IMST Institute ofM aterials Science & Testing IMST Institute ofM aterials Science & Testing T. Huber and A. Mohammed CompTest2003, Châlons-en-Champagne Jan. 2003 Thermal Expansion Behaviour of Metal-Matrix Composites T. Huber, A. Mohammed and H.P. Degischer Vienna University of Technology nstitute of Materials Science & Testing, A-1040 Vienna/Austri
27
Embed
T. Huber and A. Mohammed CompTest2003, Châlons-en-Champagne Jan. 2003 Thermal Expansion Behaviour of Metal-Matrix Composites T. Huber, A. Mohammed and.
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
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Thermal Expansion Behaviour of Metal-Matrix Composites
T. Huber, A. Mohammed and H.P. Degischer
Vienna University of TechnologyInstitute of Materials Science & Testing, A-1040 Vienna/Austria
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Part I
Thermal Expansion Behaviour of Particle Reinforced
Metal-Matrix Composites
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Metal Matrix Composites Studied – Particle Reinforced
Material P2 (P1 is similar)
MaterialCode Matrix Reinforcement
Fabrication process Condition
P1 Al99.5SiCp; 70 vol.-% particle size:
3m to 80m gas pressure infiltration
as-cast
P2 AlSi7Mg(A356)
overaged
P3 AlSi7Mg(A356)
SiCp; 55 vol.-%particle size: ~80m
centrifugalcasting
overaged
P4 AlSi10Mg(A359)
SiCp; 20 vol.-%particle size: ~12m
stir casting/extruded
overaged
Material P3
Material P4
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Section II • thermal behavior depends strongly on the reinforcement architecture• material P2: Si-bridges between SiC particles forming a percolating SiC-Si-network …• … prevent an accelerated thermal expansion
• linear to decreasing L/L0(T)
• internal stresses in the matrix surpass decreasing yield strength …• … relaxation processes (plastic flow, creep) occur• reduction in thermal expansion vanishing internal stresses• non linear increase of L/L0(T)
• with significant elastic straining• Young‘s modulus decreases with temperature
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
increasing CTEs (slope of L/L0 curve)• significant elastic straining of the matrix
• linear to decreasing L/L0(T) CTE reaches a maximum and drops• thermal expansion is more and more dominated by the densely packed ceramic• simultaneously: closure of microvoids by local plastic flow of the matrix may occur
• material P2: Si-bridges between SiC particles forming a percolating SiC-Si-network …• … prevent an accelerated thermal expansion constant CTEs up to 500°C• material P1: matrix is continuously expanding (non percolating reinforcement) increasing CTE & voids being already filled with metal matrix
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Microstructure of Al-SiC MMC – Material P2 (Section II & III)
• thermal behavior in section III depends strongly on the reinforcement architecture• material 1 & 3: Si-bridges between SiC particles forming a percolating SiC-Si-network …• … prevent an accelerated thermal expansion
SiC
SiCSiC
SiC
SiCSi
Si
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Big SiC particles surrounded by small SiCparticles connected via “Si-bridges” fromthe matrix alloy.
Compact cluster of SiC particles connectedvia “Si-bridges”.
• after dendritic solidification of the matrix alloy (-phase) remaining eutectic liquid (12.6 mass-% Si) tends to freeze around the SiC particulates• ceramic particles do not act as preferential crystal nucleation sites• “Si-bridges” between SiC particles form a percolating “SiC-Si-network”
Microstructure of Al-SiC MMC – Material P2 (Section II & III)
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Preexisting Microvoids – Material P2 (Section II & III)
voids
(a) SEM backscattered electron image (BEI)
(b) Image analysis – binary image of (a)(white area = voids)
Overlay of (a) and (b) average void volume fraction of an AlSiC MMC (result of 12 images)
1.3 0.5 vol.-%
Uncertainties of measurement:• broken SiC particles and removal of matrix alloy during metallographic preparation overestimation• closure of pores by smearing during metallographic preparation underestimation of porosity
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Extruded material• destroyed Si-network• no CTE drop down occurs
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Solution treatment530°C / 6h
Quenched inwater
Slow coolingwith 3K/min
Temp.
