Developing a second generation Laue lens prototype: high reflectivity crystals and accurate assembly
Nicolas M. Barrière, John A. Tomsick, Steven E. Boggs, Alexander Lowell, Peter von Ballmoos
aa r X i v : . [ a s t r o - ph . I M ] N ov Developing a second generation Laue lens prototype: highreflectivity crystals and accurate assembly
Nicolas M. Barri`ere a , John A. Tomsick a , Steven E. Boggs a , Alexander Lowell a , and Peter vonBallmoos ba Space Sciences Laboratory, 7 Gauss Way, University of California, Berkeley, CA 94720-7450 -USA b Institut de Recherche en Astrophysique et Planetology, UMR 5277, 9 av. du Colonel Roche,31028 Toulouse - France
ABSTRACT
Laue lenses are an emerging technology that will enhance gamma-ray telescope sensitivity by one to two ordersof magnitude in selected energy bands of the ∼
100 keV to ∼ Keywords:
Telescope, Soft gamma rays, Focusing optics, Crystals, Technological development
1. INTRODUCTION
A Laue lens is a single reflexion concentrator based on Bragg diffraction in the volume of a large number of smallcrystal slabs oriented in order to diffract incident radiation towards a common point [1] . Although the Braggdiffraction is very chromatic, the use of either mosaic crystals or crystals having curved diffraction planes allowsLaue lenses to cover broad bandpasses, of the order of a few hundreds of keV (as shown for instance in Ref. [2]and [3]) within the range from ∼
100 keV to ∼ ∗ . Another feature of Laue lensesis that they do not change the polarization of the incident signal, which can then be analyzed in the focal planeinstrument [5]. Laue lens telescopes are expected to provide an increase in sensitivity by one to two orders ofmagnitude with respect to existing instruments due to the fact that they focus from a large collecting area ontoa small detector volume, hence increasing dramatically the signal-to-noise ratio.Several fields of soft gamma-ray astronomy would benefit of the development of Laue lenses. One such topicis the study of explosion mechanism and progenitor nature of Type Ia supernovae (SNeIa). The spectroscopyand light curve of the line at 847 keV emitted by the decay chain of Ni, which is massively synthesized inSNeIa, would discriminate between the currently competing models. A Laue lens telescope as featured in theDUAL mission [6, 7] (proposed to the European Space Agency for the third medium class mission AO of the
Corresponding author: N.B.E-mail: barriere at ssl.berkeley.edu ∗ The mosaicity refers to the full width at half maximum of the angular distribution of crystallite, the tiny perfectcrystals that compose a mosaic crystal according to Darwin’s model. By extension this term is used to describe theangular spread of any crystal, mosaic crystals or crystals having curved diffracting planes. × − ph/s/cm (3 σ , 1 Ms) in the 3-% broadenedline at 847 keV, enough to observe a dozen events each year out to ∼
40 Mpc and make a breakthrough in ourunderstanding of their physics (see e.g. Ref. [8, 9, 10]).Another topic is the study of the electron-positron annihilation radiation at 511 keV. This line has beenobserved for more than 30 years from the Galactic center, yet it is still unclear whether known sources canaccount for all the 10 positrons that annihilate every second in the Galactic bulge [11]. New observationalclues are needed but both improved sensitivity and improved angular resolution are required. A Laue lenstelescope could provide the way to probe small sky regions to check for structure in the emission and probe somecandidate sources, like X-ray binaries.A Laue lens observing in a selection of narrow bands of the 100 keV to 1MeV domain could bring valuableclues concerning the emission mechanisms - yet poorly understood - of blazars and AGNs. The determination fora couple of AGNs of the turnover from synchrotron to inverse-Compton branches would constrain the emissionmodels. Many non-blazar AGN show an undisturbed power law up to 200 keV. From accretion theory, a turn-overin the spectrum is expected, but a precise measurement has not been possible in many cases. Sources like Cen Aseem to break somewhere around 500 keV [12], an energy range hardly accessible so far. A Laue lens telescopewould give important insight into the energetics close to the central engine of super massive