JPS631680B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for driving a charge-coupled device (hereinafter referred to as CCD), and particularly to a charge input method.

Since CCD was invented in 1970, it has developed rapidly along with subsequent advances in semiconductor integrated circuit technology.

A feature of CCDs is that they have a function that other devices do not have: they convert external signals into charges proportional to the amount of signals and process them inside the semiconductor in a thermally non-equilibrium state.

Furthermore, by utilizing the photoelectric conversion function that is a feature of silicon devices, and by using the signal delay function that is unique to CCDs, they are being actively applied as imaging devices, analog signal delay lines, and various analog signal processing devices.

Among the applications of these CCDs, the application as an analog signal delay line is basic, and it is used for delaying one horizontal scanning period of a television signal, a comb filter for separating color and luminance signals, a ghost canceller, and a VTR. Applications such as signal processing are expected. If we consider the case where CCDs are applied in the field of television, the driving frequency of CCDs is 10.7MHz or 14.4MHz, which is a very high frequency for a silicon MOS device. Naturally, at such a high frequency, there are problems with the driving method, distortion rate characteristics, etc. of the input section of the CCD that samples the signal. The present invention was made based on this technical background.

FIG. 1 shows a cross-sectional view of the input section of a conventional CCD. In this figure, the case of an N channel using a P-type semiconductor substrate will be explained. In the figure, 1 and 2 are first and second input gate electrodes forming an input section, and 7 is a semiconductor region having a conductivity type opposite to that of the semiconductor substrate 8, forming an input diode.

3a, 3b,..., 6a, 6b are transfer electrodes of the CCD, and 3a and 3b, 4a and 4d... form a pair of transfer electrodes, and the odd-numbered transfer electrode group and the even-numbered transfer electrode group The transfer electrode groups are connected to terminals 15 and 16 by wiring, respectively.

In addition, in a normal high-speed delay line, the input diode 7 and the transfer electrode 3a adjacent to the second input gate electrode,
3b, but they do not have to be connected in common. 13 and 14 are terminals for applying voltage to the first and second input gate electrodes. Reference numeral 17 is an insulating film, and 9 is a semiconductor region having a conductivity type opposite to that of the substrate 8, forming a buried channel. 10-1
Reference numeral 2 denotes a barrier region for realizing two-phase drive, which is usually formed of a semiconductor region of the same conductivity type as the substrate 8.

FIG. 2 shows part of a drive waveform for driving the device shown in FIG. φ ₁ , φ ₂ in Figure 2
are applied to terminals 15 and 16, respectively, in FIG. φ ₁ and φ ₂ are two-phase drive pulses that are complementary to each other and change from 0\volt to V volt.

Next, the operation of the conventional device shown in FIG. 1 will be explained.

The structure of the input section of this device is suitable for input using the potential balance method. Potential balance method input is an input method already well known to those skilled in the art. This device further uses Phase, which is a modification of this potential balance method input.
An input method called referred input (hereinafter referred to as PR input) is adopted. This PR input is characterized in that the diode 7 and one transfer electrode 3a, 3b are commonly connected, and does not particularly require a pulse for driving the input diode.

In this input method, a constant DC voltage is applied to the terminal 13, and a signal voltage superimposed on the DC voltage is applied to the terminal 14. As the pulse φ ₁ transitions from a high potential to a low potential at time t ₀ , the potential of the input diode 7 also decreases, and therefore electrons serving as signal charges are supplied from the input diode 7 directly below the input gate electrodes 1 and 2 . Ru. At time _t1 , the potential of the input diode 7 and the surface potential directly below the input gate electrodes 1 and 2 become approximately equal. Next, at time _t2 , as the pulse _φ1 transitions from a low potential to a high potential, the potential of the input diode 7 becomes high, and the excess charge existing directly under the input gate electrodes 1 and 2 is absorbed to the input diode 7 side. At the same time, signal charges approximately proportional to the potential difference between input gate electrodes 1 and 2 are transferred directly below transfer electrodes 3a and 3b, and at time _t3 all signal charges are accumulated directly below 3b, and the sampling operation is completed.

In this PR input method, when transitioning from time t ₁ to t ₂ , the potential of the input diode 7 becomes high, excess carriers are sucked out to the diode side, and at the same time signal carriers are transferred to the transfer electrode 3b. There is no period of equilibrium. Also, since φ ₁ becomes 0\volt at time t ₁ , if the transfer electrode 3a
When the threshold voltage immediately below becomes negative, charge directly leaks from the input diode to directly below the transfer electrode 3b.
In particular, the situation becomes worse when undershoot occurs in the pulse φ ₁ and the low potential of φ ₁ becomes a negative voltage. In this way, the PR input method does not require pulses to drive the input diode, so it is suitable for high-speed sampling operations, but it causes the above-mentioned disadvantages, resulting in extremely poor distortion characteristics at the input section. .

SUMMARY OF THE INVENTION An object of the present invention is to provide a new method for driving a charge-coupled device that eliminates the above-mentioned drawbacks.

