We found quantitative criteria to characterize the states of the device: i) pristine devices show at low bias I proportional to Vm with m = 0 pointing to trap filling and at higher bias m=6 pointing to tunneling. The 1/f noise is characterized by 10-7 < &agr;&mgr; (cm2/Vs) < 10-5; ii) forming state is a transition between pristine and switched-state. The time dependent soft breakdown in the Al-oxide goes hand in hand with strong discrete multi level resistive switching (RTS) with a 1/f 3/2 spectrum. Once the device is switched in the high (H-) or low (L-) conductance state it never comes back to the pristine state. iii) The H- or L-state is characterized by I proportional to Vm with either m = 1 or m = 3/2. The injection model predicts the current level and the
dependence of the 1/f noise on current. Reliable switched devices show mainly 1/f noise. In the L-state there is often a
1/f 3/2 contribution on top of the 1/f noise indicating multi level switching. Reliable switches between the L- and H-state
are characterized by a resistance R that changes for example by a factor 30 and the relative 1/f noise, fSI/I2 ≡ C1/f follows the proportionality: C1/f proportional to R with a &agr;&mgr;-value of about 3x10-2 cm2/Vs. The explanation from the noise for C1/f proportional to R is that the number of carrier in the transport switches due a change of the number of parallel conducting paths in the polymer. The onset of switching seems to be at spots of the Al / Al2O3 / polymer interface.
The 1/f noise in MOSFETs is stated to be an ensemble of many RTS with different time constants. The majority of
literature on 1/f noise is overlooking the contribution due to mobility fluctuations that are uncorrelated with number
fluctuations. Our demonstration that the so-called proofs for ΔN can also be obtained from the empirical relation is new.
The following misunderstandings and controversial topics will be addressed: 1) 1/f and RTS noise can have different
physical origins. An analysis in time domain shows that the low-frequency noise with RTS is nothing else than a
superposition of a pure two level noise with a Lorentzian spectrum and a noise with a Gaussian amplitude probability
density with a pure 1/f spectrum with different bias dependency and physical origins. 2) It is very unlikely that in a
spectrum consisting of one strong two level RTS and a pure 1/f noise, the 1/f noise is a superposition of many RTS with
different time constants. 3) The spreading in WLSI /I2 below a critical WL is not a proof for the ΔN origin. 4) The typical
shape in the double log plot of SI /I2 versus I, from sub threshold to strong inversion is also not a proof for the ΔN origin.
In this work, we have characterized three NbNx thin films deposited on sapphire substrate and compared their noise properties. The three films were measured in the same conditions.
In the first time, the films were characterized with an impedance analyzer from 20 Hz to 1 MHz. The films are then considered as a RC dipole with a resistor R in parallel with a capacitor C. With the Nyquist formula, we calculate the noise voltage spectral density SvTh of the RC dipole considering that only the resistor R exhibits thermal noise in unbiased samples.
In the second time, noise measurements were made with the samples biased. Thanks to a four contacts configuration, we checked that contact noise do not contribute to our measurements.
The difference between the measured noise and the calculated thermal noise SvTh shows an extra 1/f noise without GR noise contributions. The 1/f noise in the three films extra noise is compared. These results are also compared to the noise measured on NbN thin films deposited on silicon substrate .
In this work, we have characterized YBaCuO high Tc superconducting thin films deposited on (001) MgO substrates and compared their noise properties. The films were sputtered on substrates which were annealed at different temperatures prior to deposition. The noise measurements were performed under the same conditions:
1) Without bias, the films are at equilibrium and exhibit only thermal noise proportional to the real part of the impedance for the voltage fluctuations or proportional to the real part of the admittance for the current fluctuations.
2) With bias, the films exhibit 1/f noise due to the conductivity fluctuations.
The extra noise is compared with Hooge's empirical relation. The normalized noise spectral density (Sv / V2) measured at 300 K as a function of the substrate annealing temperature displays a bell-shaped dependence with a maximum at a critical temperature.
Low frequency noise characteristics of light-emitting diodes with InAs quantum dots in GaInAs layer are investigated. Two noise components were found in experimental noise records: RTS, caused by burst noise, and 1/f Gaussian noise. Extraction of burst noise component from Gaussian noise background was performed using standard signal detection theory and advanced signal-processing techniques. It was found that Hooge's empirical relation applied to diodes by Kleinpenning is applicable to the electric 1/f noise of quantum dot diodes as well. Two different spectra decomposition techniques are used to obtain burst noise spectra. Bias dependences of burst and 1/f noise are compared. It is concluded that the RTS noise and 1/f noise have different physical origins in light-emitting diodes with quantum dots.
The bispectrum of the 1/f noise is investigated in the present work. For the Gaussian noise it equals zero. LEDs on self-organized InAs/GaAs quantum dots and laser diodes on In0.2Ga0.8As/GaAs/InGaP quantum wells made in Russia were tested. The voltage noise was analyzed in a wide interval of currents through the diodes. Estimates of the probability density function and semi-invariants of the noise have not revealed any appreciable deviations from the Gauss law. Noise spectra Sv(f)in the range 1 Hz - 20 kHz were analyzed. In most cases the frequency exponent γs of the spectrum is close to one, the Hooge’s parameter αH has magnitude of the order 10-4. The bispectrum Bv(f1,f2of the noise is a complex function of frequencies f1 and f2. Its absolute value is decreasing while moving from the beginning of the frequency plane Of1f2. The decrease along the bisector (f1 = f2 = f) follows the power law characterized by the frequency exponent γB ≈ 1.5 γs. The dependence of the "height" of |Bv(f,f)| on the current through the diode is qualitatively similar to this one for the spectrum. The power law describes these dependences, however the exponents are essentially different.