workshop: inelastic x-ray scattering for nsls ii march 15 ...€¦ · ⇒indication of ring like...

Recent Advances in Hard X-ray Inelastic Scattering with Medium Resolution

(A Story about Water)

Workshop: Inelastic X-ray Scattering for NSLS II March 15, 2004, Upton, NY

• motivation• x-ray Raman spectroscopy (XRS)• x-ray emission spectroscopy (XES)• selective x-ray absorption (S-XAS)• resonant inelastic x-ray scattering (RIXS)

Uwe Bergmann

SSRL, Stanford Linear Accelerator Center

Anders Nilsson SSRL/Stockholm

Philippe Wernet BESSY

Lars Pettersson Stockholm

Jean-Louis Hazemann Grenoble

Pieter Glatzel UtrechtFrank de Groot

Steve Cramer LBNL/Davis

Vittal Yachandra Calvin Lab, LBNLJunko Yano

Beamline Staff SSRL 10-2, NSLS X-25, APS 18ID (BioCAT)

Collaborators

Photon-in Photon-out X-ray Spectroscopy

x-ray beamfrom synchrotron

technique experimental procedure

x-ray Raman scattering (non resonant) scan of monochromator, analyzer fixedselective x-ray absorption

x-ray emission (non resonant) scan of analyzer, monochromator fixed

resonant inelastic x-ray scattering scan of both

monochromator

sampledetector

tunable analyzer

∆E ~ 1 eV

∆E ~ 1 eV

hydrogen bonding

propertiesstructure

water“essential for life”

“organisms consist mostly of water”“most abundant substance on earth”

“only naturally occurring inorganic liquid”“third most common molecule in the universe”

…

The Local Structure of Water

criticalpoint

experimental techniques:- neutron and x-ray diffraction- infrared/fs spectroscopy- collective excitations (dynamics)

- XAS → local structure

soft

x-ray

s

Photosynthetic Oxygen Evolution

CO2 + H2O → (CH2O)n + O2ћω

photosynthesis

this process generates carbohydrates and the world supply of oxygen

4 ћω

photosystem II cytochrome oxidase

how? life cycle

oxygen evolution

2 H2O → O2 + 4 H+ + 4 e-

the oxygen is derived from water

aerobic metabolismATP + CO2 + H2O ← (CH2O)n + O2

we consume oxygen to “burn” the energy of carbohydrates to produce ATP, the biological energy currency

- OEC is a 4 Mn cluster- four states S0 – S3- EPR, K-edge, EXAFS,

Kβ, crystallography- L-edge not possible

catalyticcenter

Why water?



x-ray Raman scattering (non resonant) scan of scan of monochromatormonochromator, analyzer fixed, analyzer fixed

selective x-ray absorption



monochromator

sampledetector

tunable analyzer

∆E ~ 1 eV

∆E ~ 1 eV

Principle of X-ray Raman Spectroscopy

E0

∆E = E0 - E’

E’

sample

e-

photonphoton

6400 6500 6600 6700 6800 6900 7000 7100

elastic peak

E0 [eV]

Comptonpeak

Raman scattering

∆EIn

tens

ity

[l og

sca l

e ]

scattering probability: w ~ cos2θ sin2(θ/2) |<i|r|f>|2

for qr << 1, and |ki| ≅ |kf|

graphite

θ

why XRS?

sample specific:- true bulk probe- no vacuum requirements- in situ experiments - high temperature/pressure

technique specific:- no saturation effects- non dipole transitions at large q- pump probe experiments (LCLS)

samples

in general:- any sample with sufficient

scattering strength

in particular:- systems not suited for studies

with conventional techniques- concentrated low Z systems- liquids- reactive specimens

Motivation

Local Structure of Water

water (RT)

surface ice

ice90o CRT

data calculations

• identification of large percentage of molecules with one

‘broken’ donor bond (SD)

⇒ indication of ring like configurations

• weak temperature dependence

hydrogen bonds per moleculeTemperature 25oC 90oC MD ~3.5 ~ 3.3 experiment ~2.1 ~ 2

classical MD uses pair potentials ⇒ possibly incomplete description of hydrogen bondsneeded: ab initio MD simulations with > 500 molecules

What the data suggest

Wernet et al, Science accepted



x-ray Raman scattering (non resonant) scan of scan of monochromatormonochromator, analyzer fixed, analyzer fixed

selective x-ray absorption



monochromator

sampledetector

tunable analyzer

∆E ~ 1 eV

∆E ~ 1 eV

why XES?

- direct probe of unpaired spins in 3d transition metals⇒ oxidation state, spin state

- sensitivity to ligand type and distance- ‘orientation’ of disordered sample

future: pump probe, single shot experiments (LCLS)

samples (5-12 keV)

- any sample (very dilute sample possible) - 3d transition metals (K-emission)- rare earths (L-emission)

Motivation

Mn(II)OMn(IV)O2

KMn(VII)O4

Relative Energy [eV] 350

Inte

nsit

y [a

.u.]

