Charles Agosta, Janet Musfeldt, Meigan Aronson, Gil Clark, Larry Kevan, Martin Maley.
Future Users Committee meetings.
It is our view that the format of the Users Committee meeting should change. Too much time is spent during the presentations relating statistics and data better sent ahead in paper format. Our request is that each group in the future pick an experiment or event to highlight, and present mostly that topic. Other issues and priorities can be discussed as needed during a question and answer time. The High B/T facility report from October was a good example of an efficient report. Any statistical information on users or magnet time might be summed up on a slide but should be distributed on paper (or electronically) a week before the meeting. This new format should streamline the meeting and allow us to spend more time discussing and resolving issues while we have users and magnet lab people at the same location.
Los Alamos Issues.
We look forward to installation of the deep magnetic field modulation for the 60 T long pulsed magnet. This addition will provide a modulation field for experimental techniques that increase the signal to noise for some types of measurements and will compensate for the ripple problem during a controlled pulse. This upgrade is strongly supported by the User Committee.
We agree that an October shutdown at LANL is a good idea to allow repair, test, and development of various equipment. With the exception of this coming year when the pulsed field equipment moves to a new building, we urge the LANL group not to shut down for longer than one month per year. Extensive shut-down time will seriously impact user access to the systems.
The users committee shares the concern over the delays in the development of the high resolution 900 MHz NMR system. However we feel that the NHMFL is proceeding toward the completion of this important deliverable and should now also concern itself with the development of a user community and with the necessary user support infrastructure required to bring this unique facility on-line. We are also concerned about the absence of financial support for the user support system for the 900 MHz facility and urge the NSF to explore solutions to this problem with other federal agencies. A problem of more immediate concern is the need for a control console to complete this system and to put it into commission. We urge the NSF to provide supplementary funding to purchase this essential element in time for the projected completion of the 900 MHz spectrometer in Dec. 99, so that there is no further delay to the commencement of the scientific program.
High B/T Facility
The user committee appreciates the clear and interesting presentation and is impressed by the high quality of work including external collaborators. We think that the added operational flexibility afforded by both the top loader and second intermediate field experimental stage would be exceptionally good value and should be pursued.
The resistive magnets continue to receive the majority of the use at the Lab. The Cell 9 upgrade (30 T to 33 T) and the possibility for ``multiple power supply use" are therefore very important. The transformer limitation, which is on-line to be fixed (see below), is also a critical development. We continue to urge upgrades as needed.
The User Committee notes that Power Supply D was lost from June-Sept. 98, and that a great deal of magnet time was lost. However, most experiments were accommodated due to a reorganized schedule on the resistive magnets. The magnet operators did a great job during this time, and we thank this highly dedicated team. Since the opening of the lab, there have been 4 transformer failures which have had an important user impact. Although there have been design changes to the transformers, we suggest funds be provided so that more than one replacement can be on-site.
There have been reports that strong 60 cycle noise pickup in the far-infrared experiments is affecting User activities in field-dependent spectroscopy. Current theories point to problems with the IRLabs bolometer detectors themselves or with the grounding of the system. If there is not already a solution, we urge the Tallahassee facility to track down this problem quickly.
The User Committee recognizes the need to focus on the Renewal Proposal. We must articulate a vision for the future of high-field research which involves both science as well as facilities development. In particular, we need to identify the most exciting scientific challenges ahead as well as highlight the science that has been accomplished so far. Most important, we need to intimately connect any new high-field technology and infrastructure with underlying scientific motivation.
The following is a letter sent out by Chuck Agosta in September 98 and the responses he received concerning the development of a transverse access magnet at the NHMFL.
I am sending this email to everyone I know who might be interested in using a transverse magnet at the NHMFL.
The NHMFL will soon write a renewal proposal to the NSF, and a few projects for new facilities will be included in the proposal. There are many possible new apparatus that the magnet design group could build including:
I have done some very unscientific polling and have found a number of researchers who would like to see a transverse magnet installed at the NHMFL. If you have an experiment that would need or benefit from a transverse field magnet, would you please send me an explanation of the experiment and why a transverse magnet would be needed. If the experiment is a two axis rotation experiment, you must justify the need for a transverse magnet considering that a two axis rotator exists for the 20 T wide bore magnet. Comment on what your priorities are for new apparatus at the lab.
