XX--ray Probes of the Layer and Interface Structure ofray Probes of the Layer and Interface Structure of

NanoscaleNanoscale Films for Electronics and SpintronicsFilms for Electronics and Spintronics

Brian K Tanner (and others)

Bede plc, Belmont Business Park, Durham

and

Department of Physics, Durham University

2

Opto-electronic Materials Driver for High Resolution X-ray Diffraction

Development in 1980s

ØInxGa1-xAsyP1-y for PIN diodes for long-haul telecoms

ØAlxGa1-xAs for laser diodes

Measurement of composition

3

High Resolution X-ray Diffraction

X-ray tube

X-ray mirrors

Monochromator and conditioner crystals

Monochromatic, parallel beam

Composition of AlxGa1-xAs on GaAs

Measure splitting between substrate and layerAssume Vegard’s Law (linear interpolation) to get x

1000

10000

-100 0

Simulation with Lorentzian diffuse scatterSimulationExperiment

Specimen angle (arc sec)

Inte

nsity

(cps

)

Composition of AlxGa1-xAs on GaAs

Quadratic dependence 4 arc seconds ≅ 1% Al

B K Tanner et al Appl Phys Lett 59 (1981) 2272

S Gehrsitz et al Phys Rev B 60 (1999) 11601

0

100

200

300

400

0 0.2 0.4 0.6 0.8 1.0

r2 = 0.99933 linearr2 = 0.99984 quadratic

x

Bra

gg p

eak

split

ting

(arc

sec

onds

)

Takagi-Taupin formulationof X-ray Dynamical Diffraction Theory

hh DCDD

−+= χχ∂∂

πλ

000

0

i s

00 )(i

DCDD

hhhh

h χαχ∂∂

πλ

+−=s

αh is the deviation of the incident wave from the exact Bragg conditionso and sh are unit vectors in the directions ofKo and Kh

C is the polarisation factorχ0 and χh are the electric susceptibilities(these describe the crystal) χ

λπhe

h

rV

F= −2

Recursion relation format

A C h= −χ Bb h=

−+

( )12 2

0χ α π D = π

λγ 0

E Cb h= − γ F BB EA= −

where b= γo/γh.

Amplitude ratio X = Dh/Do,

X X' F BX'+E DF(z - w)

F AX'+B) DF(z - w)=

+ i( ) tan( ) - i( tan( )

z is depth above the depth w at which the amplitude ratio is the known value X'

M A G Halliwell, J Juler and A G Norman, Inst Phys Conf Ser 67 (1983) 365M G A Halliwell, M A G Lyons and M J Hill, J Crystal Growth 68 (1984) 523

Recursion relation format

Fit to a graded 2.3µm GaInAs layer, nominally lattice matched to InPNote mismatch of only several hundred ppm

M J Hill, B K Tanner, M A G Halliwell and M H Lyons J Appl Cryst 18 (1985) 446

9

Graded SiGe structure (blanket)

• QC200 diffractometer + bede RADS Mercury auto-fitting software

sample courtesy Hitachi-Kokusai Electric

010

110

210

310

410

510

-4000 -3000 -2000 -1000 0 1000 2000

Standardized Chi Squared: 0.013 R Factor: 8.82

Inte

nsity

Seconds

slot-23_0aa1.x01 temp.sim

Inte

nsity

(log

sca

le)

Ge profile for Slot 23

0

0.05

0.1

0.15

0.2

0.25

0.3

0 20 40 60 80 100 120

Depth / nm

Ge

frac

tion

Slot 23

10

Asymmetric relaxation: Reciprocal space maps of test structures: 10 µm × 0.5 µm

• (004) (224)GE (004)

(224)GESee Frontiers of Characterization and Metrology for Nanoelectronics Conference 2007

11

MgO Tunnel Junction for Spintronics Applications

SiSiO2

Co

NiFeMo

MgO

Ta

Co

Is there a change in the MgO interface structure when the bottom Co layer has a monolayer of CoO deposited?

