Perspectives on Materials Science in 3D and 4D D. Juul Jensen - - PowerPoint PPT Presentation

Perspectives on Materials Science in 3D and 4D

• D. Juul Jensen

Section for Materials Science and Advanced Characterization

DTU Wind Energy, Technical University of Denmark

DTU Wind Energy, Technical University of Denmark

DTU Wind Energy, Technical University of Denmark

3D (x, y, z) and 4D (x, y, z, t) are the ways forward for many types of experiments

DTU Wind Energy, Technical University of Denmark

3D techniques are not new

• Serial sectioning
• Sample dissolution
• X-ray methods
• Neutron diffraction

DTU Wind Energy, Technical University of Denmark

The materials science community has started to realize the need for 3D and 4D results

Advancing existing 3D/4D methods and developing new unique techniques Goals include:

• Much easier/less manpower-requiring operations
• Better spatial resolution
• Non-destructive methods
• Fast measurements

DTU Wind Energy, Technical University of Denmark

Techniques behind the most frequently published 3D/4D results on the micrometer scale

Advanced serial sectioning

• Semi/fully automatic mechanical sectioning
• FIB
• Laser sectioning

X-ray methods

• Tomography
• 3DXRD (monochromatic x-ray beam) in many variants
• Polychromatic x-ray microdiffraction
• Local texture techniques

TEM methods

• Tomography
• 3DOmiTEM

DTU Wind Energy, Technical University of Denmark

3DXRD Three Dimensional X-Ray Diffraction from Idea to Implementation

DTU Wind Energy, Technical University of Denmark

3DXRD – from idea to implementation

Dream 1994 To develop a technique allowing fast non-destructive

• rientation measurements in µm-sized local volumes

within the bulk of mm3 – cm3 samples Motivation The need for in-situ studies of local deformation and recrystallization phenomena occurring in the bulk

DTU Wind Energy, Technical University of Denmark

3DXRD – from idea to implementation

Requirements

• High penetration power
• High intensity

Only possible solution was X-rays from powerfull synchrotron sources

DTU Wind Energy, Technical University of Denmark

3DXRD – from idea to implementation

Henning F Poulsen

DTU Wind Energy, Technical University of Denmark

3DXRD – from idea to implementation

DTU Wind Energy, Technical University of Denmark

3DXRD – from idea to implementation

DTU Wind Energy, Technical University of Denmark

Center for Fundamental Research: Metal Structures in 4D, 2001 – 2011, 9 mio Euro

3DXRD – from idea to implementation

DTU Wind Energy, Technical University of Denmark 3DXRD Characterization and modeling

DTU Wind Energy, Technical University of Denmark

3DXRD set-up

Area detector Position and shape Orientation and strain

DTU Wind Energy, Technical University of Denmark

I II III 1 2 3 4 ML WBS LC MB WB A B C 5

DTU Wind Energy, Technical University of Denmark

3D x-ray microscopes now also at APS in USA, SPring 8 in Japan and Hasylab/Desy in Germany. Plus 3DXRD in Shanghai in the near future.

3DXRD – from idea to implementation

First version of 3DXRD at ESRF commissioned during the summer of 1999 Today: ESRF: 3D x-ray nanoscope

DTU Wind Energy, Technical University of Denmark

3DXRD (at ESRF) specifications

Experimental conditions

– Energy Range 50 – 100 keV – Flux 1011 – 1012 p/s

Measurements of:

– Position and volume – Crystallographic orientation – Elastic and plastic strain – 3D shape – Dynamics

Spatial resolution

– Mapping precision 500nm x 500nm x 1000nm –

• Min. size 30nm (no mapping)

3DXRD Characterization and modeling

DTU Wind Energy, Technical University of Denmark

Three 3DXRD modes of operation with different time resolutions

Temporal: msec –sec 30 sec 1 hour

?

