transfers them in the horizontal direction of the arrow 36.
  • Reference numeral 34 denotes an output circuit which receives the signal charge transferred from the horizontal transfer CCD and outputs it as a voltage signal.
  • Reference numeral 37 denotes an input terminal for a three-phase transfer clock pulse for vertical transfer of signal charges
  • reference numeral 38 denotes an input terminal for a horizontal transfer clock pulse for horizontal transfer of signal charges.
  • image light is incident from the back side of the paper surface of the imaging region, and a large amount of incident light is photoelectrically converted in the vicinity of the back surface and is collected by the charge collecting means on the front surface. It has the structure of.
  • FIG. 4 shows a partial planar structure for explaining the planar structure of Example 1 of the present invention in detail.
  • Reference numerals 41 and 42 indicate horizontal and vertical pitches for defining one pixel 32 surrounded by a thick black frame
  • numbers (1) to (16) surrounded by circles are assigned to each transfer stage (charge collecting means) in one pixel as shown in the figure (in the figure, these numbers are circled). However, in this description, this is indicated as a number in parentheses).
  • Reference numerals 45 and 47 denote transfer electrodes extending in the horizontal direction.
  • the transfer electrode 45 is supplied with a first-phase transfer pulse ⁇ 1 and the transfer electrode 47 is supplied with a third-phase transfer pulse ⁇ 3.
  • Reference numeral 46a denotes a transfer electrode to which a first transfer pulse of the second phase is supplied, and is formed electrically independently between (13), (14), (15), and (16). Transfer pulses ⁇ 21a, ⁇ 21b, ⁇ 21c, and ⁇ 21d are supplied.
  • Reference numeral 46b denotes a transfer electrode to which a second phase second transfer pulse is supplied, and is formed electrically independently between (9), (10), (11), and (12). Transfer pulses ⁇ 22a, ⁇ 22b, ⁇ 22c, and ⁇ 22d are supplied. Reference numeral 46c denotes a transfer electrode to which a second phase third transfer pulse is supplied, and is formed electrically independently between (5), (6), (7) and (8). Transfer pulses ⁇ 23a, ⁇ 23b, ⁇ 23c, and ⁇ 23d are supplied. Reference numeral 46d denotes a transfer electrode to which a second-phase fourth transfer pulse is supplied, and is formed between (1), (2), (3), and (4) electrically independently. Transfer pulses ⁇ 24a, ⁇ 24b, ⁇ 24c, and ⁇ 24d are supplied.
  • Reference numeral 48 supplies 18 types of transfer (control) pulses of ⁇ 1, ⁇ 21a, ⁇ 21b, ⁇ 21c, ⁇ 21d, ⁇ 22a, ⁇ 22b, ⁇ 22c, ⁇ 22d, ⁇ 23a, ⁇ 23b, ⁇ 23c, ⁇ 23d, ⁇ 24a, ⁇ 24b, ⁇ 24c, ⁇ 24d, and ⁇ 3.
  • the power supply terminal group for doing is shown.
  • Reference numeral 49 denotes an element isolation layer between the vertical transfer CCDs, and an excess charge drain (including a reset drain) is formed as necessary.
  • reference numeral 51 denotes the first pixel 32 (1)
  • reference numeral 52 denotes the second pixel 32 (1)
  • reference numeral 53 denotes the third pixel (1)
  • reference numeral 54 denotes the fourth pixel (1).
  • 55 indicates (1) of the seventh pixel, and shows a state in which the high-voltage charge collection control pulse ⁇ 24a is applied to (1) of all the first to ninth pixels 32 from the power supply terminal. .
  • the signal charge generated by the image light incident from the back surface as described above is distributed and accumulated substantially uniformly in the bulk region in each pixel (details will be described later). ).
  • the charge collection control pulse ⁇ 24a By applying the charge collection control pulse ⁇ 24a, the signal charges accumulated in the bulk of each pixel 32 are collected all at once by the charge collecting means shown in (1) of each pixel 32, and the charge transfer stage in the semiconductor surface region. Accumulated in. Let tc ( ⁇ ti) be the time until signal charges are accumulated in the bulk of each pixel 32 and the collection in (1) is completed.
  • Reference numerals 67 to 70 in FIG. 6 indicate the transfer stage (1) of each pixel 32 in the signal charge storage state after charge collection is completed.
  • a high voltage charge collection control pulse ⁇ 24b is applied to (2) of all the pixels from the power supply terminal.
  • reference numeral 61 denotes the first pixel (2)
  • 62 denotes the second pixel (2)
  • 63 denotes the third pixel (2)
  • 64 denotes the fourth pixel (2), 65.
  • FIG. 7 shows the state in which the above operation is sequentially executed from (3) to (7) of each pixel 32, and the charge collection pulse ⁇ 23d is applied from the power supply terminal in (3).
  • Reference numeral 71 is (8) of the first pixel 32
  • reference numeral 72 is (8) of the second pixel 32
  • reference numeral 73 is (8) of the third pixel 32
  • reference numeral 74 is (8) of the fourth pixel 32.
  • And 75 indicate signal charge collection states of the seventh pixel 32 in (8).
  • (1) to (7) of each pixel 32 indicated by reference numerals 76, 77, and 78 are in a signal charge accumulation (storage) state in which signal charge collection has already been completed.
  • FIG. 8 shows a state in which the above operation is further executed sequentially from (9) to (15) of each pixel 32, and the charge collection pulse ⁇ 21d is applied from the power supply terminal in (16).
  • Reference numeral 81 denotes (16) of the first pixel 32
  • reference numeral 82 denotes (16) of the second pixel 32
  • reference numeral 83 denotes (16) of the third pixel 32
  • reference numeral 84 denotes (16) of the fourth pixel 32.
  • 85 indicates each signal charge collection state in (16) of the seventh pixel 32.
  • (1) to (15) of each pixel 32 are in a signal charge storage state in which signal charge collection has already been completed.
  • the image signal charges are sequentially transferred in parallel in the vertical direction indicated by reference numeral 35 in FIG. 3, transferred line by line to the horizontal transfer CCD 33, and then output as a voltage signal from the output circuit 34 by the subsequent horizontal transfer.
  • the channel potential distribution and the movement of the signal charge from the charge collection process (16) to the charge transfer process in the vertical direction by the three-phase CCD drive are taken along the Y1-Y2 cross section (86, 87) indicated by the one-dot chain line in FIG.
  • Reference numeral 121 denotes a charge transfer electrode group
  • reference numeral 122 denotes an insulating film (generally SiO 2) formed on the semiconductor (Si).
  • Reference numeral 123 denotes the charge collection electrode of (12) that also serves as a transfer electrode
  • reference numeral 124 denotes the charge collection electrode of (8) that also serves as a transfer electrode
  • reference numeral 125 denotes the charge collection electrode of (4) that also serves as a transfer electrode
  • Reference numeral 126 denotes a charge collection electrode, and each of the charge collection electrodes of (16) also serves as a transfer electrode.
  • Each transfer electrode group is supplied with control pulses ⁇ 1, ⁇ 21d, ⁇ 22d, ⁇ 23d, ⁇ 24d, and ⁇ 3 (six types in this example) that perform transfer and charge collection as shown in the figure.
  • the operation timing of these six types of control pulses is shown in FIG.
  • Reference numeral 141 denotes ⁇ 1, 144 denotes ⁇ 21d
  • reference numeral 145 denotes a common pulse waveform of ⁇ 22d, ⁇ 23d, and ⁇ 24d
  • reference numeral 146 denotes a pulse waveform of ⁇ 3.
  • Reference numerals 142a, 142b, and 142c denote ⁇ 21a, ⁇ 21b, and ⁇ 21c, and transfer pulses that also serve as charge collection applied to electrodes adjacent in parallel to ⁇ 21d. That is, the charge collection of (13), (14), and (15) is controlled.
  • VM middle voltage
  • Reference numeral 148 indicates a low voltage (VL) in the pulse waveform, which is normally set to a predetermined negative voltage.
  • Reference numeral 149 indicates a high voltage (VH) in the pulse waveform, and is set to a sufficiently high positive voltage for charge collection. This high voltage corresponds to the charge collection control pulse.
  • Reference numeral 150 denotes a charge collection control pulse of (13), whereby signal charges are collected in (13).
  • Reference numeral 151 denotes a charge collection control pulse of (14), whereby the signal charge is collected in (14).
  • Reference numeral 152 denotes a charge collection control pulse of (15), whereby signal charges are collected in (15).
  • Reference numeral 153 is a charge collection control pulse of (16), whereby signal charges are collected in (16).
  • each of the charge collection pulses 150 to 153 is applied to each electrode with a time interval ti as shown in the figure. This time interval determines the shooting speed (frame rate) of the shot image.
  • Reference numeral 10 indicates a channel potential distribution at time t1 in FIG.
  • Reference numeral 134 denotes an M (middle) -level potential, a channel potential formed by applying a middle voltage VM to the transfer electrode directly above, and reference numeral 135 denotes an L (low) -level potential, A channel potential formed by applying the low voltage VL, reference numeral 133 indicates an H (high) -level potential, and a channel potential formed by applying the high voltage VH to the transfer electrode directly above.
  • Reference numeral 137 indicates a signal charge group that has completed the signal charge collection operation and is stored at the M-level potential of each transfer stage, and reference numeral 138 indicates the signal charge group (16) in the charge collection process by the H-level potential. Indicates.
  • Numeral 128 indicates the channel potential distribution and signal charge storage state at time t2 in FIG.
  • the charge collection process by the high level voltage VH of ⁇ 21d is completed, and all signal charges are stored at the M-level potential. From this state, vertical transfer (from right to left: the direction indicated by reference numeral 35 in FIG. 3) by the three-phase driving CCD operation is started. The principle of three-phase drive transfer will be briefly described below.
  • Reference numeral 129 in FIG. 10 indicates the channel potential distribution at time t3 in FIG. 11, ⁇ 3 changes from the low voltage VL to the middle voltage VM, the corresponding channel potential also changes to the M-level, and the signal charge is directly below the ⁇ 3 electrode. Accumulated across channels.
  • Reference numeral 130 denotes a channel potential distribution at time t4 in FIG. 11, and ⁇ 21d, ⁇ 22d, ⁇ 23d, and ⁇ 24d change from the middle voltage VM to the low voltage VL, and the corresponding channel potential shifts to the L-level. Due to this change, all signal charges move to and accumulate in the channel immediately below the ⁇ 3 electrode. Therefore, the signal charge is transferred by one transfer electrode in the process from time t2 to time t4.
  • Reference numeral 131 in FIG. 10 indicates the channel potential distribution at time t5 in FIG. 11, ⁇ 1 changes from the low voltage VL to the middle voltage VM, the corresponding channel potential also changes to the M-level, and the signal charge is directly below the ⁇ 1 electrode. Accumulated across channels.
  • Reference numeral 132 denotes a channel potential distribution at time t6 in FIG. 11, ⁇ 3 changes from the middle voltage VM to the low voltage VL, and the corresponding channel potential shifts to the L-level. Due to this change, all signal charges move to and accumulate in the channel immediately below the ⁇ 1 electrode. Accordingly, the signal charge is transferred by 2 transfer electrodes in the process from time t2 to time t6.
  • the first output time series by the first horizontal transfer is (1) (2) (3) (4) (1) (2) (3) (4)... (1) (2 ), (3) and (4) in the order of four times the number of pixels M (4 frames).
  • vertical transfer of one transfer stage is performed by the three-phase drive, and then (5) (6) (7) (8) (5) by the second horizontal transfer. ), (6), (7), (8),... (5), (6), (7), and (8).
  • FIG. 9 shows a more detailed partial planar structure in Example 1 of the present invention.
  • Reference numerals 91 and 94 denote transfer electrodes to which the transfer pulse ⁇ 1 is applied, and are formed in common to each vertical transfer channel.
  • Reference numerals 93 and 96 denote transfer electrodes to which the transfer pulse ⁇ 3 is applied, and are formed in common to each vertical transfer channel.
  • Numeral 92 is composed of electrically independent electrode groups 92a, 92b, 92c, and 92d for each vertical transfer channel, to which ⁇ 21a, ⁇ 21b, ⁇ 21c, and ⁇ 21d are applied, respectively.
  • Reference numeral 95 is composed of electrically independent electrode groups 95a, 95b, 95c, and 95d for each vertical transfer channel, to which ⁇ 22a, ⁇ 22b, ⁇ 22c, and ⁇ 22d are applied, respectively.
  • Reference numeral 97 denotes a separation part between vertical transfer channels, and 98 denotes a vertical transfer channel part.
  • Reference numerals 99 and 105 denote ⁇ 1 power supply lines, which are connected to the electrodes 91 and 94 via connection points (contacts) 111.
  • Reference numerals 100, 101, 102, and 103 denote ⁇ 21a, ⁇ 21b, ⁇ 21c, and ⁇ 21d feed lines, respectively, which are connected to the electrodes 92a and the flanges 92b, 92c, and 92d through the connection points 112, 113, 114, and 115, respectively.
  • Reference numerals 104 and 110 denote ⁇ 3 feed lines, which are connected to the electrodes 93 and 96 through the connection point 116.
  • Reference numerals 106, 107, 108, and 109 denote ⁇ 22a, ⁇ 22b, ⁇ 22c, and ⁇ 22d feed lines, respectively, which are connected to the electrodes 95a, 95b, 95c, and 95d through connection points.
  • control voltage (pulse) required in the first embodiment of the present invention can be supplied to all the electrodes.
  • the separation unit 97 plays a role of electrically separating the vertical transfer channels, but can function as an excessive or unnecessary charge discharging drain.
  • Reference numerals 117 and 118 in FIG. 9 indicate a planar structure when the drain function is added to the separating portion 97.
  • Reference numeral 118 is a drain layer composed of a high-concentration n layer, and a positive voltage is normally applied from the power supply terminal 119. Is done.
  • Reference numeral 117 is a p-layer, and is normally set to zero voltage (GND level), and plays a role of electrically separating the drain layer 118 and the n-channel layer.
  • GND level zero voltage
  • FIG. 12 the principle of signal charge collection in the first embodiment of the present invention will be described with reference to FIGS. 12, 13, 14, and 16.
  • FIG. 12 the principle of signal charge collection in the first embodiment of the present invention will be described with reference to FIGS. 12, 13, 14, and 16.
  • FIG. 12 shows a cross-sectional structure taken along the line Y3-Y4 shown in FIG.
  • Reference numerals 161, 162, 163, and 164 denote charge collection electrodes that also serve as transfer electrodes, to which control pulses ⁇ 21a, ⁇ 21b, ⁇ 21c, and ⁇ 21d are applied, respectively.
  • Reference numeral 185 denotes an insulating layer formed on the surface of the Si semiconductor, and is generally made of an Si oxide film (SiO 2 ).
  • Reference numerals 165, 166, 167, and 168 denote n-type transfer channel layers formed near the semiconductor surface under the corresponding electrodes.
  • Reference numerals 169, 170, and 171 denote high-concentration n-type drains formed between the channels.
  • P-type element isolation layers 172, 173, and 174 are formed between the n-type channel layers 165, 166, 167, and 168, respectively. ing.
  • Reference numeral 175 is a pixel determination separation layer for determining a region of one pixel, and electrically separates a signal storage layer between pixels.
  • Reference numerals 176 and 178 denote low-concentration n-type layers, in which a p-type barrier layer 177 is formed to prevent charges from moving to the CCD.
  • a p-type layer with a low concentration of 179 forms a pn junction with the low-concentration n-type layer 178.
  • Reference numeral 180 denotes a high-concentration p-type layer (p + layer) formed on the back surface of the Si semiconductor, and the junction depth from the back surface is formed extremely shallow.
  • the p-type element isolation layers 172, 173, 174, the pixel determining layer 175, and the barrier layer 177 formed on the front surface side are set to a reference potential, and a negative reverse bias voltage ⁇ VBB is applied to the high-concentration p + layer on the back surface. Is applied.
  • ⁇ VBB reverse bias voltage
  • the low-concentration pn junction layers 178 and 179 are almost completely depleted, and an electric field from the upper part to the lower part of the paper surface is formed by the space charges of the acceptor and donor ions.
  • electrons in the low concentration n layer 178 are discharged to the positive voltage n layer including drain layers 169 to 171 described later.
  • Reference numeral 184 denotes image incident light (electromagnetic wave), which enters the semiconductor from the back side of the semiconductor. That is, a back-illuminated image sensor is configured.
  • Reference numeral 181 denotes a region where incident light energy is photoelectrically converted to generate a hole-electron pair. As is well known, short-wavelength light with large photon energy is photoelectrically converted near the back surface, and long-wavelength light including near-infrared light with low photon energy penetrates from the back surface to a deep position while being attenuated, and is photoelectrically converted.
  • image incident light electrostatic wave
  • the generated hole-electron pair is separated by obtaining the electric field force, the hole moves to the P + layer 180 on the back surface and is absorbed, and the electron is attracted to the surface.
  • the barrier layer 177 suppresses movement of electrons to the surface layer. Therefore, electrons are accumulated and held in the pn junction region formed by the barrier layer 177 and the low-concentration n-type layer 178.
  • Reference numeral 182 represents a region of one pixel where electrons are accumulated as signal charges, and is referred to as a bulk accumulation layer here.
  • Reference numeral 183 denotes a signal charge bulk storage layer of an adjacent pixel. The bulk storage layer of the adjacent pixel is electrically separated by the deep junction p-type pixel determination layer 175, and signal charges are not mixed.
  • Reference numeral 186 indicates the horizontal dimension of one pixel.
  • the thickness of the semiconductor (Si) in Example 1 of the present invention is 50 ⁇ m. This is set to a thickness that can prevent incident light from entering the n layers 165 to 168 responsible for charge collection and charge transfer and generating noise charges. If the incident light is limited to visible light, about 30 ⁇ m is sufficient. However, when image light in a wavelength region including near infrared light is incident, a thickness of 50 ⁇ m or more is required.
  • the thickness of the semiconductor layer can be further reduced. Further, in order to increase the light collection and incidence efficiency on the back surface, it is preferable to form a microlens or an antireflection film.
  • a hatched area 193 in FIG. 13 represents a punch-through state.
  • the potential profile at this time is shown in FIG. 16, and details will be described later.
  • Reference numeral 191 denotes a ⁇ 21a electrode to which a charge collecting high voltage (H voltage) pulse is applied.
  • H voltage high voltage
  • the signal charge 194 of the bulk storage layer 182 is attracted to the punch-through region 193 where a high potential is formed and is instantaneously collected by the charge collection layer 192.
  • the high voltage of ⁇ 21a is released and the storage voltage becomes 0V
  • the bulk storage layer 182 and the collection region 192 are electrically disconnected, and the collected signal charge is stored and held in the charge collection layer 192.
  • Reference numeral 205 in FIG. 14 shows the charge collection layer immediately below ⁇ 21a in the accumulation state after the charge collection.
  • FIG. 14 shows the state of the cross section when ⁇ 21b has shifted to the charge collection high voltage (high voltage).
  • Reference numeral 201 denotes a ⁇ 21b electrode to which a high voltage is applied
  • reference numeral 202 denotes a charge collection layer that is at a high potential substantially in proportion to the increase in voltage of the electrode
  • 203 denotes a region in the punch-through state
  • reference numeral 204 denotes a bulk.
  • the signal charge of the storage layer is shown, drawn to the punch-through region 203 at a high potential, and instantaneously collected by the charge collection layer 202.
  • Signal charges can be collected and accumulated in all the charge collection layers arranged in the horizontal direction by sequentially applying charge collection high voltage (high voltage) pulses to the ⁇ 21c and ⁇ 21d electrodes.
  • the charge collection high voltage (high voltage VH) pulse is timed from the (1) electrode of ⁇ 24a to the (16) electrode of ⁇ 21d of each pixel.
  • VH high voltage
  • a depletion layer extends from the n + drain to the substrate side. This depletion layer completely depletes the corresponding region of the p-type barrier layer 177 and reaches the low concentration n-type layer. That is, a punch-through state in which the depletion layer penetrates the bulk storage layer 182 is obtained.
  • Reference numeral 212 denotes an area in a punch-through state.
  • the charge in the bulk accumulation layer is attracted to the punch-through region 212 in the same manner as or more than the charge collection process described above, and is instantaneously absorbed and discharged by the drains 169, 170, and 171.
  • Reference numeral 213 denotes a group of electrons absorbed by the drain. That is, the bulk storage layer 182 becomes empty and is reset to the initial state.
  • FIG. 16A shows a one-dimensional structure along the depth direction of the charge collection means.
  • Reference numeral 213 denotes a charge collection electrode
  • reference numeral 214 denotes an insulating film (oxide film)
  • reference numeral 215 denotes an n-type charge collection storage layer
  • reference numeral Reference numerals 216 and 218 denote low-concentration n ⁇ layers
  • reference numeral 217 denotes a p-type barrier layer
  • reference numeral 219 denotes a low-concentration p ⁇ layer
  • reference numeral 220 denotes a high-concentration p + layer formed on the back surface of the semiconductor.
  • FIG. 16 (b) shows a potential profile formed along (a).
  • Reference numeral 223 denotes a reference potential (0V), which means that the upper side of the page is a negative potential and the lower side is a positive potential with respect to the reference potential.
  • Reference numeral 224 indicates a negative voltage ( ⁇ VBB) applied to the p + layer 220 on the back surface, and determines the potential on the back surface of the semiconductor.
  • Reference numeral 225a denotes an electrode voltage, which is set to a VM voltage (0 V) in the signal charge accumulation state. In this state, the potential profile from the electrode 213 to the back surface 220 is represented by a curve 221.
  • the low-concentration n-layer 218 and the low-concentration p-layer 219 are in a completely depleted state, and the space charge is exposed.
  • Reference numeral 229 indicates the minimum potential of the barrier layer, which is held at the reference potential in the non-depleted state.
  • Reference numeral 226 denotes the maximum potential of the low-concentration n ⁇ layer 218, and the bulk storage layer is formed around this.
  • Reference numeral 232 denotes an electron flux that has been photoelectrically converted and stored in the bulk storage layer.
  • Reference numeral 222 indicates a potential profile from the electrode 213 to the back surface 220
  • reference numeral 225b indicates an electrode potential
  • reference numeral 227 indicates a maximum potential of the charge collection n layer 215
  • reference numeral 230 indicates a potential of the barrier layer 217.
  • a depletion layer extends from the n layer 215 in the depth direction, and the barrier layer 217 is completely depleted and coupled to the n ⁇ layer 218.
  • This profile represents the punch through. Since the barrier potential 229 changes to 230 and the barrier potential disappears, the charge 232 in the bulk accumulation layer moves to the collection layer 215.
  • a broken-line arrow 231 indicates a process of changing charges (electrons), and 233 indicates collected signal charges.
  • the signal charge in the bulk storage layer is selectively collected in the charge collection layer.
  • FIG. 17 shows the operation principle of discharging all charges of the bulk storage layer to the drain and resetting to the initial imaging state.
  • FIG. 17A shows a one-dimensional structure along the depth direction of the drain, and 261 shows an n-type high concentration drain layer.
  • Reference numerals 216, 217, 218, 219, and 220 are the same as those in FIG.
  • FIG. 17 (b) shows a potential profile formed along (a).
  • the drain regularly plays a role of discharging only an excess component of the signal charge transferred to the charge collection layer 215 (see FIG. 16), and is biased by VDR0 indicated by reference numeral 262.
  • the potential profile in this state is a curve denoted by reference numeral 264.
  • Reference numeral 268 indicates the barrier potential of the barrier layer 217, and reference numeral 267 indicates the potential of the bulk storage layer.
  • the barrier layer 217 is in a non-completely depleted state, and the reference numeral 268 is kept at the reference potential. Accordingly, the potential 267 of the bulk storage layer is also the same as the reference numeral 226 in FIG. Reference numeral 271 indicates a charge flux accumulated in the bulk accumulation layer.
  • Reference numeral 263 denotes a drain potential
  • reference numeral 269 denotes a barrier layer potential. Similar to the charge collection process in FIG. 16, the depletion layer extends from the drain layer 261, and the punch-through that reaches the bulk accumulation layer with the barrier layer 217 in a fully depleted state occurs. That is, the barrier potential of the barrier layer disappears, the charge bundle 271 moves instantaneously in the direction of the broken line arrow 270, is discharged to the drain, and the bulk storage layer becomes empty (initial state).
