By Topic

Electron Devices Meeting, 1970 International

Date 28-30 Oct. 1970

Filter Results

Displaying Results 1 - 25 of 138
  • [Front cover and table of contents]

    Page(s): c1
    Save to Project icon | Request Permissions | PDF file iconPDF (979 KB)  
    Freely Available from IEEE
  • Light-emitting diodes

    Page(s): 12 - 14
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (91 KB)  

    There is a growing interest in the development of devices that can efficiently convert electrical energy into light, particularly visible light. Of special interest is the possibility that such devices may be electrically and, hopefully, even physically compatible with existing electrical devices, particularly integrated circuits. Currently available semiconductor light-emitting diodes approach this ideal. For efficient electrical → optical power conversion at visible frequencies, the semiconductor bandgap must be ≳ 1.8 eV, which rules out silicon and even gallium arsenide. In this talk, the general physical guidelines developed recently for the selection of doped semiconductors that can produce efficient near-bandgap luminescence will be reviewed. In particular, attention will be given to the relative potentialities of direct and indirect gap semiconductors and the extent to which these potentialities have been realized, mainly in III-V compounds and their alloys. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Microwave transistors - From small signal to high power

    Page(s): 12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (80 KB)  

    New technologies have emerged in the development of high-frequency transistors, and some of the old technologies have been considerably refined and extended. New equipment and measurement techniques have been perfected that allow rapid and accurate device characterization at microwave frequencies. As a result of these developments, the list of small-signal and power transistors available for use as amplifiers and oscillators to frequencies as high as 8 GHz has grown dramatically in recent years. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Rates of energy loss by electrons in solids

    Page(s): 12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (81 KB)  

    The rate at which energetic electrons lose energy to a solid via phonons and various electronic excitations plays a key role in such devices as solid-state photomultipliers, Gunn oscillators, avalanche diodes, cold cathodes, negative-affinity emitters, and acoustoelectric amplifiers, and in such phenomena as electron mobility, current saturation, and dielectric breakdown. A semiclassical model for the rate of loss of energy will be presented that is conceptually simple and graphic and that provides a valid quantitative guide for analysis. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance limits of bucket-brigade shift registers

    Page(s): 16 - 18
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (168 KB)  

    Studies of integrated IGFET bucket-brigade charge-transfer shift registers show that charge-transfer efficiency depends on IGFET output conductance, on magnitude of the transferred charge, and on clock voltage waveforms. In a 64-device register, charge-transfer efficiency of 99.8% per stage has been achieved at 1 MHz with a structure that minimizes IGFET output conductance. An important operating constraint is that the register must always carry a small circulating charge, otherwise charge transfer efficiency is greatly impaired. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Compact COS/MOS 256-bit random-access memory

    Page(s): 16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (85 KB)  

    A 256-bit static COS/MOS (complementary symmetry MOS) random-access memory chip has been fabricated with a 30 mil2cell and a chip size (0.110 × 0.128 inch) basically comparable to that of single-channel static MOS memories. Complementary MOS layouts with both p-channel and n-channel MOS transistors have generally required more area than single-channel MOS devices. A favorable memory cell circuit symmetry and modified layout rules can, however, permit the advantages of a complementary technology without areal penalty. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A 1024 bit N-channel MOS read-write memory chip

    Page(s): 16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (85 KB)  

    The Cogar 1024-bit MOS read-write memory chip is fully decoded, has a 100-ns access time, and is only 125 × 125 mils in size. This small size is made possible by the use of a 4-device memory cell and dimensional tolerances as small as 0.15 mil. The chips are fabricated from a 2.25-inch-diameter wafer that consists of a low-resistivity p-type substrate covered by a high-resistivity p-type epitaxial layer. This epitaxial layer provides not only for precise control of surface impurity concentration, but also provides means for a reach-through type of gate protection device. In addition, the low-resistivity substrate tends to minimize unwanted noise voltages due to pulse currents flowing through the substrate. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The MNOS bipolar transistor

    Page(s): 18
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (81 KB)  

