JPS61117980A - Solid-state image pickup element of charge transfer-type - Google Patents

Abstract

PURPOSE:To reduce means and to remove completely an afterimage due to interlace by transferring two optical signals and one smear signal while the same numeric aperture as that of a conventional element is held and subtacting each optical signal and the smear signal. CONSTITUTION:Electric charges are generated in three electrodes and the remaining one is in the idle state among four electrodes. Then a vertical CCD is operated in a triple transfer mode where signals A, B and C are transferred to the idle electrode and subsequent two electrodes, whereby three-type of signals (two optical signals and one smear signal) can be transmitted with the same numeric aperture (35-45%) as that of the conventional element being held. Because signals in one group as shown by (m), (n), etc., include the same amount of smear signal components (black part in waveform in the figure) at waveforms of outputs 4-1-3 detected by electric charge detecting circuits 12-1-3 and at waveforms of outputs 14-1-2 of subtractors 13-1-2, a group of the outputs 4-1 and 4-2 and that of the outputs 4-3 and 4-2 are inputted to the subtractors 13-1 and 13-2, respectively, whereby real optical signals OA and OB free of a smear component can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a charge transfer type solid-state image sensor, and more specifically, a charge transfer type solid-state image sensor that includes a photoelectric conversion element on a semiconductor substrate and a charge transfer element that extracts optical information from each element. r
The present invention relates to an image sensor equipped with a D.

(Background of the Invention) It is necessary for an image sensor to have a resolution equal to or higher than that of an imaging electron tube conventionally used in television broadcasting. A matrix of 500 picture elements (photoelectric conversion elements) and 800 to 1000 pixels arranged in the horizontal direction and a corresponding scanning element are required.Therefore, the above-mentioned solid-state image sensor is a large-scale MOB that can be highly integrated. Made using circuit technology.Ch&rg@Coupj@a D
.

An imaging device using a MIA (hereinafter abbreviated as CCD) has been developed. This type of element is used, for example, in Proceedings of the 1980 National Conference of the Television Society of Japan, commemorating the 30th anniversary of the founding of the Society of Television Engineers, P.
33-P34).

FIG. 9 is a basic configuration diagram of a conventional CCD type image element.

119, 1 is a photoelectric conversion element made of, for example, a photodiode, 2 and 3 are a vertical CCD shift register and a horizontal CCD shift register for taking out the optical signals accumulated in the photoelectric conversion element group to output terminals, 5-1 .5-
2.6-1.6-2 are vertical CCD shift registers 2. This is a generator that generates clock pulses that drive the horizontal CCD shift register 3. Here, a two-phase clock pulse generator is shown, but four-phase l/1t,
, any three-phase clock format may be adopted. 1. In addition, 7-1 and 7-2 indicate transfer gates that send the charges accumulated in the photodiodes to the vertical shift register 2. The solid-state image sensor shown in FIG. 9 becomes a monochrome image sensor in its current form, but if a color filter is laminated on top, each party diode is provided with color information, so it becomes a color image sensor.

As is well known, solid-state imaging devices have advantages such as being small, lightweight, maintenance-free, and low power consumption, and are expected to be the next generation imaging device to replace the electron tube imaging device K.

However, the conventional CCD type solid-state image sensing device has the following two problems, which significantly deteriorate the image quality.

(1) Since interlaced scanning is performed in the vertical direction,
In the first field, odd number columns (1, 3, 5,...2
N-1) pixel signal is in the second field (i even column (
2, 4, 6, . . . 2N) pixel signal powers ζ are respectively read out. As a result, in the first field of the next frame, signals from columns that were not read in the previous field (that is, odd columns) are added to the new signals and read out (this phenomenon is generally called afterimage). . Solid-state image sensing devices have the advantage of not producing afterimages because of their fast switching speeds, but in reality, afterimages occur due to the interlaced readout method as described above.

A method to solve this problem is to provide two columns of vertical CCDs (i.e., twice as many vertical CCDs as the configuration shown in FIG.
It is conceivable to read out the signals of all columns of pixels, including odd and even numbers, in any field. However, since the number of vertical CCDs is doubled, the area allowed for the photodiode is significantly reduced (i.e., the aperture ratio is reduced from the conventional 35 to 40% to 15 to 20%), which makes the imaging element more important. When the light sensitivity, which is an item, decreases, it becomes 1.
One problem arises.

