Analytical Transmissions Electron Microscopy Part I TEM Basic
Analytical Transmissions Electron Microscopy Part I: (TEM) Basic principles Operational modes Diffraction Part II: Imaging Sample preparation Part III Spectroscopy A. E. Gunnæs Additional reading about TEM: http: //www. matter. org. uk/tem/default. htm MENA 3100 V 13
TEM is based on three possible set of techniqes Diffraction Imaging From regions down to a few nm (CBED). With spatial resolution down to the atomic level (HREM and STEM) 200 nm Spectroscopy Chemistry and elecronic states (EDS and EELS). Spatial and energy resolution down to the atomic level and ~0. 1 e. V.
Introduction EM and materials The interesting objects for TEM is not the average structure or homogenous materials but local structure and inhomogeneities Defects Interfaces Precipitates
The first electron microscope Ernst Ruska: Nobel Prize in physics 1986 • • • Knoll and Ruska, first TEM in 1931 Idea and first images published in 1932 By 1933 they had produced a TEM with two magnetic lenses which gave 12 000 times magnification. Electron Microscope Deutsches Museum, 1933 model
Electron lenses Any axially symmetrical electric or magnetic field have the properties of an ideal lens for paraxial rays of charged particles. • Electrostatic F= -e. E – Not used as imaging lenses, but are used in modern monochromators • Magnetic F= -e(v x B) – Can be made more accurately – Shorter focal length http: //www. matter. org. uk/tem/lenses/electromagnetic_lenses. htm
Basic TEM Electron source (HV= 200 k. V) Apertures Magnetic lenses Sample holder Vacuum in the column better than 10 -6 Pa Fluorescence screen Recording media (Film/CCD/ TV) Pedals for tilting the sample A. E. Gunnæs MENA 3100 V 13
Instrumentation Filament Similar components as a transmission light microscope Anode The diffraction limit on resolution is given by the Raleigh criterion: 1. and 2. condenser lenses δd=0. 61λ/μsinα, μ=1, sinα~ α Sample Objective lens Intermediate lenses Projector lens A. E. Gunnæs MENA 3100 V 13 Compared to the lenses in an optical microscope they are very poor! The point resolution in a TEM is limited by the aberrations of the lenses. -Spherical - Chromatic -Astigmatism
Spherical aberrations • Cs corrected TEMs are now available Spherical aberration coefficient ds = 0. 5 MCsα 3 M: magnification Cs : Spherical aberration coefficient α: angular aperture/ angular deviation from optical axis r 2 α r 1 2000 FX: Cs= 2. 3 mm 2010 F: Cs= 0. 5 nm Disk of least confusion The diffraction and the spherical aberration limits on resolution have an opposite dependence on the angular aperture of the objective. A. E. Gunnæs MENA 3100 V 13
Resolution limit Year 1940 s 1950 s 1960 s 1970 s 1980 s 1990 s 2000 s Resolution ~10 nm ~0. 5 -2 nm 0. 3 nm (transmission) ~15 -20 nm (scanning) 0. 2 nm (transmission) 7 nm (standard scanning) 0. 15 nm (transmission) 5 nm (scanning at 1 k. V) 0. 1 nm (transmission) 3 nm (scanning at 1 k. V) <0. 1 nm (Cs correctors) http: //www. sfc. fr/Material/hrst. mit. edu/hrs/materials/public/Elec. Micr. htm A. E. Gunnæs MENA 3100 V 13
Effect of Cs correction Before Cs correction After Cs correction Core of the M 100 galaxy seen through Hubble (source: NASA)
Illumination system JEOL 2000 FX Mini-lens screws Specimen Intermediate lens shifting screws Projector lens shifting screws Wehnelt cylinder Filament Anode Electron gun 1. and 2. beam deflectors 1. and 2. condenser lens Condenser aperture Condenser lens stigmator coils Condenser lens 1. and 2. beam deflector Condenser mini-lens Objective lens pole piece Objective aperture Objective lens pole piece Objective lens stigmators 1. Image shift coils Objective mini-lens coils (low mag) 2. Image shift coils 1. , 2. and 3. Intermediate lens Projector lens beam deflectors Projector lens Screen
The electron source • Two types of emission guns: – Thermionic emission • W or La. B 6 – Field emission • W Cold FEG Zr. O/W Schottky FEG
Thermionic guns Filament heated to give thermionic emission -Directly (W) or indirectly (La. B 6) Filament negative potential to ground Wehnelt produces a small negative bias -Brings electrons to cross over
Field emission gun • The principle: – The strength of an electric field E is considerably increased at sharp points. E=V/r • r. W < 0. 1 µm, V=1 k. V → E = 1010 V/m – Lowers the work-function barrier so that electrons can tunnel out of the tungsten. • Surface has to be pristine (no contamination or oxide) – Ultra high vacuum condition (Cold FEG) or poorer vacuum if tip is heated (”thermal” FE; Zr. O surface tratments → Schottky emitters).
