X-ray absorption spectroscopy - ETH Z...Lambert Beer’s law dI = -μ(E)I dx I = I 0 exp(-μ(E)x)...
Transcript of X-ray absorption spectroscopy - ETH Z...Lambert Beer’s law dI = -μ(E)I dx I = I 0 exp(-μ(E)x)...
X-ray absorption spectroscopy
Jagdeep Singh
Jeroen A. van Bokhoven
Historical development
Synchrotron needed
Absorption through matter
10000 12000 140000.90
0.91
0.92
0.93
0.94
0.95
Tran
smis
sion
Energy (eV)
L3
L2
L1 Au Foil
Absorption as function of energy of the x-rayShape is structure dependent
Creation of a photo-electron
Ekin = hν - EBinding
Photo-electron has kinetic energy
EBindinghν
Photo-electron has kinetic energy
XANES2
f ii f E EP T νδ − −Ψ Ψ h
Initial state Final state
Transition operator
Fermi’s Golden Rule
Outgoing wave === backscattering === interference patternConstructive / destructive interference
„wig
glin
g in
abs
orpt
ion“
Ψfinal = Ψoutgoing + Ψback scattering
EXAFS is the wiggling part of the absorption
2d sin θ = nλ
Tuning the energyDouble crystal monochromator
Bragg angle
sample
I0 IX-rayswhite beam Crystal [Si(111)]
SLS, Villigen
Experimental Hutch
BM26 (DUBBLE), ESRF Grenoble
Lambert Beer’s lawdI = -μ(E)I dxI = I0 exp(-μ(E)x)
Transmission through 1 cm
0 2000 4000 6000 8000 10000 12000 140000.0
0.2
0.4
0.6
0.8
1.0
Tran
smis
sion
Energy (eV)
10-3 mbar
Air @ 1 bar
2000 4000 6000 8000 1000012000140000.0
0.2
0.4
0.6
0.8
1.0
Tran
smis
sion
Energy (eV)
25 μm 1 wt% Au/SiO2
hνI0
I = I0e-μx
10000 12000 140000.90
0.91
0.92
0.93
0.94
0.95
Tran
smis
sion
Energy (eV)
L3
L2L1
X-ray absorption through matter
Sample environment
pressuretemperatureenvironment
Absorption of X-rays is limiting factor
Find a good window material• Size of window• Thickness• Inertness• Temperature resistance• Pressure• Safety
0 2000 4000 6000 8000 10000 12000 140000.0
0.2
0.4
0.6
0.8
1.0
Tran
smis
sion
Energy (eV)
Transmission through25 μm and 1 mm
Al
Mylar
Be
Si K Cr K Cu K Pt L3Ag L3
In situ EXAFS cells for gas-solid reactions
Reaction gas mixture flows
around a pellet
Reaction gas flows through a catalyst
pellet
Small Glass Reactor with very thin
windows (0.01mm)
Large dead volumeGood for stationary
conditions
Critical d/l (smaller effectivity of the catalyst)
Small dead volumeOptimal d/lGood for structural changes Structure-activity relations
1 x 8 mm 1 x 8 mm
Reactor
Fluorescence detector Thermo
Couple
Exit-tube to mass spec
ID26, ESRF
XANES
Dipole transition:K edge: 1s pL edge: 2s p
2p s,d
HoweverQuadrupole transition:K edge s d
p f
Quad. Trans. probability is about 10-3 smaller, but d-DOS >> p-DOSVisible in the K pre-edges!!
Δl=±1
Δl=±2
2
f ii f E EP T νδ − −Ψ Ψ h
Initial state Final state
Transition operator
Fermi’s Golden Rule
What determines the shape of XANES spectra?- Pre-edge- Edge-energy- Shape over the edge
1.2
0.8
0.4
0.0
Nor
mal
ised
abs
orpt
ion
(μx)
552054805440X-ray energy (eV)
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
552054805440
1.2
0.8
0.4
0.0
V2O5
V6O13
VO2
NH4VO3
VOSO4.3H2O
K edge (1s 4p)of VOx compounds
Pre-edge - intensity- energy
…… there is not always a pre-edge
BUT…..