black holes.Observations of the hard X-ray tails in Galactic black-hole binaries would also bring clues on the physics ofthese objects. At high mass accretion rates when these systems are in the “Steep Power-Law” state, they exhibita power-law spectral component that extends to at least hundreds of keV [13]. No spectral break has beendetected yet, due to the lack of sensitivity of present instruments above 100 keV. The production mechanism ofthis spectral component is currently unknown. In the“Hard” state, a spectral break is often seen, but observationswith INTEGRAL indicate the presence of at least two components in the 100-600 keV bandpass [14], indicatingthat our understanding of this part of the spectrum is still incomplete.In order to meet the sensitivity requirements for these objectives, the best possible efficiency is needed for theLaue lens. That translates into crystals with high reflectivity, densely packed, and accurately oriented. We havetaken up the challenge of developing such lens at SSL. We report in this paper on the status of this effort. Insection 2, we present the work we are conducting to enable efficient crystals for high energy Laue lenses. Section3 presents the crystal orientation requirements to build scientifically viable Laue lenses. Then, in section 4, wepresent the Laue lens assembly facility that we have constructed at SSL, and finally we show in section 5 thefirst results we obtained for gluing crystals.
2. CRYSTAL STUDY
For several years, we have been working to identify crystals suitable to build efficient
Laue lenses. For thispurpose the first task was to establish a list of materials fulfilling basic requirements: the material must exist ina crystalline state at room temperature, it must not be ductile, and it must not be (too) toxic or radioactive.Then, we excluded from the list all the crystal lattices not based on a cube as they are far more difficult towork with or show lower diffraction efficiency. Crystals composed of different atoms tend to have larger latticeparameters than pure materials, which dramatically decreases their diffraction efficiency. So, we limited oursearch to pure materials and bi-component crystals. Figure 1 is an updated version of the plot published inRef. [19]. It shows the peak reflectivity of the crystals currently in our list for three different energies. Moretwo-component crystals might join this list, but it is complete regarding pure materials.Once this theoretical list was established, one had to determine wether these materials are available in singlecrystals with large volume, and, if so, with what mosaicity. Depending on the project (focal length, energyband), the optimal mosaicity ranges from 0.5 to 2 arcmin. So, over the past few years we tried to procuresamples of these crystals and go to high energy X-ray or gamma ray facilities to determine their quality. So far,we have been able to measure samples of: CaF , Si, Cu, InP, GaAs, Ge, Rh, Ag, CdTe, Ta, W, Ir, Pt, Au andPb. Amongst this list, only Ta, W, Ir and Pt were not of sufficient quality (i.e. mosaicity larger than 10 arcminand very irregular), although for most of them, only one sample was measured. These crystals are of interestfor energies above ∼
750 keV thanks to their high electron density. However, they are very difficult to growproperly due to their high melting point, and, as a consequence, are very expensive. The advent of one of these2 igure 1. Reflectivity at 150, 511 and 847 keV (respectively red, green and blue points) of a selection of crystals thatare potentially interesting for the realization of a Laue lens. The crystals are sorted from left to right by increasingmean atomic number. The reflectivity was calculated assuming they are mosaic crystals with 45 arcsec of mosaicity. Thedynamical theory of diffraction was used with different crystallite size for each energy [16, 19]: 30 µ m, 60 µ m and 90 µ mrespectively for 150 keV, 511 keV and 847 keV. The thickness is computed to maximize the peak reflectivity within thelimits 2 mm T
