Latest Developments in X-ray Photoelectron Spectroscopy ... · (Auger electron spectroscopy,...

A Shimadzu Group CompanyA Shimadzu Group Company

www.kratos.com

D. Surman Ph.D., Kratos Analytical Inc., Chestnut Ridge, NY 10977

Latest Developments in X-ray Photoelectron Spectroscopy and their Application to Nanosurfaces


Outline of Presentation

• Brief discussion of the technique– Relationship of the technique with other analytical methods– Relevance of the technique to nano-applications– Fundamentals of the technique

• What types of information does the technique provide

– Developments and Applications of the technique• General information

• Nano-scale information– Application of Angle Resolved XPS to provide nano-scale depth

information

– Extension to macro-scale depth information

• Lateral information– 2D imaging

• 3D information– Sputter depth profiling


• Properties and reactivity of the surface will depend on:

• bonding geometry of molecules to the surface

• physical topography• chemical composition

• chemical structure

• atomic structure• electronic state

No one technique can provide all these

pieces of information. However, to

solve a specific problem it is seldom necessary to use every technique

available.

photons

ions

electrons

EMISSION

TRANSMISSION

Interaction with material

EXCITATION

Surface Analysis - Techniques Available


Analytical Technique Signal Measured Elemental Rang e Depth Resolution Surface info.

SIMS Secondary Ions H-U 5 - 30 Å Chemical composition

(secondary ion mass spectrometry) Chemical structure

TOF-SIMS Secondary Ions H-U, Large Organic 2000 Å (Scanning Mode) Adsorbate bonding

(time-of-flight SIMS) Molecules / Cluster Ions

TEM Transmitted Electrons X-Rays Na-U EDX N/A

(transmission electron microscopy)

FE-SEM, EDX Backscattered or Na-U 1 - 5 micrometres

(field emission SEM) Secondary Electrons and X-Rays

ISS Ions H- U monolayer atomic structure

(ion scattering spectroscopy) chemical composition

AES/SAM Auger Electrons Li-U 1 - 5 nm chemical composition

(Auger electron spectroscopy, scanning Auger microscopy)

ESCA/XPS Photoelectrons Li-U 1 – 10 nm chemical composition

(electron spectroscopy for chemical analysis, X-ray photoelectron spectroscopy) chemical structure

RAIRS IR photons organic, some inorganics monolayer Adsorbate bonding

(reflection-absorption infra-red spectroscopy)

STM - solid surfaces upper most atoms physical topography

(scanning tunnelling microscopy)

Analytical Technique Signal Measured Elemental Range Depth Resolution Surface info,



What information do we get from XPS?

• Surface sensitivity - photoelectron signal from first 1-10 layers of atoms and molecules.

• Identification of all elements (except H & He) at concentrations >0.1atomic%*.

• Quantitative determination of the elemental composition.• Information about the chemical state (molecular environment)

of the element.• Non-destructive analysis, including depth profiles, from the top

10 nm.• Destructive depth profiles of materials for 100s of nm.• Lateral variations in surface composition at >3µm resolution.• ‘Finger printing’ of materials using valance band.

• *Sensitivity - a sample with a surface of size 1 cm2 - this will have ca. 1015 atoms in the surface layer. In order to detect the presence of impurity atoms present at the 1% level, a technique must be sensitive to ca. 1013 atoms. Contrast this with a spectroscopic technique used to analyse a 1 cm3 bulk liquid sample i.e. a sample of ca. 1022 molecules. The detection of 1013 molecules in this sample would require 1 ppb (one part-per-billion) sensitivity - very few techniques can provide anything like this level of sensitivity.


The Photoelectron Spectrum – Surveys/Wide ScansWhich elements are present and how much have we got ?

Cu 2p

O KLL Auger

O 1s

N 1s

C 1s

Cu LMM Auger

Cu 3p

Cu 3sCl 2p

Cl 2s

• The energy of the photoelectron peak is indicative of the element from which it originated.

• The number of electrons (peak area) is related to t he elemental concentration.

• Both photoemission and Auger peaks observed in a spectrum.

• Peaks are superimposed on a rising background, due to inelastically scattered photoelectrons.


O 1s

C 1s

O KLLAuger

O 1s C 1s

The Photoelectron Spectrum – High ResolutionWhat chemical state are the elements in?

High resolution spectra show the subtle shiftsin energy of photoelectrons originating formdifferent chemical environments.


• X-rays generated by accelerating high energy electrons onto an anode. The core holes created decay by emission of X-rays.

