A mathematical and experimental investigation of microcycle spectral estimates of semiconductor flicker noise

The experimental portion of this thesis tries to estimate the density of the power spectrum of very low frequency semiconductor noise, from 10^-6.3 cps to 1. cps with a greater accuracy than that achieved in previous similar attempts: it is concluded that the spectrum is 1/f^α with α approximately 1.3 over most of the frequency range, but appearing to have a value of about 1 in the lowest decade. The noise sources are, among others, the first stage circuits of a grounded input silicon epitaxial operational amplifier. This thesis also investigates a peculiar form of stationarity which seems to distinguish flicker noise from other semiconductor noise.

In order to decrease by an order of magnitude the pernicious effects of temperature drifts, semiconductor "aging", and possible mechanical failures associated with prolonged periods of data taking, 10 independent noise sources were time-multiplexed and their spectral estimates were subsequently averaged. If the sources have similar spectra, it is demonstrated that this reduces the necessary data-taking time by a factor of 10 for a given accuracy.

In view of the measured high temperature sensitivity of the noise sources, it was necessary to combine the passive attenuation of a special-material container with active control. The noise sources were placed in a copper-epoxy container of high heat capacity and medium heat conductivity, and that container was immersed in a temperature controlled circulating ethylene-glycol bath.

Other spectra of interest, estimated from data taken concurrently with the semiconductor noise data were the spectra of the bath's controlled temperature, the semiconductor surface temperature, and the power supply voltage amplitude fluctuations. A brief description of the equipment constructed to obtain the aforementioned data is included.

The analytical portion of this work is concerned with the following questions: what is the best final spectral density estimate given 10 statistically independent ones of varying quality and magnitude? How can the Blackman and Tukey algorithm which is used for spectral estimation in this work be improved upon? How can non-equidistant sampling reduce data processing cost? Should one try to remove common trands shared by supposedly statistically independent noise sources and, if so, what are the mathematical difficulties involved? What is a physically plausible mathematical model that can account for flicker noise and what are the mathematical implications on its statistical properties? Finally, the variance of the spectral estimate obtained through the Blackman/Tukey algorithm is analyzed in greater detail; the variance is shown to diverge for α ≥ 1 in an assumed power spectrum of k/|f|^α, unless the assumed spectrum is "truncated".

Characterization of hot-electron effects on flicker noise in III-V nitride based heterojunctions

We report experiments on hot-electron stressing in commercial III-V nitride based heterojunction fight-emitting diodes. Stressing currents ranging from 100 mA to 200 mA were used. Degradations in the device properties were investigated through detailed studies of the I-V characteristics, electroluminescence, Deep-Level Transient Fourier Spectroscopy and flicker noise. Our experimental data demonstrated significant distortions in the I-V characteristics. The room temperature electroluminescence of the devices exhibited 25% decrement in the peak emission intensity. Concentration of the deep-levels was examined by measuring the Deep-Level Transient Fourier Spectroscopy, which indicated an increase in the density of deep-traps from 2.7 x 10(13) cm(-3) to 4.21 x 10(13) cm(-3) at E-1 = E-C - 1.1eV. The result is consistent with our study of 1/f noise, which exhibited up to three orders of magnitude increase in the voltage noise power spectra. Our experiments show large increase in both the interface traps and deep-levels resulted from hot-carrier stressing.

Microscopic Mechanism of 1/f Noise in Graphene: Role of Energy Band Dispersion

A distinctive feature of single-layer graphene is the linearly dispersive energy bands, which in the case of multilayer graphene become parabolic. A simple electrical transport-based probe to differentiate between these two band structures will be immensely valuable, particularly when quantum Hall measurements are difficult, such as in chemically synthesized graphene nanoribbons. Here we show that the flicker noise, or the 1/f noise, in electrical resistance is a sensitive and robust probe to the band structure of graphene. At low temperatures, the dependence of noise magnitude on the carrier density was found to be opposite for the linear and parabolic bands. We explain our data with a comprehensive theoretical model that clarifies several puzzling issues concerning the microscopic origin of flicker noise in graphene field-effect transistors (GraFET).

