scispace - formally typeset
Search or ask a question

Showing papers on "Depletion region published in 1978"


Journal ArticleDOI
TL;DR: In this paper, it was shown that the 2kT current in double-heterostructure AlxGa1−xAs p−n junctions is primarily due to surface recombination at the junction perimeter.
Abstract: We show that the 2kT current in double‐heterostructure AlxGa1−xAs p‐n junctions is primarily due to surface recombination at the junction perimeter. The rate of surface recombination is evaluated by means of two luminescence experiments. Both experiments provide evidence that the rate of surface recombination increases with bias as exp(eV/2kT) and has the correct magnitude to account for the measured I‐V curves. It is shown theoretically that recombination at a depleted surface is proportional to exp(eV/2kT) at high bias. The 2kT behavior is a consequence of the nearly constant ratio of electron and hole densities at the surface. The nearly constant ratio is needed to maintain equality of surface and surface depletion layer charge. p‐n junction perimeters of cleaved, etched, and proton‐bombarded interfaces are evaluated and found to have similar rates of nonradiative recombination with the rate of surface recombination given by R=s0(np)1/2, where s0 is approximately 4×105 cm/sec.

256 citations


Journal ArticleDOI
T.H. Ning1
TL;DR: In this article, hot-electron emission from silicon into silicon dioxide is discussed and several mechanisms of importance in the emission process have been identified, such as Schottky lowering of the emission barrier, the scattering of hot electrons in the image-force potential well in the SiO2 layer, the tunneling of hot electron, and the effect of lattice temperature on electron heating.
Abstract: Recent progress in the study of hot-electron emission from silicon into silicon dioxide is discussed. Experimental techniques include avalanche injection using gated diodes and MOS capacitors, nonavalanche injection using IGFET structures with an underlying supply p - n junction, and optically induced injection using silicon-gate IGFET structures. IGFET structures allow the fields in the SiO2 layer and in the silicon depletion region to be varied independently. In addition, IGFET structures of reentrant geometry allow absolute emission probabilities of the hot electrons to be determined. Such absolute emission characteristics are useful not only for designing silicon devices but also for quantitative testing of theoretical models of the emission process. Several mechanisms of importance in the emission process have been identified. These are the Schottky lowering of the emission barrier, the scattering of hot electrons in the image-force potential well in the SiO2 layer, the tunneling of hot electrons, and the effect of lattice temperature on electron heating. There is also experimental evidence of the dependence of the hot-electron distribution on electric field gradient. At present, only phenomenological models based on the lucky-electron concept have been developed to the point where quantitative comparison with experimental results is possible. The essential features of these models are discussed.

157 citations


Journal ArticleDOI
TL;DR: Induced junction photodiodes have been fabricated and characterized in this paper, where the effect of high surface recombination is counteracted and an efficient photosensitivity is prediced at UV- and blue-wavelengths.
Abstract: The natural inversion layer occuring in thermally-oxidized p-type silicon is employed to make photodiodes. The induced np-junction implies high electric fields at the silicon surface and in the depletion layer, which aid the collection of photogenerated carriers with short penetration depth. Thus, the effect of high surface recombination is counteracted and an efficient photosensitivity is prediced at UV- and blue-wavelengths. Induced junction photodiodes have been fabricated and characterized. The photosensitivity in the 200-500 nm range is comparable or better than what is commercially available in UV-enhanced silicon photodiodes.

88 citations


Journal ArticleDOI
TL;DR: In this article, a model of metal-semiconductor ohmic contact theory is proposed and a simple approximate analytical expression for the specific resistivity is derived, which is essentially proportional to NB−1, NB being the semiconductor substrate impurity concentration.
Abstract: Most papers covering metal-semiconductor ohmic contact theory which have been published up to date consider systems with homogeneous impurity concentration in the semiconductor. However, there are techniques of ohmic contact formation on nondegenerate semiconductor where only a very shallow surface layer is impurity enriched. In this paper a model of such contacts is proposed and a simple approximate analytical expression for the specific resistivity is derived. If the impurity concentration in the surface layer is very high, the contact specific resistivity is essentially proportional to NB−1, NB being the semiconductor substrate impurity concentration. To make a good ohmic contact, it is sufficient that the width of the heavily doped surface layer be equal to the equilibrium contact depletion region width. Any further enlargement of the enriched layer practically does not influence the total sample resistance due to the dominant share of the semiconductor body resistance. Experimental results confirm these conclusions qualitatively.

