3D orientation microscopy based 3D orientation microscopy based on - - PowerPoint PPT Presentation

Max-Planck-Institut für Eisenforschung GmbH

3D orientation microscopy based 3D orientation microscopy based

• n FIB-EBSD tomography:

P t ti l d li it Potentials and limits.

• S. Zaefferer

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

p y tomography

• Application example:

Application example

– Coupling of 3D measurements with 3D modelling

• Material restrictions:
• Material restrictions:

– beam induced material changes

C l i

• Conclusions

2

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

p y tomography

• Application examples

Application examples

– Grain boundary characterization – Accurate observation of deformation structures Accurate observation of deformation structures – Microstructure characterization for 3D modelling

• Material restrictions:
• Material restrictions:

– beam induced material changes

C l i

3

• Conclusions

Max-Planck-Institute for Iron Research, Düsseldorf

The need for 3D observations P di i ll hi i f i b ibl

• S. Zaefferer: 3D orientation microscopy

Pre-condition: crystallographic information must be accessible to investigate the microstructure of crystalline matter 2D Stereology

statistical observations:

3D Non-destructive

process observations:

3D Destructive

static observations:

• grain size distribution
• grain shape (from 2

l i )

• recrystallization

(e.g. nucleation, grain

• comprehensive

morphology information 3D i i f sample sections)

• volume fraction and

distribution of 2nd growth)

• deformation (e.g.

t t f ti )

• 3D connectivity of

features

• grain boundaries

distribution of 2nd phase constituents

• texture-microstruc-

texture formation)

• phase transformation

(e g variant selection)

• grain boundaries
• input data for modelling
• 3D deformation

texture microstruc ture relations (e.g. variant selection) 3D deformation structures

4

Many problems can be solved by 2D statistical observations but for some 3D observations are essential

Max-Planck-Institute for Iron Research, Düsseldorf

Serial sectioning methods

S ti i

• S. Zaefferer: 3D orientation microscopy

Sectioning:

• mechanical or chemical polishing
• FIB milling

Problems:

• FIB milling….

Observation:

BSE Mi EBSD i l

Problems:

• depth definition
• contrast definition
• BSE-Microscopy, EBSD, optical

microscopy…. contrast def n t on for segmentation

• very laborious

M.V. Kral & G. Spanos, Acta

y

serial sectioning and reconstruction of p

• Mater. 47 (1999), 711

5

reconstruction of allotriomorphic cementite by mechanical polishing

Max-Planck-Institute for Iron Research, Düsseldorf

Advantages of FIB-EBSD tomography

• S. Zaefferer: 3D orientation microscopy
• Sectioning by FIB

– accurate depth definition – flat and parallel sections (< 1° deviation ) – high resolution (< 50 nm)

• Observation by EBSD

– well-defined contrast on crystalline material well defined contrast on crystalline material – ideal for reconstruction of grains in 2D and 3D – quantitative description of microstructure quantitative description of microstructure – high resolution (~ 50 nm)

• Combination of FIB and EBSD
• Combination of FIB and EBSD

– table-top instrument “high” measurement speed

Recent reviews:

• Uchic et al., MRS Bulletin 32

( 00 ) 40 416

6

– high measurement speed – fully automatic

(2007) 408-416

• Zaefferer et al., Met. Mater.
• Trans. 39A, (2008) 374-389

Max-Planck-Institute for Iron Research, Düsseldorf

Length scale of tomographic measurements

Midgley & Weyland

• S. Zaefferer: 3D orientation microscopy

Midgley & Weyland Ultramicroscopy 96 (2003) Zaefferer, Wright & Raabe

• Mat. Trans. A (2008)

and Mulders & Day Mat Sci Forum 495-497 (2005) B C Larson et al

crystals

B.C. Larson et al., Nature 415 (2002) 887

nano

M.V. Kral, G. Spanos, Acta Mater. 47 (1999) 711 H.F. Poulsen et al.,

• J. Appl. Cryst. 34

(2001) 751

7

macroscopic texture fields

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

p y tomography

• Application examples

Application examples

– Grain boundary characterization – Accurate observation of deformation structures Accurate observation of deformation structures – Microstructure characterization for 3D modelling

• Material restrictions:
• Material restrictions:

– beam induced material changes

C l i

8

• Conclusions

Max-Planck-Institute for Iron Research, Düsseldorf

Instrument overview

• S. Zaefferer: 3D orientation microscopy
• Scanning electron

microscope (SEM)