Origin of 1/f noise in graphene produced for large-scale applications in electronics

The authors report a detailed investigation of the flicker noise (1/f noise) in graphene films obtained from chemical vapour deposition (CVD) and chemical reduction of graphene oxide. The authors find that in the case of polycrystalline graphene films grown by CVD, the grain boundaries and other structural defects are the dominant source of noise by acting as charged trap centres resulting in huge increase in noise as compared with that of exfoliated graphene. A study of the kinetics of defects in hydrazine-reduced graphene oxide (RGO) films as a function of the extent of reduction showed that for longer hydrazine treatment time strong localised crystal defects are introduced in RGO, whereas the RGO with shorter hydrazine treatment showed the presence of large number of mobile defects leading to higher noise amplitude.

Bulk-Induced 1/f Noise at the Surface of Three-Dimensional Topological Insulators

Slow intrinsic fluctuations of resistance, also known as the flicker noise or 1/f-noise, in the surface transport of strong topological insulators (TIs) is a poorly understood phenomenon. Here, we have systematically explored the 1/f-noise in field-effect transistors (FET) of mechanically exfoliated Bi1.6Sb0.4Te2Se TI films when transport occurs predominantly via the surface states. We find that the slow kinetics of the charge disorder within the bulk of the TI induces mobility fluctuations at the surface, providing a new source of intrinsic 1/f-noise that is unique to bulk TI systems. At small channel thickness, the noise magnitude can be extremely small, corresponding to the phenomenological Hooge parameter gamma(H) as low as approximate to 10(-4), but it increases rapidly when channel thickness exceeds similar to 1 mu m. From the temperature (T)-dependence of noise, which displayed sharp peaks at characteristic values of T, we identified generation-recombination processes from interband transitions within the TI bulk as the dominant source of the mobility fluctuations in surface transport. Our experiment not only establishes an intrinsic microscopic origin of noise in TI surface channels, but also reveals a unique spectroscopic information on the impurity bands that can be useful in bulk TI systems in general.

Low-frequency noise properties of GaN Schottky barriers deposited on intermediate temperature buffer layers

Metal-semiconductor-metal (MSM) structures were fabricated by RF-plasma-assisted MBE using different buffer layer structures. One type of buffer structure consists of an AlN high-temperature buffer layer (HTBL) and a GaN intermediate temperature buffer layer (ITBL), another buffer structure consists of just a single A IN HTBL. Systematic measurements in the flicker noise and deep level transient Fourier spectroscopy (DLTFS) measurements were used to characterize the defect properties in the films. Both the noise and DLTFS measurements indicate improved properties for devices fabricated with the use of ITBL and is attributed to the relaxation of residue strain in the epitaxial layer during growth process. (C) 2003 Elsevier Ltd. All rights reserved.

Low-frequency noise properties of GaN Schottky barriers deposited on intermediate temperature buffer layers

Metal-semiconductor-metal (MSM) structures were fabricated by RF-plasma-assisted MBE using different buffer layer structures. One type of buffer structure consists of an AlN high-temperature buffer layer (HTBL) and a GaN intermediate temperature buffer layer (ITBL), another buffer structure consists of just a single A IN HTBL. Systematic measurements in the flicker noise and deep level transient Fourier spectroscopy (DLTFS) measurements were used to characterize the defect properties in the films. Both the noise and DLTFS measurements indicate improved properties for devices fabricated with the use of ITBL and is attributed to the relaxation of residue strain in the epitaxial layer during growth process. (C) 2003 Elsevier Ltd. All rights reserved.

Inverse Class E amplifier and oscillator phase noise characteristics

This paper gives the first experimental characterisation of the phase noise response of the recently introduced Inverse Class E topology when operated as an amplifier and then as an oscillator. The results indicate that in amplifier and oscillator modes of operation conversion efficiencies of 64%, and 42% respectively are available, and that the excess PM noise added as a consequence of saturated Class E operation results in about a 10 dB increase in PM over that expected from a small-signal Class A amplifier operating at much lower efficiency. Inverse Class E phase transfer dependence on device drain bias and flicker noise are presented in order to show, respectively, that the Inverse Class E amplifier and oscillator follow the trends predicted by conventional phase noise theory. © 2007 EuMA.

