Carbon Alloys part 8 doc

Transmission Electron Microscopy 229 Fig. 7. (a) Debye rings, and (b) a network of diffraction spots. Fig. 8. Kikuchi pattern. 230 Chapter 14 Incident electronbeam 7 iline Fig. 9. Geometry of a lattice plane and the Kikuchi lines [5]. corresponds to the diffraction spot, and can be so indexed. Indeed, the mid-line of a Kikuchi pair of (hkl) plane corresponds to the intersection of the extension of the (hkl) plane with the film (Fig. 9). A slight tilting of the specimen, and hence of the (hkl) plane, results in a shift of the Kikuchi lines. Thus, the crystallographic orienta- tion of the specimen and deviations from the exact Bragg condition can be determined accurately. 2.3 Elemental Analysis X-Rays are generated when electrons interact with matter. With high energy incident electrons, inner shell electrons are ejected with characteristic energies and characteristic X-rays are emitted. The energy loss electrons and the characteristic X-rays are used to identify the elements which are interacting with the incident electrons. Figure 10 shows an example of elemental mapping by means of characteristic X-rays. A stacking fault (SF) lies end-on in a &-Si alloy, and the line analysis across the stacking fault indicates that copper is segregated onto the stacking fault [13]. This segregation of solute atoms onto the SF was predicted theoretically by Suzuki in 1956 and has been called the Suzuki effect. Elemental analysis using energy loss of the electrons (EELS) is discussed in detail in the following chapter. Transmission Electron Microscopy 23 1 5.0 4.5 E 4.0 - m 1 e 3.5 0 3.0 8 2.5 u 2.0 1.5 L C K 0 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 Distance (nm) Fig. 10. Elemental mapping by means of characteristic X-rays. The stacking fault lies end on in a Cu-Si alloy. 3 Specimen Preparation by FIB Preparation of thin foil specimens for TEM is a time-consuming and slow procedure. However, the technique of focused ion beam (FIB) milling is receiving attention as a new technique to prepare TEM specimens [&lo]. Figure 11 shows a schematic diagram of the FIB system. Gallium ions are emitted from a liquid gallium source. They are accelerated with a 30 kV high voltage and deflected to bombard the surface of the sample from which a thin foil specimen for TEM is to be prepared. Gallium ions sputter the sample and reduce thicknesses to those of thin foils. At the same time, secondary ions are emitted by the bombardment with gallium. The final shape of the specimen is shown in Fig. 12. Two trenches are produced, leaving behind a very thin wall which is electron transparent when tilted by 90". 232 Chapter 14 lens Beam monitor Beam deflector Objective lens -specimen Fig. 11. Schematic diagram of the FIB system. Mechanical polishing 15 Frn FIB fabrication FIB fabrication Fig. 12. The final shape of the specimen fabricated by a FIB system Transmission Electron Microscopy 233 There are at least two advantages in using FIB to prepare TEM specimens. One is that it is possible to prepare a TEM specimen with pin-point accuracy from a pre-selected area. The other is that a TEM specimen can be easily prepared from coated materials or multilayered structures composed of different substances with quite different physical and/or chemical properties. 3.1 Pin-pointing specimens One of the most successful applications of FIB to pin-pointing is the observation of a crack by TEM [ 14-17]. To carry out TEM observations of a crack, a thin foil specimen must be prepared. It can be extremely difficult to prepare such a specimen from a pre-selected region containing a crack tip. Figure 13 is an electron micrograph of a crack formed by indentation on the (001) surface of silicon. The following features are evident. Within the indent the crystal is deformed severely (P) and some dislocations (denoted by A) have glided out of the indent. Near the indent, cracks piercing the foil are observed. There are two types of cracks. One is almost parallel to the foil surface, i.e., (001) plane (L,, L,, L,, L,, L5) and has a tendency to lie on the (111) plane. The other type is perpendi- cular to the foil surface and lies on the { 110) plane (HJ. On both sides of the indent and underneath the indent, fringe contrasts can be observed (denoted by H,). The areas with fringe contrast can be considered to contain cracks. The fringes can be explained by moire fringes which result from the crystal being separated into two regions by the crack. L Fig. 13. Cross-sectional electron micrograph of a crack formed on a (001) surface of silicon. 