• Commonly used X-ray sources:anode material energy (eV) width (eV)

Mg 1253.6 0.70Al 1486.6 0.85

• Use of a monochromator prevents electrons, Bremsstrahlung, satellite X-ray lines and heat radiation striking the sample.

• The monochromator also decreases the energy spread of the X-rays. The typical energy spread is ~0.2eV, substantially enhancing the information content of the spectra.

Monochromated X-ray source


Developments in Monochromator technology

Toroidal quartz backplane

Electron gun & x-ray anode

Rowland circle diameter

Enhancements to the source design to improve small area intensity include filament design and power dissipation technology.These increase flux and therefore sensitivity (detection levels)

Modifications to the crystal geometry improve focusing of the optics for smaller irradiated areas. This increased x-ray flux density at the sample enhances sensitivity.

Changes to the Rowland circle modifies the performance of the monochromator to match energy resolution or spatial resolution requirements


Monochromated vs non-monochromated X-ray source

Monochromated Al Kα excited Ag spectrum

Non-monochromated Mg Kα excited Ag spectrum

FWHM 0.97 eV

FWHM 0.46 eV

satellite

Things to note: loss of rising background, narrower peaks, loss of satellite peaks, movement of Auger peaks

Ag 3d

Ag 3p3/2

Ag 3p1/2

Ag 3p3/2Ag 3p1/2

Ag 3d5/2

Ag 3d3/2

Ag 3d5/2

Ag 3d3/2Ag 3d

Ag 3s

Ag 3s

Ag MNV Auger

satellite


Chemical State Information

• The binding energy of an electron is dependent on the atomic orbital the electron occupies and the chemical environment of the atom.

• The variation of binding energy of a specific photoemission peakprovides information on the chemical state of the atom or ion.

Core level electron, high binding energy

Valence electron, low binding energy


Chemical State Information

• Typical binding energies for C 1s photoemission peaks from organic materials

• functional group binding energy (eV)

• hydrocarbon C-H, C-C 285.0• amine C-N 286.0• alcohol, ether C-O-H, C-O-C 286.5• fluorocarbon C-F 287.8• carbonyl C=O 288.0• 2F bound to a carbon -CH2CF2- 290.6• 3F bound to a carbon -CF3 293-294

• Typical chemical shifts for O 1s photoemission peaks from organic materials

• functional group binding energy (eV)

• carbonyl -C=O, O-C=O 532.2• alcohol, ether -O-H, O-C-O 532.8• ester C-O-C=O 533.7


Quantitative Surface Analysis of Poly(ethylene tetraph thalate) - PET Large Area Survey

O KLL

Peak Position FWHM Raw Area RSF Atomic Atomic Mass BE (eV) (eV) (CPS) Mass Conc % Conc %

C 1s 282.000 3.181 1988710.0 0.278 12.011 74.54 68.73 O 1s 530.000 3.623 2089951.0 0.780 15.999 25.46 31.27


Quantitative Surface Analysis of Poly(ethylene tetraph thalate) - PETChemical State Information

C 1s region O 1s regionO(1) 530.8eV 49 at% O(2) 532.1eV 53 at%

C(1) 285.0eV 61 at%C(2) 286.5eV 21 at%C(3) 289.2eV 18 at%

C3C2

C1 O1O2

-(-O-C- -C-O-CH2-CH2-)-= =

O On

2223

1

32

1

1


High Resolution Spectroscopy – ‘Fingerprints’

•Three isomers of PBMA•Monochromatic X-rays•Identical C1s core level•Valence band data

•‘fingerprint’ polymers•5 minute acquisition•20eV pass energy


Beyond Large Area Spectroscopy

thin film

<10 nm

thicker layer or multilayers heterogeneous surface

angle resolved XPS depth profile parallel imaging selected area spectroscopy


Angular Dependence of XPS

d

d ~ 8-10 nm

d

photoelectrons

X-rays

photoelectrons

X-rays


Angular Dependence of XPS - SiO 2 on Si wafer

0 deg (bulk sensitive)

60 degrees

45 degrees

75 degrees(surface sensitive)

Si elemental

Si oxide


Precise overlayer/oxide thickness calculation

0

1

2

3

4

0 2 4 6 8

ln(1 + IoSs/IsSo)

sec(

θθ θθ)Comparison of “oxide” signal and substrate signal yields an oxide layer thickness of 1.64nm


Development of ARXPS for nano-scale information

• Application of MEM’s methodology to ARXPS results– assumptions postulated by the model used in MEM calculations :

• the electron transport in the sample must be determined by the inelastic scattering with the constant mean free path throughout the sample and the material density must be constant throughout the sample.