Characterization and modeling of low-frequency noise in MOSFETs

For many years, RF and analog integrated circuits have been mainly developed using bipolar and compound semiconductor technologies due to their better performance. In the last years, the advance made in CMOS technology allowed analog and RF circuits to be built with such a technology, but the use of CMOS technology in RF application instead of bipolar technology has brought more issues in terms of noise. The noise cannot be completely eliminated and will therefore ultimately limit the accuracy of measurements and set a lower limit on how small signals can be detected and processed in an electronic circuit. One kind of noise which affects MOS transistors much more than bipolar ones is the low-frequency noise. In MOSFETs, low-frequency noise is mainly of two kinds: flicker or 1/f noise and random telegraph signal noise (RTS). The objective of this thesis is to characterize and to model the low-frequency noise by studying RTS and flicker noise under both constant and switched bias conditions. The effect of different biasing schemes on both RTS and flicker noise in time and frequency domain has been investigated.

Studies of high DC current induced degradation in III-V nitride based heterojunctions

We report experiments on high de current stressing in commercial III-V nitride based heterojunction light-emitting diodes. Stressing currents ranging from 100 mA to 200 mA were used. Degradations in the device properties were investigated through detailed studies of the current-voltage (I-V) characteristics, electroluminescence, deep-level transient Fourier spectroscopy and flicker noise. Our experimental data demonstrated significant distortions in the I-V characteristics subsequent to electrical stressing. The room temperature electro-luminescence of the devices exhibited a 25% decrement in the peak emission intensity. Concentration of the deep-levels was examined by deep-level transient Fourier spectroscopy, which indicated an increase in the density of deep-traps from 2.7 x 10(13) cm(-3) to 4.2 x 10(13) cm(-3) at E-1 = E-C - 1.1 eV. The result is consistent with our study of 1/f noise, which exhibited up to three orders of magnitude increase in the voltage noise power spectra. These traps are typically located at energy levels beyond the range that can be characterized by conventional techniques including DLTS. The two experiments, therefore, provide a more complete picture of trap generation due to high dc current stressing.

Investigation of Bulk and DTMOS triple-gate devices under 60 MeV proton irradiation

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Laser locking to the Hg-199 S-1(0)-P-3(0) clock transition with 5.4x10(-15)/root tau fractional frequency instability

With Hg-199 atoms confined in an optical lattice trap in the Lamb-Dicke regime, we obtain a spectral line at 265.6 nm for which the FWHM is similar to 15 Hz. Here we lock an ultrastable laser to this ultranarrow S-1(0) - P-3(0) clock transition and achieve a fractional frequency instability of 5.4 x 10(-15) / root tau for tau <= 400 s. The highly stable laser light used for the atom probing is derived from a 1062.6 nm fiber laser locked to an ultrastable optical cavity that exhibits a mean drift rate of -6.0 x 10(-17) s-(1) (-16.9 mHzs(-1) at 282 THz) over a six month period. A comparison between two such lasers locked to independent optical cavities shows a flicker noise limited fractional frequency instability of 4 x 10(-16) per cavity. (c) 2012 Optical Society of America

Crustal uplift rates of 4 GPS stations near Jakobshavn Isbrae (2006-2009)