12

First User Data from Diamond Synchrotron Light

Source

First User Data from Diamond Synchrotron

Light Source

13

X-ray Optics for Reflectivity

X-ray tube

X-ray mirrors

Monochromator and conditioner crystals

Monochromatic, parallel beam

14

X-ray Optics for Reflectivity

X-ray tube

X-ray mirrors

Slit monochromator

Monochromatic, parallel beam

ØRemove crystal monochromatorØUse slit to define wavelength

dispersion

Conditioner crystal

15

Ultra-thin transition metal films grown by Molecular Beam Epitaxy

Si (001)sample a sample b sample c

C u ( 9 0 0 Å )

N i ( 6 0 Å )Co (10Å)

C u ( 8 0 Å )

C o ( 1 0 Å )Cu (80Å)

C u ( 8 0 Å )

ØShadow masks used to create three samples with identical Cu and (where appropriate, Ni and Co) thicknessØPolarized neutron reflectivity analysis of magnetic momentsØNeed precise layer thickness to reduce number of free fitting parameters in PNR analysis

16

Ultra-thin Films - Specular Reflectivity

Si (001)sample a sample b sample c

C u ( 9 0 0 Å )

N i ( 6 0 Å )Co (10Å)

C u ( 8 0 Å )

C o ( 1 0 Å )Cu (80Å)

C u ( 8 0 Å )

100

102

104

106

108

0 0.05 0.10 0.15 0.20 0.25

C

B

A

qz (Å-1)

Cou

nt ra

te (c

.p.s

.)

840823830Cu thickness (Å)

-6164Ni thickness (Å)

1011-Co thickness (Å)

848275Cu cap thickness (Å)

Sample C

Sample BSample A

17

Scientific Conclusions

C.A.Vaz, G.Lauhoff, J.A.C. Bland, S.Langridge, D Bucknall, J Penfold,

J.Clarke, S.K.Halder and B.K.Tanner, J. Magn. Mag. Mater. 313 (2007) 89-97

ØPrecise measurement of magnetic moment possible due to restricting parameter spaceØPer Ni atom, 0.58±0.03µB and 0.59 ±0.03µB for Ni/Cu and Co/Ni/Cu samplesØNo effect of interface environment detected

Measurement of hot electron momentum relaxation times in metals by femtosecond ellipsometryV.V. Kruglyak, R.J. Hicken, M. Ali, B.J. Hickey, A.T.G.Pym and B.K.Tanner, Phys Rev B 71 (2005) 233104

Another example of need to determine layer thickness

Grazing incidence X-ray scattering• Fluorescence

Incident beam

Asymmetricdiffraction

Diffuse scatter

Diffuse scatter

Specular reflection

Evanescentwave

19

Principle of Grazing IncidenceIn-plane X-ray Diffraction

•

Beam measured in grazing incidence reflectivity

Diffracting planes

Beam measured in GIIXD

X-ray beam incident at grazing angle

Bragg angle, θ

2θ

•

Beam measured in grazing incidence reflectivity

Diffracting planes

Beam measured in GIIXD

X-ray beam incident at grazing angle

Bragg angle, θ

2θ

Beam measured in grazing incidence reflectivity

Diffracting planes

Beam measured in GIIXD

X-ray beam incident at grazing angle

Bragg angle, θ

2θ

Ø Bragg planes normal to the sample surfaceØ Probes in-plane lattice parameter (in-plane strain)Ø Probes in-plane mosaicØ Probes in-plane length scale

20

High Resolution GIIXD

0

0.35

0.70

-0.1 0.1

Optic + Ge crystalPolycapillary optic

Specimen rotation angle (°)

Nor

mal

ised

inte

nsity

(Rocking curve) scans of the Si specimen about its surface normal, with the detector set for the 220 reflection, with and without the Gemonochromator crystal. Solid lines are fits of Voigt functions to the data.

T. A. Lafford, P. A. Ryan, D. E. Joyce, M. S. Goorsky and B. K. Tanner, Phys. Stat. Sol (a) 195 (2003) 265

21

Tilt and Twist Mosaic

Tilt is misorientation out ofout of the wafer plane

Twist is the misorientation inin the wafer plane

22

Tilt and Twist Mosaic

Tilt is misorientation out ofout of the wafer plane

Twist is the misorientation inin the wafer plane

23

Twist and Tilt Mosaic as a Function of GaN Layer Thickness

0

0.2

0.4

0.6

0.8

1.0

0 5000 10000 15000 20000

Fit to Ayers' model and DE/D

S constant

Fit to Ayers' modelTwist mosaicTilt mosaic

Total thickness (Å)

Ro

ckin

g cu

rve

FWH

M (

°)

Ratio of number of threading screw to threading edge dislocations constant

T A Lafford, P J Parbrook and B K Tanner, Phys Stat Sol (c) 0 (2002) 542

24

Twist and Tilt Mosaic inSapphire/GaN/AlN/AlxGa1-xN

0.1

0.2

0.3

0.4

80 120 160 200

00.2 AlGaN (tilt)00.2 GaN (tilt)11.0 AlGaN (twist)

AlN interlayer thickness (Å)

FWH

M (

°)