Of course, please feel free to place some other development project ahead of the transverse magnet or argue why a transverse magnet is not worth the effort. And also feel free to forward this message to anyone in the community who you believe may want to contribute to this discussion.
I need to compile these responses by the end of September, so please send them to me as quickly as possible.
From: Jim Brooks <brooks@Magnet.FSU.EDU>
Chuck, I am all for #1!!!!!!!!!!1
From Azhang Ardavan (firstname.lastname@example.org)
To the NHMFL proposal committee,
I am writing to urge the committee to consider the case for a 20 Tesla transverse-field magnet at NHMFL. Such a facility would be extremely useful and would help to maintain NHMFL's position as the world's leading provider of high magnetic field facilities.
We are currently performing millimeter-wave magneto-optical measurements on organic molecular metals in order to study a new kind of magnetic resonance phenomenon, Fermi-surface traversal resonance (FTR) , arising from the traversal of carriers across open orbits on warped sections of Fermi surface. Owing to the warping of the Fermi surface, the real space velocities of the carriers oscillate as they cross the Fermi surface in the presence of an external magnetic field. This real space oscillation generates resonances in the high frequency conductivity of the material. FTR, which constitutes a generalization of cyclotron resonance in metals, is a rather important effect; it provides a new method of studying open carrier orbits in metals. A study of the angle-dependence of FTRs gives information about the direction, amplitude and periodicity of the warping components in a Fermi sheet; it can be interpreted as a direct measurement of the Fourier transform of the Fermi sheet. Open Fermi surface orbits occur frequently, even in everyday materials such as copper, but until recently the techniques for studying Fermi surface topologies, cyclotron resonance and the de Haas--van Alphen and Shubnikov--de Haas effects, have given information only about closed Fermi surface sections.
The ``natural'' way to perform FTR measurements is to fix the magnetic field magnitude and rotate its direction (see ). This would only be possible using a transverse magnetic field. It is not possible to implement the traditional rotation techniques, like those used in d.c. transport measurements, because the experiment employs waveguides to couple millimeter-wave radiation into and out of the cryostat.
So far, FTR has only been measured by sweeping the magnetic field for fixed sample angles. Since measurements at many angles are required, the sample must be repeatedly removed from the cryostat. This procedure is very time-consuming, wasteful of cryogens, and often results in the destruction of valuable samples through thermal cycling problems. The availability of a transverse-field 20 T magnet would solve these problems and make FTR a much more viable technique for Fermi surface studies.
For these reasons our group, already a regular user of NHMFL facilities, is very interested in the possibility of a 20 T transverse field facility, and we would vigorously support a proposal for its development.
 A. Ardavan et al., Phys. Rev. Lett. 81, 713 (1998).  S.J. Blundell et al., Phys. Rev. B 55, R6129 (1997).
From: "Janice L. Musfeldt" <email@example.com>
My priorities would be the following:
1 - The mini-hybrid. Accomodating more users should be a priority.
2 - The transverse magnet. Some doped spin-Peierls materials have an anisotropic phase diagram. To "move into" certain phases, one needs to control the direction of the field. Zn-doped GeCuO3 is an example.
3 - Other.
Let me know what I can do to help with the renewal process.
I'm glad you like the paper. I am rather proud of it because the work was done by a really exceptional undergraduate student. All of my undergrads are good, but this one was really hard working and independent.
From: Bruce Brandt <brandt@Magnet.FSU.EDU>
Thanks for doing this polling and search for scientific justification. No. 2 is supposedly already in the works for production and installation as soon as the 45 T hybrid insert is finished (sometime in 1999).
No. 4 might be a proposed hybrid for high field NMR and other experiments that require sitting at high fields (30 T or so in a bore at least 52 mm diameter with a 20 MW or less insert) for long times.
One person who may be interested in the transverse field, vertical access magnet is Jack Ekin at NIST Boulder. He used to do some strain measurements on superconductors. His experiments required that the field be perpendicular to the strain and that the strain by applied by a very strong probe.