DTU Wind Energy, Technical University of Denmark

100 μm 1277 grains 539 grains

E.M. Lauridsen and S.O. Poulsen

DTU Wind Energy, Technical University of Denmark

3DXRD measurements Metallurgy:

• 1. Grain orientation rotations during plastic strain
• 2. Plastic deformation/ Plastic strain
• 3. Recrystallization
• 4. Grain growth
• 5. Elastic strains in individual grains
• 6. Phase transformations
• 7. Relations to mechanical properties
• Structural biology
• Phamaceuticals
• Photochemistry
• Geology

DTU Wind Energy, Technical University of Denmark

Examples of key experiments done by 3DXRD or the related DCT technique

DTU Wind Energy, Technical University of Denmark

• A. King, G. Johnson, D. Engelberg, W. Ludwig, and J. Marrow,

Science (2008) 321, 382 - 385

Stress corrosion cracking

DTU Wind Energy, Technical University of Denmark

Construction Steel (0.21%C, 0.51% Mn, 0.20%Si) Cooling: 900°C - 600°C Low temperature: Ferrite + Cementite High temperature: Austenite

S.E. Offerman et al. Science 298 (2002), 1003-1005

Grain nucleation and growth during phase transformation

Work discussed by Aaronson in Scripta Materialia, and later by

• Spanos. See also Sharma

Acta Mater (2012) 229

DTU Wind Energy, Technical University of Denmark

Grain-resolved elastic strains in Cu

• J. Oddershede et al.

Materials Characterization 2011; 62, 651

Cu 50µm initial grain size ~ random texture In situ tensile deformation stress rig driven in position control to a plastic strain

• f 1.5% (measurements under load)

In total 871 bulk grains

DTU Wind Energy, Technical University of Denmark

Results

Accuracy: Center of mass 10 µm Volume 20% rel. error Orientation 0.05º Axial strain 10-4 Strong effect of orientation on elastic strain along tensile direction

DTU Wind Energy, Technical University of Denmark

Plastic deformation - Dislocation structures

• traditional line profiles analysis
• reciprocal space mapping with

high angular resolution 3DXRD

2mm

Al Work lead by Wolfgang Pantleon

DTU Wind Energy, Technical University of Denmark

Zoom Conventional 3DXRD High angular resolution 3DXRD

High angular resolution 3DXRD

DTU Wind Energy, Technical University of Denmark

In-situ XRD investigations with high angular resolution

peak shapes

• of individual grains
• embedded in bulk
• during deformation
• high angular resolution

0.004°=10″=710-5 rad

• B. Jakobsen, H.F. Poulsen, U. Lienert, J. Almer,

S.D. Shastri, H.O. Sørensen, C. Gundlach,

• W. Pantleon, Science 312 (2006) 889

DTU Wind Energy, Technical University of Denmark

• B. Jakobsen, H.F

. Poulsen, U. Lienert, J. Almer, S.D. Shastri, H.O. Sørensen, C. Gundlach, W. Pantleon Science 312 (2006) 889-892

= +

High angular resolution 3DXRD

DTU Wind Energy, Technical University of Denmark

Jakobsen, et al. Scripta Mater. 56 (2007) 769-772

DTU Wind Energy, Technical University of Denmark

Detector horizontally (orientation) Rocking (orientation)

y = 0° a = 1

Dynamics

DTU Wind Energy, Technical University of Denmark

Emergence of subgrains

• uninterupted test
• strain rate
• subgrains emerge

at very low strain

• structures form

during deformation

7 1

6 10 s 

 

 

2mm



DTU Wind Energy, Technical University of Denmark

Recrystallization

Need for 3D and 4D measurements

DTU Wind Energy, Technical University of Denmark

Fundamental questions

• Where do nuclei form?

– And with what orientation?

• How do boundaries move?

– And interact with dislocations?

• Do all nuclei/grains grow with the same speed irrespective of their

crystallographic orientation/misorientation?

• What are the mobilities of a boundary surrounding a recrystallizing grain

and how can it be measured experimentally

• 36

DTU Wind Energy, Technical University of Denmark

Nucleation orientation relationships

The orientation of the nuclei important for texture and microstructure (average grain size after recrystallization)

DTU Wind Energy, Technical University of Denmark

Nucleation orientation relationships

DTU Wind Energy, Technical University of Denmark

Nuclei with new orientations have been reported in literature e.g.