We analyze 2006-2009 data from four continuous Global Positioning System (GPS) receivers located between 5 and 150 km from the glacier Jakobshavn Isbrae, West Greenland. The GPS stations were established on bedrock to determine the vertical crustal motion due to the unloading of ice from Jakobshavn Isbrae. All stations experienced uplift, but the uplift rate at Kangia North, only 5 km from the glacier front, was about 10 mm/yr larger than the rate at Ilulissat, located only ~45 km further away. This suggests that most of the uplift is due to the unloading of the Earth's surface as Jakobshavn thins and loses mass. Our estimate of Jakobshavn's contribution to uplift rates at Kangia North and Ilulissat are 14.6 ± 1.7 mm/yr and 4.9 ± 1.1 mm/yr, respectively. The observed rates are consistent with a glacier thinning model based on repeat altimeter surveys from NASA's Airborne Topographic Mapper (ATM), which shows that Jakobshavn lost mass at an average rate of 22 ± 2 km**3/yr between 2006 and 2009. At Kangia North and Ilulissat, the predicted uplift rates computed using thinning estimates from the ATM laser altimetry are 12.1 ± 0.9 mm/yr and 3.2 ± 0.3 mm/yr, respectively. The observed rates are slightly larger than the predicted rates. The fact that the GPS uplift rates are much larger closer to Jakobshavn than further away, and are consistent with rates inferred using the ATM-based glacier thinning model, shows that GPS measurements of crustal motion are a potentially useful method for assessing ice-mass change models.

Contribución al diseño de lazos de realimentación electrónica para microsistemas electromecánicos (MEMS) resonantes: ruido de fase generado en lazos osciladores por sus realimentaciones

