[PPT] - Modelling and Simulation of Microalloyed Austenite During Multipass PowerPoint Presentation

SLIDE 1

Charles Hatchett Symposium – 16 July 2014

Modelling and Simulation of Microalloyed Austenite During Multipass Deformation

E.J. Palmiere

Department of Materials Science & Engineering, The University of Sheffield Sir Robert Hadfield Building, Mappin Street, Sheffield, S1 3JD, UK

SLIDE 2

Charles Hatchett Symposium – 16 July 2014

Modelling Strategies

Physical Insight Accuracy

DATA-BASED BLACK BOX e.g. ANNs PROBABILISTIC GREY BOX e.g. CA, MC-P KNOWLEDGE-BASED GREY BOX semi-empirical equations of state KNOWLEDGE + DATA- BASED HYBRID white + grey/black modules KNOWLEDGE-BASED GREY BOX e.g. FE KNOWLEDGE-BASED WHITE BOX constitutive equations

SLIDE 3

Charles Hatchett Symposium – 16 July 2014

Strain path changes non-linearly during TMP of metals.
Microstructure models were developed based on total strain,

particularly at mid-thickness, and did not take the strain path effect into account.

Comparison of the strain path angle evolution through the roll gap, calculated using 2D FE, Al-1%Mn alloy slab subjected to a 50% reduction. A.J. McLaren and C.M. Sellars, Mat. Sci. Tech, 1992, 8, 1090-1094 Centre Surface

20 40 60 80 100

0.2 0.4 0.6 0.8 1

Through Thickness

% Recrystallised

Prediction Measured

Introduction

SLIDE 4

Charles Hatchett Symposium – 16 July 2014

C Si Mn S P Ni Fe Ti N Nb

A

0.08 0.25 1.75 0.013 0.010 30.9 66.8 0.003 0.004

B

0.09 0.27 1.73 0.016 0.010 29.8 67.8 0.003 0.004 0.09

Composition (wt%)

Materials

Two Fe-30 Ni model alloys with different

Nb concentrations

Focusing on alloy B (Nb-bearing)
Large equiaxed grains (> 200 μm) with

annealing twins

2 m

SLIDE 5

Charles Hatchett Symposium – 16 July 2014

Enable for the development of microstructural models of

austenite conditioning (i.e., grain size, substructure, composition, etc.), and its role on subsequent transformation behaviour

Need for a non-

transforming C- Mn microalloyed steel analogue

Why Model Alloys?

SLIDE 6

Charles Hatchett Symposium – 16 July 2014

Equivalent Strain

0.0 0.1 0.2 0.3 0.4 0.5

Equivalent Stress (MPa)

50 100 150 200 250 300 C-Mn-Nb Fe-30Ni-Nb Temperature: 900°C Strain Rate: 10 sec-1 °C

Model Alloy Comparison

SLIDE 7

Charles Hatchett Symposium – 16 July 2014

Method – Simulated Schedule

Three different torsion schedules to study the

effect of strain path changes With the following passes: [100F] = No strain path change, 0.1 Forward

Strain path changes

[83F-17R] = 0.083 Forward – 0.017 Reverse [75F-25R] = 0.075 Forward – 0.025 Reverse Each pass has a total strain of 0.1

SLIDE 8

Charles Hatchett Symposium – 16 July 2014

D.R. Barraclough, H.J. Whittaker, K.D. Nair and C.M. Sellars, J. Testing & Evaluation, 1973, 1, 220- 226

SLIDE 9

Charles Hatchett Symposium – 16 July 2014

Temperature (° C) Time

Heat up Hold at 1250°C (Nb in solution) 20s Deformation passes of 0.1 strain Quench

Method – Simulated Schedule

14 passes between 1100°C and 800°C, each pass has a total strain of 0.1 and strain rate of 1 s-1 and a gap of 20 seconds between passes

SLIDE 10

Charles Hatchett Symposium – 16 July 2014

Simulated Schedule Results

SLIDE 11

Charles Hatchett Symposium – 16 July 2014

Simulated Schedule Results

SLIDE 12

Charles Hatchett Symposium – 16 July 2014

Simulated Schedule Results

SLIDE 13

Charles Hatchett Symposium – 16 July 2014

CCR Rolling

E.J. Palmiere, C.I. Garcia and A.J. DeArdo , Metallurgical Transactions A, 1996, 27, 951-960

Recrystallisation-Stop Temperature

Hot rolling schedules occur over many

passes at different temperatures

Rolling schedules designed to avoid

passes occurring in partial recrystallisation region (T5%<T<T95%)

Hence, determining T5% is important

SLIDE 14

Charles Hatchett Symposium – 16 July 2014

T5%

Increasing amount of effective strain per pass and/or decreasing solute supersaturation

Flow Stress Behaviour

Recrystallisation, Recovery & Work Hardening Recovery & Work Hardening

SLIDE 15

Charles Hatchett Symposium – 16 July 2014

A-100F A-83F-17R A-75F-25R

T5% increases

T5% T95%

DEFORMATION TEMPERATURE

εpas

s

STRAIN

εpas

s

Flow Stress Behaviour (without Nb)