Another project of interest is a 50 or 52 mm insert for the "existing" hybrid. I suppose it would have a field in excess of 35 T.
From: Mike Naughton <firstname.lastname@example.org>
In response to your solicitation for comment on the utility of a transverse resistive magnet at NHMFL:
If the sole charge of the NHMFL is to make bigger and stronger magnets, then there's no room for a split coil. If, on the other hand, its purpose includes enhancing scientific opportunities in strong magnetic fields, even if they don't set world records, then a transverse magnetic should be a very high priority. Regardless of what may be thought and said by any number of people who either have not or do not use or need transverse fields, there is much science that can be done in a transverse field can not be done in a vertical or simple solenoid. A very large percentage of scientifically and technologically relevant and important materials these days are anisotropic, even highly anisotropic. The electric, magnetic, optical, indeed all, properties of these materials reflect this anisotropy. Many new phenomena have been discovered in rotation experiments in anisotropic conductors in the last decade. I consider these discoveries to be the tip of the iceberg. In order to get under the water and open our eyes to what lies below, we need apparati appropriate to the task. I'd compare gross rotations in a solenoid configuration to a big ship blasting into the iceberg to see what it's made of, while fine rotations are akin to Ballard in his nuclear sub carefully looking below the surface. The differences in approaches and results they yield are nothing short of titanic.
thank you for your consideration.
still a Bills fan,
Michael J. Naughton
Department of Physics
Chestnut Hill, Mass. 02167
From: Stephen Hill <email@example.com>
I think optics, FIR and mm-wave people have some of the strongest cases for a transverse field magnet. Strangely enough, none of the FIR scientists at NHMFL are aware of the short comings of an axial magnet.
For example, if you wish to excite currents normal to the layers in high-Tc, whilst subjecting the sample to a field normal to the planes. Then< the only way to do it in a small bore (i.e. in a high field magnet) is in a split coil configuration. You simply do straight up and down reflection (polarization perpendicular to magnet axis) from the edge of the sample. It really isn't realistic to try to put together intricate systems which turn the light 90 degrees at the field center. Not controllably anyway.
This is just an example for a fixed field orientation. If you want to< rotate the field, then you really need a horizontal field. I'm talking about doing the experiment scientifically and properly. Sure, people will tell you that you can design some clever widget to rotate the sample. However, you will end up with data which you have no idea about what it means.
I've not even discussed my stuff yet. I'm attaching excerpts from a proposal to NSF to buy a split coild superconducting magnet.
PS - re-reading my proposal, I realize that you can add NMR to the list of people who should care.
An obvious experimental requirement for investigations of low-dimensional systems is the ability to thoroughly explore any anisotropy in tensorial quantities such as conductivity or susceptibility. Application of a strong magnetic field complicates this procedure considerably. To begin with, the magnetic field reduces the symmetry, and increases the consequent complexity, of the tensor to be measured. However, this is precisely the reason for applying a field since, by varying its orientation relative to the material under investigation, one can directly probe any anisotropy. Indeed, quite spectacular angular effects are observed through measurements of the normal state conductivity tensor of low-dimensional organic superconductors in strong magnetic fields . These effects have been attributed to a switching off and on of phase coherence in the least conducting directions as the magnetic field points along certain magic angles . Measurements like these have proven central to our understanding of inter-layer transport in low-dimensional conductors and superconductors.
Another problem associated with high magnetic field measurements is the geometrical and space constraints imposed by a typical magnet. In high field magnets, a homogeneous field is only achieved over a small volume (< 1 cm3), and access to this volume places considerable constraints on the design and ultimate performance of the magnet. The highest magnetic fields are achieved in electro-magnets where the field is produced inside a single solenoid [,]. The only access to the field in a solenoid is along its axis. Consequently, any experimental probe inserted into such a magnet must be co-axial with the solenoid and, therefore, co-axial with the magnetic field produced within the solenoid. As a result, the magnetic field cannot be rotated with respect to the experimental probe, which is an important requirement for measurements on low-dimensional systems, as will be outlined in this proposal.