En 1966, D. B. Leeson publicó el artículo titulado “A simple model of feedback oscillator noise spectrum” en el que, mediante una ecuación obtenida de forma heurística y basada en parámetros conocidos de los osciladores, proponía un modelo para estimar el espectro de potencia que cuantifica el Ruido de Fase de estos osciladores. Este Ruido de Fase pone de manifiesto las fluctuaciones aleatorias que se producen en la fase de la señal de salida de cualquier oscilador de frecuencia f_0. Desde entonces, los adelantos tecnológicos han permitido grandes progresos en cuanto a la medida del Ruido de Fase, llegando a encontrar una estrecha “zona plana”, alrededor de f_0, conocida con el nombre de Ensanchamiento de Línea (EL) que Leeson no llegó a observar y que su modelo empírico no recogía. Paralelamente han ido surgiendo teorías que han tratado de explicar el Ruido de Fase con mayor o menor éxito. En esta Tesis se propone una nueva teoría para explicar el espectro de potencia del Ruido de Fase de un oscilador realimentado y basado en resonador L-C (Inductancia-Capacidad). Al igual que otras teorías, la nuestra también relaciona el Ruido de Fase del oscilador con el ruido térmico del circuito que lo implementa pero, a diferencia de aquellas, nuestra teoría se basa en un Modelo Complejo de ruido eléctrico que considera tanto las Fluctuaciones de energía eléctrica asociadas a la susceptancia capacitiva del resonador como las Disipaciones de energía eléctrica asociadas a su inevitable conductancia G=1⁄R, que dan cuenta del contacto térmico entre el resonador y el entorno térmico que le rodea. En concreto, la nueva teoría que proponemos explica tanto la parte del espectro del Ruido de Fase centrada alrededor de la frecuencia portadora f_0 que hemos llamado EL y su posterior caída proporcional a 〖∆f〗^(-2) al alejarnos de f_0, como la zona plana o pedestal que aparece en el espectro de Ruido de Fase lejos de esa f_0. Además, al saber cuantificar el EL y su origen, podemos explicar con facilidad la aparición de zonas del espectro de Ruido de Fase con caída 〖∆f〗^(-3) cercanas a la portadora y que provienen del denominado “exceso de ruido 1⁄f” de dispositivos de Estado Sólido y del ruido “flicker” de espectro 1⁄f^β (0,8≤β≤1,2) que aparece en dispositivos de vacío como las válvulas termoiónicas. Habiendo mostrado que una parte del Ruido de Fase de osciladores L-C realimentados que hemos denominado Ruido de Fase Térmico, se debe al ruido eléctrico de origen térmico de la electrónica que forma ese oscilador, proponemos en esta Tesis una nueva fuente de Ruido de Fase que hemos llamado Ruido de Fase Técnico, que se añadirá al Térmico y que aparecerá cuando el desfase del lazo a la frecuencia de resonancia f_0 del resonador no sea 0° o múltiplo entero de 360° (Condición Barkhausen de Fase, CBF). En estos casos, la modulación aleatoria de ganancia de lazo que realiza el Control Automático de Amplitud en su lucha contra ruidos que traten de variar la amplitud de la señal oscilante del lazo, producirá a su vez una modulación aleatoria de la frecuencia de tal señal que se observará como más Ruido de Fase añadido al Térmico. Para dar una prueba empírica sobre la existencia de esta nueva fuente de Ruido de Fase, se diseñó y construyó un oscilador en torno a un resonador mecánico “grande” para tener un Ruido de Fase Térmico despreciable a efectos prácticos. En este oscilador se midió su Ruido de Fase Técnico tanto en función del valor del desfase añadido al lazo de realimentación para apartarlo de su CBF, como en función de la perturbación de amplitud inyectada para mostrar sin ambigüedad la aparición de este Ruido de Fase Técnico cuando el lazo tiene este fallo técnico: que no cumple la Condición Barkhausen de Fase a la frecuencia de resonancia f_0 del resonador, por lo que oscila a otra frecuencia. ABSTRACT In 1966, D. B. Leeson published the article titled “A simple model of feedback oscillator noise spectrum” in which, by means of an equation obtained heuristically and based on known parameters of the oscillators, a model was proposed to estimate the power spectrum that quantifies the Phase Noise of these oscillators. This Phase Noise reveals the random fluctuations that are produced in the phase of the output signal from any oscillator of frequencyf_0. Since then, technological advances have allowed significant progress regarding the measurement of Phase Noise. This way, the narrow flat region that has been found around f_(0 ), is known as Line Widening (LW). This region that Leeson could not detect at that time does not appear in his empirical model. After Leeson’s work, different theories have appeared trying to explain the Phase Noise of oscillators. This Thesis proposes a new theory that explains the Phase Noise power spectrum of a feedback oscillator around a resonator L-C (Inductance-Capacity). Like other theories, ours also relates the oscillator Phase Noise to the thermal noise of the feedback circuitry, but departing from them, our theory uses a new, Complex Model for electrical noise that considers both Fluctuations of electrical energy associated with the capacitive susceptance of the resonator and Dissipations of electrical energy associated with its unavoidable conductance G=1/R, which accounts for the thermal contact between the resonator and its surrounding environment (thermal bath). More specifically, the new theory we propose explains both the Phase Noise region of the spectrum centered at the carrier frequency f_0 that we have called LW and shows a region falling as 〖∆f〗^(-2) as we depart from f_0, and the flat zone or pedestal that appears in the Phase Noise spectrum far from f_0. Being able to quantify the LW and its origin, we can easily explain the appearance of Phase Noise spectrum zones with 〖∆f〗^(-3) slope near the carrier that come from the so called “1/f excess noise” in Solid-State devices and “flicker noise” with 1⁄f^β (0,8≤β≤1,2) spectrum that appears in vacuum devices such as thermoionic valves. Having shown that the part of the Phase Noise of L-C oscillators that we have called Thermal Phase Noise is due to the electrical noise of the electronics used in the oscillator, this Thesis can propose a new source of Phase Noise that we have called Technical Phase Noise, which will appear when the loop phase shift to the resonance frequency f_0 is not 0° or an integer multiple of 360° (Barkhausen Phase Condition, BPC). This Phase Noise that will add to the Thermal one, comes from the random modulation of the loop gain carried out by the Amplitude Automatic Control fighting against noises trying to change the amplitude of the oscillating signal in the loop. In this case, the BPC failure gives rise to a random modulation of the frequency of the output signal that will be observed as more Phase Noise added to the Thermal one. To give an empirical proof on the existence of this new source of Phase Noise, an oscillator was designed and constructed around a “big” mechanical resonator whose Thermal Phase Noise is negligible for practical effects. The Technical Phase Noise of this oscillator has been measured with regard to the phase lag added to the feedback loop to separate it from its BPC, and with regard to the amplitude disturbance injected to show without ambiguity the appearance of this Technical Phase Noise that appears when the loop has this technical failure: that it does not fulfill the Barkhausen Phase Condition at f_0, the resonance frequency of the resonator and therefore it is oscillating at a frequency other than f_0.