Nuclear Instruments and Methods in Physics Research B 199 (2003) 531–535 www.elsevier.com/locate/nimb

Materials science research at the European Synchrotron Radiation Facility
ke Kvick A

*

European Synchrotron Radiation Facility, BP220, F-38043 Grenoble Cedex, France

Abstract The Materials Science Beamline ID11 at the European Synchrotron Radiation Facility in Grenoble, France is dedicated to research in materials science notably employing diffraction and scattering techniques. Either an in-vacuum undulator with a minimum gap of 5 mm or a 10 kW wiggler giving high-flux monochromatic X-rays generates the synchrotron radiation in the energy range 5–100 keV. The dominant research is in the area of time-resolved diffraction, powder diffraction, stress/strain studies of bulk material, 3D mapping of grains and grain interfaces with a measuring gauge down 5  5  50 lm, and microcrystal diffraction. A variety of CCD detectors are used to give time-resolution down to the millisecond time regime. Ó 2002 Elsevier Science B.V. All rights reserved.

1. Introduction Materials science was the initial motivation for the creation of the recent high-energy synchrotron radiation storage rings and these applications now account for more than 60% of the research at these facilities. Traditionally materials synchrotron radiation research was performed at lower photon energies and studies of electronic and surface properties have provided a wealth of information. However, materials science is the discipline of finding the relationship between structures in general and the properties of the studied materials. These studies provide the basis for rational materials engineering. It was thus clear that electronic features had to be complemented by bulk studies and the knowledge of crystal structures and phase

*

Tel.: +33-4-7688-2116; fax: +33-4-7688-2542. ˚ . Kvick). E-mail address: [email protected] (A

composition of the relevant materials. A concerted effort has been done to provide this additional information and several dedicated beamlines are now in operation to provide the complete picture necessary for the construction of new materials. This review will give a brief introduction to some of the materials research facilities at the European Synchrotron Radiation Facility in Grenoble. France notably the Materials Science Beamline ID11 [1,2]. The emphasis will be given to highenergy applications.

2. The radiation source The study of bulk materials and their phase composition, crystal structure and stress/strain relationships need high photon energies. Initially this could only be achieved by the use of multipole wigglers and the early stations were all based on high-power wigglers, which produced copious

0168-583X/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 5 8 3 X ( 0 2 ) 0 1 3 9 5 - 2

532

A˚. Kvick / Nucl. Instr. and Meth. in Phys. Res. B 199 (2003) 531–535

Fig. 1. The flux of the ESRF in-vacuum undulator.

amounts of radiation in the energy range up to several hundreds of keV s. The drawback was however often severe stability problems caused by the heat-load on the optical elements; many of these problems were overcome by elegant cooling measures such as liquid nitrogen cryogenic cooling. A 10 kW multipole wiggler initially powered the ID11 Materials Science Beamline. However, the rapid development of insertion technology has created small gap undulators with superior brilliance up to energies around 100 keV. The present research is now conducted with an in-vacuum undulator [3] with a minimum gap of 5 mm and its flux is presented in Fig. 1.

3. The detector systems In order to efficiently use the very high photon fluxes a development of the detector technology was badly needed in the early 1990s. The early experiments were in many cases performed by using imaging plate detectors, which have many attractive features such as high dynamic range,

good resolution and relatively low cost. The basic physics of these detectors however prevented recording duty cycles below approximately 1–2 s and the exposure times in the millisecond range and overhead readout time in the second range prompted the search for new detectors. The ESRF pioneered the development of image intensifiers coupled to CCD detectors [4]. This development has continued and the present readout speed for a full 2000  2000 pixel frame is about 60 ms. Readout speeds to 1 ms is achievable if the CCD chip is used as a storage medium. Detector frames of 2000  20 pixels can be downshifted in the millisecond range. At the ID11 Beamline we now employ several types of these detectors and the these FRELON cameras and several commercial CCD detectors are employed. The choice of detectors is decided by the study at hand.

4. Sample environment The ID11 Materials Science Beamline is geared towards a wide range of materials science applica-

A˚. Kvick / Nucl. Instr. and Meth. in Phys. Res. B 199 (2003) 531–535

tions and a major effort had been made to provide adequate sample environment choices. The samples can now be measured at temperatures from 9 to 1900 K. A special high-pressure laboratory provides support for high-pressure applications using diamond anvil cells or large-volume presses. Stress equipment such as INSTRON distortion devices is available. The ESRF has initiated a new project to provide magnetic fields up to 10–15 T. The software control system is easily adopted to customize experiments integrating the X-ray source to the detectors or the experimental sample conditions. The experiments may be performed with a highspeed (scan speed 20°/s) KUMA eight-circle diffractometer, a Bruker CCD diffractometer or a variety of customized set-ups. In a collaboration € National Laboratory in Denmark a with RISO ‘‘3D X-ray microscope’’ has been constructed. This facility provides X-ray beams in the energy range 40–90 keV focused to X-ray intensity in the spot in the range 1012 photons/s [5]. This facility now allows bulk characterization of stress/strain relationships in real industrial samples of metals and metals alloys. The resolution also allows the simultaneous study of the behaviour of hundreds of grains in multigrain materials.

