Parabolic refractive x-ray lenses are novel optical components for the hard x-ray range from about 5 keV to about 120 keV. They focus in both directions. They are compact, robust, and easy to align and to operate. They can be used like glass lenses are used for visible light, the main difference being that the numerical aperture N A is much smaller than 1 (of order 10^-4 to 10^-3). Their main applications are in micro- and nanofocusing, in imaging by absorption and phase contrast. In combination with tomography they allow for 3-dimensional imaging of opaque media with sub-micrometer resolution. Finally, they can be used in speckle spectroscopy by means of coherent x-ray scattering. Beryllium as lens material strongly enhances the transmission and the field of view as compared to aluminium. With increased N A the lateral resolution is also considerably improved with Be lenses. References to a number of applications are given.

Parabolic refractive x-ray lenses with short focal distance can generate intensive hard x-ray microbeams with lateral extensions in the 100nm range even at short distance from a synchrotron radiation source. We have fabricated planar parabolic lenses made of silicon that have a focal distance in the range of a few millimeters at hard x-ray energies. In a crossed geometry, two lenses were used to generate a microbeam with a lateral size of 160nm by 115nm at 15.2keV at a distance of 47m from the synchrotron radiation source. First microdiffraction and fluorescence microtomography experiments were carried out with these lenses. Using diamond and boron as lens material, microbeams with lateral size down to 20nm and below are conceivable in the energy range from 10 to 100keV.

We present results on comprehensive studies of high resolution SU-8 planar refractive lenses. Lens optical properties were investigated using coherent high energy X-ray radiation. Resolution of about 270 nm was measured for the lens consisting of 31 individual lenses at energy 14 keV. Coherent properties of the set-up permit to resolve near-focus fine structure, which is determined by tiny aberrations caused by lens imperfections close to the parabola apex. This study allows understanding as far SR deep lithography as possible can maintaine to close tolerances for lens parameters. Two-dimensional focusing crossed lenses were tested and imaging experiments in projection and imaging mode were conducted. Radiation stability test was performed and conclusions on the applicability of SU-8 lenses were done.

Compound refractive lenses (CRLs) are arrays of lenses designed to focus x-rays. The advantage of a focused x-ray beam is improvement in imaging resolution for applications such as microscopy and tomography. CRLs are desirable due to their simple designs and ease in implementation and alignment. One method of fabricating CRLs is extrusion. Extrusion can be employed to produce, for example, aluminum CRLs for high-energy applications because many aluminum products are produced in this manner. Multiple lenses can be extruded in an array in a single run. This method is relatively cost effective compared to others methods of fabricating CRLs. Two generations of extruded aluminum CRLs have been manufactured to date with lens wall thicknesses of 200 and 100 μm, respectively. The first-generation CRL yielded focusing and established the potential to produce high gain if reduced wall thicknesses could be achieved. Testing of the second generation is reported here.

We have fabricated planar parabolic lenses made of silicon and boron which have a focal distance in the range of a few millimetres at hard x-ray energies. Two silicon lenses were used in a crossed geometry to generate a microbeam with a lateral size of 160x107 nm² at 15 keV. The focus size of 180 nm at 22 keV and at a distance from the synchrotron radiation source of 42 m was obtained with boron lens. The performance of the silicon lenses was improved by optimising the e-beam lithography and the etching parameters. In addition, we report on the microfabrication process of diamond as well as pyrolytic graphite lenses.