  • FIG. 18 shows an operation sequence of the first embodiment of the present invention based on the principle described above.
  • Reference numeral 241 denotes a ⁇ DR pulse for controlling the drain
  • reference numeral 242 denotes a ⁇ 24a pulse for controlling the first charge collection electrode
  • reference numerals 243, 244, 245, 246, and 247 denote the twelfth, thirteenth, and fourteenth, respectively.
  • the ⁇ 22d, ⁇ 21a, ⁇ 21b, ⁇ 21c, and ⁇ 21d pulses that control the th, fifteenth, and sixteenth charge collection electrodes are shown.
  • Reference numeral 255 denotes a high voltage pulse ⁇ DRH of ⁇ DR, which resets the imaging apparatus to an initial state.
  • Numeral 248 is a high voltage VH pulse for performing the first charge collection, and ends with a time interval of ti from the end time of ⁇ DRH.
  • Reference numerals 249, 250, 251, 252, and 253 are high voltage VH pulses for collecting the twelfth, thirteenth, fourteenth, fifteenth, and sixteenth charges, respectively, and the preceding high voltage VH end time. To ti after a time interval.
  • a video signal of M ⁇ N pixels that is applied with the three-phase drive vertical transfer pulse 254 shown in FIG. 10 and is imaged at a high speed of (1 / ti) frames / second as described above. Is output.
  • an appropriate potential barrier by a p-well and a ternary voltage of a high voltage VH for collecting charges, a middle voltage VM for storing charges, and a low voltage VL for transferring charges are applied to the collecting electrode.
  • This example consists only of a simple CCD structure. Therefore, a special backside illumination structure (shown in FIG. 2 and FIGS. 12 to 16) already realized in non-patent literature and the existing CCD technology are used to realize an imaging device capable of performing ultra-high-speed shooting immediately. be able to.
  • FIG. 19 shows a partial detailed structure of the planar structure according to the second embodiment of the present invention.
  • one transfer stage of a CCD that performs charge collection and charge transfer in the vertical direction is a so-called four-phase drive CCD having four electrodes as one unit.
  • Reference numeral 280 denotes a transfer electrode to which the transfer pulse ⁇ 1 is applied, and is formed in common to each vertical transfer channel.
  • Reference numeral 289 denotes a transfer electrode to which the transfer pulse ⁇ 4 is applied, and is formed in common to each vertical transfer channel.
  • Reference numerals 281, 282, 283, and 284 denote electrically independent electrode groups for each vertical transfer channel, to which charge collection transfer control pulses ⁇ 21a, ⁇ 21b, ⁇ 21c, and ⁇ 21d are applied, respectively.
  • Reference numerals 285, 286, 287, and 288 denote electrically independent electrode groups for each vertical transfer channel, to which ⁇ 31a, ⁇ 31b, ⁇ 31c, and ⁇ 31d are respectively applied.
  • 290 to 293 and 294 to 297 are also electrically independent electrode groups for each vertical transfer channel, and charge collection transfer control pulses ⁇ 22a to ⁇ 22d and ⁇ 32a to ⁇ 32d are respectively applied thereto.
  • Reference numeral 97 denotes a separation part between vertical transfer channels
  • reference numeral 98 denotes a vertical transfer channel part
  • Reference numeral 298 denotes a power supply terminal group for supplying a control pulse for charge collection and transfer.
  • a power supply line from each power supply terminal to each of the electrodes is indicated by a straight line, and a connection point (contact) is indicated by a black dot, which is the same as in FIG.
  • one transfer stage is composed of, for example, four electrodes ⁇ 1, ⁇ 21a, ⁇ 31a, and ⁇ 4, in this case, charge collection can be performed with two electrodes ⁇ 21a and ⁇ 31a. That is, charges are collected with an area of 1/2 of one transfer stage (in the first embodiment, an area of 1/3), so that the collection efficiency is increased. Further, since the signal charge is transferred by a well-known four-phase drive, the charge can always be accumulated in the two electrodes. When the length of one transfer stage is the same, the three-phase drive of the first embodiment is one electrode storage (1/3 area), and 1.5 times the storage (transfer) capacity is secured. can do.
  • FIG. 20 shows a third embodiment of the present invention.
  • a feature of the third embodiment is that a special energy conversion layer 300 is formed on the back surface of the semiconductor.
  • this conversion layer for example, an amplification film in which electrons generated by incident radiation are amplified by a well-known avalanche process and injected into the semiconductor back surface, the energy of incident radiation is absorbed to emit fluorescence, and the fluorescence is incident on the semiconductor.
  • a fluorescent film that generates signal charges by photoelectric conversion
  • an organic or inorganic photoelectric conversion layer that generates signal electrons by exposure to radiation in a predetermined wavelength range.
  • the bias conditions and the thickness of the film naturally vary depending on the type of the energy conversion film, and accordingly, the structure of the back surface including the electrode also varies.
  • the thicknesses of the semiconductor layers 178 and 179 can be extremely reduced depending on circumstances.
  • Embodiments of the present invention are not limited to the above description.
  • a charge discharge drain may be formed at a location different from that between the vertical transfer channels.
  • CMOS type Further, the charge collecting means in the present invention is not limited to the CCD. A so-called CMOS image sensor type may be used. In short, it is only necessary to have a plurality of charge collecting means in one pixel and independently collect the signal charge of one pixel time-divided by ti by independently controlling the charge collecting means at a predetermined time interval. In the case of the type, a barrier layer capable of controlling the voltage may be provided between the charge collection layer and the bulk storage layer.
  • the reading method may be a CMOS type instead of a CCD type.
  • signal transfer means comprising a CMOS circuit for transferring signal charges is provided in the subsequent stage of each of the charge collection means provided in each pixel.
  • the signal charge may be converted into a voltage and transferred as a voltage signal, or may be transferred as a digital signal by an AD converter.
  • a signal storage means and a signal reading means from the signal storage means to the outside of the pixel are provided after each signal transfer means.
  • the signal transfer means and the subsequent signal storage means are common.
  • CMOS type signal transfer means it is not necessary to separately provide a charge discharging stage.
  • each charge collecting means collects the charge once and then stores the k signals from the charge collecting means while the order of collecting the charges comes next. Moving to one of the means, the charge collection means can be emptied. By repeating this, m ⁇ (k + 1) continuous image signals can be recorded per pixel.
  • the arrangement and positional relationship of the m charge collecting means of each pixel are not limited to the case where they are arranged at the same spatial distance as in the first embodiment. You may arrange in a horizontal line.
  • the charge collecting means may be collected at the center of the pixel, and the transfer means and the signal storage means may be arranged radially.
  • the potential distribution toward the center of each pixel can be formed by reducing the thickness of the barrier layer on the charge collection means in the center of each pixel and increasing the periphery, and charge can be generated in a shorter time. Can collect and achieve higher shooting speed.
  • the signal storage means and the signal reading means at the subsequent stage may be formed on separate chips to form a stacked sensor.
  • Example 4 is characterized in that, among the items described in the “Other Embodiments” above, the following means described in the section “Signal storage means in the subsequent stage and signal storage outside the pixel” are provided. There is.
  • the signal After collecting the charge with one charge collecting means, while collecting the charge with the remaining (m-1) charge collecting means, the signal is transferred over several times the charge collecting time tc. To do.
  • the transfer time tt of the signal from the collecting means is four times or less than the collection time tc.
  • four of the eight charge collection means (referred to as a first group) collect charges sequentially at a time interval tc, and then the remaining four (second group) collect charges. Charges stored in the first group can be transferred to the signal storage means in the subsequent stage during 4 tc to be collected.
  • the charge collection means of the first group is empty, so that charges can be collected continuously at the time interval tc.
  • FIG. 22 shows the overall configuration of the fourth embodiment.
  • Reference numeral 341 denotes an imaging unit, which has a so-called back-illuminated structure. That is, it has a function of converting video light incident from the back side of the paper into signal charges, storing them, and further amplifying the current.
  • Reference numeral 342 denotes a control unit that supplies a signal for controlling the operation of a plurality of pixels constituting the imaging unit 341, and reference numeral 343 denotes an analog signal processing unit that performs amplification, low noise, and the like of an analog signal obtained from each pixel.
  • Reference numeral 344 denotes an AD conversion unit that converts an analog signal into a digital signal (AD conversion)
  • reference numeral 345 denotes a line memory unit that temporarily holds the converted digital signal
  • reference numeral 346 denotes a digital signal held in the line memory.
  • a communication unit 347 for transmitting to the outside in accordance with a predetermined standard (for example, a well-known LVDS system), and reference numeral 347 respectively indicate its output ports.
  • CMOS image sensors The configuration excluding the pixel portion is generally used in so-called CMOS image sensors, and need not be described.
  • FIG. 23 and FIG. 24 show the planar structure and operation principle of the pixel in Example 4 of the present invention.
  • Each block surrounded by a broken line (pixel boundary) indicates one pixel.
  • Reference numerals 351a, 351b, 351c, and 351d denote four electrically independent first charge collecting electrode groups, reference numeral 353 denotes a first charge transfer storage electrode, and reference numeral 355 denotes a third charge transfer storage electrode.
  • the signal charges collected under one of the charge collection electrodes 351a to 351d are received and accumulated.
  • Reference numerals 352a, 352b, 352c, and 352d are four electrically independent second charge collecting electrode groups, reference numeral 354 is a second charge transfer storage electrode, and reference numeral 356 is a fourth charge transfer storage electrode. The signal charges collected under the two charge collecting electrodes 352a to 352d are received and accumulated.
  • Reference numeral 357 schematically shows an output converter for converting the signal charge into a signal voltage and transmitting it to the analog signal processor.
  • Reference numerals 359 and 360 denote transfer control electrodes for selectively transferring signal charges from the third and fourth charge transfer electrodes 355 and 356 to the output converter 357, respectively.
  • Reference numeral 358 indicates a power supply terminal group for supplying an operating voltage to all the electrodes.
  • Numeral 362 denotes a separation unit for electrically separating the signal charge transfer region in the vertical direction. The operation principle of the fourth embodiment will be described below.
  • a charge collection pulse (high voltage VH) is sequentially applied to the first charge collection electrode group 351a, 351b, 351c, 351d with a time interval ti.
  • VH high voltage
  • a charge collection pulse (high voltage VH) is sequentially applied to the first charge collection electrode group 351a, 351b, 351c, 351d with a time interval ti.
  • (1) to (4) shown in FIG. 23 (in the figure, this number is enclosed in a circle, but in the present description this is indicated as a number in parentheses) is collected and temporarily stored in each electrode.
  • Signal charge In this case, signal charges are collected and temporarily accumulated in the first charge collection electrodes 351a to 351d with a time interval ti in the order of (1) ⁇ (2) ⁇ (3) ⁇ (4).
  • a charge collection pulse (high voltage VH) is sequentially applied to the second charge collection electrode groups 352a, 352b, 352c, 352d with a time interval ti, and the signal charge is changed from (5) to (6). Collected and temporarily stored in each of the second charge collection electrodes 352a to 352d with a time interval ti in the order of (7) ⁇ (8).
  • the time interval from the collection of the charge (4) to the collection of the charge (5) is also ti.
  • the signal charge group 372 of (1) to (4) temporarily stored in the first charge collection electrode during the period when the second collection electrode group is performing the collection operation is the first charge transfer storage electrode 353.
  • the period that can be allocated to this parallel transfer is 4 ti.
  • a charge collection pulse (VH) is sequentially applied to the first charge collection electrode groups 351a, 351b, 351c, and 351d with a time interval ti, and the signal charge group (9).
  • the signal charge groups 371 of (5) to (8) temporarily stored in the second charge collection electrode groups 352a to 352d are stored in the second charge collection electrode group 352a to 352d as indicated by reference numeral 374.
  • the signal charge groups (1) to (4) transferred in parallel to the first charge transfer storage electrode 353 and simultaneously stored in the first transfer storage electrode groups 351a to 351d are transferred in parallel to the third charge transfer storage electrode 355. And stored as indicated by reference numeral 376.
  • the time interval from when (8) is collected until (9) is collected is ti. Therefore, the period that can be allocated to these parallel transfers is 4ti.
  • signal charges (13) to (16) are collected as indicated by reference numeral 375 in FIG. 24C, and signal charges (9) to (12) are transferred in parallel as indicated by reference numeral 378.
  • the signal charges (17) to (20) are collected as indicated by reference numeral 379, and the signal charges of (13) to (16) are transferred in parallel as indicated by reference numeral 380.
  • the signal charges (21) to (24) are collected by the second charge collection electrode, and the signal charges (1) to (1) are respectively applied to all the electrodes in one pixel. 24) is accumulated.
  • the signal charges (1) to (24) become a video signal of one pixel in 24 frames (frames) with a time interval ti.
  • the signal charges (1) to (4) are first transferred to the output conversion unit 357, converted into an analog voltage signal, and transmitted to the analog signal processing unit. Subsequently, (5) to (8) are transferred to the output conversion unit 357 and converted into an analog voltage signal.
  • FIG. 24 (f) shows how signal charge groups are sequentially transferred to the output conversion unit 357. If each output converter 357 is composed of two independent circuits, (1) to (4) and (5) to (8) can be simultaneously transferred to the output converter. . A signal that is finally read out through such an operation is reproduced as an image of 24 frames ideally having a speed of (1 / ti) frames / second.
  • ultra-high-speed shooting at 200 million frames / second can be realized.
  • tt> 4ti so the actual frame time interval is tt / 4 (> ti).
  • the theoretical minimum frame time interval is (tt + ti) and the photographing speed is the reciprocal thereof, so that a photographing speed four times or more can be achieved.
  • Example 5 In the fifth embodiment, among the items described in the “other embodiments” above, in the “radial arrangement” section, the features are described in the section “signal storage section and signal storage outside the pixel”. As described above, the charge collecting means is collected at the center of the pixel, and the transfer means and the signal storage means are arranged radially outside the pixel.
  • the thickness of the barrier layer on the charge collecting means in the center of each pixel can be reduced and the periphery can be increased, and a potential distribution toward the center of each pixel can be formed, and the time can be reduced. Charges can be collected. This is also a major feature.
  • Example 5 The overall configuration of Example 5 is the same as that of Example 4 and is shown in FIG. Therefore, the explanation is omitted.
  • FIG. 25 shows a partial plan view of the fifth embodiment.
  • a region surrounded by a broken line denoted by reference numeral 391 indicates one pixel, and forms a square array inclined by 45 ° as illustrated.
  • Reference numeral 392 denotes a charge collecting electrode, which is composed of eight electrically independently controlled triangular electrodes.
  • Reference numeral 393 denotes a first charge transfer storage electrode
  • reference numeral 394 denotes a second charge transfer storage electrode
  • reference numeral 395 denotes a third charge transfer storage electrode.
  • the third charge transfer storage electrode 395 is composed of a plurality of electrodes constituting one transfer stage, and details will be described later with reference to FIG.
  • Reference numeral 396 denotes an output conversion unit
  • reference numeral 397 denotes a separation unit of each transfer channel for transferring signal charges in a radial direction.
  • the electrode group of each pixel has an octagonal shape, and each pixel is spread with its spatial phase shifted by 180 ° between the array rows.
  • Reference numeral 398 indicates the length p of one side of the pixel
  • reference numeral 399 indicates the horizontal pixel pitch in this pixel arrangement
  • reference numeral 400 indicates the vertical pixel pitch.
  • the pixel pitch in the horizontal and vertical directions is 1 ⁇ 2 with respect to the length of one side of the pixel. That is, it means that the spatial sampling pitch of the video is narrowed to 1 ⁇ 2, and the imaging resolution in the horizontal and vertical directions is increased by ⁇ 2 times (1.44 times). This is one of the features of the fifth embodiment.
  • the greatest feature of the fifth embodiment is that the charge collection electrode group 392 is locally arranged at the center of the pixel. This central localization enables structural improvements that have a significant effect on reducing the time required for charge collection. Specifically, it will be described later with reference to FIG.
  • FIGS. 26A to 26I show the movement of signal charges within one pixel.
  • the collection operation of the charge collection electrode 392 is performed four times in the order of (1) ⁇ (2) ⁇ (3) ⁇ (4) with an interval of time ti, and the signals of (1) to (4) Charge is collected under the collection electrode.
  • the electrode group that collects, transfers, and accumulates the signal charges (1) to (4) is defined as a first electrode group.
  • the signal charges of (5) to (8) are collected in the order of (5) ⁇ (6) ⁇ (7) ⁇ (8) as in the case of (a).
  • the signal charges (1) to (4) are transferred in parallel under the first charge transfer storage electrode of the first electrode group.
  • the time interval from the collection in (4) to the collection in (5) is also ti.
  • the electrode group that collects, transfers, and accumulates signal charges (5) to (8) is defined as a second electrode group.
  • the charge collection operation of the first electrode group is performed at the time interval, and signal charges are collected under the collection electrode in the order of (9) ⁇ (10) ⁇ (11) ⁇ (12).
  • (1) to (4) are transferred in parallel under the second charge transfer storage electrode 394 of the first electrode group, and (5) to (8) are transferred under the first charge storage electrode 393 of the second electrode group. Transferred in parallel.
  • the charge collection operation of the second electrode group is performed at the time interval, and signal charges are collected under the collection electrode in the order of (13) ⁇ (14) ⁇ (15) ⁇ (16).
  • (1) to (4) are transferred in parallel under the third charge transfer storage electrode 395 of the first electrode group, and (5) to (8) are under the second charge storage electrode 394 of the second electrode group.
  • (9) to (12) are transferred in parallel to the first charge transfer storage electrode 393 of the first electrode group.
  • the charge collection operation of the first electrode group is performed at the time interval, and the signal charges are collected under the collection electrode in the order of (17) ⁇ (18) ⁇ (19) ⁇ (20).
  • (5) to (8) are transferred in parallel under the third charge transfer storage electrode 395 of the second electrode group
  • (9) to (12) are under the second charge storage electrode 394 of the first electrode group.
  • (13) to (16) are transferred in parallel to the first charge transfer storage electrode 393 of the second electrode group.
  • the charge collection operation of the second electrode group is performed at the time interval, and signal charges are collected under the collection electrode in the order of (21) ⁇ (22) ⁇ (23) ⁇ (24).
  • (13) to (16) are transferred in parallel under the second charge transfer storage electrode 394 of the second electrode group, and (17) to (20) are under the first charge storage electrode 393 of the first electrode group. Are transferred in parallel.
  • the charge collection operation of the first electrode group is performed at the time interval, and signal charges are collected under the collection electrode in the order of (25) ⁇ (26) ⁇ (27) ⁇ (28).
  • (21) to (24) are transferred in parallel under the first charge transfer storage electrode of the second electrode group.
  • the charge collection operation of the second electrode group is performed at the time interval, and signal charges are collected under the collection electrode in the order of (29) ⁇ (30) ⁇ (31) ⁇ (32).
  • signal charges are accumulated under all the electrodes in the pixel. That is, signals for 32 frames having a time difference of ti are accumulated in each pixel.
  • the third charge transfer storage electrode 395 of the first electrode group and the second electrode group has the octagonal signal charge in the clockwise or counterclockwise direction along the electrode chain forming the outermost peripheral transfer path. It is possible to transfer the data to the output conversion unit 396 from below a predetermined electrode in the order of (1) ⁇ (2) ⁇ (3)... ⁇ (8). The voltage is converted into a voltage signal by the output converter and transmitted to the analog signal processor.
  • the signal charges (1) to (8) under the third charge transfer storage electrode 395 are all transferred to the output conversion unit through the clockwise transfer in this case, and the bottom of the third charge transfer storage electrode 395 is emptied.
  • the charges (9) to (16) under the second charge transfer storage electrode 394 are transferred all at once to the third charge transfer storage electrode 395 in parallel.
  • the signal charges (17) to (24) under one charge transfer storage electrode 393 are transferred in parallel under the second charge transfer storage electrode (394), and the charges (25) to (32) under the charge collection electrode 392 are transferred.
  • the charges are transferred in parallel under the first charge transfer storage electrode 393.
  • the charges (9) to (16) under the third transfer storage electrode are all transferred to the output converter through the clockwise transfer as in the case of (h). Further, the signal charges (17) to (24) are transferred in parallel below the empty third transfer storage electrode.
  • All signal charges (1) to (32) can be converted into voltage and transmitted to the analog signal processing unit.
  • FIG. 27 shows the detailed planar structure of the pixel.
  • Numerals 401 to 404 denote charge collection electrodes constituting the first electrode group (392 in FIGS. 25 and 26), and collection control pulses ⁇ c1 to ⁇ c4 are respectively supplied.
  • Reference numerals 405 to 408 denote charge collection electrodes constituting the second electrode group, and are supplied with collection control pulses ⁇ c5 to ⁇ c8, respectively.
  • Reference numeral 409a is supplied with the transfer control pulse ⁇ 1 by the first charge transfer storage electrode of the first electrode group
  • reference numeral 409b is supplied with the transfer control pulse ⁇ 2 by the first charge transfer storage electrode of the second electrode group. Is supplied with the transfer control pulse ⁇ 2 from the second charge transfer storage electrode of the first electrode group, and reference numeral 410b is supplied with the transfer control pulse ⁇ 1 from the second charge transfer storage electrode of the second electrode group.
  • Reference numerals 420, 411a, and 412a denote third charge transfer storage electrodes of the first electrode group
  • reference numerals 411b and 412b denote third charge transfer storage electrodes of the second electrode group, and transfer to the charge transfer storage electrodes 411a and 411b.
  • the control pulse ⁇ 31 is supplied to the charge transfer storage electrodes 420, 412a and 412b, respectively.
  • Reference numeral 413 denotes a known floating diffusion layer (hereinafter referred to as FD) that receives the signal charge transferred from the transfer electrode 420 and converts the charge amount into a voltage.
  • FD floating diffusion layer
  • a region surrounded by a broken line 414 indicates an output converter, in which a current amplifier circuit, an address circuit, a reset circuit, and the like that receive the potential of the FD are arranged.
