Conceptual Design Section from the Proposal for the GM/CA-CAT Sector

Release Date:


4.1  Sector Development Concept

The fully developed GM/CA-CAT sector is intended to include two insertion device (ID) beam lines, each one viewing a separate undulator source in the straight section, and a bending magnet (BM) beam line. Here we present conceptual plans for the beam lines, which details of our preferred full build-out may change as the project proceeds. Particularly regarding the two-undulator plan, technical and budgetary feasibility must be verified at the earliest possible stage of the project. The preferred undulator source for both ID lines is APS Undulator A (3.3 cm period), or perhaps an undulator of identical length but slightly different period ( e.g. 3.0 cm) to better line up the spectral ranges of the harmonics with the most useful photon energy ranges, or perhaps one of each design. Although initial budgets may permit build-out of just two of the three beam lines, the infrastructure for the support of all three must be in place from the start. This includes front ends capable of supporting all three lines, at minimum. The angular separation between the two ID beams is an important consideration for both the front end and the design of the two ID lines. GM/CA-CAT is well aware that the maximum potential angular separation of two ID beams radiated from the same accelerator straight section has not yet been determined and requires accelerator R&D, and is just as well aware that a front end design that can sustain, capably separate and deliver two hard x-ray ID beams to separate beam lines does not yet exist. GM/CA-CAT is eager to collaborate with APS and other CATs interested in the pursuit of these goals.

The optics designs for the three beam lines will be similar in concept, if not in detail. An important design goal is to support a MAD experimental capability on all three beam lines over as wide a range of photon energies as possible without compromise of beam attributes of the individual radiation sources. Therefore, the initial design for each beam line includes a rapidly tunable, vertically deflecting double-crystal monochromator followed by a vertically deflecting harmonic rejection mirror. Vertical deflection is especially preferred over horizontal deflection for the ID line monochromators, in order to attain the best energy resolution for a given monochromator crystal without sacrifice of intensity (an APS ID beam is 5-10 times less divergent in the vertical than in the horizontal) and in order to avoid polarization-induced intensity loss at high monochromator Bragg angles, i.e. low photon energies (for horizontal deflection, the loss is about one-fourth at 20�, one-half at 30�, and is total at 45�). The mirror for each beam line will include longitudinal coating strips to be selected for operation at particular energy ranges. Focussing in the vertical will be provided by the mirror, which will be configured as a bendable flat (bending in the meridional direction).

Focusing in the horizontal will be provided by a sagittally bent second monochromator crystal, assuming that silicon is employed. This is essentially the same optics layout presently used by SBC-CAT for both its BM and ID lines. Inclusion of an additional vertically collimating mirror upstream of the monochromator in the BM line is also being considered, in order to deliver an energy resolution limited by the source size. The vertical opening angle on the ID lines is so small that such a collimating mirror is not considered necessary. The protein crystallography BM line now under construction by IMM-CAT includes such a collimating mirror.

If we determine that steric constraints require a double-diamond-crystal monochromator on one of the ID lines, then horizontal focusing will not be provided by a sagittally bent second monochromator crystal. In this instance, the vertically-deflecting mirror could be toroidal, i.e. configured as a bendable cylinder, in order to focus in both the horizontal and vertical directions. This option would not permit the use of longitudinal coating strips on the mirror; a single coating would be employed, thereby constraining the mirror's energy range options. However, use of diamond crystals for such a beam line would already constrain the energy range capability. An alternative approach in this case, would be to use a Kirkpatrick-Baez (K-B) mirror combination, i.e. one bendable flat mirror (with meridional bending) to focus horizontally and a second one to focus vertically. In this situation, one or even both mirrors of the pair can contain coating strips. COM-CAT has just successfully implemented a double-diamond-crystal monochromator followed by a horizontal focusing mirror configured as a bendable flat on its ID line, and IMCA-CAT plans to implement a K-B mirror pair on its ID line.

At this juncture, it suffices to describe the individual beam lines together rather than separately, as they are similar in concept and final beam line designs and a construction contractor for the sector have not yet been selected. For the most part, designs to be employed will be identical to those in use by other APS CATs. The main exception will be the design of the monochromator for the secondary ID line, which must deliver a monochromatic beam offset that is relatively large, in order to provide sufficient separation of the two monochromatic ID beams in the upstream experimental station. The monochromator tanks and crystal support assemblies for each of the ID lines must be designed so as to permit unfettered propagation of the appropriate white or monochromatic beam for the adjacent ID line, as must be the mirror tanks and bending assemblies for each of these lines. Careful ray tracing will be required to determine the synchrotron radiation and bremsstrahlung shielding for the two proximal ID lines. We plan to locate all of the optics for the two ID lines upstream of both experimental stations for these lines, which should greatly simplify the shielding challenge. In so doing, common control and support areas can be arranged in the downstream end of the sector to service the two contiguous experimental stations, and common radiation enclosures, electronics and cooling services can be arranged upstream to service the optics for the two ID lines. A similar theme governs the layout of the BM line section of the sector, which is far simpler to design and implement as only one BM line is proposed.