15 mm. four crystals for use in Laue lenses would further improve the overall lens efficiency, as having several crystalsthat diffract the same energy from a given radius allows the most efficient one to be chosen.Recently, we have focused our interest to high energy Laue lenses working around 511 keV and 847 keV [16]. These lenses are composed of Rh, Ag, Pb, Cu and Ge with an option to replace Ag by Au in the 847 keVlens (these two crystals have very similar lattice parameters, and thus they can be swapped without any designchange). Since Cu and Ge have already been proven to be available for use in Laue lenses [18,20], we are focusingon enabling Ag, Rh and Pb. In all three cases, several cut samples or raw ingots were already measured withmosaicity ranging between 0.5 and 1 arcmin. We now want to establish the time, cost, and overall difficulty ofproducing cut pieces with given orientation, thickness, and mosaicity. We ordered 5 pieces of Rh111, Rh200,Rh220, Rh311, Ag111, Ag200, Ag220, Pb111, Pb200 and Pb220 (a total of 50 crystals) of 5 × and 3 to 5mm thickness from Mateck (Juelich, Germany). The aim is to obtain crystals with average mosaicity lower than1.5 arcmin, so the rejection threshold for raw ingots is set to 1 arcmin (the cut is expected to induce damages).To reach this goal, we are engaged in an iterative process between ingots production and characterization inorder to refine the ingot growth technique. We also check the mosaicity of the cut pieces as they are producedin order for Mateck to refine the cut parameters if necessary.Figure 2 shows the mosaicity as function of position along the ingot axis for three samples recently measuredat the Institute Laue Langevin (Grenoble, France) using the DIGRA beamline specifically developed for bulkcharacterization of crystals. The ingots were probed using a 5 × igure 2. Mosaicity versus position for three ingots produced by Mateck and measured in a 517 keV beam at ILL /DIGRA. On the left results are shown for a silver ingot measured across the diameter (10.7 mm) using the refelxion 200.This crystal was measured twice, rotating it of 90 ◦ about its axis between the two series of measurements. On the right,results are shown for a Rh 111 ingot (in two directions as well), measured across it diameter of 11 mm. The bottom plotshows results for a Pb 200 crystal, measured across its diameter of 12 mm.
3. CRYSTAL ORIENTATION TOLERANCES
Let’s first define the referential for the orientation of a crystal on the lens substrate. Ω B , Ω R and Ω Z are thethree axes of rotation defined as: • Ω B : in the lens plane, perpendicular to the radius connecting the crystal center to the lens center. This is theBragg angle. • Ω R : in the lens plane, parallel to the radius connecting the crystal center to the lens center. • Ω Z perpendicular to the lens plane (i.e. parallel to the optical axis of the lens).Depending on the focal length and the mosaicity of the crystals employed, the requirement on crystal ori-entation accuracy can vary dramatically. As shown on the left panel of Figure 3, the sensitivity of a Laue lenstelescope improves with its focal length. The improvement can vary slightly depending on whether the crystalsize and mosaicity are kept constant or not, but, in any case, a longer focal length produces a more sensitivetelescope. One should notice, however, that a longer focal length means a larger number of crystals, and abigger and heavier lens. The second point shown in this plot is that a smaller crystal mosaicity produces amore sensitive telescope. Again, there are subtleties (especially for focal length smaller than 10 m), but this is ageneral trend. The point is that a longer focal length and a smaller mosaicity call for very well oriented crystals,otherwise the performance degradation can be dramatic.In the following part of this discussion, we consider the case of the SNeIa Laue lens presented in Ref. [16],i.e. with a focal length of 30 m and made of 10 ×
10 mm crystals having a mosaicity of 45 arcsec. The middle4 igure 3. Left:
Sensitivity of a Laue lens telescope focusing in a given energy band (700 keV - 900 keV) as a function of itsfocal length. The crystal size is kept constant at 10 ×
10 mm , results are shown for four different crystal mosaicities: 30,45, 60 and 90 arcsec. Middle:
Crystal disorientation (1 σ , assuming a Gaussian distribution) that produces a sensitivityloss of 5, 10, 15 and 20% as function of the focal length. The calculation is done for a lens focusing in the 700 keV- 900 keV band, made of crystals of 10 ×
