Content uploaded by Amir Badiee
Author content
All content in this area was uploaded by Amir Badiee on Jun 08, 2020
Content may be subject to copyright.
Badiee, Amir (2016) An examination of the response of
ethylene-vinyl acetate film to changes in environmental
conditions. PhD thesis, University of Nottingham.
Access from the University of Nottingham repository:
http://eprints.nottingham.ac.uk/33757/1/Thesis-AB-FFinal.pdf
Copyright and reuse:
The Nottingham ePrints service makes this work by researchers of the University of
Nottingham available open access under the following conditions.
This article is made available under the University of Nottingham End User licence and may
be reused according to the conditions of the licence. For more details see:
http://eprints.nottingham.ac.uk/end_user_agreement.pdf
For more information, please contact eprints@nottingham.ac.uk
An Examination of the Response
of Ethylene-vinyl Acetate Film to
Changes in Environmental
Conditions
Amir Badiee
Thesis submitted to the University of Nottingham
for the Degree of Doctor of Philosophy
2016
I dedicate this to my father and mother for always being there for me.
Abstract
i
Abstract
Photovoltaics are used for the direct conversion of sunlight into electricity. In
order to provide useful power, the individual solar cells must be connected
together. This electrically connected and environmentally protected unit is
termed a photovoltaic (PV) module. The structure of a PV module consists of a
number of layers of various materials with different properties. The
encapsulation material is one of the critical components of a PV module. It
mechanically protects the devices and electrically insulates them, ideally for at
least the 20-25 year lifetime of the modules. The lifetime of a PV module is
generally limited by the degradation of the constituent parts. The materials
degrade and cause a decrease in the efficiency leading to eventual failure,
with the encapsulant being particularly susceptible to degradation. The most
common encapsulant material is Ethylene Vinyl Acetate (EVA) the degradation
of which leads to a significant drop in a PV module’s efficiency, durability and
lifetime. EVA undergoes chemical degradation when it is exposed to
environmental factors such as elevated temperature, humidity and Ultra
Violet (UV) radiation. Although numerous works have been done in this field
there is still a gap in knowledge to fully understand the degradation of EVA
and develop a predictive tool. This work investigates the chemical degradation
of an EVA encapsulant to understand the degradation mechanisms, develop a
predictive model and correlate the degradation with changes in the structure
and mechanical properties.
To determine the effect of environmental stresses on EVA environmental
conditions were simulated in the laboratory in order to accelerate the test
program. The ageing was classified into three main groups, namely thermal
ageing, UV ageing and damp-heat ageing. In order to investigate the effect of
elevated temperature on the mechanical and thermal properties and also to
study the thermal degradation, EVA sheets were aged in a dark laboratory
oven at 85°C for up to 80 days. To investigate the impact of UV exposure on
Abstract
ii
the properties and photodegradation of EVA the samples were exposed to UV
radiation of 0.68 W/m2 at 340 nm and 50°C. To study moisture diffusion and
the impact of absorbed moisture on the mechanical properties and
morphology, EVA sheets were aged in an environmental chamber at 85°C-85%
RH and using a potassium chloride (KCl) salt solution in a sealed chamber to
obtain 85% RH at room temperature (22±3°C).
Thermal analysis techniques including Differential Scanning Calorimetry (DSC),
Thermo-gravimetric Analysis (TGA), Dynamic Mechanical Analysis (DMA) along
with Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy
(ATR-FTIR) and Gravimetrics were used to investigate the structure,
degradation kinetics and viscoelastic mechanical properties of the EVA as a
function of ageing.
The EVA was shown to have viscoelastic properties that were highly sensitive
to the ambient temperature. Thermal ageing was shown to reduce the
storage modulus due to the changes in the structure of the EVA and reduction
in crystallinity. Over a longer time, chemical changes due to thermal activation
also occurred, hence, these were insignificant compared with transient
thermal effects. The activation energy of deacetylation was also shown not be
affected by the ageing process.
Investigation of photodegradation showed notable chemical changes as a
result of UV exposure, with FTIR absorbance peaks related to carboxylic acid,
lactone and vinyl exhibiting a sharp increase after UV irradiation. Differences
in the ATR-FTIR spectra of the UV irradiated and non-irradiated samples
showed that the intensity is depth dependant. DMA results showed UV ageing
had a significant influence on the mechanical properties of the EVA and
reduces the storage modulus. The predictive photodegradation model
showed a good agreement on the UV irradiated surface with the experimental
data where it did not agree well with the results on the non-irradiated side
which could be due to the presence of UV absorber.
Abstract
iii
The response of the EVA to damp heat was investigated at two conditions
with same the RH level (85% RH) and different temperatures (room
temperature and 85°C). The moisture diffusion coefficient was measured via
gravimetry and Water Vapour Transmission Rate (WVTR) technique which
were well-agreed. Results from DSC indicated that the crystallinity increased
due to incorporation of moisture into the structure of the copolymer but
decreased as ageing continued, showing the significant influence of elevated
temperature and thermal degradation on the structure of EVA.
A comparative study of the impact of the ageing on the structure and
mechanical properties indicated that UV has a stronger degrading influence
comparing to other degradation factors. DSC results also suggested that
property changes could be connected to structural modifications. The impact
of different degradation factors can be summarised as UV > T > DH.
Acknowledgements
iv
Acknowledgements
I would like to thank my supervisors Professor Ian Ashcroft and Professor
Ricky Wildman for their guidance throughout the entire research.
I also like to thank the members of Additive Manufacturing and 3D Printing
Research Group and Stability and Performance of Photovoltaic (STAPP)
project for their valuable help and collaboration.
Finally and for most, I would like to thank my family, father, mother, Ali,
Maryam and Manuela for everything.
Table of Contents
v
Table of contents
Abstract _______________________________________________________ i
Acknowledgements _____________________________________________ iv
Table of contents ________________________________________________ v
List of Figures _________________________________________________ ix
List of Tables __________________________________________________ xv
Nomenclature ________________________________________________ xvii
Abbreviations ____________________________________________________ xvii
Alphabetic ______________________________________________________ xviii
Greek symbols ____________________________________________________ xx
Chapter 1 ______________________________________________________ 1
Introduction ____________________________________________________ 1
1.1 Background ____________________________________________________ 1
1.2 Solar PV _______________________________________________________ 1
1.3 Stability and Performance of Photovoltaics (STAPP) project _____________ 5
1.4 Aim __________________________________________________________ 6
1.5 Objectives _____________________________________________________ 6
1.6 Research novelty _______________________________________________ 8
1.7 Thesis structure_________________________________________________ 8
Chapter 2 _____________________________________________________ 10
Literature Review ______________________________________________ 10
2.1 Introduction __________________________________________________ 10
2.2 Photovoltaic (PV) system ________________________________________ 11
2.2.1 Commonly observed PV module degradations in field _____________________ 15
2.3 Ethylene-vinyl Acetate (EVA) _____________________________________ 16
Table of Contents
vi
2.3.1 EVA storing condition and curing process ________________________________ 18
2.4 The effect of environmental stresses on the properties and structure of EVA
________________________________________________________________ 18
2.4.1 The effect of elevated temperature ____________________________________ 19
2.4.2 The UV effect ______________________________________________________ 22
2.4.3 The effect of damp-heat and moisture ingress____________________________ 28
2.4.4 The effect of combined degradation factors _____________________________ 34
2.5 Gaps in the current durability and lifetime studies of PV modules academic
research and industry ______________________________________________ 35
2.5.1 Research challenge 1 ________________________________________________ 36
2.5.2 Research challenge 2 ________________________________________________ 36
2.5.3 Research challenge 3 ________________________________________________ 37
2.6 Conclusion ____________________________________________________ 37
Chapter 3 _____________________________________________________ 38
Experimental Methods __________________________________________ 38
3.1 Introduction __________________________________________________ 38
3.2 Material and storing condition ___________________________________ 40
3.2.1 Material (sample preparation) ________________________________________ 40
3.2.2 Storage of samples __________________________________________________ 47
3.3 Ageing conditions ______________________________________________ 47
3.3.1 Thermal ageing_____________________________________________________ 48
3.3.2 UV ageing _________________________________________________________ 49
3.3.3 Damp-heat ageing __________________________________________________ 49
3.4 Experimental techniques ________________________________________ 51
3.4.1 Differential Scanning Calorimetry (DSC) _________________________________ 52
3.4.2 Dynamic Mechanical Analysis (DMA) ___________________________________ 53
3.4.3 Thermogravimetric Analysis (TGA) _____________________________________ 54
3.4.4 Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy (ATR-FTIR)
______________________________________________________________________ 54
3.5 Moisture diffusion coefficient measurement methods ________________ 56
3.5.1 Gravimetric measurments ____________________________________________ 56
3.5.2 MOCON Water Vapour Transmission Rate technique to measure diffusion
coefficient _____________________________________________________________ 57
Table of Contents
vii
3.5.3 Comparison of moisture diffusion coefficient measurement methods _________ 58
3.6 Summary _____________________________________________________ 59
Chapter 4 _____________________________________________________ 61
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal
Ageing _______________________________________________________ 61
4.1 Introduction __________________________________________________ 61
4.2 Theory of kinetics of degradation _________________________________ 63
4.3 Results and discussion __________________________________________ 66
4.3.1 Thermogravimetric Analysis (TGA) _____________________________________ 66
4.3.2 Dynamic Mechanical Analysis (DMA) ___________________________________ 71
4.3.3 Differential Scanning Calorimetry (DSC) _________________________________ 76
4.4 Conclusions ___________________________________________________ 79
Chapter 5 _____________________________________________________ 80
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
_____________________________________________________________ 80
5.1 Introduction __________________________________________________ 80
5.2 Theory of photochemical reactions ________________________________ 81
5.3 Results and discussion __________________________________________ 84
5.3.1 Fourier Transform Infrared Spectroscopy in Attenuated Total Reflectance (FTIR-
ATR) __________________________________________________________________ 85
5.3.2 Analytical investigation of photodegradation ____________________________ 89
5.3.3 Differential Scanning Calorimetry (DSC) _________________________________ 94
5.3.4 Dynamic Mechanical Analysis (DMA) ___________________________________ 98
5.4 Conclusions __________________________________________________ 103
Chapter 6 ____________________________________________________ 104
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat _____________________________________ 104
6.1 Introduction _________________________________________________ 104
6.2 Analytical Solution of the diffusion equation _______________________ 105
Table of Contents
viii
6.2.1 Fickian Diffusion ___________________________________________________ 105
6.3 Results and discussion _________________________________________ 108
6.3.1 Measurement of moisture diffusion coefficient and predicting moisture
concentration _________________________________________________________ 108
6.3.2 Differential Scanning Calorimetry (DSC) ________________________________ 110
6.3.3 Dynamic Mechanical Analysis (DMA) __________________________________ 115
6.4 Conclusions __________________________________________________ 123
Chapter 7 ____________________________________________________ 125
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions ___________________ 125
7.1 Introduction _________________________________________________ 125
7.2 Comparative study of UV, thermal and damp heat ageing impact on the
properties of EVA ________________________________________________ 126
7.3 Comparative study of influence of the UV thermal and damp-heat ageing
factors on chemical changes in EVA __________________________________ 129
7.3 Conclusion ___________________________________________________ 131
Chapter 8 ____________________________________________________ 132
Conclusion and Recommendations for Future Work __________________ 132
8.1 Conclusions __________________________________________________ 132
8.2 Recommendations for future work _______________________________ 134
References ___________________________________________________ 135
Appendix A __________________________________________________ 153
Water Vapour Transmission Rate Results _____________________________ 153
Appendix B __________________________________________________ 155
Original DSC curves _______________________________________________ 155
Appendix C ___________________________________________________ 158
Publications _____________________________________________________ 158
List of Figures
ix
List of Figures
Figure 1.1: Degradation in materials layers in roof installed PV modules in
Scuola universitaria professionale della Svizzera italiana (SUPSI), Lugano,
Switzerland. (a) Discolouration in EVA. (b) Corrosion in the back sheet. .......... 4
Figure 1.2: Work packages of STAPP project. ..................................................... 5
Figure 1.3: STAPP project program outline for the University of Nottingham. . 6
Figure 1.4: The work flow chart for the experimental programme. .................. 7
Figure 2.1: An illustration of a solar array and its components. ...................... 11
Figure 2.2: Structure of a PV module showing different material layers
(courtesy to Micro System Services (MSS)). ..................................................... 12
Figure 2.3: Chemical structure of EVA copolymer. ........................................... 17
Figure 2.4: Acetic acid formation (deacetylation). ........................................... 19
Figure 2.5: Photodegradation mechanism in EVA (Morlat-Therias et al. 2007).
.......................................................................................................................... 25
Figure 3.1: A general diagram showing the impact of the degradation factors.
.......................................................................................................................... 39
Figure 3.2: A general diagram showing the experimental methodology used in
this research. .................................................................................................... 40
Figure 3.3: Uncured EVA (courtesy to SINOVOLTAICS). ................................... 41
Figure 3.4: Schematic view of the mould used in the hot press. ..................... 42
Figure 3.5: Schematic view of the autoclave. ................................................... 44
Figure 3.6: Autoclave preparation for EVA curing. ........................................... 45
Figure 3.7: (a) Autoclave with samples, spacers and top plate, (b) Autoclave
during EVA curing. ............................................................................................ 45
Figure 3.8: Thermal ageing conditions in the laboratory oven. ....................... 49
List of Figures
x
Figure 3.9: Damp heat ageing conditions in the environmental chamber. ..... 50
Figure 3.10: Ageing conditions in the salt solution chamber. .......................... 50
Figure 3.11: Diagrammatic side view of WVTR test cell (Copyright 2012
MOCON® Inc). ................................................................................................... 58
Figure 4.1: Flowchart of investigation of EVA’s response to thermal ageing. . 62
Figure 4.2: Flowchart of the investigation the weight loss of EVA regarding the
thermal degradation only. ................................................................................ 65
Figure 4.3: TGA thermograms of EVA at different heating rates (weight
percentage versus temperature). ..................................................................... 66
Figure 4.4: Rescaled TGA thermogram of EVA at different heating rates
(weight percentage versus temperature). ....................................................... 67
Figure 4.5: Derivative of weight loss to temperature for unaged EVA at
different heating rates. ..................................................................................... 67
Figure 4.6: Plot of ln(β/Tp2) versus 1/(Tp) for unaged EVA based on Kissinger’s
method. ............................................................................................................ 69
Figure 4.7: Experimental and calculated TGA curves. ...................................... 69
Figure 4.8: Activation energy calculated for unaged and thermally aged EVA.
.......................................................................................................................... 70
Figure 4.9: Weight loss of EVA versus time in the case of exposure to 85°C
with standard deviation of 6.6521e-04. ........................................................... 71
Figure 4.10: Storage modulus vs temperature for aged and unaged EVA. ...... 71
Figure 4.11: (a) tan(δ) vs temperature for aged and unaged EVA, (b) Glass
transition temperature versus ageing duration for EVA based on tan(δ). ...... 72
Figure 4.12: Storage modulus measured at (a)-60˚C, (b) 20˚C (c), 40˚C (d) 95˚C
as a function of ageing time at 85˚C. ................................................................ 74
Figure 4.13: Average storage modulus versus ageing time for thermally aged
EVA. ................................................................................................................... 75
List of Figures
xi
Figure 4.14: Typical DSC thermograms (heat flow versus temperature) of EVA
under three steps, heating-cooling-heating-Exo Up. ....................................... 76
Figure 4.15: DSC first heating thermograms (heat flow versus temperature) for
the unaged and aged EVA (thermograms are reproduced with offsets added)-
Exo Up. .............................................................................................................. 77
Figure 4.16: Glass transition temperature versus ageing duration for EVA..... 77
Figure 4.17: Crystallinity versus ageing duration after first and second heating
for aged and unaged EVA. ................................................................................ 78
Figure 4.18: Storage modulus versus crystallinity for thermally aged EVA. .... 78
Figure 5.1: Flowchart of investigation of EVA’s response to UV ageing. ......... 81
Figure 5.2: Subtracted FTRI-ATR spectra (At - A0) in the domain 1800-1650 cm-
1. ........................................................................................................................ 85
Figure 5.3: Variation of absorbance at 1740 cm-1 as a function of exposure
time. .................................................................................................................. 87
Figure 5.4: Variation of absorbance at 1720 cm-1 as function of exposure time.
.......................................................................................................................... 88
Figure 5.5: Variation of absorbance at 1767 cm-1 as a function of exposure
time. .................................................................................................................. 88
Figure 5.6: Variation of absorbance at 910 cm-1 as a function of exposure time.
.......................................................................................................................... 89
Figure 5.7: Flowchart of the investigation the lifetime of EVA regarding the
photodegradation only. .................................................................................... 90
Figure 5.8: Variation of absorbance at 1740 cm-1 (normalized) on the
irradiated surface as a function of exposure time. .......................................... 91
Figure 5.9: Variation of concentration of ester on the irradiated surface as a
function of exposure time-validation of analytical and experimental results. 92
Figure 5.10: Variation of concentration of ester (photoreactant) on the
irradiated surface as a function of time. .......................................................... 93
List of Figures
xii
Figure 5.11: Variation of concentration of carboxylic acid versus concentration
of ester. ............................................................................................................. 93
Figure 5.12: Variation of concentration of carboxylic acid (photoproduct) on
the irradiated surface as a function of time. .................................................... 94
Figure 5.13: DSC first heating thermograms for the unaged and UV aged EVA-
Exo Up. .............................................................................................................. 95
Figure 5.14: Crystallinity versus ageing time after first and second heating for
unaged and UV aged EVA. ................................................................................ 96
Figure 5.15: Variation of crystallinity versus changes in concentration of ester
on the irradiated surface of UV aged EVA. ....................................................... 97
Figure 5.16: Variation of crystallinity versus changes in concentration of
carboxylic acid on the irradiated surface of UV aged EVA. .............................. 97
Figure 5.17: Glass transition temperature versus ageing for unaged and UV
aged EVA. .......................................................................................................... 98
Figure 5.18: Storage modulus versus temperature for unaged and UV aged
EVA. ................................................................................................................... 98
Figure 5.19: (a) tan(δ) vs temperature for unaged and UV aged EVA, (b) Tg
versus ageing duration based on Figure (5.22: a). ........................................... 99
Figure 5.20: Storage modulus measured at (a) -20˚C, (b) 0˚C (c), 20˚C (d) 40˚C
as a function of ageing time. .......................................................................... 100
Figure 5.21: Mean storage modulus versus ageing time for EVA. ................. 101
Figure 5.22: Average storage modulus versus concentration of ester on the
irradiated surface of UV aged EVA. ................................................................ 102
Figure 5.23: Average storage modulus versus concentration of carboxylic acid
on the irradiated surface of UV aged EVA. ..................................................... 102
Figure 6.1: Flowchart of investigation of EVA’s response to damp heat ageing.
........................................................................................................................ 105
List of Figures
xiii
Figure 6.2: Moisture absorption curve for EVA film at 85°C-85% RH, (a) Mt/Mq
vs time, (b) Mt/Mq vs √time/l. ......................................................................... 108
Figure 6.3: Simulated moisture concentration inside the EVA film at different
depths (X1-X10) under the damp heat condition of 85°C-85% RH. ............... 110
Figure 6.4: DSC first heating thermograms for the unaged and damp heat aged
EVA at (a) 85°C-85% RH and (b) 22±3°C-85% RH-Exo Up. .............................. 111
Figure 6.5: Glass transition temperature versus ageing time for EVA at (a)
85°C-85% RH and (b) 22±3°C-85% RH. ........................................................... 112
Figure 6.6: Crystallinity versus ageing time after first and second heating for
unaged and aged EVA at (a) 85°C-85% RH and (b) 22±3°C-85% RH. .............. 114
Figure 6.7: Crystallinity versus average concentration after first and second
heating for unaged and aged EVA at 85°C-85% RH. ....................................... 114
Figure 6.8: Storage modulus vs temperature for unaged and aged EVA at (a)
85°C-85% RH and (b) 22±3°C-85% RH. ........................................................... 116
Figure 6.9: (a) tan(δ) vs temperature for unaged and aged EVA at 85°C-85%
RH, (b) Tg versus ageing duration based on Figure (6.9: a). ........................... 117
Figure 6.10: (a) tan(δ) vs temperature for unaged and aged EVA at 22±3°C-
85% RH, (b) Tg versus ageing duration based on Figure (6.10: a). ................. 118
Figure 6.11: Storage modulus measured at (a) 0˚C, (b) 20˚C (c), 40˚C (d) 60˚C
as a function of ageing time at 85°C-85% RH. ................................................ 120
Figure 6.12: Storage modulus measured at (a) 0˚C, (b) 20˚C (c), 40˚C (d) 60˚C
as a function of ageing time at 22±3°C-85% RH. ............................................ 121
Figure 6.13: Average storage modulus versus ageing time for EVA aged at
85°C-85% RH and 22±3°C-85% RH. ................................................................. 122
Figure 6.14: Average storage modulus versus average moisture concentration
for EVA aged at 85°C-85% RH. ........................................................................ 123
Figure 7.1: Crystallinity versus ageing duration for different ageing conditions.
