Memory array with one shared deep doped region

This non-provisional application claims priority of U.S. provisional application No. 62/280,683, filed on Jan. 19, 2016, included herein by reference in its entirety.

1. Field of the Invention

This invention relates to a memory array, and more particularly, a memory array with memory cells sharing one deep doped region.

2. Description of the Prior Art

An electrically rewritable nonvolatile memory is a type of memory that retains information it stores even when no power is supplied to memory blocks and allows on-board rewriting of a program. Due to the wide range of applications for various uses, there is a growing need for a nonvolatile memory to be embedded in the same chip with the main circuit, especially for personal electronic devices having strict requirements for circuit area.

A nonvolatile memory cell of prior art comprises one floating gate transistor for retaining data, and one or two select transistors for enabling the floating gate transistor to perform corresponding operations. The floating gate may be controlled by coupling elements for program operations and erase operations. Since memory cells in different memory pages should be controlled independently, memory cells in different memory pages are usually disposed in isolated regions. However, due to the spacing rule of the manufacture, the spare area between different isolated regions can significantly increase the circuit area. Furthermore, since no elements are allowed to be disposed on the spare area, the increased circuit area is simply wasted. Therefore, how to reduce the circuit area and use the circuit area more efficiently has become an issue to be solved.

One embodiment of the present invention discloses a memory array. The memory array includes a plurality of memory pages, each memory page includes a plurality of memory cells, and each memory cell includes a floating gate module, a control element, and an erase element.

The floating gate module includes a floating gate transistor. The floating gate module can control the floating gate transistor according to a source line, a bit line and a word line. The floating gate transistor has a first terminal, a second terminal and a floating gate. The control element has a body terminal coupled to a control line, a first terminal coupled to the body terminal, a second terminal coupled to the body terminal, and a control terminal coupled to the floating gate. The erase element has a body terminal for receiving a first voltage during a program operation and a program inhibit of the memory cell and receiving a second voltage during an erase operation of the memory cell, a first terminal coupled to an erase line, a second terminal coupled to the first terminal of the erase element or being floating, and a control terminal coupled to the floating gate.

The floating gate module is disposed in a first well, the erase element is disposed in a second well, and the control element is disposed in a third well. The first well, the second well and the third well are disposed in a deep doped region. Memory cells of the plurality of memory pages are all disposed in the deep doped region. The control line is at the first voltage during the program operation, and the erase line is at the second voltage during the erase operation.

These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.

FIG. 1 shows a memory array according to one embodiment of the present invention.

FIG. 2 shows a layout of the memory array in FIG. 1 according to one embodiment of the present invention.

FIG. 3 shows a section view of the erase elements according to the layout in FIG. 2 .

FIG. 4 shows voltages of the signals during a program operation of a memory cell in the memory array of FIG. 1 .

FIG. 5 shows voltages of the signals during an erase operation of a memory cell in the memory array of FIG. 1 .

FIG. 6 shows a memory array according to another embodiment of the present invention.

FIG. 7 shows voltages of the signals during an erase operation of a memory cell in the memory array of FIG. 6 .

FIG. 8 shows a memory array according to another embodiment of the present invention.

FIG. 9 shows voltages of the signals during a program operation of a memory cell in the memory array of FIG. 8 .

FIG. 10 shows voltages of the signals during an erase operation of a memory cell in the memory array of FIG. 8 .

FIG. 11 shows a memory array according to another embodiment of the present invention.

FIG. 12 shows voltages of the signals during an erase operation of a memory cell in the memory array of FIG. 11 .

FIG. 1 shows a memory array 10 according to one embodiment of the present invention. The memory array 10 includes M memory pages MP 1 to MPM. Each memory page MP 1 to MPM includes N memory cells. For example, the memory page MP 1 includes memory cells 100 1,1 to 100 1,N , and the memory page MPM includes memory cells 100 M,1 to 100 M,N . M and N are positive integers.

In some embodiments of the present invention, memory cells in the same memory page may be coupled to a same control line, a same erase line, and a same word line, but may be coupled to different source lines and different bit lines. For example, the memory cells 100 1,1 to 100 1,N in the same memory page MP 1 are coupled to the same control line CL 1 , the same erase line EL 1 , and the same word line WL 1 . However, the memory cell 100 1,1 is coupled to a source line SL 1 and a bit line BL 1 while the memory cell 100 1,N is coupled to a source line SLN and a bit line BLN.

