US 5,506,500 AGrant
Magnetic Property Characterization System Employing a Single Sensing Coil Arrangement to Measure Both AC Susceptibility and DC Magnetization
Issue Date:1996-04-09
•25 Claims
•12 Drawing Sheets
Abstract
A coil arrangement having a primary and two balanced secondaries is used to measure ac susceptibility and, using sample movement, also to perform absolute dc moment measurements. Since the same coil arrangement is used for both AC and DC measurements, the measurements can be made successively within a cryogenic chamber without removing the sample. This is a big advantage, because the changed conditions associated with removing and replacing a sample between measurements can cause confusing, uncorrelated results. A high speed voltmeter is used to perform the signal analysis for the moment measurement. The system can be configured to yield high resolution dc moment measurements to 25 ppm and sensitivities to better than 5.times.10.sup.-5 emu.
Metadata
Assignee
- Lake Shore Cryotronics, Inc.
Inventors
- John K. Krause
- Victor Wang
- Bradley C. Dodrill
Application Information
Application Number:US 2310635
Filing Date:1994-04-22
Priority Date:1992-03-18
Art Unit:225
Classifications
IPC:
G01R 3316G01R 33035G01N 2771G01N 2772
Field of Search:
324201;202;244;205;248;259;204;228;260-262;224;226
Patent Drawings (12 sheets)
Description
FIELD OF THE INVENTION
The present invention relates to measuring the magnetic characteristics of a sample, and more particularly relates to measuring magnetic parameters using both AC and DC measuring techniques.
BACKGROUND AND SUMMARY OF THE INVENTION
AC susceptometry and DC magnetometry are two different widely used techniques for obtaining information about the magnetic characteristics of a sample. The following provides a brief background discussion of each of these well known techniques.
Briefly, in a typical dc magnetization measurement, a value for the magnetic moment m of the sample is measured for some applied dc field H.sub.dc. The magnetic moment m is a bulk sample property and is a measure of the magnetic field generated by the sample itself. The dc or static "susceptibility" is determined by dividing the magnetization by the applied field (x.sub.dc =M/H.sub.dc). When comparing different materials (or samples of the same material having different sizes), the macroscopic quantity of interest is magnetization per unit volume (or per unit mass) of the sample.
Techniques for measuring the dc moment have long been commercialized in a variety of system products such as vibrating sample magnetometers (VSMs), force magnetometers, and SQUID magnetometers. The most commonly used dc magnetometers, such as the vibrating sample magnetometer (VSM) or SQUID magnetometer, generally use a detection coil to measure the change in magnetic flux due to the presence of a magnetized sample. If the sample does not have a permanent magnetic moment, an applied field is required.
FIG. 1 schematically illustrates a typical prior art commercial DC magnetometer. In many commercial DC magnetometers, a magnet 50 (e.g., an electromagnet or superconducting solenoid) is provided to apply a constant magnetic field H.sub.dc to the sample 52 in order to magnetize the sample to a magnetic moment (m). A detection coil 54 and associated detection circuit 56 are also provided. There is no output from the detection circuit 56 until there is a magnetic flux change within the detection coil 54 (Faraday's Law). The sample flux coupled to the detection coil 54 is commonly varied by moving the sample. In a VSM, the sample 52 is vibrated near detection coil 54. As the sample 52 moves, an ac signal is generated at a frequency determined by the sample oscillation. In a SQUID system, the sample 52 is simply passed through the detection coil 54. Typically, detection circuit 56 comprises a fluxmeter, lock-in amplifier, or SQUID electronics that is coupled to the detection coil 54 in order to measure the change in flux (i.e., the magnetic moment). A so-called "extraction technique" is known wherein the fluxmeter (detection circuit) 56 deflects in an amount proportional to the flux change (moment) as the sample is removed from the sensing coil. See Cullity, Introduction to Magnetic Materials, pages 61-81, and particularly, 65-66 (Addison-Wesley Publishing Co. 1972).
Traditionally, fluxmeters, which are essentially digital voltmeters, are used for measuring secondary coil output signals over time. However, most fluxmeters have a relatively low input impedance (which can present a problem when the sensing coils to be used have a large variable resistance). In addition, fluxmeters may present potential measuring difficulties in dealing with drifts and thermal emfs.
It is generally known to use an integrating digital voltmeter as a fluxmeter. It is apparently also known to use a digital voltmeter to measure voltages to indicate DC magnetization. For example, the Rebouillat and Sasaki references cited below each appear to teach using an integrating digital voltmeter as a fluxmeter for measuring magnetization. In addition, the Werle '142 patent (cited below) teaches a magnetic fluxmeter for measuring macro flux disturbances (e.g., caused by a ship passing by) which includes a digital voltmeter type indicator for indicating the level of a signal integrated in the analog domain. The Coley '990 patent (also cited below) appears to teach a digital voltmeter arrangement in his FIG. 4 tunneling susceptometer (see components 52-66).
When properly calibrated, the output from the VSM or SQUID magnetometer yields the value of the magnetic moment for the sample. With knowledge of the sample volume (V), the magnetization (M) can be determined. Magnetization and the dc susceptibility are thus derived quantities. Usually, the moment is measured as a function of field, and the materials' magnetization curve (i.e., m or M versus H.sub.dc) is determined (see FIG. 2) by repeating the measurement for different values of H.sub.dc. That is, in a dc magnetization measurement discrete points along some characteristic magnetization curve are measured. This permits measurement of several discrete points along the magnetization curve of the sample. If the magnetic field direction can be reversed, hysteresis curves can also be generated.
DC magnetization/susceptibility measurements are extremely useful (e.g., for high field studies and for measuring hysteresis curves). However, sometimes additional (or different) information about the magnetic properties of a sample is required that is not available from a DC type measurement. For example, there are instances in which "complex" (real and imaginary) magnetic susceptibility must be measured in order to provide more complete information about the magnetic characteristics (e.g., relaxation characteristics) of a sample. AC susceptometery provides certain information (e.g., information about such "complex" parameters) that is not available from DC magnetometry. Moreover, in the AC measurement, copper wound coils can be used to generate very small amplitude AC fields (e.g, >>1 Oe) without complications arising from the remanent fields associated with iron-core or superconducting magnets. This means that the AC technique is very valuable in the study of low-field magnetic characteristics of a sample.
In addition, with the capability to vary the frequency of the drive field in an AC technique, the magnetodynamics of the magnetic system can also be studied. Further, since in the AC measurement the slope of the magnetization curve is being measured, non-linear magnetization and magnetic transitions are often best studied using an AC measurement.
AC susceptometry has thus been widely used for the characterization of magnetic materials for many years. However, prior to 1988, there was no serious commercial product available and ac susceptometer usage was almost universally "build your own" (so that, for example, many ac susceptometers were "laboratory assembled" using available components). Unlike the dc technique (where actual values for the magnetic moment m are measured), changes in m (i.e., .DELTA.m) are measured in the ac technique. Thus, the ac susceptibility gives an indication of the slope (dm/dH) of the magnetization curve. This is a fundamental difference between the ac and dc measurement techniques.
The discovery of high T.sub.c superconductors led to a rapid increase in interest in magnetic measurements. High T.sub.c materials are characterized by relatively small first critical fields, H.sub.cl, and a small full penetration field, H.sub.p. Therefore, a reliable low-field magnetization measurement technique is necessary for the full magnetic characterization of these materials.
In addition, the ac measurement can be used to differentiate between inter- and intragranular current coupling, and in determining the overall quality of a superconductor. An analysis of x" can provide information about the critical current density J.sub.c of these materials via the invocation of a suitable critical state model, and this sort of analysis has contributed to a better understanding of the mechanisms of superconductivity in these compounds.
An analysis of the complex susceptibility can also provide information about relaxational processes that may be occurring in the system under study. For example, it can be used to study spin-lattice relaxation phenomena in paramagnetic compounds, domain wall movement in metamagnetic systems, and has contributed to a better understanding of spin-glass systems.
Using the ac technique, the sample is generally centered within a detection coil and exposed to an applied AC magnetic field. The magnetic moment of the sample follows the applied field. The detection circuitry is generally balanced, with matched detection coils being provided in order to null out the changing flux due to the AC excitation. As a result, the detected change in flux is related only to the change in moment of the sample as it responds to the AC field.
Using well established principles of ac susceptometry, Lake Shore Cryotronics (the assignee of the present invention) introduced its 7000 line of ac susceptometers in the fall of 1988. FIG. 2A is a schematic block diagram showing this prior art ac susceptometer developed by Lake Shore. The basic principles of operation are described in Lake Shore's application note entitled "AC Susceptibility Measurement: Its Purpose and Process" (the disclosure of which is incorporated by reference herein for the purpose of providing discussion as to the state of the art). While no sample movement is required to perform AC susceptometery measurements, a motor and associated sample movement arrangement are provided in Lake Shore's ac susceptometer to move the sample in order to increase measurement accuracy and resolution.
