Active solid-state devices (e.g. – transistors – solid-state diode – Field effect device – Having insulated electrode
Reexamination Certificate
2001-06-28
2003-12-09
Eckert, George (Department: 2815)
Active solid-state devices (e.g., transistors, solid-state diode
Field effect device
Having insulated electrode
C257S222000, C257S291000
Reexamination Certificate
active
06661045
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a structure of a device of a MOS type solid-state imager, and more specifically, to a MOS solid-state imager having a MOS transistor with a short gate length (channel length) and a thin gate oxide film, where the problem of punch-through is likely to occur.
2. Description of the Related Art
FIG. 1
is a picture element (pixel) of a conventional MOS type solid-state imager.
The picture element is constituted of a photodiode
21
for converting an optical signal to an electric signal (charge), a read gate (MOS transistor)
22
for transferring the charge converted by the photodiode
21
to a detecting section D (detecting node), a reset gate (MOS transistor)
23
for resetting the charge (potential) of the detecting section D, a sense gate (MOS transistor)
24
for amplifying the potential of the detecting section D, and a select gate (MOS transistor)
25
for outputting the potential of the picture element selected by an address signal.
The charge photoelectrically converted by the photodiode
21
and accumulated in a signal accumulation region per unit time is transferred to the detecting section D by way of the read gate
22
, thereby changing the potential of the detection section D. The sense gate
24
amplifies the change in potential of the detection section D.
In a MOS-type solid-state imager, in order to completely transfer all the charge accumulated in the signal accumulating region of the photodiode
21
(photoelectric conversion section) to the detecting section D and further in order to stabilize the characteristics of the photodiode
21
within each of all picture elements, the impurity concentration of a semiconductor substrate (or a well region) is required to be set as low as possible.
However, if a MOS transistor is miniaturized to increase the number of picture elements (to increase the density of picture elements) in the case where the substrate contains impurities in a low amount, punch-through occurs. In other words, when the MOS transistor is reduced in size, the length of the gate (channel length) becomes shorter and the gate oxide film becomes thinner, as a natural consequence. As a result, “punch-through” occurs. “Punch through” is a phenomenon in which charges flow from the source to the drain of the MOS transistor even under the gate control.
When punch-through occurs, an unnecessary signal (unnecessary charge) flows through the MOS transistor, preventing the solid-state imager from operating normally.
Therefore, it is necessary to prevent the occurrence of punch-through. To prevent punch-through, a punch-through preventing region is provided inside a semiconductor substrate (at a position sufficiently deep from the surface) in conventional logic products.
The punch-through preventing region plays a role in preventing the leakage between the source and drain of the MOS transistor. When a MOS transistor having a p-type substrate and an n-type source/drain, a p-type punch-through preventing region is usually used. Such a punch-through preventing region is effective in the logic products. It efficiently prevents punch-through.
However, the MOS type solid-state imager has a problem. This is because a photodiode must be formed inside the semiconductor substrate at a position sufficiently deep from the surface. More specifically, a photodiode is usually formed of a p-type semiconductor substrate and an n-type signal accumulation region (impurity region. Therefore, the n-type signal accumulation region must be formed inside the semiconductor substrate at a position sufficiently deep from the surface.
In the case where the punch-through region is formed inside the semiconductor substrate, the conductive type (e.g. n-type) of an impurity (e.g., phosphorus) constituting the signal accumulation region is opposite to that (e.g. p-type) of an impurity (e.g., boron) constituting the punch-through preventing region. In addition, the signal accumulating region and the punch-through preventing region are formed at almost the same position sufficiently deep from the surface inside the semiconductor substrate, as mentioned above.
In the circumstances, if the punch-through preventing region is attempted to be formed in the MOS type solid-state imager, the punch through preventing region is first formed, and then, the signal accumulating region must be formed within the formed punch-through preventing region. To form the signal accumulating region, an impurity (n-type) must be doped in an amount sufficient to invert the conductive type (p-type) of the punch-through preventing region.
On the other hand, to completely read out all the charge photoelectrically converted and accumulated in the signal accumulation region of the photodiode, it is important to reduce the depletion potential of the photodiode as much as possible. To reduce the depletion potential, it is preferable that the signal accumulation region be stably formed with an impurity concentration as low as possible.
However, to form the signal accumulating region within the punch-through preventing region, an impurity (n-type) must be doped into the semiconductor substrate in an amount sufficient to invert the conductive type (e.g. p-type) of the punch-through preventing region. It is at least required that the concentration of the n-type impurity to be doped in the semiconductor substrate be higher than that of the p-type impurity of the punch-through preventing region.
In this case, the effect of the p-type impurity and that of the n-type impurity are considered to cancel each other out. Therefore, the impurity concentration of the signal accumulating region is given by the following equation:
Impurity concentration=dn−dp
where dp is the concentration of the p-type impurity constructing the punch-through preventing region, and dn is the concentration of the n-type impurity ion-doped in the semiconductor substrate.
However, both impurity concentrations dn and dp are extremely large. When the large numerical value (dp) is subtracted from the large numerical value (dn) to give a small value, a small change in the large numerical values results in a big change in the small numerical value. For this reason, when the punch-through preventing region is formed in an MOS type solid-state imager, it is very difficult to stably form the signal accumulating region containing the small amount of impurities.
In other words, a small variation in the concentration of the n-type impurity to be ion-doped into the semiconductor substrate results in a large variation in the impurity concentration of the signal accumulating region. Accordingly, the deletion potential of the diode is greatly changed, with the result that the charge of the signal accumulation region cannot be stably read out.
As described in the foregoing, in the MOS type solid-stage imager, with an increase in the number of picture elements (an increase of picture elements in density), the gate length of a MOS transistor becomes shorter and the gate oxide film becomes thinner. Under the circumstance, the phenomenon of punch-through has become a significant problem. On the other hand, it is very difficult to simply apply a punch-through preventing region used in a logic product to the MOS type solid-state imager, due to the presence of the signal accumulation region in the photodiode.
To transfer the charge with certainty, it is preferable that both the impurity concentration of the signal accumulating region and the depletion potential of the photodiode be set low and stable. However, when the punch-through preventing region is formed, the signal accumulating region must be formed by inverting the conductive type of the punch-through preventing region. It is therefore impossible to stably form the signal accumulating region containing a low concentration of an impurity.
To summarize, as the MOS transistor is miniaturized in a conventional MOS type solid-state imager, the problem of punch-through occurs. In this case, if a punch-through preventing region is formed to
Eckert George
Frommer & Lawrence & Haug LLP
Kabushiki Kaisha Toshiba
LandOfFree
MOS type solid-state imager and manufacturing method thereof does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with MOS type solid-state imager and manufacturing method thereof, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and MOS type solid-state imager and manufacturing method thereof will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3183867