A self-forming nanostructure—a wave-ordered structure with a controllable
period (20–180 nm)—results from the off-normal bombardment of
amorphous silicon layers by low-energy (∼ 1–10 keV)
nitrogen ions. The nanostructure has been modified by reactive-ion etching in
plasma to form a periodic nanomask on the surface of the channel region of a
metal–oxide–semiconductor field-effect transistor (MOSFET). Implantation of
arsenic ions through the nanomask followed by the technological steps
completing the fabrication of the MOSFET resulted in a periodically doped
channel field-effect transistor (PDCFET), which can be considered as a
chain of short-channel MOSFETs with a common gate. Having worse
subthreshold characteristics, PDCFETs show greater drain current and
transconductance than to MOSFETs without a periodically doped channel. This
improvement in device performance is attributed to the fact that the
channel length is cut by the length of high-conductivity doped areas in the
channel and that the voltage is distributed between the areas, depressing the
scaling rules for short-channel MOSFETs and allowing the channel to
be less doped between the areas, thus keeping drift mobility high.
The discovery of ferroelectric properties in hafnium oxide has brought back the interest in the ferroelectric non-volatile memory as a possible alternative for low power consumption electronic memories. As far as real hafnium oxide-based materials have defects like oxygen vacancies, their presence might affect the ferroelectric properties due to oxygen atom movements during repolarization processes. In this work, the transport experiments are combined with the modeling to study evolution of the oxygen vacancy concentration during the endurance and to determine the optimal defect density for a higher residual polarization in lanthanum-doped hafnium oxide.
Atomic‐layer deposition (ALD) technique in combination with in vacuo X‐ray photoelectron spectroscopy (XPS) analysis has been successfully employed to obtain fully ALD‐grown planar TiN/HfO2/TiN metal–insulator–metal structures for resistive random access memory (ReRAM) memory elements. In vacuo XPS analysis of ALD‐grown TiN/HfO2/TiN stacks reveals the presence of the ultrathin oxidized layers consisting of TiON (∼0.5 nm) and TiO2 (∼0.6 nm) at the bottom TiN/HfO2 interface (i); the nonoxidized TiN at the top HfO2/TiN interface (ii); the oxygen deficiency in the HfO2 layer does not exceed the XPS detection limit (iii). Electroformed ALD TiN/HfO2/TiN stacks reveal both conventional bipolar and complementary types of resistive switching. In the complementary resistive switching regime, each programming sequence is terminated by a reset operation, leaving the TiN/HfO2/TiN stack in a high‐resistance state. The observed feature can avoid detrimental leaky paths during successive reading operation, which is useful in the passive ReRAM arrays without a selector element. The bipolar regime of resistive switching is found to reveal the gradual character of the SET and RESET switching processes. Long‐term potentiation and depression tests performed on ALD‐grown TiN/HfO2/TiN stacks indicate that they can be used as electronic synapse devices for the implementation of emerging neuromorphic computation systems.
Nature of traps responsible for the memory effect in Si3N4 still remains the subject matter of much discussion. Based on our quantum chemical simulation results, Si–Si bonds can be identified as traps for electrons and holes with localization energies falling within the ranges of Wte=1.2−1.7 eV and Wth=0.9−1.4 eV. Within the multiphonon trap ionization model, our experimental data on Si3N4 conductivity have allowed us to evaluate the thermal ionization energies of electron and hole traps in Si3N4 as Wte=Wth=1.4 eV. The same value of 1.4 eV was obtained as half the Stokes shift of the 2.4 eV green photoluminescence line observed in Si3N4 films under excitation with 5.2 eV. Thus, the data obtained in the present study strongly suggest that Si–Si bonds are responsible for localization of electrons and holes in Si3N4.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.