Principle of X-ray Photoemission Spectroscopy (XPS), and its application

Faramarz S. Gard Department of Physics Sultan Qaboos University

History of Photoelectric effect and Electron Spectroscopy

1880: Hertz noticed that metal contacts in electrical systems, when exposed to light, exhibit an enhanced ability to spark. 1888: Hallwachs observed that negatively charged zinc plate lost its charge when exposed to UV light, but positively charged zinc plates were not affected. 1899: J.J. Thompson found that electrons were emitted from the zinc plate exposed to light. 1900: Planck introduced the concept of quantization of energy. He received the Nobel prize in 1918. 1905: Einstein explained correctly all these phenomena, using Plancks quantization of energy concept. He received the Nobel prize in 1921.

History of Photoelectric effect and Electron Spectroscopy

1914:Robinson and Rawlinson studied photoemission from X-ray irradiated gold and, photographic detection, observed the energy distribution of electron produced. 1951:Steinhardt and Sterfass first applied photoemission as a analytical tool. 1950s-60s: Kai Siegbahn (son of Manne Siegbahn (1924 Nobel Prize) developed the instrumentation and theory of Electron Spectroscopy for Chemical Analysis (ESCA), which we know today. He changed to Electron Spectroscopy for Chemical Application.

Photoelectric Effect
The electrons were emitted immediately-no time delay. The larger intensity of the incident light the larger number of photoelectrons, but not their maximum energy. Light with wavelength longer than red light will not cause the ejection of electrons, no matter what the intensity!!! A weak violet light will eject only a few electrons, but their maximum kinetic energies are greater that those for intense light of longer wavelengths.

Photoelectric Effect is energy dependent

eV = Emax,kin = h h V = e e

e h

h e

Electron Spectroscopy
Emitted Auger Electron
Free Electron Level Fermi Level 1) Electrons in, Electrons out: AES (Auger Electron Spectroscopy) involves excitation of a core electron and detection of an electron that results from the decay of the corehole. EELS (Electron Energy Loss Spectroscopy) kinetic energies of electrons are detected near the energy of the incident beam of electrons. 2) Electrons in, Photons out: 1s K APS (Appearance Potential Spectroscopy) measures the threshold energy for incident electrons to ionize a core level by detecting the onset of soft Xray emission. IPES (Inverse Photoemission Spectroscopy) involves detection of photons that are emitted when an external electrons fills an empty energy level.

Conduction Band Valence Band 2p 2s

L2,L3 L1

Electron Spectroscopy
Ejected Photoelectron
Free Electron Level Fermi Level

Incident X-ray

Conduction Band Valence Band 2p 2s 1s

3) Photons in, Electrons out: XPS (X-ray Photoemission Spectroscopy) involves emission of a core electron after absorbing a photon. UPS (Ultraviolet Photoelectron Spectroscopy) involves emission of a valance electron after absorbing a photon. AES (Auger Electron Spectroscopy) Core excitation by a photon, (a By-product of XPS)

L2,L3 L1 K

Principle of Electron Spectroscopy

Principle of Photolectron Spectroscopy: Surface is placed in a UHV environment. Irradiated with photons (X-ray, UV). Atoms at the surface emit electrons (photoelectrons) either from a core level or valance band. Electrons are counted then separated according to their energy. Energy of photoelectrons is related to atomic and molecular environment. Number of electrons emitted is proportional to the concentration of the emitting atom on the surface.

What do we learn from X-ray Photoemission Spectroscopy (XPS)?

In the outermost 10 nm of a surface, can Provide: Identification of all elements (except H and He) present at concentrations >1 - 0.1 atomic %. Semi-quantitative determination of the elemental surface composition within experimental error of 10 - 15 %. Information about the molecular environment (oxidation state, bonding atoms, etc.) Non-destructive elemental depth profiles 10 nm into the sample. Destructive elemental depth profiles several hundred nanometers into the sample using ion etching. Lateral variations in surface composition (spatial resolution of 8-100 m, instrument related).

Instrumentation, UHV

Instrumentation, UHV

Multi-element system:

electrostatic

lens

(i) Collects e's of large angular distribution - larger flux. (ii) Focuses e's at entrance slit. (ii) Retards electrons to pass energy. (iv) Can "magnify" image of sample for small spot XPS - much easier to look at small spot with analyzer than try to produce focussed x-ray beam.

Instrumentation, UHV

Pressure Degree of Vacuum Torr

102 Low Vacuum 10-1
Medium Vacuum

10-4 10-8 10-11

High Vacuum Ultra-High Vacuum

Why UHV for Surface Analysis? Remove adsorbed gases from the sample. Eliminate adsorption of contaminants on the sample. Prevent arcing and high voltage breakdown. Increase the mean free path for electrons, ions and photons.

