IMPROVEMENT OF STRUCTURAL INSTABILITY OF THE ION-SENSITIVE FIELD-EFFECT TRANSISTOR (ISFET)

Three causes involved in the instability of the ISFET are proposed in this study. First, it is ascertained that hydroxyl group resident at the surface of the Si3N4 film or in the electrolyte solution is most active and subject to gain or loss of electrons. This is one of the main causes for ISFET structural instability. Secondly, the stability of the pH-sensitive FET varies with deposition conditions in the fabrication process of the ISFET. This proves to be another cause of ISFET instability. Thirdly, the pH of the measured solution varies with the measuring process and time, contributing to the instability, but is not a cause of the instability of the pH-ISFET itself. We utilized the technique of readjusting and controlling the ratio of hydroxyl groups to amine groups to enhance the stability of the ISFET. Our techniques to improve stability characteristics proved to be effective in practice.

EFFECT OF TOTAL HYDROGEN CONTENT IN SILICON-NITRIDE SENSITIVE FILM ON PERFORMANCE OF ISFET

Quantitative determinations of the hydrogen content and its profile in silicon nitride sensitive films by the method of resonant nuclear reaction have been carried out. At a deposition temperature of 825-degrees-C, hydrogen exists in an LPCVD silicon nitride sensitive film and the hydrogen content on its surface is in the range (8-16) x 10(21) cm-3, depending on the different deposition processes used. This hydrogen content is larger than the (2-3) x 10(21) cm-3 in its interior part, which is homogeneous. Meanwhile, we observe separate peaks for the chemical bonding configurations of Si-H and N-H bonds, indicated by the infrared absorption bands Si-O (1106 cm-1), N-H (1200 cm-1), Si-H-3 (2258 cm-1) and N-H-2 (3349 cm-1), respectively. The worse linear range of the ISFET is caused by the presence of oxygen on the surface of the silicon nitride sensitive film. The existence of chemical bonding configurations of Si-H, N-H and N-Si on its surfaces is favourable for its pH response.

AN ANALYSIS OF CHARGE STATES FOR LOCAL VIBRATIONAL-MODES BY OXYGEN DEFECTS IN SEMI-INSULATING GAAS

The high-resolution spectral measurements for new local vibrational modes near 714 cm-1 due to the oxygen defect in semi-insulating GaAs are analyzed on the basis of a model calculation by self-consistent bond orbital approach. Two charge states of oxygen atom with 1 and 2 extra electrons are assigned to be responsible for these local modes. The observed frequencies are explained by the properties of Ga-O-1 and Ga-O-2 bonds and the calculated cohesive energy indicates that the O-2 state is stable. The results are in good agreement with the kinetic analysis.

Ab initio calculations for the neutral and charged O vacancy in sapphire

The energetics, lattice relaxation, and the defect-induced states of st single O vacancy in alpha-Al2O3 are studied by means of supercell total-energy calculations using a first-principles method based on density-functional theory. The supercell model with 120 atoms in a hexagonal lattice is sufficiently large to give realistic results for an isolated single vacancy (square). Self-consistent calculations are performed for each assumed configuration of lattice relaxation involving the nearest-neighbor Al atoms and the next-nearest-neighbor O atoms of the vacancy site. Total-energy data thus accumulated are used to construct an energy hypersurface. A theoretical zero-temperature vacancy formation energy of 5.83 eV is obtained. Our results show a large relaxation of Al (O) atoms away from the vacancy site by about 16% (8%) of the original Al-square (O-square) distances. The relaxation of the neighboring Al atoms has a much weaker energy dependence than the O atoms. The O vacancy introduces a deep and doubly occupied defect level, or an F center in the gap, and three unoccupied defect levels near the conduction band edge, the positions of the latter are sensitive to the degree of relaxation. The defect state wave functions are found to be not so localized, but extend up to the boundary of the supercell. Defect-induced levels are also found in the valence-band region below the O 2s and the O 2p bands. Also investigated is the case of a singly occupied defect level (an F+ center). This is done by reducing both the total number of electrons in the supercell and the background positive charge by one electron in the self-consistent electronic structure calculations. The optical transitions between the occupied and excited states of the: F and F+ centers are also investigated and found to be anisotropic in agreement with optical data.

