Semiconductor device manufacturing: process – Making field effect device having pair of active regions... – Having insulated gate
Reexamination Certificate
2000-03-21
2001-11-06
Lee, Eddie (Department: 2815)
Semiconductor device manufacturing: process
Making field effect device having pair of active regions...
Having insulated gate
C438S261000, C438S257000, C438S618000, C438S626000, C438S631000, C257S315000, C257S316000
Reexamination Certificate
active
06312991
ABSTRACT:
FIELD OF THE INVENTION
The present invention generally relates to integrated circuits and, in particular, to a method of flash memory fabrication which improves memory cell reliability and manufacturability by substantially preventing poly1 punch-through by removing the poly cap layer from the stacked gate structure in a flash memory device. Consequently, when forming contacts down to poly1 and poly2, a substantial overetch is not required which prevents the occurrence of both high resistivity poly1 contacts and punch-through down to the substrate.
BACKGROUND OF THE INVENTION
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 memory device such as a flash memory
10
comprises one or more high density core regions
12
and a low density peripheral portion
14
on a single substrate
16
. The high density core regions
12
typically consist of at least one M×N array of individually addressable, substantially identical memory cells and the low density peripheral portion
14
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
12
are coupled together in a circuit configuration, such as a NAND configuration illustrated in prior art FIG.
2
. Each memory cell
20
has a drain
22
, a source
24
and a stacked gate
26
. Each stacked gate
26
is coupled to a word line (WL
1
, WL
2
, . . . , WL
8
) while the drains
22
and sources
24
are coupled together in series to form a bit line BL through two select gate transistors
27
(SG
1
and SG
2
, respectively). Using peripheral decoder and control circuitry, each memory cell
20
can be addressed for programming, reading or erasing functions.
Prior art
FIG. 3
represents a fragmentary cross section diagram of a typical memory cell
20
and a select gate transistor
27
in the core region
12
of prior art
FIGS. 1 and 2
. The memory cell
20
typically includes the drain
22
, the source
24
and a channel
28
in a substrate
30
; and the stacked gate structure
26
overlying the channel
28
. The stacked gate
26
further includes a thin gate dielectric layer
32
(commonly referred to as the tunnel oxide) formed on the surface of the substrate
30
. The stacked gate
26
also includes a polysilicon floating gate
34
which overlies the tunnel oxide
32
and an interpoly dielectric layer
36
overlies the floating gate
34
. The interpoly dielectric layer
36
is often a multilayer insulator such as an oxide-nitride-oxide (ONO) layer having two oxide layers
36
a
and
36
b
sandwiching a nitride layer
36
c
. Lastly, a top region
38
overlies the interpoly dielectric layer
36
. The top region
38
includes a control gate layer
40
formed of polysilicon, a tungsten silicide layer (WSi
x
)
42
which overlies the control gate
40
and a poly cap layer
44
formed predominately of polysilicon.
The control gate
40
provides memory functions for the cell
20
as will be described in greater detail infra while the tungsten silicide layer
42
provides a low resistivity contact as is well known by those skilled in the art. The poly cap layer
44
overlies the tungsten silicide layer
42
and prevents the silicide layer
42
from peeling or cracking. The top regions
38
of the respective cells
20
that are formed in a lateral row (as opposed to a vertical column) share a common word line (WL) associated with the row of cells. The channel
28
of the cell
20
conducts current between the source
24
and the drain
22
in accordance with an electric field developed in the channel
28
by the stacked gate structure
26
.
The select gate transistor
27
has a stacked gate structure similar to the memory cell except that the first polysilicon layer
34
and the second polysilicon layer
40
are shorted together to form a single gate type structure. The select gate transistor
27
operates according to traditional MOS transistor principles, wherein SG
1
of prior art
FIG. 2
ensures the selectivity of the particular bit line, while SG
2
prevents the cells from passing current during the programming operation.
According to conventional operation, the flash memory cell
20
operates in the following manner. The cell
20
is programmed by applying a relatively high voltage V
G
(e.g., approximately 12 volts) to the control gate
38
and a moderately high voltage V
D
(e.g., approximately 9 volts) to the drain
22
in order to produce “hot” (high energy) electrons in the channel
28
near the drain
22
. The hot electrons accelerate across the tunnel oxide
32
and into the floating gate
34
and become trapped in the floating gate
34
since the floating gate
34
is surrounded by insulators (the interpoly dielectric
36
and the tunnel oxide
32
). As a result of the trapped electrons, the threshold voltage of the cell
20
increases by about 3 to 5 volts. This change in the threshold voltage (and thereby the channel conductance) of the cell
20
created by the trapped electrons is what causes the cell to be programmed.
To read the memory cell
20
, a predetermined voltage V
G
that is greater than the threshold voltage of an unprogrammed cell, but less than the threshold voltage of a programmed cell, is applied to the control gate
40
. If the cell
20
conducts, then the cell
20
has not been programmed (the cell
20
is therefore at a first logic state, e.g., a zero “0”). Likewise, if the cell
20
does not conduct, then the cell
20
has been programmed (the cell
20
is therefore at a second logic state, e.g. a one “1”). Consequently, one can read each cell
20
to determine whether it has been programmed (and therefore identify its logic state).
In order to erase the flash memory cell
20
, a relatively high voltage V
S
(e.g., approximately 12 volts) is applied to the source
24
and the control gate
40
is held at a ground potential (V
G
=0), while the drain
22
is allowed to float. Under these conditions, a strong electric field is developed across the tunnel oxide
32
between the floating gate
34
and the source region
22
. The electrons that are trapped in the floating gate
34
flow toward and cluster at the portion of the floating gate
34
overlying the source region
22
and are extracted from the floating gate
34
and into the source region
22
by way of Fowler-Nordheim tunneling through the tunnel oxide
32
. Consequently, as the electrons are removed from the floating gate
34
, the cell
20
is erased.
During the processing of the core portion
12
having a circuit configuration as illustrated in prior art
FIG. 2 and a
structure as illustrated in prior art
FIG. 3
, a problem sometimes occurs involving the formation of poly1 and poly2 contacts, which is illustrated in prior art
FIGS. 4
a
and
4
b
. A poly
1
contact is a contact made to the polysilicon gate
34
of the select gate transistor
27
(wherein poly1 is the first layer of polysilicon) and a poly2 contact is a contact made to the tungsten silicide layer
42
of the top region
38
of the respective memory cells. A conductive layer is then formed over the poly1 and poly2 contact regions to short circuit the first and second layers of polysilicon, as illustrated in FIG.
3
.
As illustrated in prior art
FIG. 4
a
, after the select gate transistor
27
is formed, an oxide based interlayer dielectric
46
is formed over the core portion
12
. The interlayer dielectric
46
isolates the various word lines and the select gate transistors
27
from overlying conductive layers which are subsequently formed and used to selectively interconnect the various components in the memory device. After the interlayer dielectric
46
is formed, a chemical mechanical poli
Fang Hao
Higashitani Masaaki
Wang John Jianshi
Advanced Micro Devices , Inc.
Eschweiler & Associates LLC
Lee Eddie
Richards N. Drew
LandOfFree
Elimination of poly cap easy poly 1 contact for NAND product does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Elimination of poly cap easy poly 1 contact for NAND product, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Elimination of poly cap easy poly 1 contact for NAND product will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2572950