In addition, memory cells in different memory pages but in the same column may be coupled to different control lines, different erase lines, and different word lines, but may be coupled to a same source line and a same bit line. For example, the memory cells 100 1,1 and 100 M,1 are disposed in the same column but in different memory pages MP 1 and MPM. The memory cells 100 1,1 and 100 M,1 are coupled to the same source line SL 1 and the same bit line BL 1 . However, the memory cell 100 1,1 is coupled to the control line CL 1 , the erase line EL 1 , and the word line WL 1 while the memory cell 100 M,1 is coupled to the control line CLM, the erase line ELM, and the word line WLM.

In FIG. 1 , the memory cells 100 1,1 to 100 1,N , . . . , and 100 M,1 to 100 M,N have same structures. Each memory cell includes a floating gate module 110 , a control element 120 , and an erase element 130 . The floating gate module 110 includes a floating gate 112 , a source transistor 114 , and a bit transistor 116 . The floating gate module 110 can control the floating gate transistor 112 according to a source line, a bit line and a word line.

The floating gate transistor 112 has a first terminal, a second terminal and a floating gate. The source transistor 114 has a first terminal, a second terminal, and a control terminal. The first terminal of the source transistor 114 is coupled to a corresponding source line. For example, the first terminal of the source transistor 114 of the memory cell 100 1,1 may be coupled to the source line SL 1 , and the first terminal of the source transistor 114 of the memory cell 100 1,N may be coupled to the source line SLN. The second terminal of the source transistor 114 is coupled to the first terminal of the floating gate transistor 112 , and the control terminal of the source transistor 114 is coupled to a corresponding word line WL 1 . For example, the control terminal of the source transistor 114 of the memory cell 100 1,1 may be coupled to the word line WL 1 , and the control terminal of the source transistor 114 of the memory cell 100 M,1 may be coupled to the word line WLM.

The bit transistor 116 has a first terminal, a second terminal, and a control terminal. The first terminal of the bit transistor 116 is coupled to the second terminal of the floating gate transistor 112 , the second terminal of the bit transistor 116 is coupled to a corresponding bit line, and the control terminal of the bit transistor 116 is coupled to a corresponding word line. For example, the second terminal of the bit transistor 116 of the memory cell 100 1,1 may be coupled to the bit line BL 1 , and the second terminal of the bit transistor 116 of the memory cell 100 1,N may be coupled to the bit line BLN. Also, the control terminal of the bit transistor 116 of the memory cell 100 1,1 may be coupled to the word line WL 1 , and the control terminal of the bit transistor 116 of the memory cell 100 M,1 may be coupled to the word line WLM.

The control element 120 has a first terminal coupled to the body terminal, a second terminal coupled to the body terminal, a control terminal coupled to the floating gate of the floating gate transistor 112 , and a body terminal coupled to a corresponding control line. For example, the body terminal of the control element 120 of the memory cell 100 1,1 may be coupled to the control line CL 1 , and the body terminal of the control element 120 of the memory cell 100 M,1 may be coupled to the control line CLM.

The erase element 130 has a first terminal, a second terminal, a control terminal, and a body terminal. The first terminal 134 of the erase element 130 is coupled to a corresponding erase line. For example, the first terminal of the erase element 130 of the memory cell 100 1,1 is coupled to the erase line EL 1 , and the first terminal of the erase element 130 of the memory cell 100 M,1 is coupled to the erase line ELM. The second terminal 136 of the erase element 130 is coupled to the first terminal of the erase element 130 or floating, the control terminal 138 of the erase element 130 is coupled to the floating gate of the floating gate transistor 112 , and the body terminal 132 of the erase element 130 is coupled to a well bias line WBL.

FIG. 2 shows a layout of the memory array 10 according to one embodiment of the present invention. The floating gate module 110 of the memory cell 110 1,1 can be disposed in an active region AAF 1 of a first P-well PW 1 , the erase element 130 of the memory cell 110 1,1 can be disposed in an active region AAE 1 of a first N-well NW 1 , and the control element 120 of the memory cell 110 1,1 can be disposed in an active region AAC 1 of a second P-well PW 2 . The first P-well PW 1 , the first N-well NW 1 and the second P-well PW 2 are disposed in the same deep doped region DR. In some embodiments, the deep doped region DR can be a deep N-well or an N-type buried layer.

