Active solid-state devices (e.g. – transistors – solid-state diode – Field effect device – Having insulated electrode
Reexamination Certificate
1997-08-25
2004-03-09
Nadav, Ori (Department: 2811)
Active solid-state devices (e.g., transistors, solid-state diode
Field effect device
Having insulated electrode
C257S412000, C257S413000
Reexamination Certificate
active
06703672
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to the field of semiconductor devices, and more specifically, to a process designed to improve the surface planarity of polysilicon gate layers.
2. Background Information
In order to fabricate high performance metal oxide semiconductor (MOS) transistors, it is crucial to control gate electrode linewidths. Improved control of the gate electrode linewidth allows the formation of smaller channel lengths and increases the performance of MOS transistors. Non-planar polysilicon layer surfaces degrade the ability to control linewidths during lithographic processing for gate patterning. This degradation of control is generally due to variations in photoresist thickness and irregular light reflections off the polysilicon surface.
Polycrystalline silicon (polysilicon) is a preferred gate electrode material for MOS devices because it is easy to deposit and easy to dope. Polysilicon, however, due to its grain structure, forms a relatively rough surface layer. It is this rough surface layer of polysilicon that interferes with lithographic patterning and decreases linewidth control.
FIG. 1
a
illustrates a cross sectional view of polysilicon layer
130
deposited above thin gate oxide layer
120
, isolation regions
110
&
111
, and substrate
100
. The rough surface of polysilicon layer
130
caused by the grain structure of polysilicon is also illustrated in
FIG. 1
a.
The rough surface of polysilicon layer
130
causes reflection and scattering effects during photolithographic patterning of a photoresist.
FIG. 1
b
illustrates a cross sectional view of mask
150
and photoresist layer
140
coated on the polysilicon layer
130
of
FIG. 1
a.
It is a well known process in the art to pattern polysilicon gate layers using a patterned photoresist. In order to pattern photoresist layer
140
, mask
150
is used to block the light. As shown in
FIG. 1
b
mask
150
only covers a portion of photoresist layer
140
. The uncovered regions of photoresist layer
140
are exposed to light. Exposure to light causes the uncovered portions of photoresist layer
140
to become soluble.
FIG. 1
b
illustrates mask
150
and photoresist layer
140
being exposed to light. Because the light rays are being reflected and scattered by the rough surface of polysilicon layer
130
, regions of photoresist layer underlying the mask are exposed to light and become soluble. It should be noted and it will be obvious to one with ordinary skill in the art that although a positive photoresist process is described, a negative photoresist process may also be used.
After photoresist layer
140
has been subjected to a developing solution only the insoluble portions of photoresist layer
140
remain, as shown in
FIG. 1
c.
The insoluble portions of photoresist layer are the portions of photoresist layer
140
not exposed to light. Due to the reflection and scattering of the light by polysilicon layer
130
, photoresist layer
140
is poorly patterned. The features of photoresist layer
140
have poor edge definition and have varying horizontal dimensions.
The patterned photoresist is then used to pattern the polysilicon layer
130
into a gate electrode. Polysilicon gate layer
130
is patterned using well known etch techniques, such as, reactive ion etch (RIE), to form a polysilicon gate electrode as shown in
FIG. 1
d.
Because the patterned photoresist layer
140
(in
FIG. 1
c
) has wavy edges and varying horizontal dimensions, the gate electrode
130
in
FIG. 1
d
is formed with poor edge definition and varying linewidth. Therefore, the poorly patterned photoresist layer can cause the gate electrode to be formed with varying gate-lengths.
FIG. 1
e
illustrates a top view of the gate electrode
130
in
FIG. 1
d.
The variation in gate-length of the gate electrode can cause variations in channel length. Variation of the channel length varies the electrical characteristics of an MOS device and must be carefully controlled.
Presently there are several techniques for improving gate electrode linewidth control in the manufacture of high performance MOS transistors. The first technique is known as amorphous silicon deposition. Amorphous silicon deposition eliminates the surface roughness caused by the grain structure of polysilicon, however, a degradation in the electrical performance of the gate electrode is common in pure amorphous silicon gates. Additionally, amorphous silicon deposition has problems with deposition defects and is also more difficult to fully dope.
A second technique for improving linewidth control is the use of an amorphous silicon/polysilicon composite gate electrode. An amorphous silicon layer is deposited first and then a polysilicon layer is deposited on top of the amorphous silicon layer. However, such a composite layer may still exhibit surface roughness unless an intervening oxide layer is deposited or grown above the amorphous silicon layer. Because the amorphous silicon layer is deposited first, crystal growth will occur in the deposition of the top polysilicon layer. Unless an intervening oxide layer is used there is nothing to constrain the smooth surface of the amorphous silicon layer during the deposition of the polysilicon layer. In other words, recrystallization of the amorphous silicon layer cannot be controlled. Consequently, the amorphous silicon layer may recrystallize with a rough surface. Thus, the amorphous silicon/polysilicon composite gate may exhibit the surface roughness found in a pure polysilicon layer, and the poor electrical performance, deposition defects, and decreased dopant uniformity exhibited in pure amorphous silicon gates.
A third technique for improving linewidth control is the use of anti-reflective layers as part of the lithographic process. The use of anti-reflective layers reduces the effect of the surface roughness caused by the grain structure of polysilicon. An anti-reflective layer is highly absorbing and reduces the reflection and scattering effects caused by the rough surface of the polysilicon layer. Anti-reflective layers, however, require special processing equipment and add additional steps, i.e. depositing and etching the anti-reflective layer. The need for special processing equipment significantly increases the cost of manufacturing the MOS transistors. Also, the additional steps required by the use of anti-reflective layers increases the defect level normally associated with the production of MOS transistors.
Another technique for improving linewidth control is the use of dual layer resists when patterning a polysilicon gate electrode. In a dual layer resist a first planarizing resist layer is deposited and a second resist layer is spun on top of the first resist layer. The first resist layer may contain a die such that it exhibits many of the absorbing qualities of an antireflective coating. The second resist layer is generally a constant thickness to aid in the control of thin film interference effects. However, the use of the dual resist increases the cost of manufacturing the MOS transistors and, much like the antireflective coating, requires additional processing steps. Anytime additional processing steps are added increases the likelihood that the defect level normally associated with the production of MOS transistors will increase.
Thus, what is needed is a method for the formation of a gate electrode that combines the electrical performance of polysilicon with the surface planarity of amorphous silicon thereby improving linewidth control using standard lithographic techniques without increasing the defect level already associated with the manufactured of high performance MOS transistors.
SUMMARY OF THE INVENTION
The present invention describes a method for improved linewidth control in the patterning of polysilicon layers, which are used to form gate electrodes, in the manufacture of metal oxide semiconductor (MOS) devices. The preferred embodiment of the present invention forms a composite gate electrode by depositing a layer of polysilicon above a substra
Brigham Lawrence N.
Jan Chia-Hong
Zhang Binglong
Blakely , Sokoloff, Taylor & Zafman LLP
Nadav Ori
LandOfFree
Polysilicon/amorphous silicon composite gate electrode does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Polysilicon/amorphous silicon composite gate electrode, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Polysilicon/amorphous silicon composite gate electrode will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3274376