Increasing amounts of reversal (with same total strain)

reduces the effective strain

Increasing amounts of reversal reduces driving force

for recrystallisation

Reducing amount of recrystallisation
Increasing T5%

910°C 940°C 960°C

SLIDE 16

Charles Hatchett Symposium – 16 July 2014

CA Model - Recrystallisation

1100°C – 0 sec 1095°C – 5 sec 1090°C – 10 sec 1085°C – 15 sec 1080°C – 20 sec

0.1F-0.1R 0.1F-0.1F

SLIDE 17

Charles Hatchett Symposium – 16 July 2014 990C 978C 963C

T5% falls

T5% T5%

εpas

s

STRAIN

εpas

s

B-100F B-83F-17R B-75F-25R

DEFORMATION TEMPERATURE

Flow Stress Behaviour (with Nb)

For sample with Nb, increased strain reversal influences two

factors:

Reduces driving force for recrystallisation (as with sample A)
Reduces formation of strain induced precipitates (SIPs)
SIPs inhibit grain boundary mobility and recrystallisation
The fall in T5% shows it is controlled by SIPs for sample B

SLIDE 18

Charles Hatchett Symposium – 16 July 2014

Precipitates are extremely important to properties of

microalloyed steels

Grain refinement
Precipitation hardening
Strain Induced Precipitates form on subgrain boundaries
Size and density important paramaters in how they

influence behaviour

Precipitates have been analysed based on extraction

replicas

Precipitation

110nm

B75F-25R- (thin foil- left, replica- right

SLIDE 19

Charles Hatchett Symposium – 16 July 2014

Precipitation

B75F-25R

215nm

B100

All large precipitates

measured (>100nm) by EDX consist of Ti and Nb

All precipitates measured

are FCC with a≈4.5Å consistent of NbC

Density of precipitates and

size distributions are significantly different between reverse and non- reversed samples

SLIDE 20

Charles Hatchett Symposium – 16 July 2014

Small Precipitates

Sample Size nm – mean (stdev) Volume Fraction B100 10.6 (3) 0.008 B75 9.5 (4) 0.003

Similar size values
Volume fraction significantly

larger in the sample not reversed:

less regions where they are

found

lower density in those

regions

SLIDE 21

Charles Hatchett Symposium – 16 July 2014 z y x 100 nm 250 nm

Precipitation Model

SLIDE 22

Charles Hatchett Symposium – 16 July 2014

Precipitation Model

SLIDE 23

Charles Hatchett Symposium – 16 July 2014

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Czt/Co 0.0 0.2 0.4 0.6 0.8 1.0 z/w 0.0 0.2 0.4 0.6 0.8 1.0

1 s 15 s 50 s 30 s 100 s 200 s 500 s 5 s

Precipitation Model

Dependence of concentration of niobium as a function

f fractional distance

from first generation microbands on time of holding at 950°C after single pass deformation leading to w = 720 nm

SLIDE 24

Charles Hatchett Symposium – 16 July 2014

Time (s) 20 40 60 80 100 N/No 0.0 0.2 0.4 0.6 0.8 1.0    Nucleation Time (s) 20 40 60 80 100 N/No 0.0 0.2 0.4 0.6 0.8 1.0    Nucleation Coarsening Time (s) 20 40 60 80 100 N/No 0.0 0.2 0.4 0.6 0.8 1.0    Nucleation Coarsening Time (s) 20 40 60 80 100 N/No 0.0 0.2 0.4 0.6 0.8 1.0    Nucleation Coarsening Time (s) 20 40 60 80 100 N/No 0.0 0.2 0.4 0.6 0.8 1.0    Nucleation Coarsening

Change in particle density with time on generation

ne

microbands after pass

ne, on generation two

microbands after pass two and on generation three microbands after pass three at 950ºC

Precipitation Model

SLIDE 25

Charles Hatchett Symposium – 16 July 2014

Precipitation Model

SLIDE 26

Charles Hatchett Symposium – 16 July 2014

Precipitation Model

SLIDE 27

Charles Hatchett Symposium – 16 July 2014

Precipitation Model

SLIDE 28

Charles Hatchett Symposium – 16 July 2014

Precipitation Model

SLIDE 29

Charles Hatchett Symposium – 16 July 2014

Final Microstructure: Sample A without Nb (low magnification)

Centre strain=0

strain rate=0

1000μm

0.72 Plane

strain=1.4 strain rate=1

A-100F-0R A-75F-25R

SLIDE 30

Charles Hatchett Symposium – 16 July 2014

Final Microstructure: Sample A without Nb

500μm

torsion direction

A83F-17R- Final A75F-25R- Final A100 - Final

Grains are elongated towards shear angle
With reversal compared to without:
Larger grain size
Less misorientation in grains

SLIDE 31

Charles Hatchett Symposium – 16 July 2014

Final Microstructure: Sample B with Nb

500μm

torsion direction

B83F-17R- Final B75F-25R- Final B100 - Final

Grains are elongated towards shear angle
With reversal compared to without:
Larger grain size
Less misorientation in grains

SLIDE 32

Charles Hatchett Symposium – 16 July 2014

Conclusions

Model alloys are essential to our understanding of the

physical metallurgy of austenite

Small amounts of strain reversal, significantly changes:

– T5% values

increasing or decreasing it depending on presence of microalloying

elements – Amount of recrystallisation

Almost halting it at the maximum strain reversal (0.075F + 0.025R)

studied – Strain Induced Precipitation

Reducing amount of SIPs by at least 5 times

– Stored Energy, Grain Shape and Grain Size