FIG. 3 shows a section view of the erase elements 130 according to the layout in FIG. 2 . In FIG. 3 , the erase element 130 has a structure similar to a P-type metal-oxide-semiconductor transistor. That is, the body terminal 132 of the erase element 130 is at the N-well NW, and the first terminal 134 and the second terminal 136 are two P-type doped regions P+ disposed in the N-well NW. In FIG. 3 , the well bias line WBL is coupled to the body terminal 132 directly. However, in some embodiments, the well bias line WBL may also be coupled to the N-well NW through a contact or an N-type doped region in the N-well NW. The floating gate of the floating gate transistor 112 is coupled to the control terminal 138 of the erase element 130 forming a gate structure. Since the erase line EL is coupled to the first terminal 134 of the erase element 130 , the memory cells 110 1,1 to 110 1,N , . . . , and 110 M,1 to 110 M,N can function correctly even with the their body terminals 132 of the erase elements 130 coupled to the same well bias line WBL. That is, the memory cells 110 1,1 to 110 1,N , . . . , and 110 M,1 to 110 M,N can be disposed in the same deep doped region DR, which is coupled to the same well bias line WBL.

For example, in FIG. 2 , the floating gate module 110 of the memory cell 110 1,N can be disposed in an active region AAF 2 of a third P-well PW 3 , the erase element 130 of the memory cell 110 1,N can be disposed in an active region AAE 2 of a second N-well NW 2 , and the control element 120 of the memory cell 110 1,N can be disposed in the active region AAC 1 of the second P-well PW 2 ; however, the third P-well PW 3 and the second N-well NW 2 are still disposed in the same deep doped region DR.

Also, although the floating gate modules, the control elements, and the erase elements of the memory cells 110 M,1 to 110 M,N may be disposed in different wells as shown in FIG. 2 , the different wells of the memory cells 110 M,1 to 110 M,N can still be disposed in the same deep doped region DR. Namely, memory cells 110 1,1 to 110 1,N , . . . , and 110 M,1 to 110 M,N of the M memory pages MP 1 to MPM can all be disposed in the same deep doped region DR. Since different memory pages MP 1 to MPM in the memory array 10 are disposed in one deep doped region DR, the spacing rules between deep doped regions will no longer be used to limit the circuit area of the memory array 10 , and the circuit area of the memory array 10 can be reduced significantly.

In FIG. 2 , the control elements 120 of the memory cells in the same memory page, such as the memory cells 110 1,1 to 110 1,N in the memory page MP 1 , can be disposed in the same second P-well PW 2 . The floating gate modules 110 of the memory cells 110 1,1 to 110 1,N can be disposed in two different P-wells PW 1 and PW 3 , which are disposed in opposite sides of the second P-well PW 2 . The erase elements 130 of the memory cells 110 1,1 to 110 1,N can be disposed in two different N-wells NW 1 and NW 2 , which are disposed in opposite sides of the second P-well PW 2 . Therefore, the layout of the memory array 10 will not extend to one single direction, and the layout of the memory array 10 can be more flexible. However, in some embodiments, the floating gate modules 110 of the memory cells in the same memory page can also be disposed in one P-well and the erase elements 130 of the memory cells in the same memory page can be disposed in one N-well according to the system requirements.

FIG. 4 shows voltages of the signals during a program operation of the memory cell 100 1,1 in the memory array 10 . In FIG. 4 , the first voltage VPP is substantially equal to the second voltage VEE. The first voltage VPP is greater than the third voltage VEE′, the third voltage VEE′ is greater than the fourth voltage VINH 1 , and the fourth voltage VINH 1 is greater than the fifth voltage VSS. Also, the first voltage VPP is greater than the sixth voltage VPP′, and the sixth voltage VPP′ is greater than the fifth voltage VSS.

In some embodiments, the difference between the third voltage VEE′ and the fifth voltage VSS is greater than half of the difference between the first voltage VPP and the fifth voltage VSS. The difference between the fourth voltage VINH 1 and the fifth voltage VSS is smaller than half of the difference between the first voltage VPP and the fifth voltage VSS, and the difference between the sixth voltage VPP′ and the fifth voltage VSS is smaller than half of the difference between the first voltage VPP and the fifth voltage VSS. For example, if the first voltage VPP is 18V, the second voltage VEE is in a range between 17V and 18V, and the fifth voltage VSS is 0V, then the third voltage VEE′ may be 13V, the fourth voltage VINH 1 may be 6V, and the sixth voltage VPP′ may also be 6V.

