J. Blok

Electronic noise in semiconductors

Synopsis A critical survey of some theories concerning the fluctuations in the number of current carriers in ohmic semiconductors is given and the discrepancies between them are discussed. In this paper the continuity equation is used as a basic equation to derive the fluctuations. This equation provides a good understanding of the different processes, which give rise to fluctuations: the diffusion term is responsible for the thermal noise; the generation and recombination terms give rise to true resistance fluctuations; the divergence of the conduction current causes shot noise in a narrower sence. The latter two phenomena are not exclusive as is often thought but additive. Further it is shown that there are no principal difficulties to extend this theory to more stochastic variables. By means of the Fokker-Planck formalism a matrix equation is derived from which the variances and covariances of the stochastic variables are readily calculated if all generation and recombination probabilities are known. The equation appears to be a generalisation of an expression derived recently by Burgess 7 ) and is also applicable for non-thermal equilibrium conditions e.g. for photoconductors.

A method for the computation of plural and multiple scattering corrections

Synopsis A method is given to compute the angular distribution of scattered particles starting from a function for the differential cross section; this angular distribution depends on single, plural and multiple scattering. The thickness of the scatterer has been taken about equal to the mean free path of the particles inside the scatterer. An application is given for the case of scattering of thermal neutrons by a thin sheet of paraffin. In the appendix we discuss the differential cross section in the laboratory system for the scattering of thermal neutrons by paraffin.

Measurements of noise and response to modulated light of cadmiumsulphide single crystals

Synopsis The noise of photoconducting single crystals of silver-activated and undoped cadmiumsulphide has been investigated in the range 1 c/s - 2 mc/s. The observed spectra are not of the 1/ν type but show several characteristic features, which depend on the experimental conditions, including the light intensity, the wavelength of the light and the temperature. In particular, entirely different spectra have been obtained with irradiation in the fundamental absorption band and for light of wavelengths beyond the absorption limit (5100 ). Measurements of the response of the crystals to modulated light for modulation frequencies in the range 7 c/s - 10 kc/s led to response curves, which showed much resemblance with the noise spectra under all circumstances. So a large part of the noise must be considered as photo-induced noise. The other part is ascribed to spontaneous fluctuations in the rates of recombination and trapping. The results can adequately be explained with the theoretical calculations, given in a previous paper 6).

Electronic noise in photoconducting insulators

Synopsis The noise caused by fluctuations in the number of free electrons is computed for photoconductors of the recombination type. First a discussion of the fluctuations in the incident photon stream is given and the resulting photo-induced noise is calculated. It is shown that these fluctuations do not entirely determine the ultimate sensitivity; besides optical generations of free electrons other transitions also contribute to the noise. In a model without localized levels, such as traps etc., the total noise is twice as large as the photo-induced noise, for all frequencies. Next a discussion is given for a photo-conductor according to the model of Klasens-Schon. The spectrum of the photo-induced noise does not get the same shape as the total noise spectrum. In both spectra the same relaxation times occur but the partition over the various terms is different. Finally approximations are made for excitation of electrons from activator levels and from the valence band at low and high light intensities.

Measurements on current noise in carbon resistors and in thermistors

Synopsis Measurements are reported on carbon resistors and some related types (frequency range 2 c/s — 1.2 mc/s) and on thermistors (2 c/s — 50 kc/s). For carbon resistors we found that the noise spectrum in the whole range can be represented by S(v) = Ai 2 /v 1 , with a ≈ 1, where i is the current and v the frequency. This is in contrast to measurements of Campbell and Chipman 5) and not in accordance with theories of Marfarlane and Petritz 18) 19) . Measurements on thermistors showed that the current noise is extremely weak as compared with carbon resistors. The spectrum showed deviations from the 1/ v dependence. The spectral density in the “1/ v part” of the spectrum is proportional to i 1.2 . This is in agreement with measurements of Brophy 11) , published recently. Some explanations are suggested for the other part of this spectrum.

Carrier density fluctuations in semiconductors and photoconductors with one kind of trapping centers

Abstract Applying the general theory of carrier density fluctuations in semiconductors (Van Vliet), based on the formalism of irreversible thermodynamics, expressions for the matrix elements of carrier density fluctuations and for the relaxation times are given in the case of semiconductors with one kind of trapping- or recombination centers. These expressions have been approximated in some cases by formulae that can be verified experimentally, for instance trapping of minority carriers in deep lying centers (occuring in PbS photoconductors) and trapping of carriers at shallow traps in nearly intrinsic semiconductors. With the aid of the extended g-v theorem it is finally proved that the derived expressions for the noise in semiconductors hold for photoconductors only if the centers act as traps.

Gamma lines in the decay of 202T1

The gamma-radiation of Tl/sup 202/ was reinvestigated by means of a gamma-scintillation spectrometer. Two new gamma-lines were found viz. of 523 and 965 kev. An estimation was made for the transition-energy to the ground state of Hg/sup 202/ and a level scheme of Hg/sup 202/ was set up. The ratio of the energies of the second and first excited states was compared with the values of neighboring nuclei. (auth)

Current fluctuations in PbS cells

Synopsis Measurements are reported about the noise spectrum and the photoresponse of PbS cells (Kodak Ektron detectors) in the range 1 c/s – 20 kc/s. Analysing this spectrum we can attribute it partly to contact noise partly to generation-recombination noise. We were mainly interested in the generation-recombination noise. Measurements were performed about the dependence of the spectrum on the intensity of an additional irradiation and on the temperature. Moreover photoresponse curves were measured; these curves are in agreement with the noise spectra. Comparing the results with theories based on models for the photoconductivity mechanism we find that a good description of the experimental results can be given with the majority carrier model (Petritz). Some of our results however contradict the barrier model (Slater).

On the integral width of cascaded tuned amplifiers

Synopsis A formula is derived by which the overall integral bandwidth B of an amplifier containing a number of non inductively coupled LC circuits is expressed in the effective quality factors Q i of the separate resonance circuits. The quantity B can also be expressed in the integral widths B i of the separate circuits; for two or three stages very simple relations are found.

SCATTERING OF THERMAL NEUTRONS BY BOUND PROTONS AND DEUTERONS

Synopsis With a neutron howitzer and a Ra-Be source (400 mgr Ra) some experiments were performed about the angular distribution of thermal neutrons scattered in paraffin, water, heavy water, ice and heavy ice. By using a method previously described, some data about the differential cross section for the scattering processes have been found. The differential cross section of paraffin, water and ice differ much from spherical symmetrical functions; for heavy water and heavy ice the differences are much smaller. For paraffin the distribution computed after the model of Arley deviates from the experimental values; the experimental data suggest a more forward directed scattering. From the data for water and ice may be concluded that the binding of the proton in ice is stronger than in water. Interference effects were observed; especially in the curves for heavy water and heavy ice.