Ardakani MG, Humphreys FJ. Acta metall mater 1994;42:763. Kashihara K, Tagami M, Okada T, Inoko F. Mater Trans JIM 1996;37:572. Okada T, Tagami M, Kashihara K, Inoko F. ISIJ Int 1998;38:518. Okada T, Liu W-Y, Inoko F. Mater Trans JIM 1999;40:586. Kashihara K, Tagami M, Okada T, Inoko F. Mater Sci Engng A 2000;291:207. Okada T, Takechi K, Takenaka U, Liu W-Y, Tagami M, Inoko F. Mater Trans JIM 2000;41:470. Inoko F, Okada T, Tagami M, Kashihara K. In: Hansen N, et al., (Eds.), Proc 21st Risø Int Symp. 2000, p. 365. Huang X, Wert JA, Poulson HF, Krieger Lassen NC, Inoko F. In: Hansen N, et al., (Eds.), Proc 21st Risø Int Symp. 2000, p. 359. Okada T, Ikeda L, Huang X, Wert JA, Kashihara K, Inoko F. Mater Trans 2001;42:1938 Inoko F, Mima G. Scripta metall 1987;21:1039. Inoko F, Fujita T, Akizono K. Scripta metall 1987;21:1399. Inoko F, Kobayashi M, Kawaguchi S. Scripta metall 1987;21:1405. Inoko F, Hama T, Tagumi M, Yoshikawa T. Colloque de Phys C1 1990;51:525. Inoko F, Okada T, Tagami M, Kashihara K. In: Bilde-Sørensen JB, et al., (Eds.), Proc 20th Risø Int Symp. 1999, p. 375. Liu YL, Hu H, Hansen N. Acta metall mater 1995;43:2395. Driver JH, Paul H, Glez J-C, Maurice C. In: Hansen N, et al., (Eds.), Proc 21st Risø Int Symp. 2000, p. 35. Skjervold SR, Ryum N. Acta mater 1996;44:3407.

• T. Okada, X. Huang, K. Kashihara, F. Inoko and J.A. Wert: Acta mater. Vol. 51 (2003), p. 1827.

T.J. Sabin, G. Winther and D. Juul Jensen: Acta Mater. Vol. 51. (2003), p. 3999. H.F. Poulsen, E.M. Lauridsen, S. Schmidt, L. Margulies and J.H. Driver: Acta Mater. Vol. 51 (2003), p. 2517.

DTU Wind Energy, Technical University of Denmark

Nuclei with new orientations?

It is very hard to prove new orientations experimentally Experimental problems: In-situ 2D methods: Surface – nuclei come from below? Static measurements: Lost evidence

DTU Wind Energy, Technical University of Denmark

Nucleation of recrystallization

Pure Al (99.996 %) cr 30 % 3DXRD before annealing 3DXRD after 2 min. at 320 C Aim: Do nuclei with new orientations form? Where do nuclei form in 3D in relation to the deformed microstructure?

DTU Wind Energy, Technical University of Denmark

Detector Images

Detection limits 1.0 – 1.7 mm

DTU Wind Energy, Technical University of Denmark

Sample

DTU Wind Energy, Technical University of Denmark

Nuclei with new orientation

DTU Wind Energy, Technical University of Denmark

DTU Wind Energy, Technical University of Denmark

Result: Nucleation orientation relationships

• Nuclei do form with new orientations
• The rotation between nucleus and parent grain is

related to misorientation axis across def-induced dislocation boundaries in the parent grain Future 4D investigations:

• More ”typical” samples have to be characterized
• Reactions/rotations during recovery (could be 2D)

DTU Wind Energy, Technical University of Denmark

Nucleation Sites

Important for the recrystallized grain size distribution

• Triple junctions
• Grain boundaries
• Large second phase particles
• Deformation induced bands and inhomogeneities
• R.A. Vandermeer and P. Gorden:
• Trans. TMS-AIME, 1959, 215,

577- 588.

This example to illustrate the need for 3D but not 4D characterization

DTU Wind Energy, Technical University of Denmark

Particle Stimulated Nucleation

(early work by Leslie et al. 1963)

Humphreys, Proc 1st Risø Int. Symp. 1980, 35

Orientations of the nuclei formed at particles: random or spread rolling

DTU Wind Energy, Technical University of Denmark

Direct chill-cast Al 3104 cr 80% and annealed to beginning of recrystallization

• Y. Zhang et al. / Scripta Materialia

67 (2012) 320–323

DTU Wind Energy, Technical University of Denmark

Semi automatic serial sectioning

Polished serial sectioning: step size 2 μm Characterization: EBSP and ECC

DTU Wind Energy, Technical University of Denmark

• Y. Zhang et al. / Scripta Materialia 67 (2012) 320–323

Serial sectioning EBSP and ECC characterization of each section

DTU Wind Energy, Technical University of Denmark

Results

• 2D results somewhat misleading (e.g. % nuclei at big particles: 2D=74%

while 3D=90%)

• Preferential nucleation at big particles of both rolling, random and cube
• riented nuclei
• Strong effects of the inhomogeneous distribution of small dispersoids on

the size of the nuclei/recrystallizing grains after some growth

• The very broad grain size distribution can only be understood if both the

specific nucleation site and the local growth possibilities are considered

DTU Wind Energy, Technical University of Denmark

Grain boundary migration – full 4D

In-situ measurements by synchrotron x-rays: 3DXRD of growth during recrystallization in the bulk of deformed single crystals

Schmidt, S., Nielsen, S.F ., Gundlach, G., Margulies, L., Huang, X., Juul Jensen, D., Science, 2004, 229-232.

