Hartman wind snl 2016 advances in composite materials for wind blades

1. August 31, 2016
“Advancements in Composite Materials
for Wind Blades”
Dave Hartman and Tom DeMint
Sandia National Laboratory
2016 Wind Turbine Blade Workshop
Copyright © 2016 Owens Corning. All Rights Reserved

2. Production of energy with no emission of
CO2 (wind, tidal, solar, geothermal)
Providing the basic infrastructure to deliver
clean water to excess of 5-billion people
Providing housing and infrastructure to a growing
population in developing and third-world countries
Reducing the weight of modes of transportation
to respond to increasing cost of energy
CLEAN ENERGY
WATER
INFRASTRUCTURE
URBAN
INFRASTRUCTURE
INDUSTRIAL
LIGHT WEIGHTING
COMPOSITES OPPORTUNITY - GLOBAL MEGATRENDS
2
© iStock pictures

3. DRIVERS FOR COMPOSITES IN THE WIND MARKET
3
 Longer and lighter blades
 Increased blade performance
 Development of low-wind
and off-shore sites
 Cost-of-energy reduction
 Repowering and extension of
service life

4. DRIVERS FOR COMPOSITES IN THE WIND MARKET
Lighter Stiffer
Blade Design
Productivity
Manufacturing
Details
Durability
Service Life
4
Stiffness per $

5. Composite materials
BLADE MATERIAL SELECTION BASED ON STIFFNESS PER $
Composite materials offer the best balance of stiffness and density
Source: Adolphs et.al, “Ultrablade® Fabrics - Reducing the Cost of Wind Energy” SNL 2012 Wind Turbine Blade Workshop

6. COMPOSITE MATERIAL DESIGN TO REDUCE COST PER MWh
Glass fiber reinforcement provides cost advantage for structural composites

7. 0
500
1000
1500
2000
2500
3000
3500
4000
4500
0.0 1.0 2.0 3.0 4.0 5.0 6.0
TensileStress,MPa
Tensile Strain, %
Vintage E-Glass
Advantex®
S-Glass
WindStrand®
Increase in
Modulus
Increase in
Strength
GLASS FIBER EVOLUTION FOR BLADE SPAR CAP
Source: Hartman et.al, “Advances in Blade Design and Material Technology” WindPower 2005 Technical Proceedings
Increasing performance for the Wind Industry with higher modulus glass fiber

8. HIGH PERFORMANCE UD FABRIC MANUFACTURING
Stitching Technology
- Yarn Denier
- Pattern
- Tension, etc
Winding Process
Warp Constructions
- Glass
- Micronage
- Tension, etc
Typical Stitching Patterns
Fabric Characteristics
- Uni-, biaxial, triaxial
- Architecture
- Alignment
- Skewability
- Areal weight
- UD1200
- UD1800
Each UD fabric manufacturing process step is critical to the overall performance

9. 0
200
400
600
800
1000
1200
0 10 20 30 40 50 60
LaminateStrength(MPa)
Laminate Stiffness (GPa)
56%
Vf
53%
Vf
Std. fabrics
Ultrablade® fabrics
Blade Length [m]
Thickness[m]
Z
X
Wind
Blade Length [m]
Thickness[m]
Z
X
Blade Length [m]
Thickness[m]
Blade Length [m]
Thickness[m]
Z
X
Z
X
WindWind
Source: Owens Corning Data, Optimat, Independent test reports
HIGH PERFORMANCE ULTRABLADE® FABRICS FOR SPAR CAP
Ultrablade® fabrics provide significant improvement in modulus, strength
with increased fiber volume and alignment in UD laminates

10. Ultrablade® Fabrics
Triaxial Fabrics
Biaxial Fabrics
Adhesive
FATIGUE PERFORMANCE OF ULTRABLADE® FABRICS
Load simulation identified the fatigue hot-spot for a typical wind blade
Source: Adolphs et.al, “Ultrablade® Fabrics - Reducing the Cost of Wind Energy” SNL 2012 Wind Turbine Blade Workshop

11. FATIGUE CALCULATION ACCORDING MINER’S RULE
DamageD
Mean Stress
[MPa]
Stress Amplitude
[MPa]
Goodman Diagram with Rain Flow count
Goodman Diagram up to 2x106 cycles Calculated Fatigue Damage
FATIGUE DAMAGE ANALYZED FOR IMPROVED DESIGN
Source: Adolphs et.al, “Ultrablade® Fabrics - Reducing the Cost of Wind Energy” SNL 2012 Wind Turbine Blade Workshop
Blade design optimized with weight reduction from improved laminate fatigue life