(1) On a monitor where the amount of incident light is loose, the amount of incident light is high)
This phenomenon (generally called smear), in which white vertical stripes appear above and below the subject image, greatly impairs the image quality, just like the afterimage mentioned above.

Therefore, solving the problems (1) and (1) above is a major challenge for future solid-state imaging devices.

[Purpose of the invention]

An object of the present invention is to eliminate these two problems, that is, afterimage and smear, and to provide a charge transfer type solid-state image pickup device with good image quality.

[Summary of the invention]

In order to achieve the above object, the charge transfer type solid-state image carrier of the present invention includes a group of photoelectric conversion elements and transfers optical signal charges accumulated in the element group in vertical and horizontal directions on the same semiconductor substrate. In the charge transfer type solid-state image element which integrates a vertical charge transfer element and a horizontal charge transfer element,
, three electrodes where the third signal charge is accumulated, and an empty state n
The first signal charge is applied to the empty electrode, and then the first signal charge advances and the second signal charge is applied to the empty electrode. The present invention is characterized in that a vertical charge transfer element is provided in which the second signal charge is transferred first and the third signal charge is transferred to the empty electrode repeatedly.

[Embodiments of the invention]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a basic configuration diagram of a charge transfer type solid-state imaging device showing one embodiment of the present invention.

In Fig. 1, l-1, 1-2 are photodiodes, 2'
2'-1, 2'-2, . . . are vertical CCDs 2, 2'-1, 2'-2, . . .
The electrodes 3'-1, 3'-2, and 3'-3 that constitute ' are the 31111 type signal charges Q sent by the vertical CCD 2'.
A t QB r (horizontal CCD that receives 18 and transfers each charge towards output elements 1, 4-2, 4-3;
5'-1, 5'-2, 5'-3, 5'-4 are vertical CCs
0 which is the vertical clock pulse generator driving D2'
Here, the horizontal clock pulse that drives the horizontal CCD 3' is omitted to simplify the drawing, but it may be two-phase as shown in FIG. 9, or three-phase or four-phase. . Four electrodes 2' forming vertical CCD 2'
-1, 2'-2, 2'-3, 2'-4, two series of optical signals QA# QB and smear q6 are placed under the three electrodes.
is accumulated and one remaining electrode is left empty.

Then, by using the triple transfer mode, which will be described later, the optical signal
'L@m QB is horizontal CCD 3'-1, 3-2
, 3'-3VC are sent.

FIG. 11!2 (sL) (1)) is a plan view and a cross-sectional structural view of a pixel portion constituting the CCD type solid-state image sensor of FIG. 1.

2'-1 (or 2'-3) and 2'-2 (or 2
'-4) is an electrode constituting the vertical CCD 2',
For example, 2'-1 is the first layer of polycrystalline silicon, 2'-2
is formed of the second layer of polycrystalline silicon. Each electrode 2'
An ion implantation layer 8 is formed under -1 to 2'-4, which was formed before the production of these electrodes, and the stiffness layer 8 is made of boron atoms, for example.

Here, transfer game) 7-1 (or 7-2) is C
The case where it is used also as CD electrode 2'-1 is shown, but of course CC
D[It may be formed of a material other than the pole 2'-1, for example, a third layer of polycrystalline silicon.

When the plane of Figure 2 (&) is cut along the X-X' plane, the second
The cross section is shown in Figure (bl).

Figure 2 (in bl, 9 is the CCD electrode 2' and the substrate 11
(for example, P type), and 10 is a C
It is a layer that makes the charge transfer channel of the CD a buried type (for example, U-shaped). In addition, if the channel is made into a surface shape,
this! 10 is unnecessary.

FIG. 3 is an explanatory diagram showing the triple transfer operation of the vertical CCD.

As shown in Figure 3 (→), the clock pulse generator 5'
-1 to 5'-4, each electrode 2-1゜2'-
2, 2'-3, 2'-tag, 2'-1, . . . are connected in sequence. Each army in NIK from t1 to tn! ! 2'
-1,2'-2,2'-3,2'-bottoms,... are shown in Figure 3Q), and the positions of each electrode correspond to Figure 3 (&). ing.

The triple transfer operation will be explained with reference to FIG. 3(b).