1.2
0.8
0.4
0.0
Nor
mal
ised
abs
orpt
ion
(μx)
552054805440X-ray energy (eV)
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
552054805440
1.2
0.8
0.4
0.0
V2O5
V6O13
VO2
NH4VO3
VOSO4.3H2O
VOSOVO2V6O13V2O5NH4VO
In order of increasing distortion from octahedral
4 Distorted OctahedralDistorted OctahedralDistorted OctahedralDistorted Square Pyramidal
3 Tetrahedral
Increasingpre-edge
Increasingdistortion
1.0
0.8
0.6
0.4
0.2
0.0
Nor
mal
ised
abso
rptio
n (
x)
548554805475547054655460X-ray energy (eV)
Arctangent / polynomial
548554805475547054655460X-ray energy (eV)
Isolation of the pre edgeby edge subtraction
Pre-edges intensity & energy varies (K edge)
Pure octahedral caseCentro-symmetry: no p-d mixing allowed: only quadrupole transition
very low intensity
Distortion from octahedralP-d mixing allowed: dipole transition in pre-edge + quadrupolar trans.
increasingly large intensity
Pure tetrahedral => largest pre-edge
Intensity pre-edge indicative of geometry
1.2
0.8
0.4
0.0
552054805440
NH4VO3
Vn+
P-DOS
EedgeEpre
(Pre-)edge Energy and ValenceEdge position is measure of oxidation state
1S
2p 3/2
2p 1/2
2s
1s
Ef
hν > Eo
photo-electronEkin = hν - E0
EmptyDOS
1.2
0.8
0.4
0.0
Nor
mal
ised
abs
orpt
ion
(μx)
552054805440X-ray energy (eV)
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
552054805440
1.2
0.8
0.4
0.0
V2O5
V6O13
VO2
NH4VO3
VOSO4.3H2O
For L edges > 3 keV and all K edges Shape of the whiteline
“Whiteline is first intensepeak(s) in spectrum”
VOSO4 Distorted Octahedral
VO2 Distorted Octahedral
V6O13 Distorted Octahedral
V2O5 Distorted Square Pyramidal
NH4VO3 Tetrahedral
E (eV) - EedgeE (eV) - Eedge
-60 -30 0 30 60 900.0
0.2
0.4
0.6
0.8
1.0
1.2
Nor
mal
ized
abs
orba
nce
bulk gold
-60 -30 0 30 60 900.0
0.2
0.4
0.6
0.8
1.0
1.2
Nor
mal
ized
abs
orba
nce
Au2O3
Au L3 edge
Whiteline reflects holes in d-band
Very small whiteline
Shape of the whiteline: L-edges
Totally different
11900 11925 11950 119750.0
0.2
0.4
0.6
0.8
1.0
1.2
70 °C : Au3+
200 °C : Au0
In situ Au3+ reduction in He/H2
Isobestic points
11875 11900 11925 11950 11975 120000.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Au(0)Au(III)
Nor
mal
ized
Abs
orpt
ion
Energy (eV)
Reference spectra
HAuCl4 on Al2O3
Shape of the whiteline: L-edgesPure metals:
Whiteline reflects holes in d-band
Mo0.5
Ru0.7
Rh0.8
Pd0.9
Ag1.0
Tc0.6
Ideal d-band filling
Alloying:
Whiteline reflects charge transfer
Abstract (I)Pre-edge Edge Shape over the edge- Valence - Valence - Geometry- Geometry - D-band filling (LIII-edge)
- (Non- / Anti-bonding) DOS states(- Adsorbates)
For many (many!) compounds structures and spectra are available in literature
NoteVariations in XANES may be very subtle and hardly visible in the data:
take (negative) second derivative
L edgesWhiteline intensity reflects number of holes in the d band (valence)
K edges(Pre) edge position reflects valence
Shape of XANES indicative of geometry
Typical XANES Experiment
• Catalyst samples, measured in desired conditions temperature, pressure, aggregation state
• Reference samples that likely resemble the state of the catalyst- Various oxidation states- Various coordinations
• Identification of trends, similarities in reference samples
• Comparison of trends, similarities to ‘unknowns’
• Application of theory to obtain ultimate information (expert option).