According to the present invention, the signal input section is formed on a semiconductor substrate having one conductivity type, and has a signal input section, a charge transfer section, and a signal output section, and the signal input section is connected to a semiconductor region having a conductivity type opposite to that of the semiconductor substrate. The charge transfer section is composed of a first input gate electrode provided adjacent to the semiconductor region and a second input gate electrode provided adjacent to the first input gate electrode. In a charge-coupled device configured with a plurality of transfer electrode groups including one transfer electrode adjacent to an input gate electrode, when the charge-coupled device is in operation, the surface potential of at least a partial region directly under the one transfer electrode is There is obtained a method for driving a charge-coupled device, characterized in that the threshold voltage of at least a partial region under the one transfer electrode is controlled so that the absolute value is lower than the absolute value of the potential of the semiconductor region.

Next, the present invention will be explained in detail using the drawings.

In the following explanation, the case of N channels will be explained for the sake of simplicity. FIG. 3 shows an embodiment according to the present invention, and is a sectional view of the vicinity of the input section of the CCD. In FIG. 3, the same numbers as in FIG. 1 indicate the same objects. As is clear from FIG. 3, in the present invention, for example, a semiconductor region 1 having a conductivity type opposite to that of the semiconductor substrate 8 is provided directly under one transfer electrode 3a adjacent to the second input gate electrode 2 for threshold voltage control.
8 is provided. In the case of an N-channel semiconductor region 18, it can be easily formed using boron ion implantation or the like. In the present invention, by providing the semiconductor region 18 and controlling the threshold voltage directly under the transfer electrode 3a,
Controls the surface potential just below the electric current, secures potential equilibrium time at the input section, and at the same time prevents charge from leaking directly from the input diode to the CCD transfer channel, improving distortion characteristics at the input section. can be done. The device can be driven using the drive pulse waveform shown in FIG. 2, but it is more effective to use the pulse shown in FIG. 5. The difference between FIG. 5 and FIG. 2 is that the low potential of the pulse is not 0\volt but a constant potential _VTH . In Figure 5, φ ₁ ,
The low potential for both φ ₂ is V _TH , but in reality φ ₁
Only the low potential of V _TH is sufficient. here
V _TH is the charge from the input diode transferred to the transfer electrode 3a.
This is the minimum voltage required to prevent charges from flowing directly below them into the transfer channel, and above this voltage, charges will not flow into the transfer channel. Furthermore, in addition to the driving method shown in FIG. 5, it is also possible to use the pulse shown in FIG. 2 and to set the potential of the substrate to a negative potential with respect to the input diode. FIG. 4 shows the surface potential of the input section at each time t ₀ to _{t 3} shown in FIG. 5, corresponding to FIG. 3. In FIG. 4, the potential of the input diode and the surface potential directly under the transfer electrodes 3a and 3b at each time _t0 to _t3 are 20, 23, 30, 25, 24, 31, respectively.
26, 28, 32, 27, 29, 33. 21 and 22 indicate the surface potentials directly under the first and second input gate electrodes 1 and 2, respectively, in a depletion state. 34 to 37 are each time t ₀
It shows the accumulation state of carriers (electrons) distributed corresponding to ~ _t3 . First, at time t ₀ , the potential of the input diode 7 becomes 20 during the transition of the pulse φ ₁ from a high potential to a low potential, and the first input gate electrode 1
The carrier 31 becomes shallower than the surface potential 21 directly below it.
flows directly under the first and second input gate electrodes 1 and 2. At this time, since the same pulse φ ₁ as that of the input diode 7 is applied to one transfer electrode 3a, the surface potential 23 directly under the electrode 3a also changes as φ ₁ changes. However, since boron ions are implanted directly under the transfer electrode 3a to control its threshold voltage in a more positive direction, the surface potential 23 directly below the transfer electrode 3a has a more negative value than the input diode potential 20. In other words, the potential is shallower. When the pulse φ ₁ becomes 0 volts at time t ₁ , the potential of the input diode 7 is 25, and the surface potential directly under the transfer electrode 3a is 24, and at this time, the surface potential 24 is also a more negative value than the potential 25. Because of this, the carrier 35 is not transferred to the transfer electrode 3b. In other words, the potential of the input diode 7 is 0\
Even when it becomes a bolt, the carrier 35 will not leak out into the transfer channel in advance. Furthermore, even if the pulse φ ₁ causes undershoot and the low potential of φ ₁ becomes a negative value, the surface directly under the transfer electrode 3a will be in an accumulation state with a large amount of holes accumulated, so the carrier (electrons) can recombine with the holes and prevent them from proceeding into the transfer channel.

In order to prevent such charge from entering the transfer channel, the potential of the substrate 8 is input to the diode 7.
By setting the voltage to a negative potential, the operation at the input section can be made more reliable.