  • a gray thick line 415 indicates an element isolation region for defining each charge transfer region that spreads radially.
  • Numeral 416 is a drain and is located at the center of the pixel.
  • Reference numeral 417 denotes an overflow control gate for causing excess charge under the collecting electrode to flow to the drain.
  • Broken lines denoted by reference numerals 418a and 418b are intra-pixel wiring examples of ⁇ 31 and ⁇ 32, and a small black dot 419 indicates a connection point with an electrode. The charge photoelectrically converted by the light incident from the back surface is accumulated in the Si bulk, but is absorbed by the drain by applying a predetermined high voltage to the drain 416 of each pixel.
  • the imaging operation starts from this absorption operation.
  • the state of charge transfer thereafter is as shown in FIG.
  • the third outermost charge transfer storage electrode of the pixel is composed of two electrodes 420, 411a, 411b, 412a, 412b to which ⁇ 31 and ⁇ 32 are supplied, and has a so-called two-phase drive type CCD structure. And ⁇ 31 and ⁇ 32 become two-phase transfer pulses, and in this case, charges are transferred clockwise in the outer peripheral transfer path.
  • the first and second transfer electrode groups also have a two-phase drive CCD structure, and basically transfer signal charges with two-phase transfer pulses. This transfer process will be described in detail with reference to FIG.
  • FIG. 28A shows a cut surface along the R1-R2 cut line shown in FIG.
  • Reference numeral 421 denotes a charge collection electrode (402 in FIG. 27) of the first electrode group
  • reference numeral 422 denotes a charge collection electrode (406 in FIG. 27), which are controlled by ⁇ c2 and ⁇ c4 pulses, respectively.
  • Reference numeral 423a denotes a first charge transfer electrode of the first electrode group
  • reference numeral 423b denotes a first charge transfer storage electrode of the second electrode group, which are controlled by ⁇ 1 and ⁇ 2 pulses, respectively.
  • Reference numeral 424a denotes a second charge transfer electrode of the first electrode group
  • reference numeral 424b denotes a second charge transfer storage electrode of the second electrode group, which are controlled by ⁇ 2 and ⁇ 1 pulses, respectively.
  • Reference numeral 425a indicates a third charge transfer electrode of the first electrode group
  • reference numeral 425b indicates a third charge transfer storage electrode of the second electrode group, both of which are controlled by a ⁇ 31 pulse.
  • Reference numeral 426 indicates an insulating film
  • reference numeral 427 indicates an n-type transfer channel layer
  • reference numeral 428 indicates an impurity region having a relatively low concentration relative to the n-type impurity concentration of the n-type transfer channel layer 427. Due to this impurity concentration distribution, a potential step is formed under each electrode, and a known two-phase drive CCD operation becomes possible.
  • Reference numeral 429 indicates a thick p-well layer
  • 430 indicates a thin p-well layer
  • 431 indicates a low-concentration n-type substrate.
  • FIG. 12 and the like there is a low-concentration p-type substrate further below, but this is omitted in the figure.
  • the p well and the n substrate are reverse-biased, and the photoelectrically converted electrons are attracted to the p well.
  • VH high voltage
  • a depletion layer extending from the n layer immediately below the collection electrode 421 enters a punch-through state that penetrates the thin p-well and reaches the n substrate.
  • Electrons flow into the n layer directly under the collecting electrode.
  • Reference numeral 443 indicates a punch-through region
  • reference numeral 441 indicates movement of electrons moving to the n layer immediately below the collecting electrode.
  • FIG. 28B shows the potential distribution of the n-type channel layer at each operation timing.
  • Reference numeral 432 indicates a state in which the first collection pulse is applied to the collection electrode 421 described above.
  • Reference numeral 440 indicates the potential HC of the n layer in a state where the high voltage acquisition pulse (VH) ⁇ c2 is applied, and reference numeral 439 indicates the collected signal charge (2).
  • L in the figure indicates the n-type channel potential in a state where a low voltage (VL) is applied to each electrode and H is a high voltage (VH).
  • Reference numeral 433 denotes a potential distribution of the channel layer in a state in which the collection pulse ⁇ c6 is applied to the charge collection electrode 422 facing the charge collection electrode 421, and shows a state in which the signal charge (6) is collected immediately below the charge collection electrode 422. .
  • the charge collection electrode 421 is shifted to the low voltage VL and the electrode 432a is shifted to the high voltage VH, whereby the signal charge (2) is transferred directly below the electrode 423a.
  • Reference numeral 434 denotes a potential distribution of the channel layer in a state where the collection pulse ⁇ c2 is applied to the charge collection electrode 421 again, and shows a state where the signal charge (10) is collected directly under the collection electrode 421.
  • the charge collection electrode 422 is shifted to the low voltage VL
  • the electrode 432b is shifted to the high voltage VH
  • the charge (6) is transferred immediately below the charge transfer electrode 423b
  • the charge transfer storage electrode 423a is transferred to the low voltage VL
  • the charge transfer electrode 424a is transferred.
  • Reference numeral 435 denotes a potential distribution of the channel layer in a state where the collection pulse ⁇ c2 is applied to the charge collection electrode 422 again, and shows a state in which the signal charge (14) is collected immediately below the charge collection electrode 422.
  • the charge collection electrode 421 shifts to the low voltage VL
  • the charge transfer storage electrode 423a shifts to the high voltage VH
  • the charge (10) is transferred directly below the charge transfer storage electrode 423a
  • the charge transfer storage electrode 423b is transferred to the low voltage VL, 424b. Shifts to the high voltage VH
  • the charge (6) is transferred directly below the electrode 424b.
  • the charge transfer storage electrode 424a shifts to the low voltage VL and the electrode 425a shifts to the high voltage VH
  • the charge (2) is transferred directly below the 425a.
  • Reference numeral 436 denotes a potential distribution of the channel layer in a state where the collection pulse ⁇ c1 is applied to the collection electrode 421 again, and shows a state in which the signal charge (18) is collected immediately below the charge collection electrode 421.
  • the voltage collecting electrode 422 is shifted to the low voltage VL
  • the charge transfer electrode 423b is shifted to the high voltage VH
  • the charge (14) is transferred immediately below the charge transfer electrode 423b
  • the charge transfer electrode 423a is transferred to the low voltage VL and the charge transfer electrode 424a. Shift to the high voltage VH, and the charge (10) is transferred directly below the charge transfer electrode 424a.
  • the charge transfer electrode 424b shifts to the row electrode VL and the charge transfer electrode 425b shifts to the high voltage VH
  • the charge (6) is transferred directly below the charge transfer electrode 425b.
  • Reference numeral 437 denotes a potential distribution of the channel layer in a state where the collection pulse ⁇ c2 is applied again to the collection electrode 422, and shows a state in which the signal charge (18) is collected immediately below the charge collection electrode 422.
  • the charge collection electrode 421 is shifted to the low voltage VL
  • the charge transfer electrode 423a is shifted to the high voltage VH
  • the charge (18) is transferred immediately below the charge transfer electrode 423a
  • the charge transfer electrode 423b is transferred to the low voltage VL and the charge transfer electrode 424b. Shifts to the high voltage VH, and the charge (14) is transferred immediately below the charge transfer electrode 424b.
  • Reference numeral 438 denotes a potential distribution of the channel layer in a state where the collection pulse ⁇ c1 is applied to the collection electrode 421 again, and shows a state in which the signal charge (26) is collected immediately below the collection electrode 421. At the same time, the collection electrode 422 is shifted to the low voltage VL, the charge transfer electrode 423b is shifted to the high voltage VH, and the charge (22) is transferred immediately below the charge transfer electrode 423b.
  • the charge is collected directly under the collection electrode 422 by shifting the collection electrode 421 from the charge collection potential to the high voltage VH and applying the charge collection pulse ⁇ c6 to the collection electrode 422.
  • the collection electrode 442 corresponds to the pixel drain region.
  • the charge transfer is described with a simple element isolation layer for the purpose of explaining charge transfer.
  • Example 5 charges are induced in a low potential barrier portion opened in the center of the p-well, and are distributed directly by an element isolation layer (or drain) that is maintained at a lower potential than the collection electrode portion during imaging.
  • the charges generated by the incident light incident on the most area of each pixel area can be sequentially collected by different charge collecting means by being sent to the surrounding collecting electrodes.
  • FIG. 30 shows a planar arrangement of electrodes including the charge collection electrodes 502a to 502h.
  • the pixel size is 10 ⁇ m per side.
  • the thickness of the chip is 50 ⁇ m.
  • the substrate before the process is composed of double silicon single crystal layers of low concentration p-type and low concentration n-type produced by epitaxial growth. 12 ⁇ m from the surface is an n-type substrate with a concentration of 2 ⁇ 10 ** 14 (1 / cm ** 3), and the remaining 38 ⁇ m is a p-type substrate with a concentration of 2 ⁇ 10 ** 13 (1 / cm ** 3) It is.
  • “**” indicates “power”. For example, 10 ** 14 represents 10 to the 14th power.
  • FIG. 31 shows an implant 501 for forming a p-well (reference numeral 429 in FIG. 28).
  • the p-well forms a potential barrier that prevents the signal charges (electrons in this example) from the back surface to the front surface from entering under the transfer storage electrode, and also functions to induce the signal charges to the center of the pixel.
  • the first charge collection electrodes 502a to 502h are shown as charge collection electrodes 401 to 408 in FIG. 27 and charge collection electrodes 421 and 422 in FIG. 32 and 33 are a plan view and a cross-sectional view of a potential distribution when a high voltage VH is applied to one of the charge collection electrodes 502a.
  • Both figures also show a path (504 in this case) of charges (in this case, electrons) toward the collection electrode when signal charges are generated at a position 503 on the back surface side of the farthest end point when viewed from the charge collection electrode 502a in the center of the pixel. ing.
  • charges in this case, electrons
  • the implant 501 for forming the p-well in FIG. 31 is formed by two implantations of boron ions and subsequent thermal diffusion.
  • the mask shape 505 at the time of the first layer implantation is uniform, and a hole 507 is opened at the center.
  • the mask shape 506 at the time of the second implantation consists of a collection of radial thin triangles from the periphery toward the center. Although it looks linear, due to thermal diffusion after ion implantation, a smooth ion concentration distribution shape with a dark periphery and a thin center is formed, which leads to a smooth potential distribution toward the center that is formed, resulting in a short signal charge. Collected under the collecting electrode in time.
  • Reference numeral 508 denotes a mask (element isolation region 415 in FIG. 27) for forming a channel isolation layer, and high-concentration boron ions are implanted with low energy using this mask.
  • a dense p-type ion (boron ion) having a thickness of 0.1 ⁇ m is implanted on the back surface and is instantaneously activated by an ultraviolet ultrashort pulse laser (not shown in the figure). This is the same as the high-concentration p-type layer 180 in FIG.
  • the large reverse bias voltage between the back surface and the charge collection electrode is depleted except for the layers in the vicinity of the back surface and the surface, and the generated charge is substantially perpendicular to the potential distribution at high speed on the surface side. Collected on the charge collection electrode.
  • the collection electrode group was divided into two groups.
  • the imaging device can be driven with a relatively small number of metal wires.
  • the charge is collected from the collecting electrode A to the memory region associated therewith during the period 7 tc in which the charge is collected by the other collecting electrode after collecting the charge by A. Need to be transported. Usually, the transfer time tt required for this is longer than 7tc. Therefore, the frame rate is (1 / (tt + tc / 7)) frames / second.
  • FIG. 29 shows a partial plan view of a sixth embodiment of the present invention.
  • a feature of the sixth embodiment is that an electrode group of one pixel forms an octagon and pixels are arranged in a square.
  • a region surrounded by a broken line 451 indicates one pixel.
  • an output conversion unit can be arranged in the empty area 454 in which octagonal electrode groups are spread to convert a signal into a voltage.
  • the chain of arrows 452 indicates the first transfer path
  • the chain of arrows 453 indicates the signal charge transfer direction of the second transfer path.
  • Reference numeral 455 indicates signal charges collected by the charge collecting electrode, and is collected in the order of (1) ⁇ (2) ⁇ ... ⁇ (7) ⁇ (8).
  • the structure of the sixth embodiment is used, it is easy to read out the signal charge outside the light receiving surface with only the CCD after photographing. Thereby, the low noise characteristic of CCD can be enjoyed. The effect is particularly great when cooling.
  • Example 7 A plan view of Example 7 is shown in FIG.
  • a feature of the seventh embodiment is that charges collected by the charge collecting electrode 510 are transferred to a CMOS memory (not shown in the drawing) outside the light receiving surface by the CMOS readout circuit 511 instead of the CCD memory in the pixel. Is to be preserved.
  • Reference numeral 512 denotes an output gate adjacent to the charge collection electrode.
  • Reference numeral 513 denotes an FD (floating diffusion layer) adjacent to the output gate.
  • the voltage of the electrode of the output gate is a value VM between the high voltage VH and the low voltage VL of the collecting electrode voltage, and is constant.
  • the voltage of the charge collection electrode collects the charge at the high voltage VH and then falls to VL, the collected charge is transferred to the FD over the potential barrier of the output gate. Therefore, the voltage is converted and transferred to a CMOS memory outside the light receiving surface.
  • Non-Patent Document 3 An example of such high functionality is disclosed in Non-Patent Document 3 and the like.
  • the memory in the case where the memory is used as it is with the CCD, in principle, complete charge transfer can be performed.
  • signals can be transferred and stored substantially without noise.
  • the signal is sent to the buffer memory outside the image sensor and stored, so that the signal can be amplified and transferred sufficiently high.
  • Non-Patent Document 3 in order to record the amplified signal in a small area, a large-capacity capacitor combining a normal CMOS capacitor and a capacitor between two layers of polysilicon electrodes is used as an analog signal recording element. Yes.
  • a second chip may be bonded to the back surface, a memory area may be secured on the second chip, and signals may be transferred to the surface by vertical wiring. Needless to say. In this case, not only the memory but also a power supply circuit or the like can be placed on the second chip.
  • FIG. 35 shows a sectional view of the eighth embodiment.
  • a thin p-layer is formed on the back surface.
  • the charge generation layer is composed of a low concentration thick p layer 514.
  • a feature of this example is that there is a first insulating layer 516 made of silicon oxide between the charge generation layer 514 on the back side and the charge collection layer 515 on the front side. This is a so-called SOI (Silicon-on-Insulator) structure.
  • SOI Silicon-on-Insulator
  • a hole 518 is opened at the center of the insulating layer of each pixel 517.
  • a mushroom-shaped thin n layer 519 is formed around the hole in order to induce charges to the central hole.
  • the charge collection layer 515 is n-type single crystal silicon.
  • a thick n layer 520 for channel formation is formed in a thin layer near the surface.
  • the FD 521 is formed with a very dark n layer for ohmic contact. Also, the pixels are separated by deep p-doping 522.
  • Reference numeral 523 denotes a second insulating layer made of silicon oxide. Furthermore, there is an electrode layer 524 made of polysilicon on the surface side, and a metal wiring layer 525 is formed on the surface side. The charge collection layer 515, the second silicon layer 523, the electrode layer 524, and the metal wiring layer 525 are combined to constitute a circuit layer 526.
  • planar structure of the circuit layer is the same as that of Example 7 shown in FIG. Since the voltage applied to the electrodes is the same, detailed description is omitted.
  • the center electrode (just below the hole in the figure) is not a drain but an induction electrode 527.
  • an induction electrode 527 On both sides (planarly on the circumference of the induction electrode), there is a charge collecting electrode 528, an output gate electrode 529 is formed outside, and an FD 521 and a CMOS readout circuit 530 are formed outside.
  • a light shielding layer 531 made of a tungsten layer is formed on the surface side of the insulating layer. This almost completely prevents a part of the incident light from directly entering the recording portion.
  • FIG. 36 shows a modification of the eighth embodiment.
  • the feature is that holes 518 for collecting electric charges are formed through the first insulating layer 516 on the back surface side (upper side in the figure) of the charge collecting electrodes (the number of the charge collecting electrodes is the center). 8 on the circumference of 532).
  • FIG. 37 shows a cross-sectional structure of the pixel of Example 9.
  • the surface irradiation element has a laminated waveguide structure 535 including a microlens 533 and a light guide 534 formed on the surface.
  • a photodiode 536 is formed immediately below the light guide (center portion of the pixel). The light incident on the surface is collected at the central portion of the pixel by the on-chip microlens, and further collected at the central portion of the pixel by the light guide.
  • planar structure of the layer below it is similar to that of the seventh embodiment shown in FIG.
  • incident light that has entered a large area of each pixel area is photoelectrically converted by the laminated waveguide structure.
  • charge collection electrodes 528 surrounding the photodiode, and when a high voltage VH is applied to one of them, the image signal is collected at the collection gate. Subsequent charge transfer and storage methods are the same as in the seventh embodiment.
  • This charge collecting electrode structure and control method and the laminated waveguide structure can sequentially guide the signal charge generated in the entire area of the pixel to only one of the designated collecting means.
  • Means for transferring an image signal from A to an image signal recording means belonging to A while signal charges are collected by one of the eight collecting means A and then sequentially collected by the remaining seven collecting means Is the same as in the seventh embodiment.
  • the advantage of the ninth embodiment is that it can be produced by a process of creating a normal surface irradiation image sensor.
  • the ninth embodiment is a front surface irradiation element, it is difficult to send a signal to the back surface. Therefore, a plurality of chips are stacked, and a photoelectric conversion means, charge collection means, and transfer means are placed on the first chip on which light is incident, and signal charges are accumulated on the second chip bonded to the back surface. Is difficult. It is also difficult to drive a current collecting circuit or a driving unit on the first chip by mounting a current source circuit or a driving circuit on the second chip in units of pixels or pixel blocks.
  • the tenth embodiment shown in FIG. 38 is a method using direct entry of MCP (microchannel plate) and acceleration electrons.
  • reference numeral 537 is a glass entrance window
  • reference numeral 538 is a photocathode
  • reference numeral 539 is an MCP
  • reference numeral 540 is an image sensor
  • reference numeral 541 is a light receiving surface on the back of the image sensor
  • reference numeral 542 is a package
  • reference numeral 543 is a vacuum seal.
  • a chamber for stopping, a reference numeral 544 is a pin for drawing a wiring from the vacuum chamber,
  • a reference numeral 545 is a socket for receiving (inserting) the pin, and
  • a reference numeral 546 is a circuit board.
  • Incident light 547 is converted into an electron beam 548 on the photoelectric conversion surface and enters the MCP. After electron multiplication by the MCP, an electron current 549 accelerated by a strong electric field applied between the MCP emission surface and the light receiving surface 541 of the image sensor is formed, enters the light receiving surface, and more secondary electrons are generated by impact ionization. Is generated.
  • ⁇ MCP is not required. At this time, the signal is amplified by impact ionization multiplication by direct entry of accelerated electrons.
  • Example 11 Selective Charge Collection by Transistor Embodiments of the present invention are not limited to the above examples.
  • the signal charge collected at the center of the pixel may be collected on the charge collection electrode by turning on / off a transistor arranged around the signal charge.
  • the gate voltage difference necessary for turning on and off the transistor may be 3 V or less.
  • other functions such as amplification can be added relatively easily.
  • a selection transistor may be inserted instead of the charge collection gate 510 and the output gate 512, and eight second address wirings may be added (not shown).
  • Example 12 Shared use of floating diffusion layer and charge collection electrode
  • a voltage higher than the voltage of the other floating diffusion layers is applied to one of the floating diffusion layers 513 to collect charges.
  • a high voltage is applied to the drain 552 with the reset gate 551 opened.
  • the signal charge continues to flow to the drain.
  • the potential of the floating diffusion layer is kept constant.
  • the reset gate is closed in this state, signal charges continue to flow into the floating diffusion layer. For this reason, when the signal charge is electrons, the potential of the floating diffusion layer continues to decrease. Therefore, CDS (correlated double sampling) is possible by detecting the potential after a certain time after closing the reset gate.
  • the drain voltage is supplied from a ring-shaped wiring 550 around the pixel. Therefore, the same voltage is supplied to the eight drains.
  • wiring for supplying drain voltages individually to eight drains is provided (not shown in the figure).
  • FIG. 21 shows a specific example of the imaging system (camera) of the present invention to which the imaging apparatus of the present invention is applied.
  • Reference numeral 311 denotes a solid-state imaging device of the present invention
  • reference numeral 312 denotes a lens system (condensing optical system in the case of light) for irradiating various imaging radiations (including video light) to the solid-state imaging device 311
  • reference numeral 313 denotes
  • 314 indicates a drive circuit for directly driving the image pickup device in response to an operation control signal from the AFE circuit.
  • the control timing generation circuit 313 and the drive circuit 314 are not necessary.
  • Reference numeral 319 is an enlarged view of the image pickup apparatus 311.
  • the signal output time series is in the order of reference numeral 320.
  • Reference numerals 322, 323, and 324 surrounded by broken lines indicate the first, second, and third pixels in the first row, and continue to the Mth.
  • Reference numerals 325, 326, and 327 denote pixels in the second row. Since it continues to the Nth row as a whole, a signal of M ⁇ N pixels is output in a time series of 320. Each pixel is composed of signals of 16 frames (1) to (16).
  • the signal of reference numeral 320 is input to a reference numeral 315 (rearrangement storage) having a time series rearrangement function, and converted into an array of 321.
  • Reference numerals 328 and 329 denote an image signal of the first frame (frame), an image signal of the second frame, reference numeral 330 denotes an image signal of the 16th frame, and each frame is composed of a signal of M ⁇ N pixels.
  • DSP digital signal processing circuit
  • the DSP 316 performs filtering and interpolation processing of each frame signal, correlation processing between frames, and the like, and transfers the signal to a circuit block 317 for storage or display. With this video output, it is possible to observe and reproduce a video shot at an ultra-high speed of (1 / ti) frames / second.
  • the high-speed solid-state imaging device and its operation method using the present invention are used in the industry for manufacturing solid-state imaging devices, and high-speed imaging cameras equipped with the same are widely used in the field of science and technology.

