V=I´ R.

At very low temperatures (in the cryogenic temperature realm between absolute zero and 77 Kelvin, the boiling temperature of liquid nitrogen), many materials, including combinations of metals and alloys, become superconductive by the means of disordered electron pairs becoming ordered. These electrons are called Cooper pairs, or super-electrons, because they are afforded perfect conduction in the superconductor. The density of super-electrons in the material is temperature-dependent. Operating at a lower temperature yields more Cooper pairs forming from the normal state electrons, so a lower temperature generates greater superconductive behavior (Orlando et al. 1991). At low temperature and only in weak magnetic fields, a superconducting sample has zero resistance. The temperature at which this occurs is known as the critical temperature of the sample (T_c) (Kittel 1976). In wire or film form, a strong superconducting material effectively distributes the electrical power generated by existing conventional methods.

Typically, superconductive material makes up only a small part of a superconducting sample. For a film, a metal plate is cleaned and the crystal structure of the metal aligned. A thin layer of insulating buffer material is deposited onto the surface followed by the superconducting layer. The careful crystal structure alignment of the first layer imparts this ordered orientation to all subsequent layers. The parallel orientation of grains (crystals) in a multi-crystalline structure is important to maximize the super-current passing through the sample by eliminating boundaries between the crystallites that act as barriers to current flow between adjoining grains. The elemental composition of the crystal structure and a variety of different buffer layers and metal substrates are the most prominent effects on variations in behavior of superconductive films.

The superconductive material needs to have a high degree of purity for the best grain boundary conditions; however, it must also have strategically introduced defects within the grains. The defects allow the magnetic flux passing through the film to be pinned, allowing more current to pass through the sample in a directed stream. A magnetic flux line feels a sideways Lorentz force and produces a release of energy equivalent to resistance. Through a controlled doping procedure, the sideways force on the magnetic flux lines is resisted by the grain boundaries or the introduced impurities (Sheahen 1994). Doping is the addition of other foreign metals (such as calcium), whose atom replaces an atom in the lattice (crystalline) structure.

After a sample is synthesized, its superconductivity must be measured. Because superconductors only exhibit their phenomenal behavior at low temperatures, all testing is carried out in cryogenic surroundings under vacuum conditions. Traditionally, all measurements were painstakingly taken by hand; however, now measurements of temperature, applied current, and voltage are controlled, received, and interpreted by a computer with the help of software called LabVIEW. The goal of this project was to design a system and methodology to utilize this new software and hardware technology.

LabVIEW is a graphical programming language that is manufactured by National Instruments, and is typically used to automate data acquisition in research labs and industry. To accompany the software, corresponding hardware must also be installed. Within the central processing unit of a computer, a General Purpose Interfacing Bus (GPIB) card is installed into the PCI slot. The GPIB card is the connection between GPIB-compatible instrumentation and the computer. The card uses "handshaking" to communicate between talkers, listeners, and the controller. This means that the computer acts as the controller and is able to tell an instrument to be either a talker (it "tells" the controller what value it's at) or a listener (the controller "tells" it what value to go to). The card coordinates these transfers of information.

LabVIEW is the connection between the GPIB card and the data from the experiment. A LabVIEW program works through a GPIB card to control the instruments, take measurements, and organize the results. Automated data acquisition greatly decreases the amount of human error, can be left to run on its own, and can be run regardless of the skill or experience of the user.

The long-term goal for this project was to develop a LabVIEW program to evaluate the direct relationship between temperature and resistance. To illustrate the effectiveness of this program, various superconductive samples were tested simultaneously. Thin film samples included Bi2Sr2Ca2Cu2O3/Ag (BSCCO on Ag), YBa2Cu3O7 (YBCO) and NdBa2Cu3O7 (Nd123 TFA). The vast physical size differences of all of these elements are important in the packing structure and thermal behavior of the thin film samples, and were chosen to represent a wide range of materials.

Figure 1. Varian cryoocooler, outside and inside view.

The four pins on each sample block were used to take four terminal readings of all of the samples. In a four-terminal measurement, a current is applied over a set distance (the two outer pogo pins) and two voltage readings are taken (the two middle pins). The advantage to four terminals (instead of two) is uniform, steady-state current flow that creates fewer errors from lead and contact resistances.

The pogo pins are retractable so that they fit the sample snugly, enabling stable readings. A protective shield was put into place over the entire container and the vacuum pump turned on (Figure 1). Upon reaching a vacuum pressure in the 10^-3 torr range, the helium-filled cryopump was started. The temperatures and resistances were recorded once the specified starting temperature was reached.