According to FIG. 4 , during the program operation of the memory cell 100 1,1 , the control line CL 1 is at the first voltage VPP, the erase line EL 1 is at the third voltage VEE′, the word line WL 1 is at the fourth voltage VINH 1 , the source line SL is at the fifth voltage VSS, and the bit line BL is at the fifth voltage VSS.

In this case, the control element 120 of the memory cell 100 1,1 is coupled to a high voltage by the control line CL 1 . The source transistor 114 and the bit transistor 116 are turned on so the first terminal and the second terminal of the floating gate transistor 112 of the memory cell 100 1,1 are pulled down to a low voltage. Therefore, the high voltage difference applied to the floating gate transistor 112 will induce FN (Fowler Nordheim) electron tunneling injection to the floating gate, and the memory cell 100 1,1 can be programmed. Also, to prevent leakage currents generated between the P-wells and the N-wells in the memory array 10 , the voltage of the well bias line WBL should not be smaller than the greatest voltage of all the signals. In this case, the well bias line WBL would be at the first voltage VPP.

Also, to prevent the memory cell 100 1,N in the same memory page MP 1 as the memory cell 100 1,1 from being programmed during the program operation of the memory cell 100 1,1 , the memory cell 100 1,N may perform a program inhibit operation during the program operation of the memory cell 100 1,1 . During the program inhibit operation of the memory cell 100 1,N , the control line CL 1 is at the first voltage VPP, the erase line EL 1 is at the third voltage VEE′, the word line WL 1 is at the fourth voltage VINH 1 , the source line SLN is at a fourth voltage VINH 1 , and the bit line BLN is at the fourth voltage VINH 1 .

In this case, although the memory cell 100 1,N is coupled to the same control line CL 1 , the erase line EL 1 , and the word line WL 1 as the memory cell 100 1,1 , the memory cell 100 1,N will not be programmed due to the effect of channel boost caused by the source transistor 114 and the bit transistor 116 of the memory cell 100 1,N . That is, the voltages of the first terminal and the second terminal of the floating gate transistor 112 are boosted to a voltage level higher than the fourth voltage VINH 1 , so the floating gate of the memory cell 100 1,N is not able to capture enough electrons and the memory cell 100 1,N will not be programmed. Also, since the control line CL 1 is at the first voltage VPP, the well bias line WBL is still at the first voltage VPP during the program inhibit operation of the memory cell 100 1,N .

Furthermore, during the program operation of the memory cell 100 1,1 , memory cells in unselected memory pages, such as the memory page MPM, should not be programmed. Therefore, in FIG. 4 , a control line CLM coupled to an unselected memory cell 100 M,1 in an unselected memory page MPM is at the sixth voltage VPP′, an erase line ELM coupled to the unselected memory cell 100 M,1 is at the third voltage VEE′, and a word line WLM coupled to the unselected memory cell 100 M,1 is at the fourth voltage VINH 1 .

Since the body terminal of the erase element 130 of the memory cell 100 M,1 is coupled to the well bias line WBL, which is at the first voltage VPP, the voltage of the erase line ELM cannot be too low; otherwise, the erase element 130 may breakdown. Meanwhile, the voltage of the erase line ELM cannot be too high; otherwise, the floating gate of the memory cell 100 M,1 may be programmed unexpectedly. Therefore, the erase line ELM can be at the third voltage VEE′ during the program operation of the memory cell 100 1,1 , and the difference between the third voltage VEE′ and the fifth voltage VSS can be slightly greater than half of the difference between the first voltage VPP and the fifth voltage VSS. In this case, the erase element 130 will not breakdown and the memory cell 100 M,1 will not be programmed unexpectedly.

Also, the voltage of the control line CLM should not be too low; otherwise, the memory cell 100 M,1 may be unstable. Therefore, the control line CLM is at the sixth voltage VPP′ during the program operation of the memory cell 100 1,1 , and the difference between the sixth voltage VPP′ and the fifth voltage VSS can be slightly smaller than half of the difference between the first voltage VPP and the fifth voltage VSS. In this case, the erase element 130 can remain stable.

