Closed-form analytical solutions for avalanche breakdown quantities in high-voltage diffused junctions

By using a particularly chosen exponential profile and depletion approximation, the analytical expressions for avalanche breakdown critical quantities in high-voltage diffused junctions are derived. The analytical results are in good agreement with those of the full numerical method obtained by Temple and Adler and much more accurate than the values obtained by using the one-sided abrupt junction and the linearly-graded junction depletion approximations.     

Negative resistance in p-n junctions under avalanche breakdown conditions, part I

A one-dimensional, small-signal analysis of the space-charge region of a p-n junction in which avalanche occurs uniformly is presented. The impedance is found to have a negative real part. The impedance is Well represented by a parallel connection of the depletion layer capacitance, an inductance, and a negative resistance. The admittance of the latter two is proportional to the bias current. The magnitude of the negativeQis below ten. The negative resistance is due to an intrinsic instability in the avalanching electron-hole plasma. A discussion of the instability and a traveling-wave tube-like amplification is given.     

Negative resistance in p-n junctions under avalanche breakdown conditions, part II

The small-signal impedance of the space-charge region of p-n junctions under avalanche breakdown conditions is calculated using reasonably realistic dependences of electron and hole ionization rates and drift velocities upon electric field. Two structures are analyzed: one is p⁺νn⁺structure which has a fairly uniform distribution of avalanche multiplication, and the other is a singly diffused junction which is a hybrid of an abrupt and a linear graded junction. Both structures show negative resistance when the transit time of carriers becomes appreciable. A computer program was evolved which requires, as input, the impurity profile and field dependences of ionization rates and drift velocities. The program first calculates the dc field and electron and hole currents and then solves the ac small-signal problem. Both the ac small-signal impedance and theQof the diode are calculated.     

High-voltage planar p-n junctions

A concentric ring junction has been devised to prevent surface breakdown of a planar junction. By properly choosing the spacing between the main junction and the ring, the ring junction acts like a voltage divider at the surface. In addition, the ring junction minimizes the effect of the junction curvature at the periphery of a planar junction. Devices fabricated with three such rings showed breakdown voltages of 2000 and 3200 volts on n-type silicon with impurity concentrations 6.5 × 10¹³and 2.5 × 10¹³cm^-3, respectively. That the structure operated as proposed was corroborated by comparison of the reverse leakage current with a one parameter fit to a theoretically calculated current obtained from the approximated volume of the space charge regions. These results together with the photo response measurements indicate that the field-limiting ring junction can be used successfully to obtain high-voltage planar p-n junctions.     

Input power induced thermal effects related to transition time between avalanche and second breakdown in p-n silicon junctions

The dependence of the transition time between avalanche breakdown and second breakdown on the input power to a p-n junction was studied both in transistor configurations and in single junction structures. The transition time is observed to decrease with increased input power. The dependence of the transition time on the input power indicates a constant input energy for which second breakdown occurs. The effect the thermal heating may have on conduction in a junction is also considered. The constant energy required and related occurrence of melt channels in the junction region are felt to support the thermal hypothesis.     

Infrared observation of the breakdown behavior of high-voltage p-n junctions and p-n-p structures in silicon

It is shown that the radiation emitted at breakdown from large-area high-voltage p-n junctions and from p-n-p structures can be detected using an infrared image converter in conjunction with a three-stage image intensifier. In the example given, the patterns of the emitted light consist of segments of circles and spirals, indicating an inhomogeneous breakdown across the area. The inhomogeneous breakdown is attributed to resistivity striations inside the silicon. The infrared radiation emitted from the p-n-p structure was more intense by approximately one order of magnitude than that of the p-n junction, the additional emission being recombination radiation generated in front of the forward-biased emitter.     

Effect of surface fields on the breakdown voltage of planar silicon p-n junctions

The effect of surface fields on the breakdown voltage of planar silicon diodes is studied experimentally and theoretically. It is shown that the breakdown voltage can be modulated over a very wide range by the application of an external surface field and that it tends to saturation at a maximum and at a minimum value as the gate voltage is varied in such a way as to deplete the lowly doped and highly doped sides of the junction, respectively. Both the high- and the low-voltage saturation of the breakdown voltage appear to be due to the formation of field-induced junctions which prevent further variation in the shape of the depletion region, and hence the breakdown voltage. Between these two extremes, the breakdown voltage is found to be approximately given byBV=mV_{G}+constant, where VGis the gate-to-substrate potential. The slopemapproaches unity for low substrate impurity concentrations and for small oxide thicknesses. Numerical solutions of the two-dimensional potential distribution problem give results which are in general agreement with the above experimental observations.     

D.C. characteristics of silicon p-n junctions at avalanche breakdown including selfheating

The computation of the d.c. characteristics of a silicon p-n junction at breakdown is described in this paper. In the analysis, the effect of the dissipated electric power on the junction temperature is taken into account. The influences of the thermal resistance, the ambient temperature and the low-voltage reverse current on the d.c. characteristics at breakdown are shown. A formula describing the differential resistance of a junction at breakdown is derived. The temperature coefficients of voltage and current for the d.c. characteristics under consideration are derived as well. The coordinates of the points at which these coefficients change the sign are calculated and the influence of the thermal resistance, the ambient temperature and the low-voltage reverse current on the coordinates of these points is discussed.     