At 300 K, if every N2 molecule that strikes this surface remains adsorbed, a complete monolayer is formed in about t=3 ns (atm). If p=10-3 torr (1.3x10-6 atm), t = 3x10-3 s If p=10-6 torr (1.3x10-9 atm), t = 3 s If p=10-9 torr (1.3x10-12 atm), t = 3000 s or 50 minutes

Requirement for Experiment in Vacuum: Clean surface quickly becomes contaminated through molecular collision so p must be less than about 1.3 x 10-10 atm (10-9 torr).

Instrumentation, X-ray source

Photon Sources
Photons are originated either from standard lab sources or from synchrotron radiation sources:

Laboratory sources:
Helium gas discharge lamps produce photons in UV region He1: 21.2 eV, He2: 40.8 eV, which is used for UPS. X-ray is produced by bombarding a metallic target (usually Al: 1486 eV, or Mg: 1254) with high-energy electrons, which is used for XPS. If electron falls from the 2p3/2 to the 1s then the X-ray is designated K1. If it is produced by transition from the 2P1/2 to the 1s, it is designated K2. The intensity of the K1 line is twice that of the K2 line, because 2p3/2 is populated twice that of the 2P1/2 shell. These unmonochromated sources have line widths of 0.8-0.7 eV.

Synchrotron radiation:
For radiation in the UV/soft X-ray, energy region 10-2000 eV. These sources in addition to XPS opened up many new possibilities. Properties of synchrotron radiation: High brilliance, Tunable Photon Energy, Polarized Radiation

Instrumentation, X-ray source

Instrumentation, X-ray source

Instrumentation, X-ray source

Unmonochromated X-ray source X-ray source is always unmonochromated. In addition to the main emission lines K1,2, there are also other lines associated with doubly and multiply ionized atoms all labeled K, while K emission lines associated with valence to 1s transitions. They are superimposed on broad, continuum background (Bremsstrahlng) up to the incident electron energy. XPS spectrum is dominated by K1,2, the doubly ionized K3,4 emission will give rise to photoelectron satellites of about 8% of the main intensities at kinetic energies 10 eV higher. Other lines are generally <1% of the main lines. The existence of satellite X-ray lines will complicate the photoemission spectra and could lead to some overlap with main line emission from minor species. The intrinsic line width of the dominant X-ray line places a minimum line width on photoelectron spectral features.

Instrumentation, X-ray source

Monochromatic X-ray source The monochromatic principle is based upon the Bragg diffraction of K radiation by the (10-10) planes of quartz. The crystal is set on a Rowland circle of radius R with the crystal bent to a radius 2R in the dispersive direction and to 2Rsin2 in the nondispersive direction (Johann geometry). In all cases the Bragg angle is large [ = 78.5 for Al K on quartz (10-10)]. This produces an energy width of radiation of 0.26 eV at a photon energy of 1486.6 eV (Al K) and eliminates the Bremsstrahlng as well as the satellites and other unwanted lines.

Instrumentation, X-ray source

Instrumentation, Electron Energy Analyzer

Instrumentation, Electron Energy Analyzer

Electron Energy Analyzer Most common type of electrostatic deflection-type analyzer called the concentric hemispherical analyzer (CHA) or spherical sector analyzer. Negative potential on two hemispheres V2 > V1 Potential of mean path through analyzer is

V0 =

V1 R1 + V2 R2 2 R0

An electron of kinetic energy KE = eV0 will travel a circular orbit through hemispheres at radius R0 Since R0, R1 and R2 are fixed, in principle changing V1 and V2 will allow scanning of electron KE following mean path through hemispheres. Total resolution of instrument is convolution of x-ray source width, natural line-width of peak, analyzer resolution 2 2 2 2

E m = E x + E n + E a

Primary Structure in XPS

Photoemission process often envisaged as three steps (i) Absorption and ionization (initial state effects) (ii) Response of atom and creation of photoelectron (final state effects) (iii) Transport of electron to surface and escape (extrinsic losses)

Primary Structure in XPS

Strategic Plan for an Unknown Sample If the surface composition is unknown: A broad survey scan- to identify elements present. Swap anode (if available) in order to identify XPS/AES peaks). Once the elemental composition has been determined, narrow scans of selected peaks for a more comprehensive picture of chemical composition is acquired. This procedure is also used if composition of the sample is known priori to analysis.

Primary Structure in XPS

Chemical Shift One of the most important features of XPS is that the core level energy depend on the chemical state of the atom. Changes in the local charge and potential of an atom causes shifts in core level binding energies. Chemical shift correlated with overall charge on atom (Reduced charge then increased Binding Energy) (i) Number and type of surrounding atoms (ii) Electronegativity of atoms (iii)Oxidation state, depends upon ionicity /covalency of bonding

Functional Group hydrocarbon amine alcohol, ether Cl bound to C F bound to C carbonyl