Abstract

 固体撮像素子は、M行×N列の画素391を備える。個々の画素391は、入射線の入射により電荷を発生する単一の電荷収集電極(光電変換手段)392と、電中収集電極392で生成した電荷を収集するm(≧3)個の電荷転送蓄積電極(信号収集手段)393と、電荷転送蓄積電極393で収集された電荷を電荷転送蓄積電極393から転送する電荷転送蓄積電極(信号転送手段)394,395を備える。各画素面積に入射した入射線の51%以上が電荷収集電極(光電変換手段)392により電荷に変換される。各画素391の電荷収集電極(光電変換手段)392で発生した電荷は各画素内の異なる電荷転送蓄積電極(信号収集手段)393に順次収集される。

Description

固体撮像装置
 本発明は、固体撮像装置の構造及びその動作方法に関し、特に複数枚の画像を超高速度で撮影する固体撮像装置及び撮像システムに関する。
 1画素に複数のCCDアナログメモリを接続して高速撮像を行う技術は、非特許文献1に開示されている。さらに、同技術を裏面照射構造に適用し、入射光を無駄なく光電変換せしめ、感度を飛躍的に向上させた改良は、非特許文献2に開示されている。この技術により16万5千画素に対して1600万枚/秒の超高感度ビデオカメラが実現した。
 図1に、非特許文献1に開示された高速撮像装置の部分平面を示し、図2に非特許文献2に開示された裏面照射構造の断面を示す。以下簡単に従来技術の動作原理を説明する。
 図1の符号1は信号電荷収集部を示し、信号電荷を収集・蓄積した後、CCDアナログメモリ部2の対応転送段に電荷を転送する。この状態で次のフレームの撮影時間まで待機する。これらの過程に要する時間をtiとする。CCDアナログメモリ部2はk個の転送段を有し、電荷収集部から信号電荷が転送されると矢印で示す転送方向に1段だけ転送する。次に時間tiが経過した後、次の信号電荷が電荷収集部からCCDアナログメモリ部2の対応転送段に転送され、同様に矢印の方向に1段だけ転送される。
 かかる動作をk回繰り返して、CCDアナログメモリ部2の全転送段は時間間隔tiを有するk個の信号電荷で埋め尽くされる。CCDアナログメモリ部2の最終転送段は接続ゲート8を介して垂直読み出しCCD(垂直アナログメモリ)3の対応転送段に接続されており、さらなる転送は垂直読み出しCCD3を含めた動作になる。
 1個のCCDアナログメモリので転送される信号電荷の垂直読み出しを担う垂直読み出しCCDの転送段数をjとすれば、最終的にk+j個の時間差tiを有する信号電荷がCCDアナログメモリ部2と垂直読み出しCCD3に保持されることになる。
 各画素9は信号電荷収集部1とそれに付随するCCDアナログメモリを備え、全画素で電荷収集と転送が同時に並列に行われる。従って、時間間隔tiで全画素9の映像信号電荷群(1フレーム信号電荷)が記録され、時間差tiをもつk+jフレーム分撮像装置内に保持される。
 続いて、垂直読み出しCCD3を動作せしめて、信号電荷を図の矢印の方向に転送し、水平転送CCD6に順次転送する。水平転送CCDのR個の転送段が信号電荷で満たされた後、水平CCDの水平読み出し転送(紙面の右から左への転送)を開始し、増幅・出力回路7から外部に読み出す。この動作を繰り返すことにより、時間間隔tiを有するk+jフレーム(コマ)の映像信号を外部に出力することができる。
 なお、図1の符号4は信号電荷収集部1に大量の電荷が流入した場合に、信号電荷収集部1から過剰な電荷を排出すためのドレインを示し、符号5はドレイン4への過剰電荷の排出を制御するゲートを示す。
 次に、図2の断面構造を用いて、裏面照射構造の光電変換及び信号電荷収集過程を説明する。図の断面は信号電荷収集部からCCDメモリに亘って第1図紙面の略水平方向の切断面を示す。符号11は信号電荷収集部、符号12は電荷収集を制御する収集制御電極(以後、「電荷収集電極」または「収集電極」と呼ぶ)、符号13はCCDアナログメモリ部、符号符号14はCCDの転送電極、符号15はCCDのn形転送チャネル層、符号16は素子分離層を各々示し、全て半導体(Si)の表面側に形成されている。
 Siの裏面には高濃度のp層20が形成され、バルクは低濃度のp層19と低濃度のn層の接合構造を成し、強い逆バイアス状態に保持されている。このため、バルク内の凡そ全キャリアが外部に排出された空乏状態になり、表面側から裏面に向かう電界が形成される。Siの裏面から入射した光(電磁波)21は、Si内で電子と正孔対に変換される。このとき、信号電子は前記電界によって表面側に引き寄せられ、正孔は裏面に移動して高濃度p層20に排出される。
 表面側に引き寄せられた電子が、CCDアナログメモリ部に侵入するのを防ぐためにpウェル障壁層17が形成されている。この構造によれば、光電変換された略全電子を信号電荷収集部11に収集する事ができる。この場合の電荷収集に要する時間は、数ナノ秒以下であり、表面側に引き寄せられた電子が、CCDアナログメモリ部に侵入するのを防ぐためにpウェル障壁層17が形成されている。この構造によれば、光電変換された略全電子を電荷収集部に収集する事ができる。この場合の電荷収集に要する時間は、数ナノ秒以下であり、1ナノ秒程度まで短縮することができる。
 図1を用いて説明した従来の高速撮像装置の速度(枚/秒、もしくはフレーム/秒)は、フレーム時間間隔tiの逆数になる。従って、フレーム時間間隔tiを如何に短縮できるかが、超高速撮像装置の技術的ポイントになる。
 非特許文献1また非特許文献2の技術による撮影の場合、フレーム時間間隔tiは以下の式(1)で表される。
Figure JPOXMLDOC01-appb-M000001
 ここに、tcは入射光により発生した信号電荷が電荷収集部に到達するまでの電荷収集時間、ttは電荷収集部からメモリへの転送に要する電荷転送時間、twは次のフレームの撮影までの待機時間である。
 最高撮影速度のとき、すなわち最小フレーム時間間隔tmはtwが0のときで、以下の式(1)’であらわされる。
Figure JPOXMLDOC01-appb-M000002
 前述した信号電荷収集、転送の過程において、入射光により発生した信号電荷が電荷収集部に到達する時間tcは数ナノ秒以下に短縮することができる。一方CCDへの転送時間或いはCCD転送段間の転送時間ttは10ナノ秒以上、一般的には数10ナノ秒を要する。
 即ち、フレーム時間間隔tiの最小値tmを律速しているのはCCDの電荷転送速度ttであり、さらなる撮影速度の向上にはttを短縮する、もしくはttの影響を排除することが必要である。
 このような超高速撮影手段はCCDを用いる手段に限らない。非特許文献3にはCMOS回路とCMOSメモリを用いた超高速撮像素子が開示されている。この素子を搭載した撮影装置により10万画素に対して1000万コマ/秒の撮影速度を達成している。
 この撮像素子の特徴は以下の通りである。
 (1)撮像素子チップの周辺部に各画素に対して100個のアナログメモリを作り込んでいる。
 (2)各画素列上に、該画素列上の画素数の1/4の数の信号読み出し線を配置し、4回の信号読み出し操作だけで画像1枚の全画素の画像信号を素子周辺部(受光面外)のメモリに転送し記録する。
 (3)各アナログメモリを通常のCMOSキャパシタと2層のポリシリコンで作るキャパシタの組み合わせで構成することにより、信号保存容量を大きくしている。
 (4)その他、画素ごとの電源回路の挿入や、ノイズキャンセルの工夫などが導入されている。
 この場合にも、電荷収集時間tcに比べて電荷転送時間ttがはるかに大きい。従って、電荷転送時間ttを短縮するか、もしくはその影響を排除することがさらなる高速化を達成する上でのカギとなる技術である。
 超高速撮影では極度に短いフレーム時間間隔のために入射光量の不足が非常に厳しい制約となる。非特許文献2では、非特許文献1の技術をそのまま使える特殊な裏面素子構造を発明し、開口率(画素面積に対する入射光の受光面積の比)を100%にした。
 一方、非特許文献4には、裏面照射を使わず、表面照射で、オンチップマイクロレンズと積層導波路の組み合わせで、画素に入射した光の多くを有効に光電変換する撮像素子技術が開示されている。この撮像素子は通常のデジタルカメラ用であるが、この技術は高速撮影用撮像素子にも適用できる。
江藤他、斜行直線CCD型画素周辺記録領域を持つ100万枚/秒の撮像素子、映像情報メディア学会誌、Vol.56、No.3、pp.483-486、2002. 江藤他、1,600万枚/秒の裏面照射撮像素子、映像情報メディア学会誌、Vol.65、No.3、pp.349-353、2011. 栃木他、バースト1Tpixel/sと連続780Mpixel/sの撮影速度を有するグローバルシャッタ高速イメージセンサ、映像情報メディア学会技術報告、Vol.36、No.18、pp.9-12、2012. 渡辺他、1.4μm表面照射型積層導波路構造イメージセンサ:smartFSItm、映像情報メディア学会技術報告、Vol.36、No.18、pp.37-40、2012.
 最小フレーム時間間隔tmの大部分を占める電荷転送時間ttの影響を排除し、理論的な最小コマ間時間間隔tm=tc、すなわち理論的最高撮影速度(1/tc)コマ/秒を実現する。これにより、科学技術分野等で要求される超高速現象の実体を正確に観察且つ解析できる1億コマ/秒以上の超高速撮像を可能にする新しい固体撮像技術を提供する。
 上記の課題を以下の手段で解決する。
(全画素面積で収集した光による信号電荷を複数の電荷収集手段で順次収集)
 各画素が、単一の光電変換手段と、m(≧3)個の電荷収集手段と、収集した画像信号(信号電荷もしくはそれを電圧等の他の信号に変換した信号)を転送する信号転送手段を備えるM行×N列の画素を備える撮像装置であって、前記光電変換手段がm個の電荷収集手段に個々に付随しているのではなく、m個の電荷収集手段に共通の画素面積の大部分(51%以上)に対応する広い面積に入射する入射光を光電変換する手段であり、生成した信号電荷をm個の電荷収集手段で順次収集する手段を備える。入射光をハーフミラーで2つの固体撮像手段に順次入射させることで高速撮影が可能である。この方法によれば、撮影枚数が2枚であれば全画素面積の50%の入射光を光電変換できるが、撮影枚数が3枚以上であれば全画素面積の約33%以下の入射光しか光電変換できな。高速撮影における入射光の損失は致命的であり、約33%以下の変換割合は許容し得ない。本発明では、各画素の単一の光電変換手段から各画素毎の複数の電荷収集手段に順次信号を収集することで、51%以上の変換効率を確保しつつ、撮影枚数が3枚以上の高速撮影を実現できる。
(光電変換手段から収集手段への電荷の移動経路に介在物がなく、経路が空乏化)
 さらに光電変換手段と電荷収集手段の間に、一旦信号電荷を保持する手段が介在せず、電荷収集手段の少なくとも1つで電荷を収集する状態では、光電変換手段と前記電荷収集状態にある電荷収集手段間が空乏化されており、この間を電位勾配のみによって電荷を収集する手段を備える。
(他の(m-1)個の電荷収集手段で順次信号電荷を収集中に信号電荷を転送)
 さらに、m個の電荷収集手段の中の1個から、画像信号を転送する時間内に、他の電荷収集手段の少なくとも2個で、順次、信号電荷を収集する手段を備える。
(電荷収集手段の制御は独立で信号転送手段はグループごとに共通)
 1個の電荷収集手段と1個の信号転送手段を備え、かつ必要に応じて他の信号制御回路を備える手段の組を「単位画素内回路」と呼ぶとき、m個の単位画素内回路をm/L(L≧2)個の単位画素内回路からなるL個のグループに分け、これらのグループの一つに属する電荷収集手段で順次収集した信号電荷を、他のグループに属する電荷収集手段で信号電荷を順次収集する間に一斉に転送する手段を備える。
 このためには、電荷収集手段は全て独立に制御されなければならない。一方、信号転送手段は各グループについて同時に動作して一斉にグループ内の信号を転送するので、グループ内で共通化されていなければならない。
(光電変換層と電荷収集手段の間の電位障壁を選択的にパンチスルーして電荷を収集)
 さらに、裏面照射素子構造とし、光の入射する裏面側の光電変換層と表面側の回路層の間に電位障壁層を備え、電荷を収集中の電荷収集手段においては電極に高い電圧(ホールを信号電荷とするときは低い電圧)をかけて障壁層をパンチスルーして電荷を収集する。
 m個の電荷収集手段が画素面の中央側に局在している場合は、前記電荷収集手段が局在している部分で前記障壁層の不純物濃度が低い。
(光電変換層と回路層の間の障壁が酸化膜)
 さらに、障壁層を絶縁物質としても良い。通常は絶縁層として酸化膜が使われる。電荷収集手段の位置で酸化膜層に孔を開け、この孔を通して信号電荷を収集する。
(電荷収集手段を画素の中心部分の円周上に配置し、単位画素内回路群を放射状に配置)
 電荷収集手段側を頂点としてm個の単位画素内回路を正多角形となるように敷き詰め、各画素内に内接させる。
 正6角形の場合は画素形状を正6角形として密に敷き詰めることができる。正8角形の場合は画素形状が正方形となり、各画素について1個の小さい正方形のスペースが残る。ここに各画素内の単位画素内回路に共通の制御回路を作り込む。
(導波路を用いて画素中心に集光)
 光電変換手段が、少なくとも導波路とフォトダイオードからなる。
(トランジスタにより複数の電荷収集手段で電荷を選択的に収集する。)
 トランジスタのオン・オフにより生成した信号電荷を順次収集する。
(電荷収集層がフローティング拡散層)
 電荷収集手段に絶縁層を介さずに電極をつけ、フローティング拡散層とする。
(全画素面積で収集した光により生成した信号電荷を電荷収集手段Aで収集:残りの(m-1)個の収集手段で順次電荷を収集中に電荷収集手段Aから信号電荷を転送)
 前述の通り、光電変換手段から電荷収集手段に至る信号電荷の収集時間をtc、収集手段からその外部に画像信号を転送する転送時間をtt、連続画像のフレーム時間間隔をti、最小フレーム時間間隔をtmとする。また理論的に実現可能な最小フレーム時間間隔をtuとする。
 本発明になる撮像装置によりm枚の画像を連続撮影するとき、フレーム時間間隔を最小化するには、収集時間tcごとにm個の電荷収集手段を順次切り替えて信号電荷を収集すれば良い。このときti=tc=tuとなり、理論的に実現可能な最小フレーム時間間隔による撮影ができる。ただしこの場合、連続撮影枚数がm枚に制限される。
 さらに、m個の電荷収集手段の一つ電荷収集手段Aで信号電荷を収集した後、残りの(m-1)個の電荷収集手段で順次信号電荷を収集する間に、画像信号を電荷収集手段Aの外部に転送し、電荷収集手段Aを空にする手段を備えれば、m+1個目の画像信号はまたAで収集することができる。これを繰り返せば、m×(k+1)コマの連続画像を撮影することができる。ここにkは各単位画素内回路の内部または外部にあって、単位画素内回路に属する画像信号メモリの数。また(+1)となっているのは、m個の電荷収集手段にも画像信号が保存されているからである。
 例えばm=8の場合、電荷収集手段Aで信号電荷を収集する時間はtcであり、残りの電荷収集手段で信号電荷を収集するために必要な時間は7tcである。この間に電荷収集手段Aから画像信号を転送できれば、電荷収集手段Aでは新たな信号電荷を収集できる。すなわち、7tc≧ttであれば、7tc後にはtt時間かけて電荷収集手段Aから信号電荷を転送し終えているので、Aで新たな信号電荷を収集できる。
 例えば収集時間tcが5ナノ秒で、転送時間ttが35ナノ秒以下であれば、電荷収集手段Aで1番目の信号を収集した後、35ナノ秒後には電荷収集手段Aは空になっているので、5ナノ秒かけてAで新たに9番目の信号を収集できる。例えばk=4であればtc=5ナノ秒ごとに8×(4+1)=40フレームを連続撮影できる。このときの撮影速度は1/tc=2億枚/秒である。
 一方7tc<ttの場合は、ti=tt+tc後に次のフレームの電荷が収集できる。例えばtt=45ナノ秒のとき、(tt+tc)=50ナノ秒ごとに8枚の画像が撮影できるので、フレーム時間間隔は(tt+tc)/8=6.25ナノ秒となり、撮影速度は1/((tt+tc)/8)=1億6千万枚/秒となる。これは非特許文献2に開示されている従来の世界最高撮影速度の10倍の撮影速度である。実際にはこのようにtc<<tであることが多い。
 一方、撮影時間間隔だけに注目するならば、通常の撮像素子を使い、上下左右に隣接する4個の画素グループを1個のグループ画素とみなし、4個の画素のうちの1つの画素Aで、1フレーム時間間隔だけ信号電荷を収集し、次の3フレーム時間間隔の間に生成した電荷をドレインから排出し、これを4つの画素で順次くり返せば、フレーム時間間隔tcでの撮影が可能である。引き続き3個の画素で信号電荷を取集中に画素Aから画素内、もしくは画素外のメモリに画像信号を転送することもできる。
 しかしこの場合、1フレーム撮影するときのグループ画素の受光面積は、もとの画素の1画素分となる。したがって開口率はもとの開口率の1/4以下となり、感度がもとの感度の1/4以下となる。超高速撮影では感度の低下は致命的である。
 これに対し略全画素面積で収集した光による信号電荷を画素中のm(m≧3)個の電荷収集手段で順次収集すれば、開口率を80%以上にでき、非常に高い感度で超高速撮影ができる。
(電荷収集手段の制御は独立で信号転送手段はグループごとに共通)
 例えば電荷収集手段・転送手段の組をX、Yの2グループに分け、グループXに属する電荷収集手段で順次収集した信号電荷(または画像信号)を、グループYに属する電荷収集手段で順次収集する間に一斉に転送する。この場合、電荷収集手段を制御する配線は電荷収集手段の数、すなわちm組必要である。一方、電荷の転送に必要な配線はグループの数、すなわち2種類だけで良い。したがって電荷の転送や一時保存等に必要な配線数はm組から2組に減る。これにより、信号電荷を順次収集中に信号を転送することによる撮影の高速化を享受しつつ、1画素列上の金属配線数を効果的に削減でき、画素サイズを縮小して画素数を増やすことができる。
 ただしこの場合、グループXの信号電荷を転送するために割り当て可能な時間は、電荷の収集時間tcのm/2倍、すなわちtc×(m/2)となる。一方、1つの電荷収集手段で電荷を収集後、他の(m-1)個で電荷を収集中に画像信号を転送する場合に信号電荷を転送するために割り当て可能な時間は、前述の通りtc×(m-1)となる。m≧3のとき、(m-1)>m/2であるから、1つの電荷収集手段で電荷を収集後、他の(m-1)個で電荷を収集中に画像信号を転送する場合に比べれば、画像信号転送の時間的制約は厳しくなる。
(光電変換手段から収集手段への電荷の移動経路に介在物がなく、経路が空乏化)
 信号電荷を選択的に順次異なる電荷収集手段で収集する手段としてはトランジスタのスイッチングによる方法もある。しかし信号の収集経路が空乏化していないので、ノイズが乗る。一方、収集経路が裏面表層部を除き完全空乏化していると、単調増加(ホールを信号とする場合は単調減少)する電位勾配だけで信号電荷がドリフトすることにより、高速で、かつ理論的には信号電荷を完全転送できる。
(光電変換層と電荷収集手段の間の電位障壁を選択的にパンチスルーして電荷を収集)
 このような裏面照射素子構造は非特許文献2で開示されている。この技術と各画素中に複数の電荷収集手段を導入することで、全画素面積で収集した光による信号電荷を複数の電荷収集手段で選択的に取集する技術が容易に実現できる。個別の電荷収集手段のゲート電極の電圧を上げるだけで、その上方の電位障壁が破れ、その電荷収集手段のみに電荷が集まる。
 この場合、電荷の収集経路は裏面の薄層を除き空乏化しているので、信号電荷の高速完全転送ができる。
 さらに、電荷収集手段群が各画素の中央側に局在している場合は、画素周辺の障壁層の濃度を高くし、中央部分で低くすれば以下の3つの効果が生じる。
 (1)光電変換層と障壁層の境界に、画素周辺から中心部に向かう強い電位勾配が生じ、これにより、画素の周辺部で生成した信号電荷が高速で画素の中心部に向かう。
 (2)画素中央部の障壁層を弱くし、より低い電圧でパンチスルーを起こすことができる。
 (3)画素周辺部に向かうにしたがって障壁層が強くなるので、その表面側に作り込むメモリなどへの信号電荷の迷入確率が下がる。
 この場合、画素の中央部とは、画素が円形または正多角形の場合は中心部ないし図心を意味する。また長方形の画素で、横または縦の中心線に沿って電荷収集手段を配置する場合は、中心線に沿う帯状の領域を意味する。
(光電変換層と回路層の間の障壁が酸化膜)
 絶縁層として酸化膜が使う場合はSOI(Silicon on Insulator)技術と呼ばれる汎用技術が使える。電荷収集手段の位置で酸化膜層に孔を開け、この孔を通して信号電荷を収集する。孔の表面側にある電荷収集ゲートの電極の電圧を順次上げることにより選択的に電荷を収集する。この場合も転送経路を空乏化することができる。
(電荷収集手段を画素の中心部分の円周上に配置し、回路群を放射状に配置する)
 m個の単位画素内回路を正多角形となるように敷き詰める。例えば頂角が60°であれば6個の電荷収集手段と転送手段およびその他の回路(メモリ)等の組を敷き詰めて正6角形の画素にすることができる。この場合は空きスペースなしに6角形画素を敷き詰めることができる。
 このとき6個の電荷収集手段は画素中心の周りの円周上に等間隔に配置される。画素周辺部から中心部までの距離は略等しいので、電荷収集手段ごとの電荷の収集時間は等しくなる。
 またすり鉢状の障壁層により効果的に信号電荷を収集することができる。
 一方、頂角が25度の場合は8個の単位画素内回路の集合は正8角形となる。正8角形を平面内に敷き詰めると、正8角形と同数の小さな正方形の面が残る。通常、8個の単位画素内回路に共通の回路が必要である。
 共通回路としては例えば電流源がある。また単位画素内回路をCCD構造のみで構成する場合には、共通のCMOS読み出し回路を正方形領域に作り込んでも良い(後述の実施例参照)。
 とくに千鳥配列にすると、実質の画素ピッチが正方配列の場合の1/√2となる。すなわち実質空間解像力が大きくなるという利点がある。
(導波路を用いて画素中心に光を集める)
 本発明を表面照射撮像素子に適用する場合に必要な技術である。オンチップマイクロレンズである程度画素の内側に集光した後、非常にアスペクト比の大きい(背の高い)導波路(光ではライトガイドと呼ぶ)を用いて非常に高い集光率(開口率80%程度)で入射光を画素中心のフォトダイオードに集める。
 その周囲に配置した電荷収集ゲートに順次高電圧をかけ、信号電荷を選択的に収集する。
 以上の技術により、超高速撮影に必要な高い感度と、理論的最高撮影速度である(1/tc)フレーム/秒、もしくは技術的最高撮影速度である(1/tt)フレーム/秒の撮影速度を実現できる。具体的な手段については、本発明の実施例で説明する。
非特許文献1に記載の従来技術による高速撮像素子の平面構造図である。 非特許文献2に記載の高速撮像素子の断面構造図で、光は裏面(下)から入射している。 本発明の実施例1を示す概略平面図である。 実施例1における画素の詳細平面構成図である。 実施例1における各画素の第1番目の電荷収集を示す平面図である。 実施例1における各画素の第2番目の電荷収集を示す平面図である。 実施例1における各画素の第8番目の電荷収集を示す平面図である。 実施例1における各画素の第16番目の電荷収集を示す平面図である。 実施例1における電極構成と制御電圧の給電線を示す部分平面図である。 実施例1における電荷収集と3相駆動CCDによる電荷転送原理を示す電位分布図である。 実施例1における電荷収集と3相駆動CCDの動作を制御する駆動タイミング図である。 実施例1における1画素を中心にした水平方向の断面構造図である。 実施例1における1画素の第13番目の電荷収集状態を示す断面構造図である。 実施例1における1画素の第14番目の電荷収集状態を示す断面構造図である。 実施例1における全画素の電荷を一斉にドレインに排出す状態を示す断面構造図である。 実施例1における電荷収集原理を示す半導体の深さ方向に沿った電位分布図である。 実施例1における電荷排出原理を示す半導体の深さ方向に沿った電位分布図である。 実施例1の動作シーケンス全体を示す駆動タイミング図である。 本発明の実施例2における電極構成と制御電圧の給電線を示す部分平面図である。 本発明の実施例3における1画素を中心にした水平方向の断面構造図である。 本発明を撮像システムに応用した場合のシステムブロック図である。 本発明の実施例4の全体構成である。 実施例4における画素の平面構造図である。 実施例4における画素の動作原理を示す図である。 本発明の実施例5の平面構造部分図である。 実施例5の動作原理のうち1画素内における信号電荷の動きを示す。 実施例5の画素の詳細な平面構造を示す。 実施例5の動作原理のうち転送過程を詳細に示している。 本発明の実施例6の平面構造部分図である。 実施例6の電極の平面配置図である。 実施例6のpウエル障壁層と単位画素内回路の分離層を生成するためのインプラント用マスクを示す図である。 実施例6の電荷収集電極の一つにハイ電圧VHを印加したときの電位分布の平面図である。 実施例6の電荷収集電極の一つに高電圧VHを印加したときの電位分布の断面図である。 本発明の実施例7の平面配置図である。 本発明の実施例8の断面図である。 実施例8の変形例の断面図である。 本発明の実施例9の断面図である。 本発明の実施例10の断面図である。
 以下、本発明の実施の形態を図に基づいて説明する。
(実施例1)
 図3に本発明の実施例1の平面構造を示す。符号31はM×N画素が配列されたイメージング領域を示し、小さな黒太枠で囲まれた32は1画素を示し、各画素32ごとにm×n個の電荷収集部を含む所謂CCDの電荷転送段が形成されてなる。
 符号33はイメージング部から矢印35の垂直方向に並列転送された信号電荷を1行ずつ受け取り、矢印36の水平方向に転送する水平転送CCDを示す。符号34は水平転送CCDから転送された信号電荷を受け取り、電圧信号として出力する出力回路を示す。符号37は信号電荷を垂直転送するための3相転送クロックパルスの入力端子を示し、符号38は信号電荷を水平転送するための水平転送クロックパルスの入力端子を示す。
 なお、本発明の実施例1においては、映像光はイメージング領域の紙面の裏面から入射し、多くの入射光は裏面近傍で光電変換され、表面の電荷収集手段に集められる所謂裏面照射型イメージセンサの構造を有している。