A straightforward solution to the problem of rotating the field orientation is to split the solenoid into two parts so that the experimental probe may be inserted radially. However, this compromises the performance of the magnet, since the split obviously occurs where the magnetic field would have been strongest in the magnet, i.e. the split allows flux to escape from the bore of the solenoid, leading to a reduction in the field strength at its center. Consequently, in order to maximize the attainable field, most researchers operate axial magnets and develop intricate experimental probes which allow the sample to be rotated in the magnetic field. In this proposal, we will describe a variety of experiments necessitating rotatable high magnetic fields, where the latter solution just is not possible. Hence our requirement for a high-field split-coil magnet.
The problems associated with measurements across this entire frequency range are essentially the same, and involve the coupling between the r.f. or microwave fields and the sample under investigation. In the case of microwave measurements, the sample sits in a cavity which is coupled to a network analyzer via rigid (low impedance) waveguides. All of the components in the resulting microwave circuit must be carefully impedance matched in order to optimize the coupling between the cavity (and therefore the sample) and the analyzer. The situation for high-field (i.e. high-frequency) NMR is essentially the same, except that the r.f. signal is propagated in rigid coaxial cables, and the sample is coupled to the AC fields via the coil in an LC-circuit . It cannot be over-emphasized that the coupling between the analyzer and the sample under investigation is of paramount importance to these techniques, and is a potential source of serious errors in any subsequent analysis of the data. Since the microwave and r.f. techniques are fairly similar, and since the subsequent data analysis for magnetic and conducting materials is very similar , the following discussion will be limited to problems associated with microwave conductivity measurements.
The use of resonant cavities offers many advantages in the millimeter-wave spectral range, particularly in the case of small metallic single crystals, where the radiation wavelength is comparable to the sample dimensions, thus rendering conventional optical techniques useless, e.g. simple reflectivity measurements. Under these conditions, the sample perturbs the electromagnetic field distribution within the cavity and, due to the resonant nature of the problem, the resonance condition is extremely sensitive to small changes in the sample conductivity σ(ω). Provided that the sample produces only a small perturbation of the electromagnetic field distribution within the cavity, it is relatively straightforward to relate changes in the quality (Q) factor of the resonance, and the resonance frequency (fo), to changes in the complex electrodynamic response of the sample. The specific details of the technique are highly dependent on the type of sample (i.e. whether metallic, magnetic, insulating, etc.), as well as the geometries of the sample and the cavity; an excellent account of these techniques may be found in ref. . For good conductors, changes in Q and fo are directly related to changes in the complex surface impedance, ZS = RS + iXS, of the sample which, in turn, is related to the conductivity through the expression ZS(ω) = [iμω/σ(ω)]_. In the case of a poor conductor, dissipation is related to the magnitude of the imaginary component of the dielectric function ε2 (= σ1/ωε). Dissipation generally causes changes in Q, while dispersion results in changes in fo.
Serious problems arise if re-orientation of either the sample, the cavity or both, is required. In order to carry out the research described in this proposal, it will generally only be necessary to measure a few components of either the conductivity or susceptibility tensor for a given material. This is best achieved by re-mounting the sample each time. The reason for this has to do with the way in which the sample distorts the microwave fields in the cavity, and with the subsequent data analysis. Exact methods for measuring σ(ω) are only possible (reliable) if the geometry of the sample shares some symmetry with the field distribution in the cavity, e.g. one can easily evaluate σ(ω) for a needle-like sample situated with its axis coincident with the axis of a cylindrical cavity excited in a TE 011 mode . Because of this constraint, it is sometimes necessary to use different geometry cavities to measure different components of the conductivity tensor for a sample which has a strange shape. For these reasons, the sample cannot be rotated with respect to the cavity, i.e. we do not propose any measurements where the sample orientation will be altered with respect to the resonator.
While the sample and resonator are to remain rigidly connected, it will be necessary to vary the orientation of the applied DC magnetic field relative to the sample. The orientation of this field relative to the resonator has no influence on the data analysis - only on the physical properties of the sample within the resonator. Consequently, it will not be necessary to re-mount the sample for each field orientation, provided that the field itself can be rotated relative to the resonator. This possibility will greatly simplify the research described in this proposal and will open up many new experimental possibilities.