12. HIGH PERFORMANCE FIBERS AND
FABRICS ENABLE LONGER BLADES
6,05,55,04,54,0
700
650
600
550
500
450
400
350
LOG (N)
PeakStress[MPa]
ADV
H
Fiber
Source: Owens Corning Risoe / DTU tests 2013 on UD laminates, Momentive Epoxy resin L135/H137800750700650600
Advantex® E
Windstrand® H
Fiberglass type
Higher composite stiffness and fatigue performance for
longer blade life reduces the cost of energy
Fatigue Performance at R=0.1, E-glass and H-glass Uni-directional Fabric/epoxy for spar cap
12

13. HIGH PERFORMANCE FIBERS AND
UNI-DIRECTIONAL COMPOSITE PROPERTIES
Property
Test
Method
Unit E-Glass
ECR-Glass
Boron free
H-Glass R-Glass S-Glass Carbon
Fiber and Bulk Glass Properties
Density ASTM C693 g/cm3
2.63-2.64 2,66 2,65 2.55 2,48 1.79
Refractive Index (bulk annealed) ASTM C1648 - 1.562-1.565 1.567 1.558 1.54 1,522
Conductivity ASTM C177 watts/m•K 1.0-1.3 1.22 1.34 6.83
Pristine Fiber Tensile Strength ASTM D2101 MPa 3815-3830 4050 4635 4450-4580 4830-5080 4400
Specific Pristine Strength Calculation × 105
m 1.48-1.49 1.56 1.81 1.74 2.01-2.12 2.46
Young's Modulus GPa 78-79 82 87.5 87 88 230
Specific Modulus Calculation × 106
m 3.05 3.15 3.41 3.48 3.67 12.8
Elongation at Break % 4.8 4.9 4.9 5.35 5.5 1.8
Thermal Properties
Coefficient of Thermal Expansion, 23-300 °C ASTM D696 × 10-6
cm/cm•°C 5.9-6.6 6.6 6.3 4.1 3.4 - 0.6
Specific Heat @ 23 °C ASTM C832 kJ/kg•K 0.807 0.79 0.75 0.810 1.130
Fiber Tensile Strength v. Temperature
Pristine Fiber Tensile Strength, -196 °C ASTM D2101 MPa 5310 5935 7220 7826
Pristine Fiber Tensile Strength, 22 °C ASTM D2101 MPa 3815 4050 4635 4450 5047 4400
Fiber Weight Loss @ 96 °C, 24 hours, 17µm
10% HCl % 31.68 7.88 7.59 1.53 0.05
10% H2SO4 % 32.00 6.91 6.48 1.17
1 N Nitric % 23.47 7.21 6.67 1.42
NaOH pH=12.88 % 5.40 3.24 12.6 19.34 1.10
Impregnated Strand Properties
Tensile Strength ASTM D2343 MPa 2000-2500 2200-2600 2400 -2800 3050-3400 3410-3830 4000
Tensile Modulus ASTM D2343 GPa 78-80 81-83 90 - 91 89-91 86.9-95.8 230
Toughness ASTM D2343 MPa 37 56 69 82-90
Unidirectional Composite Properties1
Tensile Strength ISO 527-5 MPa 1120 1200 1260 1560 1550 1780
Tensile Modulus ISO 527-5 GPa 46 48 52.5 51.6 53 153
Poisson's Ratio ASTM D638 - 0.29 0.33 0.33 0.32 0.27 0.28
Fiber Volume Fraction ASTM D2734 % 60 60 60 60 60 57 2
1
MGS RIM 135 epoxy + RIMH 137 hardener 2
EPON 826 DM HS-Carbon Fiber OC data pub.2011
Glass and Carbon Fiber linear-elastic behavior enables structural composites
when load sharing occurs at the fiber-matrix interphase 13

14. BLADE DESIGN AND MATERIAL INFLUENCE WEIGHT
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
0.000 0.050 0.100 0.150 0.200 0.250 0.300
CalculatedBladeWeight(tonne)
Deflection (mm)
Closed Form Solution Fiber Volume Sensitivity Analysis on Stiffness
Advantex (FVF 53) /Epoxy-5250
Advantex (FVF 57)/Epoxy-5250
Advantex (FVF 60)/Epoxy-5250
H-Glass (FVF 53)/Epoxy-5250
H-Glass (FVF 57)/Epoxy-5250
H-Glass (FVF 60)/Epoxy-5250
S-Glass (FVF 53)/Epoxy-5250
S-Glass (FVF 57)/Epoxy-5250
S-Glass (FVF 60)/Epoxy-5250
T-300 (FVF 53)/Epoxy-5250
T-300 (FVF 57)/Epoxy-5250
T-300 (FVF 60)/Epoxy-5250
Hybrid-Advantex (FVF 60) cover + T-300 (FVF 60) web
Carbon Design
Glass Design
Hybrid Design-
Glass skin, Carbon spar
Thicker Air Foil Design
Glass and carbon fiber linear-elastic behavior enables lighter blades with
load sharing depending on the aero-elastic design
Source: Owens Corning data first order approximation closed form solution of uniform bending load.