........................................................................................................................ 127
List of Figures
xiv
Figure 7.2: Comparitive effect of degradation factors on the storage modulus
of EVA at (a) 0°C, (s) 20°C, (c) 40°C versus ageing duration for control sample
and aged EVA at different conditions. ............................................................ 128
Figure App.1: Original DSC curves related to (a) Figure (4.15), (b) Figure (5.13),
(c) Figure (6.4: a), (d) Figure (6.4: b). .............................................................. 156
List of Tables
xv
List of Tables
Table 2.1: Kinetic parameters calculated by (Rimez, Rahier, Van Assche, Artoos
& Van Mele 2008). ............................................................................................ 21
Table 2.2: Activation energy calculated for EVA with 12% and 20% VAc (EVA-
12 and EVA-20) by different methods (Marín et al. 1996). .............................. 21
Table 2.3: Measured diffusion coefficient (D) in the literature. ...................... 31
Table 2.4: The characteristics of the ATR-FTIR absorption peaks. ................... 33
Table 2.5: Some DSC results for EVA with 28% VAc (Shi 2008; X. Shi et al.
2009). ................................................................................................................ 34
Table 3.1: Release films and release agents used for sample preparation. ..... 43
Table 3.2: Overview of the curing techniques. ................................................. 48
Table 3.3: Water Vapour Transmission test condition provided by RDM TEST
Equipment®. ..................................................................................................... 58
Table 4.1: Temperature of peak degradation rate for aged and unaged EVA
samples at different heating rates. .................................................................. 68
Table 4.2: The calculated kinetic parameters for unaged EVA. ....................... 69
Table 4.3: The fitting parameters for storage modulus at different fixed
temperatures based on Figures (4.12: a-d). ..................................................... 75
Table 4.4: The best fitting parameters of the average storage modulus versus
ageing duration based on Figure (4.13), fitting function: y=a1x2+a2x+c. .......... 75
Table 5.1: Attribution of infrared absorption bands of EVA film. .................... 85
Table 5.2: Line of best fit parameters (Figure (5.5) UV aged top side). ........... 88
Table 5.3: Best fit parameters (Figure (5.9)). ................................................... 91
Table 5.4: The fitting parameters Figure (5.12). ............................................... 93
List of Tables
xvi
Table 5.5: The fitting parameters for storage modulus at different fixed
temperatures based on Figures (5.23). .......................................................... 101
Table 6.1: Gravimetry test results. ................................................................. 109
Table 6.2: WVTR test results (the analyses were carried out on a MOCON®
Permatran-W Water Vapour Permeability Instrument). ................................ 109
Nomenclature
xvii
Nomenclature
Abbreviations
BIPV
Building Integrated Photovoltaic
BP
British Petroleum
DMA
Dynamic Mechanical Analysis
DSC
Differential Scanning Calorimetry
DTA
Differential Thermal Analysis
EPFL
École Polytechnique Fédérale de Lausanne
EVA
Ethylene Vinyl Acetate
Fraunhofer ISE
Fraunhofer Institute for Solar Energy Systems
FTIR
Fourier Transform Infrared Spectroscopy
GC-MS
Gas Chromatography–Mass Spectrometry
HTGPC
High Temperature Gel Permeation Chromatography
IR spectroscopy
Infrared spectroscopy
NMR
Nuclear Magnetic Resonance
NREL
National Renewable Energy Laboratory
PV
Photovoltaic
RH
Relative Humidity
RMS
Root Mean Square
Nomenclature
xviii
STAPP
Stability and Performance of Photovoltaics
SUPSI
Scuola universitaria professionale della Svizzera italiana
University of Applied Sciences and Arts of Southern
Switzerland
TGA
Thermogravimetric Analysis
UV
Ultra Violet
VAc
Vinyl acetate concent
WAXD
Wide Angle X-ray Diffraction
WVTR
Water Vapour Transmission Rate
Alphabetic
a
Pre-exponential factor
A
Absorbance
b
Path length
c(x,t)
Moisture concentration
c0
Initial moisture concentration
CI
Carbonyl Index
Cm
Concentration of medium
Cp
Concentration of product
Cr
Concentration of reactant
csat
Saturation moisture concentration
Nomenclature
xix
D
Moisture diffusion coefficient
E
Activation energy
E’
Storage modulus
E’’
Loss modulus
F(Cr)
Rate function
F()
Kinetic rate function
I
Intensity
k
Fractional decay rate
K(T)
Rate constant
l
Thickness
m0
Initial sample mass
mq
Final sample mass
mt
Sample mass at time t
R
Universal gas constant
t
Time
Tg
Glass transition temperature
Tm
Melting Temperature
Tp
Temperature at maximum degradation rate
V
Volume
Xc
Crystallinity
Nomenclature
xx
z’, t’
Variables
Greek symbols
µ(z), ϑ(t)
Independent functions
α
Molar extinction coefficient
β
Heating rate
δ
Phase Angle
λ
Stoichiometric coefficient
Conversion degree
Introduction
1
Chapter 1
Introduction
1.1 Background
Recent years have seen the UK government commit to cost effective
renewable energy as part of a diverse, low-carbon and secure future energy
mix. A key benefit of deploying renewable energy technologies is the potential
reduction in carbon emissions when compared to fossil fuels (Harris & Annut
2013). The Kyoto Protocol commits governments, including the UK, to reduce
the emission of greenhouse gases. The 2009 Renewable Energy Directive also
set a target for the UK to achieve 15% of its energy consumption from
renewable sources by 2020 (Department of Energy & Climate Change 2013).
The UK has made progress towards this challenge and total solar PV capacity
grew by 1.0 GW between July 2012 and June 2013 (Department of Energy &
Climate Change 2013).
1.2 Solar PV
Solar energy reaches the surface of the Earth with a power density of 500 to
1000 W/m2 which would ideally be used in an effective way and converted to
usable form of energy for end use customers (Czanderna & Jorgensen 1997).
Solar energy is one of the fastest-growing sources of electricity in the world.
The total solar energy absorbed by Earth's atmosphere, oceans and land
masses is approximately 3,850,000 exajoules (EJ) per year. In 2002, this was
more energy in one hour than the world used in one year (Smil 2006). Solar
radiation can be converted into electrical energy directly, without any inter-
mediate process, by the use of solar photovoltaic (PV) cells. This form of
Introduction
2
generation of electricity has considerable advantages to conventional forms
which includes (J. C. McVeigh 1983; Ken 1990; Stone 1993):
Little maintenance is required because there are no moving parts.
It is pollution free and its source is renewable.
There are no harmful waste products.
Solar panels can be installed very quickly.
Although there has been progress towards meeting the 2020 renewable
target, the scale of the increase over the next years represents a huge
challenge and will require strong contributions from the three sectors of
electricity, heat and transport. The mix of renewable energy generation
needed to meet the 2020 target will require several technologies to make a
significant contribution. Solar PV is one of the eight key technologies set out
in the Renewable Energy Roadmap Update 2012 (Anon 2014) and plays an
important role in the UK’s energy generation owing to its reliability. Solar PV is
not just important because of its energy generation potential, it can also
contribute to UK economic growth (Anon 2013).
The extensive deployment of solar panels across the UK is among the most
popular renewable energy technologies. Recently, solar received the highest
public approval rating of all renewable energy technologies, at 85%. The
major problem in using solar energy is not how to collect it but is how to
collect it in a cost effective way in order to be competitive compared to
conventional forms of energy (Claassen, R. S.; Butler 1980). The ability to
deliver further reductions in the installed costs of solar PV will determine the
level of sector growth and the ability for the levelised cost, the per-kilowatt
hour cost of building and operating a power plant over an assumed financial
life, of solar energy to become competitive with other energy sources.
PV modules are now accepted as a possible source of renewable energy for
households. There are, however, problems with this technology, primarily
that their lifetime is currently such that their cost makes benefits marginal. PV
modules work under varying climatic conditions, exposed to environmental
Introduction
3
stresses such as rain, snow, hail, temperature variation, etc. Since there are a
number of different layers of various materials used in the structure of a PV
module, the response to the environmental stresses will vary, which causes
mismatches and results in defects in the PV module structure and affects the
performance, reliability and durability of the PV module. To improve the
lifetime and durability of photovoltaic modules we need a better
understanding of the material degradation mechanisms to be able to predict
the degradation under different climatic conditions.
The encapsulant is an essential part of a PV module for mechanical protection
of the silicon solar cells and electrical insulation of the PV module. It also
protects the metal contacts and interconnects against corrosion. The
encapsulant packaging protects the solar cells from environmental damage
and mechanical stresses. This protection must insure at least 20-25 years
lifetime for the modules to be cost effective.
Recent developments in the PV industry have focused on the efficiency of the
modules and the improvement of lifetime at the material level. Among the
various material layers in a module’s structure, the performance of
encapsulant has become a subject of discussions in recent years. The subject
arose from the observed degradation and discolouration of the encapsulants
(Figure (1.1)) and degradation of generated electrical energy in PV modules
especially those deployed in dry-hot or humid-hot climates (Czanderna & Pern
1996). Therefore, it is important to be able to predict the degradation of an
encapsulant material and its lifetime under different environmental
conditions in order to optimize the durability of PV modules. Figure (1.1)
shows two types of PV module degradation in material level, the
discolouration of EVA (a) and corrosion in the backsheet (b) in a roof installed
PV module in Scuola universitaria professionale della Svizzera italiana (SUPSI),
Lugano, Switzerland.
Introduction
4
(a)
(b)
Figure 1.1: Degradation in materials layers in roof installed PV modules in Scuola universitaria
professionale della Svizzera italiana (SUPSI), Lugano, Switzerland. (a) Discolouration in EVA.
(b) Corrosion in the back sheet.
Introduction
5
1.3 Stability and Performance of Photovoltaics (STAPP)
project
The stability and Performance of Photovoltaics (STAPP) project was funded
under the India-UK Collaborative Research Initiative in Solar Energy. It is a
balanced consortium having four institutes from UK and four institutes from
India. The emphasis of the project is on the stability and performance of
photovoltaic modules and system. This project also involves many leading
Photovoltaic companies, mainly from India and the UK. The project consists of
6 work packages (WP) as illustrated in Figure (1.2).
Figure 1.2: Work packages of STAPP project.
The University of Nottingham’s contribution to STAPP focused on the
chemical degradation, mechanical modelling and lifetime behaviour
prediction of PV modules (WP2). This PhD thesis in particular investigates the
chemical degradation of the encapsulant in order to understand the
degradation mechanisms and predict ageing. Figure (1.3) shows the STAPP
project program outline for the University of Nottingham.
Introduction
6
Figure 1.3: STAPP project program outline for the University of Nottingham.
1.4 Aim
This work aims to increase understanding and predict the degradation of the
encapsulant material, EVA, under different environmental conditions and the
associated changes in mechanical properties by examining the link between
the chemistry, the structure and the intrinsic mechanical behaviour. This will
enable module degradation to be understood and lead to better production
of more durable PV modules.
1.5 Objectives
1) To understand the degradation mechanisms.
2) To identify the suitable characterisation techniques.
3) To develop reliable accelerated ageing methods.
4) To correlate chemical degradation and morphological changes with
mechanical properties.
5) To investigate the kinetics of the thermal degradation of EVA and develop
a predictive model to be able to predict weight loss of EVA when it is
exposed to elevated temperature only.
Introduction
7
6) To investigate the photodegradation of EVA in order to develop a
predictive model to predict the degradation of EVA when it is exposed to
UV only.
In order to meet the aim and the objectives of the project, a series of tests
and experiments were designed to age the material under controlled
conditions and perform thermal, optical and mechanical analysis on the aged
material to study the effect of the ageing on the material properties,
characterise the aged material and investigate the kinetic parameters in order
to calculate the lifetime of the encapsulant and correlate it with mechanical
properties. Figure (1.4) illustrates the work flow chart for the experimental
programme. STAPP project emphasizes on the stability and performance of PV
modules and system and as a part of STAPP this PhD thesis focuses on
investigating the chemical degradation of encapsulant to understand the
degradation mechanisms to be able to develop a predictive model for lifetime
behaviour prediction.
Figure 1.4: The work flow chart for the experimental programme.
Introduction
8
1.6 Research novelty
There is number of previous works on the durability and reliability of PV
modules; including an extensive description of the hydro-photo-thermal
degradation mechanisms. There is, however, a missing link between the
ageing and degradation process and their consequences within the context of
morphological changes, mechanical behaviour and prediction of degradation
at isolated UV and isolated thermal ageing conditions. Therefore, the novelty
of this research can be summarised as
Correlating of the material’s degradation with its lifetime under in-service
climatic conditions.
Developing a predictive model in order to predict Photodegradation and
thermal degradation of EVA.
Investigation of link between moisture absorption and the changes in
crystallinity and viscoelastic properties of EVA.
Investigation of link between the chemical degradation, structure and
mechanical behaviour of EVA.
1.7 Thesis structure
The thesis is structured as below:
Chapter 2 gives an introduction to photovoltaic (PV) systems and introduces
Ethylene Vinyl Acetate (EVA) as a commonly used encapsulant. This chapter
also reviews the effect of environmental stresses (elevated temperature,
Ultraviolet (UV) and humidity) on the structure and behaviour of EVA and
gaps in the knowledge are discussed.
Chapter 3 describes the methodology, equipment, techniques, material,
sample preparation and artificial ageing process including elevated
temperature, UV radiation and damp-heat used throughout the research
project in order to conduct experiments.
Introduction
9
Chapter 4 contains a detailed study on the effects of elevated temperature
and thermal ageing on the structure, mechanical properties and chemical
degradation of EVA. In particular, a correlation between chemical and
mechanical degradation is presented. A general procedure for finding kinetic
parameters is used in order to predict the material lifetime.
Chapter 5 discusses the impact of UV on the chemical degradation, structure
and mechanical properties of EVA. Morphological changes, mechanical and
chemical degradation are correlated and presented.
Chapter 6 describes the effect of the moisture/humidity ingress and damp-
heat ageing on the structure and viscoelastic properties of EVA. It also
discusses the measurement of moisture diffusion coefficient.
Chapter 7 Comparatively studies the ageing impact on the structure and
mechanical properties EVA under different ageing conditions.
Chapter 8 presents conclusions from the thesis and recommendations for
future work.
Literature Review
10
Chapter 2
Literature Review
2.1 Introduction
The global solar sector is going through a period of great change and it is
important that its full potential is grasped, along with the wide economic and
environmental benefits that it can bring. One of the challenges in this respect
is to make PV modules cost effective in order to support the growing energy
demand. Currently, however, PV module take-up and installation is
dependent upon government subsidy, owing to the marginal economic
benefit to the user as a consequence of the high capital cost and relatively low
lifetime (Anon 2013).
The lifetime of a PV module is generally limited by the degradation of the
constituent parts. The materials age and degrade and there is a continuous
decrease in the efficiency leading to eventual failure. One part that is
particularly susceptible to degradation is the encapsulant. The encapsualant
packages the silicon cells into a weatherproof structure with a front sheet,
usually of glass. This packaging is expected to protect the solar cells from
environmental damage. The most common encapsulant material is Ethylene
Vinyl Acetate (EVA) and degradation of this layer can lead to a significant drop
in a PV module’s efficiency, durability and lifetime.
Since EVA undergoes chemical degradation when it is exposed to heat,
humidity and Ultra Violet (UV) radiation, there is the possibility of a complex
interaction of several different ageing mechanisms. In order to be able to
predict and control the behaviour of EVA under various environmental
Literature Review
11
conditions it is necessary to understand the degradation mechanisms and
mechanical behaviour under different climatic conditions.
Determining the effect of environmental stresses and artificial ageing on
polymeric materials is of concern in many engineering applications and has
been the subject of significant research (Liu, Wildman, Ashcroft, et al. 2012a;
Liu, Wildman & Ashcroft 2012; Elmahdy et al. 2010; Ashcroft et al. 2012;
Ashcroft et al. 2001). In this chapter we look to explore the state of the art
before discussing the gap in knowledge and the proposed programme of
work.
2.2 Photovoltaic (PV) system
Photovoltaics is an approach to direct conversion of sunlight into electricity
(Stone 1993). A solar cell is an electrical device that converts the sun’s light
energy into electricity. In order to provide useful power for any application,
the individual solar cells must be connected together to give the appropriate
current and voltage levels and they must also be protected from damage from
the environment in which they operate. This electrically connected and
environmentally protected unit is usually termed a photovoltaic (PV) module.
A series of multiple photovoltaic modules form a photovoltaic array. Figure
(2.1) illustrates the components of a solar array.
Figure 2.1: An illustration of a solar array and its components.
Literature Review
12
The structure of PV modules consists of various material layers. There is a top
sheet, which is usually low iron glass, encapsulant surrounding the silicon
wafers, a back sheet which can be glass or polymer and a junction box. The
basic construction of a typical PV module is illustrated in Figure (2.2). This
package is the PV module, or solar panel. In other words a PV module is a
collection of individual solar cells integrated into a package which protects
them from environmental factors including rain, snow, dust, thermal and
mechanical stresses.
Figure 2.2: Structure of a PV module showing different material layers (courtesy to Micro
System Services (MSS)).
A standard PV module has an area of 0.5 m2 to 1 m2 and thickness of 5-20 mm
(Carlsson et al. 2006). The structure of the PV module is dictated by several
requirements; including the electrical output, which determines the number
of cells incorporated; the transfer of as much light as possible to the cells, the
cell temperature, which should be kept as low as possible, and the protection
of the cells from exposure to environmental stresses.
Environmental conditions play a crucial role in the durability, performance
and lifetime of the PV systems. A PV module’s performance deteriorates when
it is exposed to outdoor environmental conditions, which affect the reliability
Literature Review
13
of the PV modules in the long term (Vázquez & Rey‐Stolle 2008). PV modules
are required to survive 20 to 25 years to achieve economic break even. The
most important factor in PV electricity generation is to make it economically
viable and assure maximum power output for the lifetime of PV modules (S.-
H. Schulze, S. Dietrich, M. Ebert 2008).
Among the various materials in a PV module, the encapsulation material is
one of the most critical components in terms of durability. The encapsulant
serves a number of different purposes. It mechanically protects the devices
and electrically insulates them (Jorgensen & McMahon 2008). It also packages
the silicon cells into a weatherproof structure. Packaging cost is 50% of the
total material cost in a PV module, which shows its economic importance
(Osterwald & McMahon 2009).
The solar panel must be protected against these factors (Osterwald &
McMahon 2009):
Corrosion of metallic materials,
Water vapour permeation/moisture ingress,
Delamination of encapsulant,
Damage from physical environmental factors such as wind, hail, etc.,
Thermal expansion mismatch,
UV exposure,
Damage to external parts such as the junction box.
To avoid or limit moisture ingress, good sealing and adhesion are required.
The adhesion quality is mainly dependant on factors such as (Lange et al.
2011):
The cleanliness of the material sheets, e.g., the glass panel.
The condition of encapsulant material prior to lamination i.e. storing.
The lamination process, with process parameters including
temperature, pressure, duration and homogeneity of the distribution
of these factors.
Literature Review
14
The homogeneity of the lamination temperature profile, which is
important to achieve a high electrical yield and a long module lifetime.
Due to the importance of PV module durability there is increasing demand on
the performance of the glass/encapsulant laminates (A.K. Plessing 2003). In
numerous publications, delamination has been reported as one of the major
failure mechanisms in PV modules, with the emphasis on the
glass/encapsulant interface (Pern 1996; Pern & Glick 2003). However,
delamination at the backsheet interface has also been reported to
significantly contribute to failure initiation in PV encapsulation (Oreski &
Wallner 2005).