It has been established above that the sample cannot be rotated in the cavity during a measurement. However, can the cavity and sample be rotated together with respect to the DC magnetic field? The short answer to this is "no", especially in the case of high magnetic fields, where space constraints are severe. To rotate the cavity would risk altering the coupling between the cavity and the rigid waveguides coming from the analyzer. This coupling must reproduce faithfully from one measurement to the next, e.g. as the field orientation is altered. Consequently, the only sensible solution is to keep the microwave part of the experiment absolutely rigid, and then rotate the magnetic field with respect to it. As discussed in the introduction, this is only possible if the magnet is of the split-coil variety.
From: "W. Gilbert Clark" <firstname.lastname@example.org>
Here are my responses to your message.
At 09:37 PM 9/3/98 -0400, you wrote:
> The NHMFL will soon write a renewal proposal to the NSF, and a few
>projects for new facilities will be included in the proposal. There are
>many possible new apparatus that the magnet design group could build
>1. A transverse magnet (field perpendicular to the bore) that could reach
There are many condensed matter NMR experiments that need a transverse
and probably can not be done with the two-axis rotator in the current wide
bore 20 T magnet. They all involve some degree of rotation.
A. Two-axis rotation -- There are lots of NMR experiments to be done a
field that need two-axis rotation in a cryogenic environment. Although
single-axis rotation can be carried out with the probe in a standard magnet
(my group has already done it), I think the complexity of the electrical
circuitry of a high field NMR probe makes it very unlikely that one can use
the two-axis rotator of the wide bore magnet for this application. The
natural way to do it is to combine our (or someone else's) horizontal axis
rotation in the probe with a second rotation of the experiment about a
vertical axis in the split coil magnet.
B. Some NMR experiments need rotation and a dilution refrigerator. Although we MIGHT be able to combine single axis rotation with the portable fridge, I think it would be nearly impossible to do two axis rotations for the reasons given in A above. If rotation in the dilution refrigerator is not practical, one could think of having at least single-axis rotation using the fridge in the split coil magnet.
Regarding this latter point, for NMR it wouldn't surprise me if it would be cost effective to go with a small fridge without top loading. I had some recent experience on this in Grenoble that could be discussed. If the warmup/cooldown time is short enough, the corresponding simplicity and lowered capital cost might make such a fridge more desirable for some measurements than the top-loading system.
>2. A special purpose magnet that could accept a high homogeneity
>set of gradient coils or other custom field profiles.
Such a special purpose magnet should be quite useful for a variety of experiments. If the Keck magnet lives up to its projected specs, that might dilute the importance of a high homogeneity insert, except for power and perhaps other considerations. On the other hand, there are lots of reasons for having a large gradient available (that I will not list here for lack of time).
>3. A mini-hybrid magnet that could reach 30 T using only 10 MW,
>more users to run simultaneously.
I think the mini-hybrid is a very important item and could become one of the real workhorses of the lab for the future. But my recollection is that it should go into the 35-37 T range, depending on the homogeneity.
> I have done some very unscientific polling and have found a number of
>researchers who would like to see a transverse magnet installed at the
>NHMFL. If you have an experiment that would need or benefit from a
>transverse field magnet, would you please send me an explanation of the
>experiment and why a transverse magnet would be needed. If the experiment
>is a two axis rotation experiment, you must justify the need for a
>transverse magnet considering that a two axis rotator exists for the 20 T
>wide bore magnet. Comment on what your priorities are for new apparatus at
Of the three items above, I would place the mini-hybrid first, the split coil second, and the specialty magnet third.
That's all for now. See you in October.
Professor W. Gilbert Clark
Department of Physics and Astronomy,
University of California at Los Angeles
Box 951547, Los Angeles, CA 90095-1547
UCLA Tel:(310) 825-4079 (ofc), (310) 825-1641 (lab)
UCLA Fax:(310) 825-5734; Email: email@example.com
Return to NHMFL Home Page
Return to NHMFL Operations Home Page