10 mm and 45 arcsec of mosaicity. Right:
Sensitivity loss as function of thedisorientation of the two angles Ω R and Ω Z assuming a disorientation of 10 arcsec for Ω B . The axes show the 1 σ values,assuming a Gaussian distribution. Each contour represents a sensitivity degradation of 4%, the white area being the bin0% - 4%. panel of Figure 3 shows the angular deviation of the Bragg angle (Ω B ), which leads to sensitivity loss of 5, 10,15 and 20%, as a function of the focal length. This assumes a Gaussian distribution of the angular deviation.For a 30-m focal length lens, a standard deviation of 10 arcsec already degrades the sensitivity by 5%.Keeping this value of 10 arcsec for Ω B , we estimate now the tolerance for the other angles. The right panel ofFigure 3 shows that the requirements on Ω R and Ω Z are somewhat more relaxed. The sensitivity degradation islower than 4% if the angular deviations are within 15 arcmin for both Ω R and Ω Z . This is not too constrainingas crystals can be cut quite easily with orientation accuracy of the order of 10 arcmin. It means that we can relyon the external faces of the crystals to orient Ω R and Ω Z properly.
4. ASSEMBLY METHOD DEVELOPED AT SSL
As briefly mentioned in Ref. [16], a 13-m long beamline fully dedicated to the development of Laue lens wasconstructed at SSL (Figure 4). It is composed at one end of an X-ray generator (XRG) working at 150 kVand 450 µ A (lent by P. von Ballmoos’ group, IRAP, France) and at the other end by a cross-strip high puritygermanium detector. The spectrum incident on the samples is a continuum from ∼
50 keV to 150 keV peakingat ∼
100 keV (the low energies are attenuated by a 1.5-mm thick Cu filter). In the XRG, the electron spot on thetungsten target is ∼ ×
19 voxels,with the possibility to extract the spectrum out of a selected region and fit a Gaussian peak. A repeatability of0.035 keV was observed in the fit of the diffracted peak of a perfect 5 × × Si 111 crystal at 100 keV doing200-s integrations, which turns into an orientation accuracy of 1.5 arcsec.The Laue lens assembly station is composed of a crystal holder (Figure 5) and a substrate holder (Figure 6).The crystal holder is composed of a vacuum chuck on top of a set of three rotation stages and one translationstage. It allows for the crystals to be held firmly (but temporarily) in place while they are oriented with respectto the beam and brought against the substrate for gluing. The substrate holder is composed of a translationstage (allowing for changes in the radius of the lens to be populated) and a rotation stage with an axis thatdefines the optical axis of the lens. This setup is actually incomplete as we are missing a rotation stage to controlprecisely the orientation of the substrate with respect to the beam. However, our goal was to demonstrate that5he gluing of crystals can be done with high accuracy with respect to the X-ray beam, which doesn’t require thesubstrate to be aligned with respect to the beam, as long as it is kept fixed.The following steps describe the method we used to orient and glue crystals:1. The crystal is setup at the tip of the vacuum chuck. The two angles Ω R and Ω Z (see section 3) rely on theaccuracy to which the crystals are cut and are set manually as the crystal is placed on the vacuum chuck.2. The crystal is pre-oriented within a few keV of the desired energy by looking at its diffracted peak energyand brought in front of the hole in the substrate where it will be glued. For this trial, we used a distanceof ∼ µ m at the closest point.3. The orientation of the substrate is measured with an autocollimator (Its optical axis is the reference, andconsequently it is assumed fixed).4. The crystal orientation (Bragg angle, Ω B ) is then finely tuned by rocking it until the desired diffractedenergy is observed on the detector. Integrations of 200 s were done to insure the aforementioned angularaccuracy.5. Some glue is injected through the substrate hole. Two-part epoxy MasterBond EP30-2 that featuresextremely low shrinkage upon cure of 3 × − was used for this trial.6. The crystal is maintained in position for 6 hours before releasing slowly the vacuum and sliding back thevacuum chuck.7. The orientation of the substrate is measured optically, and the energy diffracted by the crystal is measured.