C-H, C-C C-N C-O-H, C-O-C C-Cl C-F C=O

Binding Energy (eV) 285.0 286.0 286.5 286.5 287.8 288.0

Primary Structure in XPS

Chemical Shift Usually chemical shifts are thought of as initial state effect. Example: Ti 2p1/2 and 2p3/2 chemical shift for Ti and Ti4+. Charge withdrawn from Ti4+ comparing to Ti, so 2p orbital relaxes to higher Binding Energy. Note: Spin-orbit splitting is approximately constant - confirming SOS is largely an initial state effect Chemical shift information very powerful tool for functional group, chemical environment, oxidation state

Primary Structure in XPS

Effect of surface charging: Electrical insulators cannot dissipate charge generated by photoemission process. Surface picks up excess positive charge - all peaks shift to higher BE. Can be reduced by exposing surface to neutralizing flux of low energy electrons - "flood gun" or "neutralizer BUT must have good reference peak

Primary Structure in XPS Electronic effects Spin-Orbit coupling

Spin-orbit splitting is an initial state effect, for any electron in orbital with orbital angular momentum, coupling between magnetic fields of spin (s) and angular momentum (l) occurs. But how many spin-orbit split levels at each j value? Degeneracy = 2j + 1

Observations: - s orbitals are not spin-orbit split - singlet in XPS - p, d, f orbitals are spin-orbit split - doublets in XPS - BE of lower j value in doublet is higher (BE 2p1/2 > BE 2p3/2) - Magnitude of spin-orbit splitting increases with Z - Magnitude of spin-orbit splitting decreases with distance from nucleus (increased nuclear shielding)

Primary Structure in XPS

C 1s

Cu 2p

2p3/2

Orbital=s l=0 s=+/-1/2 ls=1/2

2p1/2

Orbital=p l=1 s=+/-1/2 ls=1/2,3/2

19.8 : 2 925

Peak Area 290 288 284 280 Binding Energy (eV) 276 965

955 945 935 Binding Energy (eV)

Ag 3d 3d3/2

3d5/2

Au 4f

4f7/2 4f5/2

6.0 Peak Area 2 378 : 3 362

Orbital=d l=2 s=+/-1/2 ls=3/2,5/2

Peak Area 91

Orbital=f l=3 s=+/-1/2 ls=5/2,7/2

3.65 : 4 79

374 370 366 Binding Energy (eV)

87 83 Binding Energy (eV)

Quantification
Quantification of XPS Most analyses use empirical calibration constants (called atomic sensitivity factors) derived from standards:

I measured Ca = ASF

ASF for H, He very small - undetectable in conventional XPS! XPS spectrum will show all peaks for each element in same ratio. All XPS peaks for an element same intensity (in area ratio proportional to ASF's) -choose peak with largest ASF to maximize sensitivity. Sensitivity for each element in a complex mixture will vary.

Quantification

Concentration Measurements Accuracy better than 15 % using ASF's. Use of standards measured on same instrument accuracy better than 5 %. In both cases, reproducibility (precision) better than 2 %.

Surface sensitivity of electron spectroscopy

Inelastic Mean Free path and sampling X-ray can penetrate easily through, while electrons exhibit less ability to do so. X-rays of 1 KeV will penetrate 1000 nm, or more into matter while electrons of this energy will only penetrate only 10 nm. Definitions for electron transport in materials: IMPF (), Inelastic Mean Free Path: The average distance that an electron with a certain energy travels between successive inelastic collisions. ED, Escape Depth: The distance normal to the surface at which the probability of an electron escaping without significant energy loss due to inelastic processes drops to e-1 (38%) of its original value. AT, Attenuation Length: The average distance that an electron with a given energy travels between successive inelastic collisions, elastic electron scattering is considered. SD, Sampling Depth (3) : Information Depth (ID) where percentage of detected electrons is 95%.

Surface sensitivity of electron spectroscopy

Electron Attenuation Length
Electron loose its energy through solid in inelastic processes such as: 1. Excitation of plasmons (conduction electrons). 2. Excitations of valance electrons. 3. Ionization of core levels. Sampling Depth = Attenuation Length Inelastic mean-free-pass > attenuation Length Because elastic scattering effect is neglected. The difference is about 30%. Electron attenuation length as function of kinetic energy for a variety of elements is shown here, which reveals the variation from one elements to other is relatively small. Maximum surface sensitivity at 50 eV, attenuation length is about 4 A. Large attenuation below 50 eV is because of decreasing the probability of exciting plasmons for lower kinetic energies.

Surface sensitivity of electron spectroscopy

Depth sensitivity: effect of take-off angle

Si

Si

C Si

C Si

Large take-off angle: Bulk signal is attenuated, surface signal remains same

XPS Application

XPS Application

XPS Application

Windows are coated with complex Multilayer thin films to meet demands Energy Conservation Appearance Durability

XPS Application
Pb3O4

Egyptian Mummy 2nd Century AD World Heritage Museum University of Illinois

PbO2 C O 150 145 140 135 130

Binding Energy (eV)

Pb Pb

N Ca Na Cl

Pb

500

400

300

200

100

XPS analysis showed that the pigment used on the mummy wrapping was Pb3O4 rather than Fe2O3

Binding Energy (eV)