UPS Applications-2

Louis Scudiero

http://www.wsu.edu/~scudiero; 5-2669

scudiero@wsu.edu

http://www.wsu.edu/~scudiero

MoS2 photoelectron energy distribution spectra intensity as a functionof "binding” energy and photon energy.

Photon energies: Ne (I) 16.8 eV, He (I)21.2 eV, Ne(II) 26.9 eV, He (II) 40.8eV, and He (IIβ) 48.4 eV.

Note that the relative intensity inpeaks changes in the distributions withphoton energy. This is due to cross-section σ changes for the “d” and “p”region of the VB.

How is this related to the electronic structure of the material, what is thesignificance of the intensity variations?

To answer these questions, let us again look at the 3-step model that was firstproposed by Berglund and Spicer (C. N. Berglund and W. E. Spicer, Phys. Rev. A136,1030, 1964).

Initial Block state with energy Ei(allowed state with a periodicitydetermined by the periodic potential ofthe crystal lattice). These electrons arethen excited (1) by the photon into aconduction band state at Ej (2). Theexcited electron finally escape from thesurface (matching at the vacuum interfacebetween the excited state and somefree electron state in vacuum). This statecan be represented by a single plan wavee ip.r≈ (3).

ik

jk

jk

'jk

The contribution to the photocurrent I (hν,ε) with some initial stateenergy Ei = ε (analyzer set to ε) can be expressed as follows:

Where T (k) is some function expressing the probability of transportand escape (steps 2 and 3) and where the dipole matrix element

expresses the optical excitation (step 1). A is the vector potential, ∇gradient and the δ functions express the fact that Ei and Ej areseparated by hν and the second that the energy analyzer is “tuned “ topick out the initial state Ei.Illustration of several importance effects:

1. Density of states 2. Matrix elements 3. Transport effects

( ) ) )dkE(εhEδ(EkAkT(k),hIiij

2

fi−−−∇•∞∫ δνεν

2

ijkAk ∇•

Theoretical calculations versus photoemission spectrum

Let us assume that we are slowly varying transport functions and matrix elements sothat it can be given by:

The two δ functions define 2 intersecting surfaces in reciprocal space and if theintegral is transformed to a line integral around the line of intersection thephotocurrent is expressed as:

The integral diverges when both gradient terms go to zero (band becomes essentially“flat”). But (1/∇(E)) dE is the density of states in an energy interval dE.

So, if we neglect the singularities, the photoemission measures the overall density ofstates in the valence band of the solid and can be compared directly with calculatedband diagram.

Density of States

) )dkE(εhEδ(EIiij

−−−∞∫ δν

∫ ∇∇∞

)()(dlI

ikjkEE

Very good agreement betweentheory and experiment in the “p”region. No contribution of the “s”states is seen in the He II data. Thisis due to a very small value of thedipole matrix element.

No “s” contribution because

02≈∇• ij kAk

Dipole Matrix ElementsPhotoionization Cross-section

The absence of the ‘s’ state in the SnSe2 He (II) spectrum can beexplained by looking at the matrix element

At a given photon energy hν and for a fixed final state

the matrix element will vary depending on whether is s, p, d orf like.

In practice, the ‘s’ cross-section is extremely low above about 10 eVuntil we reach soft x-ray energies. Therefore only p, d and felectrons will be detected in the energy range of 0 - 50 eV (UPS).

2

ijkAk ∇•

fk

ik

Calculated cross-section

for certain rare-gas shells

So, back to molybdenum disulfide, the VB statesare formed from Mo 4d and S 3p electrons. Theband near the EF is purely ‘d’ like and theremaining structure comes from covalent bondingband based on the S 3p electrons. Some mixingp/d character is also expected.

Spectra between 5 and 40 eVBased on the plot cross-section vs. PE energyshown in the previous slide the 3p states dominatesbelow 25 eV and the 4d states above 25 eV.Bands 2 and 4 are purely S 3p and finally bands 3and 5, expected to have a mixed character p/d.

Atomic character of the states causeintensity changes

54

3

21

Energy loss introduced by electron-electron interactions (scattering andsecondary emission).If these electrons escape thespectrum will display peaks that willmove as the photon energy ischanged.

Structure which ‘moves’ betweenNeI and HeI spectra is assigned tosecondary electron emission(identified by bars)

Transport Effects

Band Theory and PE Spectrum(electron energy distributions)

He II

“s”

2s 1s forbidden but 2p 1s allowed

“p”

XPS

Because the C – C bond distance ingraphite ~ 1.4 Å strong interatomicinteractions > than in diamond. The VBelectron distribution is importanttheoretically and experimentally.

UPS spectrum is closer to the DOS dataP. M. Williams, Handbook of x-ray and ultraviolet photoelectron spectroscopy, edited by D. Briggs, Heyden&Son Ltd, 1977.

C. Bandis, L. Scudiero et al, Surface Science, 442 (1999) 413-419

Dispersion MeasurementsFar more useful is the direct E ---> k

(dispersion) measurement obtained

by angular photoemission studies.

