Method for simulating long-term performance of a non-volatile memory by exposing the non-volatile memory to heavy-ion radiation
View Patent ↗Testing a non volatile memory by exposing the non volatile memory to particle radiation (e.g. xenon ions) to emulate memory cell damage due to data state changing events of a non volatile memory cell. After the exposing, the memory cells are subjected to tests and the results of the tests are used to develop reliability indications of the non volatile memory. Integrated circuits with non volatile memories of the same design are provided. Reliability representations of the integrated circuits can be made with respect to a number of data state charging events based on the exposure and subsequent tests.
1. A method of developing a design of a non volatile memory comprising:
exposing a non volatile memory made according to a first design to particle radiation of a species having an atomic mass of 50 or greater for sufficient time and of sufficient energy to emulate memory cell dielectric damage due to read/write cycles of a plurality of the non volatile memory cells of the non volatile memory and for the species to pass through the non volatile memory to emulate memory cell dielectric damage due to program/erase cycles of a plurality of non volatile memory cells of the non volatile memory;
performing tests on the non volatile memory made according to the first design after the exposing to generate results of the tests; and
determining a reliability specification of the non volatile memory made according to the first design using the results of the tests, wherein the reliability specification is based on at least 100,000 program/erase cycles.
2. The method of claim 1 further comprising:
changing the design from the first design to a second design;
exposing a non volatile memory made according to the second design to particle radiation to emulate memory cell dielectric damage due to program/erase cycles of a plurality of non volatile memory cells of the non volatile memory made according to the second design;
performing tests on the non volatile memory made according to the second design after the exposing to generate results of the tests; and
determining the reliability specification of the non volatile memory made according to the second design using the results of the tests.
3. The method of claim 2 wherein the performing tests on the non volatile memory made according to the second design includes storing a charge in each of the plurality of non volatile memory cells and measuring a voltage threshold of each of the plurality of non-volatile memory cells after the charge is stored.
4. The method of claim 3 wherein performing tests on the non volatile memory made according to the second design includes measuring the voltage threshold for a second time of each of the plurality of memory cells after the measuring.
5. The method of claim 1 wherein the performing tests includes performing operational life tests on the non volatile memory.
6. The method of claim 1 wherein the exposing the non volatile memory includes exposing the non volatile memory to xenon ion particle radiation.
7. The method of claim 6 further comprising:
generating the reliability specification for the non volatile memory based on the determining, wherein the reliability specification is given with respect to a number of data state changing events, wherein the number corresponds to characteristics of the exposing, wherein the characteristics of the exposing include a species dosage of the xenon particle radiation.
8. The method of claim 7 further comprising:
providing a plurality of integrated circuits with a non volatile memory based on the first design;
representing the plurality of integrated circuits with a non volatile memory as meeting the reliability specification.
9. The method of claim 1 wherein the performing tests include performing program disturb tests on the non volatile memory.
10. The method of claim 1 wherein the performing tests include performing read disturb tests on the non volatile memory.
11. The method of claim 1 wherein the performing tests include performing sector disturb tests on the non volatile memory.
12. The method of claim 1 wherein the exposing includes exposing to emulate tunnel dielectric damage due to data state changing events of the plurality of non volatile memory cells.
13. The method of claim 12 wherein each of the plurality of non volatile memory cells is characterized as a 1 T floating gate memory cell that includes a floating gate and a tunnel dielectric between the floating gate and a substrate.
14. A method of developing non volatile memory, the method comprising:
exposing a non volatile memory to particle radiation of a species having an atomic mass of 50 or greater for sufficient time and of sufficient energy to emulate memory cell dielectric damage due to read/write cycles of a plurality of the non volatile memory cells of the non volatile memory and for the species to pass through the non volatile memory, wherein the non volatile memory including a plurality of non volatile memory cells and each non volatile memory cell of the plurality of non volatile memory cells includes a charge storage structure;
measuring a voltage threshold of each of the plurality of non volatile memory cells after the exposing; and
determining a reliability specification of the non volatile memory using results of the measuring, wherein the reliability specification is based on a number of read/write cycles of at least 100,000, wherein the number corresponds to characteristics of the exposing, wherein the characteristics of the exposing include species dosage of a particle radiation.
15. The method of claim 14 further wherein the non volatile memory is of a first design, the method further comprising:
providing a plurality of integrated circuits with a non volatile memory based on the first design;
representing the plurality of integrated circuits with a non volatile memory as meeting the reliability specification.
16. The method of claim 14 wherein the exposing the non volatile memory includes exposing the non volatile memory to xenon ion particle radiation.
17. A method of developing non volatile memory, the method comprising:
exposing a non volatile memory to particle radiation of a species having an atomic mass of 50 or greater for sufficient time and of sufficient energy to emulate memory cell dielectric damage due to read/write cycles of a plurality of the non volatile memory cells of the non volatile memory and for the species to pass through the non volatile memory, the non volatile memory being of a first design, each volatile memory cell of the plurality of non volatile memory cells including charge storage structure and a dielectric;
storing a charge in each of the plurality of memory cells after the exposing;
measuring a voltage threshold of each of the plurality of non volatile memory cells after the storing;
determining a reliability specification of the non volatile memory using results of the measuring, wherein the reliability specification is given as at least five years with respect to a number of read/write cycles, wherein the number corresponds to characteristics of the exposing, wherein the characteristics of the exposing include species dosage;
providing a plurality of integrated circuits with a non volatile memory based on the first design; and
representing the plurality of integrated circuits with a non volatile memory as meeting the reliability specification.