effects of moisture absorption on polymer composites and...
TRANSCRIPT
Effects of Moisture Absorption on Polymer Composites and Adhesives
W R BroughtonNational Physical Laboratory
TMAN EventMeasurement & Control of Moisture in Materials
4th February 2009
Tuesday, 17 March 2009
2
Background
Composites
Adhesive Joints
Conclusions
Content
Tuesday, 17 March 2009
3
Boyne BridgeGRP Cladding
Wind TurbineOff-shore
Chemical Plant + Storage
Courtesy of White Young Green
Chemical + Water TreatmentOver 30 year service
Applications
Tuesday, 17 March 2009
4
Applications
Tuesday, 17 March 2009
5
Growing demand for manufacturers to guarantee product life expectancy
– Particularly where inspection and/or maintenance can be difficult or failure catastrophic
Effective use of composites and adhesives for use in hostile (hot/wet) environments hinges on the availability of validated design data that is applicable for the entire structure service life (often 20-50 years, or longer)
Reliable in-situ, real-time techniques for measuring moisture content and distribution in composite and bonded structures
Requirements
Tuesday, 17 March 2009
6
Osmotic BlisteringGlass Fibre-Reinforced Plastic (GRP)
GRP Boat Hull
Tuesday, 17 March 2009
7
Osmotic BlisteringGlass Fibre-Reinforced Plastic (GRP)
Tuesday, 17 March 2009
8
Osmotic BlisteringGlass Fibre-Reinforced Plastic (GRP)
Water vapour permeates the structure
As water diffuses into the composite it reacts with any hydrolysable components inside the laminate to form tiny cells of concentrated solution
Under this osmotic process, more water is drawn through the semi-permeable membrane of the laminate in an attempt to dilute the solution
The water can increase the fluid pressure of the cell substantially (up to 50 atmospheres), which eventually distorts or bursts the laminate and leads to a blistering of the surface
Damage can be very extensive requiring major repair or the replacement of the structure
Tuesday, 17 March 2009
9
BRE Study
Crazing of surface
Deterioration of gel-coat
Un-even discolouration
Tuesday, 17 March 2009
10
Environmental DegradationPossible Outcomes
Time
Prop
erty
Required Value
acceptable
unacceptable
catastrophic
Tuesday, 17 March 2009
11
Moisture Effects on Thermoset Resins
Swelling, weight gain and cavity formation
PlasticisationReduction in glass transition temperatureLower operating temperaturesReduction in stiffness and strength properties
Chemical LeachingLoss of fillers, catalysts, hardeners, pigments or fire retardants
Tuesday, 17 March 2009
12
Glass Transition TemperatureMoisture Conditioned Polyester Resin
0.0 0.5 1.0 1.5 2.0 2.50
10
20
30
40
50
60
70
25oC immersion
40oC immersion
60oC immersion line of best fitG
lass
Tra
nsiti
on T
empe
ratu
re (o
C)
Moisture Content (%)
Tuesday, 17 March 2009
13
E-Glass (electrical)Relatively low cost (£1-2/kg)Most common form of fibre-reinforcement used for PMCsSensitive to moisture and other chemicals
C and E-CR Glass (chemical)Good chemical resistance to chemical attack
Carbon FibresHigh resistance to corrosion, moisture, creep and fatigueLow production volumes/relatively high price
Aramid FibresSensitive to moisture
Fibre Types
Tuesday, 17 March 2009
14
Leaching of alkali oxides (sodium and potassium oxide) from the fibre surfaceFormation of surface micro-cracks ⇒ stress concentratorsGlass fibres slowly decompose/dissolvePermanent loss of strength (even after drying)