Another crucial feature the encapsulant material must have is the ability to
transmit light under long term UV exposure. Although the degradation may
have a minor effect on light transmission it needs additional tests and
research (Michael D Kempe 2008).
Since one of the PV encapsulant’s roles is to serve as a mechanical support, it
is essential to study the mechanical properties of the encapsulant. Kemp’s
investigation (M D Kempe 2005) showed that EVA experiences a phase
transition over the temperature range of -40°C to 80°C where the elastic
modulus decreases by a factor of 500. This transition can be attributed to EVA
having a melting temperature >65°C and glass transition temperature (-15°C).
The mechanical stability of PV modules should ensure resistance against wind
(up to 2.4 kPa) and snow (up to 5.4 kPa). To reach such mechanical stability,
good support of the module glass and mounting of the module is required.
However, reducing the support and mounting material is a key factor to
lowering the cost of the module, which has led to the construction of modules
without a frame being quite common. The complexity of the mechanical
behaviour of PV modules requires us to consider various factors such as time,
temperature and the age dependent properties of material. It is also reported
that 68% of recorded module failures have occurred either at the free edge or
the clamp, which can be correlated with stress concentration regions based
Literature Review
15
on Finite Element (FE) simulations (S. Dietrich, M. Sander, M. Ebert 2008;
Sascha Dietrich, Matthias Pander 2009).
2.2.1 Commonly observed PV module degradations in field
The field degradation of PV modules can be classified into five categories:
1) Degradation of packaging materials such as glass breakage, discolouration
of the encapsulant and back sheet cracking. In the case of thin film
modules, 90% of the failures are related to the packaging materials
(McMahon 2004).
2) Loss in adhesion strength or delamination. Field experience has shown
delamination on the front side of the module is more common than the
back side. This can also cause optical decoupling and prevent effective
heat dissipation. Module reliability is inextricably related to the cohesion
and adhesion of all material layers in the module (Jorgensen & McMahon
2008).
3) Degradation of cell/module interconnects.
4) Degradation caused by moisture ingress, which causes corrosion in
metallic parts and increases current leakage. Moisture permeation also
results in delamination (Dhere 2000; Dhere, Neelkanth G. and Pandit,
Mandar 2001; Osterwald et al. 2003).
5) Degradation of the semiconductor devices.
These various degradations can lead to termination of the ability of the PV
module to produce useful safe electricity and are dependent on various
factors, some of which are difficult to simulate in the laboratory (Quintana et
al. 2002). Data collection from field experiments started in the 1970s, but this
has not been well coordinated. Different field data show different outcomes,
which is due to different environmental and measurement conditions
(Quintana et al. 2002). For instance a system tested for a period of ten years
from the mid-eighties showed a power loss of 1% to 2% per year (A. L.
Rosenthal, M. G. Thomas 1994). Field data collection from multi-crystalline
modules for a period of eight years in Sandia, USA, showed 0.5% loss in
Literature Review
16
performance per year (King et al. 2000). A study performed by National
Renewable Energy Laboratory (NREL) shows both single and multi-crystalline
field modules to degrade at about 0.7% per year (Osterwald et al. 2002).
These differences in the field data show the variability and dependency on
environmental and measurement conditions.
In the next section, the properties and structure of the most common PV
module encapsulant, Ethylene Vinyl Acetate (EVA), will be reviewed.
2.3 Ethylene-vinyl Acetate (EVA)
Copolymers were developed to expand industrial utilization of polymeric
materials due to the limited use and restricted properties of homopolymers
(Çopuroğlu & Şen 2004). EVA is a copolymer of ethylene and vinyl acetate
with the structure shown in Figure (2.3). EVA is the most commonly used
encapsulant in the PV industry due to its low cost, high adhesion strength and
high transparency. EVA’s flexibility has made it popular in many industries,
including foot ware (Allen et al. 2005; Chiu & Wang 2007), the toy industry
(Globus et al. 2004), agricultural (Abrusci et al. 2012) and in PV applications
(Czanderna & Pern 1996).
EVA has been in commercial use in PV modules since 1981 (Wohlgemuth &
Petersen 1991). The EVA used in the PV manufacturing industry is 28%-33%
weight percentage vinyl acetate compounded with additives such as a curing
agent, UV absorber, a photo antioxidant and a thermo antioxidant (Klemchuk
et al. 1997). In the range of 400 nm-1100 nm, this EVA shows nearly the same
optical transmission as glass. The properties of the EVA which make it a good
encapsulation choice are high electrical resistivity, low polymerization
temperature, low water absorption ratio and good optical transmission
(Czanderna & Pern 1996; Schulze et al. 2009).
The vinyl acetate content (VAc) has a substantial effect in EVA’s properties
and structure. It is reported that the lower the amount of VAc the more the
copolymer is akin to polyethylenes, with higher crystallisation, but increasing
Literature Review
17
VAc prevents adjacent polyethylene chains crystallizing and EVA’s structure
tends to be amorphous and rubbery (Rodríguez-Vázquez et al. 2006) this is
mostly seen when the VAc exceeds beyond 43% by weight (Kamath &
Wakefield 1965).
Figure 2.3: Chemical structure of EVA copolymer.
Recently the PV industry has focused on the improvement of lifetime at the
material level, and the encapsulant has become the subject of research. EVA
undergoes degradation when it is exposed to harsh environmental conditions.
The main degradation product of the photo-thermo-chemical degradation of
EVA is acetic acid, which causes corrosion in the metallic parts of PV modules
and deteriorates the adhesion properties of the EVA. UV also causes
discolouration in EVA, which affects the efficiency of PV systems. Moisture
ingress causes problems such as delamination and corrosion. These problems
show the important effect of environmental and climatic conditions on the
durability of PV modules.
PV systems are used in many different continents and regions with very
different environmental conditions. Therefore, it is important to be able to
predict the potential failure modes of the system in different climates to
achieve the desired reliability, durability and economical break even. Thus,
there is a need to investigate and understand the degradation factors and
mechanisms involved in this degradation.
Literature Review
18
2.3.1 EVA storing condition and curing process
Virgin (uncured) EVA is usually provided commercially in sheet form. As
mentioned in section (2.2), the pre-lamination condition of EVA has a great
influence on the quality of the lamination. The EVA manufacturers
recommend an optimum temperature of 22°C (generally below 30°C) as a
storage temperature and a relative humidity (RH) of 50% (Krauter et al. 2011).
The virgin EVA should be kept out of direct sunlight and should not be stored
for more than 6 months. The technical manual of Photocap® shows the
adhesion strength of EVA reduces as the storage duration increases (Krauter
et al. 2011).
In order to achieve suitable mechanical and optical properties EVA should be
cured. The recommended curing level ranges between 70% and 90% which
depends on the type of EVA and backsheet. A good combination of curing and
lamination can slow down the ageing of solar cells and the delamination
process. During curing cross-linking occurs and the EVA is converted from a
thermoplastic into a thermosetting material, which provides the protection of
the active PV elements in the module (Lange et al. 2011).
Previous research into the impact of the various EVA degradation factor is
reviewed in the next section.
2.4 The effect of environmental stresses on the
properties and structure of EVA
PV modules are exposed to various harsh environmental stresses such as UV
radiation, heat, humidity, pollutants and thermal cycles. Additional physical
climatic factors, e.g. rain, dust, wind and hail, can also have a deteriorating
effect. All these factors deteriorate the durability, stability, reliability and
performance of the PV modules (Czanderna & Pern 1996; Oreski & Wallner
2005). It is necessary, therefore, to understand the degradation mechanisms
caused by the main environmental stresses in order to predict the lifetime
and change to performance of a PV module in-service.
Literature Review
19
2.4.1 The effect of elevated temperature
EVA undergoes a two-step thermal degradation when it is exposed to
elevated temperatures which is investigated by several researchers (Häußler,
L. , Pompe, G., Albrecht, V., Voigt 1998). The first stage of the thermal
decomposition of EVA is an acetate pyrolysis; evolving mainly acetic acid in a
so-called deacetylation process, as shown in Figure (2.4). The second stage of
the degradation is the degradation of any remaining partially unsaturated poly
ethylene polymer and the production of a large number of straight-chain
hydrocarbon products by breakdown of the hydrocarbon backbone
(McGrattan 1994; Sultan & Sörvik 1991a; Sultan & Sörvik 1991b; Sultan &
Sörvik 1991c; Camino et al. 2000; Zanetti et al. 2001; Riva et al. 2003; Kaczaj &
Trickey 1969; Gilbert et al. 1962).
Figure 2.4: Acetic acid formation (deacetylation).
Allen et al. (2001) investigated the thermal degradation of EVA in the
presence of oxygen (thermal oxidation) and reported that the main
degradation of EVA is the loss of acetic acid (deacetylation), which is followed
by oxidation and main chain breakdown. They also reported that the rate of
degradation is greater in the presence of oxygen.
Numerous researchers (Trombetta et al. 2000; Gulmine et al. 2002; Bauer et
al. 1999; Post et al. 1995; Price & Church 1997; McNeill et al. 1976) have used
Fourier Transform Infrared Spectroscopy (FTIR) combined with other
techniques to investigate pyrolysis of polymers. (Marcilla et al. 2005) studied
the thermal pyrolysis of EVA; using combined thermogravimetry (TG)/FTIR to
investigate evolved gas during the degradation. A TG analysis was used to
decompose three commercial EVA copolymers in an inert atmosphere which
was connected to a FTIR in order to study the evolved gas. Their experimental
results clearly show a two-step decomposition. The FTIR spectra of the
Literature Review
20
evolved gas revealed that during the first stage of decomposition the evolved
gas were mainly acetic acid and small quantities of CO, CO2 and CH4. The
spectra from the second stage of the decomposition showed traces of alkene
and alkane mixtures, plus small amounts of aromatic compounds.
Costache et al. (2005) used Gas Chromatography-Mass Spectroscopy (GC-MS)
and TGA/FTIR to investigate the impact of nanoclay in the degradation of EVA.
Their results also show a two-stage degradation. The TGA/FTIR results showed
the evolution of acetic acid, CO2 and water on the early stages of degradation.
The presence of the nanoclay catalysed the loss of acetic acid in the first stage
of the degradation. They reported that the thermal degradation process of
EVA in the presence and absence of nanoclay are very similar, however, there
are subtle changes, shown by differences in the quantity and quality of the
products which are formed during the degradation. They finally concluded
that nanoclay can promote one of the degradation pathways at the expense
of another and thus lead to different volatilization rates and products.
Rimez et al. (2008) investigated the thermal degradation of EVA with 50%
vinyl acetate (VAc) content through an experimental study. In order to
investigate both the volatiles and the solid degradation products different
techniques were used such as solid state Nuclear Magnetic Resonance (NMR),
TG coupled with Mass Spectrometry (MS) and Differential Thermal Analysis
(DTA). The results showed that the deacetylation occurs between 300°C and
400°C.
Rimez at al. (2008) studied the kinetics of the degradation of EVA with 33% of
vinyl acetate according to a mechanistic approach in which isothermal, linear
heating and High Resolution (Hi-Res) TGA were used. Their calculated values
for activation energy (E) and pre-exponential (a) factor by the Kissinger
method are shown in Table (2.1).
Literature Review
21
Table 2.1: Kinetic parameters calculated by (Rimez, Rahier, Van Assche, Artoos & Van Mele
2008).
Process
Parameter
Value
Deacetylation
log(a)
12.7 (s-1)
E
177.5 (kJ/mol)
Marín et al. (1996) investigated the thermal degradation of EVA in granule
form and focused on calculating the activation energy of the deacetylation
process by establishing the polymer weight loss with temperature. Their TG
results show that the deacetylation starts at about 286°C and finishes at
396°C. The maximum reaction rate occurred at around 317°C. They also
reported that as the vinyl acetate concentration increases the degree of
crystallinity deceases and the end-product shows behaviour similar to
thermoplastic rubbers. The decomposition of EVA depends on the heating
rate; the maximum deacetylation temperature increases with increasing
heating rate. The calculated values of activation energy for EVA with different
VAc (12% and 20%) by different methods are presented in Table (2.2).
X. E. Cai (1999) studied the two-step degradation of EVA with 43.8% VAc by
TG/FTIR. They reported that heating rates from 5°C to 40°C gave good TG
curves and used the Kissinger method to calculate 221 kJ/mol for activation
energy.
Table 2.2: Activation energy calculated for EVA with 12% and 20% VAc (EVA-12 and EVA-20)
by different methods (Marín et al. 1996).
Method
EEVA-12 (kJ/mol)
EEVA-20 (kJ/mol)
Van Krevelen
165±16
185±18
Horowitz-Metzger
168±17
188±19
Coats-Redfern
158±16
177±18
Literature Review
22
Palacios et al. (2012) have investigated the thermal degradation kinetics of
EVA nanocomposites. They reported a two-step degradation with the first
step (deacetylation) occurred in the range of 300°C-400°C. They calculated
179 kJ/mol for the deacetylation activation energy and 329 kJ/mol for the
second stage (chain session).
Despite all the previous research, there is still a missing link between the
ageing process, lifetime and the consequences of that ageing within the
context of morphological changes and mechanical behaviour.
2.4.2 The UV effect
EVA in a PV module exposed to outdoor conditions undergoes photothermal
degradation, which may result in discolouration, developing from a light
yellow to dark brown in about five years or more depending on the site,
module configuration and operating temperature (Czanderna & Pern 1996).
Data from Carrisa solar farm in California showed discolouration is also
responsible for performance degradation at a rate of 10% per year (Pern
1993; Wenger et al. 1991).
Pern (1993) studied the optical characteristics and properties of EVA before
and after weathering. His findings show that the uncured EVA develops
yellowing even in the dark at room temperature after around one year while
some cured EVA films have developed a lighter yellowing in two or three
years. The degree of yellowing for uncured EVA is four times greater that of
cured EVA in five years. The front side of the EVA which was exposed to direct
UV, degraded earlier, and to a greater extent and the results clearly show that
the degradation of EVA starts from the top layer at the front side and extends
to the grid side. Pern (1993) also reported that the discolouration of EVA is
accompanied by acetic acid formation (deacetylation) and crosslinking.
Jin et al. (2010) focussed on the impact of UV on behaviour of EVA with
different VAc. A Xenon lamp with an intensity of 0.53 W/m2 at wavelength of
340 nm and constant temperature of 65°C was used to age samples up to 800
hours. The techniques used in their study include Attenuated Total
Literature Review
23
Reflectance-Fourier Transform Infrared Spectroscopy (ATR-FTIR), High
Temperature Gel Permeation Chromatography (HTGPC), TGA, Differential
Scanning Calorimetry (DSC) and mechanical tensile testing. In the case of EVA
with 14 wt% VAc the ATR-FTIR results showed absorption shoulder at 1715
cm-1 and 1175 cm-1 associated with C=O stretching. This might indicate
acetaldehyde evolution in the Norrish lll photoreaction (Allen et al. 1994) or
H2O deprivation of hydroperoxide (Copuroglu & Sen 2005; Çopuroğlu & Şen
2004). In (Czanderna & Pern 1996; Morlat-Therias et al. 2007; Lacoste et al.
1991; Glikman et al. 1986) the major sequences of photothermal degradation
including Norrish reactions are described and these ae shown schematically in
Figure (2.5). In the case of EVA with 18 wt% VAc the maximum carbonyl
formation shifts to 1163 cm-1, indicating the chain scission and greater
damage to the EVA with the higher VAc (Çopuroğlu & Şen 2004; Allen et al.
2001). Growth was also seen in the 1735 cm-1 peak and the emergence of a
new carbonyl band at 1780 cm-1, which can be attributed to lactone formation
(Allen et al. 2001). Variation was also seen at 1175 cm-1, 1715 cm-1 and 1163
cm-1 due to formation of carboxylic acid, especially in the first 400 hours, with
the higher VAc EVA showing greater variation comparing to the EVA with
lower VAc. This is because vinyl acetate is vulnerable to UV and can easily
form reactive radicals which can weaken the polymers resistance to UV
irradiation. Degradation can extend from the surface to the entire polymer by
the accumulation of radicals and penetration of oxygen. HTGPC results in this
study showed that EVA with higher VAc had greater molecular weight
reduction on ageing due to having shorter ethylene chain segments lengths.
The DSC results showed that both EVA types exhibited multiple melting
endotherms, starting at about 41.3°C which are stable during the ageing. The
endotherms at the lower and medium temperatures can vary during the
ageing and are due to secondary crystallizations while the major peak can be
attributed to the primary crystallization (Chen et al. 2009b; X. Shi et al. 2009;
Lee & Kim 2008; K.A. Moly et al. 2005; Brogly et al. 1997). Mobile vinyl acetate
units can easily absorb photon energy and generate unstable radicals, where
Literature Review
24
photochemical degradations can then initiate (Jin et al. 2010a). TGA of the UV
aged EVA showed a two-step weight loss where the first completed at about
403°C and the second one finished within 405°C and 510°C (Costache et al.
2005; Tambe et al. 2008). Mechanical testing showed a decreasing tendency
in the tensile strength, elongation at break, and modulus at 100% elongation
of EVA before and after ageing. However, EVA with lower VAc exhibited more
rigid mechanical properties (tensile strength, elongation at break, and 100%
stretching stress) which is due to the higher crystallinity (Jin et al. 2010a).
Literature Review
25
Figure 2.5: Photodegradation mechanism in EVA (Morlat-Therias et al. 2007).
Morlat-Therias et al. (2007) studied the influence of carbon nano-tubes on
photodegradation of EVA and reported that the main acetate absorption band
appears at 1740 cm-1. There are two shoulders next to this main peak with
maxima at 1718 cm-1 and 1780 cm-1, which are attributed to the formation of
carboxylic acid and lactone respectively. An unsaturated compound is a
Literature Review
26
chemical compound that contains carbon-carbon double bonds or triple
bonds. In photodegradation, unsaturations are responsible for discolouration
which corresponds to vinyl bonds at 909 cm-1 and 990 cm-1. There is also a
growing peak at 3520 cm-1 attributed to alcohols associated with the acetate
group. It is also reported that the rates of photooxidation of EVA in different
samples can be compared by measuring the increase of absorbance at 1718
cm-1 with the ageing time. Esterification is also important in EVA degradation
since it results in acid formation which causes corrosion in the PV modules
and affects the mechanical properties of EVA. Badiee et al. (2014) investigated
the effect of UV ageing on the chemical degradation of EVA and found that UV
makes significant chemical changes compared to other degradation factors,
including the formation of carboxylic acid and lactone.
A typical way that PV modules fail is to lose optical transmission due to
discolouration and yellowing, which is, however, less of a problem since the
1990s, due to improvements in UV absorbers and antioxidants (Pern & Glick
1997; Kempe et al. 2011; W.H. Holley, S.C. Agro & Yorgensen 1996). Although
modern PV modules pass various UV exposure tests it should be noted that
there might still be a reduction in optical transmission considering in-service
as the standard test is around 6 months in an environmental chamber using
highly UV transmissive glass, is much less than the desired PV lifetime (20-25
years) (Michael D Kempe 2008). Myer Ezrin at al. (1995) investigated the
discolouration of EVA and reported that this is due to the developing long
chain on conjugated double bonds as a result of loss of acetic acid from vinyl
acetate units.
Klemchuk et al. (1997) investigated the degradation of EVA in field aged PV
modules. They reported that UV-B transmission (285 nm-330 nm) is mostly
responsible for EVA degradation and discolouration, with the spectrum of
terrestrial sunlight ending at about 290 nm. Their findings showed that the
discolouration started with a very thin layer at the interface under the front
glass. Berman et al. (1995) and Berman & Faiman (1997) studied the effect of
discolouration on module efficiency in field. They observed that the mean
Literature Review
27
maximum power value of 43.7 W for a set of discoloured modules was 5%
below the manufacturer’s guaranteed value. Kojima & Yanagisawa (2004)
similarly made a comparative investigation on the degradation of EVA in the
range of 280-380 nm. They observed slight yellowing in the EVA film after
around 500 hours. Isarankura et al. (2008) investigated the photodegradation
behaviour of EVA and the impact of antioxidants on the photostability of the
material. They reported that EVA predominately degraded under UV
irradiation via a chain scission process and the type and amount of
antioxidants had a significant effect on the photostability of the EVA. They
also found that some antioxidants caused the tensile strength of EVA to
decrease after UV irradiation.
