Quantum Metrology Division

Technical Highlights

• Multilayer Reflective X-Ray Optics. The Division's thin film production and characterization facilities have produced several multilayer x-ray optics and characterized their reflectivities at design wavelength and at 0.154 nm. The individual thin films are truly thin (1 nm to 10 nm) and the largest multilayer constructions thus far have been approximately 250 nm thick. This is materials engineering at the atomic level and microstructural investigations are important to the controlled production, intelligent measurement, and application of these constructions.
  
  In order to improve the quality of these x-ray elements, we are systematically investigating and improving control of process parameters. We have in place projects to develop substrate preparation and characterization methods, new multilayer fabrication technologies, and advanced characterization methods. The reflectivity at design wavelength, surface roughness, and the (visible) optical figure are evaluated in addition to the above-mentioned high resolution low angle diffractometry. In furtherance of this effort, we have obtained contract support for development and delivery of several high quality focusing optics for telescope studies of the sun at several wavelengths in the EUV region. (J. Pedulla and A. Bartos)
• First Measurements at NIST EBIT. We have obtained first measurements at the newly commissioned NIST Electron Beam Ion Trap (EBIT). This facility, operated by the Atomic Physics Division received first users in October of 1993. The EBIT is designed to produce, trap, and excite highly-charged ions for atomic physics studies. Compared to other sources of highly charged ions, the ions in EBIT are cold and Doppler corrections to line positions are not significant. Neon-like and sodium-like barium (see Figure 1) and xenon have been trapped and observed at high resolution. Such ions as helium-like krypton and hydrogenic argon have also been observed. A variable radius focussing crystal spectrometer with unique capabilities for accurate measurement, originally developed for accelerator-based measurements was rapidly adapted to the EBIT. While not ideally suited to the EBIT geometry, several useful spectra have been acquired, particularly of neon-like barium (n=3→2 in Ba⁴⁶⁺). Barium has also been observed at other EBIT laboratories as a native contaminant from the electron gun cathode. Accordingly, we are in position to link the spectral lines of barium to the standards chain for use as markers in spectroscopic studies of trapped ions here and at other facilities around the world. Indeed, we can provide an absolute energy scale of unprecedented accuracy for spectra emitted by these highly-charged ionic systems. Collaborative developments with the Atomic Physics Division and the Naval Research Laboratory (NRL is using a NIST spectrometer developed earlier in this Division for similar purposes) enables measurements of polarization, energies, and production processes for spectral lines interesting in astrophysical and laboratory investigations. Spectral identification is aided by modern atomic-structure theory which is in turn itself put to new tests. In the coming year we plan a development program to produce crystals with well-controlled departures from ideality. This aims to significantly increase the throughput of the spectrometer, enhancing the statistics of measured spectra and decreasing accumulation times at the cost of an acceptable loss in resolution.

  

  Figure 1: High-resolution spectrum of sodium- and neon-like barium acquired at the newly commissioned NIST EBIT.

  In addition to the measurements with the EBIT, accurate results for Au and U have been obtained from recombination processes in the electron cooler at GSI and are in publication. Some progress has been made toward an improved angle-tuned Doppler spectrometer at the internal gas target of the GSI storage ring. The detector bodies have been built, the microstrip position-sensitive elements of the detectors have been manufactured, and a first prototype detector has been assembled and tested. The accelerator staff at GSI has succeeded in accelerating a uranium beam, stripping it to bare uranium ions, and decelerating again, as required in our proposal. We await the installation of a new target chamber and improvements in the gas target at GSI, planned for this coming year. In the mean time, our colleagues at GSI have developed and proposed intermediate measurements based on a beam-velocity-tuned Doppler spectrometer. (C. Chantler, J. Schweppe, and L. Hudson)
• X-Ray Resonant Inelastic Scattering. In a University of Washington (Y. Ma)/Argonne (P. Cowan) project at X-24A, X-ray resonant inelastic scattering (XRIS) was applied to crystalline Si samples. The resonant Kβ x-ray scattering line shape (but not the non-resonant line shape) was seen to depend on the Si crystal orientation [(100), (110) or (111)]. Data were analyzed on the basis of an x-ray inelastic scattering model by requiring conservation of crystal momentum of the electron states involved, as well as energy conservation. The dependence of resonant line shape on crystal orientation can be related to the Si valence band structure, and the XRIS technique offers the potential application to band structure determination in solids. In comparison with band studies involving electrons, XRIS has the advantage of larger penetration depths typical of photon-in/photon-out techniques. This application of XRIS to crystalline samples is an extension of previous studies (performed on X-24A) of resonant x-ray emission from gas-phase atoms and molecules and related theoretical developments. (S. Southworth)
• NEXAFS Studies of Passivation Breakdown. Near Edge X-ray Absorption Fine Structure (NEXAFS) studies have been carried out at the K-absorption edge of chlorine ions involved in the breakdown of passivating coatings on iron. Beamline X-24A is uniquely capable in this region and has been used for many studies involving x-ray absorption and emission from chlorine-containing samples. In this case, users from the NIST MSEL (G. Long) and the Johns Hopkins University (J. Kruger) attempted to distinguish between two models of the breakdown, one involving small holes in the coating allowing direct access of the chloride ions to the metal substrate and the case of chemical modification of the passivating film. Experimental techniques were developed and preliminary results obtained in these initial exploratory studies. A second set of measurements are scheduled on X-24A, and it appears that NEXAFS will resolve the breakdown mechanism with obvious implications concerning emphasis of future development work on passivating coatings. (S. Southworth)
• Spectrometry for Mammography. The original development of a Laue-case diffraction spectrometer for kV (and spectral) determination in the mammographic region has been granted patent protection as has its extension to the curved crystal case. Foreign filings are in progress and the first journal article has been published. The NIST decision to grant exclusive license for commercial development has been noted in the Federal Register; protests or additional bids should be resolved by the time of the Panel meeting. Interest has been widespread and enthusiastic leading to about ten talks and half as many requests for prototypes (even using present film registration). One important limitation of the initial prototype has been overcome. Specifically, with the flat transmission crystal used in the first designs, the spectrometric resolution obtained in the image plane is limited by the focal spot size. This limitation is not debilitating in the case of modern mammographic x-ray sources whose focal spots are near 150 µm. On the other hand, many units remaining in clinical use have spot sizes closer to 300 µm micrometers. To circumvent this problem, the third model of the spectrometer for kV measurement employs a carefully bent crystal to achieve symmetric Cauchois focussing. A prototype device has now been shown to give spectroscopic resolution independent of the focal spot size.
  
