1. Field of the Invention
The present invention refers to a stacked type of memory cell structure. In particular, the memory cell described is of the ferroelectric non-volatile type, but the same structure can be used for DRAM cells.
2. Description of the Related Art
As is known and shown in
The cell is capable of storing binary information thanks to the hysteresis characteristics of the ferroelectric material which is sandwiched between the plates 7 and 8 and which, when there are no applied voltages, can assume two bias states depending on the sign of previously-applied voltage across the capacitor 3 terminals.
All currently-known ferroelectric cells can be classified into two families: strapped cells and stacked cells.
In strapped cells, an embodiment of which is shown in
In stacked cells, an embodiment of which can be seen in
A titanium/titanium nitride region 26 runs below the first plate 7 to help the adhesion of the first plate 7 of the capacitor 3 on the protective layer 24.
The architecture of a array 28 of ferroelectric stacked or strapped cells 1 is shown in FIG. 4. It will be noted that the ferroelectric cells 1 are placed on rows and columns and are coupled so that the cell pairs 27 are placed parallel to bit lines BL; the transistors 2 of each cell pair 27 have common source regions, connected to the same bit line BL; and the capacitors 3 belonging to the cell pairs 27 adjacent in a parallel direction to the bit lines BL are connected to adjacent plate line pairs PL.
Ferroelectric stacked cells 1 are currently preferred, since they are the only ones capable of meeting the scalability requirements of new CMOS technologies. In stacked cells, the layout rules on the capacitor 3 design are crucial for the optimization of the cell.
There are therefore several known embodiments for stacked cells, apart from that shown in
In all these cases, the connection of at least one of the plates 7,8 with the same mask used for the connection of the ferroelectric material composing the dielectric region 17 is critical; for example, during connection, slightly volatile components are formed, and these can be redeposited along the capacitor edge and damage its active zone, causing a decay in the ferroelectric properties of the capacitor, with an increase in edge losses and lower voltage strength.
On the other hand, the separate definition of the three parts constituting the capacitor 3 (first and second plates 7,8 and dielectric region 17), which would allow the problem presented by current manufacturing processes to be solved, causes an increase in overall dimensions that is in conflict with present trends towards miniaturization. In fact, in making definition masks, account must be taken of both manufacturing tolerances (at present, with a 0.35 μm process, equal to 0.2 μm) and the minimum distances between the lower adjacent plates and the upper adjacent plates (for example, equal to 0.4 μm). In particular, with the conditions given above, it would be necessary for the lower plate 7 to be wider, with respect to the dielectric region 7, by an amount at least equal to the manufacturing tolerance (at least 0.2 μm) on each side; similarly, it would be necessary for the dielectric region 17 to be wider, with respect to the upper plate 8, by the same amount; therefore, taking into account the minimum distance between the lower plates 7, the resulting overall dimensions for the capacitors 3 and consequently for the ferroelectric cells are excessive.
An object the present invention is making a stacked memory cell, without the described disadvantages. According to the present invention, a stacked memory cell is realized as claimed in claim 1.
In practice, according to the invention, the dielectric regions of at least two adjacent cells in the direction of the bit lines are no longer separate. In particular, the dielectric region 17 can be continuous and shared between the two adjacent capacitors belonging to pairs of adjacent cells. In this way, the layout definition rules for capacitor scalability are given only by the distance between two lower adjacent electrodes and by the lateral space (enclosure) between upper electrode and lower electrode. This allows, with the same cell area, maximizing the working area of the capacitor compared with the layout of the arrangement where three different masks for definition of the capacitor are used and the dielectric region is divided between adjacent cells in the direction of the bit lines. This causes an increase in signal amplitude (proportional to the active area of the capacitor) supplied by each cell to the sense amplifier during reading. Alternatively, it is possible to obtain a reduction in the area occupied by the ferroelectric cells.
For a better understanding of the invention, an embodiment thereof will now be described, merely as a non-limiting example and with reference to the enclosed drawings, in which:
In detail,
Less doped drain extension regions 39 are formed in the substrate 31 below spacers 38 and a protective oxide layer 40 covers the surface of the substrate 31.
A first insulating layer 41 (for examples made of Boron Phosphorous Silicon Glass, or BPSG) runs above the protective oxide layer 40 and has openings into which run first and second contacts 43, 44 made of conductive material, for contacting the drain regions 33 and, respectively, the source regions 34.
Above the first insulating layer 41, there are formed lower plates 50 (corresponding to the first plates 7 in
Above the bands of ferroelectric material 51, there are first bands of conductive material 52 forming upper plates (corresponding to the second plates 8 in
Above the first insulating layer 41, there are also formed first contact regions 54, overlying and directly connected to the second contacts 44; the first contact regions 54 are also made preferably using a double layer of titanium and platinum.