Needless to say, much work has been done in the past in regard to magnetic characteristic measuring techniques. The following documents relate to techniques for measuring the magnetic characteristics of a sample:
Rillo et al, "Multipurpose a.c. and d.c. Equipment for Low Temperature Magnetic and Electric Measurements of Solids" (Abstract of paper presented at S ONR Workshop in May 1991);
U.S. Pat. No. 3,528,001--Yntema
U.S. Pat. No. 3,454,875--Bol et al
U.S. Pat. No. 4,861,990--Coley
U.S. Pat. No. 2,975,360--Bell
U.S. Pat. No. 4,037,149--Foner
U.S. Pat. No. 4,005,358--Foner
U.S. Pat. No. 4,238,734--Steingroever et al
U.S. Pat. No. 4,849,695--Muller et al
U.S. Pat. No. 5,008,621--Jiles
U.S. Pat. No. 3,863,142--Werle
Goldfarb et al, "Alternating-Field Susceptometry and Magnetic Susceptibility of Superconductors" (Office of Naval Research Workshop on Magnetic Susceptibility of Superconductors and Other Spin Systems, Berkeley Springs, W.V. May 1991);
Rebouillat, "High Resolution Automatic Magnetometer Using a Superconducting Magnet: Application to High Field Susceptibility Measurements", IEEE Trans. on Magnetics, v. MAG-8, n. 3 pp. 630-33 (3 Sept. 1972);
Sasaki, "Simple Precision Fluxmeter", 76 Nuclear Instruments and Methods n. 1 pp. 100-2 (1 Dec 1969);
Edwards et al, "Magnetometer for Surface Flux Density Measurement in MPI", 29 British Journal of Non-Destructive Testing n. 5, pp. 304-306 (Sept. 1987);
Beckley et al, "Simplified Electronic Permeameter Suitable for Routine and Standards Use", 9 Measurement and Control n.10, p. T65-T70 (Oct. 1976);
Marinaccio, "Op Amp Converts DVM To Fluxmeter", 48 Electronics n.10, pp. 112-113 (May 15, 1975);
De Mott, "Integrating Fluxmeter with Digital Readout", IEEE Journal on Magnetics v. MAG-6 n.2 pp. 269-71 (2 Jun. 1970); and
Press Release, "DOWTY RFL Offers The Only Digital Fluxmeter With IEEE-488 Bus Operation", Dowty RFL Industries, Boonton N.J. (Mar. 24, 1986).
Jiles teaches a measuring device capable of measuring various magnetic parameters (e.g., fluxmeter, gaussmeter, strain indicator), and uses the same overall transducer assembly for various measurements. The Dowty RFL press release describes a digital fluxmeter which measures "both flux density and total flux".
In addition, AC susceptometers, DC magnetometers and Vibrating Sample Magnetometers are commercially available from several companies. For example:
Quantum Design Inc. of San Diego, Calif. has been marketing a DC magnetometer since 1985, and as early as 1990 announced an option for making complex ac susceptibility measurements using elements of the SQUID detection system in its dc magnetometer. The means for making the ac measurements is not clear, nor has this group been able to qualify performance characteristics. In fact, it appears that no units have actually been delivered as of the filing date of the subject application.
Cryogenic Consultants Ltd., London England has been marketing, for approximately two years, a DC magnetometer employing a SQUID detection system. This device has no ac measurement capabilities. Cryogenic Consultants also markets a vibrating sample magnetometer.
Metronique Ingenierie of Le Bourget, France (no longer a going concern) introduced a SQUID (dc) Magnetometer in December 1989. They offered an ac measurement option which consisted of a separate set of sensing coils. The customer had to place a separate ac measurement insert into the dewar (refrigerator) system.
EG&G Princeton Applied Research (PARC), Princeton N.J., long ago introduced a vibrating sample magnetometer (VSM) that makes dc magnetization measurements. This instrument in found in many installations around the United States.
Princeton Measurements Corporation of Princeton New Jersey introduced an Alternating Gradient Force Magnetometer (AGFM) about two years ago--possibly to compete against VSMs and room temperature applications for SQUID magnetometers.
Phasetrack Instruments of Santa Clara, Calif. markets an ac susceptometer on a small scale. This instrument is not capable of making dc magnetization measurements.
See also generally the following papers relating to magnetic characteristic measurement: John K. Krause and Jeffrey R. Bergen, "Understanding magnetic measurement techniques," Superconductor Industry, vol. 3, no. 4, pp. 23-26, 1990;
Jiles, Magnetization and Magnetic Materials, pages 47-68 (Chapman & Hall 1991);
Goldfarb, "Thermoremanent magnetization and superparamagnetism in nickel-manganese alloys", Ph.D. dissertation (Colorado State University 1979);
Khoder et al, "Calibration constant calculations for Magnetic Susceptibility";
Couach et al, "Study of Superconductors by AC Susceptibility", Cryogenics Vol. 25, pp. 695-99 (1985);
Goldfarb et al, "Calibration of ac susCeptometers for cylindrical specimens", Rev. Sci. Instrum. vol. 55, pp. 761-64 (1984):
Zieba et al, "Superconducting Magnet Image Effects Observed With A Vibrating Sample Magnetometer", Rev. Sci. Instrum. Vol. 54, pp. 137-45 (1983); and
Rillo et al, "On the Sensitivity of High-TC Superconducting Ceramics as Magnetic-Field Sensors", Sensors and Actuators A-Physics, Vol. 27, N1-3 pp. 775-80 (1991).
For some time there has been a desire in the field to develop a practical, cost-effective commercial instrument capable of providing both ac and dc measurement techniques. As mentioned above, for a complete study of the magnetic properties of at least certain samples it is desirable to perform both ac and dc measurements. However, despite such desire, no one in the past has developed a practical, cost-effective instrument that is capable of accurately measuring the characteristics of a sample using both ac and dc techniques.
The present applicants have developed a preferred embodiment magnetic measurement system that is versatile, highly accurate, and can measure both AC susceptibility and DC moment. The instrument provided in accordance with a presently preferred exemplary embodiment of the present invention includes various components (i.e., two oppositely wound secondary coils, a source of AC excitation current coupled to a primary winding, and a stepping motor for moving the sample between the coils) to perform AC susceptibility measurements. The instrument also includes a source of DC current that can be coupled to the primary winding. The motor and associated sampling positioning arrangement is also used to provide the motion needed for "extraction type" DC magnetization measurement. A high speed digital voltmeter monitors and records the output of the secondary coil(s). The recorded output of the digital voltmeter is numerically processed (using a computer) to yield the voltage integral indicative of magnetic moment.
Thus, a presently preferred exemplary embodiment of the present invention provides a combination of the extraction technique for DC moment measurement with the AC susceptibility measurement using a common sensing structure--all within an instrument providing state-of-the-art electronics and computer control. By way of non-limiting example, applicants' presently preferred embodiment system provides the following advantages:
A single instrument, requiring no hardware reconfiguration between measurement modes, that is capable of measuring both AC and DC magnetic response of materials;
AC and DC measurements can be made without removing sample from dewar--thereby eliminating possible errors due to changes in set-up, changes in sample condition, etc.;
Use of common components (e.g., coil assembly) permits common calibration factors to be used for both AC and DC measurements;
Calibration constants for both AC and DC measurements can be calculated from considerations of coil geometry alone (unnecessary to calibrate with standard magnetic materials--since the coil assembly is precision wound and the system has been designed to avoid extraneous effects);
Stepper motors and associated sample suspension/mounting structure used for DC magnetization extraction technique also used for moving sample during AC measurements in order to cancel out differences between the two coils;
Dual opposed secondary coils cancel magnetic field noise;
Non-zero voltage offsets within the digital voltmeter during DC measurements are cancelled during subsequent noise reduction analysis--permitting higher speed digital voltmeter data acquisition in order to minimize "dead time" between measurements and thereby increase measurement sensitivity;
AC susceptibility measurement sensitivities comparable to or greater than those achievable using SQUID magnetometers (e.g., 2.times.10.sup.-8 emu);
High dc measurement sensitivity levels (e.g., 5.times.10.sup.-5 emu) can be achieved with an effective dynamic range extending to >10.sup.3 emu, comparable to Vibrating Sample Magnetometers;
Broad dynamic range (10.sup.-8 to >10.sup.3) to permit a wide array of material properties to be studied;
AC susceptibility measurement over a wide range of temperatures (e.g., <4.2K. to 325K.), amplitudes (e.g., 0.1 A m.sup.-1 ()0.00125 Oe) to 1600 Am.sup.-1 (20 Oe) RMS), and frequencies (e.g., 1 Hz to 10 kHz);
DC moment measurement over a wide range of temperatures and DC fields (e.g., 1.0 tesla or 5.0 tesla, plus or minus);
Capability of measuring harmonic susceptibilities, and AC and DC resistance (e.g., Hall effect, Transport J.sub.c, magnetoresistance);
For AC measurements, primary coil is driven with an AC current source so that the resultant AC field depends only on the "constancy" of the current source output--thereby eliminating complicated phase relationships dependent on measurement frequency or temperature;
Virtual elimination of effects of eddy current generation in conductive materials or generation of persistent currents in superconductive solenoids inductively coupled to the secondary coils;
Fully automated for unattended operation with data acquisition and control software that is quickly and easily tailored to address specific research requirements;
Possible to input sample parameters (e.g., volume, mass) and demagnetization factors to assure that the resultant measurement is as accurate as possible;
Wide range of materials and applications. The system is well suited for studying paramagnetic and ferromagnetic materials, amorphous alloys and diluted magnetic semiconductors, organic ferromagnets and organic superconductors (C.sub.60 compounds), conductive polymers, thin film recording media, etc.