0

0.1

0.2

0.3

0.4

0.5

0.1 0.2 0.3 0.4 0.5

00.2 AlGaN (tilt)00.2 GaN (tilt)11.0 AlGaN (twist)

Al fraction xFW

HM

(°)

T. A. Lafford, P. J. Parbrook and B. K. Tanner, Appl Phys Lett 83 (2003) 5434

AlN interlayer between micron thick GaN and AlxGa1-xN layers. MOVPE. No change in the tilt mosaic (threading screw dislocation density) with thickness or Al fraction x of the upper layer. A linear increase in the twist mosaic (threading edge dislocation density) was observed as a function of interlayer thickness and x. For all samples the twist mosaic of the AlGaN was significantly greater, by at least a factor of two, than that of the GaN layer.

25

Durham reflectometer with fluorescence attachment

26

Spin Valve Structure for Spintronics Applications

SiBuffer

Ni80Fe20

Cu

Ta

Ni80Fe20

MnFe

1nm

27

Grazing Incidence Fluorescencefrom Spin Valve Structures

• Deconvolution from Gaussian peaks• Measure area under peaks

0

0.2

0.4

5 6 7 8 9 10

IncidentBeam

Fe Kβ Cu K

β

Ni Kβ

Cu Kα

Ni Kα

Mn Kβ

Fe Kα

Mn Kα

Energy (keV)

Nor

mal

ised

Int

ensi

ty

28

Grazing Incidence FluorescenceAnalysis of Bi as a Surfactant

• Evanescent wave below critical angle• Used MoK radiation to excite Bi

0.00 0.25 0.50 0.75 1.00101

102

103

104

105

106

Cu Kα1 Mo Kα1

Pene

tratio

n D

epth

(Å)

α (deg)

Penetration depth L is

L = λ/{π(αc2 - α2)]½}

where αc is the critical angle for total external reflection

As the incidence angle rises, penetration of wave increases.

Around critical angle, very high sensitivity to depth

29

Normalized Grazing Incidence Fluorescence Data

Ø Ni fluorescence almost independent of Bi thicknessØ Cu fluorescence shows step at critical angle from

surface layer

30

Normalized Grazing Incidence Bi Fluorescence Data

Ø Low Bi thickness – fluorescence rises at lower angle compared to thick layer

Ø Peak shifted to higher angle for thick layer

50Å

Bi moves to the surface for monolayer coverageBi stays low for thick layer

31

Grazing Incidence FluorescenceBi above and below FeMn

• 25Å Bi below FeMn, 25Å low density Bi above

• Detailed fitting difficult

0

0.2

0.4

0.6

0.8

1.0

1.2

0 500 1000 1500

Top Bi layer: 100% density, 15Å thickTop Bi layer: 80% density, 18.75Å thickTop Bi layer: 60% density, 25Å thickTop Bi layer: 40% density, 37.5Å thickExperimental Data

Sample Angle (arc-seconds)

Inte

nsity

Sample A13 - Bi Fluorescence

0

0.2

0.4

0.6

0.8

1.0

1.2

0 500 1000 1500Sample Angle (arc-seconds)

Inte

nsity

Sample A13 - Mn Fluorescence

0

0.2

0.4

0.6

0.8

1.0

1.2

0 500 1000 1500Sample Angle (arc-seconds)

Inte

nsity

Sample A13 - Fe Fluorescence

32

Fits to simulation of fluorescence yield as function

of angle

0

1.5

3.0

4.5

6.0

7.5

0 1000 2000 3000 4000

Ni Kα (x 32)Cr Kα

Sample Angle (arc seconds)

Pea

k A

rea

(abr

itrar

y un

its)

Peak Areas from 200Å Cr Single Film

0

0.2

0.4

0.6

0 1000 2000 3000 4000

198.2Å Cr(1-x) Ni(x) (x=1.3%)198.2Å Cr / 3Á NiExperimental Data

Sample angle (arc seconds)

Inte

nsi

ty (

arb

itra

ry u

nit

s)

Simulations of Ni Fluorescence for Different Ni Contamination

33

Collaborators and Contributors

Bede plc

Paul RyanTamzin LaffordMatthew WormingtonStéphane GodnyKeith Bowen

Durham University

Tom HaseAlex PymDavid EastwoodStuart Wilkins (now at BNL)

Peter Parbrook (Univ of Sheffield)Mai Zhenhong (CAS, Beijing)Mark Goorsky (UCLA)Bill Egelhoff (NIST)Carlos Vaz (Univ of Cambridge)Tony Bland (Univ of Cambridge)