    Charge-storage effects in MNOS structures have been investigated with regard to application in programmable read-only semiconductor memories utilizing the insulated-gate field-effect transistor as the active circuit component. The surface-potential dependence of both base and collector current in a suitably designed bipolar transistor, however, can be combined with the charge storage effects of the MNOS gate to provide a programmable bipolar transistor. By means of the stored charge in an MNOS gate over the emitter-base junction of a bipolar transistor, the base surface beneath the gate can be stably shifted from accumulation to inversion with the inversion layer beneath the gate of a suitably designed transistor acting as an efficient extention of the emitter junction. This results in a designable increase in collector cmrent at fixed forward emitter-base bias. Examples of both n-p-n and p-n-p transistors performing in this manner are given. The principal advantage of this type of device over conventional MNOS devices lies in its inherent higher transconductance. Performance comparisons between the read speed of conventional MNOS devices and that of the bipolar MNOS will be presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A trim bipolar charge storage memory

    Page(s): 18
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (81 KB)  

    A bipolar stored-charge memory cell will be described that consists of a single transistor with an MOS capacitor between emitter and ground. The transistor serves as a bidirectional switch allowing charge to selectively flow into or out of the storage capacitor. The cell requires periodic refreshing of the stored charge since electron-hole generation in the depletion region of the emitter tends to discharge it. However, because most of the charge is stored on the MOS capacitor, storage times of several seconds at room temperature should be expected. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A high-reliability plastic encapsulation for transistors

    Page(s): 20
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (89 KB)  

    A description of the design of a reliable, inexpensive, transfer-molded, silicone plastic transistor encapsulation is given. The materials selection was based upon the needs of reliability testing and suitability for requisite manufacturing processes. The dimensions of the plastic encapsulation, which replaces the TO-18 hermetically sealed metal-glass encapsulation, were determined by analyzing the stresses at the interface of the plastic encapsulant and the metal leads. The physical and mathematical models of the lead seal and a mechanical-failure criterion were used to determine the integrity of the lead seal. Results of mechanical and electrical evaluations are given to demonstrate the capability and reliability of the encapsulation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Arsenic emitter for gold-doped shallow-junction transistors

    Page(s): 20
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (90 KB)  

    It has been known that a reduction in the isolation capacitance and the storage time of an integrated bipolar transistor could be achieved by the introduction of gold in silicon. The process of gold introduction requires a final high-temperature cycle after emitter diffusion to reactivate the gold, which in turn makes it difficult to control the base width of a shallow-junction transistor with a phosphorus emitter. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Arsenic emitter high-performance transistor

    Page(s): 22
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (120 KB)  

    Silicon n-p-n transistors with gain-bandwidth, fT, of 13 GHz and current gain β of 280 have been fabricated by using arsenic capsule diffusion to produce the emitter region. Devices have been fabricated With the same epitaxial and diffusion processes up to emitter diffusion, and then emitter regions were formed by either arsenic or phosphorus diffusion. For the same mask geometry, devices with an arsenic emitter showed 60% improvement over those with a phosphorus emitter in fTand β under the condition of similar intrinsic base sheet resistance, which is a measure of total impurity in the active base region under the emitter. For comparable base width, arsenic-emitter devices show much higher collector punch-through voltage. Experimental data on a comparison between various dc and ac characteristics and circuit speed with these two types of devices will be presented as will data on various emitter geometries with arsenic emitter diffusion. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detailed modeling of nonlinear distortion in bipolar transistors

    Page(s): 22 - 24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (200 KB)  

    Nonlinear distortion in a bipolar transistor amplifier is calculated using a new compact model to represent the transistor. Nonlinear distortion is particularly important for transistor applications in long-haul analog communications repeater amplifiers. In that case the internally generated output powers at the second and third harmonics relative to the power at the fundamental frequency should be as small as possible, The compact model is capable of predicting with any desired sensitivity the experimentally observed second- and third-order distortion coefficients. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Physical origins of burst noise in transistors

    Page(s): 24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (94 KB)  

