Laue - Topography at the Synchrotron Light Source "ANKA"
A. N. Danilewsky, J. Wittge
P. McNally, D. Allen (RINCE, Dublin City University, Dublin, Ireland)
P. Vagovic, Z. Li, T. Baumbach (ISS, Karlsruhe Institute of Technology)
At the synchrotron – light source ANKA (Angström Karlsruhe) the white beam X-ray topography beam line is in operation [1]. A small source point of 0.5 x 0.2mm2 at a bending magnet of the 2.5 GeV storage ring provides light with a critical wavelength of 2 Å. Characteristic features of the beam line are the first slit system in the front end, no optical components inside the 30 m long beam line and only one Be-window directly in front of the experiment shown in fig. 1. The topographs are collected on photographic films or by a digital X-ray camera [3]. Various geometries can be easily adjusted:
. Large Area Transmission-Topography Thin Section Transmission-Topography
. Large Area Back Reflection-Topography Section Back Reflection-Topography
. Grazing Incidence Topography
Fig. 1:The topography experiment: (1) Cooled Be-Window, (2) Fast shutter,
(3) Sample on two -circle-goniometer and
x-/z-translation up to 300 mm
(4) Film camera for transmission geometry
(5) Digital camera system for transmission geometry on x-/z-translation
Every diffraction vector which fulfils Bragg’s law results in one topograph from the same volume of the sample during one single exposure and results in a Laue pattern of topographs [2]. The topographs are magnified with conventional light microscopy and digitalised by a CCD-camera.A typical diffraction pattern for InP (211) is shown in fig. 2 with a Laue simulation for indexing the reflections.
Fig. 2: Laue pattern of InP (211), tilted by 10° to find the 111 - reflection
With a digital X-ray camera high resolution topographs from a single diffraction vector can be collected directly [3]. Fig. 3 shows InP:S as an example with straight V-shaped lines which are 60° dislocations of the type b = 1/2a [110]. Even single curved dislocations at the border of the crystal with very high dislocation density (about 10 5cm -2) are resolved. Dopant inhomogeneities (striations) produce long range strain of several milimeters.
Fig. 3: Large area transmission topography of highly doped InP taken with the digital camera, exposure time 60 minutes, distance crystal – film 56 cm, diffraction vector 111:
By image post processing of the same original TIFF-file, the topographs are magnified and opimised with respect to high dynamic range or for high sharpness. Visible are straight 60- dislocations with Burgers-vector
b = 1/2a <110>.
Curved dislocations can be resolved up to a dislocation density of 10^5-10^6 cm^ -2.
Dopant inhomogeneities (striations) produce long range strain of several milimeters.
X-ray topography of large wafers, e.g. 300 mm Silicon [4]
At the Topo-Tomo beamline at ANKA the area for a single topograph is limited to about 5 x 8 mm 2. The standard size of Si-wafer is actually 300 mm in diameter therefore about 2.200 topographs are needed to map one wafer completely. Using conventional X-ray film this becomes impossible because every single film has to be changed manually and needs in addition to 1-2 seconds exposure time another 8 minutes developing time.
The speed can be increased drastically if a digital camera set-up [3] is used instead of the X-ray films. A complete 300 mm Si wafer can be automatically mapped in high resolution within about 2.5 hours. The experiment is shown in fig. 4.
Fig. 4: Set-up for 300 mm wafer topography
The system was tested using a (001)-oriented 300 mm Si wafer which was double side polished. To find a 220 reflection in transmission mode (fig. 1), the wafer was tilted by 12°. The camera was adjusted in a 130 mm distance to collect the diffracted beam with the scintillating crystal CdWO 4. By opening the aperture of the macroscope for maximum brightness an exposure time of 1 second or even shorter can be chosen where defects are still visible.
With camera integration times of 0.1 seconds the time for full 300 mm wafer mappings are reduced to less than 1 hour and the next-generation of 450 mm Si wafers will be handeled easily.
High temperature in-situ X-ray topography [5], [6], [7]
White beam X-ray diffraction imaging (topography) with an optimised CCD-detector system is used to monitor in-situ and in real time the nucleation, growth and movement of dislocations in silicon at high temperatures. It can be shown, that damage like microcracks and the surrounding strain fields in a wafer act as sources for dislocation loops which end in slip bands far away from the source. The dislocations are arranged in channels of parallel {111} glide planes which become visible as bands of parallel surface steps when the dislocations slip out on the back or front side of the wafer. The width of such a channel or band depends on the dimensions of the damaged volume where the dislocations nucleate. Fig. 5 shows the mirror heater set-up at the Topo-Tomo beamline.
Fig 5: Mirror heater at Topo-Tomo beamline:
(1) Window for diffracted beam
(2) Mirror shell
(3) Lamp housing
(4) Translation/rotation
(5) Power supply
(6) Temperature control
Fig. 6 shows the open mirror heater with the windows for incoming white and the diffracted beams respectively, the sample in the centre and one of the two halogen lamps in the background.
