CMP Seminar MSU 10/18/ 2000 1 What makes Surface Science “surface” science ? R. J. Smith Physics Department, Montana State Univ. Work supported by NSF.

CMP Seminar MSU 10/18/ 2000CMP Seminar MSU 10/18/ 200011

What makes Surface Science “surface” science ?

What makes Surface Science “surface” science ?

R. J. Smith R. J. Smith Physics Department, Montana State Univ.Physics Department, Montana State Univ.

Work supported by NSF (DMR) Work supported by NSF (DMR) http://www.physics.montana.eduhttp://www.physics.montana.edu

CMP Seminar 10/18/ 2000CMP Seminar 10/18/ 2000 22

Outline

Motivation for surface sensitivity - thin film devicesMotivation for surface sensitivity - thin film devices General comments on surface sensitive techniquesGeneral comments on surface sensitive techniques Electron spectroscopy Electron spectroscopy

define the attenuation length (AL)define the attenuation length (AL) empiracle results for attenuation lengthempiracle results for attenuation length simple model for overall shape of ALsimple model for overall shape of AL

Recent analyses based on attenuation lengthsRecent analyses based on attenuation lengths


Metal-metal Interface Structure

Understand overlayer growth and alloy formationUnderstand overlayer growth and alloy formation Chemical composition and structure of the interfaceChemical composition and structure of the interface Applications: magnetoresistive devices, spin electronicsApplications: magnetoresistive devices, spin electronics

Surface energy (broken bonds)Surface energy (broken bonds)

Chemical formation energyChemical formation energy

Strain energyStrain energy

A

B0int AB

energyformation ABBA

energystrain )()( equilobs dEdE

interface


Metal-metal systems studied...

Substrates: Al(111), Al(100), Al(110)Substrates: Al(111), Al(100), Al(110) Metal overlayers studied so far:Metal overlayers studied so far:

Fe, Ni, Co, Pd (atomic size smaller than Al)Fe, Ni, Co, Pd (atomic size smaller than Al) Ti, Ag, Zr (atomic size larger than Al)Ti, Ag, Zr (atomic size larger than Al)

All have surface energy > Al surface energyAll have surface energy > Al surface energy All form Al compounds with All form Al compounds with HHformform < 0 < 0

Use resistively heated wires ( ~ML/min)Use resistively heated wires ( ~ML/min) Deposit on substrate at room temperatureDeposit on substrate at room temperature


Comments on surface sensitive techniques...

Sensitivity to the surface is intrinsic to the Sensitivity to the surface is intrinsic to the technique - not based only on probing depthtechnique - not based only on probing depth

Electron spectroscopy: Distance traveled by Electron spectroscopy: Distance traveled by the electron before losing characteristic the electron before losing characteristic information (Attenuation length) is shortinformation (Attenuation length) is short

Low energy ions - neutralization and strong Low energy ions - neutralization and strong Coulomb interationCoulomb interation

High energy ions - geometric shadowingHigh energy ions - geometric shadowing


Define the attenuation length

Measure decrease in beam intensity dI(x) after Measure decrease in beam intensity dI(x) after transmission through a film of thickness dxtransmission through a film of thickness dx

1( )

cos

dxdI x I

/( cos )( ) e xoI x I I

dx

I(x) < I

86% of the signal originates from depth of 286% of the signal originates from depth of 2 What is value of What is value of ? What determines this value?? What determines this value?


Other discussion…and surprises

Idzerda, Journal Vac. Sci Technol A7(3), 1341 (1989)

S. Ossicini, et al. JVST A3, 387 (1985) - growth models S. Tanuma, et al. JVST A8, 2213 (1990) - attenuation lengths


Observations of Attenuation Length for electrons in solids

Depends on Depends on electron kinetic electron kinetic energy (KE)energy (KE)

KE depends on KE depends on parameters of the parameters of the technique:Auger, technique:Auger, XPS, SEMXPS, SEM

Varies with Varies with materialsmaterials EEAL 54.0/1430 2

F

F

E

EEEdEEdEEEEEW

dEEEP

)'(')'(]''))'(''()''([

')',(

E

E’E’’+(E-E’)

E’’

EFFilled e-

states

Empty e-states

Low Energies: 1 eV < E < 100 eV

E = Kinetic Energy; Elastic Scattering

Transition rate (like Golden Rule)

W=(Transition matrix element)2

=density of filled or empty states

Total scattering probability/sec

[1/]E

EFdEEEPEP ')',()(

E

E’E’’+(E-E’)