In addition, since memory cells in the same column but different memory pages are coupled to the same source line and the same bit line, the word line may be at the fourth voltage VINH 1 for reducing the gate-induced drain leakage (GIDL) current. For example, during the program operation of the memory cell 100 1,1 and the program inhibit operation of the memory cell 100 1,N , the source line SLN and the bit line BLN coupled to the memory cell 100 M,N are at the fourth voltage VINH 1 . If the word line WLM is at the fifth voltage VSS, the big voltage difference may cause GIDL currents at the source transistor 114 and the bit transistor 116 of the memory cell 100 M,N . However, the word line WLM at the fourth voltage VINH 1 can avoid the GIDL currents efficiently while not affecting the operations of other memory cells.

FIG. 5 shows voltages of the signals during an erase operation of the memory cell 100 1,1 in the memory array 10 . During an erase operation of the memory cell 100 1,1 , the erase line EL 1 is at the second voltage VEE, the word line WL 1 is at the fourth voltage VINH 1 or the fifth voltage VSS, the source line SL 1 is at the fourth voltage VINH 1 , the bit line BL 1 is at the fourth voltage VINH 1 , and the control line CL 1 is at the fifth voltage VSS.

In this case, the high voltage of the erase line EL 1 can cause FN electron tunneling ejection so the memory cell 100 1,1 can be erased. In addition, since the erase line EL 1 has the greatest voltage, that is, the second voltage VEE, among all the signals during the erase operation of the memory cell 100 1,1 , the well bias line WBL would be at the second voltage VEE.

In some embodiments of the present invention, the memory array 10 can be erased by page. That is, memory cells at the same memory page, such as the memory cells 100 1,1 to 100 1,N in the memory page MP 1 , will be erased at the same time. In this case, the source lines SL 1 to SLN and the bit lines BL 1 to BLN coupled to the memory cells 100 1,1 to 100 1,N may all be at a same rather low voltage. For example, the source lines SL 1 to SLN and the bit lines BL 1 to BLN may all be at the fourth voltage VINH 1 or at the fifth voltage VSS. In this case, the difference between the fourth voltage VINH 1 and the fifth voltage VSS can be smaller than half of the difference between the second voltage VEE and the fifth voltage VSS.

In addition, during the erase operation of the memory cell 100 1,1 , memory cells in unselected memory pages, such as the memory page to MPM, should not be erased. For example, to prevent the memory cell 100 M,1 in the unselected memory page MPM from being erased, the voltage of the erase line ELM should not be too high. However, since the well bias line WBL is at the second voltage VEE, the voltage of the erase line ELM cannot be too low; otherwise, the erase element 130 of the memory cell memory cell 100 1,1 may breakdown. Therefore, according to FIG. 5 , the erase line ELM can be at the third voltage VEE′. The difference between the third voltage VEE′ and the fifth voltage VSS can be slightly greater than half of the difference between the second voltage VEE and the fifth voltage VSS.

In this case, the voltage of the erase line ELM is not high enough to erase the memory cell 100 M,1 and is not low enough to make the erase element 130 breakdown. The control line CLM is at the sixth voltage VPP′ so that the memory cell 100 M,1 will not be programmed or erased unexpectedly by the erase line ELM. In FIG. 5 , the difference between the sixth voltage VPP′ and the fifth voltage VSS is smaller than half of the difference between the second voltage VEE and the fifth voltage VSS. Similarly, the word line WLM, the source line SL 1 , and the bit line BL 1 can be at approximate voltages so that the memory cell 100 M,1 will not be programmed or erased unexpectedly by the erase line ELM while the GIDL current can be prevented. In some embodiments, the word line WLM, the source line SL 1 , and the bit line BL 1 can be at the fourth voltage VINH 1 .

In some embodiments of the present invention, the memory array can be erased by sector. That is, memory cells in the memory array can all be erased at the same time. FIG. 6 shows a memory array 20 according to one embodiment of the present invention. The memory arrays 10 and 20 have similar structures. The difference between these two is in that the memory cells 200 1,1 to 200 1,N , . . . , and 200 M,1 to 200 M,N are all coupled to the same erase line EL 0 , so the memory cells 200 1,1 to 200 1,N , . . . , and 200 M,1 to 200 M,N in the memory array 20 will all be erased at the same time.

FIG. 7 shows voltages of the signals during an erase operation of the memory cell 200 1,1 in the memory array 20 .