234 Chapter 14 3.2. Diamond coating on SijV4[18] Diamond coating is used as a cutting tool. The performance of such a tool is governed mostly by the microstructure of the interface between diamond and the substrate. To characterize such an interface a cross-sectional TEM study would be informative. However, because diamond is very hard, it is very difficult to prepare a cross-section for a TEM examination using the traditional techniques. The FIB technique can be used to prepare such a cross-section. Figure 14 is an example of a cross-sectional TEM micrograph of microstructure near to the interface between diamond and Si,N,. The coating Si,N, was made using a chemical vapor deposition (CVD) method. The microstructures of diamond, of the Si,N, and of the interface between them are clearly seen. Fig. 14. Cross-sectional TEM micrograph of microstructure near an interface between diamond and Si,N,. Transmission Electron Microscopy 235 4 In-Situ Heating Experiment [19] 4.1 Heating Holder Figure 15 is a schematic diagram of a heating holder developed especially for use with a side-entry type goniometer of a transmission electron microscope. The heating element used is a fine wire (diameter 20-30 pm) of a refractory metal such as tungs- ten. The heating element is bridged across a gap of the holder and heated by direct current from dry batteries. The use of a dry battery as a power source is essential, because it provides a very stable current without any fluctuations which would affect the temperature of the sample and, consequently, the quality of a TEM image. The temperature of the heating element is calibrated either using an optical pyrometer outside the microscope or observing the melting of a pure metal such as gold or aluminum inside the microscope. Figure 16 gives an example of a calibration curve. Temperatures as high as 1800 K can be achieved with a current as small as 250 mA as the thermal mass of the heating element is small. Because of this small thermal mass, the drift of a specimen due to instability of temperature during heating is also small. This permits the recording of an image in the high-resolution electron micro- scope (HREM) on an ordinary film even when the specimen is at a high temperature. Fig. 15. Schematic diagram of the heating holder. Samples are directlymounted onto the heating element. 1,000 -1 Heating current (mA) Fig. 16. Temperature versus heating current curve. 236 Chapter 14 After mounting a mixture of fine particles of reactants, in this case graphite and Si, directly onto the heating element, excess reactant was blown away so that only attached particles were transferred into the electron microscope. 4.2 Results and Discussion Figure 17a shows a low magnification micrograph, taken at room temperature, of a silicon particle attached to a particle of graphite itself attached to the heating element (not shown in the figure). Figure 17b was taken 5 min after the temperature had reached 1673 K. It can be seen that particle S, which was originally a single crystal of silicon, has now fragmented into many small particles. At the same time, the area denoted by SC in Fig. 17b, which is the same area denoted by C in Fig. 3a, became darker. Figures 18a and b are diffraction patterns taken at room temperature and at 1673 K, respectively. Before heating (Fig. Ha), a typical diffraction pattern from a single crystal of silicon and typical Debye rings from graphite were observed. At 1673 K (Fig. 18b), extra Debye rings are seen in addition to those of the graphite. The lattice spacing of the extra Debye rings corresponded to cubic p-Sic. These changes in diffraction patterns during the in-situ heating experiment show definitely that at 1675 K silicon atoms diffuse into graphite and react with carbon to form Sic. Furthermore, examination of the Debye rings of Figs. 18a and b shows that the spacing of the c-planes of graphite did not change during the reaction. Figure 19 reproduces a sequence of HREM images showing the process of formation of Sic. In Fig. 19a a silicon