5. Materials science applications The scientific program at the ID11 spans over a wide range including time-resolved studies of

533

chemical reactions, microcrystal structure determination, grain growth and nucleation, stress/strain studies and studies of polymer systems. Below a few illustrative examples are given. 5.1. Self-propagating high temperature synthesis Self-propagating high temperature synthesis [6] is an extremely rapid, low cost and easy method to fabricate a large range of materials such as intermetallics and composites. The synthesis is exothermic and the reaction proceeds through a compacted powder sample as a reaction front emanating from a starting point, usually initiated by a short external heat pulse. The reaction front, which can reach temperatures well above 2000 °C, proceeds with speeds up to 25 cm/s. We have concentrated our studies on the systems Al–Ni–Ti–C. Our studies have focussed on following the mixture of the powder components, with the goal of understanding the mechanisms of reinforcement [7]. In order to monitor the reaction in bulk we have used 0:2  0:2 mm beams at high energies (45 keV) to perform diffraction studies in transmission. The beam was centred on the middle of the sample and the diffraction through the 2 mm thick compressed powder samples was recorded with a FRELON CCD camera. The experimental set-up is illustrated in Fig. 2. An initiating current was supplied to a W wire at the bottom of the sample thereby starting the reaction and also provided a starting pulse to the

Fig. 2. The experimental set-up for SHS time-resolved diffraction experiment.

534

A˚. Kvick / Nucl. Instr. and Meth. in Phys. Res. B 199 (2003) 531–535

Fig. 3. The diffraction patterns recorded during the SHS reaction.

data collection system. The exposure time for each 1024  1024 pixel frame was 35 ms and the total readout and storage time was 65 ms/frame. In excess of 100 frames with 100 ms resolution were thus collected monitoring the initial, preheated, reaction and post-heated regions. Fig. 3 gives an example of a series of recorded powder patterns. The complete reaction could be followed from the initial melting of the aluminium through the transformation of Ti from the hexagonal to the cubic phase and the subsequent appearance the intermediate compounds NiAl and AlNi2 Ti, which themselves melt as the temperature increases. As Ti forms TiC the reactions finally terminates after a few seconds with the formation of AlNi and TiC forms. 5.2. Studies on working batteries The performance of solid-state batteries depends in a large part on the structural evolution of the electrodes during the intercalation/de-intercalation processes, which provide the batteryÕs power. An electrode may pass through several

solid solution or two-phase regions in the working range of the battery. It is thus of interest to understand the structural evolution of batteries during their working cycle in order to tune electrode materials for the desired voltage ranges and power densities. At the ESRF we have carried out studies [8–11] on real, working batteries during controlled electrochemical cycling by simultaneously measuring the diffraction pattern and electrochemical response. For example, a series of experiments carried out on Lix CoO2 based batteries has allowed us to elucidate in detail the entire phase sequence, in particular the mechanisms which lead to the gradual destruction of the batteries about 4.2 V. Furthermore, the excellent quality of the data makes it possible to follow on the second time scale the phase evolution and to obtain to high precision lattice and structural parameters (Fig. 4), as well as to solve previously unknown structures. 5.3. Crystal nucleation An understanding of crystal nucleation is important in order to produce materials of desirable

A˚. Kvick / Nucl. Instr. and Meth. in Phys. Res. B 199 (2003) 531–535

535

heating rate of 1 K/s. The crystallisation and final melting was followed by X-ray diffraction in transmission at an energy of 90 keV to allow transmission through the 2 mm thick ingots. A careful analysis of the results shows that a metastable phase is present during a few seconds of importance for understanding the amorphisation process. Similar experiments during rapid cooling show no presence of metastable phases.

References

Fig. 4. Bond lengths (above) and inter-layer spacings (below),
, for Lix CoO2 determined by X-ray powder diffraction in A during an in situ reduction cycle.

properties. Real-time studies of these phenomena were in the past only possible by indirect methods such as calorimetry or magnetic measurements since the nucleation process is too rapid to be followed by conventional X-ray methods. The highenergy X-rays at ID11 have been used to study nucleation in the novel Zr based bulk metallic glasses [12]. Bulk amorphous Zr55 Cu30 Al10 Ni5 prepared by casting from the liquid alloys were subsequently heated from room temperature at a


. Kvick, M. Wulff, Rev. Sci. Inst. 63 (1992) 1073. [1] A
. Kvick, W. Schwegle, Rev. Sci. Inst. 66 [2] M. Krumrey, A (1995) 1715. [3] J. Chavanne, P. Elleaume, P. Van Vaerenbergh, in: Proceedings of the PAC99 Conference, 1999, p. 2662. [4] J.P. Moy, Nucl. Instr. and Meth. 66 (1995) 1715.
. Kvick, Proc. Am. Inst. [5] U. Lienert, H.F. Poulsen, A Aeronaut. Astrophys. 1387/1485 (1999) 2067. [6] A.G. Merzhanov, Combustion and Plasma Physics of Materials, VCH, NY, 1990, p. 1. [7] C. Curfs, I.G. Cano, G.B.M. Vaughan, M.A. Rodriguez,
. Kvick, Int. J. Self-Propag. High Temp. X. Turillas, A Synth. 9 (2000) 331. [8] M.R. Palacin, G. Amatucci, M. Anne, Y. Chabre, L. Seguin, P. Strobel, J.M. Tarascon, G.B.M. Vaughan, J. Power Sources 81–82 (1999) 627. [9] L. Seguin, G. Amatucci, M. Anne, Y. Chabre, P. Strobel, J.M. Tarascon, G.B.M. Vaughan, J. Power Sources 81–82 (1999) 604. [10] M.R. Palacin, F. Le Cras, L. Seguin, M. Anne, Y. Chabre, J.M. Tarascon, G. Amatucci, G.B.M. Vaughan, P. Strobel, J. Solid State Chem. 144 (1999) 361. [11] J.M. Tarascon, G.B.M. Vaughan, Y. Chabre, L. Seguin, M. Anne, P. Strobel, G. Amatucci, J. Solid State Chem. 147 (1999) 410. [12] A.R. Yavari, A. Le Moulec, A. Inoue, W.J. Botta, G.B.M.
. Kvick, Mater. Sci. Eng. A 304–306 (2001) 34. Vaughan, A