– observation of microstructure

G

• Scanning Ga+-ion

microscope (FIB = focused ion beam)

EBSD system: TSL with Hikari camera

(FIB = focused ion beam)

– sputtering of material for serial sectioning

• Quantitative images with

EBSD and EDX

SEM & FIB: Zeiss Crossbeam 1540

EBSD and EDX

– quantitative characterisation of i t t

9

microstructure

Max-Planck-Institute for Iron Research, Düsseldorf

Principle of serial sectioning & orientation microscopy

ion milling

• S. Zaefferer: 3D orientation microscopy

ion milling electron beam alignment marker SEM objective lens e- tilt 34 ° lens e to EBSD detector Ga+ sample in cutting position (36° tilt)

e-

EBSD (36 tilt) sample in EBSD position (70° tilt) camera

“tilt set-up” Zaefferer, Wright, Raabe,

5 µm

• Mat. Trans. A (2008)

10

Max-Planck-Institute for Iron Research, Düsseldorf

Geometrical set-up alternatives for FIB-EBSD

• S. Zaefferer: 3D orientation microscopy

54°

SEM

54

FIB cross-

• ver point

EBSD

70° 36°

EBSD sample

70°

EBSD

static set-up

+ no stage movement required

tilt set-up

+/- medium tilt positioning accuracy

rotation set-up

+ high stage positioning accuracy required + highest possible positioning accuracy + unconventional but non- accuracy + tilt inaccuracies create linear distortions + simple software accuracy +/- rotation inaccuracies create shear distortions +/- software correction unconventional but non problematic EBSD set-up + high measurement speed simple software correction possible + freely selectable milling position / software correction more complex +/- every milling position requires a different

11

p

Zaefferer et al., Met. Mater.

• Trans. 39A, 374-389 (2008)

q ff holder

Mulders, Day, Mat. Sci. Forum 495-497, 237-242 (2005)

Max-Planck-Institute for Iron Research, Düsseldorf

EBSD & FIB-sliceing: 3D microstructure of pearlite

• S. Zaefferer: 3D orientation microscopy

X Y X Z 20 µm 30 µm

12

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

p y tomography

• Application examples

Application examples

– Grain boundary characterization – Accurate observation of deformation structures Accurate observation of deformation structures – Microstructure characterization for 3D modelling

• Material restrictions:
• Material restrictions:

– beam induced material changes

C l i

13

• Conclusions

Max-Planck-Institute for Iron Research, Düsseldorf

The cube texture in Fe 36% Ni

• S. Zaefferer: 3D orientation microscopy

cold rolled material: Cu-type rolling texture recrystallized material: sharp cube texture

• Origin of the cube texture: oriented nucleation
• Possible reasons for texture selection:

yp g p

Possible reasons for texture selection:

– stored energy differences (Etter et al. Scripta Mat. 46 (2002) 311) – grain boundary properties (e g 40° <111>) (“micro- grain boundary properties (e.g. 40 <111>) ( micro

• riented growth”, (Duggan et al., Acta metall. mater. 41 (1993) 1921))

– differences in mobility of dislocations in different

14

ff r nc s n mo ty of s ocat ons n ff r nt

• rientations (differences in recovery rate) (Rhida &

Hutchinson, Acta metall 30 (1982) 1929)

Max-Planck-Institute for Iron Research, Düsseldorf

3D-orientation microscopy on cold rolled material

• S. Zaefferer: 3D orientation microscopy

Orientation map

RD ND

KAM map

RD ND TD TD

non cube grain cube grain

20 µm 20 µm 20 µm 20 µm

Cube grains have low internal orientation fluctuations

15

Cube grains have low internal orientation fluctuations. Some other orientations do have that as well.

Max-Planck-Institute for Iron Research, Düsseldorf

Microstructure after 1 min annealing

• S. Zaefferer: 3D orientation microscopy

possibility to reconstruct original neighbourhood of grown

• riginal

cube band neighbourhood of grown grains grown cube cube nucleus

• only small cube-grain areas with

40°<111> orientation relation

• original neighbourhood of grown

16

• original neighbourhood of grown

grain does not show any special boundaries with cube band

Max-Planck-Institute for Iron Research, Düsseldorf

Direct observation of the nucleation process

• S. Zaefferer: 3D orientation microscopy
• In-situ observations are difficult

(b t N ll t l R X& GG2 (2004)) (but see: Nowell et al. ReX& GG2 (2004))