This report describes the optimization, first experimental data and the performance outlook for a special refractive lens for the focusing of x-rays. The lens was obtained by applying the strategy of Fresnel to lighten a lens by removing efficiently optically passive material. This strategy was applied to objects, which were produced by deep x-ray lithography and which can thus be shaped in only one dimension. Consequently this class of lenses can focus in only one dimension. While in the normal Fresnel lenses as much material as possible is removed, in the present lens the remaining segments were kept as large as possible, in order to finally obtain a rigid and deep structure. The resultant structure is composed of small prisms of almost identical shape, which are combined such that the final lens looks like an hour glass, i.e. two large prisms touch each other at one of their tips. Such a lens has better transmission than a normal refractive lens of equal focal length. And even though it has worse transmission than the normal Fresnel lenses, it could be produced with an unprecedented geometrical aperture of 2.6 mm for 8 keV photon energy. The uniformly etched structure depth was found to be at least 0.4 mm, over which the lens can focus free of aberrations. A lens with focal length 2.183 m was found to provide after very rapid alignment a focus size of 2.8 μm, while slightly better 1.73 μm were ideally expected.

One application of Kinoform Fresnel Lenses is to generate small focal spots of hard X-ray photons with high gain for micro-diffraction experiments. A Kinoform lens can be obtained from a refractive lens by deleting material such that at the design photon energy, the deleted regions correspond to with modulo 2π phase-shifts in the phase front. At photon energies different from the design photon energy, the phase jumps are no longer 2π, and the diffractive properties of the kinoform become more significant. We present measurements and calculations of spot size versus photon energy.

Kinoform lenses avoid the absorption losses from a comparable refractive lens by removing all material causing redundant 2π phase shifts. Such optics allow high resolution imaging with a theoretical 94% focusing efficiency. While fabrication of kinoform lenses for two-dimensional focusing is difficult, standard lithographic processes can be utilized to fabricate optics in silicon which produce a line focus. By putting two single-dimension kinoform lenses in a crossed-pair arrangement, a two-dimensional spot is achieved. First attempts at imaging with a crossed pair of kinoform lenses are presented.

To image weakly absorbing materials (e.g. biological specimens, thin films, etc.) with hard x-ray photons, phase-contrast methods have to be applied to enhance the image contrast. Micro-fabricated Fresnel prisms in silicon have been manufactured to enable wavefront division of the incoming x-ray beam for phase-contrast applications. To maximize the efficiency and aperture of these optics, multiples of 2π phase-shifting regions in a conventional prism structure have been deleted, leading to structures that are arrays of micro-prisms. We show preliminary results of x-ray beam deflection using a variety of micro-prism arrays at the NSLS X13B undulator beamline at 11.3 keV.

The Xeus mission is designed to explore the X-ray emission from objects in the Universe at high redshifts, and the success of the mission depends critically on the deployment of a 10 square metre class telescope system in a suitable orbit for science observations. The minimisation of the telescope mass and volume becomes of critical importance for such a large facility. We describe developments of novel light weight optics that enable a reduction in mass per unit area of more than an order of magnitude, compared with traditional replication optics technology. With such a large collection area, image confusion limits become a scientific driver as well, demanding arcsecond class resolution. We describe measurements that demonstrate the improvement in resolution that gives very high confidence that these requirements can be met. Some implementation details of the mission are briefly mentioned.

Producing the next generation of X-ray optics, both for large astrophysics missions and smaller missions such as planetary exploration, requires much lower mass and therefore much thinner mirrors. The use of pore structures allows very thin mirrors in a stiff structure. Over the last few years we have been developing ultra-low mass pore optics based on microchannel plate technology in glass, resulting in square, open-core glass fibres in a concentric geometry. The surface roughness inside the pores can be as low as 0.5 nm due to the extreme stretching of the surface during production. We show how improvements in the production process have led to an improved quality of the fibers and the quality of stacking the fibers in the required geometry. To achieve een higher imaging quality as required for XEUS we have developed in parallel a novel pore optics technology based on silicon wafers. The production process of silicon wafers is extremely optimised by the semiconductor industry, leading to optical qualities that are sufficient for high-resolution X-ray focussing. We have developed the technology to stack these wafers into accurate X-ray optics, set up automated assembly facilities for the production of these stacks and present very promising X-ray test results of 5.3 arcsec HEW from single reflection off such a stack, showing the great potential of this technology for XEUS and other high-resolution low mass X-ray optics.