.

DTU Wind Energy, Technical University of Denmark

Boundary migration away from facets

3DXRD – 40% cr Al (AA1050) single crystal Optical micrograph – 20% cr super pure Al EBSP – 96% cr Ni (99.996%) ECC – 50% cr Al (99.99%) a) b) c) d)

DTU Wind Energy, Technical University of Denmark

DTU Wind Energy, Technical University of Denmark

Curvature driving forces

Fsmax Num. FD=13.6MJ/m3

Zhang YB et al. Computers, Materials and Continua. 2009, 14: p197-207.

DTU Wind Energy, Technical University of Denmark

Revision of equation for boundary migration v=M FD (classic equation) v=M(FD+Fs) (revised equation)

FD: stored energy in the deformed matrix Fs: driving force contribution by local boundary curvature

DTU Wind Energy, Technical University of Denmark

Phase Field Simulations

Moelans N et al. Physical Review

• B. 2013, 88, 054103 1-20 and

Risø 2015

DTU Wind Energy, Technical University of Denmark

Summary: Grain Boundary Migration

Protrusions form primarily on high energy boundaries and relate to local deformation microstructures of high stored energy. Phase field simulations: Protrusions lead to a net positive additional driving force, so boundaries with protrusions move faster than otherwise identical boundaries without protrusions Dream Experiment Follow in-situ in 4D the local grain boundary migration through the deformed microstructure with approx.100nm spatial resolution

DTU Wind Energy, Technical University of Denmark

An experimental way forward: Polychromatic X-Ray Microdiffraction

Technique developed by the Oak Ridge Group lead by G. Ice and implemented at APS synchrotron source High spatial and low temporal resolution compared to 3DXRD

Advanced Photon Source, Argonne National Laboratory, XOR-UNI Beamline 34-ID Sponsored by U.S. DOE Basic Energy Sciences, Division of Materials Sciences

Spatially-Resolved Polychromatic X-ray Microdiffraction

DTU Wind Energy, Technical University of Denmark

White beams give Full Laue diffraction pattern for each submicron length segment

Area detector

Reference: B.C. Larson et al. Nature (2002)

Depth Profiling

Take Laue patterns as beam-blocking Pt wire (50 mm) is translated in small (~1 mm) steps just above sample. Subtract successive pictures. Difference tells where beam came from.

Wire acts as depth-resolving aperture.

DTU Wind Energy, Technical University of Denmark

First results on pure Al cr 86% and partly recrystallized – ex-situ investigation

Zhang ,Juul Jensen et al. to be published.

DTU Wind Energy, Technical University of Denmark

MD modelling by Eric Homer

DTU Wind Energy, Technical University of Denmark

0% 11%

1: Number of grains 2: Degre of deformation:

Limitations of 3DXRD

1 grain 100 grains 10,000 grains

DTU Wind Energy, Technical University of Denmark

NEW IDEA for and recent development of the 3DXRD technique: Dark Field Transmission X-ray Microscopy

Inspired by dark field electron microscopy Aim is to avoid spot overlap and look at one or a few bulk grains individually

DTU Wind Energy, Technical University of Denmark

Dark field imaging and LARGE samples

• Monochr. beam

Detector Grain 1 Grain 2

DTU Wind Energy, Technical University of Denmark

Dark field imaging and LARGE samples

• Monochr. beam

Detector Objective Grain 1 Grain 2

DTU Wind Energy, Technical University of Denmark

Demonstration of dark field imaging

Experimental Sample ID06: 15 keV Si powder

A.King, W. Ludwig, A. Snigirev, I. Snireva, G. Vaughan, H.F. Poulsen Deflector 8 m Sample Lens Spatial resolution 100nm

DTU Wind Energy, Technical University of Denmark

Concluding Remark on non-destructive 3D experimental techniques

Synchrotron sources, spatial resolution 0,1-1um Newest laboratory x- ray tomography, spatial resolution 5um 3D OmiTEM, spatial resolution 2nm

