Speaking at the IMEC Technology Forum here on Thursday (May 23) Keyvan Esfarjani, vice president technology & manufacturing at Intel and co-CEO of IMFT revealed some of the thinking behind IMFT’s 3-D NAND strategy.
An industry-wide transition for the nonvolatile NAND flash memory technology from memory cells in a 2-D array to strings of NAND transistors integrated monolithically in the vertical direction is now anticipated. These 3-D memories are expected to be arranged as a 2-D array of vertical semiconductor channels with many levels of gate-all-around (GAA) structures forming the multiple voltage level memory cell transistors.
Toshiba is leading the charge towards 3-D NAND processes with its p-BiCS (pipe-shaped Bit Cost Scalable) technology, which it has presented at numerous learned conferences over several years. Towards the end of last year Toshiba announced that it had 16-layer devices based on a 50-nm diameter vertical channel. Samples are due this year and volume production in 2015. Toshiba’s p-BiC technology arranges the transistor string in a U-shape.
But Esfarjani, while acknowledging there is a scaling limit for 2-D NAND flash, indicated in one of his slides that 2-D NAND flash can scale to two more nodes at about 15- and 10-nm. The slide showed that the first 3-D NAND generation is likely to be brought up alongside that 15-nm 2-D node. Esfarjani added that 16 layer NAND flash ICs will not be enough to provide an economic benefit. "You need 64 or at least 32 layers," he said.
Click on image to enlarge.
The transition to 3-D NAND according to Keyvan Esfarjani, vice president technology & manufacturing at Intel and co-CEO of IM Flash Technologies LLC, in a presentation at the IMEC Technology Forum.
Esfarjani said that the planar floating-gate high-K metal gate cell introduced by IMFT at the 20-nm node to replace a "wrap-around" cell used at 34- and 25-nm would scale further. IMFT introduced a 128-Gbit NAND flash memory at the 20-nm during 2012 (see Intel, Micron offer 128-Gbit NAND flash memory) leveraging its HKMG experience gained in logic circuits.
Among the techniques being used to extend the 2-D NAND flash generation are the addition of a nitride film and nanodots, said Esfarjani. "But a working cell is not enough. You also need high endurance of 10^5 cycles and higher," he told the IMEC Technology Forum audience.
"The transition to 3-D is not limited by lithography," Esfarjani said indicating that a 40-nm diameter semiconductor channel was about right. The challenges are in metrology, in finding materials that can withstand the temperatures of multilayer semiconductor processing and the high aspect ratio etch to drive in the vertical channel. This has to be controlled at an angle of taper of precisely 89.8 degrees, Esfarjani said.
"The NAND market will be with us and growing for years to come. 2-D NAND continues with the planar floating gate cell and 3-D NAND will carry us beyond the 10-nm limit," concluded Esfarjani.
During questions at the end of his talk Esfarjani was asked if he could share information about the materials that IMFT is using to create the dielectric and gates within the vertical channel. He confirmed that IMFT is working on new materials but said that some information is too good to share.
New materials are being developed for 3-D NAND, Esfarjani told the IMEC Technology Forum.