Top Banner
Differential scanning Differential scanning calorimetry calorimetry as a tool for as a tool for analysis of defects in materials analysis of defects in materials Daria Setman 1st BioTiNet Workshop: Advanced Methods for Materials Characterization 23-27 October 2011, Ljubljana-Slovenia, Hotel Slon
67

Differential scanning calorimetry as a tool for analysis of defects in materials

Mar 15, 2022

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Differential scanning calorimetry as a tool for analysis of defects in materials

Differential scanning Differential scanning calorimetrycalorimetry

as a tool for as a tool for analysis of defects in materialsanalysis of defects in materials

Daria Setman

1st BioTiNet

Workshop:Advanced

Methods

for

Materials Characterization

23-27 October

2011, Ljubljana-Slovenia, Hotel Slon

Page 2: Differential scanning calorimetry as a tool for analysis of defects in materials

2

OutlineOutline

Differential Scanning

Calorimetry

Defect

analysis

Activation

Enthalpy

Q by

DSC

More

applications

for

DSC

Page 3: Differential scanning calorimetry as a tool for analysis of defects in materials

3

Differential Differential ScanningScanning

CalorimeterCalorimeter

Perkin Elmer DSC7 Netzsch DSC 204

-

Temperature range: -170°C …

700°C -

Heating rates: 0.1 to 100 K/min

-

Argon atmosphere -

max. sample size: 6 mm diameter,

1.5 mm thickness -

Accuracy: 0.05 J/g, corresponding to

minimum dislocation density: N >

3 x 1014

m-2

minimum vacancy concentration: c > 3 x 10-5

Page 4: Differential scanning calorimetry as a tool for analysis of defects in materials

Design of Design of typicaltypical

powerpower

compensationcompensation differential scanning calorimeterdifferential scanning calorimeter..

Experiments are performed in atmosphere of inert gas to avoid oxidation of samples

Presenter
Presentation Notes
Design of typical power compensation differential scanning calorimeter
Page 5: Differential scanning calorimetry as a tool for analysis of defects in materials

SevereSevere

PlasticPlastic

DeformationDeformation

High Pressure

Torsion HPT

Equal

ChannelAngular

Pressing

ECAP

Presenter
Presentation Notes
Piston – hydrostatic medium – sample pressured thru a die – cooling at entrance In my departmen we are also producing nanomaterials but using hydrostatic extrusion which is also SPD method. In HE method a sample in the container is surrounded with the pressure medium. During extrusion the penetrating seal compresses the pressure medium until the sample starts to extrude- the priston don´t touching a extruded materials. The extrusion started after overflow a volues of critical pressure which is a characteristic value of diffrent materials and depended also of the factor (coefficient) of reduction transver field befor and after extrusion. p=Kc lnR. Material is extruded throu a hole in a die. The fraction is very small becouse of pressure medium and lubricant on the material. The negligible friction provide a minimum of losses(waste) of energy. Temperature in material in die depend of pressure, density and specific heat of extruded material. It is a big problem because of high temperature it is possible of recovery and recrystalization. Because of this dynamic structural process materials could have lower mechanical properties. To avoid this The extrusion products were cooled at the die exit. Due to negligibly small friction, high deformation homogenity and very high strain rate are assured. (strain rate during HE has achieved 2x102s-1 At the beginning HE was primarily successfully applied to the consolidation of nanostructured powders but research show that can be also used to obtain nanostruced metals starting from bulk materials.
Page 6: Differential scanning calorimetry as a tool for analysis of defects in materials

MeltingMelting

of of InIn--samplesample

forfor

calibrationcalibration

140 145 150 155 160 165 170 175Temperature /°C

1

2

3

4

5

6

7

8

9

DSC /(uV/mg)

In - 10 K/min

Onset: 156.3 °C

Peak: 159.0 °C

Area: 114.7 µVs/mg[1

↓ exo

=J/g

Page 7: Differential scanning calorimetry as a tool for analysis of defects in materials

7

DSC curve from deformed Cu

DSC curve from deformed Ni

1 –

vacancies

2 –

vacancy

agglomerates

3 –

dislocations

The energy stored by all defects corresponds to the area integrated

from one single peak emerging during heating

Different types

of defects are represented by separate peaks

HeatHeat

flowflow

curvescurves

forfor

different SPD different SPD nanometalsnanometals

Presenter
Presentation Notes
Heat flow curves for different metals
Page 8: Differential scanning calorimetry as a tool for analysis of defects in materials

8

Data obtained directly from Data obtained directly from DSC DSC --

heat flow curvesheat flow curves

peak temperature activation enthalpy

defect type

peak area energy stored

defect density

Page 9: Differential scanning calorimetry as a tool for analysis of defects in materials