63 citations


Journal ArticleDOI
TL;DR: In this paper, a two-beam (one pump-one probe) spectroscopic method was used to detect the presence of carrier recombination centers in liquid junction solar cells.
Abstract: Photocurrent spectra obtained by a two-beam (one pump--one probe) spectroscopic method on the semiconductor electrode of liquid junction solar cells can vary with the pump irradiance. In cells with n-type CdS, CdSe, CdTe, and GaAs photoanodes and chalogenide anion solutions this irradiance dependence results from and sensitively detects the presence of carrier recombination centers. With semiconductors showing no detectable recombination centers by this technique, cells with external solar to electrical conversion efficiencies of 8 to 9 percent have been made. Classical single beam photocurrent spectroscopy reveals that poor short-wavelength response in semiconductor liquid junction solar cells is due to surface or near surface recombination centers and resembles p-n junction solar cells in this respect. Lowered long-wavelength response is associated with shrinkage of the depletion region of imperfect and overdoped semiconductors.

50 citations


Journal ArticleDOI
TL;DR: In this article, the shortcircuit current generated by the electron beam of a scanning electron microscope when it impinges on the N-P junction of a solar cell is known to be dependent on the configuration used to investigate the cell's response, and the situation for one specific configuration is analyzed.
Abstract: The short-circuit current generated by the electron beam of a scanning electron microscope when it impinges on the N-P junction of a solar cell is known to be dependent on the configuration used to investigate the cell's response, and the situation for one specific configuration is analyzed. This configuration is the case in which the highly collimated electron beam strikes the edge of a planar junction a variable distance away from the edge of the depletion layer. An earlier treatment is generalized to encompass the ohmic contact at the back surface. The analysis employing Fourier and Wiener-Hopf techniques shows that it is impractical to determine the bulk diffusion length of a solar cell by a SEM used in the studied configuration unless the ohmic contact is partially removed.

46 citations


Journal ArticleDOI
TL;DR: In this paper, the flat-band potential of the n- and p-InP single crystal electrode in contact with indifferent electrolyte solution was determined, showing that a depletion region is present over a certain voltage range.

40 citations


Patent
03 Aug 1978
TL;DR: In this article, an integrated-circuit MOS capacitor has a first doped semiconductive region forming one of its plates, and this first semiconductor region is electrically isolated from the semiconductor substrate by a second DSA region, which is of opposite conductivity type to and intervenes between the substrate and the first SSA.
Abstract: An integrated-circuit MOS capacitor has a first doped semiconductive region forming one of its plates. This first semiconductor region is electrically isolated from the semiconductor substrate by a second doped semiconductor region, which is of opposite conductivity type to and intervenes between the substrate and the first semiconductive region. The present invention resides in the combination of such an MOS capacitor and a circuit for maintaining the potential in the second semiconductor region of the MOS capacitor structure substantially the same as that appearing on the first semiconductor region for minimizing leakage current flow between them. This circuit also provides the leakage current flowing between the second semiconductor region and the substrate, so the leakage current need not flow from the first semiconductor region.

39 citations


Patent
22 Dec 1978
TL;DR: In this article, the authors proposed a method of reducing the switching time of certain semiconductor devices and particularly gain-operated semiconductor device by adjusting the energy level of the radiation source to provide the depth of maximum defect generation adjacent a blocking PN junction.
Abstract: A method of reducing the switching time of certain semiconductor devices and particularly gain-operated semiconductor devices. The depth of maximum defect generation in a given type of semiconductor devices having a block PN junction is determined on irradiation with a given radiation source emitting particles with molecular weight of at least one (1), preferably protons or alpha particles; and the energy level of the radiation source adjusted to provide the depth of maximum defect generation adjacent a blocking PN junction of the type of semiconductor device. At least one semiconductor device of said given type of semiconductor device is positioned with a major surface thereof to be exposed to the adjusted radiation source, and thereafter irradiated with the adjusted radiation source to a given dosage level to reduce the switching time of the semiconductor device. Preferably, the semiconductor device is positioned and the energy level of the radiation source adjusted to irradiate through the major surface of the device closer the higher impurity region adjoining the blocking PN junction, and to provide the maximum defect generation in said higher impurity region adjoining the blocking PN junction but spaced from the PN junction beyond a depletion region formed at the PN junction on application of a specified blocking voltage across the PN junction of the semiconductor device.

