M. Ahlskog, P. Hakonen, T. Lehtinen, R. Lindell, M. Paalanen, A. Paila, L. Roschier, M. Sillanpää, R. Tarkiainen, V. Vaskelainen, F. Wu
Visitors: J. Delahaye, Yu. Makhlin, E. Sonin, T. Tsuneta and A. Zyuzin
In nanophysics research we make small, less than one micron samples by processes developed in the semiconductor industry and study their electrical conductivity at low temperatures. In both normal and superconducting nanosamples quantum mechanical wave character of the electrons and their Coulomb repulsion lead to new phenomena, which we try to utilize in new sensor applications. We have developed, among others, record-sensitive SET-components made out of carbon nanotubes and nearly back-action-free, reactively read superconducting electrometers. In addition, we have developed a novel, low-noise current amplifier, Bloch oscillating transistor, which lies between the superconducting quantum interferometer (SQUID) and the SET according to its characteristics. The same circuit has been employed for measurements of very small noise currents and their higher order moments.
J. Delahaye, P. Hakonen, T. Heikkilä, R. Lindell, M. Paalanen, M. Sillanpää, E. Sonin, and T. Yamaguchi
We have continued measurements of conductance versus current for solitary, resistively confined small Josephson junctions. Our results show, for the first time, that the Cooper pair blockade is strongly sensitive to the non-Gaussian nature of shot noise. In our most recent measurements shot noise of a SET is connected through a small coupling capacitor to the detector junction. We find both symmetric and antisymmetric effects that can be related to the second and third moments, respectively. It is not quite clear yet, what is the actual quantity that is measured for the third moment in the case of quantum noise but our data provide a unique testing ground for theoretical calculations.
P. Hakonen and R. Lindell
Bloch Oscillating Transistor (BOT) is a novel mesoscopic transistor (three terminal device) in which a large supercurrent is controlled by a small quasiparticle current. The operating principle of a BOT utilizes the fact that, Zener tunneling up to a higher band will lead to a blockade of Cooper-pair tunneling (Bloch oscillation) in a suitably biased Josephson junction. Bloch oscillation is resumed only after the junction has relaxed to the lowest band. Using a quasiparticle control current, this process can be made faster. Since, one quasiparticle triggers several cycles of Bloch oscillations, a high current gain can be achieved.
We have studied the device characteristics of BOTs and compared them with simulations based on time dependent phase-fluctuation theory. We have concentrated our studies to small ratios of Josephson and Coulomb energies EJ/EC ~ 0.1 – 0.3 at which the theory based on perturbation expansion works the best. We have measured current gain, power gain, linear range, input impedance, output impedance, and noise temperature, and found a reasonable agreement between theory and experiment. For example, we can explain the measured noise temperature of 0.4 K and the current gain of 30 within a factor of two.
P. Hakonen, L. Roschier, and M. Sillanpää
We have made semiclassical analysis of the charge sensitivity of the inductive “L-SET” read-out scheme of superconducting Cooper pair transistors (SCPT). In this method, the charge-induced change of SCPT inductance is determined using reflection measurements at frequencies around 700 MHz. Our model agrees well with the measured data. We have been able to show that the L-SET is able to reach the quantum limited resolution using regular aluminum samples in the regime of EJ/EC ~ 0.1. We have also estimated the back action noise from an L-SET and found that a very large S/N ratio of ~ 1000 can be observed for its internal qubit.
P. Hakonen, T. Lehtinen, Yu. Makhlin, A. Paila, L. Roschier, and M. Sillanpää
In addition to the development of ultra sensitive charge detector, the L-SET, we have worked on its dual circuit C-SET, which is based on the quantum capacitance of superconducting SET. This is a sensitive phase detector which, according to our estimations, is a good candidate for a read-out device for charge-phase qubits. Our C-SET measurements present the first determination of capacitance renormalization in a macroscopic quantum system, the split Cooper pair box (CPB) over its phase-gate bias plane. Our radio-frequency reactive measurement scheme allows to probe purely the capacitive susceptibility due to the CPB band structure. We were able to account for the results using the standard CPB description with parameters determined independently by spectroscopic means. In addition, we could show that the method offers an efficient way to do non-demolition readout of the CPB quantum state, as well as to do studies of fast phase fluctuations at a sensitivity of 1 mrad/Hz1/2.
The applicability of the L-SET and C-SET depend very much on the achievable sensitivity. Our analysis indicates that the quantum limit of energy sensitivity (hbar per unit band width) can be reached rather well in L-SET, better than in the ordinary rf-SET. Both L-SET and C-SET are integrated qubit-detector systems in which reactive measurements of the qubit state can be performed, with good prospects of making it in a quantum nondemolition way .
P. Hakonen, T. Lehtinen, and M. Sillanpää
In order to reach the quantum limited performance of rf-SET electrometers, preamplifiers with a noise temperature of 100-200 mK are needed. One way to achieve such a sensitivity is to use SQUIDs as amplifiers. Our first SQUID amplifier design gave a gain of 20 dB and a band width of 100 MHz. The biggest problems were connected with a large capacitive feedback and the large ground inductance connecting the input and output sides of the amplifier. We have made a new design in which these problems should be eliminated.
M. Ahlskog, P. Hakonen, M. Paalanen, R. Tarkiainen, and T. Tsuneta, F. Wu, and A. Zyuzin
Our studies of very disordered, catalytically grown CVD multiwalled carbon nanotubes (MWCNT) were completed during the past year. Resistance vs. temperature measurements on CVD tubes with good-quality contacts (Rc ~ 1 kW) and resistance of ~ 30 kW/mm displayed rather large conductance corrections which we have analyzed in terms of the interaction effects. As a function of voltage, heating effects tend to dominate, and the dependence can be best modeled by using the equation for diffusive heat transport. The density of states of these tubes has been studied using high impedance Al-AlOx-NT contacts (Rc ~ 100 kW). We have compared our results with the theoretical calculation on tunneling into 1-dimensional disordered system, and obtained good agreement with the results beyond the first order corrections.
During the past year, the main emphasis of our carbon nanotube work was shifted towards high frequency transmission/reflection measurements as well as to studies of current-current fluctuations. A new measurement setup was constructed for electrical noise measurements at frequencies of 600 – 900 MHz. It is intended for high impedance samples with a built in tunnel junction calibration source. We have made our first measurements of shot noise both in the tunneling regime as well as in quantum transport regime. We obtained an unexpectedly small Fano-factor ~ 0.1 for MWCNTs, and clear indications that electron-phonon interactions play a role at large bias.
Within the ELENA consortium funded by the Academy of Finland, we have extended our collaboration with the Aerosol-group of Prof. Esko Kauppinen. Our joint goal is to make CVD grown SWNT samples on TEM grids so that we could measure, on the same sample, both conductance and noise as well as the chirality. First samples using the tubes of the Aerosol-group were manufactured.
A second new direction with CNTs - superconductivity induced by the proximity effect - was also started. The goal is to make large-bandwidth charge sensitive devices. This can be achieved either by FET-type of devices (rf-SET etc.) or hybrid devices based on superconductivity. Superconductivity can be induced in to CNTs by proximity effect which would result in a strong reduction of the impedance of the tube. Once the impedance is close to 50 Ohms, no matching circuitry is needed and broad band operation is achieved.