(1) At t-t, the signal charges QA + QB accumulated in the photodiodes l-1 and 1-2 due to the light incident during one field period rise to a high voltage (''H'' level). Clock pulse vl #Vl is applied, and transfer gate 7-1. is in a conductive state (potential is lowered).
It is sent to the vertical CCD 2' side through 7-2 and accumulated under the electrodes 2'-1 and 2'-3, which have the lowest potential.

(1) At t - t, the clock pulse applied to each electrode becomes a low voltage ("L" level), and the potential under each electrode becomes high, but the ions formed in a part under each electrode Due to the barrier VB formed by the injection layer 8, the charge QA + QB increases during time 1-1. Same as when - K. Denzawa
iF is multiplied.

@ 1-1. Now, since the vertical scanning period begins, charges QAIQB are transferred toward the horizontal CCD. In addition, 1
-+1. ．． 1. is the time within the vertical hoisting period.

First, the charge QB accumulated on the electrode 2'-3 is transferred to the lower part of the electrode 2'-4 by the intermediate voltage (1M'' level) K which is multiplied by the electrode 2'-4.

Nt-t, then the lettuce smear electric current N (L, F, ``M'' level) accumulated on the electrode 2'-2 is transferred to the bottom of the electrode 2'-3 to which the pick pulse V is applied.

Here, the smear charge is generated when the charge generated by light incidence leaks from the vertical CCDIIK.
Since C also leaks during the subsequent transfer period (from 1 to t!I), the amount of smear charge increases as time passes. On the other hand, this leakage charge enters any electrode, and the smear charge also mixes into the signal QA + QII, and the smear component in the signal also increases with the passage of time.

(y) At t-t, an "M" level clock pulse V is applied to the electrode 2'-2, and the charge QA accumulated on the electrode 2'-1 is transferred to the bottom of the electrode 2'-2. Ru.

Due to the above operation for the time t, ~1s, the charge QB e+1
．． This means that the IQ is transferred to the horizontal CCD side by one electrode.

In this way, charges exist in three of the four electrodes, and the electrode 1(II is left empty, and the empty charge 1it
(Signal ^, then the signal A goes first and the signal B2 goes first, and then the signal C goes to the empty electrode. This is a transfer operation. The operation will hereinafter be referred to as a triple transfer operation.

■1-1. Now, the transfer of charge QB starts again, and 1
Due to the intermediate voltage (M-level) applied to 012'-1, the charge QB accumulated on electrode 2'-4tC is transferred to electrode 2'
Transferred under -1. In this way, by sequentially repeating the triple transfer operation, the three signals Q! I + (
, *QA are transferred to the horizontal CCD side, and when they reach the final electrode of the vertical CCD, the signal QB+q@yQA is sent to each horizontal CCD 3'-1, 3'-2, 3'-3. Feeding charges to the horizontal CCD is performed during the horizontal scanning period, and once the horizontal scanning period begins, signals Q, q, and QA simultaneously flow within the horizontal CCD toward the outputs -1, 4-2, and 4-3. be transferred.

Note that the horizontal CCDs 3'-1, 3'-2, and 3'-3 are provided one for each signal, so the horizontal CCDs 3'-1, 3'-2, and 3'-3 are
The charge is transferred by the same single transfer operation as the conventional horizontal CCD (
It is sufficient to perform the transfer using the expression (expression of normal operation for triple transfer operation).

Further, the state at time ts-tn in FIG. 3(6) shows that the amount of smear charge increases as the transfer progresses over time (the thick black line is the amount of smear charge).

In this case, within the adjacent group (for example, (Q) +
(lyu), QA (Shun) group, (QII (11)
+1)? '@m (n-H) l QA (Zl+
1) The smear charges in (Goop) in ) are almost the same amount. That is, the amount of smear charge leaked to the signal QAll)'QE6x) is equal to q,(1)IC.

Next, subtraction of smear charges will be described. FIG. 4 is a diagram showing the basic configuration of the smear #* circuit.

12-1.12-2.12-3 is horizontal CCD 3'-1
, 3'-2, 3'-3, and a general VcMOS source follower circuit is used. 4-1.4-2.4-3 is the output of the charge detection circuit,
13-1.13-2 is! It is a 1 calculator. This subtractor 13
-1.13-2 may be provided outside the element by using a commercially available differential amplifier, or may be formed by integrating a <M□s differential amplifier inside the element.

FIG. 5 shows the optical signal charges Q, l(L, TQ
It is a figure which shows the output waveform of B.