We propose an X-ray all-sky monitor sensitive in soft X-ray band with the potential to work up to ~ 10 keV. The limiting flux of the proposed system is ~ 10^-12 erg/s/cm² and the angular resolution ~ 4 arc min. The system consists of a number of wide-field modules based on the Multi-Foil Optics.

Collimators capable of higher resolution and optimized for greater sensitivity can significantly improve the imaging quality of gamma-cameras for single-photon-emission computed tomography of small animals. We are applying deep x-ray lithography and gold electroforming techniques to fabricate high-resolution collimators with continuous, smooth, and thin septa. Negative SU-8 photoresist was used for mold fabrication. To be efficient, collimators for gamma-cameras designed to image 140 keV gamma-rays should be over 1.5 cm tall. The height of the collimator can be achieved by stacking the appropriate number of layers.

Systematic experimental investigations of Bragg-Fresnel gratings are discussed. Gold and nickel masks with periods of 0.4 μm n 5 μm were evaporated on the surfaces of Si [111] symmetric and asymmetric crystals. These have been used to obtain x-ray diffraction in the energy range of 8000 eV n 8500 eV. A theoretically calculated maximum of diffraction efficiency of the order of 30 % was measured experimentally for gratings with grooves parallel to the beam direction (sagittal gratings). Diffraction effects in the crystalline substrate for the gratings with grooves perpendicular to the beam direction (meridional gratings) limit the diffraction efficiency on the order of a few percent. Experimental data are compared with the theoretical calculations dispersion and efficiency of Bragg-Fresnel gratings.

The concept of the design and fabrication of X-ray diffraction focusing elements is discussed. This concept includes both reflection and transmission types of optics as well as equipment necessary for their fabrication.

We introduce a new design of tilted linear zone plates, which are named tapered tilted linear (TTL) zone plates. The purpose of the design is to increase efficiency while at the same time keeping the focal plane perpendicular to the optical path. In order to accomplish this, the zone radius and number of zones must become a function of position along the structure. Simulation work described in this paper shows improved optical performance over regular tilted linear zone plates.

Using Fresnel zone plates, a spatial resolution between 20 nm for soft x-rays and 70 nm for hard x-rays has been achieved. Improvement of the spatial resolution without loss of efficiency is difficult and incremental due to the fabrication challenges posed by the combination of small outermost zone width and high aspect ratios. We describe a novel approach for high-resolution x-ray focusing, a multilayer Laue lens (MLL). The MLL concept is a system of two crossed linear zone plates, manufactured by deposition techniques. The approach involves deposition of a multilayer with a graded period, sectioning it to the appropriate thickness, assembling the sections at the optimum angle, and using it in Laue geometry for focusing. The approach is particularly well suited for high-resolution focusing optics for use at high photon energy. We present a theory of the MLL using dynamic diffraction theory and Fourier optics.

In the framework of TWINMIC, a project for the development of a multipurpose compact X-ray microscopy station capable of both scanning and full-field imaging, fabrication methods for tantalum zone plates are being developed at the Laboratory for Micro- and Nanotechnology of the Paul Scherrer Institut. Tantalum is deposited on supporting silicon or silicon nitride membranes by magnetron sputtering. The zone-plate patterns are transferred into the tantalum layer by reactive ion etching. Electron-beam lithography with continuous path control using a Leica LION LV1 e-beam machine has been used to make zone plates with diameters between 250 and 500 μm and thicknesses between 200 and 300 nm, all with an outermost zone width of 80 nm.

Easy to produce thin film waveguides can confine incident x-ray beams in one direction in guiding layers as thin as 10 nm. Consequently they can provide attractive beam dimensions for microscopy purposes. This report presents a simple model and analytical equations for the transmission calculation, which provide results consistent with the rigorous calculations based on a recursion technique. By use of these results the waveguide transmission can be compared directly with other microscopy objectives. Ideally x-ray waveguides can transmit the spatially coherent part of an incident radiation beam. The transmissions measured for state-of-the-art one-and two-dimensional waveguides are found to correspond to experimental efficiencies of the order of 0.5 for each confinement direction. Waveguides with thinner guiding layers cannot efficiently be used at smaller photon energy in highly collimated beams, instead the beam divergence in unfocused beamlines at state-of-the-art synchrotron radiation sources may eventually have to be increased to the larger angular acceptance of these waveguides by use of other focusing optics.