  A proposal for clinical and modelling studies with three cooperating medical facilities is pending with the Army Breast Cancer Research Fund. A second proposal was prepared and submitted to the NCI/NASA program on technology transfer from federal laboratories in the area of digital mammography; this proposal involves the Georgetown University Medical Center Imaging Physics Division and the company being licensed for commercialization of the NIST device. (R. Deslattes)
• X-ray Scattering Factors. Accurate values for atomic scattering factors are required for crystallographic structure determinations (conventionally in the energy range 8 keV to 30 keV, but also up to 60 keV) and for estimations of the performance of multilayer x-ray optics (from 50 eV to 15nbsp;keV). Analysis of the internal structure of multilayers by short wavelength x-rays in order to estimate their optical performance at much longer wavelengths is particularly sensitive to weakness in the x-ray optical constants database. Revised formulae lead to significant qualitative and quantitative improvement, particularly above 30-60 keV energies, (e.g., see Figure 2) near absorption edges and at 30 eV to 3000 eV energies. Selected predictions of a revised theoretical computation are in agreement with experiment, as opposed to results from earlier syntheses. The principal output of this research has been the energy-dependent but angle-independent contribution to Re(f) over the energy range from 30-100 eV to 200 keV, for all elements. A new printed tabulation and an electronic database are planned. Already one commercial vendor of optical design software packages has offered to enter a royalty agreement to include this database in the company's product. (C. Chantler)



Figure 2: Form factor f₁ for uranium near K- and L-edges indicating above-edge errors of standard derivation. Solid line indicates revised formulation.

• Lattice Transfer Measurements for 7 Silicon Crystals Completed. Samples for lattice parameter comparison of all of the silicon crystals available for gamma-ray diffraction at the ILL have been prepared and measured with respect to the lattice spacing of the NIST XROI silicon crystal. Typical profiles obtained in these measurements are shown in Figure 3. Seven samples from three different manufacturers were included in this set of comparisons. A guest researcher from KRISS (Cheon Eom) was a major contributor to these measurements and KRISS also supplied one of the samples. The uncertainty of the comparisons is a few parts in 10⁸. In addition, preparation has begun on Ge and Si samples which are needed to compare the lattice spacings of Ge crystals used for gamma-ray diffraction to that of the NIST XROI silicon crystal. A detailed manuscript describing the lattice comparison facility is scheduled for publication in the NIST Journal of Research.

  

  Figure 3: X-ray profiles obtained using the delta-d lattice comparator. The experimental profile (upper curve) is well described by dynamical diffraction theory (lower curve). The fine structure results from using two crystals of nearly equal thickness (in this case 0.455 mm) and is a significant aid in determining the profile center. These profiles are for the Si(440) reflection of Ag Kα radiation.

  The NIST XROI measurement of the lattice period of a highly perfect Si monocrystal which serves as the standard for the lattice transfer measurements continued to address the puzzling super-period residues (amplitudes greater than the 0.01 ppm performance expected of the apparatus) found in "good" results from previous years. The recent focus has been on the auxiliary Michelson interferometers which track changes in direction of the moving crystal as the system is scanned through its range of motion, typically 3 × 10⁵ to 4 × 10⁵ x-ray periods. A principal outcome of this year's efforts has been the slowly developed association of super-period errors with aperture diffraction in the auxiliary interferometers. The needed remedy, namely a new polarizing beamsplitter having higher optical quality (permitting larger beam diameters at the mirrors and hence smaller beam diameters in the collimating telescope) was designed and a contract let which is, unfortunately, now six months in arrears of a promised three month delivery; a second, reflective optics approach, is proceeding in parallel. Nonetheless, data obtained with a relatively low intensity effort show that by careful adjustment of the existing system, the superperiod errors have been reduced to an amplitude comparable with the statistical variation of x-ray phase measurements. Although the numerical results are quite reasonable and could be claimed to have higher accuracy than those previously available, the importance of this step is such that we contemplate no publication prior to installation of new optics and acquisition of a significant corpus of new data. (E. Kessler and R. Deslattes)