FIG. 5 shows a charge detection circuit 12-1, 12-2゜12-
The waveform of the output 4-1.4-2.4-3 detected at 3 and the output 14-1.14- of the subtracter 13-1.13-2
2 waveforms are shown.

As mentioned above, the signals of the same group indicated by m, n, etc. do not contain the same amount of smear signal components (the black part of the waveform in Fig. 5 is the smear signal), so the output 4- 1 and 4
-2! From the input K to the 1 calculator 13-1, the output 1
4-1 is the true optical signal OA without smear components, and by inputting outputs 4-3 and 4-2 to subtraction W13-2, output 14-2 is almost free of smear components. A true optical signal OB is obtained respectively.

FIG. 6 is a configuration diagram of a CCD type image sensor showing another embodiment of the present invention.

In the embodiment of FIG. 1, three signal charges QA? 'Ll
#Q, is transferred by three horizontal CCDs, but the sixth
In the figure, 311 signal charges QAF+1. IQB
is transferred by two horizontal CCDs. In this case, two signals (e.g. QA e Q!I) are detected by one horizontal CCD.
is sent, and the other horizontal CCD outputs one signal (for example,
q, ) can be sent.

Therefore, the horizontal CCD 3'-
The horizontal CCD 3'-1, which transfers one signal, is caused to perform a single transfer operation. The double transfer operation causes a charge to be present on two of the three electrodes connected to the pin from the pulse generator;
When one electrode is in an empty state, a signal A is applied to this empty electrode.
This is a transfer operation in which the signal person moves forward and transfers the signal B to the empty electrode.

FIGS. 7 and 8 respectively show further embodiments 11 of the present invention.
FIG. 6 is a configuration diagram of a CCD type image sensor showing 6ifl.

In FIG. 7, three signals Qk1+1. IQB 1
The data is transferred using horizontal CCDs in columns.

In this case, the horizontal CCD is the same as the vertical CCD.
It is necessary to perform triple transfer operation.

When five or more rows of horizontal CCDs are arranged as in the embodiments of FIGS. 1 and 6, it is not necessary that the channel widths of each horizontal CCD be the same.

For example, since the amount of smear charge is generally one or more orders of magnitude smaller than the amount of optical signal charge, as shown in FIG. It may be smaller than the channel width W1.W of the CCD.Here, the following 3N cases can be considered as the magnitude relationship of the channel width of each horizontal CCD.

W, < W, m W, (make the channel widths of QA and QB equal) W, <Wl< W, (make the channel width of QA smaller than QB) W, < W, <W (make the channel width of QA Furthermore, in the embodiment shown in FIG. 1, the horizontal CCD 3'-2 for smear charge transfer is placed between the horizontal CCDs 3'-1 and 3'-30 that transfer the signal charge QA r QB. However, in addition to the above-mentioned (QB = q,, QA), the signal charge transfer order of the vertical CCD 2' is changed to (qs++ Q) by changing the driving method.
B# QA)l or (QA.

QB, q, ), etc.

Therefore, for example, there is no problem even if the horizontal CD3'-1 is used for smear charge transfer and the horizontal CCD3'-2 and 3'-3 are used for signal charge transfer.

In addition, in FIGS. 1, 6, 7, and 8, the vertical CCD has three signals QA+QB, ct.

However, ultimately the optical signal charge QA
Using only r QB, smear charge q, is vertical CCD
An absorption drain may be provided at the final stage for disposal. Of course, an absorption drain can be provided at the final stage of the horizontal CCD to discard the smear charges. In this case, the subtractor shown in Figure 4 becomes unnecessary, but! [Since I don't do calculations,
It becomes impossible to remove the smear charges contained in the optical signal.

Further, in the same way as the conventional element shown in FIG. 9, the optical signals of the photodiodes in one column (odd columns in the first field, even columns in the second field) may be read out. In this case, since one optical signal QA or QB and the smear signal qlI are transferred, the number of horizontal CCDs may be two or less. However, although the smear charge contained in the optical signal QA or QB can be removed, an afterimage due to interlacing occurs as in the conventional element.

In the present invention, by operating the vertical CCD in triple transfer mode, the aperture ratio (35~
40%), three types of signals (two optical signals and one smear signal) can be transferred. and,
These 211! By subtracting the smear signal from each type of optical signal, two types of true optical signals (that is, odd-numbered columns and even-numbered columns) without smear signals can be obtained.

