Orientation Microscopy and Ion Backscattering

A utomated C rystal O rientation M icroscopy (ACOM), or in short Orientation Microscopy (OM), on bulk surfaces is usually based on B ackscatter K ikuchi D iffraction (BKD) in the SEM. A commercial acronym for this technique is EBSD. Spatial resolution is in the range of <0.05 µm, accuracy is better than 0.5°, and speed of on-line measurement is actually between 40 000 patterns/hour with conventional analog detectors and more than 3 million patterns/hour with Fast EBSD.

An alternative technique yielding higher spatial resolution is microbeam electron diffraction in the TEM using spot or transmission Kikuchi patterns. Thin foils, however, are difficult to prepare for TEM observation and, even worse, are sources of orientation errors due to wrinkling which is frequently manifested by bend contours.

The study of fine grain and heavily deformed materials in general, of nanomaterials, of recrystallization, grain growth and the characterization of grain boundaries demand for a substantially higher spatial resolution than is achieved with EBSD in the SEM.

Introduction to Ion Blocking Patterns

Diffraction patterns can be produced not only by electrons, but as well by ions of some ten keV kinetic energy when impinging on a crystalline surface. The first Ion Blocking Patterns (IBP) have been published by A.F. Tulinov in 1965 [1, 2]. Patterns have been indexed in this paper and in further papers published in the same year by A.F. Tulinov and co-workers, so providing evidence of the crystallographic nature of the patterns and of the mechanism of their generation by channeling the ions along low-index lattice planes and crystallographic directions. The patterns have been produced by proton beams of 150, 200 and 500 keV steered onto a single crystal of W respectively Mo at an angle of incidence of 30°. They have been recorded on special photographic plates, placed parallel with the surface of the crystal, to cover a large solid angle.

Backscatter Kikuchi Pattern from Cu at 20 keV.

Proton blocking pattern from W at 200 keV [1].
 The permission for reproduction is gratefully acknowledged to
Prof. A.F. Tulinov, Lomonosow State University Moscow,
and Uspekhi Fizicheskikh Nauk, Moscow.

IBP have, at a first glimpse, an appearance substantially different from Backscatter Kikuchi Patterns (BKP):
1. Instead of broad Kikuchi bands, they show narrow straight bands of high contrast, nearly black lines, on a more even and continuous background. Band profiles are described by almost symmetrical cusps.
2. The band width is given by the "critical channeling angle" rather than by the Bragg angle.
3. The band widths also depend on the atomic number of the target atoms of the crystal. This is not the case in electron diffraction.
4. No high-order lines are seen. (High-order Kikuchi lines are very useful in the precise determination of orientation and lattice parameters.)
5. Dark patches mark low-index zone axes.
6. Intensity is high also in backward scattering direction whereas BKP have a strong intensity maximum in forward scattering direction.

Kikuchi patterns are clearly generated by diffraction, whereas IBP could perhaps be better interpreted by the classical ballistic model of channeling. Precise experimental measurements are still missing so as to decide whether Bragg's equation is valid here. The background is produced by blocked ions according to Rutherford scattering. There is no significant difference in backscattering from an amorphous and from a crystalline solid for ions that impinge at directions off the critical angles. At kinetic energies in the range of several ten keV, their mean free paths are less than a few atomic layers. Only the channeled ions are missing in case of a crystal, they penetrate deep in the crystal and form sharp cusps of low intensity on the background. So the background intensity can be used as signal. This information about the crystal structure comes from the utmost atom layers, like Auger electrons. As a consequence, spatial resolution in IBP is only limited by the diameter of the primary beam probe. On the other hand, BKP are formed by electrons that are backscattered from the interaction volume that extends to a depth beneath the surface of the mean free path of energetic electrons. Hence spatial resolution in EBSD is limited to some ten nanometer, despite that the electron beam can easily be focused to a significantly smaller probe size.

The geometrical basis of IBP and BKP, however, is quite similar. The center line of a band corresponds to the section line of the - imaginarily extended - diffracting lattice plane with the recording plane, and the crossing points of bands, named poles, represent the positions of zone axes of the crystal. So the crystal structure and crystal orientation of the diffracting volume can be determined from the intensity distribution and positions of the bands in an IBP in quite a similar way as in a BKP. As an application, C.S. Barrett and co-workers [3] have used proton blocking patterns at 100 keV for phase differentiation. Crystallographic aspects of ion beam scattering have been discussed in a review article [4]. A commercial instrument for the identification and rapid alignment of single crystals was manufactured by Edwards High Vacuum Intern. [5].

[1] A.F. Tulinov: On an effect accompanying nuclear reactions in single crystals and its use in various physical investigations. Soviet Physics - Doklady 10 (1965) 463-465 (English translation of the original article of A.F. Tulinov, Doklady Akademii Nauk SSSR 162 (1965) 546-548
[2] A.F: Tulinov, V.S. Kulikauskas and M.M. Malov: Proton scattering from a tungsten single crystal. Physics Letters 18 (1965) 304-307
[3] C.S. Barrett, R.M. Mueller and W. White: Proton blocking patterns for hcp and wurtzite structures. Transactions MS AIME 245 (1969)427-429
C.S. Barrett: Line intensities in proton scattering. Transactions MS AIME 245 (1969) 429-430
[4] C.S. Barrett: Ion Beam Scattering applied to crystallography. Naturwissenschaften 57 (1970) 287-295
[5] R.G. Livesey: A 30 keV instrument for ion surface interactions studies. Vacuum 22 (1972) 595-597.