5. RESULTS OF THE FIRST TRIAL
Following the method described in section 4, we glued 10 crystals over the course of 5 consecutive days of August2011 (see Figure 6). Perfect silicon 111 crystals of 5 × × × produced at IKZ (Germany) were used for thistrial. The substrate is a 1-inch thick aluminum plate carved from the back to leave 1 mm behind the crystals.The aim of this trial was to orient the crystals to diffract at 100 keV.The crystals are arranged in two partial concentric rings: For the inner ring the substrate was set roughlyperpendicular to the beam, which means that there is a wedge of glue of about 1.13 ◦ behind each crystal due tothe Bragg angle. For the second ring, however, the substrate was brought nearly parallel to the oriented crystals,making the bond line nearly parallel. The aim was to test two different substrate geometries, either flat (and theBragg angle is created by a wedge of glue) or paraboloid, as first proposed by Lindquist [21], where the normalto the substrate stays parallel to the diffracting planes of crystals at any radius (Figure 7). In the case of crystalscleaving along their crystalline planes, the latter case would in theory allow for just laying the crystals on thesubstrate and having them be correctly oriented. Unfortunately, we do not know any crystal that is efficient fordiffraction at high energy that also cleaves.In the flat substrate case, we noticed that the glue shrinkage upon cure induced a very reproducible deviationof − ± ◦ implies for a 5-mm large crystal a wedge of ∼ µ m. According to the glue manufacturer, the glue shrinkage should then be of 0.03 µ m, which corresponds to ananglular deviation of 1.2 arcsec. We need to investigate why we observed a deviation 50 times larger. However,the very positive result is the 3 arcsec standard deviation over these 5 crystals. This shows that our methodproduces reproducible results.In the ’paraboloid’ substrate case, the average deviation over the five crystals is − ± igure 4. Bird’s eye view of the X-ray beamline in the high bay at SSL. The X-ray generator is enclosed in a lead casingon the right hand side of the room (wooden boards) and the Laue lens assembly station is in the plexiglass enclosure onthe left hand side of the room.Figure 5. Left:
Laue lens assembly station. From the left to the right, we can see the Ge detector (golden cryostat),the autocollimator, the crystal tower, the substrate tower and the tungsten slits defining the beam size illuminating thecrystal.
Right:
Crystal holder composed of a vacuum chuck on top of a set of rotation and translation stages.Figure 6.
Left:
Crystal being held in position onto the substrate while the glue is curing.
Right:
The substrate populatedwith 10 Si crystals. The reference mirror viewed by the autocollimator is visible in its upper-right corner. igure 7. Left:
Comparison between the two Laue lens substrate options tested. On the left, the substrate is a flat discwhile on the right it is a paraboloid.
Right:
Angular deviation with respect to the desired orientation for each of the 10crystals glued on the aluminum substrate. in the range 22.0 ◦ C - 29.5 ◦ C during the gluings. Although the influence of the temperature variation was notclearly established, we believe a light thermal control would increase the stability of the setup.Three crystals over the course of 24 hours could be glued with the present method and glue. Setting upa crystal at the tip of the vacuum chuck, getting a diffraction spectrum (which tells us if this crystal is goodenough as well as its orientation), then bringing it in gluing position and injecting the glue takes less than 30min. With a glue curing in one hour, a dozen crystals per day could be mounted. Now that we demonstratedthat it is possible to fix crystals with accuracy better than 10 arcsec and dense packing factor, our next endeavorwill be to speed up the process.