TimeRT
Influence of Heat Treatments on the thermal behaviour – Material P2
Heat treatment400°C / 30 min
Quenched inwater
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
0 50 100 150 200 250 300 350 400 450 500 5504
6
8
10
12
14Material P2 - 1st Heating up
Solution HT/cooling(3K/min) to RT
Inst
anta
neou
s C
TE
[ppm
/K]
Temperature [°C]
0 50 100 150 200 250 300 350 400 450 500 5504
6
8
10
12
14Material P2 - 1st Heating up
Inst
anta
neou
s C
TE
[ppm
/K]
Temperature [°C]
0 50 100 150 200 250 300 350 400 450 500 5504
6
8
10
12
14Material P2 - 1st Heating up
Solution HT/cooling(3K/min) to RT Solution HT/quenched in water
Inst
anta
neou
s C
TE
[ppm
/K]
Temperature [°C]
0 50 100 150 200 250 300 350 400 450 500 5504
6
8
10
12
14Material P2 - 1st Heating up
Solution HT/cooling(3K/min) to RT Solution HT/quenched in water 400°C for 30 min/quenched in water
Inst
anta
neou
s C
TE
[ppm
/K]
Temperature [°C]
„Peak“ within 1st heating up cycle Gradual formation of equilibrium phase Mg2Si precipitates (incoherent, overaged)in the A356 alloy, which causes a smalladditional volume change!
Influence of Heat Treatments on the thermal behaviour – Material P2
• Formation of incoherent Mg2Si precipitates 70% of L• Formation of incoherent Mg2Si precipitates 70% of L
• relief of internal stresses from quenching 30% of L
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Conclusions – Particle Reinforced MMC CTE curves of P1 & P2 up to 500°C can be divided into three sections:
section I: matrix with significant elastic straining
section II: elastic straining decreases owing to decreasing yield strength of the matrix internal stresses can at least partly relax by local plastic flow
section III: behavior depends on reinforcement architecture Al99.5 matrix: CTE increases again AlSi7Mg matrix: Si-bridges between SiC
forming a SiC-Si-network prevent an accelerated expansion
Eutectically segregated Si in an Al alloy (e.g. A356) forms a percolating Si network Thermal behaviour of the as-cast alloy is similar to the composite P2 Destruction of the Si network (plastic deformation), lead to a thermal behaviour like matrix with insulated reinforcement (increasing CTEs with rising temperature) These observations on the unreinforced matrix alloy verify the effect of the percolating SiC-Si-network on the thermal expansion of the composite P2
Solution quenched samples of Al-matrix-composites lead to a small additional expansion during a slow 1st heating up cycle, due to formation of overaged incoherent Mg2Si precipitates in the AlSi7Mg matrix alloy & relief of internal stresses from quenching.