37 citations


Journal ArticleDOI
TL;DR: In this paper, single grain-grain junction measurements of the currentvoltage characteristic of ZnO ceramic varistors are interpreted in terms of electron transmission through depletion layer barriers at the grain interface.
Abstract: Single grain‐grain junction measurements of the current‐voltage characteristic of ZnO ceramic varistors are interpreted in terms of electron transmission through depletion layer barriers at the ZnO grain interface The highly nonlinear varistor conduction process is associated with an abrupt thinning of the ZnO depletion layer due to valence‐band hole injection when the bottom of the ZnO conduction band in the grain interior drops below the top of the valence band at the grain surface

33 citations


Journal ArticleDOI
TL;DR: In this article, the authors show that the low effective mass leads to a small density of states and hence several subbands may be occupied even at moderate surface carrier densities, even in the case of Si.

Journal ArticleDOI
01 Mar 1978
TL;DR: Experiments with ion-selective membranes support the conjecture that noise is coincident with the increase of the electric resistance by the depletion process, and that when the current density is reduced, it reaches a critical value, below which the convection current changes from turbulent to laminar, and the noise disappears.
Abstract: Flicker noise of electric currents through ion-selective membranes is explained. It is attributed to the depletion of salt on one side of the membrane, which creates a thin layer of high resistance. Joule heating in this depletion layer and the ensuing temperature gradient, as well as the concentration gradient, give rise to buoyant forces which may create a turbulent convection current. The turbulence mixes the depletion layer so that the electric resistance fluctuates, and consequently the current flickers. Experiments with ion-selective membranes support this conjecture. They show that 1) Noise is coincident with the increase of the electric resistance by the depletion process. 2) When the current density is reduced, it reaches a critical value, below which the convection current changes from turbulent to laminar, and the noise disappears. 3) Noise reduces with temperature, because the expansion coefficient of water decreases with temperature, and its viscosity increases. 4) A non-ionic water-soluble polymer added to the compartment on the side of the depletion layer reduces the noise, by increasing the bulk viscosity of the solution. 5) Noise depends on the membrane's orientation in the gravitational field. 6) The convection-current in the depletion layer can be observed directly, using a laser-beam, by adding latex particles which create optical noise as they drift with the convection current across the beam. The optical noise is observed only coincidently with the current noise.

Journal ArticleDOI
TL;DR: In this paper, a series of experiments utilizing the C-t method, the MOS capacitor surface potential measurement, and the charge-coupled device (CCD) is described, which provides evidence for electrically generated electrons in the substrate of n-channel MOS structures.
Abstract: Impact‐ionization current during saturation mode operation is widely known in MOS devices. Although not noted in previous work, minority carriers also may be observed in the substrate, together with hole current. These minority carriers can degrade the MOS depletion layer lifetime, thus limiting the performance of MOS dynamic devices. A series of experiments utilizing the C‐t method, the MOS capacitor surface potential measurement, and the charge‐coupled device (CCD) is described, which provides evidence for electrically generated electrons in the substrate of n‐channel MOS structures.

Journal ArticleDOI
TL;DR: In this article, the depletion layer capacitance of a Schottky barrier on a GaAs FET film can only be measured in series with the resistance of the undepleted portion of the film.
Abstract: The depletion layer capacitance of a Schottky barrier on a GaAs FET film can only be measured in series with the resistance of the undepleted portion of the film. This inherent series resistance may be significant at all values of bias and causes large errors in C-V profile determinations. By treating the depletion layer capacitance and the series resistance as a distributed RC transmission line, it is possible to define an effective series resistance which can be related directly to the resistivity of the film. Using parameters typical of epitaxial films grown for GaAs FET applications, general criteria are developed for the profilability of these films. It is shown that, in general, films with small pinchoff voltages (i.e., films intended for low-noise FET applications)