Breakdown walkout in planar p-n junctions

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.     

Avalanche breakdown of diffused silicon p-n junctions

Using impact ionization rates of Moll et al. [23] and Lee et al. [24], the avalanche breakdown voltages of diffused silicon p-n junctions were calculated by assuming an error function diffused impurity distribution. The theoretical results were verified experimentally with samples having breakdown voltages ranging from 100 volts to 9000 volts. Good agreement was found between the experimental breakdown voltages and those calculated with the data of Moll et al. This was seen even at high breakdown voltages where the ionizing fields were far lower than those for which ionization rates have been quoted.     

Properties of Be-implanted planar GaAs p-n junctions

The fabrication and characteristics of planar junctions in GaAs formed by Be ion implantation are discussed. The critical processing step is shown to be the use of a carefully deposited oxygen-free Si/SUB 3/N/SUB 4/ encapsulation during post-implantation annealing. Forward and reverse characteristics are presented for Be-implanted junctions formed by encapsulating with SiO/SUB 2/, Si/SUB x/O/SUB y/N/SUB z/, or Si/SUB 3/N/SUB 4/ layers prior to annealing at 900/spl deg/C. Junctions which exhibit leakage current density of ~2/spl times/10/SUP -7/ A/cm/SUP 2/ at 80 V reverse bias and breakdown voltage >200 V have been fabricated using RF-plasma deposited Si/SUB 3/N/SUB 4/ layers as the encapsulant.     

Thermal breakdown delay time in silicon p-n junctions

The geometric effects on the applied power dependence of the delay time preceding thermal breakdown in p-n junctions are predicted in terms of a linear heat-flow model and temperature-dependent reverse current. Measurements of the delay time on silicon planar p-n junctions of various areas are compared to the predictions and found to be in reasonable accord.     

Avalanche drift instability in planar passivated p-n junctions

Some factors which affect the breakdown voltage of planar passivatedp-njunctions and which influence breakdown drift instability are discussed. Experiments describing these phenomena are reported. An activation energy for walk-out recovery is given as approximately 0.35 eV. It is argued that charge motion parallel to and probably within the oxide-semiconductor interface is occurring. Experiments performed to verify this parallel charge drift are related which confirm the effect. Finally, evidence for a mobile negative species is given.     

Statistical delay of microplasma breakdown in GaP p-n junctions

The mechanism for switching on a microplasma in gallium phosphide p-n junctions is investigated. It is shown that changing the distribution function of the statistical breakdown delay with respect to distance makes it possible to determine the energy spectrum of deep levels localized in the microplasma channels. In these experimental studies, commercial gallium phosphide AL102 red-light LEDs were used. In the temperature range 100–380 K the influence of a number of energy levels was detected. In these diodes, deep levels were observed to have an unusually strong effect on the statistical breakdown delay when their charge states are changed by a fractional decrease in the voltage across the p-n junction. Fiz. Tekh. Poluprovodn. 33, 1345–1349 (November 1999)     

Microplasma breakdown in high-voltage p?n junctions

The voltage-breakdown characteristics of controlled-avalanche junctions demonstrate the presence of microplasmas whose bistable current-switching properties are described. Microplasmas are exhibited by both diffused and alloyed junctions. The relationship between the breakdown voltage of the first microplasma with temperature for 1500?1800 V junctions is similar to that reported for 7?128 V junctions.     

Temperature dependence of breakdown voltage in silicon abrupt p-n junctions

Temperature dependence of breakdown voltage in silicon abrupt p⁺-n junctions has been calculated using a modified Baraff theory [1]-[3] and measured experimentally from 77°K to 500°K, with substrate doping from 10¹⁵cm^-3to 10¹⁸cm^-3. Experimental data are in good agreement with the results of theoretical calculations. These results strongly substantiate the validity of the modified Baraff theory which has been pointed out by Sze and Crowell.     

A fast method of calculating avalanche breakdown voltage of semiconductor p-n junction

A fast method of calculating the avalanche breakdown voltage of semiconductor p-n junction is described. A simple technique of calculating the integral from the stored values of the integrand is illustrated for Silicon step junctions. This results in considerable saving of computational time.     

Zener and avalanche breakdown in As-implanted low-voltage Si n-p junctions

Implanted-diffused As layers in Si have been well-characterized and have been used in fabricating low-voltage n-p junctions. It is shown that these As layers form linearly graded junctions with a uniform B-doped background (ρ ≃ 0.006 Ω.cm). The grade constant of the As profile at the junction is known sufficiently well as a function of As dose, diffusion time, and temperature to allow quantitative use of existing tunneling and avalanche theories for the calculation of the reverse I-V curves. Following a verification of the calculated I-V curves and their temperature dependence as a function of grade constant, calculated curves are presented which correlate As implant dose and diffusion with junction breakdown voltage, breakdown impedance, and temperature coefficient of reverse voltage. The temperature coefficient is shown to change from negative to positive as the transition from tunneling to avalanche occurs. In addition, the relative importance of tunneling and multiplied-generation current as a function of current density is elucidated for any particular As layer grade constant.