U.S. patent application number 11/491328 was filed with the patent office on 2006-11-16 for ultra high density flash memory.
This patent application is currently assigned to Micron Technology, Inc.. Invention is credited to Leonard Forbes, Wendell P. Noble.
Application Number | 20060258096 11/491328 |
Document ID | / |
Family ID | 25395348 |
Filed Date | 2006-11-16 |
United States Patent
Application |
20060258096 |
Kind Code |
A1 |
Noble; Wendell P. ; et
al. |
November 16, 2006 |
Ultra high density flash memory
Abstract
An ultra high density flash EEPROM provides increased
nonvolatile storage capacity. A memory cell array includes densely
packed memory cells, each cell having a semiconductor pillar
providing shared source/drain regions for four vertical floating
gate transistors that have individual floating and control gates
distributed on the four sides of the pillar. Mutually orthogonal
first gate lines and second gate lines provide addressing of the
control gates. First source/drain terminals are row addressable by
interconnection lines disposed substantially parallel to the first
gate lines. Second source/drain terminals are column addressable by
data lines disposed substantially parallel to the second gate
lines. Both bulk semiconductor and silicon-on-insulator embodiments
are provided. If a floating gate transistor is used to store a
single bit of data, an area of only F.sup.2 is needed per bit of
data, where F is the minimum lithographic feature size. If multiple
charge states (more than two) are used, an area of less than
F.sup.2 is needed per bit of data.
Inventors: |
Noble; Wendell P.; (Milton,
VT) ; Forbes; Leonard; (Corvallis, OR) |
Correspondence
Address: |
SCHWEGMAN, LUNDBERG, WOESSNER & KLUTH, P.A.
P.O. BOX 2938
MINNEAPOLIS
MN
55402
US
|
Assignee: |
Micron Technology, Inc.
|
Family ID: |
25395348 |
Appl. No.: |
11/491328 |
Filed: |
July 21, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09866938 |
May 29, 2001 |
|
|
|
11491328 |
Jul 21, 2006 |
|
|
|
09035304 |
Feb 27, 1998 |
6238976 |
|
|
09866938 |
May 29, 2001 |
|
|
|
08889554 |
Jul 8, 1997 |
5973356 |
|
|
09035304 |
Feb 27, 1998 |
|
|
|
Current U.S.
Class: |
438/257 ;
257/E21.209; 257/E21.422; 257/E21.682; 257/E21.703; 257/E27.103;
257/E27.112; 257/E29.302; 438/259 |
Current CPC
Class: |
H01L 21/84 20130101;
H01L 29/42336 20130101; H01L 27/1203 20130101; H01L 27/115
20130101; H01L 29/66825 20130101; H01L 27/11521 20130101; H01L
27/11556 20130101; H01L 29/40114 20190801; H01L 29/7881
20130101 |
Class at
Publication: |
438/257 ;
438/259 |
International
Class: |
H01L 21/336 20060101
H01L021/336 |
Claims
1. A method of forming a memory array on a substrate, comprising:
forming a first source/drain layer at a surface of the substrate;
forming a semiconductor epitaxial layer on the first source/drain
layer; forming a second source/drain layer at a surface of the
epitaxial layer; etching, in a first direction, a plurality of
substantially parallel first troughs in the epitaxial layer;
forming a first gate dielectric layer substantially adjacent to
sidewall regions of the first troughs; forming a first conductive
layer in the first troughs; removing a portion of the first
conductive layer in the first troughs such that floating gate
regions are formed along the sidewall regions therein and separated
from the sidewall regions by the first gate dielectric layer;
etching a portion of the substrate underlying a portion of the
first troughs between the floating gate regions; forming a first
intergate dielectric layer on exposed portions of the floating gate
regions in the first troughs; forming first gate lines in the
underlying etched portion of the substrate between opposing
floating gate regions in the first troughs; and forming control
gate regions in the first troughs between opposing floating gate
regions and separated therefrom by the first intergate dielectric
layer.
2. The method of claim 1, further comprising: etching, in a second
direction that is substantially orthogonal to the first direction,
a plurality of substantially parallel second troughs in the
epitaxial layer; forming a second gate dielectric layer
substantially adjacent to sidewall regions of the second troughs;
forming a second conductive layer in the second troughs; removing a
portion of the second conductive layer in the second troughs such
that floating gate regions are formed along the sidewall regions
therein and separated from the sidewall regions by the second gate
dielectric layer; forming a second intergate dielectric layer on
exposed portions of the floating gate regions in the second
troughs; and forming control gate regions and second gate lines
between opposing floating gate regions in the second troughs by the
second intergate dielectric layer.
3. The method of claim 1, wherein the substrate further comprises a
bulk semiconductor.
4. The method of claim 1, wherein the substrate further comprises a
semiconductor on insulator portion.
5. The method of claim 1, further comprising: forming an insulating
layer undercutting semiconductor regions between the first
troughs.
6. The method of claim 2, further comprising: forming a thin
silicon nitride oxidation barrier layer by chemical vapor
deposition on the sidewall regions of the second troughs;
anisotropically etching the thin silicon nitride oxidation barrier
layer to expose bottom portions of the second troughs; forming a
bottom insulation layer on the bottom portions of the second
troughs by thermal oxidation; and stripping the thin silicon
nitride oxidation barrier layer from the sidewall regions of the
second troughs by a brief phosphoric acid etch.
7. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer, the epitaxial layer being formed on
the first source/drain layer and comprising P-silicon; forming a
thin layer of silicon dioxide on the second source/drain layer;
forming a layer of silicon nitride on the thin layer of silicon
dioxide; etching a plurality of substantially parallel troughs in
the epitaxial layer; forming at least two floating gate regions
along sidewall regions of the troughs and separated from the
sidewall regions by a gate dielectric layer; forming gate lines
between opposing floating gate regions in the troughs; and forming
control gate regions in the troughs between opposing floating gate
regions and separated therefrom by an intergate dielectric
layer.
8. The method of claim 7, wherein forming the first source/drain
layer further comprises forming the first source/drain layer with
an approximate thickness ranging between 0.2 microns and 0.5
microns.
9. The method of claim 7, wherein forming the thin layer of silicon
dioxide further comprises forming the thin layer of silicon dioxide
with an approximate thickness of 10 nanometers.
10. A method, comprising: forming a first source/drain layer at a
surface of a substrate, the first source/drain comprising N+
silicon formed by epitaxial growth of silicon upon the substrate;
forming a semiconductor epitaxial layer on the first source/drain
layer; forming a second source/drain layer at a surface of the
epitaxial layer by ion implantation, the second source/drain layer
comprising N+ silicon and having an approximate thickness of 150
nanometers; forming a thin layer of silicon dioxide on the second
source/drain layer, the thin layer of silicon dioxide having an
approximate thickness of 10 nanometers; forming a layer of silicon
nitride on the thin layer of silicon dioxide, the layer of silicon
nitride having an approximate thickness of 200 nanometers; etching
a plurality of substantially parallel troughs in the epitaxial
layer; forming at least two floating gate regions along sidewall
regions of the troughs and separated from the sidewall regions by a
gate dielectric layer; forming gate lines between opposing floating
gate regions in the troughs; and forming control gate regions in
the troughs between opposing floating gate regions and separated
therefrom by an intergate dielectric layer.