 図4に本発明の実施例1の平面構造を詳細に説明するための部分平面構造を示す。符号41と42は太い黒枠で囲まれた1画素32を確定する水平及び垂直方向のピッチを示し、符号43と44は1転送段及び一つの電荷収集手段を確定する水平及び垂直方向のピッチを示す。従って、この場合1画素32内に4×4=16個の転送段(電荷収集手段)が形成されてなる。ここで、説明の便宜上、1画素内の各転送段(電荷収集手段)に図示のように円で囲んだ番号(1)~(16)を付与する(図中ではこれの番号は円で囲んでいるが、本説明においてはこれを括弧付の数字として標記する)。
 符号45及び47は水平方向に延在する転送電極を示し、この場合、転送電極45には第1相の転送パルスφ1が、転送電極47には第3相の転送パルスφ3が供給される。符号46aは第2相の第1の転送パルスが供給される転送電極を示し、(13),(14),(15),(16)間において各々電気的に独立に形成されてなり、各々転送パルスφ21a、φ21b、φ21c、φ21dが供給される。
 符号46bは第2相の第2の転送パルスが供給される転送電極を示し、(9),(10),(11),(12)間において各々電気的に独立に形成されてなり、各々転送パルスφ22a、φ22b、φ22c、φ22dが供給される。符号46cは第2相の第3の転送パルスが供給される転送電極を示し、(5),(6),(7),(8)間において各々電気的に独立に形成されてなり、各々転送パルスφ23a、φ23b、φ23c、φ23dが供給される。符号46dは第2相の第4の転送パルスが供給される転送電極を示し、(1),(2),(3),(4)間において各々電気的に独立に形成されてなり、各々転送パルスφ24a、φ24b、φ24c、φ24dが供給される。
 符号48は前記φ1、φ21a、φ21b、φ21c、φ21d、φ22a、φ22b、φ22c、φ22d、φ23a、φ23b、φ23c、φ23d、φ24a、φ24b、φ24c、φ24d、φ3の18種類の転送(制御)パルスを供給するための給電端子群を示す。符号49は垂直転送CCD間の素子分離層を示し、必要に応じて過剰電荷排出ドレイン(リセットドレインを含む)が形成されてなる。
 次に図5~図8を用いて、本発明の第1の実施例の動作原理を説明する。図5の符号51は第1の画素32の(1)、符号52は第2の画素32の(1)、符号53は第3の画素の(1)、54は第4の画素の(1)、55は第7の画素の(1)を示し、第1~第9の全ての画素32の(1)に、高電圧の電荷収集制御パルスφ24aが、給電端子から印加された状態を示す。
 電荷収集制御パルスφ24aが印加される直前においては、前述のように裏面から入射した映像光によって生成された信号電荷は、各画素内のバルク領域に略均一に分布蓄積されている(詳細は後述)。電荷収集制御パルスφ24aが印加されることにより、各画素32のバルクに蓄積された信号電荷は、一斉に各画素32の(1)に示す電荷収集手段によって収集され、半導体表面領域の電荷転送段に蓄積される。信号電荷が前記各画素32のバルクに蓄積され、(1)での収集が終了するまでの時間をtc(≦ti)とする。
 図6の符号67~70は、電荷収集が終了し、信号電荷保存状態にある各画素32の転送段(1)を示す。次に高電圧の電荷収集制御パルスφ24bが、給電端子から全画素の(2)に印加される。この状態において、符号61は第1の画素の(2)、62は第2の画素の(2)、63は第3の画素の(2)、64は第4の画素の(2)、65は第7の画素の(2)を示す。φ24aの電荷収集パルスの印加が終了してから時間間隔tiを経て、φ24bの電荷収集パルスの印加が終了する。
 図7は、以上の動作を各画素32の(3)から(7)まで順次実行し、(3)に電荷収集パルスφ23dが給電端子から印加されている状態を示す。符号71は第1の画素32の(8)、符号72は第2の画素32の(8)、符号73は第3の画素32の(8)、符号74は第4の画素32の(8)、符号75は第7の画素32の(8)における、各々の信号電荷収集状態を示す。この時、符号76、77、78に示す各画素32の(1)~(7)は、既に信号電荷の収集を完了した信号電荷蓄積(貯蔵)状態にある。
 図8は、以上の動作をさらに各画素32の(9)から(15)まで順次実行し、(16)に電荷収集パルスφ21dが給電端子から印加されている状態を示す。符号81は第1の画素32の(16)、符号82は第2の画素32の(16)、符号83は第3の画素32の(16)、符号84は第4の画素32の(16)、符号85は第7の画素32の(16)における、各々の信号電荷収集状態を示す。この時、各画素32の(1)~(15)は、既に信号電荷の収集を完了した信号電荷保存状態にある。
 各画素32の(16)が信号電荷の収集を完了し、電荷保存(蓄積)状態に移行した時点で、各画素の全転送段(1)~(16)は信号電荷で満たされ、(1)→(2)→(3)→(4)→(5)→(6)→(7)→(8)→(9)→(10)→(11)→(12)→(13)→(14)→(15)→(16)の順番で各時間間隔tiをもって撮影された16フレームの画像信号電荷を蓄積した状態になる。
 この画像信号電荷を図3の符号35に示す垂直方向に順次並列転送して、1ラインずつ水平転送CCD33に転送し、続く水平転送により出力回路34から電圧信号として出力される。(16)の電荷収集過程から3相CCD駆動による垂直方向の電荷転送過程に至るチャネル電位分布と信号電荷の動きを、図8に1点鎖線で示すY1-Y2断面(86、87)に沿って図10に示す。符号121は電荷転送電極群を示し、符号122は半導体(Si)上に形成された絶縁膜(一般的にはSiO2)を示す。符号123は、転送電極を兼ねた前記(12)の電荷収集電極、符号124は、転送電極を兼ねた前記(8)の電荷収集電極、符号125は、転送電極を兼ねた前記(4)の電荷収集電極、符号126は、転送電極を兼ねた前記(16)の電荷収集電極を各々示す。
 各転送電極群には、図示の如く転送及び電荷収集を行う制御パルスφ1、φ21d、φ22d、φ23d、φ24d、φ3(この例では6種類)が給電される。この6種類の制御パルスの動作タイミングを図11に示す。符号141はφ1、144はφ21d、符号145はφ22d、φ23d、φ24dの共通パルス波形、符号146はφ3のパルス波形を示す。符号142a、142b、142cはφ21a、φ21b、φ21cを示し、φ21dに並列に隣接する電極に印加される電荷収集を兼ねた転送パルスを示す。即ち、各々(13),(14),(15)の電荷収集を制御する。
 図11の符号147はミドル電圧(VM)を示し、通常GND電圧に設定される。符号148は、パルス波形におけるロウ電圧(VL)を示し、通常所定の負電圧に設定される。符号149は、パルス波形におけるハイ電圧(VH)を示し、電荷収集を担うに十分高い正電圧に設定される。このハイ電圧が前記電荷収集制御パルスに相当する。符号150は、(13)の電荷収集の制御パルスで、これにより(13)に信号電荷が収集される。符号151は、(14)の電荷収集の制御パルスで、これにより(14)に信号電荷が収集される。符号152は、(15)の電荷収集の制御パルスで、これにより(15)に信号電荷が収集される。符号153は、(16)の電荷収集の制御パルスで、これにより(16)に信号電荷が収集される。
 ここで、符号150から符号153までの各電荷収集パルスは、図示のように時間間隔tiをもって各電極に印加される。この時間間隔が、撮影画像の撮影速度(フレームレート)を決定する。次に、図11の動作タイミングで図10のCCDを動作させた時のCCDのチャネル電位分布と信号電荷の動きを説明する。
 図10の符号127は、図11の時間t1におけるチャネル電位分布を示す。符号134はM(ミドル)-レベル電位を示し、直上の転送電極にミドル電圧VMを印加することで形成されるチャネル電位、符号135はL(ロウ)-レベル電位を示し、直上の転送電極にロウ電圧VLを印加することで形成されるチャネル電位、符号133はH(ハイ)-レベル電位を示し、直上の転送電極にハイ電圧VHを印加することで形成されるチャネル電位を各々示す。符号137は信号電荷収集動作を完了し、各転送段のM-レベル電位に保存されている信号電荷群を示し、符号138はH-レベル電位によって電荷収集過程にある(16)の信号電荷群を示す。
 符号128は、図11の時間t2におけるチャネル電位分布と信号電荷の保存状態を示す。φ21dのハイレベル電圧VHによる電荷収集過程が終了し、全信号電荷はM-レベル電位に保存される。この状態から3相駆動CCD動作による垂直転送(右から左方向:図3の符号35に示す方向)が開始される。以下簡単に3相駆動転送原理を説明する。
 図10の符号129は、図11の時間t3におけるチャネル電位分布を示し、φ3がロウ電圧VLからミドル電圧VMに変化し、対応するチャネル電位もM-レベルになり、信号電荷はφ3電極直下のチャネルに広がって蓄積される。符号130は、図11の時間t4におけるチャネル電位分布を示し、φ21d、φ22d、φ23d、φ24dがミドル電圧VMからロウ電圧VLに変化し、対応するチャネル電位がL-レベルに移行する。この変化により、信号電荷は全てφ3電極直下のチャネルに移動し蓄積される。従って、時間t2から時間t4の過程により、信号電荷は1転送電極だけ転送された事になる。
 図10の符号131は、図11の時間t5におけるチャネル電位分布をしめし、φ1がロウ電圧VLからミドル電圧VMに変化し、対応するチャネル電位もM-レベルになり、信号電荷はφ1電極直下のチャネルに広がって蓄積される。符号132は、図11の時間t6におけるチャネル電位分布を示し、φ3がミドル電圧VMからロウ電圧VLに変化し、対応するチャネル電位がL-レベルに移行する。この変化により、信号電荷は全てφ1電極直下のチャネルに移動し蓄積される。従って、時間t2から時間t6の過程により、信号電荷は2転送電極だけ転送された事になる。
 同様の動作をさらに1回行う事で、図11の時間t7において、全信号電荷は3転送電極、即ち1転送段転送される。
 以上の動作原理に従って、垂直方向に少なくとも1転送段転送して、図3の水平転送CCD33に並列転送し、前述のごとく、出力回路34から映像信号を出力する。この場合、第1の水平転送による第1の出力時系列は、(1)(2)(3)(4)(1)(2)(3)(4)・・・・(1)(2)(3)(4)の順番で画素数Mの4倍の信号(4フレーム分)から成る。第1の水平出力が全て終了すると、前記3相駆動により、1転送段の垂直方向転送が行われ、続いて第2の水平転送により、(5)(6)(7)(8)(5)(6)(7)(8)・・・・(5)(6)(7)(8)の順番で4フレーム×Mの第2の水平出力が得られる。この動作をさらに2回繰り返すことで、水平方向に配列された1画素行の(1)~(16)の全16フレームの映像信号が出力される。以上の一連の動作をN回繰り返すことにより、M列×N行の画素数の映像信号が16フレームにわたり出力される。
 図9に、本発明の実施例1における、さらに詳細な部分平面構造を示す。符号91、94は転送パルスφ1が印加される転送電極を示し、各垂直転送チャネルに共通に形成されてなる。符号93、96は転送パルスφ3が印加される転送電極を示し、各垂直転送チャネルに共通に形成されてなる。
 符号92は垂直転送チャネルごとに電気的に独立した電極群92a、92b、92c、92dからなり、各々φ21a、φ21b、φ21c、φ21dが印加される。符号95は垂直転送チャネルごとに電気的に独立した電極群95a、95b、95c、95dからなり、各々φ22a、φ22b、φ22c、φ22dが印加される。符号97は垂直転送チャネル間の分離部、98は垂直転送チャネル部を示す。
 符号99と105はφ1の給電線を示し、接続点(コンタクト)111を介して電極91、94に接続される。符号100、101、102、103は、各々φ21a、φ21b、φ21c、φ21dの給電線を示し、接続点112、113、114、115を介して電極92a、 92b、92c、92dに接続される。符号104、110は、各々φ3の給電線を示し、接続点116を介して電極93、96に接続される。符号106、107、108、109は、各々φ22a、φ22b、φ22c、φ22dの給電線を示し、接続点を介して電極95a、95b、95c、95dに接続される。
 以上の配線群により本発明の第1の実施例において必要とされる制御電圧(パルス)を全電極に供給することができる。
 前記分離部97は、垂直転送チャネル間を電気的に分離する役割を担っているが、過剰又は不要電荷排出ドレインとしての機能を付加することができる。
 図9の符号117、118は、分離部97に前記ドレイン機能を付加した場合の平面構造を示し、符号118は高濃度のn層からなるドレイン層で、通常は給電端子119から正電圧が印加される。符号117はp層からなり、通常はゼロ電圧(GNDレベル)に設定されてなり、ドレイン層118とnチャネル層を電気的に分離する役割を担う。ドレイン層118に所定の高電圧パルスを重畳印加することで、バルク内に存在する全電荷を一斉にドレイン層に吸収することができる。また、通常の正電圧印加時は、前記転送段から溢れ出した過剰電荷を排出するドレインの役割を担うこともできる。
 次に、図12、図13、図14、図16を用いて、本発明の実施例1における信号電荷収集原理を説明する。
 図12は、図9に示すY3-Y4切断線に沿った断面構造を示す。符号161、162、163、164は転送電極を兼ねた電荷収集電極を示し、各々制御パルスφ21a、φ21b、φ21c、φ21dが印加される。符号185はSi半導体表面に形成された絶縁層で、一般にはSiの酸化膜(SiO)からなる。符号165、166、167、168は対応する前記電極下の半導体表面付近に形成されたn形転送チャネル層を示す。符号169、170、171は前記チャネル間に形成された高濃度n形ドレインで、n型チャンネル層165、166、167、168との間にはp形素子分離層172、173、174が形成されている。
 符号175は1画素の領域を確定するための画素確定分離層で、画素間の信号蓄積層を電気的に分離する。符号176、178は低濃度n形層で、この中に電荷がCCDに移動するのを抑止するp形の障壁層177が形成されている。符号179低濃度のp形層で低濃度のn形層178とpn接合を成す。符号180はSi半導体の裏面に形成された高濃度のp形層(p+層)で、裏面からの接合の深さは極めて浅く形成されてなる。
 ここで表面側に形成されたp形素子分離層172、173、174、画素確定層175及び障壁層177を基準電位に設定して、裏面の高濃度p+層には負の逆バイアス電圧-VBBが印加される。この逆バイアス電圧-VBBによって、低濃度pn接合層178、179は略完全に空乏化され、アクセプタ及ドナーイオンの空間電荷により、紙面上部から下部に向かう電界が形成される。撮像開始或いは撮像中に、低濃度n層178の電子は後述のドレイン層169~171を含めた正電圧n層に排出される。
 符号184は映像入射光(電磁波)を示し、半導体の裏面から半導体内に入射する。即ち、裏面照射型のイメージセンサが構成されている。符号181は入射光のエネルギーが光電変換されて、正孔と電子の対が生成される領域を示す。周知のように、光子エネルギーの大きな短波長光は裏面近傍で光電変換され、光子エネルギーが小さい近赤外光を含む長波長光は、減衰しながらも裏面から深い位置まで進入し、光電変換される。
 生成された正孔と電子の対は、前記電界の力を得て分離され、正孔は裏面のP+層180に移動して吸収され、電子は表面に引き寄せられる。一方、障壁層177は電子が表面層に移動するのを抑制する。従って、電子は障壁層177と低濃度n形層178で形成されるpn接合領域に蓄積され、保持される。
 符号182は電子が信号電荷として蓄積される1画素の領域を示し、ここではバルク蓄積層と称する。符号183は隣接画素の信号電荷バルク蓄積層を示す。隣接画素のバルク蓄積層とは、深い接合のp形画素確定層175により、電気的に分離され、信号電荷が混じり合うことはない。符号186が1画素の水平寸法を示す。
 本発明の実施例1における、前記半導体(Si)の厚さは50μmである。これは、入射光が前記電荷収集と電荷転送を担うn層165~168に侵入して雑音電荷を発生させるのを防ぎ得る厚さに設定される。もし、入射光が可視光に限られるのであれば30μm程度で十分であるが、近赤外光を含む波長領域の映像光を入射させる場合には50μm以上の厚さが必要になる。
 なお、図面では割愛したが、裏面上に周知のRGB等のカラーフィルタを形成した場合には、半導体層の厚みをさらに薄くすることができる。また、裏面上には集光及び入射効率を高めるためには、マイクロレンズや反射防止膜を形成するのが良い。
 次に、本発明の実施例1における信号電荷収集原理を説明する。図13を参照すると、φ21aの電極161に高電圧が印加されるとn形チャネル層165の電位も略比例して上昇し、基板側に空乏層が伸長する。この空乏層は、p形障壁層177の対応領域を完全空乏化して、低濃度n形層に達する。即ち、バルク蓄積層182に空乏層が貫通するパンチスルー状態になる。
 図13の斜線領域193がパンチスルー状態を表している。この時の電位プロファイルを図16に示すが、詳細は後述する。191は電荷収集高電圧(H電圧)パルスが印加されたφ21a電極を示す。この状態にあっては、バルク蓄積層182の信号電荷194は、高電位が形成されたパンチスルー領域193に引き寄せられ、瞬時に電荷収集層192に収集される。φ21aの高電圧が解除され、0Vの蓄積電圧になるとバルク蓄積層182と収集領域192は電気的に切り離され、収集された信号電荷は、電荷収集層192に蓄積され、保持される。図14の205が電荷の収集を終えて、蓄積状態にあるφ21a直下の電荷収集層を示す。
 図14は、φ21bが電荷収集高電圧(ハイ電圧)に移行した時の断面の状態を示す。符号201は高電圧が印加されたφ21bの電極、符号202は電極の高電圧化に略比例して高電位になっている電荷収集層、203は前記パンチスルー状態にある領域、符号204はバルク蓄積層の信号電荷を示し、高電位にあるパンチスルー領域203に引き寄せられ、瞬時に電荷収集層202に収集される。φ21bの高電圧が解除され、0V(ミドル電圧)の蓄積電圧になるとバルク蓄積層182と収集領域202は電気的に切り離され、収集された信号電荷は、電荷収集層202に蓄積され、保持される。
 電荷収集高電圧(ハイ電圧)パルスをφ21c、φ21dの電極に順次印加することにより、水平方向に並ぶ全ての電荷収集層に信号電荷を収集し蓄積することができる。
 撮像動作シーケンスの全体では、図5から図8に示すように、各画素のφ24aの(1)の電極からφ21dの(16)の電極まで、電荷収集高電圧(ハイ電圧VH)パルスを時間間隔tiをもって順次印加することで、(1/ti)のフレームレートで16フレームの画像が得られることになるが、これだけでは不十分である。即ち、撮像開始となる初期状態を設定する必要がある。
 前記初期状態の設定方法は幾つか考えられるが、ここでは、前記バルク蓄積層182,183の全電荷を図12の符号169、170、171に示す高濃度n形ドレインに排出する方法を取り上げ、その動作原理を図15で説明する。
 図15の符号211は前記ドレインに電圧φDRを供給するための給電端子を示す。各ドレイン211に前述の電荷収集パルスと同様の高電圧が印加されると、n+ドレインから基板側に空乏層が伸長する。この空乏層は、p形障壁層177の対応領域を完全空乏化して、低濃度n形層に達する。即ち、バルク蓄積層182に空乏層が貫通するパンチスルー状態になる。符号212がパンチスルー状態の領域を示す。
 この状態になると前述の電荷収集過程と同様かそれ以上にバルク蓄積層の電荷はパンチスルー領域212に引き寄せられ、瞬時にドレイン169、170、171に吸収され、排出される。符号213はドレインに吸収される電子群を示す。即ち、バルク蓄積層182は空になり、初期状態にリセットされる。
 次に、図16と図17を用いて、前記パンチスルーによってバルク蓄積層の電荷が収集あるいは排出される原理を説明する。図16(a)は電荷収集手段の深さ方向に沿った1次元構造を示し、符号213は電荷収集電極、符号214は絶縁膜(酸化膜)、符号215はn形電荷収集蓄積層、符号216と218は低濃度n-層、符号217はp形障壁層、符号219は低濃度p-層、符号220は半導体裏面に形成された高濃度p+層を各々示す。
 図16(b)は(a)に沿って形成される電位プロファイルを示す。符号223は基準電位(0V)で、基準電位に対して紙面上方が負の電位、下側が正の電位を意味する。符号224は裏面のp+層220に印加された負電圧(-VBB)を示し、半導体裏面の電位を決定する。符号225aは電極電圧で、信号電荷蓄積状態では、VM電圧(0V)に設定される。この状態における電極213から裏面220に至る電位プロファイルは符号221の曲線で表される。
 低濃度n-層218と低濃度p-層219は完全空乏状態で、空間電荷が露出している状態にある。符号229は障壁層の最小電位を示し、非空乏化状態の基準電位に保持されている。符号226は低濃度n-層218の最大電位を示し、ここを中心にバルク蓄積層が形成される。符号232は光電変換されてバルク蓄積層に蓄積されている電子束を示す。
 次に、電荷収集電極213にハイ電圧VHの電荷収集パルスが印加された状態の電位プロファイルを説明する。符号222は電極213から裏面220に至る電位プロファイルを示し、符号225bは電極電位、符号227は電荷収集n層215の最大電位、符号230は障壁層217の電位を示す。図示のように、n層215から深さ方向に空乏層が伸長し、障壁層217を完全空乏化してn層218に結合する。
 このプロファイルが、前記パンチスルーを表す。障壁電位229が符号230に変化し、障壁電位が消滅するので、バルク蓄積層の電荷232は、収集層215に移動する。符号231の破線矢印は電荷(電子)の異動の過程を示し、233は収集された信号電荷を示す。
 1個の電荷収集電極に VHが印加された場合でも、隣接する電荷収集電極はVMが印加されているので、隣接電極から半導体裏面に至る電位プロファイルは、221と同一である。従って障壁層217の電位229が障壁になって、バルク蓄積層226と電荷収集蓄積層は電気的に分離された状態を保持する。228は、既に電荷収集動作を完了し、電荷収集層215に蓄積されている電荷束を示す。
 以上の動作原理により、バルク蓄積層の信号電荷は、選択的に電荷収集層に収集される。
 図17に、バルク蓄積層の全電荷をドレインに排出して、撮像初期状態にリセットする動作原理を示す。図17(a)はドレインの深さ方向に沿った1次元構造を示し、261はn形高濃度ドレイン層を示す。符号216、217、218、219、220は、図16(a)と同一である。
 図17(b)は、(a)に沿って形成される電位プロファイルを示す。ドレインは定常的には、電荷収集層215(図16参照)に転送された信号電荷の過剰成分だけを排出する役割を担い、符号262に示すVDR0でバイアスされている。この状態における電位プロファイルは符号264の曲線になる。符号268は障壁層217の障壁電位を示し、符号267はバルク蓄積層の電位を示す。
 図16の場合と同様に障壁層217は、非完全空乏状態にあり、符号268は基準電位に保たれる。従ってバルク蓄積層の電位267も図16の符号226と同一電位になる。符号271はバルク蓄積層に蓄積された電荷束を示す。
 次に、ドレインに初期化用高電圧パルスVDRHが印加されると、電位プロファイル264は、電位プロファイル265に変化する。符号263はドレインの電位、符号269は障壁層の電位を示す。図16における電荷収集過程と同様に、ドレイン層261から空乏層が伸長し、障壁層217を完全空乏状態にしてバルク蓄積層に達する前記パンチスルーが起こる。即ち障壁層の障壁電位が消滅し、電荷束271は破線矢印270の方向に瞬時に移動し、ドレインに排出され、バルク蓄積層は空状態(初期状態)になる。
 以上説明した原理に基づく、本発明の実施例1の動作シーケンスを図18に示す。符号241はドレインを制御するφDRパルス、符号242は第1番目の電荷収集電極を制御するφ24aパルス、以下符号243、244、245、246、247は、各々第12番目、第13番目、第14番目、第15番目、第16番目の電荷収集電極を制御するφ22d、φ21a、φ21b、φ21c、φ21dパルスを示す。符号255はφDRの高電圧パルスφDRHを示し、本撮像装置を初期状態にリセットする。
 符号248は第1番目の電荷収集を行うハイ電圧VHパルスで、φDRHの終了時間からtiの時間間隔を経て終了する。符号249、250、251、252、253は各々第12番目、第13番目、第14番目、第15番目、第16番目の電荷収集を行うハイ電圧VHパルスで、各々先行するハイ電圧VH終了時間からtiの時間間隔を経て終了する。
 この一連の電荷収集が終了した後、図10に示す3相駆動の垂直転送パルス254が印加され、前述のように(1/ti)フレーム/秒で高速撮像されたM×N画素の映像信号が出力される。
 本実施例では、pウエルによる適切な電位障壁と、収集電極に、電荷収集のためのハイ電圧VH、電荷保存のためのミドル電圧VM、電荷転送のためのロウ電圧VLの3値の電圧をかけることにより、各画素面積の大部分の面積に入射した入射光により生成した電荷を、指定した電荷収集手段の1つに導き、かつある電荷収集手段で電荷収集中であっても、他の収集手段で既に収集した電荷を保持することを可能にしている。
 本例は単純なCCD構造のみで構成されている。したがって非特許文献で既に実現されている特殊な裏面照射構造(図2および図12から図16に示されている)と既存のCCD技術を使い、ただちに超高速撮影が可能な撮像素子を実現することができる。
(実施例2)
 図19に本発明の実施例2の平面構造の部分詳細構造を示す。
 実施例1との違いは、電荷収集及び垂直方向の電荷転送を行うCCDの1転送段が、4個の電極を1単位とした所謂4相駆動CCDである点にある。
 符号280は転送パルスφ1が印加される転送電極を示し、各垂直転送チャネルに共通に形成されてなる。符号289は転送パルスφ4が印加される転送電極を示し、各垂直転送チャネルに共通に形成されてなる。
 符号281、282、283、284は垂直転送チャネルごとに電気的に独立した電極群を示し、各々電荷収集転送制御パルスφ21a、φ21b、φ21c、φ21dが印加される。符号285、286、287、288は垂直転送チャネルごとに電気的に独立した電極群で、各々φ31a、φ31b、φ31c、φ31dが印加される。以下290~293及び294~297も同様に垂直転送チャネルごとに電気的に独立した電極群で、各々電荷収集転送制御パルスφ22a~φ22d及びφ32a~φ32dが印加される。