    Burst noise in bipolar junction transistors has been attributed to leakage current through crystallographic defects located in proximity to the base-emitter junction. The current is modulated by the random occupancy of a single flaw located near the defects. The previous theoretical work has only described defects located in the emitter-base space-charge region away from the surface. This paper extends the noise model to include the regions close to the surface where the high incidence of defects makes burst noise sources more likely. Experimental data is presented on typical devices, some of which were gold doped. The proposed model is in agreement with all experimental measurements. The data, in conjunction with the noise model, is used to determine physical parameters such as capture cross sections and energy level of the flaw which correlate well with published values. The investigation suggests several techniques which are useful in reducing the incidence ot burst noise. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Some factors influencing 1/F noise in silicon planar transistors

    Page(s): 24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (94 KB)  

    1/fnoise in n-p-n silicon planar transistors is shown to have two sources. One of these sources is associated with the recombination centres that determine low-current hFEand are situated where the emitter-base depletion region meets the silicon-silicon-dioxide interface. The same relationship between hFEand 1/fnoise is established for, (1) batches of production transistors with suitable processing, (2) devices in which the gain and 1/fare changed as a result of biasing a gate electrode over the emitter base oxide, and (3) devices in which the gain and 1/fnoise have been changed as a result of emitter-base avalanche stressing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimization of transistor neutron tolerance

    Page(s): 26
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (83 KB)  

    A detailed analysis has been performed for neutron-irradiated p+-n junction diodes to determine the influence of the recombination model, doping profile, and operating point on the minority carrier recombination throughout the devices. The calculations give the recombination rate, carrier distributions, and electric field in the various device regions and separate the device current into its space-charge-region recombination and diffusion components. At low bias levels, where space charge region recombination dominates, both the calculations and experimental data on irradiated devices show that the recombination current can be greatly reduced by employing heavy base doping in the vicinity of the junction. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High average power pulse generation in CW Nd:YAG

    Page(s): 28
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (78 KB)  

    Over 1000 watts of CW power has been generated with a multisection krypton arc lamp pumped Nd:YAG laser. Applying electro-optical Q-switching techniques that are unique to this high-gain CW device, pulses with peak powers greater than 5 megawatts have been produced at repetition frequencies of 400G pps. The dynamics of the rapid pulse growth results in a pseudo-mode-locked pulse envelope. Through a combination of Q switching and injection of seed pulses from a CW mode-locked Nd:YAG laser, peak powers of several hundred megawatts can be produced, with average powers greater than 500 watts. Application of this output format to high average power second-harmonic generation will be discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A 100-watt average power liquid laser

    Page(s): 28 - 30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (168 KB)  

    A pulsed liquid laser operating at 5 pulses per second and an average power of 100 watts has been developed. The laser material consists of Nd-trifluoroacetate dissolved in POCl3acidified with ZrCl4. The concentration of Nd is 0.3 mole/liter The fluorescence lifetime of the solution is in excess of 300 µsec and the dynamic loss in the circulating medium is 0.0035 cm-1. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Erbium laser materials and systems

    Page(s): 28
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (78 KB)  

    The results of an erbium laser materials and systems investigation are presented. Studies include optical properties and laser system parameters of yb3+-sensitized YAG:Er and glass: Er and their relationship to the optimization of laser host materials and laser system design. Optical properties discussed include absorption, fluorescence, and excitation spectra, as well as fluorescence lifetimes; these properties will be related to actual laser performance data. An experiment was conducted to determine the emission cross section for YAG:Er, Yb, which is compared with cross sections of ruby and YAG:Nd. the Q-switched laser performance of the optimized compositions of YAG:Er,Yb and glass:Er, Yb are presented. Projections are made concerning the potential of the erbium laser in reducing the eye-safety problem and the expectation for applications to target range finding and illumination. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Diode-pumped YAG-Nd lasers

    Page(s): 30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (90 KB)  

    In this paper we report on work, using both cooled(sim77degK)and room-temperature incoherent GaAs1-xPxdiodes (obtained from Texas Instruments, Inc.), that demonstrates the feasibility of obtaining room-temperature CW operation of a YAG:Nd laser with presently available diodes. Measurements have been made with diodes emitting at 0.81 µm that are coupled to a YAG:Nd laser rod by a semi-elliptic cylindrical reflector. The reasons for this choice of coupling geometry will be discussed together with some other possibilities. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High-repetition-rate low-energy-pulse pumping of Nd-YAG lasers

    Page(s): 30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (90 KB)  