9

Stored energy of dislocationsStored energy of dislocations

( ) ⎟⎠⎞⎜

⎝⎛=

−12 ln4

NbNGbEstor πκ

G…

shear modulusN.. dislocation densityb…

absolute value of the Burgers vector of the dislocations

k… the arithmetic average of 1 and (1-ν) assuming equal parts of edge an screw dislocations

Poisson’s ratio: ν(Ni) = 0.343

Page 10: Differential scanning calorimetry as a tool for analysis of defects in materials

10

Cv = Evac

/ (Qf

⋅ ν ⋅ Na

)

Evac

total energy stored by vacancies

Qf

energy of vacancy formation

ν

amount of substance

(ν=1/M, M-

molar mass)

Na

Avogadro’s number

(Na

=6,023•1023)

Stored energy of vacanciesStored energy of vacancies

Page 11: Differential scanning calorimetry as a tool for analysis of defects in materials

11

DSC DSC --

curves for different heating ratescurves for different heating rates HPT Cu: HPT Cu: γγtt

≅≅

60; 4 60; 4 GPaGPa

139°C

50 K/min183°C

5 K/min

30 K/min169°C

-3.8

-3.3

-2.8

-2.3

-1.8

-1.3

-0.8

-0.3

0.2

50 100 150 200 250 300 350 400 450 500

T (°C)

heat

flow

(W/g

)

Cu42_5KCu42_50KCu42_30K

Page 12: Differential scanning calorimetry as a tool for analysis of defects in materials

12

.1lnmax

2max

constTR

QT

+−=⎟⎟⎠

⎞⎜⎜⎝

⎛ Φ

Φ

-

heating rate

Tmax

-

peak

temperature

R -

gas constant

Q

-

activation enthalpy

TheThe

Kissinger PlotKissinger Plot

[1/K]Cu 4GPa 0.5rot

y = -8900,8x + 12,218R2 = 0,9978

-11

-10,5

-10

-9,5

-9

-8,5

-8

-7,50,00225 0,0023 0,00235 0,0024 0,00245 0,0025 0,00255

1/Tmax

ln( Φ

/Tm

ax2 )

Page 13: Differential scanning calorimetry as a tool for analysis of defects in materials

ResultsResults

forfor

NiNi

Page 14: Differential scanning calorimetry as a tool for analysis of defects in materials

EffectEffect

of of impuritiesimpurities

in Niin Ni HPT HPT deformeddeformed

4 GPa ; 0,5rot ; 4 GPa ; 0,5rot ; εε

= 8= 8

100 150 200 250 300 350 400Temperature /°C

-5

0

5

10

DSC *10-3 /(mW/mg)DSC *10-3 /(mW/mg)

Ni 99.99

Ni 99.998

[1.1] Ni_4_05_10Kref.sdd DSC

[2.1] Ni4_4_05_10_B.sdd DSC

HPT - 4GPa - 0.5rot

Area: -0.04979 J/g

Area: -2.273 J/g

Peak: 223.0 °C

Area: -0.02618 J/g

Area: -1.54 J/gPeak: 257.5 °C

[1.1]

[2.1]

↓ exo

Page 15: Differential scanning calorimetry as a tool for analysis of defects in materials

EffectEffect

of of deformationdeformation

degreedegree

in Niin NiNi 99,998 ; HPT 4GPa ; ε=6 and ε=20

100 150 200 250 300 350Temperature /°C

0

5

10

15

20

25

30

35

40

DSC *10-3 /(mW/mg)DSC *10-3 /(mW/mg)

2 rot.e = 20

0.4 rot.e = 6

Ni 99.998HPT - 4GPa10 K/min

Peak: 222.6 °C

Area: -1.949 J/g

Peak: 235.9 °C

Area: -1.774 J/g

[1.1]

[2.1]

↓ exo

Page 16: Differential scanning calorimetry as a tool for analysis of defects in materials

16

Ni Ni peakpeak

temperaturestemperatures

Page 17: Differential scanning calorimetry as a tool for analysis of defects in materials

Results for CuResults for Cu

Page 18: Differential scanning calorimetry as a tool for analysis of defects in materials

18

Q of Cu, as a function of shear strain Q of Cu, as a function of shear strain and hydrostatic pressureand hydrostatic pressure

for

Cu: Q ≈

0.48 .... 0.78 eV

const

!!