particle is in the left bottom corner of the micrograph and has partially reacted with graphite to form Sic (indicated by arrow). However, most of the graphite remained unchanged. In Fig. 19c, which was taken 2 min after Fig. 19a, almost the whole region, of what had been graphite, had reacted with silicon to form Sic crystals. Figure 19b, taken 1 min after Fig. 19a, is an example of the stage between these two extremes showing a decrease in the contrast of lattice fringes of graphite. before heating 1400°C for 5 rnin (b) graphite Fig. 17. (a) TEM image of a single crystalline silicon particle (denoted by S) sitting on graphite (denoted by C) before heating. (b) The same location as (a) but heated for 5 min at 1673 K. Transmission Electron Microscopy 237 Fig. 18. Diffraction patterns taken (a) before heating, and (b) at 1673 K. Fig. 19. A sequence of reactions between solid silicon and graphite observed at near-atomic resolutions. 238 References Chapter 14 1. 2. 3. 4. 5. 6. 7. 8. 9. P.B. Hirsch, A. Howie, R.B. Nicholson, D.W. Pashley and M.J. Whelan, Electron Micros- copy of Thin Crystals. Butterworths, London, 1965. M.H. Loretto and R.E. Smallman, Defect Analysis in Electron Microscopy. Chapman and Hall, London, 1975. D.B. Williams and C.B Carter, Transmission Electron Microscopy. Plenum Press, New York & London, 1996. B. Howe and J.M. Howe, Transmission Electron Microscopy and Diffractiometry of Mate- rials. Springer, Berlin, 2001. H. Saka, Electron Microscopy of Crystals. Uchidarokakuho, Tokyo, 1997 (in Japanese) R.J. Young, E.C.G. Kirk, D.A. Williams and H. Ahmed, Fabrication of planar and cross-sectional TEN specimens using a focused ion beam. Mater. Res. SOC. Symp. Proc., K-H. Park, Cross-sectional TEM specimen preparation of semiconductor devices by fo- cused ion beam etching, Mater. Res. SOC. Symp. Proc., 199: 271-280,1990. T. Ishitani and T. Yaguchi, Microsco. Cross-sectional sample preparation by focused ion beam. Res. Tech., 35: 320-333,1996. H. Saka, Transmission electron microscopy observation of thin foil specimens prepared by means of a focused ion beam. J. Vac. Sci. Technol. By 16: 2522-2527,1998. 199: 205-215,1990. 10. T. Ishitani, Y. Taniguchi, S. Isakozawa, H. Kioke, T. Yaguchi, H. Matsumoto and T. Kamino, Proposals for exact-point transmission electron microscopy using focused ion beam specimen-preparation technique. J. Vac. Sci. Technol. B, 16: 2532-2537,1998. 11. D.J.H. Cockayne, I.L.F. Ray and M.J. Whelan, Investigations of dislocation strain fields us- ing weak beams. Phil. Mag., 20: 1265-1270,1969. 12. T. Katata and H. Saka, High-resolution electron-microscopic in situ observation of the mo- tion of intervariant interfaces in Cu-Zn-AI thermoelastic martensite. Phil. Mag., A 59: 677-687,1989. 13. T. Kamino, Y. Ueki, H. Hamajima, K. Sasaki, K. Kuroda and H. Saka, Direct evidence for Suzuki segregation obtained by high-resolution electron microscopy. Phil. Mag. Lett., 66: 27, 1992. 14. H. Saka and G. Nagaya, Plan-view transmission electron microscopy observation of a crack tip in silicon. Phil. Mag. Lett, 72 251-255, 1995. 15. H. Saka and S. Abe, FIB/HVEM observation of the configuration of cracks and the defect structure near the cracks in Si. J. Electron Microsc., 46: 45-57,1997. 16. Suprijadi and H. Saka, On the nature of a dislocation wake along a crack introduced in Si at the ductile-brittle transition temperature. Phil. Mag. Lett., 78: 435443,1998. 17. A. Muroga and H. Saka, Analysis of rolling contact fatigued microstructure using focused ion. Scripta Met., 33: 151, 1995. 18. K. Kuroda, M. Takahashi, H. Itoh, H. Saka and S. Tsuji, Application of focused ion beam milling to cross-sectional TEM specimen preparation of industrial materials including heterointerfaces. Thin Solid Films, 319 92,1998. 19. T. Kamino, T. Yaguchi and H. Saka, in situ study of chemical reaction between silicon and graphite at 1400°C in a high resolution/analytical microscope. J. Electron Microsc., 43: 104-110,1994. [...]