Comments for the Roadmap Scenarios
Scenario A or D sounds more reasonable choices without big stepback to the microdiffraction community. D is the best choice for new XIS sector for the whole surface/interface X-ray community. It is flexiblly extendable for incorporating a couple of Argonne initiatives. But unfortunately, there is only one option of sector 25 as future home of XIS sector. We may think of adding more options for that choice.
The Liquid-Surface Scattering science community at the APS is interested in road map options that allow us to take advantage of new technical and scientific developments, as well as address the beamtime needs of a growing community. Although the technical aspects of Road Map options A, B, D-I are unspecified at this time, we would support these options with the following priorities.
- We place highest priority on securing access to 100% of the beamtime on an ID straight-through line that would have full energy tunability. Such full tunability may not be possible with a side-kick canted undulator source, which displaces the beamline laterally on the experimental floor.
- Our second highest priority would be securing access to 100% of the beamtime on a side-kick canted ID line with only partial energy tunability. Our understanding is that this can be achieved at either 9-ID or 15-ID, though the extent of the tunability may differ at these two sectors. This issue plus other advantages and disadvantages of locating a side-kick canted undulator line at one or the other of these two sectors needs to be explored further.
- We place lowest priority on maintaining the status quo at 50% of the beamtime on the 9-ID beamline because a) it cannot meet the present beamtime needs of our community and b) upgrades for enhanced future capabilities may be compromised by sharing with another instrument.
Irrespective of the chosen upgrade option, we hope that liquid surface scattering at 9-ID remains operational until funding for a new line is secured and the plans for the construction are well underway.
We look forward to working with the APS on the development of new capabilities in liquid surface scattering.
From the point of view of the magnetic spectroscopy (MS) beamlines, there are four different considerations.
The first is that there are currently accelerator physics questions as to whether the two MS beamlines can reside in the same sector without compromising both scientific programs, as both will have helical undulators for polarization switching in APS-U. Further clarification on this is needed from the accelerator physics side. However, if it is the case that two helical undulators cannot reside in the same straight section, scenarios A-E cannot be accomplished.
The second practical consideration is that there are space restrictions that limit the viability of canted sectors where a BM has been built out. While 4-ID-C once resided in sector 2, since it was moved to sector 4 over 10 years ago, it has experienced a substantial expansion in scope and in instrumentation. Four new instruments have been added, including a high-field magnet chamber, PEEM, XPS, and an octopole magnet chamber that requires large power supplies that occupy an ~5x2m area. These supplies need to be located close to the octopole chamber. Therefore, the ability to fit the beamline and its scientific program into sector 2 (or any sector with a built-out BM) now is doubtful, without a dramatic reduction in the scope outlined in CD-1. This suggests that if the MS-S beamline is moved, the green field option (scenario G) may be the best.
Thirdly, it’s worth noting that there is a benefit to the MS-H beamline (4-ID-D) from moving the MS-S beamline that is not reflected in the Roadmap Scenarios grid. With 4-ID-D in its own sector, it may use a dual undulator enabling faster polarization switching. Since none of the scenarios involve moving this beamline, this benefit is not made clear.
Finally, if a green field option is chosen, it may make more sense to locate MS-S in sector 28, rather than 25, due to synergies with 29-ID.
I feel that the “scope reductions” being considered in scenarios B and C for micro- and nano-diffraction (S3DD) represent a large step backward rather than forward, and should not be implemented. My reasons are: 1. The proposed facility was rated in the highest “very strongly recommended” category in the Scientific Prioritization for the APS Upgrade. The scientific case is broad and compelling, and the facility should not go backward in terms of available beamtime. 2. Sector 34-ID-E already has one of the highest demands at the APS. You currently need a score of 1.4 to get beamtime and some excellent proposals never receive any beamtime. 3. Other high-energy synchrotrons worldwide are following the APS lead and either have built or are designing similar facilities. The APS can keep the worldwide lead, particularly if developments with nano-focusing optics are continued. To me, these reasons make it clear that a “scope reduction” for this facility does not make sense.
