No abstract

No abstract

Silicon nanowire field effect transistor sensors with SiO(2)/HfO(2) as the gate dielectric sensing surface are fabricated using a top down approach. These sensors are optimized for pH sensing with two key characteristics. First, the pH sensitivity is shown to be independent of buffer concentration. Second, the observed pH sensitivity is enhanced and is equal to the Nernst maximum sensitivity limit of 59 mV/pH with a corresponding subthreshold drain current change of ∼ 650%/pH. These two enhanced pH sensing characteristics are attributed to the use of HfO(2) as the sensing surface and an optimized fabrication process compatible with silicon processing technology.

We report on physical and electrical characterization of ultrathin (3–10nm) high-κHfO2 gate stacks deposited on Ge(100) by atomic-layer deposition. It is observed that uniform films of HfO2 can be deposited on Ge without significant interfacial growth. The lack of an interlayer enables quasiepitaxial growth of HfO2 on the Ge surface after wet chemical treatment whereas a nitrided interface (grown by thermal oxynitridation in ammonia) results in an amorphous HfO2. The stacks exhibit surprisingly good thermal stability, up to temperatures only 150°C below the melting point of Ge. In terms of electrical properties, HfO2 on Ge shows significantly reduced (up to 4 decades) gate leakage currents in the ultrathin regime of equivalent electrical thickness down to ∼1.4nm due to the high-dielectric constant of ∼23. Nitrided interface is observed to be important for good insulating properties of the stack.

We investigated the kinetics of charge detrapping in high-κ gate stacks fabricated with ultrathin HfO2 dielectric films grown by atomic layer deposition and a polycrystalline silicon gate electrode. It was observed that charge trapped after electron injection in the high-κ stack was unstable and slowly decayed over time. The decay does not follow a simple first-order exponential law suggesting complex detrapping mechanism(s), possibly involving more than one type of trap present in the stack. The detrapping rate was found to depend strongly on gate voltage, temperature, and light illumination.