Strength reduces by 20% in de-ionised water at room temperature after 3 weeks exposure
Degradation accelerates as temperature and stress increaseDe-ionised water more aggressive than tap water and saltwater
Moisture Effects on E-glass Fibres
Tuesday, 17 March 2009
15
F922 Epoxy Unidirectional E-glass/F822
Stress Rupture of E-glass Fibres
fUTS
APP tlogk1−=σσ
Tuesday, 17 March 2009
16
Physical EffectsCapillarity effects on fluid uptakeFibre/matrix interfacial breakdownSwelling and weight gainMicro-crackingFibre/matrix interfacial breakdownLoss of fillers, catalysts, hardeners, pigments or fire retardantsOsmotic blistering/cavity formation
Mechanical/Physical Property EffectsReduction in glass transition + operating temperaturesReduction in stiffness and strength propertiesReduction in fatigue performanceShortened life expectancy
Moisture Effects on PMCs
Tuesday, 17 March 2009
17
F922 Epoxy Unidirectional E-glass/F822
Moisture
h
Moisture
y
xz
( ) %100xW
WWMDRY
DRYWET −=
Moisture Diffusion
Tuesday, 17 March 2009
18
F922 Epoxy Unidirectional E-glass/F822
Fickian Diffusion
( )( )
2
12
12
ttMMMh
16D ⎟
⎟⎠
⎞⎜⎜⎝
⎛
−−π
=∞
0 600
4
t11/2 t2
1/2
M2
M1
moisture equilibrium
Moi
stur
e C
onte
nt (w
t %)
Square Root of Time (t)1/2D is Diffusion coefficient
M is moisture content (wt%)
M∞ is equilibrium moisture concentration
t is exposure time
Tuesday, 17 March 2009
19
0 20 40 600
1
2
3
4
moisture equilibrium
Experimental Fickian Curve
Moi
stur
e C
onte
nt (w
t %)
Square Root of Time (hrs)1/2
0 20 40 600.0
0.2
0.4
0.6
0.8
1.0
moisture equilibrium
Experimental Fickian Curve
Moi
stur
e C
onte
nt (w
t %)
Square Root of Time (hrs)1/2
F922 Epoxy Unidirectional E-glass/F822
F922 Epoxy Unidirectional E-glass/F822
Fickian Diffusion
Tuesday, 17 March 2009
20
F922 Epoxy Unidirectional E-glass/F822
Diffusion coefficients parallel and perpendicular to the fibres may be estimated using:
Dm = diffusivity of the matrix, Vf = fibre volume fraction
Unidirectional E-glass/F922
Diffusion Coefficients- Anisotropic Solids
Dm = 8.3 x 10-1 m2/s
D22 = 6.3 x 10-2 m2/s (experimental)
D22 = 7.3 x 10-2 m2/s (predicted)
( )
mf
22
mf11
DV21D
DV1D
⎟⎟⎠
⎞⎜⎜⎝
⎛
π−=
−=
Tuesday, 17 March 2009
21
Moisture Distribution
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
M∞1 year
1 month
1 week
1 day
Moi
stur
e C
onte
nt (w
t %)
Position through plate thickness, h (mm)
⎥⎥⎦
⎤
⎢⎢⎣
⎡⎟⎠⎞
⎜⎝⎛−−=
−−
=∞
75.0
2i
i
hDt3.7exp1
MMMMG
Tuesday, 17 March 2009
22
Moisture Distribution - GRP Laminate (10 mm) Water Immersion @ 23°C
(1) 10 days
(2) 100 days
(3) 500 days
(4) 1000 days
(5) 2500 days
Tuesday, 17 March 2009
23
F922 Epoxy Unidirectional E-glass/F822
0 20 40 60 800.0
0.5
1.0
1.5
2.0
Temperature Effects on Moisture Diffusivity Unidirectional T300/924 Carbon/Epoxy
25 oC 40 oC 60 oC
Moi
stur
e C
onte
nt (w
t %)
Square Root of Time (hrs)1/2
Tuesday, 17 March 2009
24
F922 Epoxy Unidirectional E-glass/F822
T/ko expDD −=
Moisture diffusivity increases with temperature
Do = 0.327 mm2s-1 and k = 4,669 K
3.0 3.1 3.2 3.3 3.41E-8
1E-7
1E-6
Temperature Effects on Moisture Diffusivity Unidirectional T300/924 Carbon/Epoxy
Diff
usiv
ity, D
(mm
2 s-1)
1/T x 103 (1/K)
Tuesday, 17 March 2009
25
F922 Epoxy Unidirectional E-glass/F822
0.0 0.5 1.0 1.50
100
200
300
400
500
Moisture Effects on Tg
T300/924 E-Glass/913G
lass
Tra
nsiti
on T
empe
ratu
re (K
)
Moisture Content (wt%)
Tuesday, 17 March 2009
26
F922 Epoxy Unidirectional E-glass/F822
Tgd is the Tg of the dry material