6. CONCLUSIONS
At SSL, we are taking up the challenge of the development of the first scientifically viable gamma-ray optic, basedon densely packed, accurately oriented, efficient crystals. The very first test of our Laue lens assembly setupand process showed excellent results. Soon, a second trial will take place with a refined substrate (paraboloid).The glue we have been using takes six hours (at least) to cure. We estimate that it is necessary to decrease thistime down to at the most two hours in order to be able to assemble an astronomical Laue lens within one to twoyears. Indeed, in most current projects, the lens is composed of at least 5000 crystals. Now that we have proventhat the required precision can be met, we will work to speed up the process.In parallel we are conducting studies with Mateck to enable high efficiency crystals for high energies. We hopethat this activity will demonstrate the availability of pure rhodium, silver and lead crystals and consequentlyraise their Technology Readiness Level. The 50 crystals we ordered will be delivered in Spring 2012.Laue lenses telescopes are not designed to be general purpose observatories, but rather to answer very specificquestions requiring one to two order of magnitude sensitivity improvement with respect to existing instruments.Black holes (both stellar mass and super-massive), Galactic positron origin and Type Ia supernovae are threeexamples of topics that would greatly benefit from the advent of Laue lens telescopes.
ACKNOWLEDGMENTS
The authors wish to thank Dr. Nikolay Abrosimov (IKZ, Germany) for providing the Si crystals that were usedin this study. The authors also wish to thank Dr. Michael Jentschel, Gilles Roudil and Dr. Pierre Bastie for theirprecious help during the characterization of the silver, rhodium and lead ingots at ILL on the DIGRA beamline.8
EFERENCES
1. N. Lund. “A study of focusing telescopes for soft gamma rays”,
Exp. Astronomy , :259–273 (1992).2. N. Barri`ere, L. Natalucci and P. Ubertini. ”Hard X-ray/soft gamma-ray polarimetry using a Laue lens”, in”X-ray Polarimetry: A New Window in Astrophysics” Cambridge university press edition , p. 88 (2010).3. F. Frontera, A. Pisa, P. de Chiara, and et al. “Exploring the hard X-/soft gamma-ray continuum spectrawith Laue lenses”. , vol. of ESA Special Publication , p. 323 (2005).4. N. M. Barri`ere, L. Natalucci, N. Abrosimov, P. von Ballmoos, P. Bastie, P. Courtois, M. Jentschel,J. Kn¨odlseder, J. Rousselle, and P. Ubertini, III. “Soft gamma-ray optics: new Laue lens design and perfor-mance estimates”.
Proc. SPIE , : 74370K–74370K–11 (2009).5. N. Barri`ere, L. Natalucci, and P. Ubertini. “Soft gamma-ray polarimetry using a Laue lens telescope: Thegamma-ray imager”, Proceedings of Science , PoS(CRAB2008)019 (2008).6. P. von Ballmoos, T. Takahashi, and S. E. Boggs. “A DUAL mission for nuclear astrophysics”,
NuclearInstruments and Methods in Physics Research A , :431–433 (2010).7. S. Boggs, C. Wunderer, P. von Ballmoos, T. Takahashi, N. Gehrels, J. Tueller, M. Baring, J. Beacom,R. Diehl, J. Greiner, E. Grove, D. Hartmann, M. Hernanz, P. Jean, N. Johnson, G. Kanbach, M. Kippen,J. Kn¨odlseder, M. Leising, G. Madejski, M. McConnell, P. Milne, K. Motohide, K. Nakazawa, U. Oberlack,B. Phlips, J. Ryan, G. Skinner, S. Starrfield, H. Tajima, E. Wulf, A. Zoglauer, and A. Zych. “DUALGamma-Ray Mission”, astro2010: The Astronomy and Astrophysics Decadal Survey (2010).8. N. M. Barri`ere, J. A. Tomsick, and S. E. Boggs. “Unveiling physical processes in type ia supernovae with aLaue lens telescope.” PoS(INTEGRAL 2010)108 (2010).9. J. Kn¨odlseder, P. von Ballmoos, F. Frontera, A. Bazzano, F. Christensen, M. Hernanz, and C. Wunderer.“GRI: focusing on the evolving violent universe”,
Exp. Astronomy (1):121–138 (2009).10. M. D. Leising. “Focusing supernova gamma rays.”, Exp. Astronomy :49–55 (2005).11. J. Kn¨odlseder, P. Jean, V. Lonjou, G. Weidenspointner, N. Guessoum, W. Gillard, G. Skinner, P. vonBallmoos, G. Vedrenne, J.-P. Roques, S. Schanne, B. Teegarden, V. Sch¨onfelder, and C. Winkler. “The all-sky distribution of 511 keV electron-positron annihilation emission.” Astronomy & Astrophysics :513–532(2005).12. A. A. Abdo, et al. . “Fermi Large Area Telescope View of the Core of the Radio Galaxy Centaurus A.”