Numerous studies have been carried out at National Renewable Energy
Laboratory (NREL), USA (Pern 1996; Pern & Glick 1997; Pern et al. 1996; Pern
et al. 2000). These studies showed that the module service life can be greatly
improved and extended if UV≤350 nm is removed, as this was seemed to
reduce the UV-induced discolouration. They summarised the main
discolouration factors as:
The UV light intensity.
UV filtering effect of glass.
Lamination quality/delamination and Curing treatment and process.
Film thickness.
The photodecomposition of Cyasorb (UV absorber).
The loss rate of UV absorber.
EVA formulation and additives used.
Despite all the previous studies, there is still a missing link between the ageing
process, prediction of the EVA’s photodegradation and the consequences of
that photodegradation within the context of mechanical behaviour and
structural changes.
Literature Review
28
2.4.3 The effect of damp-heat and moisture ingress
The exposure of PV modules and especially the polymeric materials to
permanently changing climatic conditions is one of the critical degradation
factors. The components of the gaseous atmosphere, such as water vapour,
cause hydrolysis and other chemical reactions which affect and reduce the
lifetime of a PV module.
Moisture ingress in PV modules is via absorption with water molecules
diffusing through the polymeric materials. The process includes adsorption of
water molecules at the surface of the polymer followed by diffusion along a
concentration gradient through the material. This process continues until
equilibrium with the ambient condition, and the diffusion parameters can be
approximated with Fick’s second law (Vieth 1991; Crank 1968).
Moisture ingress into a PV module can deteriorate the performance and
durability significantly during the module’s lifetime in various ways. Moisture
causes corrosion in metallic parts, increase the current leakage by increasing
the conductivity of the encapsulant (Quintana et al. 2002), delamination and
bubble formation. The moisture ingress in PV modules in Miami, Florida, has
been correlated with failure rate, especially in hot and humid (damp)
conditions (King et al. 2000). This plus other reports show the importance of
moisture induced failures (Visoly-Fisher et al. 2003; Malmström et al. 2003).
Investigations by Kempe (Kempe 2006; M.D. Kempe 2005) showed the rate of
moisture ingress into PV modules and reported that the diffusivity of water
has an Arrhenius dependence on temperature. The diffusion was found to be
Fickian within the experimental uncertainty. He reported that in modules with
a breathable back-sheet the moisture absorbed at night is desorbed during
the day due to the higher temperature of the module and higher ambient
water vapour pressure. However, the module can saturate in a rainy climate,
especially when there is a drop in the temperature, which can be a substantial
problem owing to droplet formation by trapped water. Impermeable
Literature Review
29
backsheets are better moisture barriers, however, water can still penetrate
and reach the centre of the module in a few years via the encapsulant.
Hülsmann et al. (2010) measured water vapour permeation into PV modules
by a permeation test device developed and built in the Fraunhofer Institute
for Solar Energy Systems (ISE), which is described in (Hülsmann et al. 2009).
They conditioned the EVA at 85°C-85% RH, which is a commonly used
condition for qualification testing for PV applications. Numerous researchers
Hulsmann and colleagues (Hülsmann et al. 2013; Kim & Han 2013; Hülsmann
et al. 2010) similarly investigated the water vapour ingress into PV modules
under different climatic conditions. They reported that the water
concentration in the encapsulant between the glass and solar cells is highly
dependent on the climatic conditions. Another important factor is the
temperature dependency of the mass transfer and the combination of
encapsulant and back sheet materials.
Kapur et al. (2009) investigated the rate of moisture ingress through EVA
laminated between impermeable substrates in damp heat conditions (85%
RH, 85°C) for 1000 hours. The measurement was determined by the FTIR
method and it was reported that the moisture level in the encapsulant could
be determined by integration of the Infrared (IR) bands at 1880 nm and 1990
nm. It was seen that the experimental data compared well with that predicted
by a Fickian based diffusion model. Table (2.3) presents a summary measured
moisture diffusion coefficient in the literature.
Iwamoto & Matsuda (2005) and Rashtchi et al. (2012) investigated the
moisture ingress and water concentration in the encapsulant within a PV
module. They found that there are notable changes in the FTIR spectrum of
EVA between 3400 and 3700 cm-1 which is a good indicator of the presence of
water.
IEC 61215 is a qualification test that PV modules should pass in order to
participate in the solar energy market. Modules that have passed the
Literature Review
30
qualification test are much more likely to survive in the field. Numerous
researchers (Markus Bregulla, Michael Köhl, Benjamin Lampe, Gernot Oreski,
Daniel Philipp, Gernot Wallner 2007) investigated the degradation of EVA at
85°C, 85% RH, which is comparable to test 10.13 (damp-heat test) of IEC
61215 Ed. 2 for PV modules. The ATR_FTIR spectra of standard cured EVA
showed a growth in the peak at around 3500 cm-1, which is due to water
diffusion.
E. U. Reisner et al. (2006) reported that the peak temperature of an insulated
Building Integrated Photovoltaic (BIPV) module was 60°C to 80°C while the
peak temperature of a free standing module was 40°C to 50°C, where the
ambient temperature ranges between 30°C and 35°C in Belem, Brazil. In the
first six hours of night the encapsulant absorbed water. After sunrise the
module desorbed water due to increasing module temperature which is
greater than the ambient temperature. It should be noticed that the influence
of relative humidity is not considered in this study.
X. Shi et al. (2009) and Chen et al. (2009) reported the effect of damp heat
ageing on the structure and properties of EVA with different vinyl acetate
content (VAc). In this study EVA was aged at 40°C-93% RH and analysed with
various techniques including ATR-FTIR, DSC and Wide Angle X-ray Diffraction
(WAXD). Based on ATR-FTIR results the characteristics of the absorption peaks
were identified as presented in Table (2.4). For the sake of quantitative
comparison a Carbonyl Index (CI = A1735/A720) was introduced and during the
eight-week ageing the CI decreased in the first week of the ageing but then
increased to equal or more than that of the original values on further ageing.
This variation is due to a pyrolysis including a two-step decomposition;
acetate pyrolysis evolving acetic acid, followed by the breakdown of the
hydrocarbon backbone. However, the presence of oxygen causes ketone
formation via acetaldehyde formation and then formation and destruction of
aldehydes to form carboxylic acids. It can be concluded that the degradation
of all EVA samples, with different VAc, can be characterised by the loss and
formation of O=C group which explains the decrease and increase of CI. The
Literature Review
31
WAXD results suggested there was no significant change observed in the
crystalline phase during the ageing. Their DSC results showed the EVA with
lower VAc has longer average ethylene sequence length and therefore forms
bigger crystals which require a higher temperature to melt. Some DSC results
for EVA with 28% VAc (EVA28) showing onset melting temperature (
), final
melting temperature (
) and crystallinity (Xc) are presented in Table (2.5).
Their hardness results showed that hardness goes up with decreasing VAc and
can be correlated to crystallization. Tensile tests results surprisingly indicated
that damp heat ageing had no effect on the tensile properties of the EVA
copolymers used in this study.
Table 2.3: Measured diffusion coefficient (D) in the literature.
Material
Method/device
Conditions
D
(m2/s)
Other
information
Reference
EVA
Illinoise 7001
50°C,
35°C,
20°C,
@ 60% RH
3×10-10
2×10-10
1.5×10-10
-
(Kim & Han
2013)
EVA
Permeation test
device (developed by
Fraunhofer ISE
(Hülsmann et al.
2009))
80°C,
75°C,
55°C,
50°C,
40°C,
30°C,
20°C
7×10-9
8×10-9
9×10-9
9.5×10-9
1×10-10
5×10-10
7×10-10
-
(Hülsmann
et al. 2013)
EVA
Gravimetric
60°C,
40°C,
20°C
@ 85% RH
3.2×10-10
1.2×10-10
0.8×10-10
Thickness: 1
mm,
Laminated
with TPT
backsheet
(Hülsmann
et al. 2010)
EVA
Self-developed
permeation test
device in Fraunhofer
ISE
38°C,
30°C,
22°C
@ 90% RH
1×10-10
6.7×10-11
4.2×10-11
-
(Hülsmann
et al. 2009)
EVA
Mocon Permatran-
W® 3/31
85°C,
60°C,
40°C,
25°C
@ 100% RH
7.5×10-10
4×10-10
1×10-10
7×10-11
1000 hours,
Thickness:
0.46-2.84
mm
(Kempe
2006)
Literature Review
32
Oreski & Gernot M. Wallner (2010) investigated the thermal and thermo
mechanical behaviour of EVA before and after damp heat ageing by DMA and
DSC under three conditions (85°C-85% RH, 65°C-85% RH, 85°C-30% RH). They
reported that at certain temperatures the aged materials became softer,
showing smaller storage modulus values. The DSC results in this study showed
a broad melting area between -20°C and 80°C with a single peak at 45°C and a
shoulder at 55°C. There was a secondary melting point observed in the first
heating whereas in the second heating the peak at 45°C disappeared and a
regular melting peak at 55°C was seen. The secondary melting was assigned to
the secondary crystallization, which happens in the area between primary
crystals during storage at ambient temperature or exposure to an elevated
temperature. Some have attributed the secondary melting point of EVA to a
less organised crystal phase (Brogly et al. 1997). Since the melting process of
ethylene copolymers starts immediately above the glass transition
temperature (Tg) it is suggested that structural changes and rearrangements
are possible at room temperature (Loo et al. 2005; Androsch 1999). It was also
observed that several changes in the first 250 hours of exposure which
strongly depended on the temperature. Whereas the humidity level has only a
slight impact on the material properties. The changes observed include
recrystallization and increase in storage modulus, which can cause some
problems during the module’s service life associated with delamination,
thermal expansion or external physical stresses such as wind, hail, etc. The
stiffening of the EVA can also result in silicon solar cell cracking.
Literature Review
33
Table 2.4: The characteristics of the ATR-FTIR absorption peaks.
Ethylene segment
Absorption peak
Wave number
(cm-1)
Symmetrical
stretching vibration
of methylene
2920
Asymmetrical
stretching vibration
of methylene
2852
Deformation
vibration of
methylene
1465
Flexural vibration of
methyl
1370
Inner rocking
vibration of
methylene
721
Vinyl acetate
groups
Stretching vibration
of C=O
1735
Asymmetrical
stretching vibration
of C-O
1237
Symmetric stretching
vibration of C-O-C
1020
Literature Review
34
Table 2.5: Some DSC results for EVA with 28% VAc (Shi 2008; X. Shi et al. 2009).
Sample
Ageing duration (week)
(°C)
(°C)
Xc (%)
EVA28
0
48.4
81.5
10.10
1
47.8
81.5
11.66
8
50.8
82
13.5
In spite of all the previous work, there is still a missing link between the
consequences of damp heat ageing, moisture diffusion and moisture
concentration gradient within the context of mechanical behaviour and
changes in the crystallinity of EVA.
2.4.4 The effect of combined degradation factors
As mentioned in the previous sections, the reliability and performance of PV
modules are highly influenced by the behaviour of the encapsulant. The
degradation of encapsulant is dependent on the environmental conditions,
mainly humidity, elevated temperature and UV (Pern et al. 1996; Pern 1997;
Kempe 2010; W. Herrmann, N. Bogdanski 2010).
Peike et al. (2012) investigated the impact of permeation properties and
backsheet-encapsulant on the reliability of PV modules. They conditioned the
sample under climatic conditions including combined ageing (85°C-85% RH,
85°C-90% RH-190 kWh/m2, 60 kWh/m2) up to 1000 hours. The encapsulants
used in this study were EVA, Thermoplastic Silicon Elastomer (TPSE), ionomer
and Polyvinyl Butyral (PVB). The backsheet materials included Tedlar-PET-
Tedlar (TPT) foil, polyamide (PA) sheet and Polyethylene Terephthalate (PET)
composite film. The results of the study showed that the average Water
Vapour Transmission Rate (WVTR) and Oxygen Transmission Rate (OTR) were
higher for the encapsulants in comparison to the backsheets. Hence, the
backsheets can be atmospheric gas ingress resistors. They also reported that
Literature Review
35
after combined UV/DH ageing, yellowing was three to five times greater in the
TSPE/TPT, the TPSE/PA and the PVB/PA laminates in comparison to other
combination of backsheet/encapsulant.
Kojima (2004) also investigated the effect of combined ageing on the
degradation of EVA film. They reported that UV absorbance of EVA is
dependent on surface coarseness or the state of moisture absorption. An
increase was also seen in the transmissivity of EVA in the region of 250-400
nm which could be attributed to a reduction in the UV absorber used for
suppressing photodegradation.
These previous studies have illustrated the influences of moisture/water
vapour ingress on the behaviour, properties and structure of EVA. However,
there is still a gap in the current knowledge which is the relation between field
conditions and the consequences of field ageing on the performance of EVA
as the encapsulant. This will be explained in the next section.
2.5 Gaps in the current durability and lifetime studies of
PV modules academic research and industry
The proceeding literature review has demonstrated the significant previous
work on the durability and reliability of PV modules at a material level, and in
the encapsulation material in particular; including an extensive description of
the hydro-photo-thermal degradation mechanisms. In particular, it was seen
that one of the main degradation products is acetic acid which causes
numerous problems such as corrosion in metallic parts and delamination.
There is, however, a missing link between the ageing process and the
consequences of that ageing within the context of mechanical behaviour and
prediction of the EVA photodegradation and thermal degradation. Previous
research on EVA has considerably shown that it undergoes degradation
following exposure to environmental factors (e.g. UV, humidity and elevated
temperature). Despite the information available in the literature on the
degradation mechanisms, no work has been reported that correlates the
Literature Review
36
material’s degradation under sever in-service climatic conditions, which is
crucial for evaluating the performance of PV modules in a specific
environment. There is also very limited work published on the link between
the chemical degradation and mechanical behaviour of EVA which needs
further investigation. These identified gaps in the literature form the basis for
the research challenges identified for this thesis.
2.5.1 Research challenge 1
During operation in the field the temperature of modules can reach up to
85°C. Continuous long term exposure of EVA to high temperature causes
thermal degradation. Therefore, it is necessary to isolate this degradation
factor in order to predict the lifetime of EVA when it is exposed to the working
condition of PV modules.
This research seeks to address the need to understand changes in the EVA as
a result of thermal exposure by examining the link between the evolving
chemistry, structure, mechanical behaviour and predict the weight loss as a
result of thermal degradation.
2.5.2 Research challenge 2
Sunlight is a necessity to produce electricity via PV modules. However the
sunlight spectrum includes UV which causes EVA to photochemically degrade.
The main product of the photochemical degradation is acetic acid and the
reaction also causes discolouration, which affects the durability and
performance of the PV modules. Despite the numerous studies on the
mechanisms of photodegradation of EVA, there is limited published work on
correlating the photodegradation of EVA to structural changes and
mechanical property degradation seen when exposing a PV module to
operating conditions.
This study aims to address the need to understand the changes that are the
result of UV exposure and how they affect the durability and mechanical
properties over time. It will examine the link between the chemistry of
Literature Review
37
photodegradation, the structure, the intrinsic mechanical behaviour and
prediction of photodegradation throughout the duration of an accelerated UV
ageing degradation test.
2.5.3 Research challenge 3
One of the degradation factors the installed PV modules are exposed to,
especially in environments like the UK’s, is humidity. While PV modules are on
site moisture can penetrate into the modules. The moisture ingress causes
various problems such as corrosion, current leakage and delamination.
Although there are studies which have investigated the moisture diffusion
coefficient and moisture induced degradation, these studies have been
performed in idealistic environment which does not represent the in-service
environment of the installed PV modules. Thus, it is necessary to investigate
the moisture diffusion properties, structure and behaviour of EVA in damp
heat.
The present study seeks to address the need to understand the changes in
mechanical properties as result of damp heat exposure by examining the link
between moisture concentration, structural changes and the mechanical
behaviour throughout the duration of an accelerated damp heat ageing test.
2.6 Conclusion
EVA has a complex response to environmental ageing, with noticeable
changes in mechanical behaviour, structure, thermal and optical properties.
Numerous researchers have studied the effect of environmental factors on
the chemical degradation of EVA. In order to predict EVA’s thermal and
photodegradation in isolation and investigate the moisture absorption
behaviour of EVA and its consequence of the structure and mechanical
properties of EVA, a comprehensive analysis of the response of EVA to
changes in environmental conditions is still required.
In the next chapter the experimental methods adopted for the achievement
of the objectives of this research project will be presented.
Experimental Methods
38
Chapter 3
Experimental Methods
3.1 Introduction
The aim of the chapter is to describe the material, experimental techniques
and general research approaches which were used in this research in order to
address the research questions proposed in section (1.5). In particular, it
describes:
the materials used and their storage conditions,
environmental ageing conditions,
experimental techniques for the evaluation of the viscoelastic
properties of EVA,
a characterisation procedure to investigate the changes on the
crystallinity of EVA as a result of ageing;
a procedure to investigate the rate of thermal degradation and
lifetime of EVA under PV operating conditions;
a procedure for the evaluation of the chemical changes in the
structure of the EVA;
a procedure to investigate the photodegradation and lifetime of EVA,
an experimental procedure to investigate the moisture absorption
and calculate the moisture absorption coefficient.
As mentioned in the previous chapters, EVA undergoes chemical degradation
when it is exposed to the degradation factors of elevated temperature, UV
and humidity, which influences the mechanical properties of the material and
causes changes in the structure of EVA (Figure (3.1)). In this research the
general approach to address the research questions and achieve the research
Experimental Methods
39
aim (section (1.4)) is firstly to fully characterise the unaged EVA, which had
been isolated from environmental degradation factors. Once the
characteristics of the material were understood, the EVA was aged using
artificial ageing equipment including a laboratory oven, UV chamber and
environmental chamber. In order to understand the degradation mechanisms,
each of the degradation factors was isolated and its influence on EVA
investigated. With the purpose of investigating the impact of ageing and
environmental factors on the properties, characteristics and structure of EVA,
the characterisation of EVA was separated into three parts. First, the
mechanical properties and morphological changes (crystallinity) as a function
of temperature and artificial ageing were determined. Secondly, the chemical
degradation of the material was investigated and the degradation rate was
determined as a function of temperature/time. Thirdly, the rate of moisture
absorption was determined as a function of time and artificial ageing
conditions. Then all three factors were combined to investigate how these
parts are related and influence each other. Figure (3.2) shows the general
experimental approach and the methodology used in the research.
Figure 3.1: A general diagram showing the impact of the degradation factors.
Experimental Methods
40
Figure 3.2: A general diagram showing the experimental methodology used in this research.
3.2 Material and storing condition
3.2.1 Material (sample preparation)
Uncured EVA is not suitable to be used as an encapsulant material in PV
modules due to low transparency and poor mechanical properties. Figure
(3.3) shows uncured EVA which has low transparency. The usual method in
the PV industry for curing EVA is to use a laminator which provides a
controlled vacuum atmosphere with suitable heating/cooling rates and
pressure. Due to the unavailability of a laminator at the start of this research a
number of techniques were investigated to prepare good quality samples,
including a hot press and autoclave. A description of the various methods
investigated for preparing samples is given below, followed by an evaluation
of the quality samples manufactured.
Experimental Methods
41
Figure 3.3: Uncured EVA (courtesy to SINOVOLTAICS).
3.2.1.1 Hot press
The hot press was made up of two hot plates: the top one being fixed and the
bottom one movable. In order to achieve the desired thickness a bespoke
mould was designed, as illustrated schematically in Figure (3.4). The mould
was of a rectangular shape with dimensions of 80 x 80 x 0.5 mm; consisting of
a top plate for uniform distribution of the force, bottom plate and spacer. A
release film was used to make sure the EVA samples did not stick to the
mould and spacer. The spacer was used to control the thickness of the
samples. Before using the mould, all the mould parts and the release film
were cleaned with laboratory grade acetone. Uncured EVA (Figure (3.3)), was
cut into small sheets with dimensions of 80 × 80 mm to fit in the mould. The
samples were put in the mould as separate pieces to prevent air entrapment
and bubble formation. The load was applied to the mould by a hydraulic
system.
Experimental Methods
42
Figure 3.4: Schematic view of the mould used in the hot press.