 符号97は垂直転送チャネル間の分離部、符号98は垂直転送チャネル部を示す。符号298は前記電荷収集転送の制御パルスを供給するための給電端子群をしめす。各給電端子から各前記電極への給電線を直線、接続点(コンタクト)を黒点で示すが、図9と同様であり、説明は割愛する。
 実施例2では、1転送段が、例えばφ1、φ21a、φ31a、φ4の4電極で構成されるので、この場合φ21aとφ31aの2電極で電荷収集を行う事ができる。即ち1転送段の1/2の面積(実施例1では1/3の面積)で電荷を収集するので、収集効率が高くなる。さらに、信号電荷の転送に際しては、周知の4相駆動で行うので、常に2電極に電荷を蓄積できる。1転送段の長さを同一とした場合、第1の実施例の3相駆動が1電極蓄積(1/3の面積)であるのに対し、1.5倍の蓄積(転送)容量を確保することができる。
 但し、4列/画素の本実施例では、1転送段あたり10本の給電線(実施例2では6本の給電線)が必要になり、配線密度が高くなる。従って、画素サイズと必要フレーム(コマ)数及び最大信号電荷量によって、実施例1と使い分けるのが良い。
(実施例3)
 図20に本発明の実施例3を示す。実施例3における特徴は、半導体裏面に、特別なエネルギー変換層300が形成されているところにある。
 この変換層としては、例えば入射放射線によって生成された電子を、周知のアバランシェ過程により増幅して半導体裏面に注入する増幅膜、入射放射線のエネルギーを吸収して蛍光を発し、蛍光を半導体に入射させて光電変換により信号電荷を生成する蛍光膜、所定の波長域の放射線に感光して信号電子を生成する有機又は無機の光電変換層等々、多くの種類が考えられる。
 ここでは詳細な説明を省略するが、エネルギー変換膜の種類によってそのバイアス条件や、膜の厚さも当然異なり、それに伴って、電極を含めた裏面の構造も異なってくる。また、半導体層178、179の厚さも、場合によっては非常に薄くすることも可能である。
(その他の実施形態)
 本発明の実施形態は、以上の説明範囲に留まらない。
(裏面からの電荷排出)
例えばバルク蓄積層182に蓄積された電荷を排出しリセットする手段として、裏面の逆バイアス電圧-VBBを正電圧に変化させ、裏面から正孔を注入せしめて再結合により瞬時に消滅させる方法もある。
 さらに実施例1と類似であるが、前記垂直転送チャネル間とは別の場所に電荷排出ドレインを形成しても良い。
(CMOS型)
 さらに本発明における電荷収集手段はCCDに限定されるものではない。所謂CMOSイメージセンサタイプでも良い。要するに1画素内に複数の電荷収集手段を具備し、所定の時間間隔で電荷収集手段を独立に制御することにより、tiで時分割された1画素の信号電荷を独立に収集できれば良いので、CMOS型の場合にも、電荷収集層とバルク蓄積層の間に、電圧制御可能な障壁層を設ければ良い。
 また、読み出し方法はCCD型でなくCMOS型であっても良い。CMOS型では各画素が備える電荷収集手段の各々の後段に信号電荷を転送するCMOS回路からなる信号転送手段を備える。信号電荷は電圧に変換して電圧信号として転送しても良いし、ADコンバーターでデジタル信号にして転送しても良い。
(後段の信号保存手段と画素外での信号保存)
 さらに、各信号転送手段の後段に信号保存手段と、さらに信号保存手段から画素外への信号読み出し手段を設ける。
 CCD型の信号転送手段では信号転送手段と後段の信号保存手段が共通になる。
 またCMOS型の信号転送手段では、電荷排出段を別途設ける必要がない。
 さらに各信号転送手段に対してk(k≧1)個の信号保存手段を設けるのが望ましい。
 各画素の電荷収集手段の個数mが十分大きい場合、1個の電荷収集手段で電荷を収集した後、残りの(m-1)個の電荷収集手段で電荷を収集している間に、電荷収集時間tcの数倍の時間をかけて信号を転送することができる。この過程を各電荷収集手段で順次繰り返すことにより、各電荷収集手段で、1度電荷を収集し、次に電荷を収集する順番が回ってくる間に電荷収集手段からk個の信号を信号保存手段の一つに移し、電荷収集手段を空にすることができる。これを繰り返すことにより1画素あたりm×(k+1)個の連続画像信号を記録することができる。
 各画素のm個の電荷収集手段の配列や位置関係は実施例1の空間的に等距離に配置する場合に限らない。横一線に並べても良い。
(放射状の配置)
 また電荷収集手段を画素の中央に集め、転送手段や信号保存手段を放射状に配置しても良い。この場合、各画素の中央部の電荷収集手段上の障壁層の厚さを薄くし、周辺を厚くすることにより、各画素の中央に向かう電位分布を形成することができ、より短い時間で電荷を収集し、より高い撮影速度を達成することができる。
(画素の確定)
 また電荷収集手段が空間的に一様に分布している場合には、画素を確定する必要はない。各フレームにおいて、電荷を収集する電荷収集手段の中点を結ぶ線が自動的に各画素領域を規定する。したがって画素位置はフレームごとに少しずれる。このずれはポスト画像信号処理により、容易に補正することができる。
(積層型センサ)
 また、信号保存手段とその後段の信号読み出し手段は別チップ上に作り、積層型センサとしても良い。
(電子増倍機能の導入)
 超高速撮影では入射光不足を補うために、できるだけ感度をできるだけ高くする必要がある。感度の向上のためには、入射光の光電変換から、画像信号を撮像素子チップにいたる過程のどこかで、電子アバランシュ現象や衝突イオン化現象を利用した信号電荷の増倍機能を導入するのが好適である。そのための手段としては、加速電子の直入による衝突イオン化、MCP(Micro Channel Plate)、アバランシュフォトダイオード、EMCCD(CCD上の多段衝突イオン化増倍)等がある。いずれも良く知られた方法である。
(ホールを信号電荷とする場合)
 言うまでもなく電子でなく、ホールを信号電荷としても良い。その場合は用いる半導体の型(p型とn型)、電気的極性(+と―)を逆にする必要がある。
 以下、その他の実施形態の具体例について説明する。
(実施例4)
 実施例4は、前掲の「その他の実施の形態」で説明した項目のうち、「後段の信号保存手段と画素外での信号保存」の項で説明した以下の手段をそなえているところに特徴がある。
 1個の電荷収集手段で電荷を収集した後、残りの(m-1)個の電荷収集手段で電荷を収集している間に、電荷収集時間tcの数倍の時間をかけて信号を転送する。
 第4の実施例は8個の電荷収集手段を持つ(m=8)。
 説明を簡単にするために収集手段からの信号の転送時間ttが収集時間tcの4倍以下の場合を考える。実施例4では、8個の電荷収集手段のうちの4個(第1グループと呼ぶことにする)で時間間隔tcで順次電荷を収集した後、残りの4個(第2グループ)で電荷を収集する4tcの間に、第1グル―プに保存されている電荷を後段の信号保存手段に転送することができる。
 この時点で第1グループの電荷収集手段は空になっているので、引き続き時間間隔tcで電荷を収集することができる。
 これを繰り返すことにより、電荷収集手段の個数である8個の整数倍の連続画像を、時間間隔tcで撮影することができる。これは理論的な最高撮影速度である。
 以下、第4の実施例について具体的に説明する。図22に第4実施例の全体構成を示す。
 符号341はイメージング部を示し、所謂裏面照射型構造からなる。即ち、紙面裏側から入射した映像光を信号電荷に変換し、蓄積しさらには電流増幅する機能を有する。
 符号342はイメージング部341を構成する複数の画素の動作を制御するための信号を供給する制御部、符号343は各画素で得られたアナログ信号の増幅、低ノイズ化等を行うアナログ信号処理部、符号344はアナログ信号をデジタル信号に変換(AD変換)するAD変換部、符号345は変換されたデジタル信号を一時的に保持するラインメモリ部、符号346はラインメモリに保持されたデジタル信号を所定の規格(例えば周知のLVDS方式)で外部に伝送するための通信部、符号347はその出力ポートを各々示す。
 画素部を除いた構成は、所謂CMOSイメージセンサで一般的に用いられているものであり、説明するまでもない。
 図23、図24に本発明の実施例4における画素の平面構造と動作原理を示す。破線(画素境界)で囲まれた区画が、各々1画素を示す。符号351a、351b、351c、351dは4個の電気的に独立した第1の電荷収集電極群、符号353は第1の電荷転送蓄積電極、符号355は第3の電荷転送蓄積電極で、各々第1の電荷収集電極351a~351dの下に収集された信号電荷を受け取り蓄積する。
 符号352a、352b、352c、352dは4個の電気的に独立した第2の電荷収集電極群、符号354は第2の電荷転送蓄積電極、符号356は第4の電荷転送蓄積電極で、各々第2の電荷収集電極352a~352dの下に収集された信号電荷を受け取り蓄積する。
 符号357は信号電荷を信号電圧に変換してアナログ信号処理部に伝送するための出力変換部を模式的に示す。
 符号359、360は各々第3、第4の電荷転送電極355、356から出力変換部357に選択的に信号電荷を転送するための転送制御電極を示す。符号358は全電極に動作電圧を供給するための給電端子群を示す。
 符号362は信号電荷の転送領域を垂直方向に電気的に分離するための分離部を示す。以下、第4の実施例の動作原理を説明する。
 先ず、第1の電荷収集電極群351a、351b、351c、351dに時間間隔tiをもって電荷収集パルス(ハイ電圧VH)を順次印加する。図23に示す(1)~(4)(図中ではこれの番号は円で囲んでいるが、本説明においてはこれを括弧付の数字として標記する)は、各電極に収集され一時蓄積されている信号電荷を示す。この場合、(1)→(2)→(3)→(4)の順番で時間間隔tiをもって信号電荷が第1の電荷収集電極351a~351dに収集され一時蓄積される。
 続く動作を図24を併せて参照して説明する。図24の(a)において、第2の電荷収集電極群352a、352b、352c、352dに時間間隔tiをもって電荷収集パルス(ハイ電圧VH)が順次印加され、信号電荷が(5)→(6)→(7)→(8)の順番で時間間隔tiをもって第2の電荷収集電極352a~352dの各々に収集され一時蓄積される。
 この場合、(4)の電荷が収集されてから(5)の電荷が収集されるまでの時間間隔もtiである。第2の収集電極群が収集動作を行っている期間に、第1の電荷収集電極に一時蓄積されていた(1)~(4)の信号電荷群372は、第1の電荷転送蓄積電極353に並列転送される。この並列転送に割り当て得る期間は4tiとなる。
 続いて、図24の(b)に示すように、第1の電荷収集電極群351a、351b、351c、351dに時間間隔tiをもって電荷収集パルス(VH)を順次印加し、信号電荷群(9)~(12)を順次収集する期間に、第2の電荷収集電極群352a~352dに一時蓄積されていた(5)~(8)の信号電荷群371は、符号374に示すように、第2の電荷転送蓄積電極353に並列転送され、同時に第1の転送蓄積電極群351a~351dに蓄積されていた(1)~(4)の信号電荷群は第3の電荷転送蓄積電極355に並列転送され、符号376に示すように蓄積される。この場合にも、(8)が収集されてから(9)が収集されるまでの時間間隔はtiである。従って、これらの並列転送に割り当て得る期間は4tiである。
 同様にして、図24(c)で符号375に示すように(13)~(16)の信号電荷が収集され、符号378で示すように(9)~(12)信号電荷が並列転送される。さらに、図24(d)では、符号379で示すように(17)~(20)の信号電荷が収集され、符号380で示すように(13)~(16)の信号電荷が並列転送される。図24(e)では、符号381で示すように(21)~(24)の信号電荷が第2の電荷収集電極に収集され、1画素内の全電極の各々に信号電荷(1)~(24)が蓄積された状態になる。信号電荷(1)~(24)は、時間間隔tiの24フレーム(コマ)における1画素の映像信号になる。
 これらの信号電荷は、先ず(1)~(4)が出力変換部357に転送され、アナログ電圧信号に変換され、アナログ信号処理部に伝送される。続いて(5)~(8)が出力変換部357に転送され、アナログ電圧信号に変換される。
図24(f)は信号電荷群が順次出力変換部357に転送される様子を示す。なお、各出力変換部357が、2個の独立した回路から構成されている場合には、(1)~(4)と(5)~(8)を同時に出力変換部に転送することができる。かかる動作を経て最終的に外部に読み出される信号は、理想的には(1/ti)フレーム/秒の速度を有する24コマの映像として再生される。
 例えばti=5ナノ秒で4個の信号を同時に転送する時間ttが4tiよりも短ければ(理想的な場合)、2億コマ/秒の超高速撮影を実現できる。実際にはtt>4tiであることが多いので、実際のフレーム時間間隔はtt/4(>ti)となる。本発明の技術を用いない場合は、理論的最小フレーム時間間隔は(tt+ti)であり、撮影速度はその逆数であるから、その4倍以上の撮影速度を達成できる。
(実施例5)
 実施例5は前掲の「その他の実施の形態」で説明した項目のうち、「後段の信号保存手段と画素外での信号保存」の項で説明した特徴とともに、「放射状の配置」の項で説明したように、電荷収集手段を画素の中央に集め、その外側に転送手段や信号保存手段を放射状に配置するところに特徴がある。
 この場合、各画素の中央部の電荷収集手段上の障壁層の厚さを薄くし、周辺を厚くすることができ、各画素の中央に向かう電位分布を形成することができ、より短い時間で電荷を収集することができる。これも大きな特徴である。
 実施例5の全体構成は実施例4と同様で図22に示されている。したがってその説明は割愛する。
 図25に実施例5の平面構造部分図を示す。
 符号391の破線で囲まれた領域が1画素を示し、図示の如く正方配列を45°傾けた配列を成す。符号392は、電荷収集電極を示し、この場合8個の電気的に独立に制御される3角形の電極からなる。符号393は第1の電荷転送蓄積電極、符号394は第2の電荷転送蓄積電極、符号395は第3の電荷転送蓄積電極を各々示す。
 第3の電荷転送蓄積電極395は、1転送段を成す複数の電極で構成されるが、詳細は図27を用いて後述する。符号396は出力変換部、符号397は信号電荷を放射状の方向に転送するための各転送チャネルの分離部を示す。
 レイアウト的に見ると、各画素の電極群は各々八角形を成し、各画素は配列行間でその空間的位相が180°ずれて敷き詰められている。この敷き詰められた八角形の間にできた正方形の空き領域に出力変換部396が配置することで、平面利用効率が高いレイアウトが得られる。
 符号398は画素の1辺の長さpを示し、符号399は、この画素配列における水平方向画素ピッチ、符号400は垂直方向画素ピッチを各々示す。
 図からも明らかなように、水平方向及び垂直方向の画素ピッチは、画素の1辺の長さに対し1√2になる。即ち、映像の空間サンプリングピッチが1√2に狭くなり、水平垂直方向の撮像解像度が√2倍(1.44倍)高くなることを意味する。これは、実施例5の特徴の一つである。
 実施例5の最大の特徴は、電荷収集電極群392が画素の中心部に局在配置される所にある。この中心部局在化は、電荷収集に要する時間の短縮に大きな効果をもたらす構造の改良を可能にする。具体的には、図28を用いて、後述する。
 次に、第5の実施例の動作原理を図26、図27、図28を用いて説明する。
 図26(a)~(i)は1画素内における信号電荷の動きを示す。先ず(a)において電荷収集電極392の収集動作を時間tiの間隔をもって、(1)→(2)→(3)→(4)の順番で4回行い、(1)~(4)の信号電荷を収集電極下に収集する。この(1)~(4)の信号電荷を収集、転送、蓄積する電極群を第1電極群とする。
 次に(b)に示すように、(5)~(8)の信号電荷を(a)の場合と同様に(5)→(6)→(7)→(8)の順番で順次電荷収集電極392下に収集する。この収集期間(4ti)に、(1)~(4)の信号電荷を第1電極群の第1の電荷転送蓄積電極下に並列転送する。(4)の収集から(5)の収集の時間間隔もtiである。ここで、(5)~(8)の信号電荷を収集、転送、蓄積する電極群を第2電極群とする。
 (c)において、第1電極群の電荷収集動作が前記時間間隔で行われ、(9)→(10)→(11)→(12)の順番で信号電荷が収集電極下に収集される。同時に、(1)~(4)は第1電極群の第2の電荷転送蓄積電極394下に並列転送され、(5)~(8)は第2電極群の第1電荷蓄積電極393下に並列転送される。
 (d)において、第2電極群の電荷収集動作が前記時間間隔で行われ、(13)→(14)→(15)→(16)の順番で信号電荷が収集電極下に収集される。同時に、(1)~(4)は第1電極群の第3の電荷転送蓄積電極395下に並列転送され、(5)~(8)は第2電極群の第2の電荷蓄積電極394下に並列転送され、(9)~(12)は第1電極群の第1電荷転送蓄積電極393に並列転送される。
 (e)において、第1電極群の電荷収集動作が前記時間間隔で行われ、(17)→(18)→(19)→(20)の順番で信号電荷が収集電極下に収集される。同時に、(5)~(8)は第2電極群の第3の電荷転送蓄積電極395下に並列転送され、(9)~(12)は第1電極群の第2の電荷蓄積電極394下に並列転送され、(13)~(16)は第2電極群の第1電荷転送蓄積電極393に並列転送される。
 (f)において、第2電極群の電荷収集動作が前記時間間隔で行われ、(21)→(22)→(23)→(24)の順番で信号電荷が収集電極下に収集される。同時に、(13)~(16)は第2電極群の第2の電荷転送蓄積電極394下に並列転送され、(17)~(20)は第1電極群の第1の電荷蓄積電極393下に並列転送される。
 (g)において、第1電極群の電荷収集動作が前記時間間隔で行われ、(25)→(26)→(27)→(28)の順番で信号電荷が収集電極下に収集される。同時に、(21)~(24)は第2電極群の第1の電荷転送蓄積電極下に並列転送される。
 (h)において、第2電極群の電荷収集動作が前記時間間隔で行われ、(29)→(30)→(31)→(32)の順番で信号電荷が収集電極下に収集される。以上の動作により画素内の全電極下に信号電荷蓄積された状態になる。即ち、tiの時間差をもった32フレーム分の信号が各画素に蓄積されたことになる。
 第1電極群及び第2電極群の第3の電荷転送蓄積電極395は、図25に示すように八角形を最外周転送路を成す電極連鎖に沿って、右回り或いは左回りに信号電荷を転送することができ、所定の電極下から出力変換部396に(1)→(2)→(3)…→(8)の順番で順次転送することができる。出力変換部で電圧信号に変換してアナログ信号処理部に伝送する。
 第3の電荷転送蓄積電極395下の(1)~(8)の信号電荷が、この場合右回り転送を経て、全て出力変換部に転送され、第3の電荷転送蓄積電極395下が空になった状態で、(i)に示すように、第2の電荷転送蓄積電極394下の(9)~(16)の電荷が第3の電荷転送蓄積電極395下に一斉に並列転送され、第1の電荷転送蓄積電極393下の信号電荷(17)~(24)が第2の電荷転送蓄積電極(394)下に並列転送され、電荷収集電極392下の電荷(25)~(32)が第1の電荷転送蓄積電極393下に並列転送される。
 第3の転送蓄積電極下の電荷(9)~(16)は、(h)の場合と同様に前記右回り転送を経て、全て出力変換部に転送される。さらに、空になった第3の転送蓄積電極下に信号電荷(17)~(24)が並列転送される。
 以下同様の動作を繰り返すことによって。(1)~(32)の全ての信号電荷を電圧に変換してアナログ信号処理部に伝送することができる。
 図27に画素の詳細な平面構造を示す。
 符号401~404は第1電極群を構成する電荷収集電極を示し(図25、図26では392)、それぞれに収集制御パルスΦc1~Φc4が供給される。符号405~408は第2電極群を構成する電荷収集電極を示し、それぞれに収集制御パルスΦc5~Φc8が供給される。符号409aは第1電極群の第1の電荷転送蓄積電極で転送制御パルスΦ1が供給され、符号409bは第2電極群の第1の電荷転送蓄積電極で転送制御パルスΦ2が供給され、符号410aは第1電極群の第2の電荷転送蓄積電極で転送制御パルスΦ2が供給され、符号410bは第2電極群の第2の電荷転送蓄積電極で転送制御パルスΦ1が供給されてなる。
 符号420、411a、412aは第1電極群の第3の電荷転送蓄積電極、符号411b、412bは第2電極群の第3の電荷転送蓄積電極を示し、電荷転送蓄積電極411aと411bには転送制御パルスΦ31が、電荷転送蓄積電極420、412a、412bには転送制御パルスΦ32が各々供給されてなる。
 符号413は、転送電極420から転送される信号電荷を受け取り、電荷量を電圧に変換する周知のフローティング拡散層(以下FDと称す)を示す。
 符号414の破線で囲まれた領域は出力変換部を示し、この中にFDの電位を受ける電流増幅回路、アドレス回路、リセット回路等が配置される。グレーの太線415は放射状に広がる各電荷転送領域を確定するための素子分離領域を示す。
 符号416はドレインで画素の中心部に位置する。符号417は収集電極下の過剰電荷をドレインに流しだすためのオーバフロー制御ゲートを示す。符号418a、418bの破線はΦ31とΦ32の画素内配線例で419の黒小点が電極との接続点を示す。裏面から入射した光で光電変換された電荷は、Siバルク内に蓄積されるが、各画素のドレイン416に所定の高電圧を印加することによって、全てドレインに吸収される。
 図15と同様に、この吸収動作から撮像動作がスタートする。その後の電荷の転送の様子は図26の通りである。
 画素の最外周の第3の電荷転送蓄積電極は図示のように、Φ31とΦ32が供給される2個の電極420,411a,411b,412a,412bからなり、所謂2相駆動型CCDの構造を有し、Φ31とΦ32が2相転送パルスになることで、この場合、外周転送路を右回りに電荷転送する。
 一方、第1、第2の転送電極群も2相駆動CCDの構造を有し、基本的には2相転送パルスで信号電荷を転送する。この転送過程を、図28を用いて詳述する。
 図28(a)は、図27に示すR1-R2切断線に沿った切断面を示す。
 符号421は第1電極群の電荷収集電極(図27では402)、符号422は第2電極群の電荷収集電極(図27では406)を示し、各々Φc2、Φc4パルスで制御される。符号423aは第1電極群の第1の電荷転送電極、符号423bは第2電極群の第1の電荷転送蓄積電極を示し、各々Φ1、Φ2パルスで制御される。符号424aは第1電極群の第2の電荷転送電極、符号424bは第2電極群の第2の電荷転送蓄積電極を示し、各々Φ2、Φ1パルスで制御される。符号425aは第1電極群の第3の電荷転送電極、符号425bは第2電極群の第3の電荷転送蓄積電極を示し、共にΦ31パルスで制御される。
 符号426は絶縁膜、符号427はn形転送チャネル層を示し、符号428はn形転送チャネル層427のn形不純物濃度に対し、相対的に低濃度の不純物領域を示す。この不純物濃度分布により、各電極下に電位段差が形成され、周知の2相駆動CCD動作が可能になる。
 符号429は厚いpウェル層、430は薄いpウェル層、431は低濃度のn型基板を示す。実際にはの図12等に示すように、さらにその下に低濃度のp型基板があるが、図では省略している。
 図12で説明したように、pウェルとn基板は逆バイアスされており、光電変換された電子はpウェルに引き寄せられる。ここで収集電極421に高電圧(VH)の収集パルスが印加されると、収集電極421直下のn層から伸長する空乏層が薄いpウェルを貫通してn基板に到達するパンチスルー状態になり、電子は収集電極直下のn層に流れ込む。符号443はパンチスルー領域を示し、符号441は収集電極直下のn層に移動する電子の動きを示す。
 図28(b)は、各動作タイミングにおけるn形チャネル層の電位分布を示す。符号432は前述の収集電極421に1番目の収集パルスが印加された状態を示す。符号440は高電圧収集パルス(VH)Φc2が印加された状態のn層の電位HCを示し、符号439は収集された信号電荷(2)を示す。
 なお、図中の符号Lは各電極にロウ電圧(VL)、Hはハイ電圧(VH)が印加された状態におけるn形チャネル電位を各々示す。
 符号433は電荷収集電極421と対峙する電荷収集電極422に収集パルスΦc6が印加された状態におけるチャネル層の電位分布で、信号電荷(6)が電荷収集電極422の直下に収集された状態を示す。同時に電荷収集電極421がロウ電圧VLに、電極432aがハイ電圧VHにシフトすることにより、信号電荷(2)は電極423aの直下に転送される。
 符号434は電荷収集電極421に収集パルスΦc2が再度印加された状態におけるチャネル層の電位分布で、信号電荷(10)が収集電極421の直下に収集された状態を示す。同時に電荷収集電極422がロウ電圧VLに、電極432bがハイ電圧VHにシフトして電荷(6)が電荷転送電極423b直下に転送され、電荷転送蓄積電極423aがロー電圧VL、電荷転送電極424aがVHにシフトして電荷(2)が電荷転送電極424a直下に転送されることにより、信号電荷(2)は電荷転送電極423aの直下に転送される。
 符号435は電荷収集電極422に収集パルスΦc2が再度印加された状態におけるチャネル層の電位分布で、信号電荷(14)が電荷収集電極422の直下に収集された状態を示す。同時に電荷収集電極421がロウ電圧VLに、電荷転送蓄積電極423aがハイ電圧VHにシフトして電荷(10)が電荷転送蓄積電極423a直下に転送され、電荷転送蓄積電極423bがロウ電圧VL、424bがハイ電圧VHにシフトして電荷(6)が電極424b直下に転送される。同時に電荷転送蓄積電極424aがロウ電圧VL、電極425aがハイ電圧VHにシフトし、電荷(2)は425aの直下に転送される。
 符号436は収集電極421に収集パルスΦc1が再度印加された状態におけるチャネル層の電位分布で、信号電荷(18)が電荷収集電極421の直下に収集された状態を示す。同時に電圧収集電極422がロウ電圧VLに、電荷転送電極423bがハイ電圧VHにシフトして電荷(14)が電荷転送電極423b直下に転送され、電荷転送電極423aがロウ電圧VL、電荷転送電極424aがハイ電圧VHにシフトして電荷(10)が電荷転送電極424a直下に転送される。同時に電荷転送電極424bがロウ電極VL、電荷転送電極425bがハイ電圧VHにシフトし、電荷(6)は電荷転送電極425bの直下に転送される。