• The necessary condition for the consistency of experimentally measured data with the MEM model (successful fit) is that the Laplace transform of the compositional depth profile (LTCDP) calculated from the experimentally measured normalized intensity is a monotonically decreasing function for all measured elements.

– Enhancements to the MEM’s methodology

• Development of an efficient algorithm which can estimate the LTCDP for elements with different inelastic mean free paths and thus provide an independent estimate of the noise scaling calculated by the MEMSYS algorithm and verify whether the MEM model is appropriate for the measured data.


Etched Oxidized SiC film at different take-off angle s

Surface

Bulk


0°Take-off angleBulk

75°Take-off angleSurface

More C-C present on the surface

Higher Oxidised states of carbon

Si-C

Analysis of the chemistry of nano-layers


Surface C-C rich noted (see component C2 in red)

Si-C present below the hydrocarbon contamination (C1 component in black)

Note : C3 and N1s concentrations have been scaled (x5)

0

20

40

60

80

100

0 2 4 6 8 10

Ato

mic

Con

cent

ratio

n [%

]

Depth [nm]

Reconstructed depth profiles alfa =10.4093 c = 4.34217

C1C2

C3 * 5 O 1s

N 1s * 5 Si 2p

Chemical nano-scale depth information


Development of lateral information

• Implementation of spherical mirror analyzer opened up new avenues for the detection of laterally resolved information– Spherical mirror provides a real-time energy resolved image of the

distribution of photoelectrons from the surface.• Initial detection systems provided only a qualitative distribution

• Development and implementation of the Delay Line Detector enabled directly quantifiable images to be obtained

• Development and implementation of mathematical models have enhanced the information content of the images to provide quantitative information from single pixels.


Parallel Imaging Mode of the Spherical Mirror Analy zer

Outer hemisphere of HSA

Charge neutraliser

Spherical mirror analyser (SMA)

Hemispherical analyser (HSA)

Magnetic lensSample

Selected area aperture

Objective lens

Delay Line Detector

Retarding projectorlens

• Objective lens produces magnified photoelectron image I 1

• Projector lens produces image I 2retards to pass energy E 0

• Magnification at detector variable from <5x to >100x

• Field of view on sample from >2mm to <100µµµµm

• Lateral resolution to <3 µµµµm• Operation in FAT mode and provides

energy resolution comparable to HSA.

I1

I2


High Spatial Resolution of SMA

2.2 µm edge

25µm from centre of bars

After acquisition of the image a line scan can be generated. The line-scan can be processed to provide edge measurements, giving an indication of lateral resolution


Set of images as a function of Binding Energy (eV)

x (mm)

y(m

m)

256x256 pixels

270eV

300eV

Spectra from Images With Spherical Mirror Analyzer

• The experiment requires the acquisition of a sequence of images over any energy range.• 300 - 270eV with 0.2 eV steps corresponding to C 1s region.• 30 sec per image - total acquisition 75 mins

• Sum intensity of pixels within defined area to obtain spectral information form that area


Application of mathematical modelling (PCA)

SignificantComponents

Noise

AF 1 AF 2 AF 3

AF 4 AF 5 AF 6

PCA

PCA methods allow the reduction of noise in the images substantially enhancing the chemical and spatial information


Spectra from images

As generatedspectrum generated from significant PCA components

It is possible to use the image dataset to reconstruct a spectrum. The example below shows a spectrum generated from a single pixel (blue) using the raw images. When the noise has been removed from the image dataset using the PCA approach a much more convincing spectrum is shown.

This spectrum was defined from a single pixel 1.4 x 1.4um area ! It is possible to generate a spectrum from any of the 256 x 256 pixels on the imaged area – over 65,500 spectra !!


C 1s spectrum generated from 5x5 pixels

CC,CH

CO

C=O

-CF2

-CF3

Enhanced data processing improves spectra from images

• After generating spectrum from images it is then possible to deconvolute the data and fit with components.

• An image may then be generated corresponding to a specific component.

– Has the advantage that peak shifts may be accommodated in by the model.

– Integrating area under the peak allows chemical state images show relative concentrations.