Ability to flexibly configure a common instrument platform for AC measuring capabilities, DC measuring capabilities, or both--without sacrificing accuracy of either measurement mode.
Expandable configuration to permit purchasers to upgrade measuring system by adding additional measurement mode capabilities subsequent to purchase.
Applicants have thus developed a single instrument that incorporates both a dc moment measurement scheme and an ac susceptometer. The instrument provides a sensitive dc measurement capability which can be simply implemented, without sacrificing any of the performance characteristics of the ac measurement.
After considering a number of standard dc moment measurement schemes, applicants chose an extraction technique. Extraction refers to many variations of a basic method, but generally involves moving (extracting) a magnetized sample from within a sensing coil. The voltage induced in the coil is detected and integrated over time to yield the total flux change in the coil. The flux change is directly related to the magnetic moment of the sample. An extraction method is attractive for this application since all required experimental hardware is already in place for the ac susceptibility measurement. The only missing component is the means to detect the voltage induced in the sensing coils. After reviewing experimental requirements and instrument specifications, applicants decided to use a high speed digital voltmeter (DVM). The DVM specifications indicated that performance comparable to, or better than, a commercial fluxmeter could be achieved.
One aspect of the present invention thus provides a relatively simple arrangement to provide adc moment measurement capability in an ac susceptometer requiring minimal effort and hardware changes. Performance is comparable to many systems constructed solely as adc magnetometer. An important element to the measurement in the preferred embodiment is the use of a high speed digital voltmeter (DVM) for the signal analysis. In addition to providing the required resolution and sensitivity needed for the moment measurement, the features of the DVM can be used to add further capabilities to the system at minimal expense. For example, with the addition of a sample probe, a dc resistance measurement can now be performed.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other features and advantages of the present invention will be better and more completely understood by referring to the following detailed description of a presently preferred exemplary embodiment in conjunction with the drawings, of which:
FIG. 1 is a schematic illustration of an exemplary prior art DC magnetometer;
FIG. 2 is a graphical illustration of exemplary magnetization curves for two different samples;
FIG. 2A is a schematic block diagram of an exemplary prior art ac susceptometer arrangement;
FIG. 3 is a schematic block diagram of a presently preferred exemplary embodiment of a magnetic measuring system in accordance with the present invention;
FIG. 4 is a schematic illustration of an exemplary sample coil arrangement shown in FIG. 3;
FIG. 5 is a schematic diagram of an exemplary cryostat as shown in FIG. 3 mounted within a helium dewar;
FIG. 6 is an elevated view of an exemplary sample mounting arrangement shown in FIG. 5;
FIG. 7 is a flowchart of exemplary program control steps performed by the preferred embodiment computer to provide a dc magnetization measurement;
FIG. 8 is an exemplary graphical illustration of a voltage pulse recorded as the sample is moved from one secondary coil to another in the FIG. 3 embodiment;
FIGS. 9-12 are schematic diagrams of exemplary menu option breakdowns suitable for providing dc measurement capabilities in the FIG. 3 embodiment;
FIGS. 13 and 14 are, respectively, a plot of moment as a function of applied field for a NIST MnF.sub.2 standard sample, and a plot showing deviation of such data from a linear fit, each of these plots being generated by the presently preferred exemplary embodiment of the present invention;
FIG. 15 is a plot of actual experimental results showing moment as a function of temperature for an exemplary single crystal YBCO sample at an applied field of 10 Oe; and
FIG. 16 is a plot of exemplary hysteresis curve experimental results for a thin film YBCO sample at 4.2K (open circles), 25K (solid circles), and 50K (triangles) collected during a single run of the preferred embodiment under automatic operation.
DETAILED DESCRIPTION OF A PRESENTLY PREFERRED EXEMPLARY EMBODIMENT
FIG. 3 is a block schematic diagram of a presently preferred exemplary embodiment of a magnetic measuring system 100 in accordance with the present invention. FIG. 3 uses common reference numerals to designate elements that are similar or identical to those shown in prior art FIG. 2A. However, the fact that common reference numerals are used between the two Figures does not necessarily mean that elements designated with common reference numerals are identical in all respects. For example, even though computer 112 appears in both prior art FIG. 2A and in FIG. 3, the FIG. 3 computer executes different software to provide additional/different functionality (e.g., capability to measure dc magnetization) such that it is not identical to the FIG. 2A computer.
Referring to FIG. 3, system 100 includes a cryostat 102 housing a coil assembly 104; a stepping motor 106 mechanically coupled to a sample probe 108; a control unit 110; a computer 112; a superconducting magnet 114 and associated magnet power supply 116; a temperature controller 118 and associated temperature probe 120; an AC lock-in amplifier 122; and a digital voltmeter (DVM) 124.
Referring now to FIG. 3 and to FIG. 4 (a simplified cutaway view of coil assembly 104), two sensing (secondary) coils 126a, 126b are identical but oppositely wound on a sapphire tube 128 in the preferred embodiment. Each coil 126a, 126b is perfectly wound and contains approximately 1600 turns with a mean diameter of 1.1 cm and a length of 1.9 cm in the preferred embodiment. Secondary coils 126a, 126b are positioned coaxially with a center-to-center distance of 3.8 cm in the preferred embodiment. A 1300 turn primary coil 130 is wound directly over the top of the two secondary coils 126a, 126b in the preferred embodiment. Thus, the three coils 126a, 126b and 130 are all coaxial with one another, with primary coil 130 being wound over and covering both of secondary coils 126a, 126b. As will be appreciated, the spacing between secondary coils 126a, 126b is such that when sample 140 is within the space defined within the interior of secondary coil 126a, it is outside of the interior space within (and is also magnetically isolated from) secondary coil 126b. However, so long as sample 140 is at least partially positioned within the interior space within any of secondary coils 126a, 126b, the sample is fully within the interior space within primary coil 130.
The primary coil 130 is used in the preferred embodiment to generate the ac field for ac susceptibility measurements, but it can also be used to generate low level dc fields for the extraction type DC magnetization measurement. The overall design of the coil assembly 104 in terms of its physical size and number of turns is critical for the ac measurement (as those skilled in the art will readily recognize), as it determines the frequency response and ac field range.
Temperature sensors 120 and control heaters 132 are also mounted with the coils 126, 130 on the outside of the sapphire tube 128. The entire coil assembly 104 is wrapped with coil foil and superinsulation and sealed at one (i.e., the lower) end 134 in the preferred embodiment. The sapphire tube 134 includes an open end 136 through which the sample probe may descend into an interior space 138 within the center of coil assembly 104. As will be explained in detail below, the sample probe 108 is positioned and moved by stepping motor 106 so as to precisely position and move a sample 140 within sample space 138 with respect to secondary coils 126a, 126b.
FIG. 5 is an exemplary cut-away view of cryostat 102 shown in FIG. 3 mounted within a helium dewar (refrigerator) 142. As shown in FIG. 5, the open end 136 of the sapphire tube 128 is attached to the lower end 144 of a stainless steel tube 146 inside a vacuum jacket 147. This permits sample space 138 to be isothermal and thermally isolated from the cryogenic (e.g., liquid helium) bath 148. The sample space 138 is normally operated with helium exchange gas inside to provide the necessary thermal coupling between the sample 140 and the thermometers 120. Controlled temperatures from below 4.2K to 330K can be achieved. A vacuum line/wire feed 150 is coupled to a vacuum isolation space 152 within vacuum jacket 147 to permit vacuum to be drawn from the isolation space and to provide a path for electrical conductors coupling external equipment to the superconducting magnet 114, the coil assembly 104, and the other electrical components within cryostat 102.