ApJ :1433-1444 (2010).13. J. E. Grove, W. N. Johnson, R. A. Kroeger, K. McNaron-Brown, J. G. Skibo, and B. F. Phlips. “Gamma-RaySpectral States of Galactic Black Hole Candidates”.
ApJ :899-+ (1998).14. L. Bouchet, M. del Santo, E. Jourdain, J. P. Roques, A. Bazzano, and G. DeCesare. “Unveiling the HighEnergy Tail of 1E 1740.7-2942 With INTEGRAL”.
ApJ :1871-1876 (2009).15. W. Coburn, S. E. Boggs, S. Amrose, R. P. Lin, M. T. Burks, M. Amman, P. N. Luke, N. W. Madden, andE. L. Hull. “Results of charge sharing tests in a ge-strip detector.”,
IEEE Transactions on Nuclear Sciences :226–229 (2001).16. N. M. Barri`ere, J. Tomsick, S. E. Boggs, J. Rousselle, and P. von Ballmoos. “Development of efficient Lauelenses: experimental results and projects.”, Proc. SPIE :773226–773226–11 (2010).17. E. C. Bellm, S. E. Boggs, M. S. Bandstra, J. D. Bowen, D. Perez-Becker, C. B. Wunderer, A. Zoglauer,M. Amman, P. N. Luke, H.-K. Chang, J.-L. Chiu, J.-S. Liang, Y.-H. Chang, Z.-K. Liu, W.-C. Hung, C.-H.Lin, M. A. Huang, and P. Jean. “Overview of the Nuclear Compton Telescope”,
IEEE Transactions onNuclear Sciences :1250–1256 (2009).18. P. von Ballmoos, H. Halloin, J. Evrard, G. Skinner, N. Abrosimov, J. Alvarez, P. Bastie, B. Hamelin,M. Hernanz, P. Jean, J. Kn¨odlseder, V. Lonjou, B. Smither, and G. Vedrenne. “CLAIRE’s first light”, NewAstronomy Review :243–249 (2004).19. N. M. Barri`ere, J. Rousselle, P. von Ballmoos, N. V. Abrosimov, P. Courtois, P. Bastie, T. Camus,M. Jentschel, V. N. Kurlov, L. Natalucci, G. Roudil, N. Frisch Brejnholt, and D. Serre. “Experimentaland theoretical study of the diffraction properties of various crystals for the realization of a soft gamma-rayLaue lens”, Journal of Applied Crystallography (5):834–845 (2009).90. J. Rousselle, P. von Ballmoos, M. Jentschel, N. V. Abrosimov, N. M. Barri`ere, P. Courtois, and G. Roudil.“Cu and SiGe gradient crystal productions and measurements for a Laue lens application”, submitted toExperimental Astronomy .21. T. R. Lindquist and W. R. Webber. “A focusing x-ray telescope for use in the study of extraterrestrial x-raysources in the energy range 20-140 kev.”, Canadian Journal of Physics46