    Several recent studies have compared xenon and krypton discharges as optical pumps for CW Nd:YAG lasers. Above 50 joules, xenon is reportedly more efficient, and in the 15-50 joule range, krypton has been found to be more efficient. We have compared the laser efficiencies obtained using pulsed discharges in xenon and krypton at pressures up to 3000 Torr and repetition rates up to 350 pps. Input energy was varied over the 0.5 to 10 joule range at pulse widths of 20, 100, 240, and 500 microseconds. Effects of pump-bore diameters of 1.7, 2.3, and 5 mm and bore lengths of 30 and 50 mm were evaluated using a quartz envelope--moly ribbon foil seal tube design and a keep-alive mode of operation. A 6-inch-diameter gold-coated sphere was used as the cavity for laser rods of 3, 5, and 6.25 mm diameter. Efficiency was found to increase with gas pressure up to 3000 Torr. A maximum efficiency of 1.8% was obtained with krypton, which, depending on current density, was generally found to be 20% better than xenon. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Compensation for thermally induced birefringence in Nd-YAG lasers

    Page(s): 32
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (56 KB)  

    Thermal gradients that occur ill optically pumped YAG lasers result in bifocusing and birefringence. Methods to compensate for these induced aberrations have been suggested by Osterink and Foster. We have employed one of these methods that uses 90° rotation of the polarization. A crystal quartz rotator is placed between two pumped YAG rods so that the part of the mode that is radially polarized in the first rod is tangentially polarized in the second. The experimental set-up consists of a 50-cm convex mirror, two flat ended 3 × 50 mm krypton-pumped Nd: YAG rods separated by the rotator and a 10-m concave output mirror. The dimensions of the cavity are such that the TEM00mode should be apertured by the rod diameters when their focal lengths are 50 cm each. The rod-lamp combinations have this focal length at a 1.5 kW input to each. The multimode output power at this input is 9 W with long-radius mirrors. With the above set-up the output is 6 W in the TEM00mode. Insertion of a Brewster plate results in a linearly polarized mode in a ratio of 30:1 with less than 10% loss of power. Variation of the power input to one of the lamps results in an increase in the crossed polarization component and shows that compensation is achieved by this technique. It is possible to obtain similar polarizations without loss of power for the multimode operation. In an attempt to reduce focusing effects, the pump radiation was filtered with a pass band from 0.66 to 1.33 microns, but no significant change in the observed focal length of the YAG rod was observed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Techniques for fabricating gun electron injection semiconductor hybrid amplification devices

    Page(s): 34 - 36
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (168 KB)  

    Techniques have been developed that have made it possible to combine (in a miniature ceramic-metal vacuum envelope) a boron-implanted silicon diode, which serves as the output end of the structure, and a grid-controlled planar electron-tube structure, which is used as an electron-beam input source. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Picosecond-rise-time pulse applications of electron-beam-semiconductor devices

    Page(s): 34
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (84 KB)  

    In this paper basic analytical and experimental results are presented for ultrafast amplifiers, pulse generators, and pulse-detection systems in which a modulated electron beam controls the current in a semiconductor target. The lumped target amplifier is discussed frst, and results are summarized from a simplified analysis of target response to step-function beam excitation. The optimization of amplifier rise time and peak output level is illustrated and predicted target gain, rise time, peak output current, voltage, and power capabilities are shown graphically. The feasibility of a subnanosecond-rise-time pulse amplifier with a peak output power of 100 kW is shown. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computer simulation of large-signal characteristics of an electron-beam-excited semiconductor

    Page(s): 36 - 38
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (176 KB)  

    A computer program has been developed that permits detailed calculations of the electrical output and corresponding internal effects in the operation of an electron-beam-excited semiconductor device. The computer analysis complements the simple analytical treatment of target operation previously available, and makes possible the evaluation of high-level target operation with any or all of the following effects: (1) arbitrary input waveform, (2) arbitrary pair creating profile, (3) operation where electrons and holes contribute significantly to the target current, (4) non-constant carrier velocity, (5) arbitrary profile of drift region doping, and (6) non-resistive load impedance. These effects often appear in practical target design, especially for high-current, picosecond-rise-time amplifying devices. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.