HPT Cu

0,4

0,5

0,6

0,7

0,8

0,9

1

1,1

0 10 20 30 40 50γT

Q (

e V

)

8 GPa2 GPa

Page 19: Differential scanning calorimetry as a tool for analysis of defects in materials

19

Q of Cu, as a function of shear strain Q of Cu, as a function of shear strain and hydrostatic pressureand hydrostatic pressure

for

Cu: Q ≈

0.48 .... 0.78 eV

const

!!

HPT Cu

0,4

0,5

0,6

0,7

0,8

0,9

1

1,1

0 10 20 30 40 50γT

Q (

e V

)

8 GPa2 GPa

tracer

diff.

Page 20: Differential scanning calorimetry as a tool for analysis of defects in materials

Very high defect concentrations in Cu Very high defect concentrations in Cu after ball millingafter ball milling

0

1

2

3

4

5

6

7

8

9

0 5 10 15 20 25 30ball milling time [h]

stor

ed e

nerg

y [J

/g]

C25 SA

1*1017

m-2

5*1016

m-2

1*1016

m-2

in HPT 4*1015 m-2 to 1*1016 m-2

Page 21: Differential scanning calorimetry as a tool for analysis of defects in materials

If we subtract the stored energy of dislocation density measured by XPA from the stored energy measured by DSC and assume that the remaining energy corresponds to the annealing of vacancy type defects the resulting vacancy concentration cv

is:

cv

= 1*10-3 to 3*10-3

Combining DSC with XPA:Combining DSC with XPA:very high very high vacancyvacancy concentrations concentrations

discovereddiscovered

Page 22: Differential scanning calorimetry as a tool for analysis of defects in materials

Thermal Thermal stabilizationstabilization

of of deformationdeformation

inducedinduced

latticelattice

defectsdefects

byby

HydrogenHydrogen

in Pdin Pd

Page 23: Differential scanning calorimetry as a tool for analysis of defects in materials

100 200 300 400 500Tem p eratu re /°C

-4

-3

-2

-1

0

1

2

3

4

5

DS C * 10 -3 /(m W /m g)DS C * 10 -3 /(m W /m g)

Peak: 209 .4 °C

P eak: 396 .6 °C

Area: -1.211 J/gArea: -0 .3529 J /g

Area: -0 .1547 J /g

P eak: 249 .4 °C

[2.1]

[5 ]

↓ exo

DSC of hydrogenated and deformed PdDSC of hydrogenated and deformed Pd

8 GPa 3 rot: ε=38

hydrogenated

without Hydrogen

single vac. H - vac. clusters dislocations

Page 24: Differential scanning calorimetry as a tool for analysis of defects in materials

TEM for peak explanationTEM for peak explanation

100 200 300 400 500Temperature /°C

-4

-3

-2

-1

0

1

2

3

4

5

DSC *10-3 /(mW/mg)DSC *10-3 /(mW/mg)

Peak: 209.4 °C

Peak: 396.6 °C

Area: -1.211 J/gArea: -0.3529 J/g

Area: -0.1547 J/g

Peak: 249.4 °C

[2.1]

[5]

↓ exo

RT 500°C250°C

Page 25: Differential scanning calorimetry as a tool for analysis of defects in materials

Measured vacancy concentrationsMeasured vacancy concentrations

Stored Energy Vac. concentration

Single Vac. CR 0.04 J/g 2.4*10-5

Single Vac. HPT 0.04 J/g 2.4*10-5

HPT ε=38,

with H 1.21 J/g 7.4*10-4

(30 times higher!)

Page 26: Differential scanning calorimetry as a tool for analysis of defects in materials

Measured dislocation densities in HPT metalsMeasured dislocation densities in HPT metals

Stored Energy disl. density

Pd 0.15 J/g 0.9*1015 m-2

Pd hydrogenated 0.35 J/g 1.5*1015 m-2

Ni 2.10 J/g 10 *1015 m-2

Page 27: Differential scanning calorimetry as a tool for analysis of defects in materials

MartensiticMartensitic

transformationtransformation

afterafter

deformationdeformation

in in NiTiNiTi

-50 0 50 100 150Temperatur /°C

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

DSC /(mW/mg)

NiTi

deformed to e=0.77

deformed to e=1.81

[7[1

↓ exo

Page 28: Differential scanning calorimetry as a tool for analysis of defects in materials

Phase shift after ECAP deformation of AM60

Page 29: Differential scanning calorimetry as a tool for analysis of defects in materials

ECA ECA ––

Pressed Pressed ФФ

= 120= 120°°, 10 passes, 10 passes 1 regime: at 1 regime: at 350350°°CC;;

2 regime: at 2 regime: at 210210°°CC;; 3 regime: at 3 regime: at 150150°°CC..