11. The method of claim 10, wherein forming the first source/drain
layer further comprises forming the first source/drain layer with
an approximate thickness ranging between 0.2 microns and 0.5
microns.
12. The method of claim 10, wherein forming at least two floating
gate regions further comprises forming the at least two floating
gate regions along the sidewall regions of the troughs and
separated from the sidewall regions by the gate dielectric layer,
the gate dielectric layer having an approximate thickness that
ranges between 5 nanometers and 10 nanometers.
13. A method, comprising: forming a first source/drain layer at a
surface of a substrate, the substrate comprised of a
semiconductor-on-insulator portion, and the a first source/drain
layer comprising N+ silicon formed by ion implantation of donor
dopants into the substrate; forming a semiconductor epitaxial layer
on the first source/drain layer, the semiconductor epitaxial layer
comprising P-silicon and having an approximate thickness of 0.6
microns; forming a second source/drain layer at a surface of the
epitaxial layer by ion implantation, the second source/drain layer
comprising N+silicon and having an approximate thickness of 150
nanometers; forming a thin layer of silicon dioxide on the second
source/drain layer; forming a layer of silicon nitride on the thin
layer of silicon dioxide; etching a plurality of substantially
parallel troughs in the epitaxial layer; forming at least two
floating gate regions along sidewall regions of the troughs and
separated from the sidewall regions by a gate dielectric layer;
forming gate lines between opposing floating gate regions in the
troughs; and forming control gate regions in the troughs between
opposing floating gate regions and separated therefrom by an
intergate dielectric layer.
14. The method of claim 13, further comprising: forming a
conductive layer in the troughs.
15. The method of claim 14, further comprising: removing a portion
of the conductive layer in the troughs; and etching a portion of
the substrate underlying a portion of the troughs between floating
gate regions.
16. The method, comprising: forming a first source/drain layer at a
surface of a substrate, the first source/drain comprising N+silicon
formed by ion implantation of donor dopants into the substrate;
forming a second source/drain layer at a surface of an epitaxial
layer, the epitaxial layer being formed on the first source/drain
layer and comprising P-silicon; forming a thin layer of silicon
dioxide on the second source/drain layer; forming a layer of
silicon nitride on the thin layer of silicon dioxide; etching a
plurality of substantially parallel troughs in the epitaxial layer;
forming an insulating layer undercutting semiconductor regions
between the troughs; forming at least two floating gate regions
along sidewall regions of the troughs and separated from the
sidewall regions by a gate dielectric layer; forming gate lines
between opposing floating gate regions in the troughs; and forming
control gate regions in the troughs between opposing floating gate
regions and separated therefrom by an intergate dielectric
layer.
17. The method of claim 16, wherein forming at least two floating
gate regions further comprises forming at least two floating gate
regions along the sidewall regions of the troughs and separated
from the sidewall regions by the gate dielectric layer, the gate
dielectric layer having an approximate thickness that ranges
between 5 nanometers and 10 nanometers.
18. The method of claim 16, wherein forming control gate regions
further comprises forming control gate regions in the troughs
between opposing floating gate regions and separated therefrom by
an intergate dielectric layer, the intergate dielectric layer
having an approximate thickness that ranges between 7 nanometers
and 15 nanometers.
19. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer, the epitaxial layer being formed on
the first source/drain layer; etching a plurality of substantially
parallel troughs in the epitaxial layer; forming a thin silicon
nitride oxidation barrier layer by chemical vapor deposition on
sidewall regions of the troughs; anisotropically etching the thin
silicon nitride oxidation barrier layer to expose bottom portions
of the troughs; forming a bottom insulation layer on the bottom
portions of the troughs by thermal oxidation; forming at least two
floating gate regions along sidewall regions of the troughs and
separated from the sidewall regions by a gate dielectric layer; and
forming control gate regions in the troughs between opposing
floating gate regions and separated therefrom by an intergate
dielectric layer.
20. The method of claim 19, further comprising: planarizing the
first conductive layer using a chemical mechanical polish.
21. The method of claim 19, further comprising: stripping the thin
silicon nitride oxidation barrier layer from the sidewall regions
by a phosphoric acid etch.
22. The method of claim 19, wherein forming a first source/drain
layer further comprises: forming the first source/drain layer at
the surface of the substrate, wherein the substrate is a bulk
semiconductor.
23. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming a first bottom insulation layer on bottom portions
of the first troughs by thermal oxidation; forming first floating
gate regions along sidewall regions of the first troughs and
separated from the sidewall regions by a first gate dielectric
layer; forming first control gate regions between opposing first
floating gate regions, the first control gate regions being
separated from the first floating gate regions by a first intergate
dielectric layer; etching, in a second direction substantially
orthogonal to the first direction, a plurality of substantially
parallel second troughs in the epitaxial layer; forming second
floating gate regions along sidewall regions of the second troughs
and separated from the sidewall regions by a second gate dielectric
layer; and forming second control gate regions in the troughs
between opposing second floating gate regions, the second control
gate regions being separated from the second floating gate regions
by a second intergate dielectric layer.
24. The method of claim 23, further comprising: forming a second
bottom insulation layer on bottom portions of the second troughs by
thermal oxidation.
25. The method of claim 23, further comprising: forming the second
dielectric layer by deposition of oxynitride using chemical vapor
deposition.
26. The method of claim 23, wherein forming first floating gate
regions further comprises: forming the first floating gate regions
along the sidewall regions of the first troughs and separated from
the sidewall regions by the first gate dielectric layer, the first
gate dielectric layer having an approximate thickness that ranges
between 5 nanometers and 10 nanometers.
27. The method of claim 23, wherein forming second control gate
regions further comprises: forming the second control gate regions
along the sidewall regions of the second troughs and separated from
the sidewall regions by the second gate dielectric layer, the
second gate dielectric layer having an approximate thickness that
ranges between 5 nanometers and 10 nanometers.
28. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming a first dielectric layer along sidewall regions of
the first troughs; forming first floating gate regions along
sidewall regions of the first troughs and separated from the
sidewall regions by the first gate dielectric layer, the first
floating gate regions including a first conductive layer of N+
doped polysilicon; forming first control gate regions between
opposing first floating gate regions, the first control gate
regions being separated from the first floating gate regions by a
first intergate dielectric layer, and the first control gate
regions including N+ doped polysilicon; forming first gate lines in
the first troughs between opposing first floating gate regions;
etching, in a second direction substantially orthogonal to the
first direction, a plurality of substantially parallel second
troughs in the epitaxial layer; removing material at intersecting
portions of first troughs and second troughs to separate the first
floating gate regions into first isolated floating gates; forming a
second gate dielectric layer along sidewall regions of the second
troughs; forming second floating gate regions along the sidewall
regions of the second troughs and separated from the sidewall
regions by a second gate dielectric layer, the second floating gate
regions including a second conductive layer of N+ doped
polysilicon; forming second control gate regions between opposing
second floating gate regions, the second control gate regions being
separated from the second floating gate regions by a second
intergate dielectric layer, and the second control gate regions
including N+ doped polysilicon; and forming second gate lines in
the second troughs between opposing second floating gate
regions.
29. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming first floating gate regions along sidewall regions
of the first troughs and separated from the sidewall regions by the
first gate dielectric layer, the first floating gate regions
including a first conductive layer of N+ doped polysilicon; forming
first control gate regions between opposing first floating gate
regions, the first control gate regions being separated from the
first floating gate regions by a first intergate dielectric layer;
etching, in a second direction substantially orthogonal to the
first direction, a plurality of substantially parallel second
troughs in the epitaxial layer; forming second floating gate
regions along sidewall regions of the second troughs and separated
from the sidewall regions by a second gate dielectric layer, the
second floating gate regions including a second conductive layer of
N+ doped polysilicon; and forming second control gate regions
between opposing second floating gate regions, the second control
gate regions being separated from the second floating gate regions
by a second intergate dielectric layer.
30. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming first floating gate regions along sidewall regions
of the first troughs and separated from the sidewall regions by the
first gate dielectric layer; forming first control gate regions
between opposing first floating gate regions, the first control
gate regions being separated from the first floating gate regions
by a first intergate dielectric layer, and the first control gate
regions including N+ doped polysilicon; etching, in a second
direction substantially orthogonal to the first direction, a
plurality of substantially parallel second troughs in the epitaxial
layer; forming second floating gate regions along sidewall regions
of the second troughs and separated from the sidewall regions by a
second gate dielectric layer; and forming second control gate
regions between opposing second floating gate regions, the second
control gate regions being separated from the second floating gate
regions by a second intergate dielectric layer, and the second
control gate regions including N+ doped polysilicon.
31. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming first floating gate regions along sidewall regions
of the first troughs and separated from the sidewall regions by a
first gate dielectric layer; forming first control gate regions
between opposing first floating gate regions, the first control
gate regions being separated from the first floating gate regions
by a first intergate dielectric layer; forming first gate lines in
the first troughs between opposing first floating gate regions;
etching, in a second direction substantially orthogonal to the
first direction, a plurality of substantially parallel second
troughs in the epitaxial layer; forming second floating gate
regions along sidewall regions of the second troughs and separated
from the sidewall regions by a second gate dielectric layer;
forming second control gate regions between opposing second
floating gate regions, the second control gate regions being
separated from the second floating gate regions by a second
intergate dielectric layer; and forming second gate lines in the
second troughs between opposing second floating gate regions.
32. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming first floating gate regions along sidewall regions
of the first troughs and separated from the sidewall regions by a
first gate dielectric layer; forming first control gate regions
between opposing first floating gate regions, the first control
gate regions being separated from the first floating gate regions
by a first intergate dielectric layer, and the first control gate
regions are formed together with first gate lines by depositing N+
polysilicon in the first troughs; etching, in a second direction
substantially orthogonal to the first direction, a plurality of
substantially parallel second troughs in the epitaxial layer;
forming second floating gate regions along sidewall regions of the
second troughs and separated from the sidewall regions by a second
gate dielectric layer; and forming second control gate regions
between opposing second floating gate regions, the second control
gate regions being separated from the second floating gate regions
by a second intergate dielectric layer, and the second control gate
regions are formed together with second gate lines by depositing N+
polysilicon in the second troughs.
33. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming first floating gate regions along sidewall regions
of the first troughs and separated from the sidewall regions by a
first gate dielectric layer; forming first control gate regions
between opposing first floating gate regions, the first control
gate regions being separated from the first floating gate regions
by a first intergate dielectric layer; etching, in a second
direction substantially orthogonal to the first direction, a
plurality of substantially parallel second troughs in the epitaxial
layer; removing material at intersecting portions of first troughs
and second troughs to separate the first floating gate regions into
first isolated floating gates; forming second floating gate regions
along sidewall regions of the second troughs and separated from the
sidewall regions by a second gate dielectric layer; and forming
second control gate regions between opposing second floating gate
regions, the second control gate regions being separated from the
second floating gate regions by a second intergate dielectric
layer.
34. A method, comprising: forming a first source/drain layer at a
surface of a substrate; forming a second source/drain layer at a
surface of an epitaxial layer; etching, in a first direction, a
plurality of substantially parallel first troughs in the epitaxial
layer; forming a first gate dielectric layer along sidewall regions
of the first troughs; forming first floating gate regions along the
sidewall regions of the first troughs and separated from the
sidewall regions by a first gate dielectric layer; forming first
control gate regions between opposing first floating gate regions,
the first control gate regions being separated from the first
floating gate regions by a first intergate dielectric layer;
etching, in a second direction substantially orthogonal to the
first direction, a plurality of substantially parallel second
troughs in the epitaxial layer; forming a second gate dielectric
layer along sidewall regions of the second troughs; forming second
floating gate regions along the sidewall regions of the second
troughs and separated from the sidewall regions by the second gate
dielectric layer; and forming second control gate regions between
opposing second floating gate regions, the second control gate
regions being separated from the second floating gate regions by a
second intergate dielectric layer.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a Divisional of U.S. application Ser.
No. 09/866,938, filed May 29, 2001, which is a Divisional of U.S.
application Ser. No. 09/035,304, filed Feb. 27, 1998, which is a
Divisional of U.S. application Ser. No. 08/889,554, filed Jul. 8,
1997, now issued as U.S. Pat. No. 5,973,356, all of which are
incorporated herein by reference.
[0002] This application is related to U.S. Pat. No. 5,936,274,
which disclosure is herein incorporated by reference.
TECHNICAL FIELD OF THE INVENTION
[0003] This invention relates generally to integrated circuits, and
particularly to floating gate transistor structures for use in
nonvolatile semiconductor memories such as in flash EEPROM memory
cells.
BACKGROUND OF THE INVENTION
[0004] Electrically erasable and programmable read only memories
(EEPROMs) are reprogrammable nonvolatile memories that are widely
used in computer systems for storing data both when power is
supplied or removed. The typical data storage element of an EEPROM
is a floating gate transistor, which is a field-effect transistor
(FET) having an electrically isolated (floating) gate that controls
electrical conduction between source and drain regions. Data is
represented by charge stored on the floating gate and the resulting
conductivity obtained between source and drain regions.
[0005] Increasing the storage capacity of EEPROM memories requires
a reduction in the size of the floating gate transistors and other
EEPROM components in order to increase the EEPROM's density.
However, memory density is typically limited by a minimum
lithographic feature size (F) that is imposed by lithographic
processes used during fabrication. For example, the present
generation of high density dynamic random access memories (DRAMS),
which are capable of storing 256 Megabits of data, require an area
of 8F.sup.2 per bit of data. There is a need in the art to provide
even higher density memories in order to further increase storage
capacity.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] In the drawings, like numerals describe substantially
similar components throughout the several views.
[0007] FIG. 1 is a schematic/block diagram illustrating generally
an architecture of one embodiment of a nonvolatile memory,
according to the teachings of the invention, including an array
having a plurality of memory cells.
[0008] FIG. 2 is a schematic diagram illustrating generally one
embodiment of an array of memory cells according to the teachings
of the invention.
[0009] FIG. 3 is a perspective view illustrating generally one
embodiment of a portion of an array of memory cells according to
the teachings of the invention.