Characteristics of principal electron sources at 200 k. V W La. B 6 FEG Schottky (Zr. O/W) FEG cold (W) Electron source size (µm) 50 10 0. 1 -1 0. 010 -0. 100 Emission current (µA) 100 20~100 Brightness B (A/m 2 sr) 5*109 5*1010 5*1012 Energy spread ΔE (e. V) 2. 3 1. 5 0. 6~0. 8 0. 3~0. 7 Vacuum pressure (Pa)* 10 -3 10 -5 10 -7 10 -8 Vacuum temperature (K) 2800 1800 300 * Might be one order lower
W Advantages: La. B 6 advantages: FEG advantages: Rugged and easy to handle High brightness Extremely high brightness Requires only moderat vacuum High total beam current Long life time, more than 1000 h. Good long time stability Long life time (500 -1000 h) High total beam current W disadvantages: La. B 6 disadvantages: FEG disadvantages: Low brightness Fragile and delicate to handle Very fragile Limited life time (100 h) Requires better vacuum Current instabilities Long time instabilities Ultra high vacuum to remain stable
Electron interaction with the specimen e- Backscattered electrons Auger electrons Cathodoluminescence Secondary electrons X-rays Absorbed electrons EBIC Specimen Elastically scattered electrons Inelastically scattered electrons Transmitted electrons Gas Heating Cooling
TEM specimens TEM grids The interesting objects for EM is not the average structure or homogenous materials but local structure and inhomogeneities on n me eci iti s o p Cooling Sp Standard Heating 3 mm A. E. Gunnæs MENA 3100 V 13
Operating modes Convergent beam Parallel beam Can be scanned (STEM mode) Specimen Spectroscopy and mapping (EDS and EELS) Imaging mode or Diffraction mode A. E. Gunnæs MENA 3100 V 13
Simplified ray diagram Parallel incoming electron beam 3, 8 Å Si Sample 1, 1 nm Objective lense Diffraction plane (back focal plane) Image plane A. E. Gunnæs MENA 3100 V 13
Image or diffraction mode Filament Anode 1. and 2. condenser lenses Spesimen Objective lens Bi-prism Objective aperture Selected area aperture Diffraction plane Image plane Intermediate lenses Projector lens Viewing screen A. E. Gunnæs STEM detectors (BF and HAADF) Image or diffraction pattern MENA 3100 V 13
Simplified ray diagram Parallel incoming electron beam 3, 8 Å Si Sample 1, 1 nm Objective lense Diffraction plane (back focal plane) Image plane A. E. Gunnæs MENA 3100 V 13
Electron diffraction Inelastic scattered electrons Elastic scattered electrons Direction and magnitude of v change. Only the direction of v is changing. (Bragg scattering) Elastic scattering is due to Coulomb interaction between the incident electrons and the electric charge of the electron clouds and the nucleus. (Rutherford scattering). The elastic scattering is due to the average position of the atoms in the lattice. Energy is transferred to electrons and atoms in the sample. -It is due to the movements of the atoms around their average position in the lattice. - It give rise to a diffuse background in the diffraction patterns. Reflections satisfying Braggs law: 2 dsinθ=nλ Electrons interacts 100 -1000 times stronger with matter than X-rays -more absorption (need thin samples) -can detect weak reflections not observed with x-rays A. E. Gunnæs MENA 3100 V 13
Selected area diffraction Parallel incoming electron beam Specimen with two crystals (red and blue) Objective lense Diffraction pattern Selected area aperture Image plane Pattern on the screen
Selected area electron diffraction • Parallel incoming electron beam and a selection aperture in the image plane. • Diffraction from a single crystal in a polycrystalline sample if the aperture is small enough/crystal large enough. • Orientation relationships between grains or different phases can be determined. • ~2% accuracy of lattice parameters – Convergent electron beam better Image plane
Kikuchi pattern Incoherently and inelastically (ΔE~15 -25 e. V) scattered electrons give rise to diffuse background in the ED pattern. -Angular distribution of inelastic scattered electrons falls of rapidly with angle. I=Iocos 2α Kikuchi lines are due to: -Diffusely + Bragg scattering event Deficient θB Excess θB 2θB Objective lens Deficient line Excess line Diffraction plane Excess line Deficient line 1/d http: //www. doitpoms. ac. uk/index. html http: //www. doitpoms. ac. uk/tlplib/diffraction-patterns/kikuchi. php
Used for determination of: -crystal orientation Kikuchi pattern -g -lattice parameter 000 g Kossel cones Ig=I-g -accelerating voltage -Burgers vector Sg<0 Sg=0 g and –g Kikuchi lines Effect of tilting the specimen Parabolas http: //www. umsl. edu/~fraundorfp/nanowrld/live 3 Dmodels/vmapframe. htm
Diffraction with large SAD aperture, ring and spot patterns Poly crystalline sample Four epitaxial phases Similar to XRD from polycrystalline samples. The orientation relationship between the phases can be determined with ED.
Why do we observe many reflections in one diffraction pattern? A. E. Gunnæs MENA 3100 V 13
The Ewald Sphere is flat (almost) Cu Kalpha X-ray: = 150 pm => small k Electrons at 200 k. V: = 2. 5 pm => large k
Zone axis and Laue zones Zone axis [uvw] (hkl) uh+vk+wl= 0
ED and form effects Real space Resiprocal space
Higher order reflections, Laue zones 2 d sinθ = nλ The intensity distribution around each reciprocal lattice point is spread out in the form of spikes directed normal to the specimen λ 200 k. V = 0. 00251 nm Θ~1 o I(k’-k)I=(2/λ)sinθB=g From one set of planes we only get one reflected beam -The Bragg angle increases with increasing order (n) -Tilt sample or beam to satisfy Bragg condition of higher order reflections. First order Laue zone Zero order Laue zone Ewald sphere (Reflecting sphere) k 2θ k’ k=1/λ g
Indexing diffraction patterns The g vector to a reflection is normal to the corresponding (h k l) plane and Ig. I=1/dnh nk nl (h 2 k 2 l 2) - Measure Ri and the angles between the reflections - Calculate di , i=1, 2, 3 - Compare with tabulated/theoretical calculated d-values of possible phases - Compare Ri/Rj with tabulated values for cubic structure. - g 1, hkl+ g 2, hkl=g 3, hkl (vector sum must be ok) - Perpendicular vectors: gi ● gj = 0 - Zone axis: gi x gj =[HKL]z Orientations of corresponding planes in the real space (=K/Ri) All indexed g must satisfy: g ● [HKL]z=0
Camera constant R=L tan 2θB ~ 2 LsinθB 2 dsinθB =λ ↓ R=Lλ/d Camera constant: K=λL Film plate
Example: Study of unknown phase in a Bi. Fe. O 3 thin film Metal organic compound on Pt Bi. Fe. O 3 Heat treatment at 350 o. C (10 min) to remove organic parts. Pt Ti. O 2 Lim Si. O 2 Si 200 nm Goal: Bi. Fe. O 3 with space grupe: R 3 C and celle dimentions: a= 5. 588 Å c=13. 867 Å Process repeated three times before final heat treatment at 500 -700 o. C (20 min). (intermetallic phase grown)
Determination of the Bravais-lattice of an unknown crystalline phase Tilting series around common axis 27 o 15 o 50 nm 10 o 0 o
Determination of the Bravais-lattice of an unknown crystalline phase Tilting series around a dens row of reflections in the reciprocal space 0 o 50 nm 19 o Positions of the reflections in the reciprocal space 25 o 40 o 52 o
Bravais-lattice and cell parameters 011 111 001 c 101 b 010 a 110 [011] [100] [101] d=Lλ/R 100 6. 04 Å From the tilt series we find that the unknown phase has a primitive orthorhombic Bravias-lattice with cell parameters: a= 6, 04 Å, b= 7. 94 Å og c=8. 66 Å 7. 94 Å 6 8. 6 Å α= β= γ= 90 o
Chemical analysis by use of EDS and EELS Ukjent fase Bi. Fe 2 O 5 Bi. Fe. O 3 O-K Fe - L 2, 3 Bi. Fe. O 3 500 e. V forskyvning, 1 e. V pr. kanal Ukjent fase
Published structure A. G. Tutov og V. N. Markin The x-ray structural analysis of the antiferromagnetic Bi 2 Fe 4 O 9 and the isotypical combinations Bi 2 Ga 4 O 9 and Bi 2 Al 4 O 9 Izvestiya Akademii Nauk SSSR, Neorganicheskie Materialy (1970), 6, 2014 -2017. Romgruppe: Pbam nr. 55, Bi Fe Fe O O 4 g 4 h 4 f 4 g 8 i 4 h 2 b celleparametre: 7, 94 Å, 8, 44 Å, 6. 01Å x 0, 176 0, 349 0 0, 14 0, 385 0, 133 0 y 0, 175 0, 333 0, 5 0, 435 0, 207 0, 427 0 z 0 0, 5 0, 244 0 0, 242 0, 5 Celle parameters found with electron diffraction (a= 6, 04 Å, b= 7. 94 Å and c=8. 66 Å) fits reasonably well with the previously published data for the Bi 2 Fe 4 O 9 phase. The disagreement in the c-axis may be due to the fact that we have been studying a thin film grown on a crystalline substrate and is not a bulk sample. The conditions for reflections from the space group Pbam is in agreement with observations done with electron diffraction. Conclusion: The unknown phase has been identified as Bi 2 Fe 4 O 9 with space group Pbam with cell parameters a= 6, 04 Å, b= 7. 94 Å and c=8. 66 Å.
- Slides: 41