Kα1

Kβ''

Kβ'

Kβ2,5

Kβ1,3

Kα2

10-5

10-4

10-3

10-2

1.5 1.7 1.9 2.1 2.3

Rela

tive

Int

egra

ted

Inte

nsit

y

Mn-O Distance [A]o

-25 -20 -15 -10 -5 0 5

K2MnF

6Mn salen NitridoKMnO

4

Energy Shift [eV]

Nor

mal

ized

Int

ensi

ty [a

.u.]

FO N

chemical sensitivity of K fluorescence

X-ray Emission Spectroscopy

valencelevels

3p

2p

1s

Kα1.2

Kβ1,3, Kβ’

Kβ2,5, Kβ’’

level diagram

Bergmann et al, Chem Phys Lett 302, 1999

mono-µ-oxomanganese di-µ-oxo

manganese

di-µ-oxo (IV-IV)-(III-IV)

XANES difference spectra

Kβ XES difference spectra

XANES spectra

(III-III)(III-IV)(IV-IV)

(III-IV)(IV-IV)

mono-µ-oxo(III-IV)-(III-III)

mono-µ-oxo(IV-IV)-(III-IV)

Visser et al, JACS 123 7031, 2001

Comparison: XANES versus Kβ1,3 XES

Kβ1,3 XES on Photosystem IIdifference spectra

Messinger et al, JACS 123, 7804, 2001



x-ray Raman scattering (non resonant)

selective x-ray absorption scan of scan of monochromatormonochromator, analyzer fixed, analyzer fixed



monochromator

sampledetector

tunable analyzer

∆E ~ 1 eV

∆E ~ 1 eV

Motivation

why selective selective EXAFS?

- chemical in addition to elemental specificity- extended k-range ⇒ higher resolution

samples

- any sample- e.g. 3d metals, rare earths

Extending the Resolution of EXAFS from 0.15 to 0.10 Å

S1 : 2 to 3 Mn-Mn at 2.72 - 2.84 Å

S2 : 2 to 3 Mn-Mn at 2.71 - 2.77 Å

PS II S1 State

Selective X-ray Absorption Spectroscopy



x-ray Raman scattering (non resonant)

selective x-ray absorption scan of scan of monochromatormonochromator, analyzer fixed, analyzer fixed



monochromator

sampledetector

tunable analyzer

∆E ~ 1 eV

∆E ~ 1 eV

Principle of RIXS (1s,2p)

Tota

l Ene

rgy

hν = ωhν = Ω

Ω − ω

ground state

intermediate states

final states

K-edge

1s 3dexcitation

2p 1sdecay

1s23dn

1s3dn+1

2p53dn+1

state diagram

48 49 50 51 52 538

9

10

11

12

13

Incident Energy [eV]

Ener

gy T

rans

fer [

eV]

ΓK

ΓL

ener

gy t

rans

fer

Ω −

ω

excitation energy Ω

excitation e

nergy Ω[eV]

energy transfer Ω-ω [eV]

RIXS plane

Mn(II)

-isolate LUMO resonances (3d) → orbital population, local symmetry

- L-edge like information → sensitivity to spin

- less lifetime broadening- less radiation damage than L-edge

Why RIXS?

Incident Energy [eV]8328 8330 8332 8334 8336

(d)

example 2: high-spin Ni(III)

Glatzel et al, JACS comm, 124, 9668, 2002

RIXS plane

CET versus K-edge XANES CIE versus L-edge XANES

8326 8328 8330 8332 8334 8336 8338

848

850

852

854

856

858

860 hs-Ni(II)

8330 8332 8334 8336 8338 8340 8342

852

854

856

858

860

862

864

ls-Ni(II)

example 1: low spin/high-spin Ni(II)

E: MnII(acac)2(H2O)2 F: MnIII(acac)3 G: MnIV(sal)2(bipy)

Glatzel et al, JACS, submitted

RIXS Spectra of Mn models and PSII

1st moment positions

• XRS: low Z systems, non dipole DOS, EXAFS possible

• XES: spin state, chemical state, neighbor distance and type

• S-XAS: extended k-range, chemical/spin specificity

• RIXS: isolate LUMO resonancesL-edge/M-edge like information with hard x-rays

Conclusions

Future

• better resolution, more efficient spectrometers• lots of photons

planned ‘Advanced Spectroscopy and Inelastic Scattering Facility’at SPEAR3, first workshop see:http://www-ssrl.slac.stanford.edu/conferences/ssrl30/IXS_report.pdf

End of Show

Comparison of Mn(II), Mn(III) and Mn(IV) 1s,2p RIXS

workshop: inelastic x-ray scattering for nsls ii march 15 ...€¦ · ⇒indication of ring like...

Documents