In the process of the pulse φ ₁ transitioning from a low potential to a high potential at time t ₂ , the potential 26 of the input diode 7 becomes deeper than the surface potential 21 directly under the first input gate electrode 1 and the potential 26 of the input diode 7 becomes deeper than the surface potential 21 directly below the first input gate electrode 1 . Excess carriers among the carriers 35 accumulated directly under the gate electrodes 1 and 2 are swept out to the input diode 7 side, corresponding to the difference in voltage applied to the first and second input gate electrodes 1 and 2. Carrier will stay. At this time, in the conventional device shown in FIG. 1, the surface potential immediately below the transfer electrode 3a also deepens, so that the carriers are transferred to the transfer channel before the excess carriers are sufficiently swept out to the input diode 7 side. This worsened the distortion rate. However, in the present invention, since the threshold voltage directly below the transfer electrode 3a has a higher positive value, the surface potential 28 remains at the potential of φ ₁ even in the process where the potential of the input diode deepens due to a change in φ ₁ . It does not follow changes immediately, but changes after a certain delay time. Furthermore, the effect of this threshold voltage control allows surface potential 28 to be located at a shallower value than potential 26. Therefore, compared to conventional devices, more time is required for potential equilibrium, leading to improvement in distortion rate.

At time _t3 , _φ1 becomes a high potential, and all of the carriers 33 are transferred directly below the transfer electrode 3b, completing the sampling operation. This carrier 33 is transferred and output by normal CCD operation.

As described above, the present invention is characterized by controlling the threshold voltage directly below the transfer electrode 3a, and it should be noted that the value of this threshold voltage is set at the transfer electrode when the pulse φ ₁ is at a low potential. When the surface potential directly under the transfer electrode 3a is high enough to prevent carriers from directly flowing into the transfer channel from the input diode 7, and when the pulse φ ₁ is at a high potential, the surface potential directly below the transfer electrode 3a becomes the second input. Gate electrode 2
It is necessary that the surface potential be controlled to be higher than the surface potential directly below it. Such control can be sufficiently achieved, for example, if boron ion implantation is used and the dose is selected on the order of 10 ¹¹ to 10 ¹² /cm ² . Naturally, the threshold voltage can also be controlled by other means, such as partially controlling the dielectric constant or film thickness of the insulating film 17.

As described above, according to the present invention, a CCD having high performance input characteristics even during high-speed operation can be obtained.

Further, in the above-mentioned device, an N-channel device has been described, but it is clear that the gist of the present invention can be similarly applied to a P-channel device. Furthermore, although the above description has been made using a two-phase drive device, other drive methods such as three-phase or four-phase drive may be used.

The present invention is also applicable to surface channel devices that do not use buried channels.

[Brief explanation of the drawing]

Figure 1 is a sectional view of the input section of a conventional CCD, Figure 2
The figure shows an example of a driving pulse waveform of a CCD, and FIG. 3 shows an embodiment of the present invention.
FIG. 4 shows the third data at each time _t0 to _t3 shown in FIG.
The potential and carrier distribution of each part of the CCD shown in the figure are shown, and FIG. 5 is a diagram for explaining the drive pulse waveform. In the figure, 1 and 2 are first and second input gate electrodes, 3a, 3b, ...,
6a and 6b are transfer electrodes of the CCD, and 7 is a semiconductor region having a conductivity type opposite to that of the semiconductor substrate 8, forming an input diode. 9 forms a buried channel with a semiconductor region having a conductivity type opposite to that of the semiconductor substrate 8;
10 to 12 are means for imparting directionality to CCD transfer; 13 to 16 are first and second input gate electrodes 1 and 2, and transfer electrodes 3a, 3b, . . .
..., 6a, 6b, terminals for applying a predetermined voltage, 17 an insulating film, 18 a threshold voltage control means, which in this example is formed of a semiconductor region having the same conductivity type as the semiconductor substrate 8. . 20 to 27 are the potentials of the input diode at each time, and 21 and 22 are the first and second input gate electrodes 1 and 2 in the depletion state.
The lower surface potentials 23 to 29 and 30 to 33 indicate the surface potentials immediately below the transfer electrodes 3a and 3b at each time, and 34 to 37 indicate the carrier distribution at each time.

Claims (1)

[Claims] 1 formed on a semiconductor substrate having one conductivity type,
a signal input section, a charge transfer section, and a signal output section; the signal input section includes a semiconductor region having a conductivity type opposite to that of the semiconductor substrate; and a first input gate electrode provided adjacent to the semiconductor region; a second input gate electrode provided adjacent to the first input gate electrode, and the charge transfer section includes a plurality of transfer electrode groups including one transfer electrode adjacent to the second input gate electrode. In the charge-coupled device thus configured, in the operating state of the charge-coupled device, the absolute value of the surface potential of at least a partial region directly under the one transfer electrode is equal to or greater than the absolute value of the potential of the semiconductor region. A method for driving a charge-coupled device, characterized in that the threshold voltage of at least a partial region under the one transfer electrode is controlled so as to be lowered.