1 3 2

DTU Wind Energy, Technical University of Denmark

Challenges and suggestions for the future success of 3D Materials Science

DTU Wind Energy, Technical University of Denmark

Challenges and suggestions for the future success of 3D Materials Science

Indications of success already exist:

• Papers published proving 2D is not enough
• Some new physical mechanisms have been suggested
• 3D/4D data are used as input/validation for models
• Several of the techniques are in the (++) version
• The 3D/4D community is growing (beyond the groups behind the

techniques) and includes some industries

• Conferences and conference series are held focusing on 3D Materials

Science– eg Risø Symposia and 3DMS

DTU Wind Energy, Technical University of Denmark

Challenges and suggestions for the future success of 3D Materials Science

How do we get over ”the valley of death”?

Is commercialization needed?

DTU Wind Energy, Technical University of Denmark

Challenges and suggestions for the future success of 3D Materials Science

There are many different 3D/4D methods

• Beyond the raw data treatment, analysis and

representation could/should be united

The 3D/4D community is scattered

• Establish firmer networks (technique development,

science-technique discussions, data representation and visualisation)

• Web-based sharing of tools

DTU Wind Energy, Technical University of Denmark

Challenges and suggestions for the future success of 3D Materials Science using large facilities

Some 3D/4D methods only operate at large international facilities - with focus on first of a kind experiments

• Make faster/more standard measurements – block

allocation system

• Always be very well prepared with optimal and well

characterized samples Data analysis may take ”forever”

• Focus on On-line data visualization
• Make the right compromise between technique

advancements and more standard measurements

DTU Wind Energy, Technical University of Denmark

Concluding remarks

3D/4D measurements are essential for some investigations We have to bring 3D/4D beyond the demonstration stage and produce series of important scientific results Ease data analysis Prescribing joint protocols for exchange of data between different platforms Find ways for visualization of 3D/4D data in journals

DTU Wind Energy, Technical University of Denmark

Objective Lens aperture 20 images/second

+ sample tilts (ω)

3D orientation mapping in TEM (3D-OMiTEM)

DTU Wind Energy, Technical University of Denmark

DTU Wind Energy, Technical University of Denmark

For development of 3D-OMiTEM a non-destructive 3D orientation mapping technique with spatial resolution down to 1 nm

Microscopy Today 2012 Innovation Award

DTU Wind Energy, Technical University of Denmark

Extinction spot

monochromatic X-rays sample detector plane

diffraction spot DCT raw data

(1)

• J. Appl. Cryst. (2008). 41, 302-309, W. Ludwig, S. Schmidt, E. M. Lauridsen, H. F. Poulsen

(2)

• J. Appl. Cryst. (2008). 41, 310-318, G. Johnson, A. King, M. G. Honnicke, J. Marrow and W. Ludwig

(3) REVIEW OF SCIENTIFIC INSTRUMENTS (2009), 80, 033905, W. Ludwig, P. Reischig, A. King, M- Herbig, E. M. Lauridsen, G. Johnson, T.J. Marrow and J.Y. Buffiere

Diffraction contrast tomography

DTU Wind Energy, Technical University of Denmark

Kinetics in inhomogeneous microstructures

Copper DPD to  = 2.0 Anneal 1h at 120°C

Lin et al., Acta Mater 2014

DTU Wind Energy, Technical University of Denmark

Large grain-to-grain variations

Copper DPD to  = 2.0

Lin et al., Risø 2012

DTU Wind Energy, Technical University of Denmark

Kinetics in inhomogeneous deformed microstructures

Lin et al., Acta Mater 2014

DTU Wind Energy, Technical University of Denmark

3D X-ray Diffraction (3DXRD) 3D Microscope for in-situ characterization

DTU Wind Energy, Technical University of Denmark

Kinetics in inhomogeneous deformed microstructures

Lin et al., Acta Materialia See also Doherty et al., Risø 1986

DTU Wind Energy, Technical University of Denmark

Summary – Recrystallization kinetics Recrystallization kinetics is strongly affected by:

–Inhomogeneous deformation microstructures –Spatial distribution of nucleation sites –Time dependent and texture dependent growth rates –Each recrystallizing grain has its own kinetics

Wide distribution of apparent activation energies observed experimentally (3DXRD) may serve as basis for advancing modelling