FIG. 6 is an elevated side view of an exemplary preferred embodiment sample probe 108. Sample probe 108 descends, in the preferred embodiment, through stainless steel tube 146 into sample space 138. The preferred embodiment provides access to the secondary coils 126a, 126b and sample space 138 from above through a room temperature load seal assembly 156 and associated sample probe load seal (not shown) using a sample rod 154. Sample probe load seal assembly 156 in the preferred embodiment includes an O-ring seal 158 located on a lower portion thereof that makes a vacuum tight seal between the probe seal and the load seal. Side-mounted opposed thumb screws (not shown) in the load seal pull the probe seal 156 downward to compress the O-ring, and also hold the seal snugly in place to prevent any relative movement between the load seal and the probe seal during sample movement.
Sample rod 154 in the preferred embodiment is a 0.64 cm diameter polished stainless steel rod with a nylon extension 160 and sample mount 162. The sample rod 154 slides through vacuum-tight Teflon O-ring probe seal assembly 156, permitting the sample 140 (which is disposed within sample mount 162) to be moved up and down while maintaining a vacuum seal in the sample space 138. This arrangement allows samples 140 to be exchanged while the system is at cyrogenic temperatures. In the preferred embodiment, sample mount 162 includes a nylon (e.g., Delrin) bushing 164, a Delrin sample holder 166, and a holder lid (not shown, may comprise a #10-24 threaded Delrin rod for mechanically coupling bushing 164 to sample holder 166).
Once the sample 140 has been inserted into the sample probe 108 and the sample probe has been inserted into the sample space 138 within the secondary coils 126, an upper, threaded end 168 of the sample rod 154 is attached to stepping motor 106 via a finder nut 170. Stepping motor 106 positions and moves the sample 140 between the two secondary coils 126a, 126b under control of control unit 110. In the preferred embodiment, the ac measurement technique uses a two-position measurement scheme to eliminate uncertainties which arise from slight imbalances between the two secondary coils 126a, 126b. In the preferred embodiment, this same movement mechanism (i.e., stepping motor 106 and associated mechanical coupling) is used in the dc moment measurement.
Referring once again to FIGS. 3 and 5, a 7.6 cm diameter 1 tesla superconducting magnet 114 is mounted on the outside of the vacuum space 152 in direct contact with the liquid helium bath 148. In the preferred embodiment, the magnet 114 is designed to have a uniformity of better than .+-.0.1% over the full axial length of the secondary coils 126a, 126b. The large diameter is used to minimize inductive coupling to the primary and secondary coils 126, 130 which adversely affects the ac measurement. The magnet 114 is used in the preferred embodiment both for applying, to the sample, the magnetic field required for the dc moment measurement; and also for applying a dc bias field to the sample when making ac measurements. In order to vary fields easily and rapidly, a persistent switch is not used in the preferred embodiment.
Referring once again to FIG. 3, system 100 is designed for automatic operation and all control is done through computer 112. The sample movement, temperature control, field control, and data acquisition are all executed under software control by computer 112. Magnet power supply 116 selectively provides power to superconducting magnet 114 under control of computer 112 (e.g., via an RS-232 serial link 171). In the preferred embodiment, magnet power supply 116 comprises a convention Lake Shore Model 610 or 612 magnet power supply.
In addition, computer 112 is connected, in the preferred embodiment, via an IEEE 488 interface bus 172 to permit it to control the operations of temperature controller 118, control unit 110, AC lock-in amplifier 122, and DVM 124. In the preferred embodiment computer 112 comprises a general purpose off the shelf personal computer (e.g., the Hewlett-Packard VECTRA (TM) computer and associated monitor and keyboard) that includes an IEEE-488 (GPIB) board and internal hard disk (not shown). Software stored on the internal hard disk permits computer 112 to control the operations of system 100.
In the preferred embodiment, AC lock-in amplifier 122 comprises a commercially available EG&G Model 5209 Lock-in amplifier that is controlled by computer 112 via bus 172. Briefly, lock-in amplifier synchronizes with an AC reference (excitation) signal provided by control unit 110 in order to "lock in" to and amplify a version of the ac excitation signal received by secondary coils 126a, 126b. Lock-in amplifier 122 provides a highly accurate measurement of the parameters of the AC signal it is "locked in" to (synchronous detection being used to eliminate noise effects and to make the instrument phase sensitive so as to permit imaginary susceptibility components to be detected, as is well known).
Control unit 110 preferably comprises a Lake Shore model 140 or 710 ACS Control Unit that includes an internal DC current source 110a, AC current source 110b, and motor control 110c. In the preferred embodiment, the outputs of the AC and DC current sources 110a, 110b are coupled to primary coil 130, and these current sources are operable under direct, programmable control by computer 112. Thus, computer 112 may cause an adjustable amount of AC and/or DC current to primary coil 130 by controlling DC current source 110a and AC current source 110b via bus 172. In the preferred embodiment, AC current source 110b provides a reference current output that is applied to a "REF AC In" terminal of Lock-In Amplifier 122 (as described above).
Computer 112 also controls the operation of stepping motor 106 by writing appropriate information to control unit motor control block 110c. Stepper motor 106 in the preferred embodiment includes a movement motor head (not shown) providing a conventional stepping motor coupled to axially displace a threaded shaft, the threaded shaft being mechanically coupled to sample probe finger nut 170. The movement motor head preferably includes travel limit switches (not shown) of conventional design to limit the vertical travel of sample probe 108.
In the preferred embodiment, the secondary coils 126a, 126b are connected, in series, alternately and selectively to either lock-in detector amplifier 122 for the ac measurement; or to digital voltmeter 124 for dc measurement. Such alternate connections are accomplished in the preferred embodiment via connector network 123 (a switching network is avoided in the preferred embodiment in order to reduce noise; instead, the user manipulates cables to manually disconnect coils 126a, 126b to either the lock-in amplifier or to the digital voltmeter depending upon which measurement, ac or dc, is desired). In the preferred embodiment, DVM 124 comprises a modified Keithly Model 182 Sensitive Integrating Digital Voltmeter. This DVM 124 provides an output which is numerically processed by computer 112 to provide an integrated voltage value.
The Keithly Model 182 in unmodified form has excellent sensitivity, but does not sample rapidly enough to provide desired sensitivity in the preferred embodiment because it provides a substantial "dead time" between samples during which auto-zeroing compensation routines are performed (i.e., in order to compensate for non-zero voltage offsets generated within the meter itself so as to provide absolute voltage readings). The preferred embodiment employs a Keithley meter that has been modified to reduce "dead time" between samples by eliminating zero offset compensation functions (such that the DVM does not provide absolute voltage readings but instead provides only relative voltage readings). Thus, the preferred embodiment modified DVM 124 spends a higher percentage of the time making measurements (thereby increasing the amount of time during a given measuring sequence during which the output voltages provided by sensing coils 126a, 126b are being sensed). As a consequence of this modification, the modified DVM may, when connected to a zero input voltage, tend to drift between slight positive and slight negative voltage readings due to drifting within the input amplifiers and other associated stages of the meter itself. The preferred embodiment compensates for such meter offset voltage drifting - and at the same time compensates for background DC voltage picked up by secondary coils 126a, 126b - such that the absolute voltage readings of the Keithly DVM can be sacrificed for reduced "dead time" realizable by decreasing the amount of overhead processing performed in the "dead time" between successive samples. Although the modified meter drifts slightly over time, it provides sufficient stability over the time a measurement is actually being made so that errors in the relative voltage measurements it provides during a particularly magnetic characteristic measurement are relatively negligible. While not presently preferred due to its decreased sensitivity relative to the modified Keithley DVM, it is also possible to use a Hewlett Packard Model HP 3458A integrating DVM since it may give satisfactory performance and may have its own specific advantages depending upon what is desired from the measurement. In the preferred embodiment, "integrating" DVM 124 integrates the output of sensing coils 126a, 126b over relatively short time periods during measurement to provide many successive samples to computer 112. Computer 112 numerically processes these successive samples (as described below) to provide a value indicating dc moment.