... sp3Amorphous Materials In investigations of diamond-like carbon, one aim is to fabricate highly sp3-bonded amorphous carbon with properties comparable to those of crystalline diamond Thus, several highly $-bonded materials, having more than 80 % sp3,were obtained with, for instance, 92% sp3material [ 18] , 87 % sp3[9], 85 % sp3[19] and 83 % sp3 [20] For 87 % sp3[9] and 73% sp3[7] materials, the 1s+z* transition... c - - I - ’ 1 0 I 180 285 20 30 40 50 I I I 290 295 300 I 305 I 310 Energy Loss (eV) Energy Loss (eV) Fig 10 High-resolution EELS spectra of (a) plasmon region and (b) inner shell excitation region, k-edge of carbon, for C , fullerene, the present amorphous diamond, crystalline diamond For comparison, plasmon 8, spectra for 73% sp3 [7] and for 85 % sp3 [ ] and k-edge spectra for 87 % sp3 [9] and 73%... and structure of very hard carbon films produced by cathodic-arc deposition with substrate pulsed biasing Appl Phys Lett., 68: 779- 781 ,1996 7 V.I Merkulov,D.H.Lowndes,G.E.Jellison,A.A PuretzkyandD.B Geohegan, Structure and optical properties of amorphous diamond films prepared by ArF laser ablation as a function of carbon ion kinetic energy Appl Phys Lett., 73: 2591-2593,19 98 8 S.D Berger, D.R McKenzie... structure of diamond-like carbon Diamond Relat Mater., 6: 212-2 18, 1997 18 D.R Mckenzie, D Muller and B.A Pailthorpe, Compressive-stress-inducedformation of thin-film tetrahedral amorphous carbon Phys Rev Lett., 67: 773-776,1991 19 C Ronning, E Dreher, J.-U Thiele, P Oelhafen and H Hofsass, Electronic and atomic structure o undoped and doped-ta-films Diamond Relat Mater., 6: 83 0 -83 4, 1997 f 20 M Chhowalla,... amorphous carbon Diamond Relat Mater., 6: 207-211,1997 21 J Lee, R.W Collins, V.S Veerasamy and J Robertson, Analysis of the elipsometric spectra of amorphous carbon thin films for evaluation of the sp’-bonded carbon content Diamond Relat Mater., 7: 999-1009,19 98 22 M Weiler et al., Preparation and properties of highly tetrahedral hydrogenated amor- 256 Chapter I5 phous carbon Phys Rev B, 53: 159416 08, 1996... to carbon materials, a natural crystalline diamond was examined Representative high-resolution EELS spectra for C,, fullerene, amorphous diamond, and natural crystallinediamond in the plasmon region (0-50 eV) and in the inner-shell excitation region ( 280 -2 98 eV) are shown in Fig 10a(top) and Fig 10b(top), respectively For the C,, fullerene, the peaks around 6.5 eV and 285 eV 30.9 eV 73% sp3 ta-c 85 %s#... [lo] also found a (&x&)R30"honeycomb structure on VCo ,8( ll and considered that a carbon atom was missing at 1) the dark depression and that there was one carbon atom in each corner of the honeycomb skeleton Each dark depression in the honeycomb structure corresponds to a carbon atom in the Mo2C(0001)-(&x&)R30" structure, as explained below Each surface carbon atom located at the honeycomb hole is imaged... kinetic energy Appl Phys Lett., 73: 2591-2593,19 98 8 S.D Berger, D.R McKenzie and P.J Martin, EELS analysis of vacuum arc-deposited diamond-like films Phil Mag Lett., 57: 285 -290,1 988 9 S Ravi et al., Nanocrystallites in tetrahedral amorphous carbon films Appl Phys Lett., 69: 491493,1996 10 H Hirai, M Terauchi, M Tanaka and K Kondo, Band gap of essentially four-fold coordinated amorphous diamond Phys Rev... for some carbons [6,9, 18, 22,23].The density of the amorphous diamond 254 Chapter 15 diamond a & 4.0 Q 30 c c d f 2.0 : fi Y c a 10 0 0.0 - 35 ‘ “0 E diamond + b e 3.0 h A c r( * 9 n 0 25 A 20 ” ” * A A - A @ O* e A A ” ” ’ Fig 13 (a) Variation of optical (band) gap versus the sp3 fraction for amorphous carbon materials Solid rhombus: the present amorphousdiamond; open circle: Merkulov (19 98) [7];... to carbon monoxide, indicated that carbon monoxide adsorbs preferentially on a step edge which has exposed molybdenum atoms If the 1x 1) unit mesh of the exposed molybdenum atoms is extended to the ( f i x 3)R3Oo domain, dark depressions of the honeycomb correspond to three-fold hollow sites Thus, the carbon atoms are considered to occupy a three-fold hollow site of the (1x 1) molybdenum layer The carbon . Microscopy 23 1 5.0 4.5 E 4.0 - m 1 e 3.5 0 3.0 8 2.5 u 2.0 1.5 L C K 0 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 Distance (nm) Fig. 10. Elemental mapping by. Phil. Mag. Lett., 78: 435443,19 98. 17. A. Muroga and H. Saka, Analysis of rolling contact fatigued microstructure using focused ion. Scripta Met., 33: 151, 1995. 18. K. Kuroda, M. Takahashi,. of almost 100% sp'-bonded carbon close to crystalline diamond, as compared with other highly spj-bonded amorphous carbons. Keywords: EELS, Amorphous carbon, Amorphous diamond, Band