- Joining XSD mdiff and S3DD nano makes very much sense, and should be done (as in scenarios A and D-J).
- Scenarios B and C would be very detrimental to microscopy and micro/nanodiffraction programs by curtailing existing and SAC approved new programs; they should be avoided.
- Placing the existing mFluor program on a cant is possible (scenarios A-F, H-J), but to minimize the loss of independence and corresponding capabilities, the source should consist of two short collinear IDs to feed the two independent endstations to either maximize coherent flux or flexibility depending on experiment requirements.
- mFluor has complex FOE optics, and needs to be located far away from the source, limiting the possible partners. MS-S is probably a good partner, and partnering at 2-ID is probably very cost effective.
- The in situ nanoprobe requires complex FOE optics, and ideally should be joined with a program (in a canted sector) that has less space requirements in the FOE; BCDI may be an excellent choice.
- Paring the in situ nanoprobe with the magnetic diffraction (MD) program (scenario F) seems not feasible: both programs require relatively large lateral space, need to be far away from the source, and therefore are not very compatible.
- Instead of moveing BCDI to 32-ID, and rebuilding 34-ID-1 to accommodate S3DD nano and XSD mDiff (scenarios A, D-J), it may be very cost effective to directly locate S3DD nano and XSD mDiff at a newly canted 32-ID
- We propose scenario K and derivatives, (see excel spreadsheet) to address several of the above issues.
General comments on the existing scenarios:
- The plan to move and upgrade XSD mdiff together with S3DD nano into a new endstation makes perfect sense, we strongly support this (a suggested name for the new, joined program is ‘nanodiffraction’).
- This could be done in a new endstation 34-ID-F, as implied in scenarios A, D-J, and the new scenario K. These scenarios also imply a move of BCDI to 32-ID-1.
- A different option could be to build XSD mdiff and S3DD nano immediately at 32-ID-1 (last station), saving a move, saving rebuilding a station (32-ID), and saving building a new station (34-ID-F), although this would forgo advantages for BCDI being located at the very end of 32-ID (good match with from a BL layout perspective TXM, as well as scientific commonalities (nano-morphology)).
- Scenario B would relocate to XDF mdiff to join a cant with BCDI on 34-ID. This would still require a new hutch (34-ID-F), since mdiff needs full access to reciprocal space, and therefore must be the last station, making no room for S3DD nano. It is not possible to locate XSD mdiff to 34-ID-C while maintaining its current capabilities because of the close proximity of the canting beam pipe. It would also reduce the BCDI program by 50% and not realize the gains promised by joining forces between XSD mdiff and S3DD nano.
- Canting 2-ID (or relocating the existing XSD mfluor program to a new cant) is possible but would limit the current ability to run the hard x-ray programs of 2-ID-D and –E quasi independently
- Much of this independence would be restored by using 2 short (1.2m), collinear IDs with 2.7 and 3.0 cm periodicity as sources for the D and E endstation programs, with the ability to run at the same energy when opportune for maximum flux, and at different energies when maximum flexibility is required.
- Scenario C would not only cant 2-ID, but keep the XSD diff at 2-ID. This would be very detrimental to the rather successful XSD mfluor program, since it would cut about 50% of beamtime, severely affecting the user base (many of which have come to rely on the mfluor program for their research (NIH, DOE, NSF funded)). It would also not realize the expected gains of joining the XSD mdiff program and the S3DD nano.
- The in situ nanoprobe is probably best paired with a beamline that has fairly simple FOE requirements, and does not require to be far away from the source
We propose a new option (see also scenario K and its derivatives):
- There are advantages to placing the in situ nanoprobe (ISN / INP) together with Bragg CDI (BCDI) to the two cants on sector 9.
- ISN front end optics are fairly complex, BCDI optics very straightforward. This is a much easier (and cheaper) match than partnering with the 2-ID program, which would require an very complex first hutch.
- High hutches and larger source to sample distance at sector 9 are ideal for BCDI.
- Proximity of ISN and BCDI provides opportunities for scientific synergy between the programs (biofuels, batteries, solar cells, etc).