Quantitative Elemental and Chemical State Imaging

CH, CC CO, CN C=O CF2

-CF3 Fluorine Nitrogen Oxygen


Advantages of imaging for biologically active surfa ces

• Surface and bulk compositions not always the same– Micro-contact angle measurements produce different results from

that predicted by bulk composition– Surface properties are critical in determining bonding of proteins

etc.– Imaging XPS can detect and indicate subtle changes in the surface

composition which lead to changes in the material’s ability to separate different proteins


Image stitching – high spatial resolution over large image areas

glass slide

poly(hydroxyethyl methacrylate)(pHEMA)

Acrylate, diacrylate, dimethylacrylate and triacrylate

combinations (as above)

D.G. Anderson, S. Levenberg & R. Langer Nature Biotechnology 22 (7) 863-866

Monomers used to generate sample array


2mm 2mm 2mm

Fluorine (peak-background) Oxygen (peak-background) Carbon (peak-background)

Image stitching – high spatial resolution over large image areas

● Peak-background images – not yet corrected for transmission function, or relativesensitivity factor (RSF)

● Thermal scale simply provides intensity / counts per pixel

10 x 10 image array 64x64 pixels per image acquired for 10 secondsTotal acquisition time per element 3000 sec


Quantitative elemental stitched images

2mm 2mm 2mm

Fluorine (peak-background) Oxygen (peak-background) Carbon (peak-background)

● Quantitative elemental images corrected for transmission function and RSF● Thermal scale now indicates relative atomic concentration at a pixel

- each pixel represents 6.25um on the surface


Quantitative elemental stitched images

Fluorine11.4 at%

0 at%

90.6 at% 60.8 at%Carbon

Oxygen32.5 at%

7.1 at%

2mm


Extending 2D information into 3D

• Several methods available for extending the top surface information into 3D

– Sputter depth profiling• Most common method used

• Utilizes either inorganic or organic molecules as the sputtering agent– Ar ions are typically used for inorganic materials– Organic molecules such as Coronene or C60 are used for organic

materials

• This method enables nanometer information to be extended to 100’s of nm’s or even into the µm range

– Mechanical methods• Not precise but can be used to investigate large depths


Si 2p region

Si (elemental)

Si (carbide)Si (oxide)

Si 2p region

SiC/Si : high resolution snapshot spectra


SiC/Si : Elemental depth profile

Oxygen

Silicon

Carbon

750 V Ar+ ions3x3mm rastered crater


SiC/Si : Si chemical state depth profile

Si (

elem

enta

l)

Si (

carb

ide)

Si (

oxid

e)

Si (

elem

enta

l)

Si (

carb

ide)

Si (

oxid

e)

Si (

elem

enta

l)

Si (

elem

enta

l)

Si (

carb

ide)

2 41 3

1

2

3

4

Si (elemental)Si (oxide)Si (carbide)

surface

bulk

SiC/Si interface

SiC sub-surface


20kV Electron impact, mass filtered, polyatomic ion source for Organic material profiling

Mass filter

Focussing optics with retardation lens

Source vacuum isolation valve

Ionisation chamber

Differential Pumping

Sublimation oven

Neutral suppression


12kV PAH, 197nm PLGA on Si

12kV C60, 197nm PLGA on Si

Retention of chemical information using Coronene on PLGA

Slight loss of stoichiomery

Stoichiometrymaintained to interface


134 nm PLGA on Si: 12kV Coronene

C 1s

O 1sSi 2p


Profile through modified Carbon film: Cluster ion s ource depth profile

surface

bulk

bulk

surface

CF3 CF2 O-C=OC-O

C-C O-C=OC-O

C-C


Modified Carbon film: Cluster ion source depth prof ile

0

10

20

30

40

50

60

70

80

90

0 2 4 6 8 10 12 14

etch time (seconds)

conc

entra

tion

(%)

carbon

oxygenfluorine

Carbon oxygen fluorine

0

5

10

15

20

25

30

35

40

45

50

0 2 4 6 8 10 12 14

etch time (seconds)

conc

entra

tion

(%)

C-C aromatic

C-C aliphatic

C-O

O-C=O

C=OCF2

CF3


Conclusions

• XPS is an ideal technique for the investigation of material surfaces providing information on the nanometer scale

– It provides elemental information– Chemical information– Quantitative information– Detection levels on the order of 1ppb (cf bulk analysis techniques)– Spatially resolved information at the micron level

• Quantitative imaging

– 3D distributions of both elements and chemical states• Use of Polyatomic species enhances organic material information


Acknowledgements

• All my colleagues at Kratos Analytical for their contributions both to this talk and the continued development of the instrumentation that makes this all possible.

• Neil Fairley – CASA XPS

• Dr. Morgan Alexander – Nottingham University, UK

• Dr. Alex Shard – NPL, Teddington, UK

Latest Developments in X-ray Photoelectron Spectroscopy ... · (Auger electron spectroscopy,...

Documents

Transcript of Latest Developments in X-ray Photoelectron Spectroscopy ... · (Auger electron spectroscopy,...