[0010] FIG. 4 is a plan view from above of a working surface of a
substrate, which illustrates one embodiment of one of a memory cell
according to the teachings of the invention.
[0011] FIGS. 5-20 illustrate generally various stages of one
embodiment of a method of forming an array of memory cells
according to the teachings of the invention.
[0012] FIG. 21 is a perspective view of a structure dating from
another embodiment of a method of forming the array of memory cells
according to the invention, using semiconductor-on-insulator (SOI)
techniques.
DETAILED DESCRIPTION OF THE INVENTION
[0013] In the following detailed description, reference is made to
the accompanying drawings which form a part hereof. and in which is
shown by way of illustration specific embodiments in which the
invention may be practiced. These embodiments are described in
sufficient detail to enable those skilled in the art to practice
the invention, and it is to be understood that the embodiments may
be combined, or that other embodiments may be utilized and that
structural, logical and electrical changes may be made without
departing from the scope of the present invention. In the following
description, the terms wafer and substrate are interchangeably used
to refer generally to any structure on which integrated circuits
are formed, and also to such structures during various stages of
integrated circuit fabrication. Both terms include doped and
undoped semiconductors, epitaxial layers of a semiconductor on a
supporting semiconductor or insulating material, combinations of
such layers, as well as other such structures that are known in the
art, including bulk semiconductor and semiconductor-on-insulator
(SOI) substrates. In the drawings, like numerals describe
substantially similar components throughout the several views. The
following detailed description is not to be taken in a limiting
sense.
[0014] FIG. 1 is a schematic/block diagram illustrating generally
an architecture of one embodiment of a memory 100 according to the
present invention. In the embodiment of FIG. 1, memory 100 is a
nonvolatile ultra high density electrically erasable and
programmable read only memory (EEPROM) allowing simultaneous
erasure of multiple data bits, referred to as flash EEPROM.
However, the invention can be applied to other semiconductor memory
devices, such as static or dynamic random access memories (SRAMs
and DRAMS, respectively), synchronous random access memories or
other types of memories that include a matrix of selectively
addressable memory cells.
[0015] Memory 100 includes a memory cell array 105, having memory
cells therein that include floating gate transistors, as described
below. Y gate decoder 110 provides a plurality of first gate lines,
YG1, YG2, . . . , YGN for addressing floating gate transistors in
array 105, as described below. X gate decoder 115 provides a
plurality of second gate lines, XG1, XG2, . . . , XGN for
addressing floating gate transistors in array 105, as described
below. Y source/drain decoder 120 provides a plurality of first
source/drain interconnection lines YS1, YS2, . . . , YSN, for
accessing first source/drain regions of the floating gate
transistors in array 105, as described below. In an embodiment in
which commonly connected first source/drain interconnection lines
YS1, YS2, . . . , YSN are used, Y source/drain decoder 120 may be
omitted. X source/drain decoder 125 provides a plurality of data
lines, XD1, XD2, . . . , XDN for accessing second source/drain
regions of the floating gate transistors in array 105, as described
below. X source/drain decoder 125 also typically includes sense
amplifiers and input/output (I/O) circuitry for reading, writing,
and erasing data to and from array 105. In response to address
signals A0-AN that are provided on address lines 130 during read,
write, and erase operations, address buffers 135 control the
operation of Y gate decoder 110, X gate decoder 115, Y source/drain
decoder 120, and X source/drain decoder 125. The address signals
A0-AN are provided by a controller such as a microprocessor that is
fabricated separately or together with memory 100, or otherwise
provided by any other suitable circuits. As described in detail
below, the address signals A0-AN are decoded by Y gate decoder 110,
X gate decoder 115, Y source/drain decoder 120, and X source/drain
decoder 125 to perform reading, writing, and erasing operations on
memory cells that include a number of vertical floating gate
field-effect transistors (FETs) formed on the sides of a
semiconductor pillar on a substrate.
[0016] FIG. 2 is a schematic diagram illustrating generally one
embodiment of array 105 in more detail. In FIG. 2, each memory cell
205 comprises four floating gate transistors 200, e.g. four
field-effect transistors (FETS), each having an electrically
isolated (floating) gate that controls electrical conduction
between source and drain regions. The floating gate transistors 200
are arranged in cells 205, such as cells 205AA, 205BA, . . . ,
205NA, in a first direction, e.g. in the Y-direction of the first
source/drain interconnection lines YS1, YS2, . . . , YSN, and in
cells such as 205AA, 205AB . . . , 205AN in a second direction,
e.g. in the X-direction of the data lines, XD1, XD2, . . . , XDN.
In the embodiment of FIG. 2, each cell 205 includes four floating
gate transistors 200 that share a common first source/drain region,
such as a source region coupled to one of the first source/drain
interconnection lines YS1, YS2, . . . , YSN. The floating gate
transistors 200 of each cell 205 also share a common second
source/drain region, such as a drain region coupled to one of the
data lines, XD1, XD2, . . . , XDN. Each cell 205 has first and
second source/drain regions that are fabricated using a common
semiconductor pillar on a substrate, as explained below.
[0017] FIG. 3 is a perspective view illustrating generally one
embodiment of a portion of array 105, including portions of two
cells 205 of floating gate transistors 200, such as illustrated in
FIG. 2. In FIG. 3, the substantially identical cells 205 are
illustrated by way of example through cells 205AA and 205BA. Cells
205AA and 205BA each include a semiconductor pillar 300, initially
of a first conductivity type such as P-silicon, fabricated upon a
monolithic substrate 305. In one embodiment, substrate 305 is a
bulk semiconductor, such as P-silicon. In another embodiment, a
semiconductor-on-insulator (SOI) substrate 305 includes an
insulating layer, such as silicon dioxide (SiO.sub.2), as described
below.
[0018] Each pillar 300 includes a first source/drain region of a
second conductivity type, such as N+ silicon source region 310,
formed proximally to a sub-micron dimensioned interface between
pillar 300 and substrate 305. Each pillar 300 also includes a
second source/drain region of the second conductivity type, such as
N+ silicon drain region 315, that is distal to substrate 305, and
separated from source region 310 by a first conductivity type
region, such as P-body region 320.
[0019] Each pillar 300 provides a source region 310, a drain region
315, and a body region 320 for the four floating gate transistors
200 of a particular memory cell 205. In one embodiment, the
physical dimensions of each pillar 300 and the doping of P-body
region 320 are both sufficiently small to allow operation of the
floating gate transistors 200 that is characteristic of fully
depleted body transistors. First source/drain region
interconnection line YS1 electrically interconnects the source
region 310 of each pillar 300. of cells 205AA, 205BA, . . . ,
205BN. In one embodiment, the first source/drain interconnection
lines YS1, YS2, . . . , YSN, comprise a conductively doped
semiconductor of the second conductivity type, such as N+ silicon,
disposed at least partially within substrate 305. For example,
dopants can be ion-implanted or diffused into substrate 305 to form
the first source/drain interconnection lines YS1, YS2, . . . , YSN.