- Proximity is highly beneficial for local scientific support (eg, one person can much easier take care of technical support for both programs).
- Vibration issues at 9-ID would need to be addressed (probably straightforward, pumps on mezzanine are strong suspects).
- This will free up 32-ID-C, a fairly large hutch, for other programs such as MD.
- XSD’s microdiffraction program (XSD mdiff; Zhonghou Cai), would move together with nano-S3DD into the newly built 34-ID-F (downstream of 34-ID-C). This would be one instrument doing both type of experiments, making excellent use of synergy between science and methods of both programs. The close proximity to the micro-S3DD would be very beneficial for both programs.
- XSD mfluor (the existing hard xray fluorescence program at 2-ID) would stay at 2-ID.
- The ideal uncanted scenario (K and K’), avoids downtime for the existing user program, permits early experiments towards in situ capability development, and retains opportunities for future upgrades.
- Alternatively (K”), XSD mFluor could be paired (canted) with MS-S, and two short (1.2m), collinear IDs (2.7 and 3.0 cm) for mFLuor would continue to make it possible to run two independend experiments at each station, with the ability top choose between maximized flux and m,aximised flexibility depending on which experiments are to be run.
- In case the 9-ID option above was ruled out, one could consider moving the XSD udiff together with nanoS3DD to 32-ID instead of 34-ID, to save costs of rebuilding endstations and the effort of moveing. The downside would be that the BCDI program would remain ‘un-optimized’ in its current location (34-ID-C), and that the synergy of running both S3DD nano and micro programs on the same (canted) beamline is missed. Similarily, the synergy of running BCDI and in situ nanoprobe on the same (canted) beamline would not be realized.
My understanding is that 3-D x-ray facilities were among experiment stations given the highest priority for upgrades. As a user of facilities 34-ID-E, (as well as 1-ID, and 6-ID-D), I have gained much important data that has had considerable impact in the research communities in which I participate. My research is advancing understanding of heterogeneous deformation and damage nucleation at and near grain boundaries (using titanium as a model material), lead-free solder joints, and high purity Nb for particle accelerators. The research of my group has considerable impact, as my papers are well cited (my ISI h-index is 30). I was at the TMS conference last week, and a prominent researcher in lead-free solder from Taiwan commented to me that he needed to find something to do that could make a similar impact, and he is using synchrotron measurements now. In addition we recently received a prestigious second generation materials world network grant from NSF to collaborate with German colleagues in our investigation of heterogeneous deformation in commercial purity Ti, and the 3-D x-ray work was and is important for this work. We also have a grant from DOE/BES to investigate these issues in Ti alloys. We have also used beamline 34-ID-E to characterize deformation system activity in pure Nb, a BCC metal that is used for particle accelerators. BCC metals are notorious for the difficulty to make sense of dislocation activity, and the data we have obtained have and will be used to clarify these issues.
Regarding work on 34-ID-E, I am under the impression that scenarios B and C will disrupt the operation and reduce user access time for this beamline for some period of time. This would hamper our progress considerably, as we have been gathering information about underlying grains in regions where very interesting damage nucleation is occurring, and this underlying grain information is crucial for us to eliminate speculation about what is occurring where we can’t see in any other way. A disruption in our access to beam time will hamper the progress in our research and negatively affect the Ph.D. dissertation of two Ph.D. students working on titanium. The student working on solder found a very exciting result from beam time last fall that was presented last week, too, and this opens up new opportunity to assess mechanisms of continuous dynamic recrystallization that precede damage nucleation in solder joints (this research topic underlies our entire modern infrastructure, as half of electronic system failures are solder joint failures). We have considerable momentum in discovery of mechanisms responsible for damage nucleation that depend on the data and analysis of information obtained on 34-ID-E, and I hope this will not be slowed. There are other users who have focused on obtaining precise information about defect content that require essentially a week of beam time for a particular patch using monochromatic frequency sweeps to obtain the needed information. As much as we would like to do similar work, we have not had the opportunity to do this, as we are seeking out information over a larger area/volume, and the lack of beam time has limited what we can accomplish and discover.