 符号437は収集電極422に収集パルスΦc2が再度印加された状態におけるチャネル層の電位分布で、信号電荷(18)が電荷収集電極422の直下に収集された状態を示す。同時に電荷収集電極421がロウ電圧VLに、電荷転送電極423aがハイ電圧VHにシフトして電荷(18)が電荷転送電極423a直下に転送され、電荷転送電極423bがロウ電圧VL、電荷転送電極424bがハイ電圧VHにシフトして電荷(14)が電荷転送電極424b直下に転送される。
 符号438は収集電極421に収集パルスΦc1が再度印加された状態におけるチャネル層の電位分布で、信号電荷(26)が収集電極421の直下に収集された状態を示す。同時に収集電極422がロウ電圧VLに、電荷転送電極423bがハイ電圧VHにシフトして電荷(22)が電荷転送電極423b直下に転送される。
 さらに、収集電極421を電荷収集電位からハイ電圧VHにシフトし、収集電極422に電荷収集パルスΦc6を印加することで、電荷を収集電極422直下に収集する。
 以上の動作により、断面に沿った全電極に信号電荷を蓄積することができる。収集電極442は、画素ドレイン領域にあたるが、同図では電荷転送の説明を趣旨とするので、単なる素子分離層に置き換えて示してある。
(実施例5)
 実施例5では、pウエルの中央に開けた電位障壁の低い部分に電荷を誘導し、その直下にあり、撮影中は収集電極部分より低い電位に保たれた素子分離層(またはドレイン)で振り分け、周囲の収集電極に送ることで、各画素面積の大部分の面積に入射した入射光により生成した電荷を、異なる電荷収集手段で順次収集することを可能にしている。
 次に実施例5の画素の具体的なサイズや不純物濃度について説明する。
 図30に電荷収集電極502a~hを含む電極の平面配置を示す。画素サイズは1辺10μmである。チップの厚さは50μmである。
 プロセス前の基板は、エピタキシャル成長で生成した低濃度のp型と低濃度のn型の2重のシリコン単結晶層からなる。表面から12μmは濃度が2×10**14(1/cm**3)のn型基板で、残りの38μmは濃度が2×10**13(1/cm**3)のp型基板である。ここに「**」は「乗」を示す。例えば10**14は10の14乗を表す。
 図31にpウエル(図28の符号429)を形成するためのインプラント501を示す。このpウエルは、裏面から表面に向かう信号電荷(本例では電子)の転送蓄積電極下への迷入を防ぐ電位障壁を形成するとともに、信号電荷を画素の中央に誘導する機能を果たす。図30で第1の電荷収集電極502aから502hは、図27では電荷収集電極401から408、図28では電荷収集電極421、422で示されている。図32、図33に電荷収集電極の一つ502aに高電圧VHを印加したときの電位分布の平面図と断面図を示す。両図には画素の中心部分にある電荷収集電極502aから見て最も遠い端点の裏面側の位置503で信号電荷が発生したときの収集電極に向かう電荷(この場合電子)の経路504も併記している。
 図31のpウエルを形成するためのインプラント501は、2回のホウ素イオンの打ち込みと、その後の熱拡散で形成される。1回目の層の打ち込み時のマスク形状505は一様で、中央に孔507が開いている。2回目の打ち込み時のマスク形状506は、周辺から中央に向かう放射状の細い3角形の集まりからなる。直線的に見えるが、イオン打ち込み後熱拡散により、周辺が濃く、中央が薄い滑らかなイオン濃度分布形状が形成され、これにより形成される中央に向かう滑らかな電位分布に導かれて信号電荷が短時間で収集電極下に収集される。符号508はチャンネルの分離層形成のためのマスク(図27の素子分離領域415)で、このマスクを用いて低エネルギーで高濃度のホウ素イオンが打ち込まれる。
 プロセスの最終段階で裏面に0.1μmの厚さの濃いp型のイオン(ホウ素イオン)を打ち込み、紫外超短パルスレーザーにより、瞬間的に活性化されている(図には記されていない。図12の高濃度のp形層180と同じ。)。
 裏面には―25V、電荷収集電極502a~502hのうち電荷収集電極502b~502hにはロウ電圧VL=+2V、電荷収集電極502aには高電圧VH=+10Vが印加されている。
 裏面と電荷収集電極の間の大きな逆バイアス電圧により、裏面の近傍と表面の近傍の層を除いて空乏化されており、生成した電荷は実質的に電位分布に直交して高速で表面側の電荷収集電極に収集される。
 図32、図33に示すシミュレーション結果によれば電荷収集電極502aから見て最遠点にあたる画素端の裏面で発生した電荷であっても0.95ns(<1ns)で収集電極に到達する。すなわち、理論的には10億枚/秒で連続撮影できる。
 実施例5でも、収集電極群を2群に分けた。これにより、比較的少ない金属配線数で本撮像素子を駆動できる。一方、1画素内の32個の電極を全て独立に駆動することも可能である。この場合はCMOS読み出し回路も全て独立に駆動しなければ意味がない。このためには1画素上に70本程度の金属線を配線する必要がある。微細CMOSプロセスを使えば、1μm当り3本程度の金属線を乗せることは容易である。したがって画素サイズが25μm程度あれば、8個の電荷収集電極と読み出し回路を独立に駆動することができる。この場合は、8個の電荷収集電極のうちの1個Aで電荷を収集したのち、残りの7個で順次信号を収集する間にAから信号を転送することができる。また8個の収集電極間の電荷の収集順序も任意に決めることができる。
 このとき、理論的最高速度を達成するためには、Aで電荷を収集した後、他の収集電極で順次電荷を収集する7tcの間に、収集電極Aからこれに付随するメモリ領域に電荷を移送する必要がある。通常はこれに要する転送時間ttの方が7tcよりも大きい。したがってフレームレートは(1/(tt+tc/7))フレーム/秒となる。
(実施例6)
 図29に本発明の実施例6の平面構造部分図を示す。実施例6の特徴は、1画素の電極群が八角形を成し、且つ画素が正方配列されてなる点にある。破線451で囲まれた領域が1画素を示す。本実施例においても、実施例5と同様に、八角形の電極群を敷き詰めた空き領域454に出力変換部を配置して、信号を電圧に変換する事ができるのは自明である。
 一方、本実施例においては前記出力変換部を使用しないで、最外周の転送路を連結させて所謂垂直転送CCDを構成することも容易にできる。この場合、各画素の垂直中心線456の右側を第1転送路、左側を第2転送路として分割し、信号電荷を垂直方向に蛇行しながら並列転送することができる。
図中の矢印452の連鎖が第1転送路、矢印453の連鎖が第2転送路の信号電荷の転送方向を各々示す。455は電荷収集電極によって収集される信号電荷を示し、(1)→(2)→…→(7)→(8)の順番で収集される。
 図の(1)→(4)を収集し転送蓄積する電極群を第1電極群、(5)→(8)を収集し転送する電極群を第2電極群として動作させることにより、(1),(3),(5),(7)を第1転送路、(2),(4),(6),(8)を第2転送路に分けて並列転送することができる。
 第6の実施例の構造を使えば、撮影後にCCDだけで受光面外に信号電荷を読み出すようにすることも容易である。これによりCCDの低ノイズ特性を享受できる。とくに冷却する場合はその効果が大きい。
(実施例7)
 実施例7の平面図を図34に示す。
 実施例7の特徴は、電荷収集電極510で収集された電荷が、画素内のCCDメモリではなく、CMOS読み出し回路511によって、受光面外のCMOSメモリ(図には書かれていない)に転送され保存されることである。
 符号512は電荷収集電極に隣接する出力ゲートである。符号513は出力ゲートに隣接するFD(フローティング拡散層)である。
 出力ゲートの電極の電圧は、収集電極の電圧のハイ電圧VHとロウ電圧VLの間の値VMであり一定である。
 電荷収集電極の電圧が高電圧VHで電荷を収集した後、VLに下がると、収集した電荷は出力ゲートの電位障壁を超えてFDに転送される。そこで電圧変換されて受光面外のCMOSメモリに転送される。
 CCDのかわりにCMOS回路を使って転送することの利点は、CMOS回路に様々な工夫を加えてノイズ低減やユーザーフレンドリな操作のための高機能化ができることである。このような高機能化の例は、非特許文献3等で開示されている。
 一方、CMOS回路で転送する場合には、転送時に乗るノイズを見かけ上減らすために、信号を増幅して送る必要があり、そのため1画像信号当たりのアナログメモリの必要面積が大きくなることである。本発明においては撮影中は、撮像素子のチップ上に作ったメモリ領域上に信号を保存するので、メモリに割り当てることのできるチップ上面積には限りがある。したがって、連続撮影枚数を減らすか、もしくは信号の増幅率を下げてSN比の低下を容認するか、いずれかの選択肢を採る必要がある。
 一方、実施例1から実施例4に示すように、CCDでそのままメモリする場合には原理的に電荷の完全転送ができる。とくに冷却する場合には実質的にノイズレスで信号を転送・保存できる。撮影後の読み出し時には、撮像素子外のバッファメモリに信号を送って保存するので、信号を十分高く増幅して転送することができる。
 一方、非特許文献3では、増幅された信号を小さい面積で記録するために、通常のCMOSキャパシタと、2層のポリシリコン電極間のキャパシタを組み合わせた大容量キャパシタをアナログ信号記録要素として使っている。
 メモリ領域を同一チップ上の受光面外に作る代わりに、裏面に第2のチップを接着し、第2のチップ上でメモリ領域を確保し、面に垂直配線により信号を転送しても良いことは言うまでもない。この場合には第2のチップ上には、メモリだけでなく、電源回路等を乗せることもできる。
(実施例8)
 図35に実施例8の断面図を示す。裏面には薄層の濃いp層が形成されている。電荷生成層は低濃度の厚いp層514からなる。この例の特徴は裏面側の電荷生成層514とその表側の電荷収集層515の間に酸化ケイ素からなる第1の絶縁層516があることである。いわゆるSOI(Silicon-on-Insulator)構造となっている。
 各画素517の絶縁層の中央に孔518が開いている。孔の周囲には電荷を中央の孔に誘導するためにキノコ型の薄いn層519が形成されている。
 電荷収集層515はn型の単結晶シリコンである。表面に近い薄い層にチャンネル形成のための濃いn層520が作られている。FD521にはオーミックコンタクトのための非常に濃いn層が形成されている。また濃いpドーピング522で画素が分離されている。
 符号523は酸化ケイ素からなる第2の絶縁層である。さらにその表面側にはポリシリコンからなる電極層524があり、その表面側に金属配線層525が形成されている。電荷収集層515、第2のシリコン層523、電極層524、金属配線層525を併せて、回路層526を構成している。
 回路層の平面構造は、図34に示す実施例7と同様である。電極に印加する電圧等も同じであるので詳細な説明は省略する。
 ただしこの場合は、中央(図では孔の直下)の電極がドレインではなく誘導電極527となっている。その両サイド(平面的には誘導電極の円周上)に電荷収集電極528があり、出力ゲート電極529がその外部に、さらにその外部にFD521、CMOS読み出し回路530が形成されている。
 また絶縁層の表面側にはタングステン層からなる遮光層531が形成されている。これにより入射光の一部が直接記録部に入射することをほぼ完全に防いでいる。
 図36に実施例8の変形が示されている。特徴は電荷収集電極の裏面側(図では上側)に、第1の絶縁層516を貫いて電荷を収集するための孔518が電荷収集電極の数だけ開けられていることである(中央のドレイン532の円周上に8個)。
(実施例9)
 図37に実施例9の画素の断面構造を示す。実施例9の特徴は、表面照射素子で、表面にマイクロレンズ533とライトガイド534からなる積層導波路構造535が形成されていることである。ライトガイドの直下(画素の中心部分)にはフォトダイオード536が形成されている。表面に入射した光はオンチップマイクロレンズで画素の中央部分に集められ、さらにライトガイドによって画素の中心部分に集められる。
 その下の層の平面構造は図34に示す第7の実施例のそれと類似している。
 すなわち第9の実施例では積層導波路構造によって、各画素面積の大部分の面積に入射した入射光を光電変換する。
 フォトダイオードを取り囲むように8個の電荷収集電極528があり、そのうちの一つに高電圧VHを加えると、画像信号はその収集ゲートに集まる。それ以後の電荷の転送と保存方法は第7の実施例と同じである。
 この電荷収集電極構造および制御方法と、積層導波路構造の組み合わせで、画素の全面積で生成した信号電荷を指定した収集手段の1つのみに順次導くことができる。
 8個の収集手段の一つAで信号電荷を収集した後、残りの7個の収集手段で順次信号電荷を収集する間に、Aから画像信号をAに属する画像信号記録手段に転送する手段については第7の実施例と同じである。
 第9の実施例の長所は、通常の表面照射撮像素子を作成するプロセスで作ることができることである。
 一方、第9の実施例は表面照射素子であるので、裏面に信号を送ることは難しい。したがって、複数のチップを積層し、光の入射する第1のチップに光電変換手段、電荷の取集手段、転送手段を乗せて、裏面に接着した第2のチップ上で信号電荷を蓄積することは難しい。また第2のチップ上に各画素単位あるいは画素ブロック単位で電流源回路や駆動回路を乗せ、第1のチップ上の電荷の収集手段や転送手段を駆動するようにすることも難しい。
 また光以外の、イオン等が入射する場合には使えない。
(実施例10:電子倍管の導入)
 図38に示す実施例10では、MCP(マイクロチャンネプレート)と加速電子の直入による方法である。図中、符号537はガラスの入射窓、符号538は光電面、符号539はMCP、符号540は撮像素子、符号541は撮像素子裏面の受光面、符号542はパッケージ、符号543はこれらを真空封止するためのチャンバー、符号544は真空チャンバーから配線を引く出すためのピン、符号545はピンを受ける(差し込む)ソケット、符号546は回路ボードである。
 入射光547は光電変換面で電子ビーム548に変換され、MCPに入射する。MCPで電子増倍されたのち、MCP出射面と撮像素子の受光面541の間にかけられた強い電場で加速された電子流549となり、受光面に入射し、衝突イオン化でさらに多数の2次電子を生成する。
 MCPは無くても良い。このときは加速電子の直入による衝突イオン化増倍で信号を増幅する。
(実施例11:トランジスタによる選択的電荷収集)
 本発明の実施例は以上の例に限らない。例えば、画素中心に集めた信号電荷を周囲に配置したトランジスタをオン・オフして電荷収集電極に集めても良い。電荷の移動経路が空乏化されていない等の不利な点もあるが、トランジスタのオン・オフに必要なゲート電圧差が3V以下で良い、等の利点がある。またCMOS回路では増幅等の他の機能を比較的容易に追加できる。
 そのためには図34において、電荷収集ゲート510とアウトプットゲート512の代わりに選択トランジスタを入れ、第2のアドレス配線8本を追加すれば良い(図は省略する)。
(実施例12:フローティング拡散層と電荷収集電極の共用)
 実施例12の概要を図34を用いて説明する。この実施例ではフローティング拡散層513で直接電荷を収集することを特徴とする。この実施例では電荷収集ゲート510とアウトプットゲート512がない。
 フローティング拡散層513の一つに他のフローティング拡散層の電圧より高い電圧を印加して電荷を収集する。このためにはリセットゲート551を開けた状態で、ドレイン552に高い電圧を印加する。このとき信号電荷はドレインに流れ続けている。しかしフローティング拡散層の電位は一定に保たれている。この状態でリセットゲートを閉じると、フローティング拡散層に信号電荷が流入し続ける。このため信号電荷が電子の場合はフローティング拡散層の電位が下がり続ける。したがって、リセットゲートを閉じた後、一定時間後の電位を検出することでCDS(相関2重サンプリング)が可能である。
 図34ではドレイン電圧は画素周辺のリング状の配線550から供給されている。したがって8個のドレインに同じ電圧が供給されている。実施例12では、8個のドレインに個別にドレイン電圧を供給する配線を備える(図には記されていない)。
 以上の技術により、超高速撮影に必要な高い感度と、理論的最高撮影速度である(1/tc)フレーム/秒、もしくは技術的最高撮影速度である(1/tt)フレーム/秒の撮影速度を備える撮像手段を提供する。具体的な手段については、本発明の実施例で説明する。
撮影システム
 図21に、本発明の撮像装置を応用した本発明の撮像システム(カメラ)の具体例を示す。符号311は本発明の固体撮像装置、符号312は種々の映像放射線(映像光を含む)を固体撮像装置311に照射するためのレンズ系(光の場合には集光光学系)、符号313は撮像装置の動作を制御する制御タイミング発生回路(TG)と撮像装置から出力される信号のアナログ処理(例えば所謂CDS回路によるノイズリダクション)及び信号のデジタル変換を行うA/D変換回路を搭載したAFE(アナログフロントエンド)集積回路、符号314はAFE回路からの動作制御信号を受けて撮像装置を直接駆動する駆動回路を各々示す。
 但し撮像装置内に駆動回路及びAFE相当回路が内蔵されている場合(デジタル出力CMOSイメージセンサ等)には、制御タイミング発生回路313、駆動回路314は不要である。
 符号319は撮像装置311の拡大図で、本発明の実施例1に従えば、その信号出力時系列は符号320の順番になる。破線で囲んだ符号322、323、324は第1行の第1番、第2番、第3番の画素を示し、第M番まで続く。符号325、326、327は第2行の画素を示す。全体では第N行まで続くので、M×N画素の信号が符号320の時系列で出力される。各画素は、(1)~(16)の16フレームの信号からなる。
 符号320の信号は時系列並べ替え機能を有する符号315(並べ替えストレージ)に入力され、321の配列に変換される。328、329は、第1コマ(フレーム)の画像信号、第2コマの画像信号、符号330は第16コマの画像信号を示し、各コマはM×N画素の信号からなる。このように、フレームにまとめられた信号をデジタル信号処理回路(DSP)316に入力する。
 DSP316は各フレーム信号のフィルタリングや補間処理、及びフレーム間の相関処理などを行い、信号をストレージあるいは表示するための回路ブロック317に転送する。この映像出力によって、(1/ti)フレーム/秒の超高速で撮影された映像をスロー再生して観察することができる。
 本発明を使用した高速固体撮像装置とその動作方法は、固体撮像装置を製造する産業に利用され、これを搭載した高速撮像カメラは、科学技術分野等に広く活用される。
 1 信号電荷収集部
 2 CCDアナログメモリー
 3 垂直読み出しVCCD
 4 ドレイン
 5 過剰電荷の排出を制御するゲート
 6 水平転送CCD
 7 出力回路
 8 コネクションエレメンツ
 11 電荷収集部
 12 電荷収集を制御する収集制御電極
 13 CCDアナログメモリー部
 14 CCDの転送電極
 15 CCDのn型転送チャネル
 16 素子分離層
 17 pウェル障壁層
 18 電荷収集層
 19 低濃度のp層
 20 裏面の高濃度のpの薄層
 21 Sの裏面から入射した光(電磁波)
 31 イメージング領域
 32 1画素
 33 水平転送装置
 34 出力回路
 41 1画素の水平寸法
 42 1画素の垂直寸法
 43 垂直転送チャネル幅
 44 1転送段の長さ
 45、47 3相駆動CCDの電極
 46 収集電極兼3相駆動CCDの電極
 51~55 1番目の電荷収集が行われている転送段
 61~65 2番目の電荷収集が行われている転送段
 66~70 信号電荷が蓄積されている転送段
 71~75 8番目の電荷収集が行われている転送段
 76~78 信号電荷が蓄積されている転送段
 81~85 16番目の電荷収集が行われている転送段
 91、93、94、96 チャネル間で一体化された垂直転送電極
 92、95 独立に給電される電荷収集電極兼転送電極
 97 チャネル間分離領域
 98 Nチャネル領域
 99~110 制御電圧の給電線
 121 転送電極
 122 絶縁膜
 123~126 電荷収集電極兼転送電極
 127 電荷収集状態のチャネル電位プロファイル
 128~132 電荷転送チャネル電位プロファイル
 150~153 高電圧電荷収集パルス
 161~164 電荷収集電極(兼転送電極)
 165~168 電荷収集兼電荷転送N形チャネル層
 169~171 高濃度n形ドレイン
 172~174 p形素子分離層
 175 画素確定p形層
 176、178 低濃度n形層
 177 p形障壁層
 179 低濃度p形層
 180 高濃度p形層
 182 信号電荷バルク蓄積領域
 185 絶縁膜
 191 1番目の電荷収集高電圧印加電極
 193 1番目のパンチスルー領域
 201 2番目の電荷収集高電圧印加電極
 203 2番目のパンチスルー領域
 211 ドレインへの給電端子
 212 ドレインからのパンチスルー領域
 221 電荷蓄積状態の深さ方向電位プロファイル
 222 電荷収集状態の深さ方向電位プロファイル
 223 基準電位
 224 半導体裏面逆バイアス電位
 225 電極電圧
 226 バルク蓄積層の電位
 227 電荷収集層の電位
 229、230 障壁電位
 241 ドレイン制御パルス
 242~247 電荷収集、電荷転送制御パルス
 262、263 ドレインの電位
 264 ドレインからの深さ方向電位プロファイル(定常時)
 265 ドレインからの深さ方向電位プロファイル(初期化時)
 268、269 障壁電位
 280、289 チャネル間で一体化された垂直転送電極
 281~288 独立に給電される電荷収集電極兼転送電極
 300 エネルギー変換層
 311 本発明の固体撮像装置
 312 レンズ系
 313 AFE(アナログフロントエンド)集積回路
 314 駆動回路
 315 時系列並べ替えストーレージ
 316 デジタル信号処理回路(DSP)
 317 表示ストーレージ
 318 システムブロック図
 320 出力時系列
 321 フレーム毎に並べ替えられた信号群
 322~324 第1行の第1番、第2番、第3番の画素
 325~327 第2行の画素
 328~329 第1コマ(フレーム)の画像信号、第2コマの画像信号
 330 第16コマの画像信号
 341 イメージング部
 342 制御部
 343 アナログ処理部
 344 AD変換部
 345 ラインメモリ
 346 通信部
 347 出力ポート
 351a~351d 第1の電荷収集電極群
 352a~352d 第2の電荷収集電極群
 353 第1の電荷転送蓄積電極
 354 第2の電荷転送蓄積電極
 355 第3の電荷転送蓄積電極
 356 第4の電荷転送蓄積電極
 357 出力変換部
 358 給電端子群
 359 第3の電荷転送電極
 360 第4の電荷転送電極
 361 画素境界
 362 分離部
 371~381 信号電荷群
 391 1画素(破線で囲まれた領域)
 392 電荷収集電極
 393 第1の電荷転送蓄積電極
 394 第2の電荷転送蓄積電極
 395 第3の電荷転送蓄積電極
 396 出力変換部
 397 各転送チャネルの分離部
 398 画素の1辺の長さ
 399 水平方向画素ピッチ
 400 垂直方向画素ピッチ
 401~404 第1電極群を構成する電荷収集電極
 405~408 第2電極群を構成する電荷収集電極
 409~412 電荷転送蓄積電極
 413 フローティング拡散層
 414 出力変換部
 415 素子分離領域
 416 ドレイン
 417 オーバフロー制御ゲート
 418 画素内配線
 419 電極との接続点
 420 転送電極
 421 第1電極群の電荷収集電極
 422 第2電極群の電荷収集電極
 423~425 電荷転送電極
 426 絶縁膜
 427 n形転送チャネル層
 428 2層駆動転送のためのn型低濃度の不純物領域
 429 厚いpウェル層
 430 薄いpウェル層
 431 低濃度のn形基板
 432~438 電位分布の時間変化の図
 439 収集された信号電荷
 440 高電圧収集パルス(VH)Φc2が印加された状態のn層の電位HC
 441 収集電極直下のn層に移動する電子の動き
 442 ドレイン
 443 パンチスルー領域
 451 1画素(破線で囲まれた領域)
 452 第1転送路の信号電荷の転送方向
 453 第2転送路の信号電荷の転送方向
 454 出力変換部
 455 電荷収集電極によって収集される信号電荷
 456 各画素の垂直中心線
 501 pウェルを形成するためのインプラント
 502 第1の電荷収集電極
 502a VHを印加した第1電荷収集電極
 503 電荷収集電極502aから見た最遠点
 504 電荷の最遠点から収集電極502aに向かう信号電荷のパス
 505 pウェル形成のための第1のマスク形状
 506 pウェル形成のための第2のマスク形状
 507 第1のマスクに開けられた孔
 508 分離層形成のためのマスク
 510 電荷収集電極
 511 CMOS読み出し回路
 512 出力ゲート電極
 513 フローティング拡散層
 514 電荷生成層
 515 電荷収集層
 516 第1の絶縁層
 517 画素
 518 絶縁層の孔
 519 キノコ状のn層
 520 チャネル形成のためのn層
 521 FD(フローティング拡散層)
 522 画素分離層
 523 第2の絶縁層
 524 電極層
 525 金属配線層
 526 回路層
 527 誘導電極
 528 電荷収集電極
 529 出力ゲート電極
 530 CMOS読み出し回路
 531 遮光層
 532 ドレイン
 533 マイクロレンズ
 534 ライトガイド
 535 積層導波路構造
 536 フォトダイオード
 537 ガラスの入射窓
 538 光電面
 539 MCP
 540 撮像素子
 541 撮像素子裏面の受光面
 542 パッケージ
 543 これらを真空封止するためのチャンバー
 544 真空チャンバーから配線を引く出すためのピン
 545 ピンを受ける(差し込む)ソケット
 546 回路ボード
 547 入射光
 548 電子ビーム
 549 加速された電子流
 550 リング状のドレイン電圧配線
 551 リセットゲート
 552 ドレイン