In another embodiment, the first source/drain interconnection lines
YS1, YS2, . . . , YSN are formed above substrate 305. For example,
a doped epitaxial semiconductor layer can be grown on substrate
305, from which first source/drain interconnection lines YS1, YS2,
. . . , YSN are formed. Alternatively, an undoped epitaxial
semiconductor layer can be grown on substrate 305, and dopants then
introduced by ion-implantation or diffusion to obtain the first
source/drain interconnection lines YS1, YS2, . . . , YSN of the
desired conductivity.
[0020] Each pillar 300 is outwardly formed from substrate 305, and
is illustrated in FIG. 3 as extending vertically upward from
substrate 305. Each pillar 300 has a top region that is separated
from substrate 305 by four surrounding side regions. A floating
gate 325 is formed substantially adjacent to each side surface of
pillar 300, and separated therefrom by a gate dielectric 330, such
that there are four floating gates 325 per pillar 300, though FIG.
3 omits some of the floating gates 325 for clarity of illustration.
Each floating gate 325 has a corresponding substantially adjacent
control gate 335, from which it is separated by an intergate
dielectric 340. Except at the periphery of array 105, each control
gate 335 is interposed between two approximately adjacent pillars
300 and shared by two floating gate transistors 200, each of these
floating gate transistors 200 having portions in one of the two
approximately adjacent pillars 300.
[0021] Also interposed between approximately adjacent pillars 300,
except at the periphery of array 105, are first gate line YG1, YG2,
. . . , YGN that are substantially parallel to each other in the
first direction, e.g. the Y-direction. Each of the first gate lines
YG1, YG2, . . . , YGN interconnects ones of the control gates 335.
For example, first gate line YG1 electrically interconnects control
gates 335 of floating gate transistors 200 in cells 205AA, 205BA, .
. . , 205BN. In the embodiment of FIG. 3, the first gate lines YG1,
YG2, . . . , YGN are disposed at least partially within substrate
305, as described below.
[0022] Also interposed between approximately adjacent pillars 300,
except at the periphery of array 105, are second gate lines XG1,
XG2, . . . , XGN that are substantially parallel to each other in
the second direction, e.g. the X-direction. Each of the second gate
lines XG1, XG2, . . . , XGN interconnects ones of the control gates
335. For example, second gate line XG2 electrically interconnects
control gates 335 of floating gate transistors 200, in which the
control gates are shared between pairs of cells 205, e.g. 205AA and
205BA, 205AB and 205BB . . . , 205AN and 205BN. In the embodiment
of FIG. 3, the second gate lines XG1, XG2, . . . , XGN are disposed
above substrate 305, as described below.
[0023] Drain regions 315 of the pillars 300 are interconnected by
data lines XD1, XD2, . . . , XDN that are substantially parallel to
each other in the second direction, e.g. the X-direction. FIG. 3
illustrates, by way of example, data lines XD1 and XD2, which are
shown schematically for clarity. However, it is understood that
data lines XD1, XD2, . . . , XDN comprise metal or other
interconnection lines that are isolated from the underlying
topology, e.g. pillars 300, floating gates 325, control gates 335,
first gate lines YG1, YG2, . . . , YGN, and second gate lines XG1,
XG2, . . . , XGN, by an insulating layer through which contact
holes are etched to access the drain regions 315 of the pillars
300.
[0024] FIG. 4 is a plan view, looking toward the working surface of
substrate 305, illustrating generally by way of example one
embodiment of one of cells 205 of four floating gate transistors
200, such as cell 205BB. In FIG. 4, each of the four floating gates
325 is adjacent to one side of pillar 300, and separated therefrom
by gate dielectric 330. Each control gate 335 is separated from a
corresponding floating gate 325 by an intergate dielectric 340, and
is integrally formed together with one of the first gate lines YG1,
YG2, . . . , YGN or second gate lines XG1, XG2, . . . , XGN. The
control gates 335 that are integrally formed together with ones of
the first gate lines YG1, YG2, . . . , YGN protrude upwardly
therefrom such that an overlap capacitance is created with floating
gates 325 that are disposed on either side thereof.
[0025] The center-to-center spacing ("pitch") between adjacent
first gate lines YG1, YG2, . . . , YGN, such as between YG2 and
YG3, or between adjacent second gate lines XG1, XG2, . . . , XGN,
such as between XG2 and XG3, is twice the minimum lithographic
feature size F. Since four floating gate transistors 200 are
contained within a cell 205 having an area of 4F.sup.2, an area of
only F.sup.2 is needed per bit of data. In another embodiment,
multiple charge states (more than two) are used to obtain
correspondingly higher data storage densities, such that an area of
less than F.sup.2 is needed per bit of data, since more than one
bit of data can be stored on a single floating gate transistor 200.
In one embodiment, four charge states are used to store two bits of
data per floating gate transistor 200, corresponding to eight bits
of data per memory cell 205. One example of using more than two
charge states to store more than one bit of data per transistor is
set forth an article by T.-S. Jung et al., entitled "A 117-mm.sup.2
3.3-V Only 128-Mb Multilevel NAND Flash Memory For Mass Storage
Applications," IEEE J. Solid-State Circuits, Vol. 31, No. 11,
November 1996. In a further embodiment, a continuum of charge
states is used to store analog data in array 105.
[0026] In one embodiment, programming of one of the floating gate
transistors 200 is by hot electron injection. For example, a
voltage of approximately 10 volts is provided, such as by one of Y
gate decoder 110 or X gate decoder 115, through a particular one of
the first gate lines YG1, YG2, . . . , YGN or second gate lines
XG1, XG2, . . . , XGN to a particular control gate 335. A resulting
inversion region (channel) is formed in the body region 320 at the
surface that is approximately adjacent to the particular one of the
first gate lines YG1, YG2, . . . , YGN or second gate lines XG1,
XG2, . . . , XGN. A voltage of approximately 5 Volts is provided,
such as by X source/drain decoder 125, through a particular one of
data lines XD1, XD2, . . . , XDN to a particular drain region 315.
A voltage of approximately 0 Volts is provided, such as by Y
source/drain decoder 120, through a particular one of first
source/drain interconnection lines YS1, YS2, . . . , YSN, to the
particular source region 310 of the floating gate transistor 200.
Electrons are injected onto the floating gate 325 interposed
between the control gate 335 and the pillar 300 in which the
particular drain region 315 is disposed. The exact value of the
voltages provided to the particular control gate 335 and drain
region 315 will depend on the physical dimension of the floating
gate transistor 200, including the thickness of the gate dielectric
330, the thickness of the intergate dielectric 340, and the
separation between source region 310 and drain region 315.
Alternatively, if higher voltages are provided to control gate 335,
and the gate dielectric 330 and intergate dielectric 340 are made
thinner, the floating gate transistor 200 may be programmed instead
by Fowler-Nordheim tunneling of electrons h m the body region 320,
source region 310, or drain region 315.
[0027] Addressing a particular memory cell 205 for reading data
includes selecting a particular one of data lines XD1, XD2, . . . ,
XDN and also selecting a particular one of first source/drain
interconnection lines YS1, YS2, . . . , YSN. Addressing a
particular floating gate transistor 200 within the particular
memory cell 205 for reading data further includes selecting a
particular one of first gate lines YG1, YG2, . . . , YGN or second
gate lines XG1, XG2, . . . , XGN.