4.2  Elements of the Design

Proceeding from upstream to downstream and beginning with the first components beyond the main accelerator shield wall and beryllium window(s), each of the beam lines will include the following major components (the potential BM line collimating mirror is described last, although it appears far upstream in the BM line):

4.2.1 A white beam primary aperture to be positioned ahead of the monochromator.

Standard APS designs for the BM and ID beam lines will be employed, but the aperture for the secondary ID line will require special attention so as not to interfere with the adjacent primary ID white beam. In fact, due to space constraints (the two ID white beams are likely to be separated by just a few cm at the location of the aperture for the secondary ID line) it may not be feasible for the primary aperture of the secondary ID line to have a full complement of translational degrees of freedom, in which case fixed shield settings may be required.

4.2.2 A vertically deflecting, double-crystal monochromator, preferably using silicon crystals with cryogenic cooling of the first crystal for each of the two ID lines and water cooling of the first crystal for the BM line.

Provision for sagittal bending of the second monochromator crystal is to be included (assuming that silicon is used), to provide horizontal beam focusing or collimation. The monochromators should provide standard beam offsets presently adopted on APS beam lines (3 to 5 cm) in order to be compatible with standard APS designs for downstream photon shutters and bremsstrahlung shields, as well as designs of monochromator systems commonly used on APS beam lines. The vertical offsets of the double-crystal monochromators of the two ID lines will be oppositely directed to increase beam separation. The monochromator for the secondary ID line will require a rather large beam offset of perhaps 15 cm or greater, for acceptable separation of the two ID monochromatic beams at the location of the upstream experimental station.

If APS accelerator studies determine that a horizontal 1-mrad angular deviation of the two ID beams cannot be achieved, then there may not be sufficient clearance inside the upstream monochromator tank (of the secondary ID line) for a cryogenically cooled first crystal and its support that avoid interference with the adjacent primary ID white beam. This would force a very compact first crystal and its cooling support for the secondary ID line, which should be feasible if water-cooled diamond is employed, as is presently done in the monochromators for the ID lines of SRI-CAT Sector 3 and COM-CAT. Use of diamond as the crystal material would preclude the capability of sagittal bending of the second crystal, unless the second crystal is chosen to be silicon or germanium instead ( i.e. Si(220) or Ge(220) to "match" approximately the d spacing of diamond(111). However, mismatched crystal pairs are not an option as they would result in variable vertical direction for the monochromatic beam as the photon energy is changed, sacrificing the goal of rapid tunability. It is intended that, for all beam lines, the beam is to be fixed in position at the experiment location, regardless of chosen photon energy, except in certain unusual instances which we will discuss below.

Otherwise, interchangeable Si(111) and Si(220) crystal pairs are planned for all monochromators in order to span, as much as possible, the desired photon energy range of about 3.5 to 35 keV. Because convenient and rapid exchange of cryogenically cooled first crystals for the ID line monochromators may not be feasible, one of the ID line monochromators may permanently employ a Si(111) pair to cover from 3.5 keV to ~20 keV, and the other a Si(220) pair to cover from ~6 keV to 35 keV. If this turns out to be the case, then the lower energy Si(111) pair should be used to simplify the design of the non-standard monochromator that will provide the large beam offset, because the crystals do not need as large a separation between them at lower energies as they would at higher energies. Monochromator tanks of both ID lines will be designed to allow for propagation of both ID line beams within them, i.e. the secondary ID line monochromator tank will allow for passage of the primary ID white beam, and the downstream primary ID line monochromator tank will allow for passage of the secondary ID monochromatic beam. Naturally, if cryogenic cooling of the first crystal can be accommodated for both ID line monochromators, then it does not matter which of the monochromators is upstream.

4.2.3 A vertically deflecting, bendable flat harmonic-rejection mirror with longitudinal coating strips, to provide vertical beam focusing or collimation.

The design of this mirror would be similar for all three beam lines. For each of the ID lines, the mirror would deflect the beam in the same direction as deflected by the appropriate double-crystal monochromator, in order to increase the separation between the two ID monochromatic beams. This means that the mirror for one ID line will reflect the beam upward and the mirror for the other ID line will reflect the beam downward. Just as for the monochromator tanks and crystal supports, both of the mirror tanks and mirror benders must be designed to allow one ID monochromatic beam to propagate through undisturbed even when the mirror is aligned to reflect the other ID monochromatic beam. Achieving this goal may require that the design of each mirror support cradle must have the mirror's reflecting surface face toward the cradle instead of away from it as conventionally pursued, in order to avoid interference of the cradle structure with the other ID monochromatic beam which passes by overhead or underneath. Coating strips which are envisioned are platinum (for operation at high energies), palladium or rhodium (for operation at mid-range energies between about 7 and 20 keV where most experimental measurements are expected to take place), and no coating at all (for operation at low energies) in which case the substrate itself serves as the reflecting surface. The substrate is envisioned to be silicon of length roughly 1 meter. The glancing angle of incidence upon the mirrors is expected to be about 3 mrad, but can be adjusted as needed if circumstances warrant ( e.g. to 2.4 mrad if use of the platinum coating at 35 keV is desired; this angle adjustment would result in a vertical displacement of the beam in the experimental station). The mirror benders will have sufficient range and strength to bend the mirrors to the required meridional radii to focus the beam at hand. Appropriate degrees of freedom will be included in the mirror mounts or tanks, including a horizontal translation to center the different coating strips in the beam path and a vertical translation to remove the mirror from the beam path entirely if its use is uncalled for ( e.g. to diagnose the beam emanating from the monochromator).