By tilting the sample 12° with respect to the incoming beam, the 0-22 reflection is projected to the detector
Fig. 6:View into the open mirror heater with inserted sample and geometry of diffraction. At a tilt of 12° the 0-22 diffracted beam from a (100) Si sample hits the camera system
The optimisation of optics and camera allows integration times of less than 1 s which produce topographs with sharp contrasts even at high temperature and during fast temperature changes. At frame frequencies of 1 – 10 Hz a real time monitoring of dislocation dynamics becomes possible.Fig. 7 shows a sequence of topographs from the real time movie. The contrasts
originate from 4 indents with 400 mN and 4 indents from 600 mN loads (from left to right) with 1 mm distances (fig. 7a) and are the sources of dislocation loops (fig. 7b).The loops multiplay and increase with time (fig. 7c) and finally intense slip bands arrive in the field of view from the edge of the sample (figs. 7c,d).
Fig. 7:equence of transmission topographs from video 1 showing contrasts originating from 4 indents with 400 mN and 4 indents from 600 mN loads (from left to right) with 1 mm distances (sample no. 9, contrast inverted and rotated by 180°): (a) at t = 0 s and 953°C: first dislocation loops become visible (b) at t = 10 s and 981 °C: multiplication of dislocation loops (c) at t = 25 s and 994 °C: multiplication of dislocation loops and dense slip bands arising from the edge (d) at t = 40 s and 1001 °C: screw segments from the adjacent indents reach equivalent dislocations and annihilate because of inverse Burgers vectors or slip out at the back side of the wafer
And here the movie (shockwave flash format, quick motion):
The mechanism of slip band formation is explained in fig. 8:
Fig. 8:Formation of slip bands from dislocation loops around indents schematic:
(a) Crystallographic situation with 2 out of the 4 {111} glide planes in a (100) Si wafer with 2 indents at the front surface
(b) First dislocation loop formed around an indent consisting of pairs of half loops on the inclined (111) and
(1-1-1 ) glide planes containing the [0-1 1] direction. Each half loop consits of two inclined 60° and one screw dislocation with an initial single Burgers vector a/2[0-1 1]
(c) Multiplication of dislocation loops of the same type. The screw segments slip parallel to the surfaces of the wafer
(d) Screw segments slipped out at the backside of the wafer. The remaining inclined 60° disloactions running from front to back side of the wafer form sets of slip bands in some distance of the dislocation source
Literatur:
[1] A. N. Danilewsky, R. Simon, A. Fauler, M. Fiederle, K. W. Benz,
Nuclear Instruments and Methods in Physics Research NIM-B, vol.199, 71-74 (2003): White Beam X-Ray Topography at the Synchrotron Light Source ANKA, Research Centre Karlsruhe
[2] T. Tuomi, K. Naukkarinen, P. Rabe, phys. stat. sol.(a) 25 (1974), 93: Use of synchrotron radiation in X-ray diffraction topography
[3] A. N. Danilewsky, J. Wittge, A. Rack, T. Weitkamp, R. Simon,
Nuclear Instruments and Methods in Physics Research NIM-B 266(9) (2008), 2035-2040: White Beam Synchrotron Topography Using a Digital X-ray Imaging Detector at TOPO-TOMO Beamline at ANKA
[4] A. N. Danilewsky, J. Wittge, A. Rack, T. Weitkamp, R. Simon, T. Baumbach and P. McNally Mater. Sci-Mater.: El. 19(Supl.1) (2008), 269-272: White Beam Topography of 300 mm Si -Wafer
[5] A. N. Danilewsky, J. Wittge, A. Hess, A. Cröll, D. Allen, P. McNally, P. Vagovic, A. Cecilia, Z. Li, T. Baumbach, E. Gorostegui-Colinas and M. R. Elizalde, Nuclear Instruments and Methods in Physics B, NIM-B 268(3-4) (2010), 399-402, DOI: http://dx.doi.org/10.1016/j.nimb.2009.09.013 : Dislocation Generation Related to Microcracks in Si-Wafers: High Temperature In-situ Study with White Beam X-Ray Topography"
[6] J. Wittge, A. N. Danilewsky, D. Allen, P. McNally, Z. Li, T. Baumbach,
E. Gorostegui-Colinas, J. Garagorri, M. R. Elizalde, D. Jaques, M. C.
Fossati, D. K. Bowen and B. Tanner, J. Appl. Cryst. 43(5) (2010),
1036-1039: Dislocation sources and slip band nucleation from indents on silicon wafers
[7] A. N. Danilewsky, J. Wittge, A. Hess, A. Croell, D. Allen, P. McNally,
P. Vagovic, T. dos Santos Rolo, Z. Li, T. Baumbach, E.
Gorostegui-Colinas, J. Garagorri, M. R. Elizalde, M. Fossati, D. K.
Bowen and B. Tanner, J. Crystal Growth 318 (2011), 1157-1163, DOI: http://dx.doi.org/10.1016/j.jcrysgro.2010.10.199: Microcracks as Sources for Dislocations and Slip Bands in Silicon: In Situ Study by X-ray Diffraction Imaging
