Chapter 3: Experimental Methods
3.2 Structural characterization .1 X-ray diffraction
After Röntgen discovered X-ray in 1895, people finally obtained a way into the un-visible world, which could not be observed before, for instance, the matters in atomic scale.
X-ray is one kind of electromagnetic radiation with the wavelength in the range of 0.01–100 Å, and the basic element that can interact with X-ray is electron. When X-ray incidents into a matter and meets an atom including Z electrons, the electron will be excited due to the resonance with electromagnetic wave, and then irradiate another electromagnetic wave, called the exiting wave or scattered wave. The scattering amplitude is highly dependent on the electron density of the atom, which call atomic form factor
fat
rk
. For a crystalline material, the atoms are regularly arranged to construct the lattice. These atoms (or other basic component, e.g. molecules) are called basis. And then the crystal is in definition the53
combination of lattice and basis. Therefore, the generation of diffraction peak is decided by structure factor rules, which describes the sum of the scattering wave in terms of atoms and lattice:
When the scattering vector satisfies the Laue condition ( k g, g is the reciprocal lattice vector), diffraction will occur, which means that it also has to satisfy the Bragg’s law
2 sind
[7]. It is important to understand the meaning of reciprocal lattice because each reciprocal space point represents a corresponding diffraction peak that we can observe from the experiments.The X-ray structural analysis is the most common method to determine the crystallines of a material. For a powder sample or polycrystalline films, the typical diffraction feature is a ring pattern, which can be viewed as coalition of a huge of diffraction spots from the randomly distributed small crystal grain in space. For an epitaxial thin film grown on a single substrate, it also shows the diffraction spots feature and has broader width comparing to the single crystal, caused by some defects and mosaic structure. To understand the crystalline orientation of the thin films and epitaxy between films and substrate, some techniques are usually used, such as 2 scans, rocking curves, and scans. 2 scans, or called normal scans, are the scans always satisfy the Bragg’s law, which the angle from incident beam to sample surface
always keep half the angle from incident beam to detector
2 .And thus, the diffraction signal only provides the information of the diffracted planes parallel to the sample surface because there is only variation of out-of-plane components in reciprocal lattice vectors, for example such as the (001) oriented films, only the (00l) family planes in the sample can be observed in 2scans. The rocking curve usually means the scans of ranging around Bragg angle, which can describe the mosaic degree of the epitaxial films. The
scans, or the azimuthal scans, means the rotation of the sample itself. To get the scans, one should first find the so-call asymmetry reflections of substrate and films (the reciprocal lattice vector including the in-plane component), and then rotate sample stage. For a good expitaxy, the films should present as same numbers of peaks as substrate, which means they have the same symmetrical fold. The techniques mentioned above are all the line scans;
however, sometimes line scans cannot reveal the real shape of the diffraction patterns in
reciprocal space, especially those in the crystal with tilted and distorted structure. Therefore, the advanced technique called reciprocal space mapping (RSM) needed to resolve these non-regular structures. Reciprocal-space mapping measurement requires setting of the measuring range of 2 / scans, and setting of the measuring range of off-set angle , shown in Figure 3.2. For a double-axis diffractometer, the off-set angle is the difference of the θ from out-of-plane direction, and for a triple-axis, can be the varied by one additional freedom of degree, defined by, where is an angular circle perpendicular to incident beam. And then the RSM is constructed by a lot of 2 / scans with different offset angles.
Figure 3.2 a) The typical scheme of double-axis diffractometer (left), and the corresponding motion for each angular motor in real space (right). (Figure courtesy of Panalytical B.V) b) The high-resolution triple-axis diffractometer (Huber 8 circle diffractometer) at NSRRC Beamline 17B1 (left), and the scheme of the corresponding motion of each angular motor in
55
real space (right) [15].
X-ray diffraction characterization of the films in this dissertation was conducted in two kinds of instruments. One is the commercial machine, which is the PANalytical X’Pert Pro 4-circle x-ray diffractometer with Cu Kα1 radiation (1.54Å). The other is the Huber high resolution four-circle diffractometer using a synchrotron as source, where the energy is set as 10KeV (1.24Å); these RSM experiments were mostly performed at wiggler beamline BL13A, BL17A or BL17B1 at the National Synchrotron Radiation Research Center (NSRRC), Taiwan.
The incident X-rays were made monochromatic with a Si (111) double-crystal monochromator. With four pairs of slits between sample and detector, the typical scattering vector resolution in the vertical and horizontal scattering planes was set to ~10-4 nm-1 to dramatically decrease the noise single and raise the resolution during experiments.
3.2.2 Transmission electron microscopy
Transmission electron microscopy (TEM) was performed to characterize the microstructures of the films, such as morphological and crystallographic features. The operation of TEM is similar to the way of optical microscope, which also obeys the lens formula. The detailed setup of TEM instrument is shown in Figure 3.3 a). The electron beam is emitted from the filament accelerated by a high voltage (100kV - 1000kV) and then focused on the sample by electromagnetic lenses (condenser lenses). When the beams pass through the specimen, two kinds of modes can be observed. One is “image mode”, where all the transmitted and diffracted beams are focused by object lens to form the first magnified image;
the image can be viewed as “virtual object” and focused by the intermediate lens to magnify again. The final image is presented at the fluorescent viewing screen (Figure 3.3 b)). Because the electron beams behave similarly as light, it also produces diffraction patterns while interacting with specimen. Therefore, the second mode is “diffraction mode”. The change from image mode to diffraction mode can be performed by decreasing the electrical current in intermediate lens, and thus the focal length of intermediate lens will extend backward to the
“back focal plane”. At this time, the diffraction patterns become the virtual object and follow the artificial dashed line to form the magnified diffraction image on the screen. Based on these two modes, more advanced imaging and diffraction techniques are developed to get the detailed structural information, including the bright-field and dark-field image, selected area
electron diffraction (SAD); convergent-beam electron diffraction (CBED), phase-contrast imaging (high-resolution TEM, HRTEM); and Z-contrast imaging [8].
Figure 3.3 a) The detailed setup of a common TEM instrument. b) the concept of the image mode in TEM observation. c) The concept of the diffraction mode in TEM observation [8].
Since electrons scattering is strongly dependent on the atomic electronic potential of the sample, samples should be prepared in very thin region, which less than 1 μm for TEM characterization. In this dissertation, the focused ion beam (FIB, SII Nanotechnology SMI 3050SE) was used to fabricate the specimen for TEM cross-section observation. Before cutting the sample, SiO2 thin film of 100 nm had been deposited by E-gun system as a passivation layer, preventing the damage from the bombardment of Ar ion beam. Then Pt thin film around 20 nm capped onto the SiO2 layer for first observation by scanning electron microscopy (SEM). First, the Ar beam with the energy of 30 keV bombarded the surface of sample to localize the chosen area, and then the beam was tuned to lower energy of 5 keV to thin the chosen area finely until the thickness achieved around 50nm. Finally, the specimen was picked up to a Cu grid for following TEM observation. The microscopy that was used is the JEOL JEM-2010F operated at 200 V.
57
3.3 Techniques for measuring physical properties