Journal ArticleDOI
TL;DR: In this article, a study of neutron damage mechanisms in charge transfer devices has been performed with emphasis placed on investigation of dark current increases, where MOS capacitors were used to determine damage coefficients that are applicable to the radiation response of CCDs.
Abstract: A study of neutron damage mechanisms in charge transfer devices has been performed with emphasis placed on investigation of dark current increases. MOS capacitors were used to determine damage coefficients that are applicable to the radiation response of CCDs. Measurements of dark current density in CCDs were made following neutron bombardment. A unique value for generation-lifetime damage coefficient was determined (Kg = 7.0 × 106 n-sec/cm2) using MOS capacitors on both n- and p-type silicon substrates and this value was then used to calculate the expected change in dark current density with neutron fluence in a charge transfer device. The calculated value is in good agreement with the present experimental value of ~4×l0-11 nanoamps per neutron. A qualitative explanation is given to account for the nearly two-orders-of-magnitude difference between Kg and recombination-lifetime damage coefficients which is based in part on the nature of neutron damage in a depletion region. An explanation is also given to account for the more than six-orders-of-magnitude difference between Kg and storage-time damage coefficient. It is demonstrated that a moderate reduction in the operating temperature of a charge transfer device should result in substantial radiation tolerance in terms of neutron-induced increases in dark current.

Journal ArticleDOI
TL;DR: In this paper, a study of a new type varactor diode which consists of a semiconductor multilayer structure has been performed, and it is shown that such a varactor has much stronger nonlinearity in capacitance variation with applied bias voltage than does the conventional varactor due to the two-dimensional expansion of the depletion region.
Abstract: A study of a new type varactor diode which consists of a semiconductor multilayer structure has been performed. It is shown that such a varactor diode has much stronger nonlinearity in capacitance variation with applied bias voltage than does the conventional varactor due to the two‐dimensional expansion of the depletion region. First, the internal behaviors of this diode are investigated in detail by two‐dimensional computer simulation. Based on the physical insight obtained by the simulation, an approximate analytical expression for the voltage dependence of capacitance is derived for the purpose of practical design. Then, the proposed structure has been experimentally made from Si, whose capacitance‐voltage characteristics are compared with those of the conventional structure made from the same material. Those results substantiate the theoretical predictions.

Journal ArticleDOI
TL;DR: The resistivity of both chlorine-compensated and undoped cadmium telluride grown by the THM or programmed solution growth techniques was measured by dc and ac Hall effect and by C -V evaluations as mentioned in this paper.

Journal ArticleDOI
TL;DR: In this article, the collection efficiency of polycrystalline solar cells is analyzed and the extent of the low collection efficiency region surrounding grain boundaries is shown to be very much smaller in the depletion layer than in the bulk quasi-neutral regions of the cells.
Abstract: The collection efficiency of carriers optically generated within the depletion region of polycrystalline solar cells is analysed. The extent of the low collection efficiency region surrounding grain boundaries is shown to be very much smaller in the depletion layer than in the bulk quasi-neutral regions of the cells. For the optimum case where the grain boundaries are perpendicular to the edges of the depletion layer, it is shown that, near the centre of the layer, the low collection efficiency area extends only a fraction of an extrinsic Debye length from the grain boundary.

Journal ArticleDOI
TL;DR: In this paper, a theory has been developed to explain hot electron injection and trapping in the oxide and its effect on the breakdown voltage, and a comparison of results predicted by theory, with the experiments has also been carried out.
Abstract: Junction breakdown walkout in p-n junctions has been investigated in this paper. It has been shown that walkout is closely related to the avalanching in the junction. During the time the junction is subjected to the reverse breakdown, because of avalanching, hot electrons are generated in the depletion region. Some of the hot electrons have enough energy to cross the oxide-silicon barrier and to go into the conduction band of the oxide. The electrons are trapped in the traps and charge the oxide negatively, resulting in reduction of electric field intensity in the surface depletion region of the p-n junction. This results in an increase of the breakdown voltage. A theory has been developed to explain hot electron injection and trapping in the oxide and its effect on the breakdown voltage. A comparison of results predicted by theory, with the experiments has also been carried out.