As a result of experiments conducted by the present inventors, the interlace afterimage of about 50%, which was 11% with the conventional element, has completely disappeared (to 0%), and the smear has been reduced compared to the conventional element. It has become clear that approximately 40+IB can be reduced.

In the embodiment, an interline type CCD element, which is typical as a conventional solid-state image sensor, was explained. However, the present invention is not limited to this type, and may be applied to other types, such as 7-rem transformer 7-area. The CCD system of Ayoko et al. can perform 'iM eyebrows in exactly the same way.

(Effect of brightness) As explained above, according to the present invention, two optical signals and one smear signal are transferred while maintaining the same aperture ratio as the conventional d1 element, and each optical signal and smear signal are Since subtraction is performed, smear can be significantly reduced compared to the conventional method, and the image quality of the CCD solid-state image sensor can be improved by completely removing the afterimage caused by one interlace.

[Brief explanation of the drawing]

FIG. 1 is a basic configuration diagram of a charge transfer type solid-state imaging device showing an embodiment of the present invention, and FIG. 2 is a plan view and cross-sectional structural diagram of a pixel portion of the CCD type solid-state imaging device shown in FIG. 1. 3 is a diagram showing the triple transfer operation of the vertical CCD, FIG. 4 is a basic configuration diagram of the smear removal circuit of the present invention, FIG. 5 is a diagram showing the output waveform of the optical signal charge in FIG. 1, and FIG. , FIG. 7, and FIG. 8 are block diagrams of a CCD type solid-state image pickup device showing other embodiments of the present invention, and FIG. 9 is a basic block diagram of a conventional CCD type solid-state image pickup device. 1-1.1-2: Photodiode, 2': Vertical CCD,
2'-1, 2'-2, 2'-3, 2'-47 electrode, 3'
-1,3'-2,3'-3,3'-2,3"': Horizontal C
CD. Flea, 4-1.4-2.4-3: Output terminal, 7-1°7-
2: Transfer gate, 5-1.5-2.6-1° 6-2: Clock pulse generator. B, - Leg' Figure 2 Figure 3 Straw Ox Figure Straw 5 Figure 6 Figure 7 Figure 2' Summary Figure 8

Claims (4)

[Claims] (1) Charge transfer in which a group of photoelectric conversion elements, a vertical charge transfer element and a horizontal charge transfer element that transfer optical signal charges accumulated in the element group in the vertical and horizontal directions are integrated on the same semiconductor substrate. In the type solid-state image sensor, the first, second,
It has multiple sets of three electrodes in which third signal charges are accumulated and one electrode in an empty state, and the first signal charge is applied to the empty electrode, and then the first signal charge is Vertical charge that repeats the operation of transferring the second signal charge to the electrode where the signal charge went first and became empty, and then the third signal charge to the electrode where the second signal charge went first and became empty. A charge transfer type solid-state image sensor characterized by being provided with a transfer element. (2) Charge transfer in which a group of photoelectric conversion elements, a vertical charge transfer element and a horizontal charge transfer element that transfer optical signal charges accumulated in the element group in the vertical and horizontal directions are integrated on the same semiconductor substrate. In the type solid-state image sensor, the first, second,
It has multiple sets of three electrodes in which third signal charges are accumulated and one electrode in an empty state, and the first signal charge is applied to the empty electrode, and then the first signal charge is Vertical charge that repeats the operation of transferring the second signal charge to the electrode where the signal charge went first and became empty, and then the third signal charge to the electrode where the second signal charge went first and became empty. A transfer element, and a subtracter disposed after the output of the horizontal charge transfer element to subtract third signal charges leaking during signal transfer from the transferred first and second optical signal charges, respectively. A charge transfer solid-state image sensor featuring: (3) The horizontal charge transfer element transfers the first, second, and third signal charges using three horizontal CCD columns, or transfers the first, second, and third signal charges using two horizontal CCD columns.
3. The charge transfer type solid-state image pickup device according to claim 2, wherein the charge transfer type solid-state image pickup device is characterized in that transfer is performed by a CD or by a single row of horizontal CCDs. (4) The charge transfer type solid-state image pickup device according to claim 2 or 3, wherein when two or three rows of horizontal CCDs are provided, a size relationship is provided for channel width.