Tgw is the Tg of conditioned (or wet) material
M is the amount of moisture absorbed (wt %)
g is the temperature shift (in K) per unit moisture absorbed
g is 28.9 K for E-glass/913 and and 36.8 K for T300/924
Corresponding values of Tgd are 430 K and 482 K
gMT = T gdgw −
Moisture Effects on Tg
Tuesday, 17 March 2009
27
F922 Epoxy Unidirectional E-glass/F822
0 10 20 30 40 500
200
400
600
800
1000
1200
1400
Moisture Effects on Tensile StrengthUnidirectional E-glass/913 and E-glass/F922
E-glass/913 E-glass/F922R
esid
ual T
ensi
le S
tren
gth
(MPa
)
Exposure Time (days)
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.40
200
400
600
800
1000
1200
1400
Res
idua
l Ten
sile
Str
engt
h (M
Pa)
Moisture Content (wt%)
Tuesday, 17 March 2009
28
F922 Epoxy Unidirectional E-glass/F822P is material property at the test temperature T
Po is the initial property (un-aged) value of the dry material at room or reference temperature To
n is an empirical constant experimentally derived
n is 1 for E-glass/913 and 0.3 for T300/924
Tgd > To and Tgw > T
Temperature Effects on Transverse Flexure Properties
n
ogd
gw
o TTTT
PP
⎟⎟⎠
⎞⎜⎜⎝
⎛
−−
=
Tuesday, 17 March 2009
29
F922 Epoxy Unidirectional E-glass/F822
t1/2 is time required for the tensile strength to degrade to half its original value (half-life) at each temperature
A and k are material constants determined experimentally
T is the ageing (or conditioning) temperature
Unidirectional E-glass/Polyester
Temperature Effects on Tensile Strength
A = -430 and k = 1.48 x 105
T/A + k = t ln 2/1
Tuesday, 17 March 2009
30
F922 Epoxy Unidirectional E-glass/F822
0.0028 0.0030 0.0032 0.00340
10
20
30
40
50
Tensile Strength Half-Life vs. TemperatureE-glass/Polyester
t 1/2 (d
ays)
1/T (K-1)
Tuesday, 17 March 2009
31
F922 Epoxy Unidirectional E-glass/F822
2.0x10-5 3.0x10-5 4.0x10-5 5.0x10-50
1
2
3
4
Tensile Strength Half-Life vs. DiffusivityE-glass/Polyester
ln t 1/
2 (d
ays)
Transverse Diffusivity (mm2s-1)
Tuesday, 17 March 2009
32
Empirical ApproachNon-Dimensional Temperature
P Material property (e.g. longitudinal tensile strength)
PO Initial property value of the dry material at room or reference temperature To (296 K)
T Test temperature (K)
Tg Glass transition temperature of the material
g Temperature shift (in K) per unit moisture absorbed
M amount of moisture absorbed (wt %)
gMTTwhereTTTT
PP
gdrygwet
n
ogdry
gwet
o
−=⎟⎟⎠
⎞⎜⎜⎝
⎛
−−
=
Tuesday, 17 March 2009
33
F922 Epoxy Unidirectional E-glass/F822
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
Transverse PropertiesUD E-glass/913 Epoxy
modulus strength
P/P o
(Tgw - T)/(Tgd - To)
Tuesday, 17 March 2009
34
F922 Epoxy Unidirectional E-glass/F822
Kitagawa power-law relationship between shear yield stress, τ, and shear modulus, G, for glassy polymers is:
To is the reference temperature (K) – (e.g. 23°C or 296K)
τo is the shear yield stress at To
G is shear modulus
n is an empirical constant experimentally derived
Kitagawa Power-Law RelationshipShear Properties
n
o
o
o
o
TGGT
TT
⎟⎟⎠
⎞⎜⎜⎝
⎛=
ττ
Tuesday, 17 March 2009
35
Shear Modulus Shear Strength
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
AS4/3501-6 XAS/914 T300B/R23 APC-2
P/P o
(Tgd-Topr)/(Tgd-Tr)
0.0 0.2 0.4 0.6 0.8 1.00.0
0.2
0.4
0.6
0.8
1.0
AS4/3501-6 XAS/914 T300B/R23 APC-2
P/P o
(Tgd-Topr)/(Tgd-Tr)
In-Plane Shear PropertiesUnidirectional Laminates
Tuesday, 17 March 2009
36
Kitagawa Power-LawIn-Plane Shear PropertiesUnidirectional CFRP Laminates