Journal ArticleDOI
TL;DR: In this paper, the admittance of Al-anodic oxide-InSb MOS-capacitors has been studied in the temperature range from 5K to 300K.
Abstract: The admittance of Al-anodic oxide-InSb MOS-capacitors has been studied in the temperature range from 5K to 300K. Capacitance-voltage characteristics reproducibly show a frequency-dispersion in accumulation. By assuming an incompletely oxidized interlayer, it was possible to fit the capacitance-frequency and the conductance-frequency characteristics in accumulation by an appropriate equivalent circuit. Including the semiconductor space charge layer, an observed background of the conductance voltage characteristics can be explained. Furthermore a high frequency capacitance voltage analysis atT=5K was performed: Because of the relative low state density of the InSb conduction band in comparison to, e.g., Si, interface state analysis could be expanded into the bottom of the conduction band. A peak of interface state density around the conduction band edge was found.

Journal ArticleDOI
TL;DR: In this article, the effects of mechanical stress on the currentvoltage characteristics of metal-piezoelectric semiconductor interface are investigated for both Au-CdS and Au-GaP Schottky diodes.

Journal ArticleDOI
TL;DR: In this article, the authors obtained photocapacity spectra for Schottky barriers on natural semiconducting diamond using photon energies between 0.8 and 4.0 eV, indicating the presence of a deep level approximately 1.4 eV above the valence band.
Abstract: Photocapacity spectra have been obtained for Schottky barriers on natural semiconducting diamond using photon energies between 0.8 and 4.0 eV. The results indicate the presence of a deep level approximately 1.4 eV above the valence band. In reaching this conclusion independent evidence regarding the donor nature of this level has been used, and it has been assumed that electrons are excited from the valence band to ionised donors in the depletion layer in a two-step process via ionised acceptors. This is the reverse process to donor-acceptor pair recombination.

Patent
Albert W. Vinal1
03 May 1978
TL;DR: In this article, a high sensitivity, low noise, broad bandwidth, twin channel conduction Lorentz channel coupled semiconductive field sensor device is described, where magnetic fields may be used to create a LorentZ voltage in a region between the two conductive channels to vary the amount of current received at the two drains by utilizing the depletion width modulation effects of the Lorenz voltage upon the boundaries defining the conductive channel portions.
Abstract: A high sensitivity, low noise, broad bandwidth, twin channel conduction Lorentz channel coupled semiconductive field sensor device is described. The conductive channels are configured to create exceptionally narrow, undepleted conduction zones of approximately filamentary form. The filamentary conductive channels so formed are provided with a common source at one end of each channel and a separate drain at the other end thereof. The independent drains are spaced apart by a narrow area of semiconductive material. Magnetic fields may be utilized to create a Lorentz voltage in a region between the two conductive channels to vary the amount of current received at the two drains by utilizing the depletion width modulation effects of the Lorentz voltage upon the boundaries defining the conductive channel portions. Modulation of the depletion zone widths and depths along the channel sides effectively move the streams of carriers and the conductive channel areas to conduct more current to one drain more than another. This develops a differential drain current balance which can be utilized to provide an output signal. Width and length criteria for defining the filamentary channel structures are described for the ultimate desired configuration and size which are to be obtained. As noted, operation of the device is based upon Lorentz voltage modulation of the width and depth of the depletion zone boundaries defining the conductive channel. The Lorentz voltage is created in an area of semiconductive material coupling the two filamentary channels. An increased signal output is obtained by reducing the width of the filamentary channels to eliminate excess carriers normally found in wide channel devices and, further, by making the depletion zones as large a portion of the total channel widths as can be obtained.

Journal ArticleDOI
TL;DR: In this article, an admittance spectroscopy was carried out for the single-crystal CdS-Cu2S heterojunction in which the Cu2S was grown chemically on the Cd and S plane of a single crystal.
Abstract: The experiment of the admittance spectroscopy was carried out for the single‐crystal CdS‐Cu2S heterojunction in which the Cu2S was grown chemically on the Cd and S plane of CdS single crystal. The results showed that the concentration of the trapped electrons at the several trapping levels in the depletion layer of the former heterojunction decreases with increasing temperature. The opposite effect, however, was obtained for the latter heterojunction at the temperature below 210 K. The causes for the difference in the heterojunctions will be discussed in this paper.

