Last Update: October 2000 |
ESUH (Genova, Italy)
1.Goal of the experiment
When a Josephson junction is irradiated with a radiation of frequency f, the response of the supercurrent gives rise to constant-voltage Shapiro steps in the dc I-V curve at voltages
Vn = n hf/2e (1)
n (the order of the step) being an integer and the step amplitude hf/2e is, for example, 31mV if the radiation frequency is 15 GHz. The goal of the experiment is to find how nearly equal are the voltages (1), produced by two junctions irradiated with the same radiofrequency f.
By directly comparing the voltages across the two junctions, biased on the induced step of the same order n, one make a differential measurement connecting the junctions by a full superconducting loop and reading the current flowing in this loop by a SQUID (its input coil is superconducting). A difference DV in voltage, between the sources, produces a current I, in this no resistance circuit, according with the equation
where L is the circuit inductance. This experiment is very sensitive to differences in voltage, because we can wait for tens hours to detect the change of the current.
The helium bath temperature is stabilized within 1e-4 K. A mmetal shield, to reduce the earths field to less than 100 mG, surrounds the cryostat. The experiment is mounted inside a niobium box, made of two separate compartments, one holds the RF board with the junctions and a second (very well shielded from the other) holds the DC SQUID and the superconducting switch. The good quality, of this RF shielding, allows connecting the input coil of the SQUID directly inside the superconducting loop. It is not requested any superconducting transformer to couple the SQUID with the loop.
We use custom made silicon chips with two junctions; one junction electrode is connected to the electrode of the other junction by a superconducting niobium sputtered line and the other two electrodes are separately connected, by the same type of line, to two superconducting pads. The superconducting loop is closed connecting the SQUID input coil with these two pads.
We polarize both junctions on the same step, opening the superconducting loop by a superconducting switch and we turn off the switch heater to have again a full superconducting loop, when the step comparison starts.
2. Physics achievements during 1999 -2000
The first comparison measurement, between the Shapiro steps of second order (2 x 29.5mV = 59mV), has been made with sensitivity better than 1e-17V. This limit comes from the magnetic coupling between the junction DC polarization currents and the superconducting loop, made by the two junctions and their connections with the SQUID input coil. We proved that the polarization currents are stable, but the small change (parts on ten thousands) of their mutual inductance with the loop is responsible of the drift of the circulating current I of about 80nA (0.5 quantum flux inside the SQUID washer) in 19 hours.
The sensitivity of this comparison measurement is
L (DI/Dt) = (3.3 e-6 H) (80e-9 A/68e3 sec) = 4 e-18 V
It is possible to measure the superconducting loop inductance L, causing the entrance of a well-known number of quantum fluxes inside the loop and measuring, by the SQUID, the relative change of the circulating current I. This prove also that the contacts, between the pads on the chip and the rest of the loop, are superconducting too. We also measured that the circulating current does not change for long time (hours), having both the junctions on the step zero (no RF power applied).
The mutual inductances, between the polarization currents and a superconducting loop, having the two junctions belonging to different chips, are almost an order of magnitude grater than the ones with the loop having both the junctions on a single chip. It is possible to produce very good and stable steps with junctions having a not hysteretic I_V characteristic. The reported comparison has been made between junctions belonging to different chips, because no of the tested chips had both junctions able to produce good steps. We have other chip to be tested and we are having contacts to build two not hysteretic junctions on a chip having the same circuit design of the present.
The rest of the apparatus works and could make the comparison between the steps, with the sensitivity near 1e-20V.
3. INFN contribution to the experiment in terms of manpower and financial support
-Manpower: 2 researchers (1 FTE), 0.33 technician and the support of the mechanics workshops, the mechanical design and liquid helium services of the Genoa section.
-Financial support: 381MLit in 8 years
4. Number of publication in refereed journals: 0
5. Number of talks to conferences: 0
6. Number of undergraduate and doctoral thesis on the experiment:
-N.8 Laurea diplomas (One is attending the graduate school and all the others are working in the electronic and communication industry)
7. Leadership role in the experiment
The experiment has been designed and built in Genoa.
We wrote the specifications for the chip with the junctions and the RF filters. Dr. R. Cantor, Star Cryoelectronics, Los Alamos (USA) designed its layout and the company Hypres, N.Y (USA) made them.
8. Innovative instruments
The measured drift of the circulating current I is 5.8e-15 A/sec (equivalent to a change of 1.4e-4 quantum flux per hour inside the SQUID washer) is essential for the sensitivity of this experiment. One of the necessary feature of the apparatus to achieve this result is the bath temperature stabilized within 1e-4 K; we use the liquid helium vapor pressure as a control variable, because we cannot supply current to a heater inside the magnetic shielded cryostat.
9. Competing experiments
This experiment has been done the first time by Clarke [1] and after by Tsai, Jain, Lukens [2], [3], an interesting review of the physical problem has been written by Mc Donald on the Science magazine [4]. We do not know of other experiments of this kind in progress now.
[1] J.Clarke, Phys. Rev. Lett 21, 1566 (1968)
[2] Tsai, Jain, Lukens, Phys. Rev. Lett. 51, 316(1983)
[3] Jain, Lukens, Tsai, Phys. Rev. Lett. 58, 1168 (1987)
[4] D.G. Mc Donald, Science 247, 177(1990)
10. No international committee has reviewed this experiment
11. Results of special relevance for other scientific disciplines or of special impact for applied research and industrial applications.
We designed and built a susceptometer to make the in vivo measurement of the iron overload. Now we can measure (in collaboration with Prof. S. Parodi of the Medical Department and National Institute for Cancer Research -IST- of University of Genoa) the average iron concentration inside the body of living rats, injected with iron dextran, with a sensitivity of about 0.4mg/cc. We can show that it is possible to build a similar apparatus having the same or a better sensitivity, suitable to do the same in vivo measurement on the human body.
In order to accomplish the removal of excess iron from the body of the Thalassemia patients, it is necessary to ascertain its concentration. At present only two instruments are available worldwide (Cleveland, USA and Hamburg, Germany), which use Superconducting Quantum Interference Device (SQUID). Their utilization requires very expensive and cumbersome criogenic apparatuses, whose geometry renders the measurements strongly dependent on the magnetic properties of the most external abdominal surfaces. This work was performed to study the feasibility of a susceptometer planned to work only with components at room temperature. The "Associazione Veneta per la Lotta alla Talassemia" and the "Arcispedale SantAnna - Azienda Ospedaliera di Ferrara" are paying the salary of a young physicist working full time for this project. A patent request is in progress.