15. 15Source: Joseph Cariveau approved reference to LM Website: LM Wind Power layup of LM 88.4 P spar cap, reprinted with permission.
BUILDING THE LONGEST WIND TURBINE BLADE

16.  High fiber modulus and strength for blade root
 Fabric form influences blade root joint design
 Simulation of fiber, fabric, and laminate
property on blade root joint durability
HIGH PERFORMANCE FIBERS AND FABRICS
ENABLE WIND BLADE ROOT JOINT DURABILITY

17. Design simulation predicted higher modulus fiber/fabric reduced the
bearing load transferred to the bolt. The lower axial stress in the bolt
substantially increased the blade joint bolt fatigue life durability.
HIGH PERFORMANCE FIBERS AND FABRICS
ENABLE WIND BLADE ROOT JOINT DURABILITY
®®

18. Simulation of axial stress in the joint tension bolt and laminate load in bearing
assumes good matrix cohesion and adhesion at the fiber-matrix interface
Laminate load sharing in bearing
HIGH PERFORMANCE FIBER-MATRIX INTERPHASE
IMPROVES BLADE ROOT JOINT DURABILITY

19. 19
Acoustic and fracture surface analysis of 45o tension in Advantex® glass/epoxy lamina
show the improved fiber-matrix adhesion leads to a higher transverse strength
Source: Owens Corning WindStrand® fibers and data. Panels dry-wound roving and infused using Momentive epoxy RIMR 135/H137
E-glass UD/epoxy WindStrand® UD/epoxy
Higher composite fiber-matrix adhesion for durability
HIGH PERFORMANCE FIBER-MATRIX
INTERPHASE IMPROVES BLADE DURABILITY

20. THE DESIGN, RELIABILITY AND DURABILITY OF POLYMER
COMPOSITE MATERIALS IS ENABLED BY INTERFACE SCIENCE
Interface science from physical chemical bonding mechanisms to micro-macro
structure-property relationship is required for theoretical and analytical
approaches to mimic composite material performance

21. • Molecular Dynamics predict water molecules break Si-O-Si bonds creating
a high concentration of Si-OH silanol groups on the glass surface
• Surface flaw crack initiation ~1µm fractures at a lower stress in tension
than the glass fiber intrinsic strength, accelerated by high temperature
• Experimental validation: liquid nitrogen immobilizes water and testing
shows up to 35% higher fiber median strength, and
• Vacuo treatment with time reverses water effect on glass which
enables 25% higher fiber median strength
• Silane adsorption bonding glass and adhesion to matrix, protects glass
3500
3700
3900
4100
4300
4500
4700
4900
1 10 100 1000 10000 100000
AverageStress(MPa)
LogTime (Minutes)
Impact on Fiber Strength of HoldingSample In Vacuo
and Testingin Ambient
0
1000
2000
3000
4000
5000
6000
7000
-200 -100 0 100 200 300 400 500 600 700 800
Stress(MPa)
Heat Treatment Temperature C
Strength of FibersTreatedatTemperature and
Tested at RoomTemperature
GLASS COMPOSITION AND SURFACE PROTECTION REDUCE
STRESS CORROSION FROM MOISTURE AND TEMPERATURE
Source: Owens Corning data single fiber testing, glass stress corrosion simulation

22. MACRO COUPON TESTING COMMON FOR DESIGN AND
MANUFACTURING DETAILS, MICRO FOR INTERFACE SIMULATION
Validation of micro-macro property correlation is important
for predicting composite material performance

23. 0
500
1,000
1,500
2,000
2,500
3,000
0 10 20 30 40 50 60
TensileStrength(MPa)
Glass Fiber Manufacturer
MACRO FIBER STRENGTH IMPROVES BY 3X FOR GLASS COMPOSITION, 2X
FOR ALL PROCESS/PRODUCT PARAMETERS,~1.3X FOR FIBER DIAMETER/TEX
1,000
1,200
1,400
1,600
1,800
2,000
2,200
2,400
2,600
2,800
3,000
10 15 20 25 30 35
TensileStrength(MPa)
Fiber Diameter (microns)
1,000
1,200
1,400
1,600
1,800
2,000
2,200
2,400
2,600
2,800
3,000
0 1000 2000 3000 4000 5000 6000 7000 8000
TensileStrength(MPa)
Strand Tex (g/Km)
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0 1 2 3 4 5 6
TensileStrength(MPa)
Tensile Strain (%)
Vintage E-Glass
State-of-Art E-Glass
S-Glass
Source: Hartman et.al, “Advances in Blade Design and Material Technology” WindPower 2005 Technical Proceedings