0.0 0.1 0.2 0.3 0.40.0
0.1
0.2
0.3
0.4
n = 1
-log
(T0τ
/Tτ 0)
-log (T0G/TG0)
AS4/3501-6 XAS/914 T300B/R23 APC-2
Tuesday, 17 March 2009
37
GRP Pultruded Rods
Fibre Volume Fraction (Vf)Well bonded: 56.2 ± 0.7Poorly bonded: 55.8 ± 0.8
Glass Transition Temperature (Tgdry)Well bonded: 118.2 °CPoorly bonded: 122.2 °C
Tuesday, 17 March 2009
38
GRP Pultruded Rods – Flexure PropertiesMaterial Moisture Content
(%) Flexural Modulus
(GPa) Flexural Strength
(MPa) Dried at 50 °C Well Bonded Poorly Bonded
0.00 0.00
33.8 ± 0.8 30.1 ± 1.1
853 ± 39 371 ± 56
1 Month Well Bonded Poorly Bonded
0.16 ± 0.07 0.27 ± 0.15
36.0 ± 1.1 29.2 ± 0.8
871 ± 61 281 ± 6
3 Months Well Bonded Poorly Bonded
0.27 ± 0.04 0.83 ± 0.22
36.1 ± 1.4 28.3 ± 1.9
866 ± 52 298 ± 31
Flexural stiffness and strength reduced due to poor fibre/matrix interfacial strengthPoorly bonded systems tend to absorb higher levels of moisture – deionised water at 23°C
Tuesday, 17 March 2009
39
Fibre Volume Fraction (%)Well bonded: 56.2 ± 0.7Poorly bonded: 55.8 ± 0.8
Flexural Modulus (GPa)Well bonded: 33.8 ± 0.8Poorly bonded: 30.1 ± 1.1
Flexural Strength (MPa)Well bonded: 853 ± 39Poorly bonded: 371 ± 56
Interlaminar Shear Strength (MPa) – core materialWell bonded: 54.6 ± 3.1Poorly bonded: 20.0 ± 1.4
Moisture Content (%) – 3 months in deionised water at 23°CWell bonded: 0.27 ± 0.04Poorly bonded: 0.83 ± 0.22Poor Good
GRP Pultruded Rods – Test Data
Tuesday, 17 March 2009
40
GRP Pultruded Rods
Poor Good
Blisters or residue sizing droplets on surface of the fibre
Interphase region< 50 nm (well bonded)100-350 nm (poorly bonded)
Tuesday, 17 March 2009
41
5 µm
0 µm
2.5 µm
5 µm0 µm 2.5 µm
0.00 nm
57.33 nm
5 µm
0 µm
2.5 µm
5 µm0 µm 2.5 µm
5.410 V
7.331 V2 µm
0 µm
1 µm
2 µm0 µm 1 µm
5.623 V
7.302 V1 µm
0 µm
0.5 µm
1 µm0 µm 0.5 µm
5.577 V
7.268 V
1 µm
0 µm
0.5 µm
1 µm0 µm 0.5 µm
0.00 nm
24.84 nm
2 µm
0 µm
1 µm
2 µm0 µm 1 µm
0.00 nm
29.99 nm
Topography
Phase Images
5,2 and 1µm Phase Images of a portion of a unidirectional GFRP specimen with poor interfacial bonding
Phase Image of interface region for poorly bonded sample
5µm scan size 2µm scan size 1µm scan size
Glass
Matrix
Glass
Matrix
Tuesday, 17 March 2009
42
Property Predictions - E-glass/Vinylester
Undamaged – core materialE11c = 48.1 ± 1.3 GPaM/41.90 GPaP
S11c(T) = 1216 ± 83 GPaM
G12 = 3.88 GPaP
S13 = 54.6 ± 3.1 MPaM
S11c(F) = 1216 ± 83 GPaM
Degraded by 100% – core material E11c = 43.5 ± 1.3 GPaM/41.87 GPaP
S11c(T) = 540 ± 52 GPaM (0.44S11c(T))G12 = 1.44 GPaP (0.37G12
P)S13 = 20.0 ± 1.4 MPaM (0.37S13)
Measured – m, predicted - p
Tuesday, 17 March 2009
43
Steam Autoclave - 2.2 bar/136 °CMaterial Moisture Content
(wt %) Tensile Strength
(MPa) E-glass/F922 Dry 0.00 1087 ± 29 70 °C/85% RH (6 weeks) 1.00 763 ± 42 Steam autoclave (24 hrs) 2.09 654 ± 22 Steam autoclave (48 hrs) 2.94 579 ± 47 Steam autoclave (72 hrs) 2.83 585 ± 50 HTA/F922 Dry 0.00 1684 ± 132 70 °C/85% RH (6 weeks) 1.06 1728 ± 132 Steam autoclave (24 hrs) 2.12 1691 ± 91 Steam autoclave (48 hrs) 1.90 1725 ± 42 Steam autoclave (72 hrs) 2.20 1784 ± 94
Tuesday, 17 March 2009
44
Single-Lap JointAluminium/Epoxy
Load
Gauge length: 112.5 mm
50 mm 50 mm12.5
0.25 mm1.6 mm
(Width: 25 mm)
Tuesday, 17 March 2009
45
Moisture Effects on Joint StrengthAluminium/Epoxy Single-Lap Joint
0 10 20 30 40 500
50
100
150
200
250
300
350
25 oC
40 oC
50 oC
60 oC
70 oC
Failu
re L
oad
(N/m
m)
Exposure Time (days)
Tuesday, 17 March 2009