A temperature of 150°C and duration of 20 minutes were suggested as
optimum curing conditions by the EVA supplier. The influence of different
pressures and release films on bubble formation, surface flatness and
uniformity of thickness was then investigated. PET film was found to be the
best release film, but it needed to be removed in hot water for best results.
However, a uniform thickness was not achieved and the thickness fluctuates
around 10%. Table (3.1) presents a comparative investigation of different
release films.
Using a hot press was not an ideal technique due to the uncontrollable
heating and cooling rate, inaccurate load control system (± 100 N) and most
importantly the imperfect and non-parallel top and bottom plates which
prevented the load to be applied to the mould uniformly, resulting in large
thickness variation and poor quality cured samples. Therefore, further
investigation was carried out to find a better method for EVA curing.
Experimental Methods
43
Table 3.1: Release films and release agents used for sample preparation.
After further investigation to find a better curing technique due to
unsuitability of the hot press technique, the autoclave was found to be a
robust technique, as discussed in the next section.
3.2.1.2 Autoclave
Autoclaves are used for numerous applications, such as to cure coatings, grow
crystals under high temperatures and pressures, manufacture carbon fibre
composite, due to the capabilities below.
Air removal (vacuum)
Controllable heating rate
Controllable pressure
An autoclave creates a vacuum atmosphere where the mould is and is
separated from the pressure chamber with a vacuum bag. The autoclave
enables the user to control various parameters including, the temperature of
the hot plate, the heating rate and the pressure applied to the mould.
Release Film
Problem
Solution
Picture of EVA samples
Polytetrafluoroethylene
(PTFE) film
Very
sticky,
Leaves
residues
on the
sample
Polyethylene
terephthalate (PET) film
Very sticky
Peel off
the film in
hot water
First try
After applying
solutions
Forms
bubbles
Increase
the
pressure
Release mould spray –
Long lasting PTFE
mould
Very sticky
Release liquid for
epoxy-acrylique-
polyester
Rough
surface
.
Experimental Methods
44
Figure (3.5) presents a schematic view of the autoclave used in this work. The
mould is composed of a steel top plate and a steel spacer with a thickness of
0.5 mm. On each side of the spacer, there is a sheet of non-stick release film
to prevent adherence of the EVA to the metallic parts of the mould. The
mould is covered with a vacuum bag in order to create a vacuum atmosphere
around the mould to prevent bubble formation in the sample. A second
chamber over the vacuum bag was created by adding a cover plate to apply a
uniform pressure to the top of the mould.
Figure 3.5: Schematic view of the autoclave.
The following procedure was used in curing the EVA in the autoclave.
The mould was placed on the hot plate and covered with a breathable film
and the vacuum bag film,
A vacuum was applied between the vacuum bag and the hot plate,
A uniform pressure of 0.21 MPa was applied to the top plate of the mould,
The final temperature of 150°C was set at the heating rate of 4°C/min,
After reaching the final temperature of 150°C, samples were kept at this
temperature for curing,
The mould was cooled down by natural convection while exposed to room
temperature.
Experimental Methods
45
Figures (3.6)-(3.7) demonstrate the preparation of the autoclave and mould
positions on the autoclave hot plate.
Figure 3.6: Autoclave preparation for EVA curing.
Figure 3.7: (a) Autoclave with samples, spacers and top plate, (b) Autoclave during EVA curing.
The autoclave cure process took approximately two hours and two samples
with dimension of 300 × 80 × 0.5 mm were cured at a time. The quality of the
samples cured in autoclave was significantly better than that of the hot press
cured EVA. The thickness was uniform, with a variation less than 2%, and the
Experimental Methods
46
samples were transparent. However due to the unavailability of the
autoclave, uncontrollable cooling system and cooling the samples by natural
convection this method was found not to be suitable and as consistent as
laminator in curing the EVA samples. In the next section the laminator is
described.
3.2.1.3 Laminator cured EVA
After trying different curing techniques it was found that hot press was not
suitable to cure EVA due to its inability to create vacuum atmosphere,
uncontrollable cooling rate and not parallel plates. Autoclave was also not an
ideal method to cure the EVA due to uncontrollable cooling process. Finally, a
laminator was used as the most reliable technique which is the common way
to laminate PV modules in the PV industry. The base material used in this
research was a laminator cured EVA copolymer with 33% vinyl acetate, which
was supplied in 0.5 mm thick sheets (provided by Ecole Polytechnique
Fédérale de Lausanne (EPFL)). The curing of EVA was performed in a
controlled atmosphere in a 3S Swiss Solar Systems laminator S1815
manufactured by 3S Modultec, Switzerland. The curing process including the
steps below was performed by (Heng-Yu Li, Laura-Emmanuelle Perret-Aebi,
Ricardo Theron, Christophe Ballif, Yun Luo 2010; Li et al. 2013).
1. Preheating for approximately 300 seconds (plate temperature at 140°C),
2. A pressure of 0.1 MPa was applied to the laminates at the beginning of the
curing time under vacuum at about 100 Pa,
3. Curing time was set to 1300 seconds,
4. The pressure was removed and the cooling process started.
In the next section the advantage and disadvantage of the curing methods
are discussed.
3.2.1.4 Comparison of manufacturing methods
Various methods were used to cure the EVA. Table (3.2) presents the
approach taken to select the most suitable curing method in this research.
Experimental Methods
47
The evaluation of manufacturing process indicated that a laminator is the
most suitable method of manufacturing. Although hot press and autoclave
manufactured samples were used in the early stage of this work all samples
used to produce results in this study were manufactured using the laminator.
3.2.2 Storage of samples
In order to minimize stored samples from degradation factors (elevated
temperature, UV and humidity) the EVA sheets were stored in a dark
desiccator at room temperature (22±3°C) and humidity around 15% RH after
manufacture. In order to include the influence of the storage conditions on
the test results, control samples were tested along with the aged samples in
the experiments. Control samples were kept in the desiccator during all
ageing time and compared against the aged samples.
3.3 Ageing conditions
To determine the effect of environmental stresses on polymeric materials
under controlled conditions the real environmental conditions experienced by
PV modules were simulated in the laboratory. Artificial ageing was used since
ageing in field conditions is prohibitively time consuming and difficult to
control. Since the main degradation factors are elevated temperature, UV and
humidity/moisture, to conduct a thorough investigation and cover all the
environmental degradation factors the ageing has been classified in to three
main groups as thermal ageing, UV ageing and damp-heat ageing, as
described below.
Experimental Methods
48
Table 3.2: Overview of the curing techniques.
Technique
Disadvantage
Advantage
Problems
Selected
Hot press
No controllable
heating/cooling rate, No
precise pressure
controlling, No parallel
pressing plates, No
vacuum atmosphere
_
Nonuniform
thickness, bubble
formation, low
transparency
No
Autoclave
No controllable cooling
rate, Limited sample
curing size
Vacuum
atmosphere,
precise pressure
controlling,
controllable
heating rate
Great amount of
time and energy
required,
properties
affected by
improper cooling
No
Laminator
_
Controllable
uniform pressure
&
heating/cooling
rate, Vacuum
atmosphere,
large curing plate
_
Yes
3.3.1 Thermal ageing
In order to investigate the effect of heat on the mechanical and thermal
properties of EVA and its thermal degradation the EVA sheets were aged in a
dark laboratory oven at 85°C for up to 80 days. Figure (3.8) shows the thermal
ageing condition. A batch of EVA samples was also aged at 65°C to be
compared to the UV aged samples. The samples were kept in dark
Experimental Methods
49
atmosphere away from light and humidity in a desiccator to exclude the
effects of UV and humidity and to isolate the effect of heat.
Figure 3.8: Thermal ageing conditions in the laboratory oven.
3.3.2 UV ageing
The effect of UV exposure on the properties and photodegradation of the EVA
was investigated by ageing in a Q-Sun, 1-Xenon up to 800 hours using a
daylight filter. The UV intensity was 0.68 W/m2 at 340 nm and 50°C. This
intensity is equal to July midday sunlight in Colorado, USA as mentioned in Q-
Sun manual. For the sake of comparison, a batch of EVA samples were kept in
a dark desiccator at room temperature as control samples and another batch
was aged in a dark laboratory oven at 65°C as stated in section (3.3.1). The
samples used for the analysis of UV effect included the UV irradiated side (top
side) of the UV aged samples, the UV non-irradiated side (back side) of the UV
aged samples, thermally aged sample (no UV) and control sample (no UV, no
humidity, room temperature).
3.3.3 Damp-heat ageing
In order to study the presence of moisture diffusion in the EVA and the impact
of absorbed moisture on its mechanical properties and morphology, samples
were aged in an environmental chamber at 85°C-85% RH which is commonly
used for material qualification testing of PV applications (Hülsmann et al.
010 20 30 40 50 60 70
70
80
90
100
Time (hr)
T (°C)
010 20 30 40 50 60 70
0
5
10
Time (hr)
RH (%)
Experimental Methods
50
2013) and by using a potassium chloride (KCl) salt solution in a sealed
chamber to obtain conditions of 85±3% RH at room temperature (22±3°C).
Figures (3.9) and (3.10) show the damp heat ageing conditions in the
environmental chamber and the salt solution chamber respectively. DMA and
DSC were conducted on samples aged up to 9 weeks. The amount of moisture
absorbed as a function of time was measured by recording the samples’
weight every 24 hours in the early ageing period and then every 48 hours. This
was to enable the study of moisture diffusion kinetics and saturation
equilibria.
Figure 3.9: Damp heat ageing conditions in the environmental chamber.
Figure 3.10: Ageing conditions in the salt solution chamber.
010 20 30 40 50 60 70 80 90 100
70
80
90
100
Time (hr)
T (°C)
010 20 30 40 50 60 70 80 90 100
70
80
90
100
Time (hr)
RH (%)
010 20 30 40 50 60 70
0
10
20
30
40
50
Time (hr)
T (°C)
010 20 30 40 50 60 70
50
60
70
80
90
100
Time (hr)
RH (%)
Experimental Methods
51
3.4 Experimental techniques
Krauter et al. (2011) reported that a single test cannot be used as the only
measure to characterise durability, for instance there have been cases with
perfectly cured EVA with poor mechanical properties. In order to carry out a
comprehensive investigation of the material properties and the ageing impact
on them the characterisation of EVA was separated into different parts. The
general approach in order to address the research aims was to investigate
various properties and characteristics of the EVA before and after ageing.
Thus, degradation rate and the effect of ageing on the EVA properties were
investigated as a function of ageing duration. These properties include:
The rate of thermal degradation as a function of temperature and ageing
duration.
Viscoelastic mechanical properties as a function of temperature and
artificial ageing.
The thermal and morphological properties, such as glass transition
temperature (Tg) and degree of crystallinity (Xc) as a function of ageing
duration.
The photodegradation rate as a function of exposure duration in the case
of UV ageing.
The rate of moisture absorption as a function of temperature.
A number of different experimental techniques were used to enable us to
investigate the above mentioned properties as listed below.
Differential Scanning Calorimetry (DSC); for investigation of the structure
and state of the material as the temperature is changed
Thermo-gravimetric Analysis (TGA) to investigate the evolution of
degradation products from a sample exposed to changes in temperature
and thus, the reaction rates
Dynamic Mechanical Analysis (DMA) to study the viscoelastic mechanical
properties of the material and their relation with the ageing conditions
Experimental Methods
52
Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy
(ATR-FTIR) to investigate the rate of the evolution of degradation products
and chemical changes as the material aged.
Gravimetric and Water Vapour Transmission Rate (WVTR) testing to
calculate the moisture absorption rate when exposed to a damp heat
environment.
The experimental techniques used in this work are explained in more detail
below.
3.4.1 Differential Scanning Calorimetry (DSC)
DSC is a calorimetry method which measures heat flow as a function of
temperature associated with thermally active transitions such as Tg and Xc. In
this study DSC was used to investigate the impact of ageing on the thermal
properties (Tg) and structural changes (Xc). The relative crystallinity, Xc, of the
samples was calculated as a function of ageing duration through
(3.1)
where
is the fusion enthalpy of a perfect polyethylene (277.1 J/g) crystal
and is the enthalpy of fusion of the EVA samples, respectively (J.
Brandrup 1999).
All DSC experiments were conducted in a nitrogen atmosphere with a flow
rate of 50 ml/min using a TA-Q10 (TA Instruments, USA). In order to study the
unaged and aged material and also the pure material after erasing the
thermal history of the samples there was a need to program the DSC test
steps. The DSC program used to evaluate the behaviour of the previously
cured and aged EVA samples was a heat-cool-heat cycle based on ASTM-D
3418-08. In this study the first heating was at 10°C/min, from -75°C to 250°C.
Experimental Methods
53
The temperature was held at 250°C for 5 min, to make sure the sample is
stabilized, and then cooled down at -10°C/min to -75°C and held at this
temperature for 5 min. Finally the sample was heated again to 250°C at
10°C/min. It should be mentioned that the first heating was to investigate the
impact of ageing on EVA and the second heating was to erase the thermal
history regarding the curing, storing consitions and the ageing inorder to
investigate the properties of EVA indipendant from of these effects. Samples
were cut into circular disc shapes weighing around 7 mg. All DSC experiments
were carried out in hermetic Aluminium pans and all the thermograms are
Exo Up.
3.4.2 Dynamic Mechanical Analysis (DMA)
DMA is a technique where a small deformation is applied to a sample in a
cyclic manner in order to investigate the properties of viscoelastic materials.
This allows the material’s response to stress, temperature and frequency to
be studied. DMA works by applying a sinusoidal deformation to a sample of
known geometry. The sample can be subjected to a controlled stress or strain.
For a known stress, the sample will then deform a certain amount. How much
it deforms is related to its stiffness. DMA measures stiffness and damping,
these are characterised by modulus and tan(δ). The storage modulus (E’) and
loss modulus (E’’) are the measures of the sample’s elastic and viscous
behaviour respectively. The ratio of the loss modulus to the storage modulus
is tan(δ) which is a measure of the energy dissipation of a material.
Since EVA is a viscoelastic material it is important to apply a technique which
is able to measure the viscous and elastic properties of EVA. In this research
DMA was performed to investigate the temperature dependant viscoelastic
properties of the EVA. Samples were loaded in tension at 1 Hz while the
temperature was ramped from -70°C to 100°C at a heating rate of 5°C/min.
DMA is tensile test used instead of shear test due to the capability of tensile
test in temperatures below the glass transition temperature. DMA shear test
is better test in temperature above the glass transition temperature.
Experimental Methods
54
3.4.3 Thermogravimetric Analysis (TGA)
TGA is a thermal analysis technique which measures the amount and rate of
change in the weight of a material as a function of temperature or time in a
controlled atmosphere. TGA measurements are used primarily to determine
the composition of materials and to predict their thermal stability up to
elevated temperatures.
Heat is one of the degradation factors which influences the performance of
EVA as encapsulation material in PV modules, therefore, it is important to
understand its thermal degradation and investigate the thermal stability. In
this regards TGA was used as the technique in the present study. All
experiments were performed in a nitrogen atmosphere with a flow rate of
100 ml/min. TGA analysis under dynamic conditions was carried out using a
TA SDT-600 (TA Instruments, USA) at heating rates of 5, 10, 15 and 20°C/min,
recording mass loss and the rate of mass loss as functions of temperature.
Samples were cut into circular disc shapes weighing around 15 mg. All TGA
experiments were carried out in platinum pans.
3.4.4 Attenuated Total Reflectance-Fourier Transform Infrared
Spectroscopy (ATR-FTIR)
Fourier Transform Infrared (FTIR) spectroscopy is a technique that is often
used to study the chemical changes that occur in a material as a function of
temperature and/or time. FTIR spectroscopy is a reliable fingerprinting
method. Many substances can be characterised, identified and also
quantified. Traditionally IR spectrometers have been used to analyse solids,
liquids and gases by means of transmitting infrared radiation directly through
the sample. Attenuated Total Reflectance (ATR) has in recent years
revolutionized solid sample analysis owing to its capability to resolve the
challenging parts of infrared analysis, namely sample preparation and spectral
reproducibility. ATR uses the property of total internal reflection resulting in
an evanescent wave. A beam of infrared light is passed through the ATR
crystal in such a way that it reflects at least once off the internal surface in
Experimental Methods
55
contact with the sample. This reflection forms the evanescent wave which
extends into the sample. The penetration depth into the sample is typically
between 0.5 and 2 micrometres.
Since the degradation factors cause structural changes in EVA it is needed to
understand these changes and investigate the result of ageing in EVA. To fulfil
this need, in this study ATR-FTIR was used to study the variation of different
chemical bonds as a result of UV and thermal ageing. ATR is not a very useful
technique to investigate chemical changes caused by moisture absorption
because information is obtained only from a few microns of the surface
(Morlat-Therias et al. 2007). ATR-FTIR spectra were obtained for the UV
irradiated side and non-irradiated side during ageing at intervals of 48-72
hours in order to monitor molecular changes due to UV ageing. The scanning
range was 500 cm-1 to 4000 cm-1 with a resolution of 8 cm-1 at 32 scans. In
order to obtain quantitative comparison of specific functional group
formation during ageing, the absorbance peak intensities were normalized
with respect to the absorbance of the methylene sequence (-CH2-CH2-CH2-) at
721 cm-1 (Eq. (3.2)). It is assumed that the amount of these sequences does
not change with respect to changes in the other bands (Copuroglu & Sen
2005).
(3.2)
In order to eliminate the effect of the air present in the spectroscopy
chamber, the spectra for air was subtracted from the final results.
Experimental Methods
56
3.5 Moisture diffusion coefficient measurement methods
3.5.1 Gravimetric measurments
Moisture diffusion in EVA films was investigated by the gravimetric method in
which weight changes in samples were measured as a function of time using
an OHAUS PA214 Pioneer Analytical Balance (Ohaus Corporation, USA) with a
precision of 0.1 mg. Samples were taken out of the environmental chamber
every 24 (early stage of ageing) and 48 hours to be weighed. Care was taken
to make sure that the balance was clean and accurately calibrated and zeroed
prior to each set of weighing.
3.5.1.1 Material and sample preparation
The unaged material was cut into square samples of 70×70 mm with a
thickness of 0.5 mm, in order to perform gravimetric tests. To calculate the
error 5 samples (free standing film) were prepared for each test.
3.5.1.2 Sorption measurements
The sorption experiment was carried out in a calibrated environmental
chamber at the conditions described in section (3.3.3). During all damp-heat
experiments the moisture contents were determined gravimetrically based on
Eq. (3.3) (Abdelkader & White 2005).
(3.3)
where Mt is the moisture content at time t, mt is the specimen’s weight at
time t and mdry is the weight of the dry specimen. The assumption, therefore,
is that all mass change in the sample can be attributed to change in the
amount of absorbed moisture only.
Experimental Methods
57
3.5.2 MOCON Water Vapour Transmission Rate technique to
measure diffusion coefficient
An alternative technique that has been used to obtain the diffusion coefficient
of EVA is by using MOCON® water vapour devices. MOCON® measures the
Water Vapour Transmission Rate (WVTR). Assuming Fickian diffusivity WVTR
can be described by Eq. (3.4).
(3.4)
where D is the moisture diffusion coefficient determined by the time required
to reach steady state, cs is saturation concentration which is determined by
the steady state WVTR, l is sample thickness, and t is time.
As Figure (3.11) illustrate the MOCON® water vapour test instrument consists
of one or more test cells. The test sample divides the cell into two chambers,
one with water vapour at a constant relative humidity (RH) and the other with
a nitrogen carrier gas to produce a dry environment. As soon as the water
vapour molecules permeate through the film they are picked up by the
nitrogen carrier stream and taken to the detector. The flow rate of the
nitrogen stream keeps the carrier side of the film at virtually 0% RH, thus
maintaining a constant flow gradient throughout the test. A pulse modulated
infrared detector is used to measure the amount of water vapour in the
nitrogen carrier stream. The WVTR is then calculated from the amount of
moisture in the stream, the carrier gas flow rate and the area of film sample
being tested. The water vapour permeability can then be calculated using the
additional input of driving force and sample thickness.
Experimental Methods
58
Figure 3.11: Diagrammatic side view of WVTR test cell (Copyright 2012 MOCON® Inc).