Principles of Operation
A. DC Magnetization
Superconducting magnet 114 is activated during DC measurements in order to apply a constant, uniform magnetic field to sample 140. Primary coil 130 is typically deactivated during DC measurements, although it is sometimes desirable to apply a controlled amount of DC current to primary coil 130 in order to provide "fine" adjustment of the applied magnetic field. Thus, in the preferred embodiment magnet power supply 116 has an output that is adjustable in coarse increments; DC current source 110a permits "fine" magnetic field bias adjustments to permit the operator to, for example, measure more (or specific user-selected) points on the sample's magnetization curve.
The "extraction method" is used during the dc measurement - such that stepping motor 106 moves the sample from the axial center of secondary coil 126b to the axial center of secondary coil 126a (and, in the preferred embodiment, back to the axial center of secondary coil 126b) during the time DVM 124 is acquiring a measurement. Sample movement results in a change in magnetic flux that is detected by secondary coils 126a, 126b and induced voltage measured by DVM 124 to provide a measure of the net magnetic moment of the sample. Secondary coils 126a, 126b are wound identically (albeit in opposite directions) and are intended to have identical characteristics so that background voltage (e.g., due to stray magnetic fields) induced in one coil is cancelled by the same background voltage induced in the other coil.
In the extraction measurement, the voltage as a function of time induced in the secondary coil by the moving sample [.upsilon.(t)] is not as important as the determination of the integral of the voltage over time. This integral gives the total magnetic flux (.PHI.) change through the coil due to the sample movement:
Note that the flux change will be independent of how the movement is executed; the result only depends on the starting and stopping points of the sample 140. Variations in the mechanics of the movement may change the shape and appearance of .upsilon.(t), but not the value of the integral. The use of two oppositely wound secondary sensing coils 126a, 126b as in the preferred embodiment will double the integral in (1), but more importantly will minimize induced noise from the environment and the applied field.
The magnetic flux can be related to the magnetic moment (m) of the sample through a calibration coefficient, .alpha.:
The calibration coefficient can be determined experimentally with known magnetic standard samples, or, if the coil and sample geometry are known, a value for .alpha. can be calculated.
As mentioned above, during the DC measurement the sample 140 is moved from the center of one of the secondary coils 126a, 126b to the center of the other secondary coil. Thus, the sample 140 remains within the region of the field uniformity created by superconducting magnet 114 and/or primary coil 130 at all times (since the secondary coils 126 each lie entirely within the interior space defined by superconducting magnet 114 in the preferred embodiment, and also each lie entirely within the interior space defined by primary coil 130). This movement also minimizes uncertainties related to sample positioning since the voltage induced by a sample at a coil center is effectively zero. Note this will not be generally true for every two-coil system but depends on the coil/sample geometry and the flux coupling between the sample and each of the two secondary coils 126a, 126b.
Initially, it is necessary to position the sample 140 to be in the center of one or the other of secondary coils 126a, 126b - and for the operator to inform computer 112 which secondary coil the sample is positioned within. To initially position the sample 140 prior to a DC measurement, the operator sets a DC field and selects a "position" routine in the control software. The operator then uses his or her "best guess" to select which secondary coil 126 (upper or lower) the sample is believed to be within. A test scan is then performed to automatically move the sample by a displacement corresponding to the distance between secondary coils 126a, 126b. This will yield voltage peak outputs of the type shown in FIG. 8. For a properly positioned sample, the voltage peaks will be perfectly symmetrical. for in incorrectly positioned sample, on the other hand, non-symmetrical peaks will result. In the preferred embodiment, an improperly positioned sample 140 should be repositioned by moving the sample up or down as needed. This process is repeated until a symmetrical voltage peak curve is obtained.
This positioning technique may be difficult to perform for samples with very low signals. When the baseline "scatter" is comparable to the signal due to small sample size, it may be possible to increase the field strength in order to increase signal level. As a least resort, manual positioning of sample rod 154 (e.g., through a process of physically measuring the length of the sample, relative to the distance between secondary coils 126a, 126b, and marking the rod accordingly) may be used to properly position the sample.
In order to obtain a measurable voltage signal (i.e., a voltage signal of a sufficiently high level to be measurable), a rapid sample movement is required. However, there is a limit as to how fast the stepping motor 106 can move the sample rod 154 through the relativeiy tight o-ring seal 156. Changing to a different movement mechanism was considered, but the benefits of a stepping motor 106 outweighed the constraints it placed on the measurement. A sample velocity of several cm/sec was determined sufficient to meet the desired moment specifications. Except as noted, all measurements were performed with a sample velocity of 2.4 cm/sec.
FIG. 7 is a high level flowchart of measuring steps performed by system 100 to provide a DC magnetization measurement. Initially, computer 112 controls magnetic power supply 116 (and possibly also DC current source 110a) to provide generate a constant magnetic field of specified intensity (block 200). Secondary coils are coupled to DVM 124 (via connection network 123), and the DVM is programmed to integrate with desired sampling times and for a desired overall integration time (block 202). The voltage induced in the secondary coils 126a, 126b is logged with high speed integrating voltmeter 124. Voltmeter 124 allows computer 112 to specify the time period through which the voltmeter integrates the signal (.tau.) and the time interval between the start of each measurement (.DELTA.t). Note, by necessity, .DELTA.t>.tau.with the difference representing the integral processing time required for the DVM instrument 124 to complete a single measurement. For best noise rejection, .tau. is set at an integral multiple of the power line cycle (PLC). For the maximum reading rate, .tau.t is set to the minimum permitted by DVM 124. All filtering, autoranging, and other features (e.g., auto zeroing) which may slow the reading rate are disabled.
In the preferred embodiment, computer 112 controls DVM to begin monitoring voltage prior to controlling stepping motor to begin moving the sample 140 (blocks 204, 206). In a single "scan" the sample is typically moved from the center of bottom coil 126b to the center of top coil 126a and then back to the center of bottom coil 126b. During a sample movement in the preferred embodiment, a series of approximately one hundred readings characterizing .upsilon.(t) are recorded and stored to the internal memory of DVM 124. Computer 112 then controls stepping motor 106 to cease moving the sample 140 (block 208), and finally, controls DVM to cease monitoring voltage (block 210). The voltage readings are then read back into the computer 112 (block 212) for processing.
The "scan" of blocks 204-212 may be repeated several (e.g., up to 10) times to provide a single averaged value (e.g., for measuring samples with low level signals or when extra precision is required). The preferred embodiment system 100 permits a "half scan" mode (e.g., each movement of the sample 140, either up or down, defines one moment measurement). Such "half scans" are useful when rapid data acquisition is required (e.g., during a rapid field sweep). Since the accuracy and repeatability of the measurement is degraded somewhat when "half scan" is used, however, this mode is recommended for large samples or when non-critical measurements are being made.
The first step of the processing by computer 112 involves the elimination of uncertainties related t thermal emf's. The thermal emf's and any zero offsets are easily removed by monitoring the voltage in the sensing coils for a short period of time both before the movement starts (the time between performing blocks 204 and 206 shown in FIG. 7 is timed in the preferred embodiment) and after the movement stops (the time between performing blocks 208 and 210 of FIG. 7 is thus also timed by computer 112 in the preferred embodiment). The readings are linearly fit with respect to time in the preferred embodiment (FIG. 7 block 214) to create a baseline voltage which is subtracted from the measured voltages. This yields the true voltage data due to only the sample movement. Typical results are shown in FIG. 8. The voltage referred to in the description below is the voltage after correction for thermal emf's.
Each voltage measurement made by the DVM(.upsilon..sub.i) is actually a mean integral value over the integration period: ##EQU1## where t.sub.i denotes the start of the measurement process. What is required from (1) is the voltage integral over the complete pulse consisting of n voltage readings. This can be determined as follows: ##EQU2## The second term in the summation, .epsilon..sub.i contains the contribution to the integral corresponding to the period .DELTA.t-.tau. the DVM is not integrating the voltage. If this time period is short, .epsilon..sub.i, can be evaluated by approximating the voltage V(t) as the means value of the two voltage readings which span the time period.
Substituting into (4), rearranging, and taking note of the fact that the initial and final voltage readings are zero yields the final result: ##EQU3##
This final expression appears exactly like a numerical integration using a trapezoidal approximation. However, what is physically occurring is not a numerical integration. This is a subtle but important distinction to understand. Since each reading of DVM 124 is actually proportional to the voltage integral over the time period .tau., the analysis using (6) will yield better results than what a strict numerical integration would imply. In a numerical integration, the error is a function of the data point spacing and curvature. The numerical error in (6) is due to the voltage approximation made in (5) and the magnitude of .DELTA.t-.tau., but not necessarily the number of data points.