46
Effects of Moisture on Elastic PropertiesEpoxy Adhesive (Water Immersion)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0 50 100 150 200 250 300
Conditioning Time (hours)
Moi
stur
e C
onte
nt, M
(%)
AV119 - Water Immersion at 60°C
1500
1700
1900
2100
2300
2500
2700
2900
3100
3300
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Moisture Content (%)
E (G
Pa)
0.30
0.32
0.34
0.36
0.38
0.40
v
Young's Modulus
Poisson's Ratio
Tuesday, 17 March 2009
47
Effects of Moisture on Tensile StrengthEpoxy Adhesive (water Immersion)
Tuesday, 17 March 2009
48
Predicted Moisture Concentration DistributionsAluminium/Epoxy Single-Lap Joint
0
2
4
6
8
10
12
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5
Distance from the edge to the centre of the overlap (mm)
Moi
stur
e C
onte
nt (%
)1 hour water immersion
6 hours water immersion
1 day water immersion
2 days water immersion
5 days water immersion
12 days water immersion
Tuesday, 17 March 2009
49
Predicted Shear Stress DistributionsAluminium/Epoxy Single-Lap Joint
0
2
4
6
8
10
12
14
16
18
20
22
24
0 1.25 2.5 3.75 5 6.25 7.5 8.75 10 11.25 12.5Distance from edge (mm)
S12
Stre
ss (M
Pa)
No conditioning
1 hour water immersion
6 hours water immersion
1 day water immersion
2 days water immersion
5 days water immersion
12 days water immersion
Tuesday, 17 March 2009
50
Predicted Plastic Strain DistributionsAluminium/Epoxy Single-Lap JointImmersed in Deionised Water @ 60°C
Unconditioned 12 days immersion
Tuesday, 17 March 2009
51
Effects of Stress on Moisture AbsorptionEpoxy Adhesive
Tuesday, 17 March 2009
52
Effects of Pressure on Moisture AbsorptionEpoxy Adhesive
Tuesday, 17 March 2009
53
Dielectric measurements to monitor water content in adhesive joints
Optic fibre systems – Embedded Fibre Bragg Gratings for in-situ sensing of chemical species (including water) – composites and adhesive joints
Improved predictive analysis – fracture mechanics based FE modelling incorporating interfacial degradation
Future Trends
Tuesday, 17 March 2009
54
Fibre/matrix interfacial bond strength is critical to structural and long-term performance
Tests exist that can be used differentiate between well and poorly bonded composite systems
Continuous Immersion in Deionised Water
Analytical/semi-empirical relationships - relate degree of material property degradation with level of degrading agent
Enabling extrapolation for longer exposure times under more benign (realistic) conditions
Embedded sensors, such as Fibre Bragg gratings are now being used to determine expansion or swelling due to moisture and chemical species concentrations
Conclusions
Tuesday, 17 March 2009
55
Concern regarding the relationship of the results of accelerated ageing tests for polymeric materials and the actual service performance
Few test methods or standards exist that predict life expectancy with confidence
A major challenge is to ensure that performance testing for determination of chemical resistance (level of degradation) is based upon a set of “quantitatively”measurable criteria
Conclusions
Tuesday, 17 March 2009
56
AcknowledgementsSpecial thanks to:UK Department of Innovation, Universities and Skills National Measurement System Policy UnitJohn Hartley (Exel Composites Ltd)Peter Thornburrow (Owens Corning)NPL Colleagues: Bruce Duncan Sam Gnaniah Maria LodeiroNeil McCartney Gordon Pilkington Graham SimsTim Young
Tuesday, 17 March 2009
57
Thank You
Questions?Comments