In order to obtain the diffusion coefficient of our EVA via the WVTR technique,
samples were sent to RDM TEST EQUIPMENT® laboratory, the UK. All the
samples were analysed on a Mocon Permatran-W Water Vapour Permeability
Instrument (see appendix A). The test conditions are presented in Table (3.3).
Table 3.3: Water Vapour Transmission test condition provided by RDM TEST Equipment®.
Test Gas
Water Vapour
Test Temperature
40 (°C)
Test Gas Concentration
NA
Carrier Gas
Nitrogen
Test Gas Humidity
100% RH
Carries Gas Humidity
0% RH
3.5.3 Comparison of moisture diffusion coefficient measurement
methods
Gravimetric measurement of moisture diffusion coefficient is a direct method
where samples are exposed to humidity and the weight changes in samples
were measured. Then moisture diffusion coefficient is determined via Fick’s
second law after reaching steady state where the diffusion is Fickian.
Experimental Methods
59
WVTR is an alternative method where the samples are placed in a MOCON®
device at a controlled environment and exposed to humidity after being
degassed. This method is performed in an environment where it is reassured
there is no humidity in the samples prior to the test and the samples are
isolated from external environmental factors such as UV and O2. The device
measures water vapour transmission rate and moisture diffusion coefficient
comes from Eq. (3.4).
The MOCON® condition is ideal and the test time is much shorter where in
gravimetric test the samples are exposed to elevated temperature and
possibly to other degradation factors for a much longer time. It should also be
mentioned that it is reported that using MOCON® devices are not suitable for
barrier films with WVTR lower that 10-3 (g/m2 d) (Hülsmann et al. 2009).
3.6 Summary
The following section summarises the previously described methodology
adopted for the achievement of the objectives of this research project. In
regard to the procedures mentioned in Sections (3.1)-(3.4), the following
considerations could be made:
Different curing techniques were investigated to prepare EVA samples
for analysis and the lamination technique was found to be the most
reliable. The EVA used in this work was provided by Ecole
Polytechnique Fédérale de Lausanne (EPFL) in form of cured sheets
with 33% of vinyl acetate and 80% of gel content.
The EVA sheets were stored with minimal degradation factors, in a
dark desiccator at room temperature and low humidity (around 15%
RH) prior to testing or ageing.
In order to investigate the influence of the environmental stresses
(elevated temperature, UV, humidity), EVA was aged using artificial
ageing under the conditions described in section (3.3).
Experimental Methods
60
Various experimental techniques were used to characterise the effect
of ageing on the EVA. These techniques enabled the investigation of
structure and state of the material (DSC), the thermal stability and
evolution of degradation products (TGA), the viscoelastic mechanical
properties of the material (DMA), the rate of the evolution of
degradation products and chemical changes as the material is aged
(ATR-FTIR) and the moisture absorption rate (gravimetric and
MOCON®).
This chapter reported the methodologies and procedures which have
been used for the achievement of the objectives (described in section
(1.5)).
Sections (3.2)-(3.4) outline the approaches that were used to achieve
the results presented in Chapters 4 to 6.
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
61
Chapter 4
Examination of the Response of Ethylene-vinyl
Acetate Film to Thermal Ageing
4.1 Introduction
The practical importance of EVA has meant that many studies have been
carried out to investigate its chemical structure and thermal degradation, but
very few studies have comprehensively investigated its performance under a
PV module’s working condition (Pern & Glick 1997; Agroui et al. 2007). The
previous studies have illustrated the main mechanisms through which
degradation of EVA occurs. However, the influence of thermal ageing on the
mechanical properties and morphology of EVA used as PV encapsulation
material, with high vinyl acetate content (>30% VAc), has not been
investigated thoroughly and there is a missing link between the ageing
process and consequences of that ageing within the context of its mechanical
behaviour. This study is motivated by the need to investigate the thermal
degradation and viscoelastic properties of unaged and aged EVA films and the
generation of inputs to models that can then predict the lifetime of
encapsulants in-service.
Thermal degradation in polymers is related to significant changes in molecular
state as the thermal energy provides a driving force for chemical reactions.
Elevated temperatures cause the components of the long chain backbone of
the polymer to separate (molecular chain scission) and react with one another
to change the properties of the polymer. Solid polymeric materials undergo
both physical and chemical changes when heat is applied; this will usually
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
62
result in undesirable changes to the properties of the material (Pielichowski &
Njuguna 2005).
This chapter seeks to address the need to understand the changes in
mechanical properties on thermal ageing by examining the link between the
chemistry, the structure and the mechanical behaviour throughout the
duration of an accelerated thermal only ageing test. Figure (4.1) illustrates the
approach taken and techniques used in this investigation.
Figure 4.1: Flowchart of investigation of EVA’s response to thermal ageing.
The approach taken in this study was to fully characterise the material
behaviour and correlate this with measures of thermal degradation by
measuring weight loss, determining the kinetics of reactions and investigating
structural changes and changes in the mechanical properties, which can be
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
63
deduced from calorimetry techniques and dynamic mechanical analysis
respectively.
4.2 Theory of kinetics of degradation
The modelling of thermally activated reactions is of concern in many
engineering applications and has been the subject of significant research
(Starink 2003; Vyazovkin 2006; Órfão & Martins 2002; Muralidhara &
Sreenivasan 2010; Morshedian 2009; Wang et al. 2004; Gu et al. 2012; Salin &
Seferis 1993; Hamed Gholami, Gholam Reza Razavi 2011). In order to
investigate the kinetics of thermal degradation of EVA and calculate the
kinetic parameters the Kissinger–Akahira–Sunose (KAS) method was used.
This is also known as the generalised Kissinger method (Starink 2003).
The rate of material conversion (d
⁄dt) of a solid state process has the form
presented by Eq. (4.1)
(4.1)
where
is the conversion degree, t is the reaction time, K is a reaction rate
which depends on the temperature (T) and f(
) is a kinetic model function
which is a first order case for thermal degradation of EVA (Bianchi et al. 2011).
In this study the conversion degree is defined by Eq. (4.2)
(4.2)
where m0, mt and m∞ are initial sample mass, sample mass at time t and
sample mass at the end of the reaction, respectively. As the reactions to be
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
64
considered here are related to thermal processes, K(T) is assumed to be of an
Arrhenius form, such that Eq. (4.1) becomes.
(4.3)
where a is a pre-exponential (frequency) factor, E is the activation energy and
R is the universal gas constant. For non-isothermal experiments carried out
with a constant heating rate,
, it is possible to arrive at Eq. (4.4).
(4.4)
E can then be obtained from non-isothermal data without choosing the
reaction model. The generalised Kissinger’s method is an established
technique to calculate the activation energy based on the rate equation at the
maximum reaction rate. At this point,
allows for the calculation of E
at the maximum rate of degradation, when the heating rate is constant, and
the time and temperature derivatives of weight loss are linearly related.
Therefore, data can be plotted as a function of time or temperature and
analysed using Eq. (4.5) which comes from Eq. (4.3) (Sánchez-Jiménez et al.
2008).
(4.5)
where Tp is temperature at the maximum degradation rate and is the
heating rate. It is possible, therefore, to plot
vs. 1/Tp and obtain the
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
65
activation energy and pre-exponential factor from the slope and intercept
respectively. Figure (4.2) shows the flowchart of lifetime calculation. The
advantage of this method is that it enables the activation energy to be
obtained independently of the kinetic model (Starink 2003).
Figure 4.2: Flowchart of the investigation the weight loss of EVA regarding the thermal
degradation only.
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
66
4.3 Results and discussion
The thermal ageing experiments were performed according to the ageing
conditions and experimental techniques described in sections (3.3) and (3.4)
respectively. The results of the TGA, DMA and DSC are presented below.
4.3.1 Thermogravimetric Analysis (TGA)
Figure (4.3) shows TGA curves corresponding to dynamic experiments carried
out at a range of heating rates. The results show evidence of a two-step
thermal degradation process. The first stage, completed at around 370°C
(rescaled in Figure (4.4)), can be attributed to a deacetylation process in the
vinyl acetate fraction, since it is reported that acetic acid is lost in the first
thermal degradation step, leaving a polyunsaturated linear hydrocarbon
(Dolores Fernández & Jesús Fernández 2007; Zanetti et al. 2001; McGrattan
1994). The (TG)/FTIR investigation of EVA’s pyrolysis (Marcilla et al. 2005)
showed that the evolved gas were mainly acetic acid and small quantities of
CO, CO2 and CH4. The second stage has previously been identified as complete
chain scission of the residual main chain (within temperature range of 380-
480°C). As should be expected, the temperature at which the reaction is
complete increases as the heating rate increases.
Figure 4.3: TGA thermograms of EVA at different heating rates (weight percentage versus
temperature).
0100 200 300 400 500 600
0
20
40
60
80
100
120
T (°C)
W(%)
heating rate: 5°C/min
heating rate: 10°C/min
heating rate: 15°C/min
heating rate: 20°C/min
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
67
Figure 4.4: Rescaled TGA thermogram of EVA at different heating rates (weight percentage
versus temperature).
Since the maximum temperature that photovoltaic systems operate is around
85°C, chain scission is not considered as significant in this work, therefore, the
focus is on the deacetylation of EVA. Figure (4.5) shows the derivative of
weight as a function of temperature, illustrating the temperature of the peak
degradation rate, which is used to calculate the activation energy and pre-
exponential factor based on the generalized Kissinger’s method, described
earlier. Table (4.1) shows Tp for the first weight loss at different heating rates
which were used in Kissinger method to calculate the activation energy.
Figure 4.5: Derivative of weight loss to temperature for unaged EVA at different heating rates.
050 100 150 200 250 300 350 400
0
20
40
60
80
100
120
T (C)
W (%)
Heating rate 5 C/min
Heating rate 10 C/min
Heating rate 15 C/min
Heating rate 20 C/min
0100 200 300 400 500 600
-0.5
0
0.5
1
1.5
2
T (°C)
d(W)/dT(%/°C)
heating rate: 5°C/min
heating rate: 10°C/min
heating rate: 15°C/min
heating rate: 20°C/min
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
68
Table 4.1: Temperature of peak degradation rate for aged and unaged EVA samples at
different heating rates.
Heating rate
(°C/min)
Temperature of peak degradation rate
TP (°C)
Unaged
Aged for 40
days
Aged for 60
days
Aged for 80
days
5
335.24
336.53
336.61
336.58
10
346.61
348.80
348.60
348.72
15
354.19
355.36
365.05
356.31
20
359.17
361.83
362.28
362.03
Figure (4.6) show the relationship between
and
for one heating rate. It
can be seen that the activation energy and pre-exponential factor were
obtained from the slope and intercept of line fitting to this plot using Eq. (4.5).
The calculated values are listed in Table (4.2). Experimental data were fit to
Eq. (4.4) in order to find the pre-exponential factor. Figure (4.7) shows this fit
and the best agreement between the experimental and calculated curves
around the peak degradation rate temperature, by definition, but beyond this
point differences are observed, which could in part be attributable to other
processes occurring, such as dehydration. The generalised Kissinger method
was also applied to thermally aged EVA. Figure (4.8) shows the calculated
activation energy for the aged EVA and indicates that there is no clear effect
of ageing on the activation energy.
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
69
Figure 4.6: Plot of ln(β/Tp2) versus 1/(Tp) for unaged EVA based on Kissinger’s method.
Figure 4.7: Experimental and calculated TGA curves.
Table 4.2: The calculated kinetic parameters for unaged EVA.
Sample
R2
Slope of the line
E (kJ/mol)
a (min-1)
Unaged
0.9995
-20944
174.1284
2.4625×1014
1.58 1.59 1.6 1.61 1.62 1.63 1.64x 10-3
-11.2
-11
-10.8
-10.6
-10.4
-10.2
-10
1/Tp (1/K)
Ln(/Tp2)
50 100 150 200 250 300 350
0
20
40
60
80
100
120
T (°C)
W (%)
Experimental result
Optimised fitting
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
70
Figure 4.8: Activation energy calculated for unaged and thermally aged EVA.
The determination of the activation energy and the pre-exponential factor
enables predictions to be made about the thermal degradation and associated
weight loss over the lifetime of a module. Assuming that the material was at a
module operating temperature of 85°C (which is the temperature of the
material near to the cells as they generate heat during the conversion of
photons to electronic potential), the determined activation energies and pre-
exponential factors were used to determine the conversion of material as a
function of time through the solution of Eq. (4.3). A failure criterion is
suggested 5% weight loss from a Thermogravimetric Analysis (TGA)
(Křižanovský & Mentlík 1978; Flynn 1995). In the case of the PV modules the
typical lifetime of a module is usually considered to be around 30 years and
Figure (4.9) shows that the weight loss in this period is predicted to be around
1%, which has little impact on the module performance. This also indicates
that a more accurate method of determining the kinetics of degradation is
unnecessary.
010 20 30 40 50 60 70 80
150
155
160
165
170
175
180
185
190
195
200
Ageing duration (day)
E (kJ/mol)
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
71
Figure 4.9: Weight loss of EVA versus time in the case of exposure to 85°C with standard
deviation of 6.6521e-04.
4.3.2 Dynamic Mechanical Analysis (DMA)
Figure (4.10) shows the storage modulus (E’) determined as a function of
temperature at a frequency of 1 Hz. As the temperature increases the curves
show the characteristic glassy, rubbery and viscous regions of a viscoelastic
material (Stark & Jaunich 2011).
Figure 4.10: Storage modulus vs temperature for aged and unaged EVA.
010 20 30 40 50 60 70 80 90 100
96.5
97
97.5
98
98.5
99
99.5
100
Weight loss (%)
Time (year)
-80 -60 -40 -20 0 20 40 60 80 100
10-1
100
101
102
103
104
T (°C)
E (MPa)
unaged
Aged for 40 days
Aged for 60 days
Aged for 80 days
Transition
Glassy Rubbery Viscous
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
72
Significant changes are seen in the storage modulus over the observed
temperature range, with 3 decades difference between the moduli at -75°C
and 95°C. There is a sharp decrease at around -30°C, which can be attributed
to the glass transition and then another stepped decrease between 40°C and
65°C, indicating some crystal melting . Figure (4.11: a) shows tan(δ) versus
temperature, which can be used to emphasise the Tg of the samples as shown
in Figure (4.11: b) (Varghese et al. 2002). It can be seen that Tg is
approximately -20°C.
(a)
(b)
Figure 4.11: (a) tan(δ) vs temperature for aged and unaged EVA, (b) Glass transition
temperature versus ageing duration for EVA based on tan(δ).
-80 -60 -40 -20 0 20 40 60 80 100
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
T (°C)
tan()
unaged
Aged for 40 days
Aged for 60 days
Aged for 80 days
010 20 30 40 50 60 70 80 90
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (day)
Tg (°C)
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
73
In order to further investigate the effect of ageing on the mechanical
properties of the EVA, the storage modulus at a given temperature was
plotted against the ageing time for the sample. Figure (4.12: a-d) shows this
for four temperatures. One can see that E’ reduces with increasing time of
ageing, although the confidence in the fit and the calculated gradient
decreases with increasing temperature as presented in Table (4.3). As a way
of approximating the general trend, E’ vs temperature for four temperature
values was consolidated into one plot as a mean where the mean E’ is the
average of E’ at the complete range of experimental temperature. This is
shown in Figure (4.13), which shows a clear monotonically decreasing storage
modulus as the material ages. Table (4.4) presents the parameters of best fit
to changes of mean E’ versus ageing duration. Further, the variation in
modulus depends on temperature; as the measurement temperature
increases, the dependence on the ageing duration weakens. That is to say, the
storage modulus reduction rate, when measured at a given temperature,
varies more slowly when measured at a higher temperature than a lower
temperature. It is worth emphasising that this is for the same ageing condition
of 85°C. It therefore appears that thermal ageing can result in significant
changes in the mechanical properties, but considering the very small changes
in weight loss, this is unlikely to be associated with chemical changes and any
subsequent material volatilisation.
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
74
(a)
(b)
(c)
(d)
Figure 4.12: Storage modulus measured at (a)-60˚C, (b) 20˚C (c), 40˚C (d) 95˚C as a function of
ageing time at 85˚C.
-10 0 10 20 30 40 50 60 70 80 90
1500
1600
1700
1800
1900
2000
2100
Time (day)
E (MPa)
-10 0 10 20 30 40 50 60 70 80 90
8
10
12
14
16
18
20
22
Time (day)
E (MPa)
-10 0 10 20 30 40 50 60 70 80 90
3
4
5
6
7
E (MPa)
Time (day)
-10 0 10 20 30 40 50 60 70 80 90
0.5
0.6
0.7
0.8
0.9
1
E (MPa)
Time (day)
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
75
Table 4.3: The fitting parameters for storage modulus at different fixed temperatures based
on Figures (4.12: a-d).
T (°C)
Gradient
Intercept
R2
-60
-5.8400
2051
0.93
20
-0.02667
5.521
0.93
40
-0.0155
2.652
0.85
95
-0.0036
0.968
0.44
Figure 4.13: Average storage modulus versus ageing time for thermally aged EVA.
Table 4.4: The best fitting parameters of the average storage modulus versus ageing duration
based on Figure (4.13), fitting function: y=a1x2+a2x+c.
a1
a2
Intercept
R2
-0.006341
-0.3843
367.1
0.9987
-10 0 10 20 30 40 50 60 70 80 90
260
280
300
320
340
360
380
Time (day)
mean E (MPa)
Experimental results
Best fit
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
76
4.3.3 Differential Scanning Calorimetry (DSC)
In order to understand the changes in morphological behaviour, DSC was
performed, consisting of a heating, a cooling and final heating stage
(described in section (3.4.1)). The results of these experiments can be seen in
Figure (4.14), which shows typical DSC thermograms for samples that have
been cycled. There are significant changes around 50°C and the peaks in this
region can be associated with internal structural changes or crystallisation
transition. This correlates with the step between 40°C and 60°C in the DMA
results (Figure (4.10)) which were associated with crystal melting. In the
second heating, it can be observed that the peaks have largely disappeared,
suggesting that in the process of going through the previous cycle the
structure has been eliminated and has not reformed during the cooling
process. In Figure (4.15) the first heating thermograms of the unaged and
aged EVA are reproduced with offsets added for the sake of comparison. The
melting transition of the ethylene segment has been accepted as being
associated with a peak with a shoulder which is observed between 40°C to
70°C (Li et al. 2013; Motta 1997). The glass transition, determined by a step
change in the heat flow typically between -30°C and 20°C, can be seen at
around -20°C for both aged and unaged samples, suggesting ageing has no
significant effect on Tg which is shown in Figure (4.16).
Figure 4.14: Typical DSC thermograms (heat flow versus temperature) of EVA under three
steps, heating-cooling-heating-Exo Up.
-100 -50 0 50 100 150 200 250
-1.1
-1
-0.9
-0.8
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
T (°C)
J (W/g)
1st heating
Cooling
2nd heating
Crystallization peak
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
77
Figure 4.15: DSC first heating thermograms (heat flow versus temperature) for the unaged
and aged EVA (thermograms are reproduced with offsets added)-Exo Up.
-10 0 10 20 30 40 50 60 70 80 90
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (day)
Tg (°C)
Figure 4.16: Glass transition temperature versus ageing duration for EVA.
Analysis of the first heating results shows that crystallinity decreases due to
ageing as shown in Figure (4.17), which when correlated with the changes in
mechanical behaviour, suggests that it is the changes in internal structure that
drive the changes in the storage modulus, E′ rather than the chemical changes
discussed in section (4.3.1) as shown in Figure (4.18). It can also be seen in
Figure (4.17) that there is no change in crystallisation in the second heating.
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
78
This reversion to a constant crystallisation content at the first heating cycle
points towards a reduction in the structure at higher temperatures, with a
consequent reduction in modulus and a weakening in the relationship
between ageing driven modulus reduction at higher temperatures.
-10 0 10 20 30 40 50 60 70 80 90
0
1
2
3
4
5
6
7
8
9
10
Time (day)
Xc (%)
First heating
Second heating
Figure 4.17: Crystallinity versus ageing duration after first and second heating for aged and
unaged EVA.
Figure 4.18: Storage modulus versus crystallinity for thermally aged EVA.