2,33602,62960,4144ECAP at 150oC

0,43950,73200,4148ECAP at 350oC

0,8400--0,7149As-cast

Dissolution of

precipitates

Dislocations annealing, subgrain

growth

Recovery of structure,

dislocations mobility

1,83702,33130,2147ECAP at 210oC

∆H, J/g

Тp

, оС∆H, J/g

Тp

, оС∆H, J/g

Тp

, оС

Peak 3Peak 2Peak 1Condition

DSC scan of the AM60 alloy performed at heating rate of 30 K min-1

Page 30: Differential scanning calorimetry as a tool for analysis of defects in materials

Phase shift performed by ECAP Phase shift performed by ECAP

Magnesium alloy AM60 Mg-6%wtAl-0.13%wtMn

Page 31: Differential scanning calorimetry as a tool for analysis of defects in materials

CristallineCristalline

lamellaelamellae

in in iPPiPP

Page 32: Differential scanning calorimetry as a tool for analysis of defects in materials

⎟⎟⎠

⎞⎜⎜⎝

⎛−Δ

σ=

0m

m

e

TT1h

2l

l=lamella thicknessσe

=surface energy ≈

0.7 J/m2

Δh=heat of melting ≈

200 J/gTm

0=melting temp. ≈

186°C (100% crystalline)

Tm

=melting temperature

-> l ≈

20nm

iPPiPP

––

crystalline lamellae crystalline lamellae embedded in amorphous matrixembedded in amorphous matrix

Presenter
Presentation Notes
l = independent from deformation degree
Page 33: Differential scanning calorimetry as a tool for analysis of defects in materials

lamellalamella

thicknessthickness

in in iPPiPP, , deformationdeformation

dependencedependence

60 80 100 120 140 160 180 200Temperature /°C

0.4

0.5

0.6

0.7

0.8

0.9

DSC /(mW/mg)

[1.2] iPP_B61_70_a.sdd DSC

[2.1] iPP_0_a.sdd DSC

iPP

e = 60

e = 0

Peak: 176.0 °C

Peak: 168.5 °C

Area: 97.49 J/g

Area: 58.16 J/g [1.2]

[2

↓ exo

ε ε

ε

Page 34: Differential scanning calorimetry as a tool for analysis of defects in materials

LimitationsLimitations

of DSC in Mgof DSC in Mg

Page 35: Differential scanning calorimetry as a tool for analysis of defects in materials

Mg Mg ……

nur 9mgnur 9mg

50 100 150 200 250 300 350 400 450Temperatur /°C

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

DSC /(mW/mg)

Mg after HPT9mg

[2]

↓ exo

Page 36: Differential scanning calorimetry as a tool for analysis of defects in materials

50 100 150 200 250 300 350 400 450Temperature /°C

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

DSC /(mW/mg)

[2] subtrHPT_2017P_41s_12.sdd70_HPT_2017P_41s_12.sdd72.mdd DSC

[5] subtr2017P10_1.sdd70_2017P10_2.sdd70.mdd~.mdd DSC

HPT

HE

2017 after HPTshear strain = 1,4

2017 after HEshear strain = 1,4

Area: -4.023 J/g

Peak: 254.1 °C

Peak: 391.6 °C

Area: -20.88 J/g

Peak: 245.8 °C

Peak: 371.5 °C

Peak: 233.6 °C

Area: -5.767 J/gArea: -27.65 J/g

[2]

[5]

↓ exo

HE HE comparedcompared

withwith

HPTHPT

––

manymany

peakspeaks

2017 Al-alloy: HPT & HE ε

=1,4 (Taylor equivalency)

Page 37: Differential scanning calorimetry as a tool for analysis of defects in materials

37

SummarySummary: : activationactivation

enthalpyenthalpy

QQ

DSC is an

easy and

fast technique for measuring defect concentrations

of deformation induced

defects

Also

activation enthalpies correlate well with literature (Molodova 2007, Cao

2008; Jiang

2000, Zhilyaev

2002) and other investigation

techniques like tracer diffusion

DSC can be used for a wide range of investigations (phase transitions, Vacancy-Hydrogen-Clusters, stabilization of defects, …) of Nanomaterials

BUT always use additional methods to verify your results

Page 38: Differential scanning calorimetry as a tool for analysis of defects in materials

38

Thank

You… for

the

invitation

and the

possibility

to talk

about

my

passion!