I have listed every publication we have made that has been supported by our beam time on 34-ID-E (I have not included publications from work done on other beamlines). Many more presentations(including 6 invited) have been made that have publicized our discoveries, including 3 last week (two with titanium, one with solder – data/analysis were added after abstract was submitted). This series of papers continues, as indicated by papers submitted. I sincerely hope that you will not choose paths that will compromise the performance and user access to 3-D x-ray diffraction.
I am aware that, I am getting myself in trouble here but, honestly how many MD (magnetic diffraction) stations do we need? There are already MD activities in sector 4 and Sector 6 with (medium and high energy stations); why do we need more? I am saying this because I see in the road map that 5 scenarios of the MD involve partial takeover of 9-ID. What is the uniqueness of such an instrument? All magnetic scattering that do not require a cryo-magnet can be performed on practically any of the abundant X-circle spectrometers at the APS. I am not sure what exactly the Bio is about, but if it is based on SAXS/WAXS versatile instrument, it seems more natural to combine them with the LSS in Sector 9. Strategically, it is better to combine SAXS-bio and LSS for the simple reason that the preparation lab needs are similar for both. Scientifically too, it makes more sense to put together closely related complementing activities.
Scenario B is not going to work! Because of the canting beam pipe, it is physically impossible to locate XSD mDiff to ID-34-C and maintain its current capabilities which are supported by a means of full access to reciprocal space.
The APS Upgrade scenarios presented are well thought through and reflect the significant effort by the APS-U team at communication with the community. Many scenarios appear to provide excellent opportunities to maintain APS’ excellence by preserving the capabilities of the existing programs, while at the same time adding powerful and visionary new flagship beamlines that will keep the APS at the scientific forefront for many years to come.
From the perspective of the In-Situ Nanoprobe beamline, a number of scenarios as presented are in principle feasible, one additional synergy could be exploited, and one scenario is not feasible.
- Scenario “F” would curtail the performance of the ISN and is therefore not feasible. “F” calls for pairing ISN and MD. These programs are incompatible, as both instruments require significant lateral space, and need to be positioned far away from the source. The ISN instrument requires large working distance to accommodate in situ environments, while, at the same time, providing a spatial resolution between 20 nm and 50 nm. This necessitates positioning the instrument as far from the source as possible, and requires partnering with a program that can be positioned closer to the source.
- One potential partnering opportunity that addresses the constraints listed above is not reflected in the existing roadmap, and is suggested as “Scenario K” in a different roadmap comment (MIC). It would combine the ISN program with the BCDI program at a canted sector. The two programs are compatible in that the BCDI FOE optics could more easily be combined with the more complex ISN FOE optics than other scenarios would allow, and in that the BCDI instrument does not have to be positioned as far away from the source as ISN instrument and can tolerate a canted beam passing by within < 50 cm laterally. There are additional synergies between the two programs that are spelled out in the complementary roadmap comment.
- Siting the ISN beamline at 25-ID (Scenario H) will eliminate the possibility of extending the beamline (and thus increasing the working distance) in the future, and is therefore less preferred.
Thanks for developing scenarios for the APS-U roadmap for future beamlines at the APS.
As for the S3DD upgrade related scenarios, I vote for scenario A, D, E, F, G, H, I, or J, but not B and C.
In these ten scenarios, option B and C would significantly affect our S3DD user program. If the S3DD upgrade results in a compromised way, both current microdiffraction platform and future nanodiffraction platform will have to share one undulator source instead of what we originally proposed to have two independent canted sources. These two scenarios would have consequence of substantial reduction of the available beam time to our microdiffraction platform users. In the last year, we canted 34-ID beamline and increased microdiffraction user operation time from 50% to 100%. If B or C scenario happens, then we have to cut microdiffraction time back to 50%. Although once the nanodiffraction platform is built, some of the microdiffraction users will naturally choose nanodiffraction, but not all user experiments could be performed on nano platform because of the limitation of working distance, beam angular divergence, and unnecessity of nanobeam in some cases. In addition, the construction and commissioning of the nanodiffraction platform would take time (maybe 2 years), which will reduce the total available beam time to our users during that period if B or C scenario happens. I just come back from the TMS conference this week, there are big user demands on our 3D diffraction microprobe facility for materials science research and will be more when the technique is more advertised.