3.544.555.566.57
260
280
300
320
340
360
380
mean E (MPa)
Xc (%)
Examination of the Response of Ethylene-vinyl Acetate Film to Thermal Ageing
79
4.4 Conclusions
The thermal degradation of EVA has been studied using techniques that
enabled the viscoelastic properties and thermal stability to be measured. Key
findings were that the activation energy of the first stage of thermal
degradation was unaffected by the ageing process whereas storage modulus
at 1 Hz was significantly reduced with increases in temperature. Ageing was
shown to reduce the storage modulus monotonically as a function of ageing
degree, though this effect was weakened at elevated temperatures. TGA
measurements showed that chemical changes due to thermal activation were
insignificant, even over the typical lifetime of the module, but examination of
DSC results suggested that property changes could be connected to structural
modifications, specifically that thermally induced decreases in crystallinity
resulted in a reduction in storage modulus. In the field, such modules will be
subjected to light and humidity as well as elevated temperatures, and the
interaction between these effects requires investigation. Hence, now that the
influence of elevated temperature on the properties, structure and lifetime of
EVA has been investigated, in the next chapter the impact of UV irradiance on
the properties and photodegradation of EVA will be investigated.
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
80
Chapter 5
Examination of the Response of Ethylene-vinyl
Acetate Film to UV Irradiation
5.1 Introduction
Photochemical reactions are of interest in diverse applications such as
polymer degradation and solar energy capture. UV irradiation is one of the
most important environmental factors affecting the deterioration of the
mechanical strength of the encapsulant and the structural integrity of a PV
module. EVA undergoes photochemical degradation when it is exposed to UV.
UV irradiation causes discoloration and photodegradation which reduces the
spectral transmission which causes a reduction in the conversion efficiency of
the PV modules. Thus, there is a need to investigate the effect of
photodegradation, on the EVA in a PV module.
It has been reported that the main photodegradation reactions in EVA are
Norrish Ι and Norrish ΙΙ (Czanderna & Pern 1996; Morlat-Therias et al. 2007;
Lacoste et al. 1991; Glikman et al. 1986), this was shown in Figure (2.4). This
comprehensive review of the photodegradation shows that the main products
of the Norrish reactions are ester, carboxylic acid, lactone, vinyl and vinylene,
which are fully investigated in this research.
This chapter investigates the photochemical degradation of EVA, develops a
model to predict degradation and also studies the influence of the
photodegradation on the structure and mechanical properties of EVA. Figure
(5.1) illustrates the approach taken and techniques used in this chapter.
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
81
Figure 5.1: Flowchart of investigation of EVA’s response to UV ageing.
The approach taken in this research was to make a comprehensive
investigation of photodegradation chemically by spectroscopic methods (FTIR-
ATR). Once the degradation mechanisms were fully understood a predictive
model was developed to predict the photochemical degradation in time. The
influence of photodegradation on the material behaviour and properties were
also investigated.
5.2 Theory of photochemical reactions
Sheats & Diamond (1988) introduced a pair of coupled partial differential
equations (PDE) the chemical transformation in photochemical reactions.
Since EVA is highly UV absorbing as reported in the literature (Pern 1996;
Skowronski et al. 1984; Czanderna & Pern 1996; Mcintosh et al. 2009; Kempe
et al. 2011; Jin et al. 2009), the introduces PDEs can be applied in this case.
=
5.1
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
82
=
5.2
where Cr, Cp and Cm are concentration of reactant, products and medium
respectively. k is the fractional decay rate per unit intensity (Dill et al. 1975), α
is the respective molar extinction coefficients, z and t refer to position and
time respectively. The molar extinction coefficient, α, used in this study is
related to the decadic molar extinction coefficient, ε, via Eq. (5.3) (Sheats &
Diamond 1988). I(z,t) is the light intensity at depth z at time t and F(Cr)
represents a rate function.
5.3
In general, the quantities Cr, Cp and I depend on position and time, resulting in
the coupling of the differential equations. Therefore, the equations can be
articulated as
=
5.4
=
5.5
which are subjected to the boundary conditions
5.6
5.7
where is the light intensity in the medium inside the surface z0 at time t.
Equations (5.4)-(5.5) and the associated boundary conditions are taken as the
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
83
basic coupled PDEs, which describe the photoreaction in the absence of
change in surface reflectivity, scattering or diffraction.
Considering the first order case, reaction rate is first order in both Eq. (5.4)-
(5.5) and only the reactant absorbs light. Therefore, Eq. (5.4)-(5.5) become
=
5.8
=
5.9
The solutions for Eq. (5.8)-(5.9) have been reported by (Herrick 1966). The
general solutions are
5.10
5.11
where and are determined by the initial conditions
5.12
5.13
The general solutions subject to boundary conditions, Eq. (5.6)-(5.7).
In order to find k the changes of the concentration on the surface is
considered which is independent of z, therefore Eq. (5.8) becomes
5.14
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
84
The solution for Eq. (5.14) is
5.15
where k can be found by fitting Eq. (5.15) to the experimental data.
The time dependent concentration of photoproduct is related to the
concentration of photoreactant via stoichiometric coefficient, , and equation
(Sheats & Diamond 1988)
)
5.16
where λ is the ratio of the stoichiometric coefficient in the reaction.
The initial concentration can be determined by Beer-Lambert law, Eq. (5.17)
using FTIR-ATR results,
5.17
where A, ε, b and C are absorbance, molar extinction coefficient, path length
of the sample and concentration respectively. The path length is 0.65 μm
based ATR catalogue (Bruker Optics 2011), the absorbance comes from the
FTIR-ATR measurements and ε is known from (Morlat-Therias et al. 2007).
5.3 Results and discussion
The results are based on the ageing conditions and experimental techniques
described in sections (3.3)-(3.4).
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
85
5.3.1 Fourier Transform Infrared Spectroscopy in Attenuated
Total Reflectance (FTIR-ATR)
FTIR-ATR was used to study the variation of different chemical bonds as a
result of UV aging. FTIR-ATR spectra were obtained for the UV aged samples
at 50°C, including the UV irradiated side, the non-irradiated side, a thermally
aged sample at 65°C and a control sample (see section (3.2.2)).
In order to perform a thorough investigation the impact of ageing on the
individual peaks these peaks should be identified. To identify the exact wave
numbers for the specific peaks the spectra of unaged EVA (A0) was subtracted
from the spectra of the aged EVA (At) as shown in Figure (5.2). The identified
peaks are presented in Table (5.1).
Figure 5.2: Subtracted FTRI-ATR spectra (At - A0) in the domain 1800-1650 cm-1.
Table 5.1: Attribution of infrared absorption bands of EVA film.
Chemical component
Wave number (cm-1)
Phenomena
Vinyl
910
Norrish I & II
Carboxylic acid
1720
Norrish I & II
Ester carbonyl stretching
1740
Norrish I & II
Lactone
1767
Norrish I & II
1650170017501800
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
Wave number (cm-1)
Absorbance
unaged
50 hours
190 hours
270 hours
360 hours
1767 cm-1
Lactone formation 1740 cm-1
Ester elimination
1720 cm-1
Carboxylic acid
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
86
To have an accurate and more detailed view of how the absorbance changes
with ageing time as a result of UV exposure plots of absorbance versus ageing
duration were produced. Figure (5.3) shows the changes in IR spectra of
unaged and UV aged EVA at 1740 cm-1 (ester carbonyl stretching), normalized
as A1740/A721 (described as Eq. (3.2)) versus exposure time. Ester is important
in this case since one of the main products of the photodegradation of EVA is
acetic acid which come via ester elimination (Copuroglu & Sen 2005). The
absorbance for the top side of the UV aged EVA shows a sharp decrease up to
around 300 hours which is due to ester elimination. This levels off after 300
hours due to the predominant aldehyde development (Copuroglu & Sen
2005). Aldehyde and ester have the same absorbance peak identification and
aldehyde formation overlaps the ester elimination, meaning that although, no
changes are observed after 300 hours it does not mean the ester elimination
has stopped. The FTIR technique is not able to separate these two
phenomena and other techniques should be used for further investigation. A
slow decrease in absorbance is also seen in the thermally aged sample but no
noticeable changes are observed on the back side of the UV aged EVA and the
control sample, despite the fluctuation in data. The results for the back side
of the UV aged sample show that the UV has not been able to penetrate and
affect it which can be due to the presence of UV absorber. Therefore, the only
difference between the back side of the UV aged sample and the thermally
aged is the ageing temperature. The thermally aged EVA has been exposed to
a higher temperature (65°C) which is the reason for the decrease in the
absorbance for thermally aged EVA. Thus, UV has the dominant ageing effect
while the effect of temperature is minor.
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
87
Figure 5.3: Variation of absorbance at 1740 cm-1 as a function of exposure time.
Figure (5.4) shows the absorbance changes at 1720 cm-1 (carboxylic acid) as
A1720/A721 for all samples. As expected, no significant changes are observed in
the case of the control sample and the non-irradiated side of the UV aged
sample. Nevertheless, there are notable changes seen in the UV and thermally
aged samples in which the rate of changes for the UV aged sample is more
significant and shows the dominant ageing effect of UV. It was expected that
the thermally aged and back side of the UV aged samples show similar
behaviour but due to elevated ageing temperatures in thermal ageing the rate
of changes for the thermally aged sample is higher. The rate of
photodegradation comes from the changes at 1720 cm-1 versus ageing time
(Morlat-Therias et al. 2007). Table (5.2) presents the best fit parameters for
the changes in A1720 for the top side of the UV aged sample.
-200 0 200 400 600 800 1000 1200 1400
7.5
7.6
7.7
7.8
7.9
8
8.1
8.2
8.3
8.4
8.5
Time (hour)
A1740/A721
Control sample
Thermally aged sample
UV aged sample, top side
UV aged sample, back side
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
88
Figure 5.4: Variation of absorbance at 1720 cm-1 as function of exposure time.
Table 5.2: Line of best fit parameters (Figure (5.4) UV aged top side).
Sample
Wave number (cm-1)
R2
Gradient
UV aged (top side)
1720
0.9699
0.00249
Figure (5.5) shows the variation of absorbance at 1767 cm-1 (lactone) by
ageing time as A1767/A721. The absorbance for the top side of the UV aged EVA
has increased dramatically while there are no changes observed in the control
sample, thermally aged sample and the back side of the UV aged sample.
Figure 5.5: Variation of absorbance at 1767 cm-1 as a function of exposure time.
-200 0 200 400 600 800 1000 1200 1400
4.5
5
5.5
6
6.5
7
7.5
Time (hour)
A1720/A721
Control sample
Thermally aged sample
UV aged sample, top side
UV aged sample, back side
-200 0 200 400 600 800 1000 1200 1400
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
1.6
Time (hour)
A1767/A721
Control sample
Thermally aged sample
UV aged sample, top side
UV aged sample, back side
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
89
Figure (5.7) shows the absorbance change at 910 cm-1 (A910/A721) versus
exposure time. A910 is the attributed wave number for vinyl group which is an
unsaturated group and responsible for discolouration of EVA, Therefore the
rate of vinyl formation can be a measure for EVA discolouration. The
absorbance for the top side of the UV aged EVA has sharply increased while
there is a very slight increase in the case of the thermally aged EVA is
observed. As expected no changes are observed in the control sample and the
back side of the UV aged EVA which, is due to being away from the
degradation factors and not exposed to a high temperature respectively. Once
again the results indicate the dominant ageing effect of the UV in chemical
degradation of EVA and particularly in discolouration of EVA.
Figure 5.6: Variation of absorbance at 910 cm-1 as a function of exposure time.
5.3.2 Analytical investigation of photodegradation
Figure (5.7) shows the flowchart of the taken approach to predict EVA’s
photodegradation only. In order to calculate k (fractional decay rate), Eq.
(5.15) was fitted to variation of experimental data in Figure (5.3). Figure (5.8)
and Table (5.3) show the fitting and its parameters respectively.
-200 0 200 400 600 800 1000 1200 1400
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
Time (hour)
A910/A721
Control sample
Thermally aged sample
UV aged sample, top side
UV aged sample, back side
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
90
Figure 5.7: Flowchart of the investigation the lifetime of EVA regarding the photodegradation
only.
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
91
Figure 5.8: Variation of absorbance at 1740 cm-1 (normalized) on the irradiated surface as a
function of exposure time.
Table 5.3: Best fit parameters (Figure (5.8)).
Sample
k (m2/J)
Intercept
R2
Irradiated surface, UV aged EVA
29400
-0.0001533
0.949
Once k is found concentration of ester (photoreactant) can be modelled as
function of time on the irradiated surface of EVA. Figure (5.9) shows
concentration of ester as function of exposure time for the first 300 hours of
the exposure time. The results show a good agreement between the
experimental and analytical results with an error in the range 5%. The source
of error may be variation of intensity in UV chamber, thermal ageing of EVA
(described in chapter 4).
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
92
Figure 5.9: Variation of concentration of ester on the irradiated surface as a function of
exposure time-validation of analytical and experimental results.
Now the concentration of photoreactant can be predicted on the surface of
EVA in time. Figure (5.10) shows the variation of concentration of ester on the
irradiated surface of EVA in 30 years. The results show that the photoreaction
is completed on the surface after around 5 years of continuous UV exposure
at the intensity of 0.68 W/m2 however, in the field due to cycles of day and
night and various weather conditions this time will be longer in the real field
conditions. These results are in agreement with the field data presented in
(Czanderna & Pern 1996).
In order to achieve the concentration of photoproduct (carboxylic acid), λ
(stoichiometric coefficient) should be known. Figure (5.11) shows the
variation of concentration of carboxylic acid versus concentration of ester
where λ comes from the gradient of the fitted line. Table (5.4) shows the
calculated λ and the fitting parameters.
050 100 150 200 250 300
49
49.5
50
50.5
51
51.5
52
52.5
53
Time (hr)
Concentration of Ester (mol/m3)
Analytical results
Experimental results
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
93
Figure 5.10: Variation of concentration of ester (photoreactant) on the irradiated surface as a
function of time.
Figure 5.11: Variation of concentration of carboxylic acid versus concentration of ester.
Table 5.4: The fitting parameters Figure (5.11).
Gradient
Intercept
R2
-0.3113
62.44
0.98
0 5 10 15 20 25 30
0
10
20
30
40
50
60
Time (year)
Concentration of Ester (mol/m3)
32 33 34 35 36 37 38 39 40
49.5
50
50.5
51
51.5
52
52.5
Concentration of Caboxylic acid (mol/m3)
Concentration of Ester (mol/m3)
Experimental data
Linear fit
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
94
Now the concentration of photoproduct can be predicted on the surface of
EVA. Figure (5.12) shows the variation of carboxylic acid concentration on the
irradiated surface of EVA in 30 years. The results show a similar trend to the
variation of ester which indicates that the photoreaction is completed on the
surface after 5 year.
Figure 5.12: Variation of concentration of carboxylic acid (photoproduct) on the irradiated
surface as a function of time.
Since the experimental data were only achieved on the surface of EVA, it is
not possible to predict the photodegradation in depth and validate the model
however, it should be mentioned that in the field the top side of EVA is more
susceptible to photodegradation and the discolouration and delamination also
start from interface of EVA-Glass. Therefore, it is concluded that the model is
valid for the surface.
5.3.3 Differential Scanning Calorimetry (DSC)
As mentioned in section (5.3.1) UV makes significant changes in EVA’s
structure owing to photodegradation. In order to investigate these changes
DSC technique was applied as described in section (3.4.1).
0 5 10 15 20 25 30
20
40
60
80
100
120
140
160
180
200
Time (year)
Concentration of Carboxylic acid (mol/m3)
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
95
Figure (5.13) shows the DSC thermograms before and after ageing. Both
unaged and aged EVAs show double melting peaks. In the case of unaged EVA
the melting peak appears around 46°C with a shoulder around 57°C. After
ageing, the peak and the shoulder separate and the peaks’ temperatures
increase as the ageing time increase. The separation of melting peaks is due to
primary and secondary crystallisation (Chen et al. 2009b; X. Shi et al. 2009;
Lee & Kim 2008; K. A. Moly et al. 2005; Li et al. 2004; Brogly et al. 1997b; Feng
& Kamal 2005; Alizadeh et al. 1999). The entire ageing process includes
physical variations in crystallization and photochemical degradation.
Figure 5.13: DSC first heating thermograms for the unaged and UV aged EVA-Exo Up.
The UV ageing process consists of photochemical degradation and physical
variation in crystallisation (Jin et al. 2010b). The relative crystallinity, Xc, of the
samples was calculated as a function of ageing duration through Eq. (3.1).
Figure (5.14) shows the changes in crystallinity versus ageing time. Analysis of
the results of the first heating shows that crystallinity decreases due to
degradation, which can influence the mechanical behaviour of EVA. It is also
observed that there is no change seen during the second heating.
-100 -50 0 50 100 150 200 250
J (W/g)
T (°C)
Unaged
UV at 65°C, Aged for 3 weeks
UV at 65°C, Aged for 6 weeks
UV at 65°C, Aged for 8 weeks
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
96
Figure 5.14: Crystallinity versus ageing time after first and second heating for unaged and UV
aged EVA.
In order to understand the impact of chemical photodegradation on the
structure of EVA crystallinity is plotted versus concentration of ester and
carboxylic acid, respectively. Figures (5.15) and (5.16) show the relation
between the concentration of ester and carboxylic acid with changes in
crystallinity of the EVA respectively. The results show the crystallinity
decreases by decreasing concentration of ester and increasing concentration
of carboxylic acid. Therefore, continuing photodegradation can damage the
material and affect the structure of it.
-1 0 1 2 3 4 5 6 7 8 9 10
0
1
2
3
4
5
6
7
8
9
10
Time (week)
Xc (%)
1st heating
2nd heating
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
97
Figure 5.15: Variation of crystallinity versus changes in concentration of ester on the
irradiated surface of UV aged EVA.
Figure 5.16: Variation of crystallinity versus changes in concentration of carboxylic acid on the
irradiated surface of UV aged EVA.
The Tg determined by a step change in the heat flow, was found to be around
-25°C for aged and unaged samples. Figure (5.17) shows Tg versus ageing time.
The results show ageing has no significant impact on Tg.
4243444546474849505152
1
2
3
4
5
6
7
8
Concentration of Ester (mol/m3)
Xc(%)
35 40 45 50 55 60
1
2
3
4
5
6
7
8
Concentration of Carbocylic acid (mol/m3)
Xc (%)
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
98
Figure 5.17: Glass transition temperature versus ageing for unaged and UV aged EVA.
5.3.4 Dynamic Mechanical Analysis (DMA)
The storage modulus (E’) was determined as a function of temperature for the
UV aged samples, as described in section (3.4.2). Figure (5.18) illustrates E’
versus temperature. The curves show the different regions and transitions of
a viscoelastic material as temperature increases (Dolores Fernández & Jesús
Fernández 2007). There is a sharp decrease in modulus at around -30°C, which
can be attributed to the glass transition and then another stepped decrease
between 40°C and 70°C, which may be due to some crystal melting. These two
stepped changes correlate well with the DSC identified Tg and melting point.
Figure 5.18: Storage modulus versus temperature for unaged and UV aged EVA.
-1 0123456789
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (week)
Tg (°C)
-80 -60 -40 -20 0 20 40 60 80 100
10-1
100
101
102
103
104
T (°C)
E (MPa)
unaged
Aged for 3 weeks
Aged for 6 weeks
Aged for 8 weeks
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
99
Figure (5.19: a) shows tan(δ) versus temperature for the UV aged samples. In
the curves the peaks are attributed to the glass transition temperature of the
samples (Varghese et al. 2002) which is shown in Figure (5.19: b). This also in
correlation with DSC results (Figure (5.17)) and does not show a significant
change in Tg.
The storage modulus at a given temperature was plotted against the ageing
time to investigate the effect of ageing on the mechanical properties of the
EVA. Figures (5.20: a-d) show this for four temperatures. These figures show
E’ reduces with increasing ageing time.
(a)
(b)
Figure 5.19: (a) tan(δ) vs temperature for unaged and UV aged EVA, (b) Tg versus ageing
duration based on Figure (5.19: a).
-80 -60 -40 -20 0 20 40 60 80 100
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
T (°C)
tan()
unaged
Aged for 3 weeks
Aged for 6 weeks
Aged for 8 weeks
-1 0123456789
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (week)
Tg (°C)
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
100
(a)
(b)
(c)
(d)
Figure 5.20: Storage modulus measured at (a) -20˚C, (b) 0˚C (c), 20˚C (d) 40˚C as a function of
ageing time.