Page 39: Differential scanning calorimetry as a tool for analysis of defects in materials

HPT Cu: HPT Cu: peakpeak

temperaturetemperature

Cu HPT

100

120

140

160

180

200

220

240

0 10 20 30 40 50γT

T [ °

C ]

2 GPa8 GPa

Page 40: Differential scanning calorimetry as a tool for analysis of defects in materials

tracertracer

diffusiondiffusion

in in ECAPECAP--CuCu

and Cu and Cu alloysalloys

Q= 0.93 eV

Q= 0.73 eV

Results

•~0.73 eV

for non-equilibrium high-angle GBs

(UFG structure)•~0.93 eV

for relaxed high-angle GBs

(CG structure)

•no ultrafast

diffusion (~0.57 eV) in HPT like in ECAP Cu (no porosity !)

Page 41: Differential scanning calorimetry as a tool for analysis of defects in materials

Comparison

of different SPD techniques

Page 42: Differential scanning calorimetry as a tool for analysis of defects in materials

Hydrostatic ExtrusionHydrostatic Extrusion

Technically pure (99,5%) AluminiumInitial sample

Final product – 3 mm wire

Hydrostatic extrusion in three consecutive passes

Presenter
Presentation Notes
In my departmen we are also producing nanomaterials but using hydrostatic extrusion which is also SPD method. In HE method a sample in the container is surrounded with the pressure medium. During extrusion the penetrating seal compresses the pressure medium until the sample starts to extrude- the priston don´t touching a extruded materials. The extrusion started after overflow a volues of critical pressure which is a characteristic value of diffrent materials and depended also of the factor (coefficient) of reduction transver field befor and after extrusion. p=Kc lnR. Material is extruded throu a hole in a die. The fraction is very small becouse of pressure medium and lubricant on the material. The negligible friction provide a minimum of losses(waste) of energy. Temperature in material in die depend of pressure, density and specific heat of extruded material. It is a big problem because of high temperature it is possible of recovery and recrystalization. Because of this dynamic structural process materials could have lower mechanical properties. To avoid this The extrusion products were cooled at the die exit. Due to negligibly small friction, high deformation homogenity and very high strain rate are assured. (strain rate during HE has achieved 2x102s-1 At the beginning HE was primarily successfully applied to the consolidation of nanostructured powders but research show that can be also used to obtain nanostruced metals starting from bulk materials.
Page 43: Differential scanning calorimetry as a tool for analysis of defects in materials

HE HE comparedcompared

withwith

HPTHPT

2017 Al-alloy: HPT & HE ε

=1,4 (Taylor equivalency)

50 100 150 200 250 300 350 400 450Temperature /°C

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

DSC /(mW/mg)

[2] subtrHPT_2017P_41s_12.sdd70_HPT_2017P_41s_12.sdd72.mdd DSC

[5] subtr2017P10_1.sdd70_2017P10_2.sdd70.mdd~.mdd DSC

HPT

HE

2017 after HPTshear strain = 1,4

2017 after HEshear strain = 1,4

Area: -4.023 J/g

Peak: 254.1 °C

Peak: 391.6 °C

Area: -20.88 J/g

Peak: 245.8 °C

Peak: 371.5 °C

Peak: 233.6 °C

Area: -5.767 J/gArea: -27.65 J/g

[2]

[5]

↓ exo

Presenter
Presentation Notes
stored energy in HE higher than HPT
Page 44: Differential scanning calorimetry as a tool for analysis of defects in materials

0 100 200 300 400 500Tem perature /°C

0.00

0.02

0.04

0.06

0.08

0.10

DSC /(m W /m g)

Peak: 164.0 °C

Peak: 79.9 °C

Area: 143.4 J/g

[1.3]

↓ exo

DSC of DSC of hydrogenhydrogen

loadedloaded

undeformedundeformed Pd Pd

Presenter
Presentation Notes
deformed sample is thinner, kinetic higher because of more grain boundaries after HPT
Page 45: Differential scanning calorimetry as a tool for analysis of defects in materials

HydrogenHydrogen

loadedloaded

HPTHPT

deformeddeformed at at --5050°°CC

andand undeformedundeformed PdPd

0 100 200 300 400 500Tem perature /°C

0.00

0.02

0.04

0.06

0.08

0.10

DSC /(m W /m g)

Peak: 164.0 °C

Peak: 79.9 °C

Area: 143.4 J/g

Peak: 127 .5 °C

Peak: 40.8 °C

Area: 140.4 J/g[1.3]

[3.3]

↓ exo

8 GPa 30 rot: ε=380

Presenter
Presentation Notes
deformed sample is thinner, kinetic higher because of more grain boundaries after HPT
Page 46: Differential scanning calorimetry as a tool for analysis of defects in materials

4 4 weeksweeks

annealingannealing

at at roomroom

temperaturetemperature

forfor

HydrogenHydrogen desorptiondesorption

Page 47: Differential scanning calorimetry as a tool for analysis of defects in materials

