1. Field of the Invention
The present invention relates to a nonvolatile switch, in particular for high-density nonvolatile programmable-logic devices.
2. Description of the Related Art
As known, programmable-logic devices are currently mainly formed by RAMs, which must be written each time the device is turned on. It is therefore necessary to provide an external memory that contains the code to be loaded at turning-on.
To eliminate the above need, programmable-logic devices, based upon nonvolatile components, have already been proposed. A solution is disclosed in U.S. Pat. No. 5,015,885, wherein a nonvolatile cell (EPROM or EEPROM) operates directly as a switch for connecting or separating horizontal and vertical segments formed by pass transistors. However, this solution is problematical as regards management of the switches, since they carry out two different functions and hence require a separate encoding for each function.
In other solutions, disclosed for example in U.S. Pat. NO. 6,625,221 and U.S. Pat. No. 5,576,568, a floating strip of polysilicon forming the gate electrode of a pass transistor is prolonged and used as the first plate of a capacitor, the second plate whereof is connected to a terminal of a coupling transistor. The polysilicon strip moreover forms the floating gate of an EEPROM cell or a plate of a further coupling capacitor to enable injection or extraction of charges from the polysilicon strip and hence programming and erasing of the cell. Also, these solutions are disadvantageous, since they require a large area and cannot be integrated in a memory array. Programming of the cells moreover requires high voltages, which are hardly from compatible with the devices and circuits integrated in the same chip.
The aim of the present invention is thus to provide a nonvolatile switch overcoming the drawbacks indicated above.
According to the present invention, a nonvolatile switch and a method for controlling a nonvolatile switch are provided, as defined in claim 1 and 9, respectively.
According to one aspect of the present invention, the switch is formed by a flash cell and a pass transistor, which have a common floating-gate region. The flash cell enables modification of the charges contained in the floating-gate region by channel hot-electron injection (writing) and extraction by Fowler-Nordheim (FN) tunneling effect (erasing), and hence modulation of the threshold of the pass transistor. The pass transistor therefore has a variable threshold, which depends upon the charge present in the floating-gate region. The flash cell and the pass transistor also share the control-gate region. The pass transistor is connected to the outside of the cell and enables or not passage of the logic signals from the input terminal to the output terminal according to the threshold programmed. If the threshold of the pass transistor is sufficiently low, once it has been selected through its control gate region, it outputs the logic signal present at input. If, instead, the threshold of the pass transistor is high, it is off even when it is selected and does not enable passage of the logic signal.
For an understanding of the present invention there are now described some preferred embodiments thereof, which are provided purely by way of non-limiting example, with reference to the attached drawings, wherein:
In detail, the memory element 2 has a drain terminal FD connected to a first bias generator 10 and a source terminal FS connected to a second bias generator 11. The pass transistor 3 has a drain terminal PD connected to a data input 12 and a source terminal PS connected to a data output 13. In addition, the control-gate region 5 is connected by a terminal CG to a third bias generator 14, and a body region B, in common both to the memory element 2 and to the pass transistor 3, is connected to a fourth generator 15.
The table of
The table of
The switch 1 can be implemented in a very compact way, as is illustrated by way of example in the layout of
In particular, in the illustrated embodiment, the memory element 2 is made in a first active area 20, and the pass transistor 3 is made in a second and a third active areas 21, 22. The active areas 20–22 extend parallel to one another in a well region 24 (see
On top of the well 24, there extend, in order (see
As may be noted in
In practice, the pass transistor 3 comprises two pass transistors formed in two adjacent active areas (second and third active area 21, 22) and parallel-connected so as to operate as a single pass transistor 3 with an area sufficient to conduct the required current.
In addition, the memory source connection line 32 is connected to a memory source region 45 formed in the first active area 20, through a third contact 46. Instead, the memory drain connection line 33 is connected to a memory drain region 47 formed in the first active area 20, through a fourth contact 48 and a memory local-interconnection line 50, formed directly on top of the substrate 24 between the memory drain region 47 and the fourth contact 48.
Analogously to what illustrated in
This solution proves advantageous when the pass transistors are to be configured in the same way and selected simultaneously (for example, for connection to a bus), thanks to the saving in the occupied area that can be obtained.
Also, the above embodiment enables different embodiments of the layout. For example, the four active areas forming the pass transistors 3a, 3b can be joined in pairs, as illustrated in
Thanks to the described structure, it is possible to obtain a nonvolatile switch with an extremely reduced area. In addition, the described switches can be made using standard technology for manufacturing flash cells, and hence using machines commonly available in the microelectronics industry and known and reliable processing steps. The use of a flash memory element further enables also negative voltages to be used, with a consequent reduction of the amplitude of the voltages applied.
In addition, it is possible to form the switch within a flash-memory array, and hence in a still more compact way. In this case, the connection lines 30–33 operate as bitlines, and the control-gate region 5 operates as a wordline.
Finally, it is clear that numerous modifications and variations can be made to switch described and illustrated herein, all of which fall within the scope of the invention, as defined in the attached claims. For example, also the memory element 2 can be formed in two distinct active areas, and the floating gate region can be shortened so as to end above the first active area 20 (see
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.
From the foregoing it will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without deviating from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
03425225 | Apr 2003 | EP | regional |
Number | Name | Date | Kind |
---|---|---|---|
5015885 | El Gamal et al. | May 1991 | A |
5247478 | Gupta et al. | Sep 1993 | A |
5576568 | Kowshik | Nov 1996 | A |
5587603 | Kowshik | Dec 1996 | A |
5625211 | Kowshik | Apr 1997 | A |
5764096 | Lipp et al. | Jun 1998 | A |
5949710 | Pass et al. | Sep 1999 | A |
RE37308 | Cappelletti et al. | Aug 2001 | E |
6456529 | Sansbury et al. | Sep 2002 | B1 |
Number | Date | Country | |
---|---|---|---|
20040233736 A1 | Nov 2004 | US |