Photoluminescence investigation of charge build-up process in the emitter of a double-barrier resonant tunneling structure

Charge build-up process in the emitter of a double-barrier resonant tunneling structure is studied by using photoluminescence spectroscopy. Clear evidence is obtained that the charge accumulation in the emitter keeps almost constant with bias voltages in the resonant regime, while it increases remarkably with bias voltages beyond resonant regime. The optical results are in good agreement with the electrical measurement. It is demonstrated that the band gap renormalization plays a certain rob in the experiment.

Verification of EL2 electronic absorption effect on charge transfer in semi-insulating GaAs

The electronic absorption of EL2 centers has been clarified to be related to the electron acid hole photoionizations, and the transition from its ground state to metastable state, respectively. Under an illumination with a selected photon energy in the near infrared region, these three processes with different optical cross sections will show different kinetics against the illumination time. It has recently been shown that the photosensitivity (measured under 1.25 eV illumination) of the local vibrational mode absorption induced by some deep defect centers in SI-GaAs is a consequence of the electron and hole photoionizations of EL2. This paper directly measures the kinetics of the electronic transition associated with EL2 under 1.25 eV illumination, which implies the expected charge transfer among different charge states of the EL2 center. A calculation based on a simple rate equation model is in good agreement with the experimental results.

An asymmetric MOSFET-C band-pass filter with on-chip charge pump auto-tuning

An asymmetric MOSFET-C band-pass filter(BPF)with on chip charge pump auto-tuning is presented.It is implemented in UMC (United Manufacturing Corporation)0.18μm CMOS process technology. The filter system with auto-tuning uses a master-slave technique for continuous tuning in which the charge pump OUtputs 2.663 V, much higher than the power supply voltage, to improve the linearity of the filter. The main filter with third order low-pass and second order high-pass properties is an asymmetric band-pass filter with bandwidth of 2.730-5.340 MHz. The in-band third order harmonic input intercept point(HP3) is 16.621 dBm,wim 50 Ω as the source impedance. The input referred noise iS about 47.455μVrms. The main filter dissipates 3.528 mW while the auto-tuning system dissipates 2.412 mW from a 1.8 V power supply. The filter with the auto-tuning system occupies 0.592 mm~2 and it can be utilized in GPS (global positioning system)and Bluetooth systems.

Simulation of a Monolithic Integrated CMOS Preamplifier for Neural Recordings

A monolithic integrated CMOS preamplifier is presented for neural recording applications. Two AC-coupied capacitors are used to eliminate the large and random DC offsets existing in the electrode-electrolyte interface. Diode-connected nMOS transistors with a negative voltage between the gate and source are candidates for the large resistors necessary for the preamplifier. A novel analysis is given to determine the noise power spectral density. Simulation results show that the two-stage CMOS preamplifier in a closed-loop capacitive feedback configuration provides an AC in-band gain of 38.8dB,a DC gain of 0,and an input-referred noise of 277nVmax, integrated from 0. 1Hz to 1kHz. The preamplifier can eliminate the DC offset voltage and has low input-referred noise by novel circuit configuration and theoretical analysis.

First charge collection and position-precision data on the medium-resistivity silicon strip detectors before and after neutron irradiation up to 2x10(14) n/cm(2)

Test strip detectors of 125 mu m, 500 mu m, and 1 mm pitches with about 1 cm(2) areas have been made on medium-resistivity silicon wafers (1.3 and 2.7 k Ohm cm). Detectors of 500 mu m pitch have been tested for charge collection and position precision before and after neutron irradiation (up to 2 x 10(14) n/cm(2)) using 820 and 1030 nm laser lights with different beam-spot sizes. It has been found that for a bias of 250 V a strip detector made of 1.3 k Ohm cm (300 mu m thick) can be fully depleted before and after an irradiation of 2 x 10(14) n/cm(2). For a 500 mu m pitch strip detector made of 2.7 k Ohm cm tested with an 1030 nm laser light with 200 mu m spot size, the position reconstruction error is about 14 mu m before irradiation, and 17 mu m after about 1.7 x 10(13) n/cm(2) irradiation. We demonstrated in this work that medium resistivity silicon strip detectors can work just as well as the traditional high-resistivity ones, but with higher radiation tolerance. We also tested charge sharing and position reconstruction using a 1030 nm wavelength (300 mu m absorption length in Si at RT) laser, which provides a simulation of MIP particles in high-physics experiments in terms of charge collection and position reconstruction, (C) 1999 Elsevier Science B.V. All rights reserved.