Claims (11)

  1.  M行×N列の画素を備え、
     各画素は、
     電磁波または荷電粒子である入射線の入射により信号を生成する単一の光電変換手段と、
     前記光電変換手段で生成した信号を収集するm(≧3)個の信号収集手段と、
     前記信号収集手段で収集された信号もしくは前記信号を他の信号に変換した信号である画像信号を前記信号収集手段から転送する信号転送手段を備え、
     前記光電変換手段は各画素面積に入射した入射線の51%以上を前記信号に変換し、
     各画素の前記光電変換手段で生成された信号を、各画素内の異なる信号収集手段で、を順次収集する手段を備えることを特徴とする固体撮像装置。
  2.  請求項1の固体撮像装置であって、該m個の信号収集手段の中の1個から、画像信号を転送する時間内に、他の信号収集手段の少なくとも2個で、順次、信号を収集する手段を備えることを特徴とする固体撮像装置。
  3.  請求項1または請求項2の固体撮像装置であって、1個の前記信号収集手段と1個の信号転送手段を備え、かつ必要に応じて他の信号処理回路を備える1組の手段を単位画素内回路と定義したときに、m個の単位画素内回路をm/L(L≧2)個の単位画素内回路からなるL個のグループに分け、これらのグループの一つに属する信号収集手段で順次収集した信号を、他のグループに属する信号収集手段で信号を順次収集する間に一斉に転送する手段を備えることを特徴とする固体撮像装置。
  4.  請求項1から請求項3のいずれ1項の固体撮像装置であって、前記光電変換によって生じる信号が電荷であることを特徴とする固体撮像装置。
  5.  請求項4の固体撮像装置であって、前記光電変換手段と前記信号収集手段の間に、一旦電荷を保持する手段が介在せず、前記信号収集手段の少なくとも1つで電荷を収集する状態では、前記光電変換手段と前記電荷収集状態にある前記信号収集手段の間を空乏化し、電位勾配のみによって電荷を収集する手段を備えることを特徴とする固体撮像装置。
  6.  請求項4または請求項5の固体撮像装置であって、光電変換手段からなる第1の層と、少なくとも信号収集手段および信号転送手段からなる第2の層である回路層、および第1の層と第2の層の間に介在し、第1の層で生成した電荷が第2の層に移行することを抑止する障壁層とからなり、前記障壁層の障壁機能を選択的に解除して、前記信号収集手段の1つのみに電荷を誘導する手段を備えることを特徴とする固体撮像装置。
  7.  請求項5または請求項6の固体撮像装置であって、前記信号収集手段が画素の中央部に局在しており、前記信号収集手段が局在している部分の前記障壁層の不純物濃度が相対的に低いことを特徴とする固体撮像装置。
  8.  請求項5から請求項7のいずれか1項に記載の固体撮像装置であって、前記障壁層が絶縁物質からなることを特徴とする固体撮像装置。
  9.  請求項1から請求項8までのいずれか1項に記載の固体撮像装置であって、各単位画素内回路が、信号収集手段側を頂点とする略三角形をなし、各画素が、m個の単位画素内回路を前記頂点を画素中心に置いて敷き詰められた正多角形を内包することを特徴とする固体撮像装置。
  10.  請求項9に記載の固体撮像装置であって、単位画素内回路の組が生成する正多角形の面と、これを内包する画素面との間のスペースに、前記単位画素内回路の組の制御に必要な共通の電子回路が組み込まれていることを特徴とする固体撮像装置。
  11.  請求項1から請求項5、請求項8から請求項10のいずれか1項に記載の撮像装置であって、前記光電変換手段が、少なくとも電磁波の導波路とフォトダイオードからなることを特徴とする撮像装置。
PCT/JP2013/055364 2012-02-29 2013-02-28 固体撮像装置 WO2013129559A1 (ja)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2014502360A JP6188679B2 (ja) 2012-02-29 2013-02-28 固体撮像装置
US14/467,605 US9503663B2 (en) 2012-02-29 2014-08-25 Solid-state imaging apparatus

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2012062673 2012-02-29
JP2012-062673 2012-02-29
JP2012-128368 2012-05-17
JP2012128368 2012-05-17

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/467,605 Continuation US9503663B2 (en) 2012-02-29 2014-08-25 Solid-state imaging apparatus

Publications (1)

Publication Number Publication Date
WO2013129559A1 true WO2013129559A1 (ja) 2013-09-06

Family

ID=49082750

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2013/055364 WO2013129559A1 (ja) 2012-02-29 2013-02-28 固体撮像装置

Country Status (3)

Country Link
US (1) US9503663B2 (ja)
JP (1) JP6188679B2 (ja)
WO (1) WO2013129559A1 (ja)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015056518A (ja) * 2013-09-12 2015-03-23 日本放送協会 裏面照射型cmos撮像素子、裏面照射型cmos撮像素子の製造方法
WO2015133350A1 (ja) * 2014-03-01 2015-09-11 江藤 剛治 撮像素子、撮影装置、及び計測装置
JP2015220255A (ja) * 2014-05-14 2015-12-07 日本放送協会 裏面照射型cmos型撮像素子、及び、裏面照射型cmos型撮像素子の製造方法
WO2016203667A1 (ja) * 2015-06-19 2016-12-22 江藤 剛治 高速撮像装置
JP2018032749A (ja) * 2016-08-24 2018-03-01 国立大学法人静岡大学 電荷蓄積素子
JP2018033008A (ja) * 2016-08-24 2018-03-01 国立大学法人静岡大学 光電変換素子及び固体撮像装置
WO2018038230A1 (ja) * 2016-08-24 2018-03-01 国立大学法人静岡大学 光電変換素子及び固体撮像装置
JP2018148097A (ja) * 2017-03-07 2018-09-20 パナソニックIpマネジメント株式会社 固体撮像素子
WO2019093208A1 (ja) * 2017-11-07 2019-05-16 江藤 剛治 高速イメージセンサ
JP2019161211A (ja) * 2017-11-07 2019-09-19 江藤 剛治 高速イメージセンサ
JP2020047780A (ja) * 2018-09-19 2020-03-26 キヤノン株式会社 光検出装置
JP2020522173A (ja) * 2017-06-01 2020-07-27 スペシャライズド イメージング リミテッド ピクセルセンサ素子、画像センサ、撮像装置および方法
JP6873336B1 (ja) * 2020-04-10 2021-05-19 株式会社オプトハブ 半導体イメージセンサ
JP7370323B2 (ja) 2017-10-27 2023-10-27 エッレファウンドリ エッセ.エッレ.エッレ. 電離放射線及び電離粒子の集積センサ

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014204364A (ja) * 2013-04-08 2014-10-27 ソニー株式会社 固体撮像素子および駆動方法、並びに電子機器
JP6217907B2 (ja) * 2013-08-02 2017-10-25 住友電気工業株式会社 光センサ装置およびスペクトルイメージング撮像システム
GB2549330A (en) * 2016-04-15 2017-10-18 Teledyne E2V (Uk) Ltd Image sensor
US20180301484A1 (en) * 2017-04-17 2018-10-18 Semiconductor Components Industries, Llc Image sensors with high dynamic range and autofocusing hexagonal pixels
KR102593949B1 (ko) * 2018-07-25 2023-10-27 삼성전자주식회사 이미지 센서
CN110855914B (zh) * 2019-11-01 2022-02-18 南京理工大学 一种通用型emccd单板相机系统
CN112366212B (zh) * 2020-10-28 2024-02-09 中国兵器工业集团第二一四研究所苏州研发中心 一种光收集效率高的emccd及其制作方法

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001053260A (ja) * 1999-08-05 2001-02-23 Sony Corp 固体撮像素子及びその製造方法
JP2004214363A (ja) * 2002-12-27 2004-07-29 Fuji Film Microdevices Co Ltd 固体撮像素子及びデジタルカメラ
JP2006269612A (ja) * 2005-03-23 2006-10-05 Shimadzu Corp 撮像素子およびそれを用いた撮像素子用装置
JP2007329722A (ja) * 2006-06-08 2007-12-20 Matsushita Electric Ind Co Ltd 固体撮像素子、デジタルカメラ
JP2008066937A (ja) * 2006-09-06 2008-03-21 Matsushita Electric Ind Co Ltd 撮像装置および撮像方法
JP2009252983A (ja) * 2008-04-04 2009-10-29 Canon Inc 撮像センサー、及び撮像センサーの製造方法
JP2009252973A (ja) * 2008-04-04 2009-10-29 Panasonic Corp 固体撮像素子およびその製造方法
JP2011035382A (ja) * 2009-07-10 2011-02-17 Shimadzu Corp 固体撮像素子
JP2011133464A (ja) * 2009-11-24 2011-07-07 Hamamatsu Photonics Kk 距離センサ及び距離画像センサ

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6661463B1 (en) * 1983-05-09 2003-12-09 David Michael Geshwind Methods and devices for time-varying selection and arrangement of data points with particular application to the creation of NTSC-compatible HDTV signals
US5355165A (en) * 1992-08-06 1994-10-11 Princeton Scientific Instruments, Inc. Very high frame rate CCD imager
TW326120B (en) * 1996-05-21 1998-02-01 Goji Eto Image sensing apparatus
JP4326050B2 (ja) * 1998-09-22 2009-09-02 ハイスペック合資会社 高速撮像素子
DE60104000T2 (de) * 2000-03-28 2005-06-30 Link Research Corp. Schnelle abbildungsvorrichtung und schnelle fotografische vorrichtung
US7091530B1 (en) * 2002-07-03 2006-08-15 Massachusetts Institute Of Technology High-speed, high-sensitivity charge-coupled device with independent pixel control of charge collection and storage
JP2007166581A (ja) * 2005-11-16 2007-06-28 Matsushita Electric Ind Co Ltd 高速撮影用固体撮像装置
US20080106625A1 (en) * 2006-11-07 2008-05-08 Border John N Multi image storage on sensor
JP4724151B2 (ja) * 2007-05-17 2011-07-13 岩手東芝エレクトロニクス株式会社 固体撮像装置
US8576319B2 (en) * 2009-12-25 2013-11-05 Japan Atomic Energy Agency Image sensor, semiconductor device and image sensing method

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001053260A (ja) * 1999-08-05 2001-02-23 Sony Corp 固体撮像素子及びその製造方法
JP2004214363A (ja) * 2002-12-27 2004-07-29 Fuji Film Microdevices Co Ltd 固体撮像素子及びデジタルカメラ
JP2006269612A (ja) * 2005-03-23 2006-10-05 Shimadzu Corp 撮像素子およびそれを用いた撮像素子用装置
JP2007329722A (ja) * 2006-06-08 2007-12-20 Matsushita Electric Ind Co Ltd 固体撮像素子、デジタルカメラ
JP2008066937A (ja) * 2006-09-06 2008-03-21 Matsushita Electric Ind Co Ltd 撮像装置および撮像方法
JP2009252983A (ja) * 2008-04-04 2009-10-29 Canon Inc 撮像センサー、及び撮像センサーの製造方法
JP2009252973A (ja) * 2008-04-04 2009-10-29 Panasonic Corp 固体撮像素子およびその製造方法
JP2011035382A (ja) * 2009-07-10 2011-02-17 Shimadzu Corp 固体撮像素子
JP2011133464A (ja) * 2009-11-24 2011-07-07 Hamamatsu Photonics Kk 距離センサ及び距離画像センサ

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015056518A (ja) * 2013-09-12 2015-03-23 日本放送協会 裏面照射型cmos撮像素子、裏面照射型cmos撮像素子の製造方法
WO2015133350A1 (ja) * 2014-03-01 2015-09-11 江藤 剛治 撮像素子、撮影装置、及び計測装置
JPWO2015133350A1 (ja) * 2014-03-01 2017-04-06 江藤 剛治 撮像素子、撮影装置、及び計測装置
JP2015220255A (ja) * 2014-05-14 2015-12-07 日本放送協会 裏面照射型cmos型撮像素子、及び、裏面照射型cmos型撮像素子の製造方法
WO2016203667A1 (ja) * 2015-06-19 2016-12-22 江藤 剛治 高速撮像装置
US10636822B2 (en) 2016-08-24 2020-04-28 National University Corporation Shizuoka University Photoelectric conversion element and solid-state image capturing device
JP2018033008A (ja) * 2016-08-24 2018-03-01 国立大学法人静岡大学 光電変換素子及び固体撮像装置
WO2018038230A1 (ja) * 2016-08-24 2018-03-01 国立大学法人静岡大学 光電変換素子及び固体撮像装置
JP2018032749A (ja) * 2016-08-24 2018-03-01 国立大学法人静岡大学 電荷蓄積素子
JP2018148097A (ja) * 2017-03-07 2018-09-20 パナソニックIpマネジメント株式会社 固体撮像素子
JP2020522173A (ja) * 2017-06-01 2020-07-27 スペシャライズド イメージング リミテッド ピクセルセンサ素子、画像センサ、撮像装置および方法
JP7370323B2 (ja) 2017-10-27 2023-10-27 エッレファウンドリ エッセ.エッレ.エッレ. 電離放射線及び電離粒子の集積センサ
WO2019093208A1 (ja) * 2017-11-07 2019-05-16 江藤 剛治 高速イメージセンサ
JP2019161211A (ja) * 2017-11-07 2019-09-19 江藤 剛治 高速イメージセンサ
US11222910B2 (en) 2017-11-07 2022-01-11 Takeharu Etoh High-speed image sensor
JP2020047780A (ja) * 2018-09-19 2020-03-26 キヤノン株式会社 光検出装置
JP7224823B2 (ja) 2018-09-19 2023-02-20 キヤノン株式会社 光検出装置
JP6873336B1 (ja) * 2020-04-10 2021-05-19 株式会社オプトハブ 半導体イメージセンサ

Also Published As

Publication number Publication date
US9503663B2 (en) 2016-11-22
JP6188679B2 (ja) 2017-08-30
US20140362268A1 (en) 2014-12-11
JPWO2013129559A1 (ja) 2015-07-30

Similar Documents

Publication Publication Date Title
JP6188679B2 (ja) 固体撮像装置
JP5023808B2 (ja) 固体撮像装置およびカメラ
KR101466845B1 (ko) 고체 촬상 장치 및 카메라
US6278142B1 (en) Semiconductor image intensifier
US8338248B2 (en) Semiconductor element and solid-state imaging device
US8569805B2 (en) Solid-state image sensor and method for producing the same
JP5671830B2 (ja) 固体撮像素子、固体撮像素子の製造方法、および電子機器
WO2014141900A1 (ja) 固体撮像素子およびその製造方法、並びに電子機器
CN104979365B (zh) 图像传感器像素及图像传感器
KR20110123667A (ko) 고체 촬상 장치, 고체 촬상 장치의 제조 방법 및 전자 기기
KR20230025932A (ko) 고체 촬상 소자 및 그 제조 방법, 및 전자 기기
EP1684351A1 (en) Solid-state image pickup device
JP2013038118A (ja) 固体撮像素子および電子機器
KR20170010034A (ko) 고체 촬상 장치와 그 제조 방법, 촬상 장치, 반도체 장치와 그 제조 방법, 및 반도체 기판
JP5167693B2 (ja) 固体撮像装置およびカメラ
US20160329367A1 (en) Back side illuminated cmos image sensor arrays
JP2008034836A (ja) 固体撮像素子
WO1997046004A1 (en) High-speed ccd imaging, image processing and camera systems
WO1997046004A9 (en) High-speed ccd imaging, image processing and camera systems
JP7029037B2 (ja) 固体撮像装置
KR101583904B1 (ko) 고체 촬상 장치, 고체 촬상 장치의 제조 방법 및 카메라 모듈
WO2019093208A1 (ja) 高速イメージセンサ
JP4753673B2 (ja) 裏面照射型撮像素子
US8669510B2 (en) Imaging device in particular of the CMOS time delay and integration type
US20160255295A1 (en) Image sensor with electron multiplication and grouped readout of pixels

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13755184

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2014502360

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13755184

Country of ref document: EP

Kind code of ref document: A1