The mirror systems for the ID lines will both be placed upstream of the two ID line experimental stations. Careful positioning of the mirrors along the ID lines to provide appropriate beam deflections at the position of the upstream experimental station, combined with the beam deflections imparted by the two ID line monochromators, should allow to attain a separation between the two ID monochromatic beams of at least 30 cm at the location of the experimental apparatus in the upstream station. This can be accomplished as follows. For the primary ID line, the monochromator can deflect the beam down by 5 cm and the mirror can shift the beam direction downward by 6 mrad which will add 3.6 cm to the overall deflection at a distance of 6 m downstream of the mirror (where the secondary ID line experimental apparatus might be located), resulting in a total shift downward by 8.6 cm from the white beam plane.

For the secondary ID line, the monochromator can deflect the beam up by 15 cm and the mirror can shift the beam direction upward by 6 mrad which will add 6.0 cm to the overall deflection at a distance of 10 m downstream of the mirror (where the secondary ID line experimental apparatus might be located), resulting in a total shift upward by 21.0 cm from the white beam plane. The combined deflections of the two beams at this location add to about 30 cm in this case. It should be feasible to accomplish even larger beam separations. Regardless, a 30 cm separation should be sufficient to guarantee safe passage of the primary ID monochromatic beam, enclosed within a long, skinny shielded beam tube, below the sample and detector axes and stay-clear zones of the diffractometer system in the upstream, i.e. secondary ID line, experimental station. The diffractometer circles would sit above the beam tube, their support (on the table) would sit behind or straddle the beam tube, and the optical table top would sit below the beam tube, all separated enough from the tube to prevent collision with it throughout the full ranges of motion of the table and diffractometer. Shifting the two-theta setting away from zero in this case must necessarily tilt the two-theta arm upward, to prevent collision of the detector with the tube. A simple shelf with sufficient clearance can be built off the tabletop to surround the beam tube, to protect it from equipment, tools, and users.

4.2.4 A photon shutter to be placed downstream of the mirror tank for each of the three beam lines, of appropriate standard APS designs.

These can be operated independently for the two ID lines, and would both be placed upstream of the two ID line experimental stations.

4.2.5 Miscellaneous items.

Appropriate monochromatic beam slits, beryllium windows, filter and photon energy calibration foil assemblies, and beam intensity and position monitors and diagnostic fluorescent screens are to be included in each of the three beam lines. All beam lines will include appropriate vacuum hardware (pumps, beam tubes and transition bellows, crosses, gate valves, gauges, etc.) as needed.

4.2.6 A primary collimating mirror may be included in the BM line, to be placed in the white beam before the monochromator.

This would be necessary if it is determined that the energy resolution requirement for the BM line is the same as for the ID lines (~10 -4) . The ID lines can provide this energy resolution without use of a collimating mirror because the vertical divergence of each ID beam is quite small (<10 �rad), whereas the vertical divergence of the BM beam is closer to 0.1 mrad which, for a Si(111) monochromator operating at 12 keV, would give rise to an energy resolution of ~6x10 -4. To attain a better resolution using the same monochromator crystals without reducing intensity with narrow vertical slits requires use of a vertical-collimating mirror. This mirror would be of similar design to the downstream vertical focusing mirrors but, being placed in the BM white beam, would require water cooling and would need to be protected on its upstream end by a cooled safety mask. Since this mirror would serve solely as a vertical beam collimator, i.e. to focus the beam at infinity image distance, it may be acceptable to pre-manufacture a finite and fixed meridional radius, thereby foregoing use of a bender mechanism to achieve the desired radius. However, a bender mechanism might still be desirable since accurate pre-manufacture of the exact several-km required radius may be difficult to ensure, and radial compensation in response to slight adjustment of the mirror's glancing angle of incidence may be needed.

Figures showing a conceptual sector layout and a proposed deflection scheme for the ID beams are attached as Appendix B. Detailed locating of the components, especially the focusing optics (monochromators and mirrors) will require ray tracing and engineering design study to determine. For the ID lines, demagnification factors are expected to be about 3:1 for the horizontal (based on placement of the sagittal focusing monochromators) and 5:1 to 6:1 for the vertical (based on placement of the meridional focusing mirrors). For the BM line, these factors are likely to be smaller (about 2:1 for the horizontal and 3:1 for the vertical), due to the distance scales involved and the desire to avoid excessive horizontal beam convergence at the sample position (which should be maintained within 2 mrad).