High-Permittivity Perovskite Thin Films for Dynamic Random-Access Memories
- PDF / 1,353,215 Bytes
- 7 Pages / 576 x 777.6 pts Page_size
- 1 Downloads / 200 Views
46
achievable. (See Reference 2 for a review and comparison of materials.) As the required area has become larger than the available surface area per cell, the solution has been to produce innovative cell structures that result in an increased effective area, such as by producing highaspect-ratio trenches in the silicon or via structures such as those shown in Figure I.3 This approach has the advantage that the simple ONO dielectric is retained, but there is great concern that the structural complexity increase necessitated by 4-Gbit devices will result in an unmanufacturable product. As a result, since the mid-1980s, there has been an increasing effort to investigate an alternative approach, namely the replacement of the ONO dielectric with an alternative dielectric having a sub-
stantially higher permittivity. While there is a natural tendency in the semiconductor industry to shy away from the incorporation of "exotic" materials into devices, the much larger permittivities available in alternative materials appear increasingly attractive as integration densities increase. Two materials have been seriously considered as alternative DRAM dielectrics: Ta2O5 and (Ba,Sr)TiO3 (abbreviated as "BST"). Ta2O5 has the advantage of being a simple binary oxide and has a relative permittivity of about 25, but capacitors that include Ta2Os films have tended to suffer from leakage currents somewhat higher than ONO, and the lower limit of thickness is larger than that of ONO. Thus reported storage capacitances for Ta 2 O 5 are around 10-20 fF/yum2.2 If utilized at all, it appears that Ta2O5 will only be appropriate for one generation of DRAMs. On the other hand, BST, while being a multicomponent oxide, appears to offer a greater potential. It is put to use in DRAM capacitors as a polycrystalline thin film on metal or conductive metal oxide electrode in a metal-insulator-metal structure. This material is a ferroelectric with the perovskite structure and has been well-studied in bulk ceramic form where the measured permittivities are well into the thousands. The utilization of the BaTiO3-SrTiO3 solid solution allows the Curie point (ferroelectric-paraelectric phase-transition temperature) to be shifted from that of pure BaTiO3 at 130°C to a r o u n d room t e m p e r a t u r e for Bao.7Sro.3Ti03, thus achieving the maximum permittivity around the operating temperature. 4 However it should be pointed out immediately that the permittivities are not as high in the case of thin films. This issue is discussed in the next section. Finally the volatilities of the solid solution's components are low compared to Pb-based ferroelectric materials,
Table I: The Road Map of DRAM Technology (Adapted From Reference 2).
16 Mbit 64 Mbit 256 Mbit 1 Gbit 4 Gbit
Minimum Feature Size (fim)
C/Area (fF/ M m 2 )
0.60 0.35 0.25 0.18 0.15
25 30 55 100 140
Capacitor Area (Atm2)
Operating Voltage (V)
Year* 1 Million Devices
1.10 0.70 0.35 0.20 0.10
3.3 3.3 2.2 1.6 1.1
1992 1995 1998 2001 2004
*Year in which 1 million devices were/are projected
Data Loading...