A second insulating layer 55 is formed above the first insulating layer 41, covering the first bands of conductive material 52; openings are formed through the second insulating layer 55, into which third contacts 56 run (FIG. 5), directly connected to the first contact regions 54, and openings into which fourth contacts 57 run (FIG. 6), directly connected to the first bands of conductive material 52. Above the second insulating layer 55 metalization lines 60 are formed that define the bit lines BL and are directly connected to the third contacts 56; in addition, second contact regions 61 are formed (
Above the second insulating layer 55, completely covering the metalization lines 60, there runs a third insulating layer 63, showing openings into which fifth contacts 65 run (
A passivation layer 70 completely covers the device.
Purely by way of example, possible values relative to the regions which compose the ferroelectric cell 1 are given hereinbelow. The area of the capacitor 2 is in the range between 1 and 5 μm2; the lower titanium portion 50a of the lower plate is 20 nm thick; the upper platinum portion 50b has a thickness between 100 and 200 nm; the layer of ferroelectric material 51 has a thickness between 50 and 250 nm, and is preferably equal to 100 nm; the first bands of conductive material 52 are approximately 200 nm thick.
The process for manufacturing the ferroelectric cells 1 shown in
The advantages that can be obtained with the present invention are as follows. First of all, the ratio between capacitor 3 area and total cell 1 area is maximized. In fact, the restrictive rules in the cell design must now take into account only the distance between the two lower plates 50 belonging to adjacent pairs of cells 27 and the enclosure between the edges of the lower plate 50 and of the first band of conductive material 52 of each cell.
In addition, critical points in the steps of photolithography and ferroelectric material connection are removed.
Studies by the applicant have shown that the parasitic ferroelectric capacitor formed between two capacitors 3 belonging to two pairs of adjacent cells 27 on the same row does not prejudice the correct operation of the ferroelectric cells 1, given the increased thickness of the dielectric between the two plates 50 and the reduced facing area of the two plates 50 themselves. In addition, the ratio between the active capacitance (capacitance of capacitors 3) and the parasitic capacitance improves with the reduction in thickness of the lower plate 50 and of the layer of ferroelectric material 51 and therefore tends to be reduced with the increase in miniaturization of the devices, expected in the next technological generations.
Another embodiment is illustrated in
Bands of conductive material 152 extend above and in contact with the bands of ferroelectric material 51; furthermore, the bands of conductive material 152 and the bands of ferroelectric material 51 have the same width and are aligned to each other. In detail, the bands of conductive material 152 as well run, without gaps, above and between adjacent pairs of lower plates 50, belonging to different, adjacent pair of cells 127; in addition, the bands of conductive material 152 continuously run above and between the lower plates 50 of adjacent cells in a direction perpendicular to the bit lines 60, as also shown in the layout of
The bands of conductive material 152 form upper plates (corresponding to the second plates 8 in
The process for manufacturing the ferroelectric cells shown in
The embodiment described with reference to
All of the above U.S. patents, U.S. patent application publications, U.S. patent applications, foreign patents, foreign patent applications and non-patent publications referred to in this specification and/or listed in the Application Data Sheet, are incorporated herein by reference, in their entirety.
Finally, it is evident that modifications and variations can be made to the described cell, still remaining within the scope of the present invention.
For example, the layer of ferroelectric material 51 can run over the whole surface of the first insulating layer 41 except for the zones where the first contact regions 54 have to be made (openings 75 in FIG. 5); the materials can differ from those described, as long as they are suitable for the functions required; the exact arrangement of the areas and their dimensions can differ from those described above.
Number | Date | Country | Kind |
---|---|---|---|
TO99A0356 | Apr 1999 | IT | national |
This application is a continuation-in-part of allowed U.S. patent application Ser. No. 09/911,637, filed Jul. 23, 2001, now U.S. Pat. No. 6,656,801, which is a divisional of U.S. patent application Ser. No. 09/365,178, filed Aug. 2, 1999, which issued as U.S. Pat. No. 6,300,650, both applications and the patent being incorporated herein by reference.
Number | Name | Date | Kind |
---|---|---|---|
5350705 | Brassington et al. | Sep 1994 | A |
5418388 | Okudaira et al. | May 1995 | A |
5519237 | Itoh et al. | May 1996 | A |
5796133 | Kwon et al. | Aug 1998 | A |
5796136 | Shinkawata | Aug 1998 | A |
5955758 | Sandhu et al. | Sep 1999 | A |
6028361 | Ooishi | Feb 2000 | A |
6063656 | Clampitt | May 2000 | A |
Number | Date | Country | |
---|---|---|---|
20040058493 A1 | Mar 2004 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 09365187 | Aug 1999 | US |
Child | 09911637 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 09911637 | Jul 2001 | US |
Child | 10621262 | US |