[0028] In one embodiment, reading data stored on a particular
floating gate transistor 200 includes providing a voltage of
approximately 5 volts, such as by one of Y gate decoder 110 or X
gate decoder 115, through a particular one of the first gate lines
YG1, YG2, . . . , YGN or second gate lines XG1, XG2, . . . , XGN to
the particular control gate 335 of the floating gate transistor
200. A voltage of approximately 0 Volts is provided, such as by Y
source/drain decoder 120, through a particular one of first
source/drain interconnection lines YS1, YS2, . . . , YSN, to the
particular source region 310 of the particular floating gate
transistor 200. A particular one of data lines XD1, D2 . . . , XDN
that is switchably coupled to the drab region 315 of the floating
gate transistor 200 is precharged to a positive voltage by a sense
amplifier in X source/drain decoder 125, then coupled to the drain
region 315 to determine the conductivity state of the floating gate
transistor 200 between its source region 310 and drain region
315.
[0029] If there are no electrons stored on the floating gate 325,
the floating gate transistor 200 will conduct between its source
region 310 and drain region 315, decreasing the voltage of the
particular one of data lines XD1, XD2, . . . , XDN toward that
voltage of its source region 310, e.g. toward a "low" binary logic
level of approximately 0 Volts. If there are electrons stored on
the floating gate 325, the floating gate transistor 200 will not
conduct between its source region 310 and drain region 315. As a
result, the sense amplifier will tend to increase the voltage of
the particular one of data lines XD1, XD2, . . . , XDN toward a
positive voltage, e.g. toward a "high" binary logic voltage
level.
[0030] In one embodiment, erasure of floating gate transistors 200
includes providing an erasure voltage difference of approximately
between -10 and -12 Volts from a source region 310 to a
corresponding control gate 335. For example, a voltage of
approximately 0 Volts is provided, such as by Y source/drain
decoder 120, to source regions 310 of floating gate transistors 200
that are interconnected by one or several first source/drain
interconnection lines YS1, YS2, . . . , YSN. A voltage of
approximately between -10 and -12 Volts is provided, such as by one
of Y gate decoder 110 or X gate decoder 115, through a
corresponding one or several of the first gate lines YG1, YG2, . .
. , YGN or second gate lines XG1, XG2, . . . , XGN to the control
gates 335 of the floating gate transistors 200 to be erased. As a
result of the negative voltage applied to the control gates 335,
electrons are removed from the corresponding floating gates 325 by
Fowler-Nordheim tunneling, thereby erasing the data from ones of
the floating gate transistors 200. In another example, a voltage of
approximately between -5 and -6 Volts is applied to the control
gates 335 and a voltage of approximately between +5 and +6 Volts is
applied to the source regions 310 in order to obtain the erasure
voltage difference of approximately between -10 and -12 Volts from
a source region 310 to a corresponding control gate 335. The exact
value of the erasure voltage difference will vary depending upon
the physical dimensions of the floating gate transistor 200 and the
thicknesses of gate dielectric 330 and intergate dielectric
340.
[0031] In one embodiment, the entire array 105 of floating gate
transistors 200 is simultaneously erased by applying approximately
between -10 and -12 Volts to each of first gate lines YG1, YG2, . .
. , YGN and second gate lines XG1, XG2, . . . , XGN, and also
applying 0 Volts to each of first source/drain interconnection
lines YS1, YS2, . . . , YSN. In another embodiment, one or more
sectors of array 105 are simultaneously erased by selectively
applying approximately between -10 and -12 Volts to one or more of
first gate lines YG1, YG2, . . . , YGN or second gate lines XG1,
XG2, . . . , XGN, and also applying 0 Volts to one or more of first
source/drain interconnection lines YS1, YS2, . . . , YSN.
[0032] FIGS. 5-20 illustrate generally one embodiment of a method
of forming memory array 105. In this embodiment, the array 105 is
formed using bulk silicon processing techniques and is described,
by way of example, with respect to a particular technology having a
minimum feature size F, which is also sometimes referred to as a
critical dimension (CD), of 0.4 microns. However, the process steps
described below can be scaled accordingly for other minimum feature
sizes without departing from the scope of the invention.
[0033] In FIG. 5, a P-silicon starting material is used for
substrate 305. A first source/drain layer 500, of approximate
thickness between 0.2 microns and 0.5 microns, is formed at a
working surface of substrate 305. In one embodiment, first
source/drain layer 500 is N+ silicon formed by ion-implantation of
donor dopants into substrate 305. In another embodiment, first
source/drain layer 500 is N+ silicon formed by epitaxial growth of
silicon upon substrate 305. On the first source/drain layer 500, a
semiconductor epitaxial layer 505, such as P-silicon of 0.6 micron
approximate thickness, is formed, such as by epitaxial growth. A
second source/drain layer 510, such as N+ silicon of 150 nanometer
approximate thickness, is formed at a surface of the epitaxial
layer 505, such as by ion-implantation of donor dopants into
P-epitaxial layer 505 or by epitaxial growth of N+ silicon on
P-epitaxial layer 505. A thin layer of silicon dioxide (SiO.sub.2),
referred to as pad oxide 515, is deposited on the second
source/drain layer 510. Pad oxide 515 has a thickness of
approximately 10 nanometers. A layer of silicon nitride
(Si.sub.3N.sub.4), referred to as pad nitride 520, is deposited on
the pad oxide 515. Pad nitride 520 has a thickness of approximately
200 nanometers.
[0034] In FIG. 6, photoresist masking and selective etching
techniques are used to form, in the first direction (e.g., the Y
direction, which is perpendicular to the plane of the drawing of
FIG. 6), a plurality of substantially parallel first troughs 600
that extend through the pad nitride 520, pad oxide 515, second
source/drain layer 510, the underlying portion of epitaxial layer
505, and at least partially into first source/drain layer 500. The
photoresist is then removed.
[0035] In FIG. 7, a thin silicon nitride oxidation barrier layer
700 is deposited by chemical vapor deposition (CVD) to protect
against oxidation of sidewalls of first troughs 600. Barrier layer
700 is anisotropically etched to expose bottom portions of first
troughs 600. A bottom insulation layer 705 of silicon dioxide is
formed on the bottoms of first troughs 600 by thermal oxidation of
the exposed bottom portions of first troughs 600.
[0036] In FIG. 8, barrier layer 700 is stripped from the sidewalls
of the first troughs 600, such as by a brief phosphoric acid etch,
which is timed to expose the sidewalls of the first troughs 600 but
which avoids significant removal of the pad nitride 520. A first
gate dielectric layer 800 such as, for example, silicon dioxide of
thickness approximately between 5 nanometers and 10 nanometers
(sometimes referred to as "tunnel oxide"), is formed substantially
adjacent to the exposed sidewalls of the first troughs 600. A first
conductive layer 805, such as N+ doped polysilicon, is formed in
the first troughs 600, such as by CVD, to fill the first troughs
600. The first conductive layer 805 is planarized, such as by
chemical mechanical polishing (CMP) or other suitable planarization
technique.
