Active solid-state devices (e.g. – transistors – solid-state diode – Field effect device – Having insulated electrode
Reexamination Certificate
2003-03-10
2004-11-02
Flynn, Nathan J. (Department: 2826)
Active solid-state devices (e.g., transistors, solid-state diode
Field effect device
Having insulated electrode
C257S288000, C257S368000, C257S411000, C257S406000, C438S585000, C438S587000, C438S588000, C438S592000
Reexamination Certificate
active
06812536
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a semiconductor device and more particularly to a structure of a MOS (i.e., metal-oxide semiconductor) transistor.
2. Description of the Background Art
The gate electrode of a MOS transistor serves as part of wiring. In this respect, it is desirable to reduce the resistance of gate electrode lines as much as possible. However, the recent semiconductor devices continue to downsize to smaller dimensions. This forcibly narrows the width of respective gate electrode lines. The resistance of gate electrode lines increases correspondingly, inducing a problem that the undesirable voltage drop occurs or the response is delayed. To solve this problem, a conventionally proposed prospective method is to employ a poly-metal gate (i.e., a combination of polysilicon and metal) as the gate electrode.
FIG. 59
is a cross-sectional diagram showing the arrangement a MOS transistor having a poly-metal gate structure which is disclosed in Japanese Patent Application Laid-Open No. 2002-76336.
As shown in the drawing, source and drain regions
52
are selectively formed in an upper layer portion of a silicon substrate
51
. Extensively overlying on the upper surface of the silicon substrate
51
is a SiO
2
film
53
which serves as a gate oxide film. A doped polysilicon layer
54
, formed on the SiO
2
film
53
, extends in a limited region corresponding to a gap between the source and drain regions
52
. Formed on the doped polysilicon layer
54
is a tungsten layer
55
. The doped polysilicon layer
54
and the tungsten layer
55
cooperatively constitute a poly-metal gate
50
.
Smile Oxidation
An edge portion of the gate electrode partly overlaps with an impurity diffusion area of the source and drain regions
52
which has a relatively high impurity concentration. When a voltage of an accumulation direction is applied on the gate electrode, for example, when the gate voltage is smaller than the drain voltage in a NMOS transistor, there is a tendency that a leak current, so-called GIDL (i.e., gate induced drain leakage current), may flow in this overlap region.
FIG. 60
is a cross-sectional diagram explaining the GIDL phenomenon. As shown in this drawing, an inner depletion layer
61
and an outer depletion layer
62
appear along the boundary of the source or drain region
52
in accordance with the potential distribution. In this case, the upper edge of the inner depletion layer
61
develops under a region of the SiO
2
film
53
located just beneath the doped polysilicon layer
54
. The upper portion of depletion layer
61
elongated in this manner under the doped polysilicon layer
54
is a high electric field region
63
positioned adjacently to the edge of the doped polysilicon layer
54
. The electric field of the gate edge portion can be simply expressed by {(Vg−Vd)/tOX}, where Vg represents a gate voltage, Vd represents a drain voltage, tOX represents a gate oxide film thickness. One of the methods for suppressing the GIDL phenomenon is a smile oxidation.
FIG. 61
is a cross-sectional diagram explaining the smile oxidation. The smile oxidation (also referred to as a poly-smile oxidation as a light thermal reoxidation) is a technique for forming a smile oxide film
56
by performing an oxidation process after forming a gate electrode. As shown in
FIG. 61
, the smile oxide film
56
has a large film thickness at the limited region near the gate edge.
Having a large film thickness at the region near the gate edge makes it possible for the smile oxide film
56
to relax the electric field in the vicinity of the gate edge portion. The GIDL phenomenon can be reduced correspondingly. Furthermore, the smile oxide film
56
has a relatively thin film thickness at the central region of the gate. Hence, the smile oxide film
56
can minimize the reduction in the drain current during an ON state.
Selective Oxidation
From the foregoing description, the ordinary person skilled in the art will simply expect that applying the smile oxidation process to the poly-metal gate structure may bring the effects of lowering the gate resistance and suppressing the GIDL phenomenon.
FIG. 62
is a cross-sectional diagram explaining a problem occurring when the smile oxidation process is applied to the poly-metal gate structure. Needless to say, it is well known that the metals, such as iron, copper, and aluminum, are readily oxidized as apparent from the generation of rust caused when the metals are oxidized.
In general, tungsten (W) is a metal material generally used for forming the poly-metal gate structure. However, compared with other metals, tungsten (W) is not an exception in that tungsten (W) easily bonds with oxygen to form an oxide having a higher resistance value. More specifically, as shown in
FIG. 62
, applying the smile oxidation process to the poly-metal gate structure causes the oxidation in the tungsten layer
55
and leaves an affected tungsten layer
55
o
. The resistance value of the poly-metal gate becomes large due to the presence of thus formed affected tungsten layer
55
o
. This is a problem to be solved in realizing an excellent poly-metal gate structure sufficiently low in the resistance value. From the view point that the capability of reducing the sheet resistance cannot be enjoyed when the metal is oxidized, this problem is fatal and will result in a negation of metal use.
To overcome this problem, there is known a conventional method according to which a selective oxidation technique is employed for performing the smile oxidation. The selective oxidation technique is characterized by an oxidation process performed in a strong reducing atmosphere, for example, containing a large amount of hydrogen. According to the selective oxidation technique, the oxide of tungsten (W), if it is once produced by the bonding of W and oxygen, promptly reduces to original W and oxygen. Thus, it becomes possible to minimize the chemical reaction to be caused between the tungsten (W) and oxygen.
However, the selective oxidation technique requires a sensitive or exquisite control for properly maintaining the balance between two directly-opposed, i.e., oxidation and reducing, phenomena. Accordingly, the required manufacturing or fabricating conditions are very severe. There is no degree of freedom. Simultaneously supplying both of oxygen and hydrogen significantly limits the temperature in the forming process for the purpose of avoiding possible dangers.
FIG. 63
is a cross-sectional diagram (Part I) explaining a problem peculiar to the selective oxidation process. As shown in
FIG. 63
, when the smile oxide film
56
having a sufficient thickness is formed in the vicinity of the edge, it is necessary to shorten the process time in view of the cost (or throughput) requirements. To this end, the process temperature needs to be maintained at a higher level. However, the above-described limit of the temperature in the forming process substantially prohibits the process being performed in such a higher temperature environment. On the other hand, if the supply of hydrogen is reduced to maintain the balance between the oxidation and reducing phenomena during the selective oxidation process, it will be unable to sufficiently suppress the oxidation of W.
From the foregoing reasons, as shown in
FIG. 63
, at least part of the tungsten layer
55
turns into the affected tungsten layer
55
o
as a result of oxidation. The gate resistance increases. In this manner, the selective oxidation technique requires a sensitive or exquisite control for properly maintaining the balance between two directly-opposed, i.e., oxidation and reducing, phenomena. From this fact, an upper limit of the film thickness of the smile oxide film, in the vicinity of the edge, is undesirably restricted to a smaller value. Furthermore, if the process time is elongated to obtain a sufficient film thickness under given conditions, it is needles to say that the throughput will be worsened.
FIG. 64
is a cross-sectional diagram (Part II) explain
Kinugasa Akinori
Nishida Yukio
Ohto Kenichi
Shiratake Shigeru
Terauchi Takashi
Flynn Nathan J.
Renesas Technology Corp.
Wilson Scott R.
LandOfFree
MOSFET with graded gate oxide layer does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with MOSFET with graded gate oxide layer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and MOSFET with graded gate oxide layer will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3352051