DSC after HPT deformation at DSC after HPT deformation at --5050°°C and C and RT annealingRT annealing

8 GPa 3 rot: ε=38

hydrogenated

without Hydrogen

100 200 300 400 500Tem p eratu re /°C

-4

-3

-2

-1

0

1

2

3

4

5

DS C * 10 -3 /(m W /m g)DS C * 10 -3 /(m W /m g)

Area: -0 .1547 J /g

P eak: 249 .4 °C

[5]

↓ exo

Page 48: Differential scanning calorimetry as a tool for analysis of defects in materials

motivationmotivation

Fukai: high vacancy concentration up to several % in quenched metals using Hydrogen

trying now to use Hydrogen for formation of deformation induced vacancies

Page 49: Differential scanning calorimetry as a tool for analysis of defects in materials

Material Material usedused•

Pd 99.95% -

from

Alpha Aesar

Hydrogen

loaded

for

12h at 20 bar and 330°C -> x=(H/Pd)=0.6

Presenter
Presentation Notes
stay in the one- phase region, to avoid stresses inside the sample. slow loading. otherwise loading only at the surface, lattice parameter changes on surfaces because of hydrogen -> stress inside sample
Page 50: Differential scanning calorimetry as a tool for analysis of defects in materials

To avoid hydrogen desorption, deformation at -50°C, cooled by solid CO2

pellets

CoolingCooling

withwith

solid COsolid CO22

Page 51: Differential scanning calorimetry as a tool for analysis of defects in materials

51

Q of Ni, as a function of shear strain and Q of Ni, as a function of shear strain and hydrostatic pressurehydrostatic pressure

for

Ni: Q(disl) ≈

const

0.95 eV, Q(vac) ≈

const

0.65 eV

dislocations

& vac.aggl.

vacancies

0

0,2

0,4

0,6

0,8

1

1,2

0,0 5,0 10,0 15,0 20,0 25,0 30,0

99.998 ‐ 2GPa ‐ Vac

99.998 ‐ 2GPa ‐ Disl

99.998 ‐ 4GPa ‐ Vac

99.998 ‐ 4GPa ‐ Disl

99.99 ‐ 4GPa ‐Vac

99.99 ‐ 4GPa ‐Disl

Q(e

V)

γT

Page 52: Differential scanning calorimetry as a tool for analysis of defects in materials

52

The value of the activation enthalpy for the single/double vacancy peak,

Qmig

= 0.65 eV, is about a factor 2 lower than that of vacancy bulk diffusion

Qmig

= 1.18 eV

[1] , which hints at the fact that single/double vacancies

move by core diffusion [2, 3].

In Ni the activation enthalpy Q for the second peak Qmig

= 0.95 eV

is lower

than that given for vacancy bulk diffusion, Qmig

= 1.18 eV

[1]. This indicates

that with the mechanism of dislocation annealing deformation induced

excess vacancies (-agglomerates) are involved.

[1]

H.J. Wollenberger, in R.W. Cahn, P. Haasen

(Eds.), Physical Metallurgy, vol. 9, Elsevier, Amsterdam, 1983,

pp. 1189-1221

[2]

R.F. Canon, J.P. Stark, J. Appl. Phys. 40 (1969) 4366-4373

[3]

K. Sassa, W. Petry, G. Vogl, Phil. Mag. 48 (1983) 41-61.

Page 53: Differential scanning calorimetry as a tool for analysis of defects in materials

53

Reasons for decreasing Q values?Reasons for decreasing Q values?

--

vacancy agglomeratesvacancy agglomerates0

γ

(as evaluated from the difference of apparent dislocation density derived from DSC and the values for true dislocation density from XPA)

HPT Cu 99.99%

Page 54: Differential scanning calorimetry as a tool for analysis of defects in materials

54re s o lv e d s h e a r s tra in γ re s

1 1 0 1 0 0

1 0

2 0

3 0

4 0

5 0

re s o lv e d s h e a r s tra in γ re s1 1 0 1 0 0

1 0

2 0

3 0

4 0

5 0

1 0

2 0

3 0

4 0

5 0

long

rang

e in

tern

al s

tress

es |Δ

τ w -

Δτc| [

MPa

]

1 0

2 0

3 0

4 0

5 0

0 .8 G P a

4 G P a

8 G P a

2 G P a …… internal stressesinternal stresses

(as measured by X-ray Profile Analysis)

Schafler

et al., Mater.Sci.Forum. 503-504 (2006)

Page 55: Differential scanning calorimetry as a tool for analysis of defects in materials

DSC measurements on 2017 AlDSC measurements on 2017 Al--alloyalloy

Defect concentrations in Al-Alloys are higher after HE than after HPT

strain rate during HE has achieved 2x102/s•

interpretation of peaks more complex because of vacancy-trapping by several alloy atoms

Page 56: Differential scanning calorimetry as a tool for analysis of defects in materials

56

vacancies

& clusters

dislo-cations

twins

& stacking

faults

Grains

(size,

distrib.)

internal

stresses

specialabilities

TEM clusters≥

2nmdens.