E’’

EFfilledstates

emptystates

To go further might assume free-electron form for the density of states

Attenuation Length at low energies

2 2( ) ( / 8)oP E N A E

)()( 0EEAE

Distance between collisions is then

gv 1g

Ev

k

2 2

2

kE

m

Simple model (for homework) 0( '')E N

3

2( )const E

No

E

E’E’’+(E-E’)

E’’

EFFilled e-

states

Empty e-states

High Energies: E > few 100 eV

Consider scattering of point charges, i.e. Coulomb Scattering

Unscreened Coulomb Potential has cross section ~ 1/E2

21~ E

N

•For small E phonon scattering may dominate

•Insulators and semiconductors have energy gap so have long AL as approach twice the gap energy


Ion scattering chamber Ion scattering chamber

High precision High precision sample goniometersample goniometer

Hemispherical VSW Hemispherical VSW analyzer (XPS, ISS)analyzer (XPS, ISS)

Ion and x-ray sourcesIon and x-ray sources LEEDLEED Metal wires for film Metal wires for film

depositiondeposition


FM growth:layer-by-layerFM growth:layer-by-layer

Non-linear Non-linear growth curvesgrowth curves

Rapid attenuation Rapid attenuation of substrate signalof substrate signal

Breaks in slopeBreaks in slope =6,6 (green) =6,6 (green)

=20,20 (red)=20,20 (red)

substrate


VW mode: Island growthVW mode: Island growth

Linear growth in Linear growth in signalssignals

Relatively slow Relatively slow decrease in decrease in substrate signalsubstrate signal

=6,6 (green) =6,6 (green) =20,20 (red)=20,20 (red)

substrate


Co on Al (100): He+ backscatteringCo on Al (100): He+ backscattering

Ion channeling Ion channeling

Only near-surface Only near-surface

Al atoms are visible Al atoms are visible

to the ion beamto the ion beam

Increase of Al peak Increase of Al peak

means Co causes Al means Co causes Al

atoms to move off atoms to move off

lattice siteslattice sites

Coverage from Co Coverage from Co

peak area (RBS)peak area (RBS)


Co on Al (100): HEIS intensitiesCo on Al (100): HEIS intensities

HEIS Al surface peak HEIS Al surface peak

vs. Co coverage vs. Co coverage

Number of visible Al Number of visible Al

increases up to 3 MLincreases up to 3 ML

Slope of 2:1 suggests Slope of 2:1 suggests

stoichiometry Alstoichiometry Al22Co Co

for the interface but for the interface but

AlAl22Co not in phase Co not in phase

diagramdiagram


Co on Al (100): XPS intensitiesCo on Al (100): XPS intensities

Measured (o ,Measured (o , ) ) Simulation (lines)Simulation (lines)

Use HEIS for Use HEIS for coveragecoverage

0-3: layered CoAl 0-3: layered CoAl 3-10: Co islands3-10: Co islands CoAl particle CoAl particle

density is ~ 1.4 x density is ~ 1.4 x Al density Al density


LEED patterns for Co on Al(100)LEED patterns for Co on Al(100)

LEED at 42.8 eV LEED at 42.8 eV (a) Clean Al(001)(a) Clean Al(001) (b) 0.5 ML Co (b) 0.5 ML Co

destroys pattern destroys pattern completelycompletely

(c) 7.6 ML A hint (c) 7.6 ML A hint of some long range of some long range order (1x1) is seen.order (1x1) is seen.

Co coverage from Co coverage from RBSRBS


Co on Al(110): HEIS intensitiesCo on Al(110): HEIS intensities

Number of visible Al Number of visible Al

increases to 5 MLincreases to 5 ML

Slope 2.3:1, suggests Slope 2.3:1, suggests

stoichiometry of Alstoichiometry of Al22Co or Co or

AlAl55CoCo22

AlAl55CoCo22 exists in bulk phase exists in bulk phase

diagram, but crystal diagram, but crystal

structure is complex so structure is complex so

less likely to form at 30 less likely to form at 30 ooC. C.


