Integration of lead zirconium titanate thin films for high density ferroelectric random access memory

Abstract

Interests are being focused on types of nonvolatile memories such as ferroelectric random access memory (FRAM), phase change random access memory, or magnetoresistance random access memory due to their distinct memory properties such as excellent write performance which conventional nonvolatile memories do not possess. Among these types of nonvolatile memories, FRAM whose cell structure and operation are almost identical to dynamic random access memory (DRAM) can ideally realize cell size and speed of DRAM. Thus FRAM is the most appropriate candidate for future universal memory where all memory functions are performed with a single chip solution. Due to the poor ferroelectric properties of downscaled ultrathin lead zirconium titanate (PZT) capacitors as well as technical issues such as hydrogen and plasma related degradation arising from embedding ferroelectric metal-insulator-metal capacitors into conventional complementary metal oxide semiconductor processes, current FRAM still falls far below its ideally attainable cell size and performance. In this paper, based upon PZT capacitor, current mass-productive one pass transistor and one storage capacitor (1T1C), capacitor over bit line (COB) cell technologies are introduced upon which cell size of 0.937 μm2 at 250 nm minimum feature size technology node has been realized. And then, most recent 1T1C, COB cell technologies are discussed from which cell size of 0.27 μm2 at 150 nm minimum feature size technology node has been realized, and finally future three dimensional capacitor technologies for the FRAM with cell size of less than 0.08 μm2 beyond 100 nm minimum feature size technology node are suggested. © 2006 American Institute of Physics.