Semiconductor device manufacturing: process – Making field effect device having pair of active regions... – Having insulated gate
Reexamination Certificate
1999-10-29
2001-01-30
Chaudhari, Chandra (Department: 2813)
Semiconductor device manufacturing: process
Making field effect device having pair of active regions...
Having insulated gate
C438S266000, C438S593000
Reexamination Certificate
active
06180454
ABSTRACT:
TECHNICAL FIELD
The present invention generally relates to flash memory devices such as EEPROMs. More particularly, the present invention relates to flash memory devices having improved poly 1—select gate contact performance.
BACKGROUND ART
Semiconductor devices typically include multiple individual components formed on or within a substrate. Such devices often comprise a high density section and a low density section. For example, as illustrated in prior art
FIG. 1
a,
a memory device such as a flash memory
10
comprises one or more high density core regions
11
and a low density peripheral portion
12
on a single substrate
13
. The high density core regions
11
typically consist of at least one M×N array of individually addressable, substantially identical floating-gate type memory cells and the low density peripheral portion
12
typically includes input/output (I/O) circuitry and circuitry for selectively addressing the individual cells (such as decoders for connecting the source, gate and drain of selected cells to predetermined voltages or impedances to effect designated operations of the cell such as programming, reading or erasing).
The memory cells within the core portion
11
are coupled together in a NAND-type circuit configuration, such as, for example, the configuration illustrated in prior art
FIG. 1
b.
Each memory cell
14
has a drain
14
a,
a source
14
b
and a stacked gate
14
c.
A plurality of memory cells
14
connected together in series with a drain select transistor at one end and a source select transistor at the other end to form a NAND string as illustrated in prior art FIG.
b.
Each stacked gate
14
c
is coupled to a word line (WL
0
, WL
1
, . . . , WLn) while each drain of the drain select transistors are coupled to a bit line (BL
0
, BL
1
, . . . , BLn). Lastly, each source of the source select transistors are coupled to a common source line Vss. Using peripheral decoder and control circuitry, each memory cell
14
can be addressed for programming, reading or erasing functions.
Prior art
FIG. 1
c
represents a fragmentary cross section diagram of a typical memory cell
14
in the core region
11
of prior art
FIGS. 1
a
and
1
b.
Such a cell
14
typically includes the source
14
b,
the drain
14
a
and a channel
15
in a substrate or P-well
16
; and the stacked gate structure
14
c
overlying the channel
15
. The stacked gate
14
c
further includes a thin gate dielectric layer
17
a
(commonly referred to as the tunnel oxide) formed on the surface of the P-well
16
. The stacked gate
14
c
also includes a polysilicon floating gate
17
b
which overlies the tunnel oxide
17
a
and an interpoly dielectric layer
17
c
overlies the floating gate
17
b.
The interpoly dielectric layer
17
c
is often a multilayer insulator such as an oxide-nitride-oxide (ONO) layer having two oxide layers sandwiching a nitride layer. Lastly, a polysilicon control gate
17
d
overlies the interpoly dielectric layer
17
c.
The control gates
17
d
of the respective cells
14
that are formed in a lateral row share a common word line (WL) associated with the row of cells (see, for example, prior art
FIG. 1
b
). In addition, as highlighted above, the drain regions
14
a
of the respective cells in a vertical column are connected together by a conductive bit line (BL). The channel
15
of the cell
14
conducts current between the source
14
b
and the drain
14
a
in accordance with an electric field developed in the channel
15
by the stacked gate structure
14
c.
The select transistors have a stacked gated structure similar to the memory cells, except that the first polysilicon layer (floating gate) and the second polysilicon layers (control gate) are shorted together to form a single gate type structure. Select gates typically ensure the selectivity of a particular bit line and prevent the memory cells from passing current during the program operation.
A poly 1 contact is made to the first polysilicon gate of the select transistor while a poly 2 contact is made to the tungsten silicide/second polysilicon gate (control gate) of the memory cells. These contacts are typically formed through an oxide layer covering the memory cells and select transistors (a so-called interlayer dielectric). These contacts are also typically formed at about the same time (or during the same processing window). However, while the depth of the poly 2 contact is smaller than the depth of the poly 1 contact, the time required to form an high quality poly 2 contact is undesirably greater than the amount of time required to form a poly 1 contact.
This phenomenon leads to deleterious overetching of the poly 1 layer (first polysilicon layer). In some instances, the overetching is severe enough that the poly 1 contact opening extends into the field oxide and/or even into the substrate. Such overetching is termed punch-through. Poly 1 contact punch-through renders the select gate inoperable, consequently resulting in failure of the flash memory device. In instances where the overetching is not so severe (where the poly 1 contact opening extends deep into the first polysilicon layer, but not completely through), there is a resultant increase in poly 1 contact resistivity. Increases in poly 1 contact resistivity can lead to parametric failure of the flash memory device, and/or latent reliability problems since the increased contact resistivity often leads to thermal heating, which, in turn, further increases the resistivity often leading to device failure.
Attempts to address punch-through involve making the first polysilicon layer thicker. But the thicker the first polysilicon layer, the more likely undesirable cracking occurs in the tungsten silicide layer and the more likely etch problems occur due to high aspect ratios and high topographies. Moreover, there are numerous other concerns associated with setting the thickness of the first polysilicon layer. For example, the thicker the floating gate, the higher the stress released on the tunnel oxide layer, improving tunnel oxide reliability, improving conductivity and thus resulting in better circuit performance. The thinner the floating gate, the more likely an increase in pinhole defects. Further, when the thickness of the Poly 1 is too thin, an HF dip cleaning step (prior to forming the ONO multilayer dielectric film) may degrade the Poly 1 and attack the tunnel oxide. Therefore, it is desirable to set the thickness of the first polysilicon layer without regard to punch-through problems, since there are other important concerns to address. It is also desirable, nevertheless to address the undesirable punch-through problem.
SUMMARY OF THE INVENTION
As a result of the present invention, a flash memory device having improved reliability is obtainable by providing an improved poly 1—select gate contact. By forming a poly 1 contact in accordance with the present invention, punch-through of the poly 1 layer and field oxide layer by the poly 1 contact is minimized and/or eliminated. Moreover, by forming a poly 1 contact in accordance with the present invention, the resistivity of the poly 1—select gate contact is reduced, thereby improving the performance and reliability of flash memory devices. Addressing punch-through and resistivity problems without regard to the thickness of the floating gate, enables the selection of a floating gate thickness based on fewer concerns, thereby better addressing the concerns associated with floating gate thickness. According, also as a result of the present invention, producing a flash memory device where the occurrence of microcracking in the tungsten silicide conductive layer is minimized and/or eliminated, where pinhole defects are minimized and/or eliminated, and/or where HF dip cleaning degradation is minimized and/or eliminated is possible.
In one embodiment, the present invention relates to a method of forming a flash memory device involving the steps of forming a gate oxide layer on a substrate; forming a first poly layer over the gate oxide layer; forming an insulating layer ov
Chang Kent Kuohua
Ou Wei-Wen
Wang John Jianshi
Advanced Micro Devices , Inc.
Amin, Eschweiler & Turocy, LLP
Chaudhari Chandra
Chen Jack
LandOfFree
Method for forming flash memory devices 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 forming flash memory devices, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for forming flash memory devices will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2529603