1014 m-2

+ + + also alloys

MXPA cl. > 5nmconcentr.> 10-5

dens. 1014 -

1018 m-2

+ + + also alloys

DSC concentr.> 10-5

dens. > 1014 m-2

- - + no alloys

but

activ. energy

RER concentr.> 10-8

dens. 1012

-

1019

m-2o <500 µm o no alloys

but

activ. energy

PAS concentr.10-7 - 10-4

dens. 1013

- 1015 m-2

+ - - also alloys,

vac.cluster

analysis

Page 57: Differential scanning calorimetry as a tool for analysis of defects in materials

57

1/Tmax

0.0015 0.0020 0.0025 0.0030

ln(Φ

/Tm

ax2 )

-11.5

-11.0

-10.5

-10.0

-9.5

-9.0

-8.5

-8.0vacanciesdislocations

Q = 0.65 eV

Q = 0.95 eV

(in agreement

with

A.Zhilyaev, M.D.Baro, T.G.Langdon, Mater.Phys.Mech. 5, 23 (2002): 0.93 eV)

vacancies: compare

with

Ni vacancy

bulk

diffusion

Qdiff

= 1.2 eV

core diffusion

dislocations:

compare

with

primary

recrystallization

Qself

= 1.5 eV

presence of vacancy agglomerates confirmed

by

Estor

- measurements

MeasuredMeasured

migrationmigration

enthalpiesenthalpies

Q in Ni 99.99(8)Q in Ni 99.99(8)

Page 58: Differential scanning calorimetry as a tool for analysis of defects in materials

58

shear strain γ0 20 40 60 80 100

(app

.) di

sloc

atio

n de

nsity

[1015

m-2

]

0

2

4

6

8

10

12 DSC 8 GPaRER 2 GPa XPA 2 GPa XPA 4 GPa

Apparent & true dislocation densities from different measuring methods

Conclusion: In Cu exists

a marked

density

of vacancy agglomerates (d ≤

3 nm)

which

is

detected

by

DSC and RER, but

not

by

XPA (only

sensitive to dislocations)

Page 59: Differential scanning calorimetry as a tool for analysis of defects in materials

59

PeakPeak temperaturetemperature

Microhardness

-- recoveryrecovery, , and ...and ...

M.Zehetbauer, D. Trattner

(1987)

Cu – cold rolled

Presenter
Presentation Notes
Recovered structures better for properties, prssure dependent ... Enhanced by pressure
Page 60: Differential scanning calorimetry as a tool for analysis of defects in materials

Advantages of Advantages of nanocrystallinenanocrystalline

metalsmetals

L. T.Klassen

et al., Z. Metallkd., 94, 610 (2003)

nanocrystalline

Presenter
Presentation Notes
high diffusion at grain boundaries
Page 61: Differential scanning calorimetry as a tool for analysis of defects in materials

the

main

peak

temperature

for

Ni and Cu is

T/Tm = 0.30

Page 62: Differential scanning calorimetry as a tool for analysis of defects in materials

Y. Fukai, J. Alloys

Comp. 356-57, 263 (2003)

Page 63: Differential scanning calorimetry as a tool for analysis of defects in materials

Y. Fukai, J. Alloys

Comp. 356-57, 263 (2003)

Page 64: Differential scanning calorimetry as a tool for analysis of defects in materials

NiTiNiTi

-100 -50 0 50 100Temperatur /°C

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

DSC /(mW/mg)

[1.2] NiTi500_2.sdd DSC

[1.3] NiTi500_2.sdd DSC

NiTi 500_2e=0.77

Fläche: 20.89 J/gOnset: 10.8 °C

Peak: 24.4 °C

Onset: -57.3 °COnset: 1.2 °C

Peak: -44.1 °C Peak: 7.0 °C

Fläche: -12.73 J/g Fläche: -6.116 J/g

[1

[1.3]

↓ exo

Page 65: Differential scanning calorimetry as a tool for analysis of defects in materials

Prof. R. Würschum, TU Graz (Austria): Preparation of small samples from large HPT

Cu samples for investigations of dilatation and positron annihilation. In parallel, DSC measurements were carried out for comparison. Common FWF project!