[0037] In FIG. 9, the first conductive layer 805 is etched back in
the first troughs 600 to approximately 100 nanometers below the
silicon surface, which is defined by the interface between the
second source/drain layer 510 and the pad oxide 515 layer. A first
spacer layer, such as silicon nitride of an approximate thickness
of 7 nanometers, is deposited by CVD and anisotropically etched by
reactive ion etching (RIE) to leave nitride first spacers 900 along
the sidewalls of the first troughs 600. A second spacer layer, such
as silicon dioxide of an approximate thickness of 90 nanometers, is
deposited by CVD and anisotropically etched by RIE to leave second
spacers 905 along the sidewalls of the first troughs 600.
[0038] In FIG. 10, a portion of the first conductive layer 805 in
first troughs 600 between second spacers 905 is removed, such as by
using spacers 905 as a mask while etching down to bottom insulation
layer 705, thereby forming from the first conductive layer 805
floating gate regions 1000 along the sidewalls of the first troughs
600. A thin oxidation barrier layer 1005, such as silicon nitride
of approximate thickness of 5 nanometers, is deposited by CVD.
Barrier layer 1005 is removed from the bottom insulation layer 705
in first troughs 600 by anisotropic etching. The remaining portions
of barrier layer 1005 protect the floating gate regions 1000 during
subsequent processing described below.
[0039] In FIG. 11, a portion of the bottom insulation layer 705 is
removed, exposing a portion of the underlying substrate 305, by an
anisotropic etch that is timed to leave enough of second spacers
905 to protect floating gate regions 1000 during a subsequent etch
of substrate 305. A portion of substrate 305 that underlies a
portion of first troughs 600 between the floating gate regions 1000
is removed by selectively anisotropically etching the substrate 305
to a depth sufficient to carry the first gate lines YG1, YG2, . . .
, YGN. A first trough insulation layer 1100 is formed on sidewall
and bottom regions of the etched portions of substrate 305
underlying the first troughs 600. Barrier layer 1005 is removed to
expose the floating gate regions 1000 in first troughs 600, such as
by wet etching.
[0040] The first intergate dielectric 340, having an approximate
thickness between 7 nanometers and 15 nanometers, is formed on the
exposed portions of floating gate regions 1000. In one embodiment,
a silicon dioxide intergate dielectric 340 is formed by thermal
oxidation of the floating gate regions 1000. In another embodiment,
an oxynitride intergate dielectric 340 is formed on the floating
gate regions 1000 by CVD.
[0041] First gate lines YG1, YG2, . . . , YGN are formed in the
etched portions of substrate 305 underlying the first troughs 600
between opposing floating gate regions 1000 in the first troughs
600. First gate lines YG1, YG2, . . . , YGN are insulated from
substrate 305 by first trough insulation layer 1100. Control gates
335 are formed in the first troughs 600 between opposing floating
gate regions 1000, and separated therefrom by the first intergate
dielectric 340. In one embodiment, first gate lines YG1, YG2, . . .
, YGN and control gates 335 are formed together by depositing N+
polysilicon to fill first troughs 600, and etching back the
deposited N+ polysilicon approximately to the top portion of the
floating gate regions 1000.
[0042] In FIG. 12, a cap layer 1200 is formed, such as by CVD of
silicon dioxide, and then planarized, such as by CMP, such that the
top surface of cap layer 1200 is substantially even with the top
surface of pad nitride 520. A masking layer 1205 is formed, such as
silicon nitride deposited by CVD to an approximate thickness of 100
nanometers. Another masking layer 1210 is also formed, such as
polysilicon deposited by CVD to an approximate thickness of 100
nanometers. A photoresist layer 1215 is formed on masking layer
1210.
[0043] FIG. 13 is a perspective view, illustrating the selective
etching, in a second direction (X-direction) that is substantially
orthogonal to the first direction (Y-direction), of a plurality of
substantially parallel second troughs 1300, as described below.
Forming second troughs 1300 includes selectively etching masking
layer 1210 and underlying masking layer 1205, such that portions of
cap layer 1200 in the second troughs 1300 are exposed. With
photoresist layer 1215 still in place, a nonselective dry etch is
used to simultaneously remove exposed silicon dioxide and
polysilicon in intersecting portions of first troughs 600 and
second troughs 1300, including the removing of: portions of cap
layer 1200, gate dielectric 800, floating gate regions 1000,
intergate dielectric 340, and the control gate 335 portions of
first gate lines YG1, YG2, . . . , YGN. The nonselective dry etch
removal proceeds at least to the depth of the interface between
floating gate regions 1000 and underlying bottom insulation layer
705, thereby separating floating gate regions 1000 into the
isolated floating gates 325. During the nonselective dry etch, the
regions between first troughs 600 are protected by the pad nitride
520 and the regions between second troughs 1300 are protected by
selectively patterned photoresist layer 1215.
[0044] In the plan view of FIG. 14, the photoresist layer 1215 has
been removed by conventional photoresist stripping techniques,
thereby exposing the underlying. selectively patterned polysilicon
masking layer 1210. An insulating layer 1400, such as silicon
dioxide deposited by CVD, is formed everywhere on the topography of
the working surface of substrate 305, thereby filling the
nonselectively dry-etched intersections of the first troughs 600
and second troughs 1300. The insulating layer 1400 is then
planarized, such as by CMP, and recess etched to a depth that is
slightly above the interface between second source-drain layer 510
and pad oxide 515, thereby leaving behind recessed portions of
insulating layer 1400 in the nonselectively dry-etched
intersections of the first troughs 600 and second troughs 1300, as
illustrated in FIG. 14.
[0045] In the plan view of FIG. 15, the exposed portions of pad
nitride 520 (e.g., between first troughs 600 and within second
troughs 1300) are removed by a selective etch of silicon nitride,
thereby exposing underlying portions of pad oxide 515. The exposed
portions of pad oxide 515 (e.g., between first troughs 600 and
within second troughs 1300) are removed by dipping into a wet
etchant, which is timed to remove the exposed portions of pad oxide
515, but to leave most of the remaining portions of the thicker
silicon dioxide insulating layer 1400 intact. The removing of
portions of pad oxide 515 exposes the second source/drain layer 510
portion of the underlying silicon epitaxial layer 505. The exposed
portions of silicon epitaxial layer 505 (e.g., between first
troughs 600 and within second troughs 1300) are removed by a
selective etching that is preferential to silicon over silicon
dioxide, thereby forming recesses 1500 in second troughs 1300
between first troughs 600. Recesses 1500, which are considered to
be part of second troughs 1300, are etched through epitaxial layer
505 and at least partially into first source/drain layer 500.
Etching recesses 1500 also removes the remaining portions of
polysilicon masking layer 1210, thereby exposing underlying silicon
nitride masking layer 1205, as illustrated in FIG. 15.
[0046] FIG. 16 is a cross-sectional view in the direction of second
troughs 1300 (e.g. such that the X-direction is orthogonal to the
plane of the illustration of FIG. 16). as indicated by the cut line
16-16 in FIG. 15. In FIG. 16, a thin silicon nitride oxidation
barrier layer 1600 is deposited by CVD to protect against oxidation
of sidewalls of second troughs 1300. Barrier layer 1600 is
anisotropically etched to expose bottom portions of second troughs
1300. A bottom insulation layer 1605 of silicon dioxide is formed
on the bottoms of second troughs 1300, such as silicon dioxide of
approximate thickness of 50 nanometers formed by thermal oxidation
of the exposed bottom portions of second troughs 1300.
