The design of electronic circuitry is described that has the capability of measuring independently the bias dependence of the first derivatives dI/dV (dynamic conductance) and dV/dI (dynamic resistance) and the second derivatives d2I/dV2 and d2V/dI2 of a tunneling current‐voltage (I‐V) characteristic over a wide range of junction impedances (less than 1 Ω to greater than 104 Ω). Connections to the sample are four terminal so that problems with lead resistance are eliminated. Resolution in the measurement of the first derivatives is better than one part in 104, and problems with long‐term drift are obviated by the use of feedback. Modulation levels at the sample junction, both at constant current and constant voltage, are conveniently monitored with a calibrated lock‐in amplifier and can be easily changed. The circuitry has been operated successfully with voltage modulation levels across the junction ranging from 1.0 μV to 2.5 mV rms. Data are presented on Al–I–Pb and Al–I–AuPb3 tunnel‐junction samples to ...

A new approach to high resolution measurements of structure in superconducting tunneling currents

Publisher Summary This chapter emphasizes on tunneling spectroscopy, which one might define, roughly, as any measurement giving structure, however weak, as a function of the injection energy eV  Normally, of course, one measures the tunneling conductance G ( V ) or its derivative with respect to V , perhaps in the presence, additionally, of a magnetic field or stress, to clarify the origin of energy-dependent structure Such measurements are usually made near 42 K, even when a normal metal counterelectrode is used, because the resolution of the spectroscopy is limited by thermal energy k B T  The apparent exclusion of superconductivity in the title may thus appear misleading; on the other hand, this chapter does not consider such purely superconductive phenomena as gap anisotropy, vortex properties, quasiparticle lifetimes, and the whole range of Josephson effects, that certainly comprise a separate field These topics have been reviewed in the recent book by L Solymar, to which the present article is complementary in a sense, although it attempts a rather greater depth in treatment of the physical content of material, at some expense in exhaustive bibliography Nevertheless, it discusses the phenomenon of spin-split superconductivity as a means of determining the spin polarization in magnetic electrodes, the superconducting proximity effect as a specific strategy for studying magnetic interactions and other properties of the normal state, and several superconducting materials of an unusual nature

Nonsuperconducting Electron Tunneling Spectroscopy

A simple four‐terminal conductance bridge circuit, which provides high‐resolution measurements of the current–voltage characteristics, dI/dV and d2I/dV2, as well as capacitance and dC/dV measurements, is presented. Current measurements in the wide range 1 nA–10 mA are possible and modulation voltages down to 1 μV can be used to achieve high resolution. The circuit has been used successfully for tunneling spectroscopy on GaAs–GaAlAs field effect transistor structures.

New high‐resolution conductance bridge for tunneling spectroscopy

Very thin films of CdSe have been prepared as tunneling barriers in Sn–CdSe–Sn tunnel junctions. The CdSe barriers were prepared at a substrate temperature of 115°C. High substrate temperatures were used since under such conditions dielectric films of CdSe can be produced. Conductance measurements on these junctions at 300°K as well as 77°K showed that the conductance depended exponentially on the thickness of the CdSe barrier within the thickness range studied, 50–100 A. This result is consistent with tunneling theory. In addition, at 1.4°K a sudden jump in current occurred at a voltage corresponding to the Sn superconducting energy gap. The magnitude of the current jump agrees with the theory of quasi‐particle tunneling. The shape of the I‐V curve as well as its weak dependence on temperature also indicate that tunneling is the conduction mechanism in these junctions. Since thin CdSe films are usually not continuous, it is necessary to oxidize those areas of the Sn electrode where no CdSe is present. A ...

Cadmium selenide films deposited at high substrate temperature as tunneling barriers between tin electrodes

The essential prerequisite for determination of peak profiles and intensities is calibrated tunneling spectra. The first sections of this chapter will introduce a simple calibration scheme which can be implemented with any type of measurement system assisted by a minicomputer, microcomputer, or sophisticated calculator. The approach will be of a tutorial nature so that the experimenter can obtain the knowledge necessary to interface his own measurement system for use with a computer-controlled data acquisition system. Practical aspects of calibration, shortcuts, and tricks will be discussed in detail along with software-hardware tradeoffs. Following the details of calibration, the subtraction of backgrounds to IET peaks will be discussed in detail along with peak profiles in both the normal and superconducting states, from which peak intensities are easily determined The use of differential inelastic electron tunneling spectra (DIETS) will be briefly mentioned.

Computer-Assisted Determination of Peak Profiles, Intensities, and Positions

A method of generating bias has been developed which greatly facilitates studying the voltage‐current characteristics of tunnel junctions. A triangle wave to sweep the bias, a dc level for offset, and a small amplitude sine wave used for taking derivatives are all summed in a dc amplifier. The junction is biased by the attenuated output of this amplifier with either a high or low impedance load line. The two transistor circuits which largely comprise this bias supply are compact and battery powered, minimizing grounding problems.

Andrew Longacre

Papers

Biasing Circuitry for Tunnel Junctions