Universidad de Cádiz Determina nati tion o n of Bi positi tions ns in GaA aAs (1-x) Bi Bi x heterost stru ructure res w s with at atomi mic co column r res esolution David d L. S . Sal ales es (david id.sa .sale les@uc s@uca.e .es) s) E. Guerre rrero ro , , J. J. F. . Rodrigo, A. A. Yá Yáñez, P. P. L. G . Galindo, S. I S. I. . Molina M. . Henini, M M. . Sh Shafi, S.V S.V. N . Novikov M.F. Ch Chis isholm lm
Introduction Framework I. Introduction • Materials Science & Engineering Group • Framework • Computational Intelligence systems • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF • Image processing • Results V. Image simulation STEM Group (Dr. Pennycook’s group) Dr. Henini’s Group 7/17/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 2 | 30
Introduction Transmission electron microscopes I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation JEOL 2010 JEOL 2011 JEOL 1200 II. Methodology & FEG LaB 6 EX Materials • Growth • HAADF • Image Aberration-corrected processing • Results V. Image simulation NION VG-HB603 VG-HB501 UltraSTEM 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 3 | 30
Introduction Previous TEM work in GaAsBi • Molecular beam epitaxy of GaAsBi I. Introduction • Framework on (311)B GaAs substrates • Tools • Previous GaAsBi works • Motivation 1 µm 1 µm II. Methodology & (b) (c) Materials GaAsBi GaAsBi • Growth • HAADF • Image processing • Results GaAs GaAs Z-contrast V. Image simulation (001) (311)B M. Henini et al. Appl. Phys. Lett. 91, 251909 2007 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 4 | 30
Introduction Previous TEM work in GaAsBi J. F. Rodrigo et al. Applied Surface Science 256 (2010) 5688–5690 I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF • Image processing • Results V. Image simulation 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 5 | 30
Introduction Other nano-tools • Focussed Ion Beam I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation FEI DUAL BEAM FEI QUANTA 200 3D II. Methodology & Materials - Nano-machinning • Growth -3D sample preparation for tomopraphy • HAADF of localized areas. • Image processing - Substrates nano-patterning • Results - As ions imaging V. Image simulation -3D tomography 7/17/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 6 | 30
17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 7 | 67
17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 8 | 67
Introduction The Motivation I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF Bi There are • Image Nanoclusters So… ¿can processing some would explain we see • Results experimental PL them? V. Image simulation evidences enhancement 7/17/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 9 | 30
Methodology and materials I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF • Image processing • Results V. Image simulation 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 10 | 30
Methodology & Materials Experimental techniques I. Introduction • Framework A • Tools • Previous GaAsBi works • Motivation A B II. Methodology & Materials B • Growth B • HAADF • Image processing C C • Results V. Image simulation B 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 11 | 30
Methodology & Materials The sample I. Introduction • Framework • Tools p Bi = 1.2·10 -7 Torr • Previous [001] GaBi x As (1-x) p As = 8.0·10 -6 Torr GaAsBi works 1 µm • Motivation T ≈ 350ºC II. Methodology & Materials • Growth • HAADF GaAs Substrate • Image processing semi-insulating (100) • Results V. Image simulation Region of near stochiometric growth x ≈ 0.03 HRXRD Henini et al. APL 91 , 251909 (2007) 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 12 | 30
Methodology & Materials Why HAADF? I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF • Image processing • Results V. Image simulation 17/07/2010 PhD Defence David Sales Lérida | 13 | 67
Methodology & Materials Why HAADF? • For a ternary alloy: I. Introduction • Framework – Linear relationship • Tools • Previous Intensity quotient (R) vs. Composition. GaAsBi works • Motivation II. Methodology & = + Materials R e ( x ) 1.005 0 . 229 x • Growth • HAADF • Image processing • Results V. Image simulation Column-by-column compositional mapping by Z-contrast imaging S. I. Molina et al. Ultramicroscopy 109 (2009) 172–176 7/17/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 14 | 30
Methodology & Materials Quantitative Compositional analysis HAADF EELS I. Introduction • Framework • Tools S27 • Previous 0 0,2 0,4 0,6 0,8 1,0 S25 S25 GaAsBi works InP P molar fraction, (1-x) • Motivation 2 nm 2 nm InP II. Methodology & Materials • Growth 1.0 S27 S27 InP • HAADF 0.8 • Image 0.6 processing Contenido en P 0.4 • Results 0.2 V. Image simulation 0.0 -0.2 InP 2 nm 2 nm 0 2 4 6 8 10 12 14 Monocapa 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 15 | 30
Methodology & Materials STEM - HAADF I. Introduction • Framework • Tools • Previous GaAsBi works vacuum • Motivation GaAsBi II. Methodology & Materials • Growth • HAADF • Image GaAs processing • Results V. Image simulation 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 16 | 30
Ga As/Bi
1. Localize intensity maxima (As/Bi columns) 2. Localize Ga columns 3. Select integration area 4. Determine average integrated intensity in every dumbbell: I Ga and I As/Bi I Ga &I As/Bi
Methodology & Materials Image processing • Determining R factors: I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & I As/Bi Materials • Growth = R(x) • HAADF • Image processing I Ga • Results V. Image simulation Minimize variations due to: • Same local thickness • Same amorphous layer • Same experimental image conditions 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 19 | 30
I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF • Image processing • Results V. Image simulation 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 20 | 67
Results Plotting R I. Introduction • Framework • Tools • Previous GaAsBi works • Motivation II. Methodology & Materials • Growth • HAADF • Image processing • Results V. Image simulation 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 21 | 30
Results Analysis • In order to relate R with x (Bi content): I. Introduction • Framework • Tools N N ∑ ∑ • Previous = + GaAsBi works R N a x • Motivation i i = = 1 1 i i II. Methodology & Materials N, the number of atomic columns, • Growth x i = Bi percentage per column, • HAADF Σ x i =2.65% total Bi percentage • Image processing • Results • Fitting equation: V. Image simulation = + R 1 . 0629 0 . 0729 x 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 22 | 30
Results Results I. Introduction • Framework 120 • Tools • Previous 4 100 GaAsBi works • Motivation Columns 80 3 II. Methodology & 60 Materials 2 40 • Growth • HAADF 20 • Image processing 1 0 • Results 0 1 2 3 4 V. Image simulation 0 Bi atoms per column 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 23 | 30
Image Simulations The next step… I. Introduction Solving the Schrödinger • Framework stationary equation • Tools • Previous GaAsBi works 2 • Motivation ∆ Ψ + − Ψ = ˆ [ ] 0 E V t 2 m II. Methodology & 0 Materials • Growth by FFT multislice method • HAADF • Image (Ishizuka’s code) processing • Results STEM image V. Image simulation simulations 17/07/2010 Workshop on Bi-Containing Semiconductors | David L. Sales | 24 | 30
The SICSTEM software A Paralell HAADF-STEM Simulation Sw Workshop on Bi-Containing Semiconductors | David L. Sales | 25 | 30
Cádiz University supercomputer I. Introduction • Framework • Tools • Previous GaAsBi works • Hewlett-Packard (2007) • Motivation – 320 Xeon Woodcrest II. Methodology & cores running at 3GHz Materials • Growth – 3.75 Tflops (position 327 • HAADF in Top500 last year) • Image processing – Each node 8 or 16 Gb • Results RAM V. Image simulation – Total RAM = 700 GB – 2.5 TB disk capacity
Recommend
More recommend