During the erase operation of the memory cells 200 1,1 , the erase line EL 0 is at the second voltage VEE, the control line CL 1 is at the fifth voltage VSS, the source line SL 1 and the bit line BL 1 are both at the fourth voltage VINH 1 or at the fifth voltage VSS, while the word line is at the fourth voltage VINH 1 or at the fifth voltage VSS.

In this case, the high voltage of the erase line EL 0 can cause FN electron tunneling ejection so the memory cell 200 1,1 can be erased. Since the memory cells 200 1,1 to 200 1,N , . . . , and 200 M,1 to 200 M,N in the memory array 20 are erased at the same time, voltages of the signals received by all the memory cells 200 1,1 to 200 1,N , . . . , and 200 M,1 to 200 M,N can be the same.

In addition, since the erase line is at the third voltage VEE′ during the program operation and the program inhibit operation, the memory array 20 can be operated with the same principle as the memory array 10 during the program operation and the program inhibit operation as shown in FIG. 4 .

Consequently, the memory cells 200 1,1 to 200 1,N , . . . , and 200 M,1 to 200 M,N of the memory array 20 can all be disposed in the same deep doped region. Since different memory pages MP 1 to MPM in the memory array 20 are disposed in one deep doped region, the spacing rules between deep doped regions will no longer be used to limit the circuit area of the memory array 20 , and the circuit area of the memory array 20 can be reduced significantly. In addition, since all the memory cells 200 1,1 to 200 1,N , . . . , and 200 M,1 to 200 1,4,N of the memory array 20 are coupled to the same erase line, the driving circuit for providing the erase line can be simplified, which can further reduce the chip area required by the memory array 20 .

FIG. 8 shows a memory array 30 according to one embodiment of the present invention. The memory array 30 has the similar structure as the memory array 10 . The difference between these two is in that each of the memory cells 300 1,1 to 300 1,N , . . . , and 300 M,1 to 300 1,4,N has a floating gate module 310 , a control element 120 and an erase element 130 .

The floating gate module 310 includes a floating gate 312 and a source transistor 314 . The floating gate transistor 312 has a first terminal, a second terminal, and a floating gate. The second terminal of the floating gate transistor 312 is coupled to a corresponding bit line. For example, the second terminal of the floating gate transistor 312 of the memory cell 300 1,1 is coupled to the bit line BL 1 , and the second terminal of the floating gate transistor 312 of the memory cell 300 1,N is coupled to the bit line BLN. The floating gate of the floating gate transistor 312 is coupled to the control element 120 and the erase element 130 .

The source transistor 314 has a first terminal, a second terminal, and a control terminal. The first terminal of the source transistor 314 is coupled to a corresponding source line. For example, the first terminal of the source transistor 314 of the memory cell 300 1,1 can be coupled to the source line SL 1 , and the first terminal of the source transistor 314 of the memory cell 300 1,N can be coupled to the source line SLN. The second terminal of the source transistor 314 is coupled to the first terminal of the floating gate transistor 112 , and the control terminal of the source transistor 314 is coupled to a corresponding word line. For example, the control terminal of the source transistor 314 of the memory cell 300 1,1 can be coupled to the word line WL 1 , and the control terminal of the source transistor 314 of the memory cell 300 M,1 can be coupled to the word line WLM.

FIG. 9 shows voltages of the signals during a program operation of the memory cell 300 1,1 in the memory array 30 .

In FIG. 9 , during the program operation of the memory cell 300 1,1 , the control line CL 1 is at the first voltage VPP, the erase line EL 1 is at the third voltage VEE′, the word line WL 1 is at the fourth voltage VINH 1 , the source line SL 1 is at the fifth voltage VSS, and the bit line BL 1 is at the fifth voltage VSS.

In this case, the control element 120 of the memory cell 300 1,1 is coupled to a high voltage by the control line CL 1 . The source transistor 314 is turned on so the first terminal and the second terminal of the floating gate transistor 312 of the memory cell 300 1,1 are pulled down to a low voltage. Therefore, the high voltage difference applied to the floating gate transistor 312 will induce FN (Fowler Nordheim) electron tunneling injection to the floating gate, and the memory cell 300 1,1 can be programmed. Also, to prevent leakage currents generated between the P-wells and the N-wells in the memory array 30 , the voltage of the well bias line WBL should not be smaller than the greatest voltage of all the signals. In this case, the well bias line WBL would be the first voltage VPP.