The numerical error is important to understand because it affects both the absolute accuracy of the measurement and the repeatability from one movement to the next. If the voltage measurements were perfectly synchronized with the sample movement, each .upsilon..sub.i would always occur at exactly the same point in the voltage pulse. The evaluation of (6) would always give exactly the same numerical result but would be in error by the approximations made in (5). However, perfect synchronization would require timing to the millisecond level between the mechanical and electronic components. The effort to guarantee this is not justified, so in practice the .upsilon..sub.i fall at different points along the pulse for each sample movement. Slightly different numerical errors then arise in the evaluation of (6) which creates a non-repeatability associated with the numerical processing.
The time required to complete a moment measurement is dependent on how the measurement sequence is defined. A single scan measurement requires about 15 seconds to execute. Multiple scans may increase the measurement time (e.g., up to 150 seconds if 10 scans is selected). Computer 112 generates a moment value at the conclusion of the moment measurement sequence (block 216 FIG. 7).
The experimental feasibility of this approach and analysis was initially tested with the HP 3458A set to a 1 PLC (16.666 msec) integration time as DVM 124. The time between measurements could be set to as low as 16.69 msec, yielding an active DVM integration time over the total pulse of 99.86%. A permanent magnet sample was used to supply an output signal well above the noise floor of the voltmeter and the sample was moved at 1 cm/sec. The voltage integral was approximately 0.0104 volt-sec. Repeated movements of the sample showed a standard deviation of less than 25 ppm in the measurement and evaluation of (6). Also, as predicted in (5), with At set to 20 msec, the standard deviation for repeated measurements increased to 150 ppm.
The effect of varying the integration time on the DVM was also tested using this same experimental set-up. The integration time was varied from 1PLC to 40 PLC while the time between readings was fixed at 0.023 milliseconds greater than the integration time. The results, consistent within .+-.20 ppm, verified the predicted independence on the number of voltage readings. Over two hundred points were recorded at 1 PLC while only 5 or 6 were recorded at 40 PLC. In all situations encountered to date, evaluating the integral by summing over a large number of readings yields the same or slightly better repeatability than using longer integration times with a fewer number of voltage readings. For this reason, a 1 PLC integration time has been adopted in the preferred embodiment.
The voltage noise floor and sensitivity limit can be determined by making repeated measurements with no sample present and comparing the standard deviation in the integral evaluations. Ideally, the results should be zero. The HP 3458A yields a lower limit of .+-.2.5.times.10.sup.-.perspectiveto. volt-seconds in the evaluation of the integral. This corresponds to a moment sensitivity of approximately .+-.10.sup.-3 emu (10.sup.-6 Am.sup.2). In the discussion of the preceding paragraphs, the standard deviation of 25 ppm reflects this noise floor and is not necessarily the limit to the reproducibility which might be achieved for large samples.
Most demands are for higher sensitivity as opposed to high resolution. For this reason, the final system configuration (as mentioned above) actually uses a modified Keithley 182 Sensitive Voltmeter instead of the HP 3458A. The Keithley instrument offers approximately a factor of ten improvement in voltage sensitivity and resolution, yielding a lower limit of .+-.2.5.times.10.sup.-8 volt-seconds in the integral evaluation. This equates to about .+-.10.sup.4 emu (10.sup.-7 Am.sup.2) in moment sensitivity. Voltages from the sensing coils of less than a microvolt can be readily detected and measured.
The increased sensitivity of the Keithley is at the expense of some speed and versatility. With the Keithley, voltage readings are limited to approximately 30 millisecond intervals. At 1 PLC integration times, the voltage is only integrated over 55% of the total time. Numerical errors associated with the approximation in (5) then limit the accuracy and repeatability of the integral determination to about .+-.0.1%. However, this is more than sufficient for most applications.
It will be understood that in some applications it may be desirable to change the magnetic field intensity between (or during) dc moment measurements. For example, it may be desirable to provide a magnetic field table to permit system 100 to automatically step through a series of predefined magnetic fields (so as to measure moment at various points along the magnetization curve of the sample). The preferred embodiment typically provides "wait"periods between field changes to permit the field to stabilize prior to making the next successive measurement, and also preferably directly monitors the current output of magnet power supply 116 to ensure accuracy. It is also possible to control magnet power supply to "ramp" at a user defined rate in order to "sweep" the magnetic field during a measurement (this may be useful to generate hysteresis curves for example).
CALIBRATION
In order to complete the moment measurement, the value of the calibration coefficient in equation (2) must be determined. The value of .alpha. is a measure of the flux coupling between the sample and the secondary coils and will vary with the sample size and geometry. A proper calibration therefore requires a separate calibration coefficient for each sample geometry measured.
However, samples are often small with respect to the sensing coils 126a, 126b and in these situations they can be approximated by magnetic dipoles. With this assumption and the simple secondary coil 126 geometry employed in this system, an expression for .alpha. can easily be derived in closed form. The calculations are similar to those used to derive an ac calibration coefficient. A magnetic dipole is assumed centered in one of the secondary coils and the field is integrated over the geometry of both the secondary coils 126a, 126b to give the total magnetic flux. The flux contained in the empty coil represents only 0.6% of this total, indicating that the two secondary coils 126a, 126b are nearly independent of each other. The total flux change when the sample is moved to the second coil is twice the calculated value. This gives a value for .alpha. of approximately 5300 emu/volt-sec (5.3 Am.sup.2 /volt-sec).
Larger samples with simple geometries, such as cylinders, can also be handled numerically. These types of calculations indicate that the dipole approximation is actually a very good approximation for most applications. For the present coil geometry and for samples up to 5 mm diameter and less than 10 mm long, the potential error due to the dipole approximation is less than 3%.
Since numerical values for .alpha. can be effectively calculated, the measurements made with preferred embodiment system 100 are absolute determinations of the moment. The accuracy is limited by the accuracy of the calibration coefficient which is estimated at 1 to 2 percent. The dimensions of the coil arrangement 104 are known to better than 1% and the effects of thermal expansion and contraction are only a few tenths of a percent. The two sensing coils 126a, 126b are matched to within 0.2% in the preferred embodiment as determined from their ac response or by measuring the induced voltage as the magnetic field is ramped. This level of calibration accuracy is comparable to what is often accomplished with standard samples.
For completeness, the calculated value for the calibration coefficient has been confirmed experimentally using NIST standard Ni and MnF.sub.2 samples. The agreement is within the experimental uncertainty.
In a properly designed system which avoids eddy currents and other ac specific problems, the ac calibration follows directly from the dc calibration and vice versa. Both calibration coefficients relate the flux coupling between the coils and the sample. In the preferred embodiment, the DC calibration coefficient used is equal to .pi. times the ac calibration coefficient. Thus, a single calibration coefficient and associated analysis can be used for both ac and dc measurements. This has been tested experimentally by verifying that ac measurements are consistent with dc measurements for samples which show no ac related effects.
Preferred embodiment system 100 also provides a capability to measure the background signal attributable to the sample mount 162, sample rod 154, and anything else that may generate a background signal (e.g., substrate material etc.) during a moment measurement. The capability for subtracting the background signal form the data being recorded with the sample present is provided by software executing on computer 112. Such "addenda data" should be recorded using fixed temperature points over an applicable temperature range, and at least two applied fields which span the intended range of use (where one field can be zero field). A special option in the calibration analysis provided by the preferred embodiment processes the "addenda" data and stored the temperature/field/moment data to a file for later use. System 100 will automatically subtract the addenda data from a measured sample moments upon demand. The operator need only check the temperature T and field H at which the sample data was recorded. System 100 then uses the temperature/field/moment data in the addenda file to determine the addenda moment at temperature T and field H (i.e., a simple two-dimensional linear interpolation is performed); and subtracts the addenda moment from the actual sample moment.
FIGS. 9-12 show exemplary software menu structures provided by software executing on computer 112 to permit preferred embodiment system 100 to make dc magnetization measurements. Appendix A provides a listing of menu breakdown and functional description of FIGS. 9-12.
EXPERIMENTAL RESULTS
The dc moment measurement is made by making one complete movement cycle in the preferred embodiment; the sample is moved from the bottom to the top coil and the returned to the bottom. Two voltage integral evaluations are made and averaged to give the final moment output for the sample. This process requires approximately 15 seconds to complete. Averaging over multiple measurements improves the sensitivity and noise with a lower limit of <5.times.10.sup.-5 emu (5.times.10.sup.-.sub.8 Am.sup.2). With the voltage ranges available in the DVM, there is no practical upper limit to the measured moment.