X-ray parabolic compound refractive lenses (PCRLs) are widely used to focus with high quality a synchrotron source radiation onto an object in a strongly demagnifying setup. This allows to generate a concentrated hard X-ray microbeam with lateral dimensions in the submicrometer range. Beryllium and aluminium PCRLs can be operated in energy range up to 60 keV, they are well adapted to the high heatload of the undulator beams at synchrotron radiation sources and used as the main optical element in various applications for microanalysis and microimaging. Unfortunately, in some experimental setups the single PCRL optical system has drawback consisted in the step-like focusing distances at different energies especially when the number of single lenses is not very large and/or the distance between the PCRL and object plane is fixed. An optical objective with varied focal length consisting of two PCRL's separated by a finite distance allows to create more flexible experimental setup and in some cases provides cost-effective solution by combination of PCRLs made of different materials (for example, all we know that Be lenses are more effective but also more expensive and dangerous than Al lenses) without significant loss in beam size, depth of field, background, flux, and gain.

Experimental results of synchrotron radiation focusing by parabolic planar compound refractive lenses made of glassy carbon are presented. The lenses with the curvature radii of 5 μm and 200 μm and with the geometric aperture of 40 μm and 900 μm correspondingly were developed with various techniques of laser evaporation. The number of bi-concave elements in the compound lenses was 4, 7 and 200. The planar lenses allow one to obtain linear focuses of the lengths comparable with the depths of their parabolic profiles. Use of two lenses in the cross geometry provides a formation of 2-dimensional focus. The experiments were performed at the ESRF at the bending magnet beamline BM-5. The minimum experimental focus size was as fine as 1.4 μm, which is larger than the theoretical estimation. The reasons of the focus broadening are discussed.

We present results of study of optical properties of extremely thick refractive lens when the thickness (or length) of a compound refractive lens is comparable with its focal distance. We tested a 2D parabolic compound refractive lens composed from 300-500 elements. Each element is a bi-concave lens made from aluminum with the curvature radius R = 0.2 mm and thickness 1 mm. Special long holders were designed and manufactured to keep up to 500 of elements. As far as the thin lens approximation is not valid we developed and used accurate theory of long parabolic compound lens for ray-tracing analysis. The experimental measurements were performed for the X-ray energies E = 20-30 keV. The measured focus distance and effective aperture correlate with the theory.

Planar parabolic refractive lenses are becoming the key optical elements for many hard x-ray microprobe and microscopy applications at third generation synchrotron radiation sources (e.g. the ESRF), as well as they are promising candidates for future X-ray free-electron lasers. In this paper we review all technological limitations taking place during fabrication of silicon and non-silicon refractive lenses and propose some approaches to overcome these limitations in order to fabricate high performance refractive lenses in terms of aperture, gain and focal spot size etc. We propose to use low-temperature silicon bonding techniques as alternative for very deep etching. Combination of two etched silicon structures by bonding of two lenses with relief to relief doubles the lens relief depth.

Sets of planar SU-8 cross lenses focusing in two directions have been fabricated by tilted deep X-ray lithography using an X-ray mask with tilted absorber structures. The profile of the absorber structures on the mask take into account the lithographic peculiarities of SU-8 resist to reproduce the designed profile of the lens elements exactly. The cross lenses are placed on one substrate and have identical focal distances, which allow to scan the spectral range from 5 keV to 30 keV by stepping the lens substrate from one lens to the next. Another set of cross lenses was developed with different quasi-parabolic profiles to obtain a large focus depth (up to several centimeters) with uniform intensity distribution in the micron focal spot. This together with the stepping possibilities between lenses satisfies the requirement of static spectroscopy experiments. For the truncated parabolic profile, these cross lenses consist of separate segments arranged in a new mosaic form. In comparison with the known “fern”-like kinoform profile, the lenses have been developed with smaller gain loss. The testing of the new sets have been performed at the undulator ID-18F and ID-22 beamlines (ESRF, Grenoble, France) and the experimental results are compared to simulations.