[0047] In FIG. 17, barrier layer 1600 is stripped from the
sidewalls of the second troughs 1300, such as by a brief phosphoric
acid etch, which is timed to expose the sidewalls of the second
troughs 1300 but which avoids significant removal of the silicon
nitride masking layer 1205. A second gate dielectric layer 1700,
such as silicon dioxide of thickness approximately between 5
nanometers and 10 nanometers (sometimes referred to as "tunnel
oxide"), is formed substantially adjacent to the exposed sidewalls
of the second troughs 1300. A second conductive layer 1705, such as
N+ doped polysilicon, is formed in the second troughs 1300, such by
CVP, to fill the second troughs 1300. The second conductive layer
1705 is planarized such as by chemical mechanical polishing (CMP)
or other suitable planarization technique.
[0048] In FIG. 18, the second conductive layer 1705 is etched back
in the second troughs 1300 to approximately at or slightly above
the level of the silicon surface, which is defined by the interface
between the second source/drain layer 510 and the pad oxide 515
layer. Thus, in the second troughs 1300, the top surface of the
second conductive layer 1705 is approximately even with the top
surface of the recessed portions of insulating layer 1400. A spacer
layer, such as silicon nitride of an approximate thickness of 100
nanometers, is deposited by CVD and anisotropically etched by
reactive ion etching (RTE) to leave nitride third spacers 1800,
along the sidewalls of the second troughs 1300, e.g. on the etched
back portions of the second conductive layer 1705 and on the
recessed portions of insulating layer 1400, and against the second
gate dielectric 1700.
[0049] In the perspective view of FIG. 19; third spacers 1800 are
used as a mask for the anisotropic etching of the etched back
portions of polysilicon second conductive layer 1705 together with
the recessed portions of silicon dioxide insulating layer 1400. By
first utilizing an etchant to remove silicon dioxide, the second
troughs 1300 are etched in insulating layer 1400 to a depth
sufficient to carry a second gate line X1, X2, . . . , XN, but not
so great as to expose the first gate lines Y1, Y2, . . . , YN
underlying the recessed portions of silicon dioxide insulating
layer 1400 in second troughs 1300. Then, the anisotropic etch is
continued using a selective etchant to remove polysilicon but not
silicon dioxide until the bottom insulation layer 1605 is exposed,
thereby forming from the second conductive layer 1705 separate
floating gates 325 along the sidewalls of the second troughs
1300.
[0050] In the perspective view of FIG. 20, a second intergate
dielectric 2000 is formed in the second troughs 1300, such that the
second intergate dielectric 2000 has an approximate thickness
between 7 nanometers and 15 nanometers and being formed by thermal
growth of silicon dioxide or deposition of oxynitride by CVD.
Control gates 335 are formed between opposing floating gates 325 in
the second troughs 1300 and separated therefrom by the second
intergate dielectric 2000. The control gates 335 in second troughs
1300 are formed together with the second gate lines X1, X2, . . . ,
XN in second troughs 1300 by a single deposition of N+ doped
polysilicon that fills second troughs 1300 and is planarized, such
as by CMP. Phosphoric acid is used to remove the remaining silicon
nitride, such as third spacers 1800, masking layer 1205, and pad
nitride 520, leaving the structure illustrated in FIG. 20. An
insulator such as silicon dioxide is then deposited, and subsequent
processing follows conventional techniques for forming contact
holes, terminal metal, and inter level insulator steps to complete
wiring of the cells 205 and other circuits of memory 100.
[0051] Though FIGS. 5-20 illustrate generally one embodiment of
forming the memory array 105 using bulk silicon processing
techniques, in another embodiment a semiconductor-on-insulator
(SOI) substrate is formed from substrate 305. In one such
embodiment, a P-silicon starting material is used for substrate
305, and processing proceeds similarly to the bulk semiconductor
embodiment described in FIG. 5-7. However, after the barrier layer
700 is formed in FIG. 7, an isotropic chemical etch is used to
fully undercut the semiconductor regions separating the first
troughs 600, and a subsequent oxidation step is used to fill in the
evacuated regions formed by the undercutting. As a result, an
insulator is formed on the bottoms of first troughs 600, bars of
SOI are formed between first troughs 600, and the topography on the
working surface of substrate 305 is separated from substrate 305 by
an insulating layer 2100 illustrated in the perspective view of
FIG. 21.
[0052] Thus, in the above described Figures, substrate 305 is
understood to include bulk semiconductor as well as SOI embodiments
in which the semiconductor integrated circuits formed on the
surface of substrate 305 are isolated from each other and an
underlying semiconductor portion of substrate 305 by an insulating
layer.
[0053] One such method of forming bars of SOI is described in the
Noble U.S. patent application Ser. No. 08/745,708 which is assigned
to the assignee of the present application and which is herein
incorporated by reference. Another such method of forming regions
of SOI is described in the Forbes U.S. patent application Ser. No.
08/706,230, which is assigned to the assignee of the present
application and which is herein incorporated by reference.
[0054] In an SOI embodiment of the present invention, processing of
first troughs 600 to carry the first gate lines YG1, YG2, . . . ,
YGN varies slightly from the bulk semiconductor embodiment
described with respect to FIGS. 10 and 11. A barrier layer 1005
need not be formed to protect the floating gate regions 1000. A
portion of the substrate 305 that underlies a portion of the first
troughs 600 between the floating gate regions 1000 is removed by
selectively anisotropically etching the silicon dioxide insulator
portion of substrate 305 to a depth sufficient to carry the first
gate lines YG1, YG2, . . . , YGN. A portion of the resulting
structure of array 105 is illustrated in the perspective view of
FIG. 21, which includes an insulating layer 2100 portion of
substrate 305, as described above.
[0055] Thus, the present invention provides an ultra high density
flash EEPROM having increased nonvolatile storage capacity. If a
floating gate transistor 200 is used to store a single bit of data,
an area of only F.sup.2 is needed per bit of data. If multiple
charge states (more than two) are used, an area of less than
F.sup.2 is needed per bit of data. The increased storage capacity
of the ultra high density flash EEPROM is particularly advantageous
in replacing hard disk drive data storage in computer systems. In
such an application, the delicate mechanical components included in
the hard disk drive are replaced by rugged, small, and durable
solid-state ultra high density flash EEPROM packages. The ultra
high density flash EEPROMs provide improved performance, extended
rewrite cycles, increased reliability, lower power consumption, and
improved portability.
[0056] It is to be understood that the above description is
intended to be illustrative, and not restrictive. Many other
embodiments will be apparent to those of skill in the art upon
reviewing the above description. For example, though the memory
cells 205 have been described with respect to a particular
embodiment having four floating gate transistors 200 per pillar
300, a different number of floating gate transistors per pillar
could also be used. It is also understood that the above structures
and methods, which have been described with respect to EEPROM
memory devices having floating gate transistors 200, are also
applicable to dynamic random access memories (DRAMS) or other
integrated circuits using vertically oriented field-effect
transistors (s) that do not have floating gates. Thus, the scope of
the invention is not limited to the particular embodiments shown
and described herein.
* * * * *