FIGS. 13 and 14 illustrate with a NIST MnF.sub.2 standard sample what level of performance can be achieved. FIG. 13 shows the measured moment versus the applied field. The data should be linear over the range of fields used, and FIG. 14 shows the data plotted as a deviation from a straight line fit. The standard deviation about the line is 6.times.10.sup.-5 emu - illustrating the excellent linearity in the measurement.
Repeated measurements over the course of an hour on a sample at fixed field gives reproducibilities of better than .+-.0.1%. When all experimental aspects are considered, including data processing, sample positioning, field setting, etc., the overall measurement reproducibility is a few tenths of a percent.
Image effects have been reported in superconducting magnets, and measurements were made to see if these effects could be detected. No effect was observed (<0.1%) which is probably due to the relatively large bore of the magnet 114 with respect to the sensing coils 126a, 126b.
FIGS. 15 and 16 show typical dc moment results to further illustrate the capability and flexibility of the system. FIG. 15 shows moment as a function of temperature for a single crystal YBCO sample at an applied field of 100e (the sample was zero field cooled and had a mass of 15.2 mg). FIG. 16 shows experimentally measured hysteresis curves for a thin film YBCO sample at 4.2K (open circles), 25K (solid circles), and 50K (triangles). Note the low dc field used in the data of FIG. 15 and the smoothness of the hysteresis loops of FIG. 16 for the samples with relatively low moments.
B. AC SUSCEPTIBILITY
System 100 performs ac susceptibility measurements in a manner that is substantially identical in principle to the ac measurement techniques used since 1988 in conjunction with Lake Shore's Model 7000 AC Susceptometer product (FIG. 2A is a block diagram of this prior art ac susceptometer arrangement showing similar or identical components to those used in the FIG. 3 preferred embodiment of the present invention).
Briefly, to perform an AC susceptibility measurement using the FIG. 3 preferred embodiment, the sample 140 is placed within sample space 140 and AC current source 110b is activated to provide a desired AC field to the sample via primary coil 130. Since the applied AC field is changing, the magnetization of the sample 140 changes in response. Secondary coils 126a, 126b measure changes in the magnetic field due to the susceptibility of sample 140 (the magnetization of which "follows" changes in the applied field may lag behind in an amount dependent on the magnetic response of the sample).
Superconducting magnet 114 does not need to be activated for AC susceptibility measurement. This is because most ac susceptibility measurements are desirably performed under low field conditions (although the superconducting magnet may be activated if high field ac measurements are desired, such as for "irreversibility line" type measurements). In addition, sample 140 is stationary during the time ac susceptibility parameters are being measured by lock-in amplifier 122.
In the preferred embodiment, stepping motor 106 is used to position sample 140 within the interior space of first one secondary coil 126b (at which time a first measurement is taken with sample 140 remaining stationary during the measurement); and is then used to position the sample within the interior space of the other secondary coil 126a (at which time a second measurement is made with the sample again remaining stationary during the second measurement). The two measurements are used by computer 112 to ascertain the magnitude and direction of the error between the two coils (such that the error can be cancelled out from one of the two measurements in Order to yield a highly accurate result). Hence, stepping motor 106 is used in the preferred embodiment during the ac susceptibility measurement process even though sample 140 is stationary during the time lock-in amplifier 122 is actually acquiring an ac measurement.
While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. ##SPC1##
Claims
What is claimed is:
1. A method of automatically characterizing ac and dc magnetic properties of a sample comprising the following steps: closely thermally linking a sensing coil arrangement to the sample; varying, using computer control, the temperature of the sensing coil arrangement and thereby also the temperature of the sample closely thermally linked thereto; automatically measuring, by coupling electrical signals to/from the sensing coil arrangement, an ac susceptibility characteristic of the sample; and automatically measuring, by coupling electrical signals to/from the sensing coil arrangement, a dc magnetization characteristic of the sample, wherein said measuring steps are automatically performed successively under said computer control without thermally unlinking said sample from said sensing coil arrangement.
2. A method as in claim 1 wherein said dc magnetization characteristic measuring step comprises using a computer to automatically move said sample from the center of a first sensing coil to the center of a second sensing coil to provide substantially symmetric signal measurements.
3. A method as in claim 1 wherein said dc magnetization characteristic measuring step comprises automatically moving said sample relative to said sensing coil arrangement under computer control.
4. A method as in claim 1 wherein said dc magnetization characteristic measuring step comprises digitally integrating, substantially in real time, sensing coil voltage over time to determine total magnetic flux change and correcting the resulting integral with a calibration factor to yield a dc moment value.
5. A method as in claim 1 further including the step of calibrating said ac characteristic measurement and said dc characteristic measurement with a common calibration.
6. A method as in claim 1 wherein said dc magnetization characteristic measuring step comprises integrating at least one voltage over time with an integrating digital voltmeter.
7. A method as in claim 1 wherein said ac magnetization characteristic measuring step comprises moving said sample relative to said sensing coil arrangement so as to magnetically couple said sample alternately to a first sensing coil and to a second sensing coil during said ac susceptibility measuring step.
8. A method as in claim 1 wherein at least one of said measuring steps comprises sampling and averaging several scans per measurement point.
9. A method as in claim 1 further including displaying a computer controlled display screen providing frequency, amplitude and harmonic measurement selection.
10. A method as in claim 1 further including performing multiple temperature-dependent susceptibility sequences at different field, followed by several dc measurements at different temperatures, for a single sample in a common measurement sequence without thermally unlinking said sample from said sensing coil arrangement.
11. A method as in claim 1 further including allowing the user to program a computer to set up a sequence of measuring conditions selected from the set of: cooling conditions, temperature conditions, field conditions, ac drive field, measurement type, harmonics, and real time plotting.
12. A method as in claim 1 wherein each of said measuring steps includes the step of recording measurement data in a computer data file.
13. A magnetic measuring system for measuring the ac and dc magnetic characteristics of a sample at cryogenic temperatures, said system comprising: a cryogenic chamber; at least one sensor coil arrangement disposed within said chamber, said sensor coil arrangement including a primary coil and plural, substantially balanced sensing coils; a computer; means for mounting said sample within said chamber and for moving said sample relative to said sensor coil arrangement under control of said computer; means for applying a constant magnetic field to said sample to make dc. magnetic characteristic measurements; and means, coupled to said sensing coils, for successively measuring dc magnetic characteristics and ac magnetic characteristics of said sample without requiring said sample to be removed from said chamber, said system further including ac applying means for selectively coupling an ac excitation signal to said primary coil during said ac magnetic characteristic measurements.
14. A system as in claim 12 wherein said mounting and moving means comprises a computer-controlled electric motor for automatically moving said sample from one to another of said plural sensing coils for at least said dc magnetic characteristic measurements.
15. A system as in claim 13 wherein sensor coil arrangement includes a sapphire coil form, said plural sensing coils being wound on said sapphire coil form.
16. A system as in claim 13 wherein said sensor coil arrangement comprises a longitudinally separated detection coil set comprising said plural sensing coils with primary coil wound around said set.
17. A system as in claim 13 wherein each of said plural sensing coils is made of copper.
18. A system for measuring ac and dc magnetic characteristics of a sample, said system comprising: a chamber; a dc field coil magnetically coupled to said chamber; at least one sensor coil arrangement disposed within said chamber, said sensor coil arrangement including an ac field coil and at least one sensing coil; electronics coupled to said dc field coil and to said sensor coil arrangement, said electronics comprising: (a) an ac signal processing arrangement that applies, during an ac measurement, an ac signal to said ac field coil and measures ac magnetic characteristics of said sample in response to signals received from said sensing coil, (b) a dc signal processing arrangement that applies, during a dc measurement, a dc signal to said dc field coil and measures dc magnetic characteristics of said sample, and (c) a computer connected to control said ac signal processing arrangement and said dc signal processing arrangement to perform said ac and dc measurements successively without requiring removal of said sample from said chamber; and a movable mechanical structure that retains said sample within said chamber and moves said sample relative to said sensor coil arrangement during said dc measurement.
19. A system as in claim 18 wherein said movable mechanical structure includes an electric motor controlled by said computer at least during said dc measurements.
20. A method for measuring ac and dc magnetic characteristics of a sample comprising the following successively performed steps: (a) disposing a sample within a chamber and magnetically coupling said sample to an ac field coil, a dc field coil and at least one sensing coil; (b) subsequently to said step (a), applying an ac signal to said ac field coil under computer control and measuring ac magnetic characteristics of said sample in response to signals received from said at least one sensing coil; (c) subsequently to said step (b), without removing said sample from said chamber, applying a dc signal to said dc field coil, moving said sample relative to said pickup coil under computer control, and measuring dc magnetic characteristics of said sample in response to signals received from said pickup coil; and (d) subsequently to said step (c), removing said sample from said chamber.
