Applied Materials' new VeritySEM 5i allows semiconductor makers to speed development of 3D FinFETs, and enable 3D flash memory cubes to be brought to the mass market.

The world's largest equipment supplier for semiconductors, displays and solar cells, Applied Materials Inc. (AM) claims to have uniquely solved the most pressing problems facing 3D flash chip stacks and 3D FinFETs for everybody. A paper on AM's technique was presented at the International Society for Optics and Photonics (SPIE) in San Jose, California.

3D NAND flash memory and 3D FinFET transistors have been enabled by critical-dimension scanning electron microscopy (CD-SEM). By applying techniques only recently proven effective in the research lab, AM's new VeritySEM 5i could not only allow semiconductor makers to speed development of 3D FinFETs, but also to finally enable 3D flash memory cubes to be brought to the mass market.

"The industry is transitioning from 2D to 3D," Ofer Adan, global product and technology manager for PDC, told EE Times. "So we have the logic guys transitioning to 3D transistors -- FinFETS -- and the memory guys are transitioning from 2D NAND flash to 3D NAND flash where you stack your NAND device vertically."

The big problem that manufacturers have with FinFETs is turning them all off at precisely the same voltage -- the only way you take advantage of the low leakage and subsequent cool running afforded by 3D FinFET transistors. To control leakage power, you need to control all the different aspects of the 3D gate -- the fin height and width, the gate height (sticking out of oxide) and the slope of the gate sides. Unfortunately, none of these parameters can be accurately measured with traditional 2D semiconductor metrology equipment.

The problem with 3D NAND devices is even worse, since the measurements to be made are inside the 3D cube, with only Samsung claiming to be manufacturing devices. Those are not commercially available, but are only being supplied to select cloud customers with no data publicly available on architecture of yields.

"The bottom line is you cannot measure 3D devices with 2D metrology. Somebody had to come up with a production-worthy high-volume in-line solution for 3D metrology, and that's what we have done."

This article was originally published on EE Times. To read the rest of the story, please click here.