I fully understand that the scenarios B and C are mostly because of that the available insertion device ports at APS now are limited and siting is a difficult task, and also some budgetary reason. In the APS-U programs, the SAC committee suggested the S3DD as a "Very Strongly Recommended". So if there is any possibility, I will not suggest that the S3DD upgrades in a compromised way.
Thank you very much for your consideration.
I am still puzzled by the fact that LSS science is considered a second-order perturbation in any road map including this one. If one of the goals of the upgrade is to come up with unique instruments then, I would argue that having a single liquid surface spectrometer on the APS floor is by definition unique (the LSS in Sector 15 is only partially available to the community). It will be a real loss if LSS science is not promoted and advanced at the APS. This is not a push-button apparatus and good experiments are in general very hard to accomplish. Therefore the recognition of its importance has been slow over the years. However, there is a growing recognition by bio-chemists, bio-physicists, materials scientists in the techniques for diverse applications, in separation chemistry (nuclear waste), bio-mineralization (nano crystals), protein/polymer interactions (medical applications, implants healing), control and growth of crystal polymorphism (pharmaceuticals) just to name a few of the topics that are currently pursued on the LSS.
After dismantling the LSS operations in Sector 6, I strongly hope that at least the current LSS capabilities at APS will not be reduced. If this road map is final, which I still hope it is not, then I would like to see option C realized – namely, maintaining the current LSS status in Sector 9. It is the option that does not require any investment anyway, and users through the PUP and GUP processes will do the actual incremental upgrades. More important than any upgrade at this time is the reliability of the apparatus. Options A or B on a canted beamline may also work but we do not know the range of tunable energy. Tunable energy is crucial for the future success of the LSS science as more and more spectroscopy techniques have now been developed and tested with LSSs for various applications.
From the viewpoint of microdiffraction and nanodiffraction science, I would recommend schemes A/D/E/F/G/H/I/J over B and C. The S3DD programs are doing very much the same kind of science as XSD mDiff: both scan samples to measure how diffraction changes with position, generally as a function of microstructure. I expect substantial synergy by combining the two programs in the same sector
I think scenario D is the best. I am concerned about moving ASD-D to Sector 30 in the other scenarios. IXS experiments at Sector 30 have to deal with very low signal strengths and generally require a lot of beam time. Moving ASD-D over could substantially reduce the available beam time to IXS users. I would further argue that Sector 30 could use an upgrade to improve the photon beam intensity to a level similar to what is available at ESRF.
Scenario D is best for the XIS project (in my opinion), as it leaves open the possibility of developing substantial lab space and offices outside of the traditional LOM structure (i.e., outside of 400 next to the CNM) with the open space near sector 25. This keeps open possibilities for coordinated development with other ANL priorities. It also allows for expansion of the beamlines into the LOM, which is important for making all of the planned beamlines and capabilities to fit into the sector footprint.
The other scenario (at sector 28) may be constraining with regards to fitting in the planned components of the sector. This option might require modifying the LOM to add a second level to accommodate all of the hutch/lab/office space that is anticipated for the sector without having to change the building footprint.
Both options (sector 25 and 28) would benefit from being near the CNM.
I suggest that the APS roadmap should consider this point, and give a plan for the locations of such type of devices.
Due to the demands for improved performance and efficiency in applications of the structural materials, more and more materials need to be tested at extreme enviroument such as high temperature (more than 1000 C) and/or high pressure (more than 100 GPa). Although it will increase the experimental cost, I think it represents the future development of the experimental devices.
I suggest that the APS roadmap should consider this point, and give a plan for the locations of such type of devices.
The various road map scenarios presented by the APS upgrade team make sure that all of APS's existing programs have a home, and that disruptions to operations are minimal, guided by a "do not harm" principle. Since real estate in the accelerator and on the floor is very limited, and since the number of programs/new facilities exceeds the ideal amount of real estate needed, such a "do not harm" approach is likely to result in compromises preventing optimization of APS programs, including upgrade programs.