21. A system as in claim 20 wherein said disposing step sensor includes coupling said sample to a pair of balanced sensing coils, and each of said steps (b) and (c) comprises having a computer control an electric motor to move said sample relative to said pair of balanced sensing coils to obtain substantially symmetrical responses from each of said sensing coils.
22. A method for measuring ac and dc magnetic characteristics of a sample comprising the following successively performed steps: (a) disposing a sample within a chamber and magnetically coupling said sample to an ac field coil, a dc field coil and at least one sensing coil; (b) subsequently to..said step (a), applying a dc signal to said dc field coil, moving said sample relative to said at least one sensing coil under computer control, and measuring dc magnetic characteristics of said sample in response to signals received from said sensing coil; (c) subsequently to said step (b), without removing said sample from said chamber, applying an ac signal to said ac field coil under computer control and measuring ac magnetic characteristics of said sample in response to signals received from said at least one sensing coil; and (d) subsequently to said step (c), removing said sample from said chamber.
23. A system for measuring ac and dc magnetic characteristics of a sample, said system comprising: a chamber; a dc field coil; at least one non-superconducting sensor coil arrangement disposed within said chamber, said sensor coil arrangement including a non-superconducting ac field coil and two coaxial, oppositely wound non-superconducting sensing coils; electronics coupled to said dc field coil and to said sensor coil arrangement, said electronics comprising: (a) ac signal processing circuitry for, during an ac measurement, applying an ac signal to said ac field coil and measuring ac magnetic characteristics of said sample in response to signals received from said coaxial, oppositely wound sensing coils, (b) dc signal processing circuitry for, during a dc measurement, applying a dc signal to said dc field coil and measuring dc magnetic characteristics of said sample, and (c) a computer controller connected to at least in part control said ac signal processing circuitry and said dc signal processing circuitry, said computer controller controlling said system to successively perform said ac and dc measurements without requiring removal of said sample from said chamber; and a movable mechanical structure that retains said sample within said chamber and moves said sample relative to said sensor coil arrangement during said dc measurement, said computer controller automatically controlling said movable mechanical structure to move relative to said oppositely wound sensing coils during said dc measurement to provide substantially symmetrical outputs from said sensing coils.
24. A method for measuring ac and dc magnetic characteristics of a sample comprising the following successively performed steps: (a) disposing a sample within a chamber and coupling said sample to a non-superconducting ac field coil, a pair of coaxial, oppositely wound non-superconducting sensing coils, and a dc field coil; (b) subsequently to said step (a), applying an ac signal to said non-superconducting ac field coil under computer control and measuring ac magnetic characteristics of said sample in response to signals received from said pair of coaxial, oppositely wound non-superconducting pickup coils; (c) subsequently to said step (b), without removing said sample from said chamber, applying a dc signal to said dc field coil, moving said sample relative to said sensing coils under computer control, and measuring dc magnetic characteristics of said sample in response to signals received from said sensing coils; and (d) subsequently to said step (c), removing said sample from said chamber.
25. A method for measuring ac and dc magnetic characteristics of a sample comprising the following successively performed steps: (a) disposing a sample within a chamber and coupling said sample to a non-superconducting ac field coil, a pair of coaxial, oppositely wound non-superconducting sensing coils, and a dc field coil; (b) subsequently to said disposing step (a) applying a dc signal to said dc field coil, moving said sample relative to said pair of coaxial, oppositely wound non-superconducting sensing coils under computer control, and measuring dc magnetic characteristics of said sample in response to signals received from said sensing coils; (c) subsequently to said step (b), without removing said sample from said chamber, applying an ac signal to said non-superconducting ac field coil under computer control and measuring ac magnetic characteristics of said sample in response to signals received from said sensing coils; and (d) subsequently to said step (c), removing said sample from said chamber.
Patent Citations (15)
| Patent | Date | Inventor | Cited By |
|---|---|---|---|
| US2975360 | 1961-03-01 | Bell | |
| US3454875 | 1969-07-01 | Bol et al. | |
| US3528001 | 1970-09-01 | Yntema | |
| US3665297 | 1972-05-01 | Yates | |
| US3863142 | 1975-01-01 | Werle | |
| US4005358 | 1977-01-01 | Foner | |
| US4037149 | 1977-07-01 | Foner | |
| US4134064 | 1979-01-01 | Jacobs et al. | |
| US4234939 | 1981-01-01 | Grossman et al. | |
| US4238734 | 1980-12-01 | Steingroever et al. | |
| US4277750 | 1981-07-01 | Bonnet et al. | |
| US4588947 | 1986-05-01 | Ketchen | |
| US4849695 | 1989-07-01 | Muller et al. | |
| US4861990 | 1989-08-01 | Coley | |
| US5008621 | 1991-04-01 | Jiles |
Non-Patent Literature (22)
- B. D. Cullity, "Introduction to Magnetic Materials," pp. 65-66 (Addison-Wesley Publishing Company), (No Month) 1972.
- Quantum Design Brochure, "MPMS2 AC Measurement and Field Cancelling Options," 2 pages, Aug. 1990.
- R. B. Goldfarb, et al., "Alternating-Field Susceptometry and Magentic Susceptibility of Superconductors," pp. 1-32 (Plemum Press), May 1991.
- J. Rebouillat, "High Resolution Automatic Magnetometer . . . Measurements," pp. 630-633 (IEEE Transactions on Magnetics), Sep. 1972.
- Reference Material for Quantum Design's Seminar (Nov. 5-6, 1990 Seoul) Woo Sin Enterprise, Inc.
- Rillo et al., "Multipurpose Cryostat for Low Temperature Magnetic and Electric Measurements of Solids," Magnetic Susceptibility of Superconductors and Other Spin Systems, Edited by R. A. Hein et al., Plenum Press, New York, pp. 1-24, May 1991.
- C. Rillo, et al., "Multipurpose a.c. and d.c. Equipment . . . Measurements of Solids", May 1991.
- H. Sasaki, "A Simple Precision Fluxmeter," pp. 100-102 (North-Holland Publishing Co.), Aug., 1969.
- C. Edwards, et al., "A Magnetometer for Surface Flux Density Measurement in MPI," pp. 304-306 (British Journal of NDT), Sep. 1987.
- P. Beckley, et al., "A Simplified electronic permeameter suitable for routine and standards use," pp. 379-384 (Measurement and Control), Oct., 1976.
- L. F. Marinaccio, "Op amp converts DVM to fluxmeter," pp. 112-113 (Electronics), May, 1975.
- E. G. DeMott, "Integrating fluxmeter with Digital Readout," pp. 269-271 (IEEE Transactions on Magnetics), Jun., 1970.
- E. Seely, "Dowty RFL Offers the Only Digital Fluxmeter With IEEE-488 Bus Operation," (Dowty RFL Industries), Mar., 1986.
- Lake Shore Cryotronics, Inc. Brochure, "AC Susceptibility Measurement--Its Purpose and Process," Nov. 1988.
- C. Rillo, et al., "On the Sensitivity of High-TC Superconducting . . . Sensors," pp. 775-780 (Sensors and Actuators A-Physical), (No Month) 1991.
- Bartolome, "Thermal, Magnetic, Magnetoelastic and Transport Characterization of Hard Magnetic Alloys," Supermagnets, Hard Magnetic Materials, Proceeedings of the NATO Advanced Study Institute on Supermagnets, Hard Magnetic Materials, II Ciocco, Italy Jun. 10-22, 1990, pp. 391-413, published by Kluwer Academic Publishers.
- Broersma, "Inductance Apparatus for Diamagnetic Measurements," The Review of Scientific Instruments, vol. 20, No. 9, Sep. 1949, pp. 660-663.
- Svendlindh, et al., "Static Scaling in an Amorphous Metallic Spin Glass," Europhysics Letters, 2 (10), pp. 805-812 (1986).
- Butera, et al., "Apparatus for the Determination of Magnetothermodynamic Properties to High Fields at Low Temperatures utilizing a Superconducting Magnet," The Review of Scientific Instruments, vol. 39, No. 8, Aug. 1968, pp. 1138-1144.
- Nave et al., "Micromagnetic susceptometer," The Review of Scientific Instruments, 51(5), pp. 591-596 (May 1980).
- Hein et al., "Differential Paramagnetic Effect in Superconductors," Physical Review, vol. 121, No. 2, pp. 407-415 (Jul. 15, 1961).
- Campbell et al., "Critical field measurements in sueprconductors using ac inductive techniques," Review ofScientific Instruments, 54 (9), pp. 1191-1198 (Sep. 1983).