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Part II
Thermal Expansion Behaviour of Fiber Reinforced
Metal-Matrix Composites
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
•Irreversible plastic deformation
CFR-MMC’s Thermal Stresses
Temperature, °C
dl/l
%
•Reversible plastic deformation
Width of hysteresis
Residual stresses
•Elastic deformation
Tcr
•Al
l•C
Fiber Matrix
0
Stresses during cooling
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax Mg/C-T300J,80/20
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax Mg/C-T300J,80/20 Mg/C-T300J,50/50
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax Mg/C-T300J,80/20 Mg/C-T300J,50/50 Mg/C-T300J,20/80
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax Mg/C-T300J,80/20 Mg/C-T300J,50/50 Mg/C-T300J,20/80 Mg/C-Tenax,trans
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax Mg/C-T300J,80/20 Mg/C-T300J,50/50 Mg/C-T300J,20/80 Mg/C-Tenax,trans Mg/C-M40B, trans
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
dl/l
%
Temperature, °C
Pure Mg Mg/C-M40B, long Mg/C-Tenax Mg/C-T300J,80/20 Mg/C-T300J,50/50 Mg/C-T300J,20/80 Mg/C-Tenax,trans Mg/C-M40B, trans Mg/C-T300J, out of plane
Thermal Mechanical Analysis (TMA) for CFR-Mg MC’s
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
dl/l
%
Temperature, °C
Pure Al
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
dl/l
%
Temperature, °C
Pure Al 0°
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
dl/l
%
Temperature, °C
Pure Al 0° 45°
-50 0 50 100 150 200-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
dl/l
%
Temperature, °C
Pure Al 0° 45° 90°
Thermal Mechanical Analysis (TMA) for CFR-Al MC’sAl/M40B/74f
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Residual stresses after thermal cycling for CFR-MMC’s
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
0.04
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
Al/M40 Mg/M40 Mg/Tenax Mg/T300,80/20 Mg/T300,50/50 Mg/T300,50/50,out of plane
dl/l
%
max
min
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
0.04
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
120°
C
200°
C
Al/M40 Mg/M40 Mg/Tenax Mg/T300,80/20 Mg/T300,50/50 Mg/T300,50/50,out of plane
dl/l
%
max
min
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Inst.CTE vs Thermal cycling
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/M40
2nd heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
4th heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/Tenax
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
4th heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/Tenax
2nd heating Mg/Tenax
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
4th heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
1st heating Mg/Tenax
2nd heating Mg/Tenax
3rd heating Mg/Tenax
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
4th heating Mg/M40
00.5
11.5
2
2.53
3.54
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K 1st heating Mg/Tenax
2nd heating Mg/Tenax
3rd heating Mg/Tenax
4th heating Mg/Tenax
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
4th heating Mg/M40
0
0.5
1
1.5
2
2.5
3
3.5
4
-50 0 50 100 150 200 250
Temperature, °C
Ins
t.C
TE
, pp
m/K
10
15
20
25
30
35
40
45
50
1st heating Mg/Tenax
2nd heating Mg/Tenax
3rd heating Mg/Tenax
4th heating Mg/Tenax
1st heating Mg/M40
2nd heating Mg/M40
3rd heating Mg/M40
4th heating Mg/M40
1st heating Mg/T300J outof plane
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Inst.CTE vs Temperature for CFR-Mg MC’s
-10123456789
10
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
Mg/M40,0
-10123456789
10
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
Mg/M40,0
Mg/Tenax
-10123456789
10
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
Mg/M40,0
Mg/Tenax
Mg/T300J,80/20
-10123456789
10
-50 0 50 100 150 200 250
Temperature, °C
Ins
t.C
TE
, pp
m/K
Mg/M40,0
Mg/Tenax
Mg/T300J,80/20
Mg/T300J,50/50
-10123456789
10
-50 0 50 100 150 200 250
Temperature, °C
Inst
.CT
E,
pp
m/K
Mg/M40,0
Mg/Tenax
Mg/T300J,80/20
Mg/T300J,50/50
Mg/T300J,20/80
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Al/M40B/74f +
Mg/M40B/70f
Polar diagrams for inst.CTE vs orientation
MgAl0.2/C-T300J/65f (0/90°,50/50%)
MgAl0.2/C-T300J/65f (0/90°,80/20%)
IMSTInstitute of Materials Science & Testing
IMSTInstitute of Materials Science & Testing
T. Huber and A. MohammedCompTest2003, Châlons-en-Champagne Jan. 2003
Conclusions
During thermal cycling, all the CFRM exhibit hysteresis of length change between heating and cooling. The level of plastic deformation is represented by the hysteresis’ width.
Total strain, width of hysteresis are strongly dependent on - the fiber properties - thermal history - fiber orientation and arrangement. Residual stresses causing macroscopic deformation are produced by cooling
from above a certain critical temperature Tcr , which depends strongly on the matrix properties at elevated temperatures.
The plastic deformation and the residual stresses are reproduced by following temperature cycles for the same temperature range.
Transverse fibers of more than 20% increase the longitudinal expansion; low expansion in longitudinal direction is compensated by high expansion transverse and out of plane respectively.
Instantaneous coefficient of thermal expansion (CTE) is not material constant for MMC.