Non-volatile memory with reduced charge fluence
View Patent ↗A method including performing a program/erase cycle on a first non-volatile memory (NVM) bit of an integrated circuit using a first fluence, wherein the first NVM bit has a first transconductance is provided. The method further includes performing a program/erase cycle on a second NVM bit of the integrated circuit using a second fluence, wherein the second NVM bit has a second transconductance, and wherein the first transconductance is greater than the second transconductance and the second fluence is greater than the first fluence.
1. A method, comprising:
performing a program/erase cycle on a first non-volatile memory (NVM) bit of an integrated circuit using a first fluence, wherein the first NVM bit has a first transconductance; and
performing a program/erase cycle on a second NVM bit of the integrated circuit using a second fluence, wherein the second NVM bit has a second transconductance, and wherein the first transconductance is greater than the second transconductance and the second fluence is greater than the first fluence, wherein the first NVM bit and the second NVM bit are coupled in parallel so that the threshold window for the first and second NVM bits is reduced
2. The method of claim 1 , wherein the step of performing a program/erase cycle on the first NVM bit is further characterized by the first NVM bit comprising a plurality of NVM cells coupled in parallel for reading.
3. The method of claim 2 , wherein the step of performing a program/erase cycle on the first NVM bit is further characterized by separately programming each of the plurality of NVM cells.
4. The method of claim 3 , wherein the step of performing a program/erase cycle on the first NVM bit is further characterized by simultaneously erasing the plurality of NVM cells.
5. The method of claim 4 , further comprising separately reading each NVM cell of the plurality of NVM cells.
6. The method of claim 2 , wherein the step of performing the program/erase cycle on the first NVM bit is further characterized by the plurality of NVM cells comprising transistors having drains permanently connected together.
7. The method of claim 1 , wherein
the step of performing a program/erase cycle on the first NVM bit is further characterized by the first NVM bit comprising a first transistor having a first channel width;
the step of performing a program/erase cycle on the second NVM bit is further characterized by the second NVM bit comprising a second transistor having a second channel width; and
the first channel width is greater than the second channel width.
8. The method of claim 7 , wherein:
the step of performing a program/erase cycle on the first NVM bit is further characterized by the first transistor having a first nanocrystal storage layer; and
the step of performing a program/erase cycle on the second NVM bit is further characterized by the second transistor having a second nanocrystal storage layer.
9. The method of claim 8 , wherein the step of performing a program/erase cycle on the first NVM bit is further characterized by removing electrons from the first nanocrystal layer to achieve erasure of the first NVM bit and adding electrons to the first nanocrystal layer to achieve programming of the first NVM bit.
10. The method of claim 1 , wherein:
the step of performing a program/erase cycle on the first NVM bit is further characterized by the first NVM bit being in a first sector in a first memory array; and
the step of performing a program/erase cycle on the second NVM bit is further characterized by the second NVM bit being in a second sector in the first memory array.
11. A memory, comprising:
a first non-volatile memory (NVM) bit having a first transconductance;
a second NVM bit having a second transconductance less than the first transconductance; wherein the first NVM bit and the second NVM bit are coupled in parallel so that the threshold window for the first and second NVM bits is reduced; and
a program/erase circuit coupled to the first and second NVM bits that performs a program/erase cycle using a first fluence on the first NVM bit and a second fluence on the second NVM bit, wherein the first fluence is less than the second fluence.
12. The memory of claim 11 , wherein the first NVM bit comprises a plurality of NVM cells coupled to a plurality of bit lines of a memory array, further comprising:
a column select circuit coupled to the plurality of bit lines; and
an address decoder coupled to the column select circuit that receives an address signal and couples the plurality of bit lines together in response to the address signal indicating that the first NVM bit is being selected.
13. The memory of claim 12 , further comprising an address match circuit, coupled to the address decoder, that determines if the address decoder is to couple the plurality of bit lines together.
14. The memory of claim 13 , further comprising a sense amplifier, wherein the address match circuit directs the column select circuit to couple only one of the plurality of bit lines to the sense amplifier in response to a configuration signal.
15. The memory of claim 12 , wherein each of the plurality of NVM cells of the firat NVM bit have a first channel width and the second NVM bit comprises a transistor having the second channel width.
16. The memory of claim 11 , wherein the first NVM bit is characterized as having less trap-up per program/erase cycle than the second NVM bit.
17. The memory of claim 11 , wherein the first and second NVM bits comprise nanocrystals for charge storage.
18. The memory of claim 11 , wherein the first NVM bit is in a first sector of a first memory array and the second NVM bit is in a second sector of the first memory array.
19. A memory, comprising:
a first non-volatile memory (NVM) bit having a first transconductance;
a second NVM bit having a second transconductance less than the first transconductance; wherein the first NVM bit and the second NVM bit are coupled in parallel so that the threshold window for the first and second NVM bits is reduced; and
means for performing an event on each of the first and second NVM bits wherein the event comprises one of a group consisting of program and erase and wherein the event causes more trap up in the second NVM bit than the first NVM bit.
20. The memory of claim 19 , wherein the means for performing is for performing the erase in which electrons are removed from nanocrystals used for charge storage in the first and second NVM bits and more trap-up in the second NVM bit is caused by more fluence in erasing the second NVM than in erasing the first NVM bit.