Journal ArticleDOI
TL;DR: In this paper, the Sah, Noyce, and Shockley model of surface recombination is replaced with an alternative model in which recombination at a depleted surface will have 2kT charater.
Abstract: The interfaces forming the perimeters of AlxGa1−xAs p–n junctions, such as cleaved surfaces or etched surfaces are regions of intense nonradiative recombination. Almost all of the p–n junction 2kT current is due to recombination at these nonradiative perimeters. When an uncontacted sample is studied in photoluminescence, these interfaces produce broad nonradiative regions known as large dark spots. The rate of interface recombination can be quantitatively evaluated from measurement of either the 2kT current or the large dark spot line shape. The 2kT current is usually explained in terms of the Sah, Noyce, and Shockley model of p–n depletion layer recombination. We develop an alternative model in which we show that recombination at a depleted surface will have 2kT charater. The rate of surface recombination is R=s0(np)1/2=s0ni exp(eV/2kT) where s0?4×105 cm s−1 for x=0.08 active layers with etched surfaces. This form of the recombination rate follows naturally from the requirement that the interface remain ...

Patent
29 Dec 1978
TL;DR: The sensitivity of the threshold voltage in MOSFET devices to changes in substrate voltage may be reduced at a given temperature by the introduction of sufficiently deep energy level, low diffusivity impurities into the depletion region under the gate.
Abstract: The sensitivity of the threshold voltage in MOSFET devices to changes in substrate voltage may be reduced at a given temperature by the introduction of sufficiently deep energy level, low diffusivity impurities into the depletion region under the gate of the MOSFET.

Patent
Gabriel Lorimer Miller1
27 Dec 1978
TL;DR: In this article, a method of changing the geometry of p-n or isotype junctions in semiconductor crystal material by laser or electron beam melting through a portion of the junction is described.
Abstract: A method of changing the geometry of p-n or isotype junctions in semiconductor crystal material by laser or electron beam melting through a portion of the junction is described. Two adjoining regions are doped at different levels and a laser or electron beam melts through a portion of one region into the other region any desired depth. Upon cooling, the melt epitaxially recyrstallizes, producing a deformation in the otherwise planar junction that extends the more heavily doped region into the more lightly doped region. In the case of reverse-biased diodes, such as zener or avalanche diodes, this can be used to either increase the field in a portion of the depletion region, or push a portion of the junction into a more heavily doped region, or both, which reduces the breakdown voltage. Also, a method of controlling a pulsed laser or electron beam while continuously monitoring the breakdown voltage of the diode to obtain the desired breakdown voltage is described. The method is also useful for reducing the channel width of field effect transistors, and for making contact to buried doped regions and creating isolated surface regions.

Patent
30 May 1978
TL;DR: In this article, a source and a drain region of the p+ type are disposed in an n semiconductor layer to form a gate region between them, and the pn junction is normal to the main face.
Abstract: In the disclosed FAMOS semiconductor non-volatile memory a source and a drain region of the p+ type are disposed in an n semiconductor layer to form a gate region between them. The main face of the semiconductor layer is coated with a silicon dioxide film in which a polycrystalline silicon gate is buried to bridge the source and drain regions. An n+ type high doped semiconductor region is disposed in the semiconductor layer only under the silicon gate to form a pn junction with the drain region. Thus the pn junction is normal to the main face of the semiconductor layer.

Journal ArticleDOI
TL;DR: In this paper, the excitation spectra of the impurity and the exciton cathodoluminescence of semiconductors are analyzed with a detailed account of the surface space charge region (SCR) for various generation functions.

Patent
07 Mar 1978
TL;DR: In this paper, an improved transistor comprising an embedded electrode formed in a semiconductor substrate and having a high resistivity semiconductor region intervening between the embedded electrode and the substrate is selected to insure that this latter region is substantially pinched off in the operative state of this transistor.
Abstract: An improved transistor comprising an embedded electrode formed in a semiconductor substrate and having a high resistivity semiconductor region intervening between the embedded electrode and the substrate. The dimension and the impurity concentration of the high resistivity region are selected to insure that this latter region is substantially pinched off in the operative state of this transistor by the depletion layer growing from either the embedded electrode or the substrate, the width of said depletion layer varying in good faith without delay with the quick changes in the voltage of the embedded electrode. This provides an effective reduction mainly in the capacitance between the embedded electrode and the substrate, and also in the conductance in high-speed operation, which jointly bring about a high speed operation and a large driving ability. This transistor is extremely useful when adopted in a semiconductor integrated circuit.