APS has a unique opportunity that presents itself only every decade or two. We can choose to heavily invest in certain areas where APS can lead the world, or compromise by not pursuing ideal conditions for these programs to be world leaders. The SPX project is one example where the goal is to optimize all aspects of the program to position it to lead the world (real estate in both accelerator and experimental floor, insertion devices, etc). This is expensive. However, a decision has been made that this is a critical program for APS/ANL and the correct path is that of not compromising the scientific program at the expense of "do not harm". While "do not harm" should be used whenever possible and feasible, priority should be given to position APS to lead in certain areas of synchrotron science in the years to come. In order to open real estate this may involve merging existing programs and resources into focused areas of growth. We should aim to build the best facilities and instruments, even if this causes some level of disruption, or even if it requires merging of existing programs to open real estate.
Placing beamlines at canted sectors is certainly suitable in many cases, but significantly compromises programs in other cases. As with the SPX program, we need to ensure that the key programs for APS science are upgraded under OPTIMAL conditions of real estate in both the accelerator and experimental floor as well as insertion devices and beamline instruments. It seems that the various scenarios may have placed more emphasis into preserving all programs and not disrupting operations than ensuring OPTIMAL conditions for key programs where APS could lead the world in synchrotron science into the next decade .
I vote for scenario C.
I think scenario D is the best suited for the overall "big picture" plan for the following reasons:
- Least amount of location unchanged. More people staying put more feasible to meet milestones and less unnecessary movement.
- No scope reduction in this scenarios.
- In this scenario the ASD-D does not relocate. I think it will be difficult to quickly upgrade the diagnostic section with very little time for planning.
Thanks for yesterday's presentation, which was lucid and informative. Looking from the point of view of Sector 3, is all I can do at this point.
I am very concerned that our Sector 3 will be forgotten in this upgrade period. We have some exciting upgrade ideas of our own, but we are delegated to the third group, hence, no resources are currently being allocated. During the proposal phase, we were asked to have someone else make the case for us. I feel I was the most qualified person to make the case for Nuclear Resonance Scattering, having been in the field for over 30 years, and recognized for my work around the world. However, that chance was taken away from me by the rules set at that time. My attempts since then to get back on the map is encountered with a polite smile. Yet, we do have some very interesting ideas of our own:
- Improving the energy resolution and stability to 0.4 meV level, with improved spectral density. This will be accomplished by combination of new undulators and cryogenically cooled monochromator Tom Toellner has developed.
- Reducing the focal spot size to 1-5 micrometer range to access above 2 Mbar pressure for geophysics work, by employing new high quality mirrors.
- Implementing a fast shutter to increase the efficiency of Mössbauer measurements by a factor of 100. This is something we already have demonsrated. We just need money to implement it at our beamline.
- Creating new space for dedicated catalysis chamber, and ultra high vacuum chamber for thin film magnetism and device applications. We have been active in nanocatalyst work, but we are limited because of lack of equipment.
- Extending the range of momentum transfer and adding new sample capabilities for Sector 3 HERIX instrument for single crystal work. This will improve productivity quite a bit, becuase phonon cross-sections have square dependence on momentum transfer.
- A small scale detector development program for array APD and stacked APD's. These detectors exist in Japan, but they won't sell to us.
I just returned from a international workshop in Germany dedicated to Nuclear Resonant Scattering. We presented 11 talks out of 30 total, on work done at Sector 3-ID. We still have leadership role. Users who come to us from ESRF and SPring-8 point out the excellence at Sector 3. Yet, recently PETRA-III has moved ahead of with 10 m long undulator, and they will get even better when they implement the remaining 15 m. Similarly, Spring-8 has finished the new IXS beamline, with 25 m long undulator by the end of next year.
I feel that the earth is moving under me, and yet, never in my life I found myself so helpless. It seems, there is no recourse, and I just wanted to let you know how I feel.
Best of luck with the upgrade project.