Co on Al(110): XPS intensitiesCo on Al(110): XPS intensities

Measured (o ,Measured (o , ) ) Simulation (lines)Simulation (lines)

Use HEIS for Use HEIS for coverage (change coverage (change at 5 ML)at 5 ML)

0-5: layered CoAl 0-5: layered CoAl growth growth

3-10: layered Co 3-10: layered Co metal growthmetal growth


Comparison of XPS Intensities for Al(100) and Al(110)Comparison of XPS Intensities for Al(100) and Al(110)

Contract (110) Contract (110) coverage to coverage to (100) MLs(100) MLs

(100) Co(100) Co

(110) Co(110) Co

(100) Al(100) Al

(110) Al(110) Al

Co overlapCo overlap Al don’t !Al don’t !


Snapshots from MC simulationsSnapshots from MC simulations

Al(110)+0.5 ML Ni Al(110)+0.5 ML Ni Clean Al(110)Clean Al(110) Al(110)+2.0 ML Ni Al(110)+2.0 ML Ni

MC (total energy) using EAM potentials for Ni, Al (Voter)MC (total energy) using EAM potentials for Ni, Al (Voter) Equilibrate then add Ni in 0.5 ML increments (solid circles)Equilibrate then add Ni in 0.5 ML increments (solid circles) Ion scattering simulations (VEGAS)Ion scattering simulations (VEGAS)


Ion scattering simulations using VEGAS and the MC snapshotsIon scattering simulations using VEGAS and the MC snapshots

Measured (o) Measured (o) Simulation (Simulation ())

Slopes agreeSlopes agree Change of slope Change of slope

at 2 ML correct at 2 ML correct Good agreement Good agreement

so use snapshots so use snapshots for more insight for more insight


XPS chemical shifts for Ni 2pXPS chemical shifts for Ni 2p

Shifts in BEShifts in BE Shifts in satelliteShifts in satellite Compare with XPS for Compare with XPS for

bulk alloys bulk alloys

(BE) (sat)(BE) (sat)NiAlNiAl33 1.05eV 1.05eV

NiNi22Al 0.75eV (8.0 eV)Al 0.75eV (8.0 eV)

NiAl 0.2 eV (7.2 eV)NiAl 0.2 eV (7.2 eV)

NiNi33Al 0.0 eV (6.5 eV)Al 0.0 eV (6.5 eV)

Ni 0.0 eV (5.8 eV)Ni 0.0 eV (5.8 eV)


Simulated XPS intensity for Ni using EAM snapshotsSimulated XPS intensity for Ni using EAM snapshots

Ni coverage from Ni coverage from RBS RBS

Fit using model Fit using model with exponential with exponential attenuationattenuation

See a change in See a change in slope for all slope for all values of values of

Best fit:Best fit:NiNi= 5.2Å= 5.2Å


XPS: Comparison of Calculated and Measured Al Intensities XPS: Comparison of Calculated and Measured Al Intensities

XPS intensity vs XPS intensity vs Ni coverageNi coverage

Best agreement Best agreement with data for with data for NiNi

= 5.2 Å = 5.2 Å AlAl = =

15 Å15 Å Universal curve Universal curve

NiNi = 13.5 Å = 13.5 Å

AlAl = 20.2 Å = 20.2 Å

Equilibrium?Equilibrium?


SummarySummary

Surface sensitivity associated with short attenuation Surface sensitivity associated with short attenuation length for electrons in solidslength for electrons in solids

Long AL at low energy associated with decreased Long AL at low energy associated with decreased availability of final states for scatteringavailability of final states for scattering

Long AL at high energy associated with decreasing Long AL at high energy associated with decreasing scattering cross section for point chargesscattering cross section for point charges

Minimum AL for KE ~ 150 eVMinimum AL for KE ~ 150 eV Modeling of film morphology can be helpful for Modeling of film morphology can be helpful for

complex interface and alloy formationcomplex interface and alloy formation


XPS: Comparison of Calculated and Measured Ni 2p Intensities XPS: Comparison of Calculated and Measured Ni 2p Intensities

XPS intensity vs XPS intensity vs Ni coverageNi coverage

Best agreement Best agreement with data for with data for NiNi

= 5.2 Å = 5.2 Å AlAl = =

15 Å15 Å Universal curve Universal curve

NiNi = 13.5 Å = 13.5 Å

AlAl = 20.2 Å = 20.2 Å

Equilibrium?Equilibrium?

CMP Seminar MSU 10/18/ 2000 1 What makes Surface Science “surface” science ? R. J. Smith Physics Department, Montana State Univ. Work supported by NSF.

Documents

e e e ee e efef

e states

e states low energies

e states high energies

small e phonon scattering

gap energy slide

energy gap

cmp seminar msu