24. MACRO FIBER-MATRIX ILSS INFLUENCED UP TO 1.4X BY MATRIX, 1.3X
BY FIBER DIAMETER, 1.5X DRY AND 1.5-3X HOT/WET AGED INTERPHASE
Shear Strength in Epoxy - All Products
10
20
30
40
50
60
70
80
0 5 10 15 20 25 30 35
Fiber Diameter (microns)
ShearStrength(MPa)
Wet Shear Strength in Epoxy - All Products
10
20
30
40
50
60
70
80
0 5 10 15 20 25 30 35
Fiber Diameter (microns)
ShearStrength(MPa)
Shear Strength in Polyester - All Products
10
20
30
40
50
60
70
80
0 5 10 15 20 25 30 35
Fiber Diameter (microns)
ShearStrength(MPa)
Wet Shear Strength in Polyester - All Products
10
20
30
40
50
60
70
80
0 5 10 15 20 25 30 35
Fiber Diameter (microns)
ShearStrength(MPa)
Source: Hartman et.al, “Advances in Blade Design and Material Technology” WindPower 2005 Technical Proceedings

25. GOOD COMPOSITE FRACTURE TOUGHNESS WITH HIGH FIBER
STRENGTH, MATRIX MODULUS, AND INTERFACIAL ADHESION
Uni-directional glass fiber-reinforced polymer interlaminar
fracture crack growth correlates to fatigue performance
Source: Hartman et.al, “Advances in Blade Design and Material Technology” WindPower 2005 Technical Proceedings

26. Zangenberg1 et al suggested that glass UD fabric/polyester fatigue failure
mechanisms are analogous to cracking in thin films proposed by Beuth2
1 J. Zangenberg et al, “Fatigue damage propagation in unidirectional glass fiber reinforced composites made of a non-crimp fabric”, Journal of Composite Materials, 2013
2 J. L. Beuth, Jr, “Cracking of thin bonded films in residual tension,” 1992; International Journal of Solids and Structures; Figures used with license.
),,(
2 1
2
h
a
G
E
h


Gss
Gss= Steady state strain energy release rate
G = non-dimensional crack area
ah Crack extension
Interply resin layer, E1
90o weft fibers
Axial fibers, E2, 0o
Crack channeling
Two types of fatigue crack propagation:
• extension
• channeling
FATIGUE CRACK GROWTH CHARACTERIZATION
Depending on interply resin layer thickness, the crack is arrested due to a decrease
in its energy release rate in a compliant material approaching a stiffer material.

27. 0
500
1000
1500
2000
2500
3000
100 150 200 250 300 350 400
Axial Stress, , MPa
StrainEnergyReleaseRate,
GssJ/m2
G1C
If Gss< G1C then there is no crack growth
Steady state strain energy release rate at 50m interply resin layer vs. axial stress
),,(
2 1
2
h
a
G
E
h


Gss
Gss= Steady state strain energy release rate
G = non-dimensional crack area
ah Crack extension
Interply resin layer, E1
90o weft fibers
Axial fibers, E2, 0o
REDUCE FATIGUE CRACK GROWTH WITH
INCREASED INTERPLY TOUGHNESS
One way to reduce fatigue crack growth is to increase interply G1C matrix
critical strain energy release rate or “toughness”

28. Strain Energy Release Rate,G (J/m2) required for crack
growth at  = 200 MPa vs interply resin thickness h, mm
0
200
400
600
800
1000
1200
1400
1600
1800
2000
20 30 40 50 60 70 80 90 100 110 120
StrainEnergyReleaseRate,J/m2
Interply Resin Layer Thickness, h, mm
Increasing the interply resin layer thickness by decreasing FVF, increases
the strain energy release rate needed for cracks to propagate between plies
REDUCE FATIGUE CRACK GROWTH WITH
INCREASED INTERPLY RESIN LAYER THICKNESS
Reduce fatigue crack growth by increasing the interply resin layer thickness with
optimizing the fabric architecture or decreasing the fiber volume fraction

29. NIST AND NW COLLABORATION
ON INTERPHASE CHARACTERIZATION
Determining the relationship between fiber-matrix interphase
and composite part performance enables more effective
development for increasingly robust service life requirements
Hypothesis: a test methodology to characterize fiber-matrix
interfacial performance that will provide insight for:
 reduced crack initiation and propagation rate in fatigue
 higher stress corrosion resistance
 Interface/interphase input for simulation
 consistent robust polymer composites
Test Methodology: determine fiber surface interphase
relationships to mimic improved composite performance
 interfacial shear strength
 fiber fragmentation and critical length measurements
 atomic force microscopy, multi-functional molecular probes
 advanced fluorescence microscopy

30. Thank You !
tom.demint@owenscorning.com
206-325-3513
dave.hartman@owenscorning.com
740-321-5399
Georg Adolphs, Marcus Liu and Richard Veit contributing
QUESTIONS ?