We report the results on experimental study of optical properties of Ni refractive lenses made by deep X-ray lithography and LIGA techniques. One- and two-dimensional lenses were tested at the ESRF ID15 beamline using wide energy spectrum from 40 keV to 220 keV. The focusing properties in terms of focal length, size of the focal spot/line and gain were studied. Sub micrometer focusing was measured in the energy range from 40 to 150 keV. The measured lens parameters were compared with ray-tracing analysis.

Si planar refractive lenses are well known and used for microfocusing and collimating synchrotron radiation. We have developed a method of X-ray lens bonding that allow to gain in a depth aperture of linear planar lenses by factor 2. Two identical Si refractive lenses were placed face-to-face with alignment precision of 3 μm and fixed by clamp. This lens system with total depth aperture of 200 μm has been tested on BM05 (ESRF, Grenoble, France) at 20 keV. Linear focusing with full width at half maximum of 8 μm and gain of 9 was measured at the distance of 270 cm from the lenses.

We present the design and test of a large aperture linear Fresnel zone plate (FZP) as a condensing lens for hard X-rays. The FZP is made of Si and is defined by electron beam lithography and chemical wet etching of <110> oriented silicon substrates. The central zone is 75 μm wide, the outermost lines have a width of 0.35 μm, and the structures have a height of 10 μm. These parameters result in an optimal focusing distance of 9 m, a phase shift of π, and a diffraction efficiency of about 30% at 7.3 keV. We have tested the FZP at the Anomalous Scattering Beamline (ID01) of the ESRF. The FZP was inserted in the vacuum chamber in the Optics Hutch at 38.5 m from the source and 9 m from the sample. We have studied the dependence of beam size and gain as a function of energy in the range from 7 to 8.5 keV. We found the optimum energy range as 7.2-7.4 keV, where the focused beam width was 100 μm and the experimental gain was 9.2, for an expected theoretical value of 9.6.

The X-ray power absorption by the Beryllium compound refractive lenses (CRL) installed in the ESRF ID10 front-end reaches 139 W. This non-negligible power leads to an excessive temperature in the lens such that the induced thermal stress is much larger than the yield stress of Beryllium. The thermal fatigue damage of the lens occurred after certain number of operation cycles. Sudden loss of focusing ability was observed recently after 6 ~ 7 years frequent operation. SEM and phase contrast images confirmed the damage of the CRL. Following these observations, optimization of some design parameters (width, and thickness of the thin part between two holes) of the CRL has been carried out as well as some operational parameters (cooling of the lens, vertical aperture of the X-ray beam on the lens). An optimized Beryllium CRL for the ID10 front-end should have a width of 10 mm instead of 2 mm and the thickness of the thin part between two holes should be increased to 0.2 mm. The temperature of the CRL can be reduced by a better cooling of the lens, for instance by improving the thermal contact between the Beryllium and the copper cooling block, or by reducing the vertical aperture of the X-ray beam from 4 mm to 2 mm (eventually to 1 mm).

The spatial resolution is a key optical parameter characterizing the performance of an imaging microscope. Zone plate based x-ray microscopy offers the highest spatial resolution over the whole electromagnet wave spectrum. Sub-20 nm resolution have been demonstrated with soft x-rays and sub-60 nm resolution have been obtained with multikeV x-rays using a laboratory source. There are two simple pathways to achieve sub-10 nm resolution x-ray imaging: (1) improving the fabrication technology to produce zone plates with an outermost zone width less than 10 nm and (2) using a higher diffraction order (such as the third diffraction order) of a currently available zone plate.

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.