Bit line pad and borderless contact on bit line stud with...

Details Bit line pad and borderless contact on bit line stud with... Bit line pad and borderless contact on bit line stud with...

2000-10-30

2004-09-07

Weiss, Howard (Department: 2814)

Active solid-state devices (e.g., transistors, solid-state diode

Combined with electrical contact or lead

Of specified material other than unalloyed aluminum

C257S774000

Reexamination Certificate

active

06787906

ABSTRACT:

BACKGROUND OF THE INVENTION
In order to make more efficient use of integrated circuit surface area, conventional two-dimensional semiconductor technology has evolved such that contemporary circuits are formed on multiple layers in a three-dimensional configuration. In such configurations, active devices and interconnects are formed in a layered relationship. During the formation of each subsequent layer, an inter-layer pathway, referred to in the art as a “plug”, or “stud” is electrically coupled between the various active devices and transmission lines of the different layers. To assist in aligning a plug, “landing pads” or “taps” are formed in lower layers to serve as a target for the plugs passing from the upper layer. The landing pads are coupled to an underlying circuit or interconnect and are generally larger in surface area than the circuit or interconnect to serve as a wider-tolerance target for the plug.
Such multi-layered technology has enabled the design of highly-integrated memory devices, for example DRAM devices, having extremely high capacity, for example above 1 gigabyte. Such DRAM devices include multiple arrays of memory cells, densely and efficiently laid out under tight design constraints. Between the cell areas are peripheral regions, which include supporting circuitry and interconnect circuitry between the cells, as well as input/peripherals, and the like.
Any misalignment between the vertical plugs and the horizontal interconnect features can cause defects and reliability problems. To ensure that the plug aligns with a feature, the features are made larger than required, for example through the use of landing pads. The area by which the feature is made larger is referred to in the art as a “border” around the vertical contact hole. Any excessive border area thus has a negative impact on circuit density.
Attempts have been made in the past to provide multiple layer interconnect, while reducing or eliminating the border area. These include circuits and fabrication procedures disclosed in U.S. Pat. Nos. 6,083,824, 5,612,254, and 4,966,870.
To a larger extent, the packing density of circuits is limited by how closely the interconnect metal between circuits can be formed without encroaching on each other. These limits are dictated by design rules that govern the separation of one level of contact from another, and by design rules for nesting tolerance or for borders used around contacts.
Other efforts have made toward reducing the high-aspect ratio of the holes made for inter-level interconnects, where the aspect ratio refers the height of a hole as compared to its width. In general, the deeper the hole, the more difficult it is to fabricate the hole. Using the line of an underlying circuit, for example a bit line of a DRAM memory device, as a landing pad, the aspect ratio of the interconnection hole can be significantly reduced.
A typical multiple-layer DRAM memory device includes a cell region and a peripheral region. The cell region includes active switching devices, coupled to vertically-oriented capacitors, that serve as data storage devices. A cell bit line serves as an interconnect to transfer data between peripheral circuit regions and the cell region. The peripheral region includes a number of bit lines, that function as local interconnects, or studs, electrically coupling between the various active devices and transmission lines of the different layers. An insulative oxide layer is formed above the bit lines, and an interconnect stud is opened through the oxide layer and connected to the bit line.
When the bit lines are used for local interconnection, for example especially in sense amplifier regions, the layers of circuits can become very dense and crowded. For example, to access the bit line from an upper layer, the region between the bit lines must be accurately etched to form a stud interconnect hole; both in a lateral direction, so as to avoid contact with adjacent bit lines, and in a vertical direction, so as to ensure that the hole is formed at the proper depth. Because the peripheral region, for example a sense amplifier region of a DRAM device, is often times densely populated with various interconnect paths, the cross-sectional area of any vertical stud interconnects should be minimized. Therefore, the above case requires subsequent formation of high-aspect-ratio studs that are difficult to achieve using contemporary fabrication processes.
Contemporary techniques of forming the interconnect stud are subject to several process limitations. These include horizontal misalignment, in a lateral direction, where the stud hole may be laterally misaligned with the underlying bit line during formation of the stud hole. Vertical misalignment can also occur, wherein the stud hole is not etched deep enough, so the stud does not make contact with the underlying bit line, or wherein the stud hole is etched too deeply, and is etched through the bit line.
To improve alignment accuracy, U.S. Pat. No. 5,895,239 discloses a technique for employing a bit line landing pad together with a bit line stud. However, this approach requires tight tolerances at either, or both, the top portions of the bit lines, including the landing pad, and bottom portions of the upper interconnect stud, so as to provide a minimal width at the top of the bit line, and a maximum width at the bottom of the upper interconnect stud. A wide stud top limits circuit density considerations, while a narrow stud bottom leads to increased contact resistance and an increased aspect ratio that is difficult to accurately fabricate. No provision is made for vertical alignment of the stud, so if the stud hole is slightly misaligned with the underlying bit line, a void can be formed in the underlying inter-layer dielectric adjacent the stud.
Another approach at multiple-layered interconnect is disclosed in U.S. Pat. No. 5,891,799. With reference to
FIG. 1
, in this approach, an etch-stop layer
206
, for example a silicon nitride masking layer (Si
3
N
4
) is formed over an inter-layer dielectric (SiO
2
)
202
, in turn formed over a metal layer
210
formed on a substrate
200
. Stud holes
213
A,
213
B are patterned through the mask layer
206
and underlying dielectric layer
202
for the deposit of studs
212
A,
212
B to connect between the upper and lower layers. Once the studs
212
A,
212
B are formed, the masking layer
206
later serves as an etching reference for studs
214
A,
214
B formed through an upper masking layer
208
and upper dielectric layer
204
. However, this technique suffers from a number of limitations. Since the Si
3
N
4
masking layer
206
,
208
is a high-stress-bearing material, and is formed indiscriminately as a layer over the entire circuit, this configuration imparts undue stress on the various layers, which may lead to warping of the circuit. Additionally, due to its high density, the masking layer prevents outgassing of impurities contained in the inter-layer dielectric, for example C, F, and Cl, during later high-temperature processes. The remaining Si
3
N
4
masking layer would prevent the introduction of H
2
and O
2
during popular alloy processes, greatly affecting the conductive adhesiveness between the upper and lower metals.
Furthermore, this process is incompatible with contemporary memory fabrication processes, because the Si
3
N
4
masking layer would be applied between bit lines. Dielectric spacers formed on each lateral side of a cell bit line prevent shorting between the cell bit line and the nearby capacitor. In order to make such spacers, it would be necessary to remove any masking layer between adjacent bit lines to allow for space for the capacitors to be inserted between the bit lines. However, this process would also remove any dielectric layer formed over the bit lines necessary for insulating the bit lines from the capacitors. This would also remove any masking layer on either side of the peripheral region bit lines, defeating the purpose of forming the masking layer in the first place.
SUMMARY OF THE INVENTION
The present invention is

Affiliated with

Also associated with

Rating

Bit line pad and borderless contact on bit line stud with... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Bit line pad and borderless contact on bit line stud with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Bit line pad and borderless contact on bit line stud with... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3227896