A Varian cryocooler powered by a Varian cryopump and a vacuum pump were used with helium as a cooling agent. Attached to the thermometer terminals of current and voltage was a LakeShore 340 Temperature Controller (later switched to a LakeShore 34CA model). Through the use of the auto-tune feature on the temperature controller, optimal Proportional/Integral/Differential (PID) parameters were found. The aim of ideal PID parameters is to quickly and precisely control the temperature so as to obtain a swift temperature reading while minimizing the drift over time. Temperature zones were programmed into the temperature controller to change PID parameters as the temperature changed to achieve maximum thermal control.

A 1992 Scanner card was installed into a Keithley 199 Digital Multi-meter/Scanner for the purpose of stepping through the four samples to make measurements. The eight channels of the card were connected to the four current lines and four voltage lines. When closed, the channel was able to pass current supplied by the Keithley 224 Programmable Current Source through a sample. The corresponding voltage channel was closed, voltage readings made, and resistance calculated (via Ohm’s Law).

The designed LabVIEW program controls the amount of current applied to each sample. This can range from 1 m A - 1 Amp, depending on the sensitivity of the sample. The two voltage readings are taken on each sample, and these are passed through the GPIB card to LabVIEW for interpretation. The values of current and voltage are then converted into measurements of resistance, and the results graphed as a function of temperature.

The LabVIEW program stores the specifications of the sample and the rates at which readings were taken from the instruments. The rate at which readings were taken became smaller as the sample neared the critical temperature. When the lowest temperature (designated by the program’s user) is reached, the program starts to increase the temperature at that same slow rate until the critical temperature is again reached. The temperature and resistances are updated, graphed, and recorded in spreadsheet format only when the criteria set by the user are met.

The temperature set-point rate of change switches when a critical point is reached in order to gather more data in the region near the critical temperature. Both the change in temperature and the change in resistance values in the program are not fixed and allow flexibility concerning the data recorded in the Excel spreadsheet. Data are recorded while temperature is decreased and again while temperature increased to determine if the behavior of the sample is consistent during cooling and warming. The difference between the graphs of warming and cooling is known as hysteresis; ideally there should be none. Originally one program controlled both phases of the scan (cooling and warming). This was subdivided into separate programs for user versatility.

Thin film samples of various types were tested in this apparatus including Bi2Sr2Ca2Cu2O3/Ag (BSCCO on Ag), YBa2Cu3O7 (YBCO) and NdBa2Cu3O7 (Nd123 TFA).

Figure 2. Resistance vs. temperature during warming and cooling.

Discussion

The room temperature resistances show a large range in the order of magnitude, especially for the neodymium samples (Table 1). The thickness of the thin film sample plays a minor role in the level of resistance; however, the position of the contacting pogo pins on the surface of the sample is also important in resistance level. Inconsistencies in the sample may be present, and this may result in different paths of current flow through the film. Both neodymium samples show resistances at a much greater magnitude than expected, most likely because a current pathway exists through the metal substrate. The order of magnitude behavior for YBCO and BSCCO (Table 1) are typical of thin film samples of this type.

For our program, the critical temperature (T_c) is defined as the inflection point of the resistance vs. temperature curve (Figure 2). In the neodymium and yttrium samples, these values are similar because of the comparable elemental position within the lattice structure. XBa₂Cu₃O₇ (where X is Nd or Y) has an orthorhombic lattice structure and consists of stacking layers of perovskite units BaCuO₃, XCuO₂, and BaCuO₂. The Bi₂Sr₂Ca₂Cu₂O_x/Ag film shows a vastly different behavior because of the different packing structure in the lattice and the use of a dissimilar metal substrate (silver). In the BSCCO film, the layers of CuO₂ are separated by calcium containing planes (Poole et al. 1995). The vast physical size differences of all of these elements are important in the packing structure and influencing the thermal behavior of the thin film samples.

A wide variety of new superconductive compounds are now being made using different methods and various substrates under an assortment of different conditions. The TESS system we have developed will be used to process this great number of samples quickly by characterizing up to four potential new superconductors simultaneously. This system is an effective means of quickly a sample for superconducting behavior. Because this system is automated by LabVIEW, once the user defines the parameters of the sample set, the program will run unmonitored for its duration (usually 2-3 hours depending upon user-defined temperature changing rates). The storing of data is also a useful feature of the system allowing the user to manipulate the measurements in a variety of ways even after the run has taken place.

The ability to operate at high temperatures makes superconductors accessible and economically feasible for use in industry. They can be used in the development of transmission lines, levitation, electric motors, and medical and aerospace applications. The rapid characterization and testing of potential superconductors is therefore important to both science and industry.

This project could not have been possible without the guidance of my mentors, Rich Kerchner and Dave Christen. I would also like to thank the rest of the Superconductivity group and the Solid State Division at Oak Ridge National Laboratory in Oak Ridge, Tenn.

The research described in this paper was performed in the Superconductivity group of the Solid State Division at Oak Ridge National Laboratory sponsored by the United States Department of Energy.