Compound refractive lenses (CRLs) are arrays of concave lenslets used to focus X-rays. For a given incident X-ray
beam energy, the focal length of a CRL depends on the material and shape of the individual lenslets, and in particular is
inversely related to the number of lenslets in the array. The throughput of a lens array is heavily affected by absorption
of the X-rays in the lens. For this reason, it is necessary to employ low-atomic-number materials and fabricate the
lenses as thin as possible, especially for low to moderate X-ray energy range (~ 5 - 20 keV) photons.
Lithium and beryllium are two of the best candidate materials for X-ray lenses due to their relatively high (real
decrement) index of refraction and low X-ray absorption. Lithium is very malleable, however, and reacts strongly with
moisture in the air, requiring a special fabrication environment and housing. Beryllium, on the other hand, is a solid
metal and is easy to machine and handle.
This paper summarizes the recent work at the Advanced Photon Source (APS) on parabolic lithium and cylindrical
beryllium lenses. These lenses have been tested on APS X-ray beamlines. Their performance in terms of the focal size
and gain is described and further improvements including tighter manufacturing tolerances and thinner lens walls are
Lithium promises to give refractive x-ray optics the highest possible
transmission, aperture and intensity gain. Room-temperature embossing of lithium with parabolic dies from polypropylene produces lenses that focus well but are not yet good enough for imaging. X-ray measurements suggest two causes of problems, one of which one can be solved easily.
To develop narrow-bandpass multilayer monochromators, we have studied small d-spacing WSi<sub>2</sub>/Si multilayers. We found that WSi<sub>2</sub>/Si is an excellent multilayer system for achieving both the desired spectral resolution and peak reflectivity. Compared to other traditional multilayer systems such as W/Si, WSi<sub>2</sub>/Si not only has a lower density and lower absorption, but also is a chemically more stable system, since WSi<sub>2</sub> is already a silicide. One thus expects better thermal stability and sharper interfaces for WSi<sub>2</sub>/Si multilayers. There are two approaches to achieve high-resolution multilayers: either decrease the d spacing or use low absorption materials. By using WSi<sub>2</sub>/Si, we can utilize both approaches in the same system to achieve good energy resolution and peak reflectivity. Another advantage of this system is that the sputtering rate for Si is much higher than other traditional low-Z materials. Several WSi<sub>2</sub>/Si multilayers have been fabricated at the Advanced Photon Source (APS) deposition lab using dc magnetron sputtering with constant currents of 0.5 A in Ar at a pressure of 2.3 mTorr. A test sample of [9.65Å-WSi<sub>2</sub>/10.05Å-Si] × 300 was studied at four institutions: using laboratory x-ray diffractometers with Cu Kα (8.048 keV) wavelength at the APS x-ray lab and at European Synchrotron Radiation Facility (ESRF), and using synchrotron undulator x-rays at 10 keV at MHATT-CAT and at 25 keV at ChemMatCARS-CAT of the APS. The measured first-order reflectivity was 54% with a bandpass of 0.46% at 10 keV and 66% reflectivity with a bandpass of 0.67% at 25 keV of undulator x-rays. Similar results were obtained from Cu Kα x-rays. This result is very attractive for the design of a multilayer monochromator for the ChemMatCARS-CAT to be used in the 20 to 25 keV range. Other small d-spacing multilayers are being studied. Comparison between WSi<sub>2</sub>/Si and W/Si multilayers will be discussed.
We have designed and fabricated a new type of focusing multilayer mirror, capable of reflecting a divergent beam over a large energy band around 9 keV. A flat energy response about 10% wide and providing a reflectivity of 50% has been achieved by a non-periodic bi-layer sequence while a lateral thickness gradient follows the varying Bragg condition over the whole mirror length. The focusing setup is based on a simple one-point bender and a pre-shaped substrate. A focal spot size of about 8 micrometers has been obtained at a distance of 285 mm from the center of the mirror using synchrotron radiation from an undulator source. Energy dependent scans have shown that this device enables focusing experiments with fixed geometry at variable energies.
Lithium is the best material for refractive x-ray lenses, with peak performance around 8 keV. To date we have built a prototype of Cederstrom's so-called alligator lens, and have tested the lens with beamline 7ID's 10 keV x-rays on the Advanced Photon Source at Argonne National Laboratories. To date we have attained only a threefold gain, most likely limited by surface roughness that is avoidable with more careful manufacturing techniques.
In our recent x-ray photon correlation spectroscopy (speckle) experiments at NSLS, one of the challenges is to increase the coherent photon flux through a pinhole, whose size is chosen to match the beam's horizontal transverse coherence length l<SUB>h</SUB>. We adopted an approach to vertically focus the x-ray beam so as to match its vertical transverse coherence length l<SUB>v</SUB> (at NSLS X13, l<SUB>v</SUB> approximately 50 l<SUB>h</SUB>, l<SUB>h</SUB> approximately 12 micrometers at 3 KeV) with l<SUB>h</SUB>. By demagnifying the vertical size by a factor of l<SUB>v</SUB>/l<SUB>h</SUB>, we expect to increase the intensity of the x-rays through the pin hole by the same factor while keeping the beam coherent. A piece of commercial 3/8' thick float glass, by virtue of its low surface roughness (approximately 3 angstroms rms), good reflectivity in the low photon energy range of interest and low cost, was chosen as the mirror material. A computer controlled motorized bender with a four point bending mechanism was designed and built to bend the float glass to a continuously variable radius of curvature from -700 m (intrinsic curvature of the glass surface) to < 300 m, measured with the Long Trace Profiler at the BNL Metrology Lab. This mirror bender assembly allows us to continuously change the focal length of the x-ray mirror down to 0.5 m under our experimental conditions. At the NSLS X13 Prototype Small Gap Undulator beamline, we were able to focus the x-ray beam from a vertical size of 0.5 mm to approximately 25 micrometers at the focal point 54 cm from the mirror center, thus increasing the photon flux density by a factor of 20. Results also show that, as expected, at an incident angle of 9 mrad, the mirror cuts off the harmonics of the undulator spectrum, leaving a clean 3 KeV fundamental for our experiments.