• Modelling on the basis of orientation

microscopy data microscopy data

– Hypothesis: abnormal subgrain growth as nucleation mechanism of recrystallisation mechanism of recrystallisation

(Humphreys, Acta Mater. (1997))

• 3D-Monte-Carlo Potts model for the simulation

3D Monte Carlo Potts model for the simulation

• f subgrain growth

– Freely selectable energy and mobility functions Freely selectable energy and mobility functions – Experimental microstructures as input data – Stored energy determined from local orientation

17

Stored energy determined from local orientation gradients

Max-Planck-Institute for Iron Research, Düsseldorf

Nucleation simulation by a 3D-Monte Carlo Potts model

• S. Zaefferer: 3D orientation microscopy

0,8 1,0

(a.u.) energy

E(Θ)

0°<Θ<15°:

⎟ ⎟ ⎞ ⎜ ⎜ ⎛ ⎟ ⎞ ⎜ ⎛ Θ − Θ = ln 1 E

IPF Map

• riginal cube

band

0,2 0,4 0,6

grain boundary energy

0°<Θ<15°: Θ>15°:

⎟ ⎟ ⎠ ⎜ ⎜ ⎝ ⎟ ⎠ ⎜ ⎝ − = 15 ln 1 15 E

1 = E

grown cube nucleus

• 10

10 20 30 40 50 60 70 80 0,0

m isorientation (°) m obility

M(Θ)

0,6 0,8 1,0

a.u.)

1 ⎤ ⎡ ⎞ ⎛

M(Θ)

4 µm

0 0 0,2 0,4

mobility (a

( ) ( ) ( ) [ ]

1 15 2 exp 1 1 9 , + ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − Θ ⋅ − + = Θ m

• 10

10 20 30 40 50 60 70 80 0,0

m isorientation (°)

Stored energy according to

18

Stored energy according to Read-Shockley approach

KAM Map (“stored energy”)

Max-Planck-Institute for Iron Research, Düsseldorf

Evolution of stored energy from MC simulations

• S. Zaefferer: 3D orientation microscopy

4 00E 06 4.50E+06 3 00E 06 3.50E+06 4.00E+06

non-cube grains cube grains

Stored energy

2 00E 06 2.50E+06 3.00E+06 rgy [J/m³] .

g difference nc-n

difference between cube and non-cube grains persists even

1 00E+06 1.50E+06 2.00E+06 Ener

grains persists even after longer annealing periods

0 00E+00 5.00E+05 1.00E+06

periods Growth advantage of

0.00E+00 200 400 600 800 MCS

the cube grains

19

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

p y tomography

• Application examples

Application examples

– Grain boundary characterization – Accurate observation of deformation structures Accurate observation of deformation structures – Microstructure characterization for 3D modelling

• Material restrictions:
• Material restrictions:

– beam induced material changes

C l i

20

• Conclusions

Max-Planck-Institute for Iron Research, Düsseldorf

Materials restrictions of FIB-milling

• S. Zaefferer: 3D orientation microscopy
• Anisotropic sputtering and curtaining

Amorphisation (beam damage)

• Amorphisation (beam damage)
• FIB-induced phase transformation
• Reaction between gallium and aluminium

damage due to Ga-Al interaction at grain

21

interaction at grain boundaries under a nano- indentation in Al

Max-Planck-Institute for Iron Research, Düsseldorf

Anisotropic sputtering & curtaining

F 3% i ll ll hi

• S. Zaefferer: 3D orientation microscopy

Fe 3% Si alloy: crystallographic

• rigin of anisotropic sputtering

Experiments and calculations on anisotropic sputtering of Cu

B.W. Kempshall et al., p , J.Vac.Sci.Tech. B19 (2001), 749

low resistance against sputtering against sputtering higher resistance against sputtering

22

easy sputtering: {100} crystal planes hard sputtering: {111} crystal planes

Max-Planck-Institute for Iron Research, Düsseldorf

Amorphisation of lattice structure (beam damage)

• S. Zaefferer: 3D orientation microscopy
• Investigation of Kato et al., J.Vac.Sci.Tech. A17(1999), 1201:

– 20 nm side-wall amorphisation after 30 keV milling on p g silicon – 8 nm after 10 keV milling amorphisation depth is proportional to ion energy amorphisation depends on Z of target material p p g

milling:

Fe Al matrix:

milling: 30 keV, 500 pA

Fe3Al matrix: excellent diffraction patterns Laves phase inclusion: complete

23

p amorphisation

Max-Planck-Institute for Iron Research, Düsseldorf

FIB-beam-induced material changes

• S. Zaefferer: 3D orientation microscopy

Beam-induced α-γ h t f ti

g g

transformation

• f metastable

phase transformation in Fe-Ni

formation of a thin a-

3mm

a

3mm

a g

austenite into

martensite

during milling

formation of a thin a layer

g g

3D orientation microscopy

• n a TRIP steel:
• n a TRIP steel:
• no residual austenite left
• “bainitic” orientation gradients

preserved preserved

24

• rientation deviation 0°…20°

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

tomography tomography

• Application examples

– Grain boundary characterization – Accurate observation of deformation structures – Microstructure characterization for 3D modelling

• Material restrictions:

– beam induced material changes

• Measurement accuracy

25

Measurement accuracy

• Conclusions

Max-Planck-Institute for Iron Research, Düsseldorf

Problems of section alignment

• S. Zaefferer: 3D orientation microscopy

Crystallographic interface analysis of martensite plates R h l S i M 55 (2006) 11 16 Cube nucleus in Fe Ni

26

Rowenhorst et al. Scripta Mater. 55 (2006) 11–16

Bad section alignment leads to significant errors in plane determination

Max-Planck-Institute for Iron Research, Düsseldorf

• S. Zaefferer: 3D orientation microscopy

Software-based improvement of resolution

• Two sources of inaccuracy for the tilt set-up:

– Tilt inaccuracies: linear expansion of measurement p f m m field Δα ≤ 1° Δl/l ≈ 1 % ( on a 10 µm field: 100 nm) – Shift inaccuracies: translations of measurement field Sh ft naccurac es translat ons of measurement f eld usually in the order of 1 image pixel ≈ 50 nm

• Correction of inaccuracies:

Correction of inaccuracies:

– Tilt: measurement of Δα by measurement of average misorientation between slices misorientation between slices correction by linear image distortion Shift: minimization of Euler angle correlation – Shift: minimization of Euler angle correlation coefficient between successive slices

!

∑

27

min ) , , ( ) , , (

, , , 2 1 , , 2 1

= − =∑

+ + j i j y i x w y x t ij

C ϕ φ ϕ ϕ φ ϕ

Max-Planck-Institute for Iron Research, Düsseldorf

Advanced section alignment

• S. Zaefferer: 3D orientation microscopy
• Non-systematic voxel shifts due to

inaccurate beam movement

– leads to locally changing misalignment of slices

• Approach:

pp

– short Monte-Carlo grain growth process – MC termination condition: MC termination condition

me grain volu constant =

g

ds dV surface grain shrinking <

g mc

d dA ds

• conserving main grain shape and size, reducing

i b d h

mc

ds

28

grain boundary roughness – conserving the internal structure of grains

Max-Planck-Institute for Iron Research, Düsseldorf

Effect of MC clean-up

• S. Zaefferer: 3D orientation microscopy

Sample: deformed TRIP steel

Raw data MCS = 5

29

effective clean-up: triple line reduction

Max-Planck-Institute for Iron Research, Düsseldorf

MC clean-up: sub-structure conservation

• S. Zaefferer: 3D orientation microscopy

MCS = 5 Raw data

30

KAM values showing the local orientation gradients

Max-Planck-Institute for Iron Research, Düsseldorf

Contents

• S. Zaefferer: 3D orientation microscopy
• The need and methods for 3D

characterization of crystalline matter

• Principle of 3D characterisation by FIB-EBSD

p y tomography

• Application examples

Application examples

– Grain boundary characterization – Accurate observation of deformation structures Accurate observation of deformation structures – Microstructure characterization for 3D modelling

• Material restrictions:
• Material restrictions:

– beam induced material changes

C l i

31

• Conclusions

Max-Planck-Institute for Iron Research, Düsseldorf

Conclusions I: general features

• S. Zaefferer: 3D orientation microscopy
• A multi-dimensional microstructure vector is
• btained at each 3D spatial position

p p

– phase, orientation, defect density, elemental composition

• Spatial resolution: 50 x 50 x 50 nm3
• Observable volume: ≈ 50 x 50 x 50 µm3

Observable volume: ≈ 50 x 50 x 50 µm

• Angular resolution: 0.5° (precision of tilt)

Ti ti t 15 60 i / l

• Time consumation per cut: 15 … 60 min /cycle

32