Also, in some embodiments, to prevent the memory cell 300 1,N in the same memory page MP 1 as the memory cell 300 1,1 from being programmed during the program operation of the memory cell 300 1,1 , the memory cell 300 1,N may perform a program inhibit operation during the program operation of the memory cell 300 1,1 . During the program inhibit operation of the memory cell 300 1,N , the control line CL 1 is at the first voltage VPP, the erase line EL 1 is at the third voltage VEE′, the word line WL 1 is at the fourth voltage VINH 1 , the source line SLN is at a seventh voltage VINH 2 , and the bit line BLN is at the seventh voltage VINH 2 .

Since the second terminal of the floating gate transistor 312 is coupled to the corresponding bit line, the bit line BLN may have to be at a rather high voltage to prevent the memory cell 300 1,N from being programmed. In this case, the bit line BLN can be at the seventh voltage VINH 2 . The difference between the seventh voltage VINH 2 and the fifth voltage VSS must be smaller than the source/drain junction breakdown voltage of the floating gate transistor 312 . For example, if the source/drain junction breakdown voltage of the floating gate transistor 312 is 9V, the seventh voltage VINH 2 may be 8V.

In this case, although the memory cell 300 1,N is coupled to the same control line CL 1 , the erase line EL 1 , and the word line WL 1 as the memory cell 300 1,1 , the memory cell 300 1,N will not be programmed due to the rather high voltages at the first terminal and the second terminal of the floating gate transistor 312 of the memory cell 300 1,N . Also, since the control line CL 1 is at the first voltage VPP, the well bias line WBL is still at the first voltage VPP during the program inhibit operation of the memory cell 300 1,N .

Furthermore, during the program operation of the memory cell 300 1,1 , memory cells in unselected memory pages, such as the memory page to MPM, should not be programmed. Therefore, in FIG. 9 , a control line CLM coupled to an unselected memory cell 300 M,1 in an unselected memory page MPM is at a sixth voltage VPP′, an erase line ELM coupled to the unselected memory cell 300 M,1 is at the third voltage VEE′, and a word line WLM coupled to the unselected memory cell 300 M,1 is at the fourth voltage VINH 1 . In FIG. 9 , the difference between the sixth voltage VPP′ and the fifth voltage VSS is smaller than half of the difference between the first voltage VPP and the fifth voltage VSS.

Since the erase line ELM is at the third voltage VEE′ during the program operation of the memory cell 300 1,1 , the erase element 130 will not breakdown and the memory cell 300 M,1 will not be programmed unexpectedly. Also, the control line CLM is at the sixth voltage VPP′ to ensure that the memory cell 300 M,1 is not programmed.

In addition, since memory cells in the same column but different memory pages is coupled to the same source line and the same bit line, the word line may be at the fourth voltage VINH 1 for reducing the gate-induced drain leakage (GIDL) current. For example, during the program operation of the memory cell 300 1,1 and the program inhibit operation of the memory cell 300 1,N , the source line SLN and the bit line BLN coupled to the memory cell 300 M,N are at the seventh voltage VINH 2 . If the word line WLM is at the fifth voltage VSS, the big reverse voltage difference may cause GIDL currents at the source transistor 314 of the memory cell 300 M,N . However, the word line WLM at the fourth voltage VINH 1 can avoid the GIDL currents efficiently while not affecting the operations of other memory cells.

FIG. 10 shows voltages of the signals during an erase operation of the memory cell 300 1,1 in the memory array 30 .

During the erase operation of the memory cell 300 1,1 , the erase line EL 1 is at the second voltage VEE, the control line CL 1 is at a fifth voltage VSS, the source line and the bit line are both at the fourth voltage VINH 1 or at the fifth voltage VSS, while the word line is at the fourth voltage VINH 1 or at the fifth voltage VSS.

In this case, the high voltage of the erase line EL 1 can cause FN electron tunneling ejection so the memory cell 300 1,1 can be erased. In addition, since the erase line EL 1 has the greatest voltage, that is, the second voltage VEE, among all the signals during the erase operation of the memory cell 300 1,1 , the well bias line WBL would be at the second voltage VEE.

