Active solid-state devices (e.g. – transistors – solid-state diode – Field effect device – Having insulated electrode
Reexamination Certificate
2001-03-21
2003-07-22
Ho, Hoai (Department: 2818)
Active solid-state devices (e.g., transistors, solid-state diode
Field effect device
Having insulated electrode
C257S411000, C257S324000, C257S325000, C438S216000
Reexamination Certificate
active
06597047
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a nonvolatile semiconductor memory device and a method for fabricating the device, and more particularly relates to measures to be taken to improve the reliability of a nonvolatile semiconductor memory device.
A known nonvolatile semiconductor memory device, like a flash EEPROM, typically has a structure in which a floating gate electrode for storing charge thereon is interposed between a control gate electrode, functioning as a gate electrode for an MIS transistor, and a channel region. Normally, information stored on the floating gate electrode can be read by determining the state of the MIS transistor as ON or OFF, which depends on whether there is any charge stored on the floating gate electrode or not. In a nonvolatile semiconductor memory device like this, information on the floating gate electrode is alterable by utilizing a charge tunneling phenomenon occurring in a tunnel insulating film located under the floating gate electrode. That is to say, the charge can be injected into, or removed from, the floating gate electrode by taking advantage of that tunneling phenomenon. A tunnel insulating film is usually an oxide film. However, it is already known empirically that the tunnel insulating film as a gate insulating film deteriorates with time, or through repetitive passage of charge through the tunnel insulating film. For that reason, various techniques have been proposed to improve the reliability of the tunnel insulating film.
Hereinafter, it will be described with reference to
FIGS. 13A through 13D
how the tunnel insulating film can have its reliability improved in a known nonvolatile semiconductor memory device.
First, in the process step shown in
FIG. 13A
, an isolation region
102
and an active region
103
, surrounded by the isolation region
102
, are defined in a p-type semiconductor substrate
101
. Next, in the process step shown in
FIG. 13B
, a tunnel insulating film
104
a
is formed to a thickness of about 10 nm on the surface of the substrate
101
. Then, in the process step shown in
FIG. 13C
, first polysilicon, ONO and second polysilicon films are stacked in this order over the substrate, and then patterned, along with the tunnel insulating film
104
a,
into a predetermined shape. In this manner, a gate electrode section
108
, including floating gate electrode
105
, interelectrode insulating film
106
and control gate electrode
107
, is formed. As used herein, the “ONO film” is a multilayer structure consisting of oxide, nitride and oxide films that have been stacked one upon the other. Finally, in the process step shown in
FIG. 13D
, a sidewall
109
is formed on the side faces of the tunnel insulating film
104
, floating gate electrode
105
, interelectrode insulating film
106
and control gate electrode
107
. Then, ions of an n-type dopant are implanted into the substrate
101
using the gate electrode section
108
and sidewall
109
as a mask, thereby defining n-type source/drain regions
110
and
111
in the substrate
101
on the right- and left-hand sides of the gate electrode section
108
.
In the prior art, a write operation is performed by injecting electrons from the channel region, which is part of the substrate
101
located under the tunnel insulating film
104
, into the floating gate electrode
105
by way of the tunnel insulating film
104
. The electrons may be injected by utilizing an FN tunneling phenomenon, for example. On the other hand, the electrons can be removed from the floating gate electrode
105
into the channel region of the substrate
101
. However, it is known that the greater the number of times the electrons pass through the tunnel insulating film
104
by the FN tunneling, the greater the number of defects (e.g., trap sites) created in the tunnel insulating film
104
and the less reliable the film
104
becomes. Thus, a proposed technique attempts to suppress the creation of defects such as trap sites by diffusing nitrogen atoms into the oxide film as the tunnel insulating film.
However, I found as a result of various experiments that even if nitrogen atoms are diffused into the tunnel insulating film, it is still difficult to suppress the degradation of the tunnel insulating film effectively. So I looked into the reasons to make the following findings.
Generally speaking, when nitrogen atoms are diffused into a silicon dioxide film (which is a thermal oxide film), the lower part of the tunnel insulating film closer to the substrate has its quality improved, whereas the upper part thereof closer to the floating gate electrode does not. This is because the nitrogen atoms exist at a relatively high density in the lower part of the tunnel insulating film near the interface with the substrate, while almost no nitrogen atoms exist in the upper part of the tunnel insulating film near the interface with the floating gate electrode.
In a known annealing process for forming a thermal oxide film, the oxide film is usually formed by a pyrolytic oxidation using oxygen and hydrogen gases. In an oxide film formed by the pyrolytic oxidation, a lot of oxygen atoms are contained. However, it is known that these oxygen atoms terminate dangling bonds included in the oxide film, thereby reducing a stress produced in the underlying semiconductor substrate and contributing to the performance enhancement of the resultant transistor. That is to say, it is known that a silicon dioxide film formed by the pyrolytic oxidation is much more reliable than a counterpart formed by a dry oxidation using oxygen gas only.
However, the experimental data I collected told me that while nitrogen atoms were being diffused into a thermal oxide film, hydrogen atoms, which had been introduced into the thermal oxide film by a pyrolytic oxidation process, might diffuse outward. This experimental data will be detailed later. And I believe that a tunnel insulating film, which has been subjected to the nitrogen diffusion process, has its quality degraded because hydrogen atoms, existing near the surface of the oxide film, diffuse outward to create charge trapping sites near the surface as a result of an annealing process at an elevated temperature. Hereinafter, it will be described how I believe the tunnel insulating film deteriorates.
FIG. 14
is a band diagram illustrating energy band structures for a cross section passing the floating gate electrode, tunnel insulating film and semiconductor substrate. Specifically,
FIG. 14
illustrates how electrons are injected from the substrate into the floating gate electrode by way of the tunnel insulating film. As shown in
FIG. 14
, while electrons are injected from the substrate into the floating gate electrode by utilizing the FN tunneling, dangling bonds may exist in the upper part of the tunnel insulating film (i.e., a thermal oxide film where nitrogen atoms have been diffused) near the floating gate electrode. This is because hydrogen atoms may have diffused outward and may be absent from that part. In that case, holes may be trapped at those dangling bonds. In the lower part of the tunnel insulating film near the substrate on the other hand, nitrogen atoms exist at a relatively high density as described above. Accordingly, it is believed that even if dangling bonds have been formed there due to the outward diffusion of hydrogen atoms, those dangling bonds are terminated with the nitrogen atoms and the probability of hole trapping is not so high there.
FIG. 15
is a band diagram illustrating energy band structures for a cross section passing the floating gate electrode, tunnel insulating film and semiconductor substrate. Specifically,
FIG. 15
illustrates a state where holes have been trapped in the upper part of the tunnel insulating film near the floating gate electrode. As shown in
FIG. 15
, if holes have been trapped in the upper part of the tunnel insulating film, then the energy band structure of the tunnel insulating film changes so that the potential level locally drops in part of the conduction band of the tunnel insulating fi
Arai Masatoshi
Hashidzume Takahiko
Ho Hoai
Le Dung A
Nixon & Peabody LLP
Studebaker Donald R.
LandOfFree
Method for fabricating a nonvolatile semiconductor device does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for fabricating a nonvolatile semiconductor device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for fabricating a nonvolatile semiconductor device will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3105929