By measuring the energy of the peak position with respect to theangle θ and using the expression k// α E sinθ. The dispersionrelation graph that plots the binding energy-wave vector dependencefor graphite can be obtained.

The data obtained here (BE vsk// ) for the first Brillouin zoneagrees well with the calculatedbands of Painter and Ellis (G.S.Painter and D. E. Ellis, Phys. Rev.B1, 4747, 1970) shown here.

Adsorption on solidsAdsorption of a monolayer of CO onpolycrystalline Pt and Ni.From this kind of data the followingschematic of molecular orbitals could beobtained:

5σ1π 4σ5σ, 1π and 4σ

contribute to the PE spectrum

Polymer (P3HT) thermally deposited on clean Ag

18 16 14 12 10 8 6 4 2 0

2 1 0

2 1 0

Inte

nsity

(abi

trar

y un

its)

Binding Energy (eV)

P3HT thickness Ag foil 0.25 nm 0.32 nm 0.44 nm 0.75 nm 1.0 nm 1.9 nm 3.1 nm

0.56 eVEF

Clean Ag

P3HT 0.44 nmHOMO cutoff

0.8 eV

4 3 2 1 0

0.55 eV

Δ = 0.56 eV (interface dipole moment) and HOMO (VBM) = 0.55 eV

L. Scudiero & al. applied Surface Sciences 255 (2009) 8593-8597

Hybrid Nanocomposites for Electronics Applications(P3HT-Ag nanoparticles)

4.5 4 3.5 3 2.5 2 1.5 1 0.5 0Binding Energy (eV)

18 16 14 12 10 8 6 4 2 0Binding Energy (eV)

P3HT

Clean Ag foil

Δ

Δ3

P3HT

(c)

Δ2

Δ1

3:1

1:1

1:3

3:1

1:1

1:3

Ag

(b)(a)

Blend film 3:1Blend film 1:1Blend film 1:3

0.55 eVHOB edge

Fermi‐level

Δ3(3:1)

ΦBlend film

(3.77 eV)

0.79 eV

Δ2(1:1)

ΦBlend film

(3.47 eV)

1.36 eV

Δ1(1:3)

ΦBlend film

(3.37 eV)

ΦP3HT film

(3.71 eV)

P3HT film

0.60 eV

0.56 eV

EF

Ag

Evac

ΦAg

(4.27 eV)

It is possible to tune the value of the barrier height ( ) from 0.55 to 1.36 eV by simply varying the composition of the blend film (P3HT/Ag ratio ranging from 3:1 to 1:3) and changing the electronic properties of nanocomposite materials.

L. Scudiero & al. ACS Applied Materials & Interfaces

Combining UV-Vis measurements and/or computational chemistry to measure or calculate band gap of materials and UPS measurements of the HOMO (VB energy) a energy diagram can be easily drawn.

Example: C60(CN)2 deposited on semiconductor by spin coating for photovoltaic applications.UV-Vis measurements give a band gap energy = 2.18 eV, Calculation (TD-DFT) gives 1.99 eV (isolated molecule, energy difference between HOMO-LUMO).UPS measurements yield HOMO (valence band max) position at 1.1 eV.

~ 1.1 eV

~ 1.82 eV

2nd Edition E. H. Rhoderick and R. H. Williams Oxford Science Publications

Metal- Semiconductor Contacts

Electron states in solids and at surfaces

Density of states for n-type Si (a)and band structure as a function ofdistance (b).

The important surface parameter forsemi-conductor is the electronaffinity χs defined as the differencein energy between an electron at restoutside the surface and an electron atthe bottom of the conduction bandjust inside the surface

I is the minimum energy needed to remove an electron from the VB.gs

EI += χ

Band bending near the surfacecould be cause by either surfacestates on a free surface or ametal overlayer.

Typical PE spectrum

for n-type SC

(Theoretical variation of cross-section).

Likelihood of ionization of anelectron from a given orbital in anatom.

Photoionization Cross-section of C

1. Elemental dependency of the cross-section on photon energy for C, N and O.

2. Difference in cross-section for different orbitals for Phosphorus.

References

• Handbook of x-ray and ultraviolet photoelectron spectroscopy, edited by D. Briggs, Heyden&Son Ltd, 1977.

• P.M. Williams same Handbook. p.313 – 340

• M. Thompson, P.A. Hewitt and D. S. Wooliscroft same Handbook, p341 -379

• B. L. Sharma, Plenum Press, New York

UPS Applications-2���Louis Scudiero��http://www.wsu.edu/~scudiero; 5-2669�� scudiero@wsu.eduSlide Number 2Slide Number 3Slide Number 4Slide Number 5Slide Number 6Slide Number 7Slide Number 8Slide Number 9Slide Number 10Slide Number 11Slide Number 12Slide Number 13Slide Number 14Slide Number 15Slide Number 16Slide Number 17Slide Number 18Slide Number 19Slide Number 20Slide Number 21Slide Number 22Slide Number 23Slide Number 24Slide Number 25Slide Number 26Slide Number 27