In addition, during the erase operation of the memory cell 300 1,1 , memory cells in unselected memory pages, such as the memory page MPM, should not be erased. For example, to prevent the memory cell 300 M,1 in the unselected memory page MPM from being erased, the voltage of the erase line ELM should not be too high. However, since the well bias line WBL is at the second voltage VEE, the voltage of the erase line ELM cannot be too low; otherwise, the erase element 130 of the memory cell memory cell 300 M,1 may breakdown. Therefore, in FIG. 10 , the erase line ELM can be at the third voltage VEE′.

In this case, the voltage of the erase line ELM is not high enough to erase the memory cell 300 M,1 and is not low enough to make the erase element 130 breakdown. According to the voltage of the erase line ELM, the control line CLM can be at a sixth voltage VPP′. In FIG. 10 , the difference between the sixth voltage VPP′ and the fifth voltage VSS is smaller than half of the difference between the second voltage VEE and the fifth voltage VSS. Also, the word line WLM, the source line SL 1 , and the bit line BL 1 can be at approximate voltages so that the memory cell 300 M,1 will not be programmed or erased unexpectedly by the erase line ELM while the GIDL current can be prevented. In some embodiments, the word line WLM, the source line SL 1 , and the bit line BL 1 can be at the fourth voltage VINH 1 . Since the erase line EL 1 coupled to the memory cell 300 1,1 is at an even higher voltage, the second voltage VEE, and the control line CL 1 coupled to the memory cell 300 1,1 is at a low voltage, the fifth voltage VSS, the memory cell 300 1,1 can still be erased normally even with the source line SL 1 and the bit line BL 1 being at the fourth voltage VINH 1 .

In addition, in some embodiments of the present invention, the memory array 30 can be erased by page. That is, memory cells at the same memory page, such as the memory cells 300 1,1 to 300 1,N in the memory page MP 1 , will be erased at the same time. In this case, the source lines SL 1 to SLN and the bit lines BL 1 to BLN coupled to the memory cells 300 1,1 to 300 1,N may all be at the fourth voltage VINH 1 during the erase operation.

In some embodiments of the present invention, the memory array can be erased by sector. That is, memory cells in the memory array can all be erased at the same time. FIG. 11 shows a memory array 400 according to one embodiment of the present invention. The memory arrays 400 and 300 have similar structures. The difference between these two is in that the memory cells 400 1,1 to 400 1,N , . . . , and 400 M,1 to 400 M,N are all coupled to the same erase line EL 0 so the memory cells 400 1,1 to 400 1,N , . . . , and 400 M,1 to 400 M,N in the memory array 40 will all be erased at the same time.

FIG. 12 shows voltages of the signals during an erase operation of the memory cell 400 1,1 in the memory array 40 .

During the erase operation of the memory cells 400 1,1 , the erase line EL 0 is at the second voltage VEE, the control line CL 1 is at the fifth voltage VSS, the source line SL 1 and the bit line BL 1 are both at the fourth voltage VINH 1 or at the fifth voltage VSS, while the word line is at the fourth voltage VINH 1 or at the fifth voltage VSS. In this case, the high voltage of the erase line EL 0 can cause FN electron tunneling ejection so the memory cell 400 1,1 can be erased.

Since the memory cells 400 1,1 to 400 1,N , . . . , and 400 M,1 to 400 M,N in the memory array 40 are erased at the same time, voltages of the signals received by all the memory cells 400 1,1 to 400 1,N , . . . , and 400 M,1 to 400 M,N can be the same.

In addition, since the erase line EL 0 is at the third voltage VEE′ during the program operation and the program inhibit operation, the memory array 40 can be operated with the same principle as the memory array 30 during the program operation and the program inhibit operation as shown in FIG. 9 .

Consequently, the memory cells 400 1,1 to 400 1,N , . . . , and 400 M,1 to 400 M,N of the memory array 40 can all be disposed in the same deep doped region. Since different memory pages MP 1 to MPM in the memory array 40 are disposed in one deep doped region, the spacing rules between deep doped regions will no longer be used to limit the circuit area of the memory array 40 , and the circuit area of the memory array 20 can be reduced significantly.

In summary, according to the memory arrays provided by the embodiments of the present invention, the memory cells of different memory pages in a memory array can all be disposed in the same deep doped region. Since different memory pages in the memory array are disposed in one deep doped region, the spacing rules between deep doped regions will no longer be used to limit the circuit area of the memory array, and the circuit area of the memory array can be reduced significantly.

Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.