Dr. S. Divinsky

und G. Wilde , Univ. Münster

(Germany): Tracer diffusion experiments in HPT-

deformed Cu. First results were presented in September 2009 on the BNM conference in Ufa (Russia) and published in Advanced Engineering Materials

Georgi

Stefanov

from Institute of Metal Science „Akad. Angel Balevski“-

Bulgarian academy of sciences in Sofia (Bulgaria): Activation enthalpies of precipitation growth in Al-alloys deformed by rolling and folding were studied by DSC. The investigations are published in Journal of Alloys and Compounds

R. Pippan

at ESI-ÖAW Leoben: DSC measurements of several steels (A220, P800, Fe3Si, ...) for investigations of the influence of HPT deformation temperature on the density and stability of deformation induced defects.

B. Verkin

Institute of UAS, Kharkov (Aleksey Podolskiy, E. Tabachnikova): interesting results concerning defect concentrations in wire-drawn and HPT deformed ultrafine grained Zr

Anastasia Sergeeva

and Vladimir Stolyarov

(Mechanical Engineering Research Institute of RAS, Moscow, Russia): DSC measurements in electro-plastically deformed NiTi.

These results were published in Mater. Sci. Forum.

Kinga

Wawer

from Poland: HE•

Olga Kulyasova

& Elena Korznikova

from Russia

national & international national & international cooperationscooperations

Page 66: Differential scanning calorimetry as a tool for analysis of defects in materials

Thesis Thesis workwork

--

publishingpublishing

conferences

(TMS Orlando: Feb. 2007, nanoSPD4: Aug. 2008, TMS San Francisco: Feb. 2009, BNM Ufa: Aug. 2009)

and publications

(4 publ., 2 acc., 1 subm., 1 in prep.)“The Presence and Nature of Vacancy type Defects in SPD Nanometals”

-

Daria

Setman, Erhard Schafler, Elena Korznikova, and Michael J. Zehetbauer, Mater.Sci.Eng.A, 493 (2008) 116–122“Lattice Defects in Hydrogenated and HPT Processed Pd“

Daria

Setman, Maciej

Krystian

and Michael J. Zehetbauer, Mater. Sci. Forum, 584-586 (2008) 355-360“Effect of Electroplastic

Deformation on Martensitic

Transformation in Coarse Grained and Ultrafine Grained Ni-Ti Shape Memory Alloy”

A.E. Sergeeva, D. Setman, M.J. Zehetbauer, V.V. Stolyarov, S.D. Prokoshkin, Mater. Sci. Forum, 584-586 (2008)127-132"Activation Enthalpies of deformation induced lattice defects in SPD nanometals

measured by Differential Scanning Calorimetry" Daria

Setman, Michael B. Kerber, Erhard Schafler, and Michael J. Zehetbauer, Metall. Mater. Trans. A, accepted in press"Formation of superabundant vacancies in nano-Pd–H generated by high-pressure torsion”

M. Krystian, D. Setman, B. Mingler, G. Krexner, M.J. Zehetbauer, Scripta

Mater., 62 (2010) 49-52" Phase evolution in rapidly solidified Al-Fе-V-Si alloys at changes of main components ratio" Georgi

Stefanov, Daria

Setman, Georgi

Avdeev, Petur

Kovachev, Roumen

Petrov, Journal of Alloys and Compounds, submitted"Plasticity and grain boundary diffusion at small grain sizes" G. Wilde, J. Ribbe, G. Reglitz, M. Wegner, H. Rösner, Y. Estrin, M.J. Zehetbauer, D. Setman, S. Divinski, Advanced Engineering Materials, submitted"High pressure torsion induced impacts on the microstructure and physical properties of Pr0.67Fe3CoSb12,

Pr0.71Fe3.5Ni0.5Sb12, and Ba0.06Co4Sb12" L. Zhang, A. Grytsiv, B. Bonarski, M. Kerber, D. Setman, E. Schafler, P. Rogl, E. Bauer, G. Hilscher, M. Zehetbauer, in prep.

Page 67: Differential scanning calorimetry as a tool for analysis of defects in materials

Collaborations with IC members and other groups Collaborations with IC members and other groups of Vienna Universityof Vienna University

iPP

Florian

Spieckermann

(IC): lamella thickness •

HPT Mg -

Bartlomiej

Bonarski

(IC): Mg too light for DSC

FeCu

and NiCo

-

Lidija

Rafailovics

(IC): nano

effects with phase transitions

HPT Co3Sb nano

thermoelectrics

-

Long Zhang (IC): phase transitions, incl. crystallization from amorphous state

Li/NaVycor

-

Abdul

Ghaffar: melting point dep. on nanopore

size•

7CB in porous Si -

Madalina

Puica: