Tessa Cooper - University of Illinois at Chicago
Download
Report
Transcript Tessa Cooper - University of Illinois at Chicago
Tessa Cooper
Materials Science and Engineering
Rutgers University
Advisors:
Dr. R. Klie and Q. Qiao
Department of Physics, University of Illinois
UIC
Physics
Project description.
Methods to be used.
Results obtained for bulk SrTiO3.
Results obtained for SrTiO3/GaAs interface.
UIC
Physics
Characterize ultra-thin SrTiO3 film on GaAs
using Transmission Electron Microscopy
(TEM), Electron Energy Loss Spectroscopy
(EELS), and multiple scattering calculations.
Determine the effects of having interfacial O
vacancies and Ti diffusion in the substrate.
Evaluate potential uses of this material in
electrical and other applications.
UIC
Physics
Molecular Beam Epitaxy is used to deposit
monolayer films of SrTiO3 on GaAs.
SrTiO3 (4 ML)
SrTiO3 (4 ML)
Ti pre-layer (0.5 ML)
GaAs support
GaAs support
Sample 1
Sample 2
Ti pre-layer Deposition
Direct Deposition
(d)SrTiO3/GaAs (1)
Intensity (arb.units)
As 3d
SrTiO3/GaAs (2)
(c)
Ti/GaAs
(b)
bare GaAs
(a)
39
44
40
41
Energy ( eV)
42
43
R.F. Klie, Y. Zhu, Applied Physics Letters, 87, 143106 (2005).
UIC
Physics
Z-contrast image, SrTiO3
Z-contrast image, SrTiO3
Schematic drawing of interface:
2.0 nm
Ga
As
Sr
Ti
O
Highly distinct interfaces are formed, which do not
display differences in atomic structure whether or
not a prelayer is used.
R.F. Klie, Y. Zhu, Applied Physics Letters, 87, 143106 (2005).
UIC
Physics
GaAs
•Semiconducting
•Highly resistive
•High electron mobility
•Direct band gap
SrTiO3
•Dielectric constant of
300
•Mature deposition
method
•Good substrate for other
oxides.
GaAs on (110) plane
SrTiO3 on (100) plane
45°
UIC
Physics
The properties of this
system make it ideal for
transistors and other
electronic applications.
Ga
As
Sr
Ti
O
Prelayer
Correct orientation
Minimized defects
UIC
Physics
Use image simulations and multiple scattering
calculations to model the atomic and electric
structures, which helps to…
Interpret experimental results.
Support theories that are not obvious
through experimentation.
16000
14000
12000
10000
8000
6000
4000
2000
0
-2000 450
460
470
480
490
500
UIC
Physics
UIC
Physics
FEFF9
relies on Full Multiple Scattering
calculations to produce x-ray or electron behavior
in a material.
Other
methods are Fourier based calculations,
which require periodic structures.
UIC
Physics
O electrons are ejected from the K shell, closest to the nucleus.
Ti electrons are ejected from LII or LIII.
UIC
Physics
Used FEFF9 to produce O K and Ti L edges
in bulk SrTiO3.
Constructed GaAs/SrTiO3 interface to use
with the multiple scattering calculations.
Used FEFF9 to produce O K and Ti L edges
at the interface of SrTiO3.
With Oxygen vacancies
Without vacancies
UIC
Physics
Theoretical vs. Experimental Results for Ti L edge
45000
FEFF9 with peak broadening
40000
Counts (arb.)
35000
30000
25000
FEFF9 without peak broadening
20000
15000
Experimental EELS
10000
5000
0
455
460
465
470
475
480
Energy loss (eV)
UIC
Physics
Theoretical vs. Experimental results for O K edge
8.00E+03
FEFF9 with peak broadening
7.00E+03
Counts (arb.)
6.00E+03
5.00E+03
FEFF9 without peak broadening
4.00E+03
3.00E+03
Experimental EELS
2.00E+03
1.00E+03
0.00E+00
535
540
545
550
555
560
565
570
Energy loss (eV)
UIC
Physics
Ga
As
Ti
Sr
O
Targeted a Ti atom at the middle of the interface from which
to eject the electron, and removed O atoms around this atom.
UIC
Physics
Ti L Edge Simulation Comparison
36000
Bulk SrTiO3
31000
Counts (arb.)
26000
21000
SrTiO3/GaAs Interface with 4 O vacancies
16000
11000
SrTiO3/GaAs Interface
6000
1000
455
-4000
460
465
470
475
480
Energy Loss (eV)
UIC
Physics
Ga
As
Ti
Sr
O
Target a specific oxygen atom at the interface, and introduce
oxygen vacancies surrounding that atom.
UIC
Physics
Simulation for Bulk vs. Interface, O K Edge
Bulk SrTiO3
8.00E+03
7.00E+03
Counts (arb.)
6.00E+03
5.00E+03
4.00E+03
3.00E+03
SrTiO3/GaAs Interface
2.00E+03
1.00E+03
0.00E+00
530
535
540
545
550
555
560
565
570
575
580
Energy loss (eV)
UIC
Physics
Ejected O electron at interface, 1 vacancy
2.50E+00
1 vacancy, 6Å distance
2.00E+00
Counts (arb.)
1 vacancy, 4Å distance
1.50E+00
1 vacancy, 2Å distance
1.00E+00
No vacancies
5.00E-01
0.00E+00
530
535
540
545
550
555
560
Energy loss (eV)
UIC
Physics
Ejected O electron at interface, 2 vacancies
2.50E+00
2 vacancies, 6Å distance
2.00E+00
Counts (arb.)
2 vacancies, 4Å distance
1.50E+00
2 vacancies, 2Å distance
1.00E+00
No vacancies
5.00E-01
0.00E+00
530
535
540
545
550
555
560
Energy loss (eV)
UIC
Physics
Ejected O electron at interface, 3 vacancies
2.00E+00
1.80E+00
3 vacancies, 4Å distance
1.60E+00
Counts (arb.)
1.40E+00
1.20E+00
3 vacancies, 2Å distance
1.00E+00
8.00E-01
No vacancies
6.00E-01
4.00E-01
2.00E-01
0.00E+00
530
535
540
545
550
555
560
Energy loss (eV)
UIC
Physics
Ejected O electron at interface, 4 vacancies
1.40E+00
1.20E+00
4 vacancies, 2Å distance
Counts (arb.)
1.00E+00
8.00E-01
6.00E-01
No vacancies
4.00E-01
2.00E-01
0.00E+00
530
535
540
545
550
555
560
Energy loss (eV)
UIC
Physics
Ga
As
Ti
Sr
O
Targeted a specific oxygen atom at the center of the crystal
structure, and introduced oxygen vacancies surrounding
that atom.
UIC
Physics
Electron ejected from center of SrTiO3 cell
5.00E+00
4.50E+00
4 vacancies
4.00E+00
3 vacancies
Counts (arb.)
3.50E+00
3.00E+00
2 vacancies
2.50E+00
1 vacancy
2.00E+00
1.50E+00
1.00E+00
No vacancies
5.00E-01
0.00E+00
530
535
540
545
550
555
560
565
570
Energy loss (eV)
UIC
Physics
Bulk SrTiO3 spectra can be reliably
calculated for O K edge and Ti L edge.
Vacancy effect occurs in both Ti L edge
and O K edge.
Oxygen vacancies can be shown by using
FEFF9.
UIC
Physics
I would like to thank the following for making
this research project possible:
The National Science Foundation, EEC-NSF Grant
# 1062943 and CMMI-NSF Grant # 1134753.
Dr. Jursich and Dr. Takoudis
The University of Illinois at Chicago
UIC
Physics