E’ vs temperature for four temperature values was consolidated into one plot
as a mean in order to approximate the general trend of E’. This can be seen in
-1 0 1 2 3 4 5 6 7 8 9
50
100
150
200
E' (MPa)
Time (week)
-1 0 1 2 3 4 5 6 7 8 9
5
10
15
20
25
E' (MPa)
Time (week)
-1 0 1 2 3 4 5 6 7 8 9
0
1
2
3
4
5
6
7
E' (MPa)
Time (week)
0 2 4 6 8
0
1
2
3
E' (MPa)
Time (week)
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
101
Figure (5.21), which shows a decreasing storage modulus as the material ages.
Further, the variation in modulus depends on temperature; as the
measurement temperature increases, the dependence on the ageing duration
weakens (Table (5.5)).
Figure 5.21: Mean storage modulus versus ageing time for EVA.
Table 5.5: The fitting parameters for storage modulus at different fixed temperatures based
on Figures (5.20).
Temperature (°C)
Gradient
Intercept
R2
-20
-15.9
194.7
0.96
0
-1.284
10.86
0.97
20
-0.4655
5.245
0.82
40
-0.2712
2.119
0.81
In order to understand the influence of photodegradation on the mechanical
properties of EVA mean E’ is plotted versus Cr and Cp. Figures (5.22) and (5.23)
show the relation between the concentration of ester and carboxylic acid with
changes in mean E’ respectively. The results show the storage modulus
decreases by decreasing concentration of ester and increasing concentration
0 1 2 3 4 5 6 7 8
0
20
40
60
80
100
120
140
mean E' (MPa)
Time (week)
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
102
of carboxylic acid. Therefore, photodegradation damages the EVA’s
mechanical properties which are due to destructive influence of
photochemical degradation.
Figure 5.22: Average storage modulus versus concentration of ester on the irradiated surface
of UV aged EVA.
Figure 5.23: Average storage modulus versus concentration of carboxylic acid on the
irradiated surface of UV aged EVA.
43444546474849505152
0
20
40
60
80
100
120
140
Concentration of Ester (mol/m3)
E' (MPa)
35 40 45 50 55 60
0
20
40
60
80
100
120
140
Concentration of Carbocylic acid (mol/m3)
E' (MPa)
Examination of the Response of Ethylene-vinyl Acetate Film to UV Irradiation
103
5.4 Conclusions
In this chapter the impact of UV irradiation on the photodegradation,
structure and properties of EVA was investigated by exposing EVA to artificial
UV ageing. For the sake of comparison some samples were kept in a
desiccator away from the degradation factors as control samples and also
some samples were aged in a dark laboratory oven at 65°C in order to isolate
the UV from the thermal effects.
The FTIR-ATR spectra of unaged and UV aged EVA showed notable chemical
changes occurred as a result of UV ageing. The absorbance peaks related to
carboxylic acid, lactone and vinyl sharply increased after the UV irradiation.
The changes on the irradiated side of the UV aged samples were more
significant compared to other samples which show the dominant effect of UV
in the chemical degradation of EVA. The differences in the FTIR-ATR spectra
on the UV irradiated and non-irradiated side of the UV aged samples show
that the intensity is depth dependant. The analytical results showed that the
photochemical reaction is completed on the irradiated surface of EVA in
around 5 year at 0.68 W/m2 of continuous UV intensity. It should be
considered that this time is much longer considering the real field condition
and the glass top layer which absorbs a great portion of UV. DMA showed
storage modulus at 1 Hz was significantly reduced with increases in
temperature due to viscoelastic nature of the copolymer. Ageing had
significant influence on the mechanical properties of the EVA and reduces the
storage modules monotonically as a function of ageing degree. DSC results
showed the crystallinity was massively affected by UV which suggested that
property changes could be connected to structural modifications. Both DSC
and DMA results showed that the photodegradation significantly affects the
structure and mechanical properties of EVA.
In the next chapter moisture absorption of the EVA is studied and the
influence of damp heat ageing on the properties of EVA in investigated.
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
104
Chapter 6
Investigation of Moisture Diffusion and the
Response of Ethylene-vinyl Acetate Film to Damp-
Heat
6.1 Introduction
The importance of moisture ingress and the impact of moisture on the
durability and performance of PV modules have motivated many researchers
to investigate moisture diffusion in the module and in particular, the
encapsulant materials. However, the majority of these investigations have
been carried out using Mocon® devices under ideal conditions and either in
the presence of elemental gas, such as nitrogen, to degas the material and
then with humidity applied to study the permeability. This method is an
indirect method to calculate the moisture diffusion coefficient from direct
measurements of the permeability and solubility.
Diffusion is the net movement of molecules from a region of high
concentration to a region of low concentration. Water sorption in polymers is
linked to the availability of free volume holes in the polymer chain networks
and polymer-water affinity. The hole availability depends on factors such as
polymer morphology, structure and crosslink density (Nogueira et al. 2001; Liu
et al. 2003). The investigation and modelling of water/moisture absorption in
polymers is of concern in many engineering applications and has been the
subject of significant previous research (Ashcroft et al. 2012; Liu, Wildman,
Ashcroft, et al. 2012b; a. Mubashar et al. 2009; A. Mubashar et al. 2009).
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
105
In this chapter the diffusion coefficient of EVA is determined as function of
relative humidity. This was performed by applying the solution of the diffusion
equation to moisture uptake experiments and extracting the diffusion
coefficient (D) via curve fitting following with prediction of moisture
concentration. Figure (6.1) illustrates the approach taken and techniques used
in this chapter in order to investigate the impact of the damp heat ageing and
moisture absorption on the structure and viscoelastic properties of EVA.
Figure 6.1: Flowchart of investigation of EVA’s response to damp heat ageing.
6.2 Analytical Solution of the diffusion equation
6.2.1 Fickian Diffusion
Moisture absorption in a plane sheet can often be simplified to a one
dimensional diffusion problem. The simplest moisture diffusion can be
modelled through the solution of Fick’s Second Law of diffusion (Siah 2010;
Shirrell 1978; Crank 1975), which is represented mathematically by Eq. (6.1).
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
106
(6.1)
where c(x,t) is the moisture concentration and D is the diffusion coefficient, t
and are the time and spatial coordinate respectively. In the case where D is
independent of the moisture concentration, Eq. (6.1) can be simplified to:
(6.2)
In the one dimensional case with uniform initial concentration of the
penetrant throughout a plane sheet, the boundary conditions can be
represented by
(6.3)
(6.4)
where and are the initial moisture concentration in the plate and the
moisture concentration of the boundary/surface respectively and is
thickness of the sheet where in this case is used due to using free standing
films in the gravimetric test.
The solution for Eq. (6.2) can be obtained by the method of separation of
variables or by the Laplace transform (Crank 1975). The solution for 1-D
Fickian diffusion in a plane sheet is
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
107
(6.5)
In our experiments it is assumed the material had no moisture content at t = 0
when the samples were placed in the environmental chamber, therefore,
and . Eq. (6.5) can then be simplified to
(6.6)
This gives the moisture content distribution, however to relate to gravimetric
experiments the sum of moisture in the film is determined by integrating over
the thickness , giving Eq. (6.7) (Crank 1975; Fan et al. 2009).
(6.7)
where and are the total mass of moisture at time t, saturated mass of
moisture in the sheet. The relation between concentration and mass is
presented as Eq. (6.8)-(6.9).
(6.8)
(6.9)
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
108
where and are the volume and the initial mass of moisture respectively.
6.3 Results and discussion
The results of gravimetric, water vapour transmission, DSC and DMA are
presented below based on the ageing conditions and experimental techniques
described in sections (3.3) and (3.4) respectively.
6.3.1 Measurement of moisture diffusion coefficient and
predicting moisture concentration
6.3.1.1 Gravimetric method
Figure (6.2) shows the absorption curve for EVA at 85°C-85% RH. As expected
the moisture content increases with time. Under this condition, the samples
were saturated after around 5 weeks, after which the sample weights were
constant. Using the data shown in Figure (6.2), the value of the moisture
diffusion coefficient were obtained by fitting Eq. (6.7) to the experimental
gravimetric data, where the summation was taken to the first 50 terms. The
resultant values for the moisture diffusion coefficient (D), the saturation
moisture concentration (cs) are presented in Table (6.1).
(a)
(b)
Figure 6.2: Moisture absorption curve for EVA film at 85°C-85% RH, (a) Mt/Mq vs time, (b)
Mt/Mq vs √time/l.
0 0.5 1 1.5 2 2.5 3 x 106
0
0.5
1
Time (s)
Mt/Mq
Fitted curve
Experimental result
0 1 2 3 4 5 6 x 106
0
0.5
1
Time /l ( s /m)
Mt/Mq
Fitted curve
Experimental result
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
109
Table 6.1: Gravimetry test results.
Sample
Condition
Thickness
(mm)
WVTR
(g/(m2.day)
Permeability
(gm.cm)/(cm2.sec.Pa)
cs
(kg/m3)
Diffusion
(m2/sec)
Solubility
(gm/(cm2.Pa)
EVA
85°C
85% RH
0.526
-
-
13.714
5.028× 10-11
_
6.3.1.2 Water Vapour Transmission Rate
As described in section (3.5.2) Water Vapour Transmission Rate (WVTR) is an
indirect technique to measure the moisture diffusion coefficient. Table (6.2)
presents the results of the WVTR test provided by RDM TEST Equipment® at
the conditions detailed in Table (3.3).
Table 6.2: WVTR test results (the analyses were carried out on a MOCON® Permatran-W
Water Vapour Permeability Instrument).
Sample
Condition
Thickness
(mm)
WVTR
(g/(m2.day)
Permeability
(gm.cm)/(cm2.sec.Pa)
cs
(kg/m3)
Diffusion
(m2/sec)
Solubility
(gm/(cm2.Pa)
EVA
40°C
100% RH
0.526
37.77
3.114 × 10-13
-
7.085 × 10-11
4.395 × 10 -7
The diffusion coefficient measured and calculated by both techniques are in
agreement with the values from the literature (Table (2.3)), however, there is
a difference of around 2 × 10-11 between the two values measured by Mocon®
and gravimetric methods which can be due to the different test conditions,
differences in the nature of tests where in the gravimetric method the EVA
film was exposed to 85°C-85% RH for around 5 weeks but in the WVTR test
the sample was exposed to 40°C and 100% RH for around 140 minutes in the
presence of nitrogen to avoid any degradation and also experimental errors
where in gravimetry the samples were not kept continuously at 85°C-85% RH
and they were taken out for weight measurement.
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
110
In this study the diffusion coefficient measured by gravimetric test was used
for predicting moisture concentration in depth.
6.3.1.3 Moisture concentration
Once moisture diffusion coefficient is determined, moisture concentration at
any spot within the film at time t can be achieved via Eq. (6.6) (Karimi 2006).
Figure (6.3) shows the moisture concentration inside the EVA film. The results
show the moisture concentration increases by time until saturation point after
around 5 weeks under constant condition of 85°C-85% RH.
Figure 6.3: Simulated moisture concentration inside the EVA film at different depths (X1-X10)
under the damp heat condition of 85°C-85% RH.
6.3.2 Differential Scanning Calorimetry (DSC)
The first heating thermograms of unaged and aged EVA at 85°C-85% RH and
22±3°C-85% RH are reproduced with offsets added for the sake of comparison
which are presented in Figures (6.4: a-b). Unaged EVA shows a melting area
between 40°C and 80°C with a single peak at 45°C and a shoulder at 55°C. The
melting transition of the ethylene segment has been accepted as showing as a
peak with a shoulder observed between 40°C to 80°C (Motta 1997; Li et al.
2013). As the ageing at 85°C-85% progresses the peak at 45°C disappears and
the melting peak shows itself at 55°C. The secondary melting point is linked to
the secondary crystallization which takes place between the primary
0 0.5 1 1.5 2 2.5 3 3.5
x 106
0
5
10
15
Time (s)
c (kg/m3)
Moisture concentration
Average moisture concentration
X1
X2X3
X10
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
111
crystallization and storage or exposure to elevated temperature (Oreski &
Gernot M. Wallner 2010). It is reported that the ethylene copolymer has a
two-stage crystallisation: primary and secondary. The secondary
crystallisation is due to rearrangement of irregular polymer chains and the
segments that are left in the amorphous phase, into the crystals (Akpalu et al.
1999; Feng & Kamal 2005; Alizadeh et al. 1999). The secondary melting point
is assigned to the less organised crystal phase considering the fact that EVA
containing 33% VAc is semi-crystalline and ageing may affect the less
crystalline phase (Brogly et al. 1997a). In the case of 22±3°C-85% RH the
thermograms do not change with ageing, which shows the significant effect of
temperature on changes in the copolymer’s structure.
(a)
(b)
Figure 6.4: DSC first heating thermograms for the unaged and damp heat aged EVA at (a)
85°C-85% RH and (b) 22±3°C-85% RH-Exo Up.
-100 -50 0 50 100 150 200 250
J (W/g)
T (°C)
Unaged
85(°C)-85% RH, Aged for 3 weeks
85(°C)-85% RH, Aged for 6 weeks
85(°C)-85% RH, Aged for 9 weeks
-100 -50 0 50 100 150 200 250
J (W/g)
T (°C)
Unaged
22±3°C-85% RH, Aged for 3 weeks
22±3°C-85% RH, Aged for 6 weeks
22±3°C-85% RH, Aged for 9 weeks
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
112
Figure (6.5: a-b) show Tg versus ageing time for the two different ageing
conditions. Tg, typically identified by a step change in the heat flow was found
to be around -25°C for aged and unaged samples, suggesting ageing has no
significant effect on Tg.
(a)
(b)
Figure 6.5: Glass transition temperature versus ageing time for EVA at (a) 85°C-85% RH and
(b) 22±3°C-85% RH.
Figures (6.6: a-b) show Xc (calculated via Eq. (3.1)) versus ageing time for the
two different ageing conditions. Analysis of the first heating results shows that
in the case of ageing at 85°C-85% RH, crystallinity had increased after 3 weeks
of ageing and then decreased to a value lower than the original crystallinity.
The increase in the crystallinity is due to the secondary crystallization which
0 1 2 3 4 5 6 7 8 9 10
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (week)
Tg (°C)
0 1 2 3 4 5 6 7 8 9 10
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (week)
Tg (°C)
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
113
continues when EVA is aged above the glass transition temperature in the
damp heat ageing and the crystals become larger (X. M. Shi et al. 2009)which
has also been observed by other researchers using different techniques
including DSC and WAXD (Oreski & Gernot M Wallner 2010; X.-M. Shi et al.
2009; Chen et al. 2009). The increasing crystallinity can also be due to the
creation of double bonded water as presented in the study performed by
(Iwamoto & Matsuda 2005; Iwamoto et al. 2003). The results of ageing at
22±3°C-85% show a similar trend but over a longer time scale, which would be
expected as both moisture absorption and thermal degradation are thermally
activated processes. It can also be seen that in both ageing environments
there is no change in crystallinity during the second heating. It can be
concluded, therefore, that moisture diffusion into the polymer structure
causes an increase in crystallinity, however, elevated temperature has a
greater influence on the structure of the EVA as the ageing time increases. It
should also be mentioned again here that the first heating was to investigate
the impact of ageing on EVA and the second heating was to erase the thermal
history regarding the curing, storing consitions and the ageing inorder to
investigate the properties of EVA indipendant from of these effects. In order
to understand the relation between the moisture absorption and the
structural changes in EVA, the changes of crystallinity were investigated as
moisture concentration increases. Figure (6.7) shows crystallinity after first
and second heating versus moisture concentration at the intervals of 0, 3, 6
and 9 weeks considering the EVA samples saturated in 5 weeks. The results of
the first heating show crystallinity increases as the samples absorbs moisture
until the saturation point. After the saturation point the crystallinity decrease
to a value lower than the initial value which show the initial impact of
moisture absorption which cause an increase in crystallinity however, with
further ageing at damp heat condition after saturation point thermal
degradation (described in chapter 4) plays a dominant role which cause a
decrease in the crystallinity. The results of the second heating do not show
any significant changes in the crystallinity due to the erased thermal history.
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
114
(a)
(b)
Figure 6.6: Crystallinity versus ageing time after first and second heating for unaged and aged
EVA at (a) 85°C-85% RH and (b) 22±3°C-85% RH.
Figure 6.7: Crystallinity versus average concentration after first and second heating for
unaged and aged EVA at 85°C-85% RH.
-1 0 1 2 3 4 5 6 7 8 9 10
0
2
4
6
8
10
12
Time (week)
Xc (%)
1st heating
2nd heating
-1 0 1 2 3 4 5 6 7 8 9 10
0
1
2
3
4
5
6
7
8
9
10
Time (week)
Xc (%)
1st heating
2nd heating
0 5 10 15
2
3
4
5
6
7
8
9
10
11
12
c (kg/m3)
Xc (%)
1st heating
2nd heating
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
115
6.3.3 Dynamic Mechanical Analysis (DMA)
Figures (6.8: a-b) show the storage modulus (E’) determined as a function of
temperature at a frequency of 1 Hz for the two damp heat ageing conditions.
As the temperature increases, E’ decreases and the curves show the typical
regions and transitions of a viscoelastic material, as described previously
(Stark & Jaunich 2011). A four-decade difference is observed in E’ between -
75°C and 95°C with a sharp decrease at around -30°C, which can be attributed
to the glass transition and then another stepped decrease between 40°C and
65°C which may be due to some crystal melting. Both transitions correlate
well with the DSC identified Tg and melting points. The results show that the
mechanical properties of EVA were affected more significantly under 85°C-
85% RH condition due to the higher temperature and changes in the
crystallinity, where E’ has a significant increase after 3 weeks and then it
drops to lower values than the original values as the ageing time increases.
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
116
(a)
(b)
Figure 6.8: Storage modulus vs temperature for unaged and aged EVA at (a) 85°C-85% RH and
(b) 22±3°C-85% RH.
Figures (6.9: a) and (6.10: a) show tan(δ) versus temperature for two damp
heat ageing conditions, in which the peak can be attributed to the glass
transition temperature of the samples as shown in Figures (6.9: b) and (6.10:
b) (Varghese et al. 2002). This also correlates well with the DSC results (Figure
(6.5)). The plots do not show a significant change in Tg with ageing for either
condition.
-80 -60 -40 -20 0 20 40 60 80 100
10-1
100
101
102
103
104
E' (MPa)
T (°C)
Unaged
85(°C)-85% RH, Aged for 3 weeks
85(°C)-85% RH, Aged for 6 weeks
85(°C)-85% RH, Aged for 9 weeks
-80 -60 -40 -20 0 20 40 60 80 100
10-1
100
101
102
103
104
E' (MPa)
T (°C)
Unaged
22±3°C-85% RH, Aged for 3 weeks
22±3°C-85% RH, Aged for 6 weeks
22±3°C-85% RH, Aged for 9 weeks
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
117
(a)
(b)
Figure 6.9: (a) tan(δ) vs temperature for unaged and aged EVA at 85°C-85% RH, (b) Tg versus
ageing duration based on Figure (6.9: a).
-80 -60 -40 -20 0 20 40 60 80 100
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
tan ()
T (°C)
Unaged
85(°C)-85% RH, Aged for 3 weeks
85(°C)-85% RH, Aged for 6 weeks
85(°C)-85% RH, Aged for 9 weeks
-1 0 1 2 3 4 5 6 7 8 9 10
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (week)
Tg (°C)
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
118
(a)
(b)
Figure 6.10: (a) tan(δ) vs temperature for unaged and aged EVA at 22±3°C-85% RH, (b) Tg
versus ageing duration based on Figure (6.10: a).
For further investigation of the effect of damp heat ageing on the mechanical
properties of the EVA, the storage modulus at a given temperature was
plotted against the ageing time for the aged and control samples. Figures
(6.11: a-d) and (6.12: a-d) show plots of modulus as a function of ageing time
measured at 0°C, 20°C, 40°C and 60°C. In the case of 85°C-85% RH after three
weeks of damp heat ageing E’ increased due to the increase in crystallinity
(Figure (6.6)). However, one can see that E’ reduces to lower than the initial
value with increased time of ageing. At 22±3°C-85% RH E’ shows a different
behaviour. E’ increases up to the sixth week of the ageing at damp
environment and start to decrease after this point which is a similar trend to
85°C-85% RH but in a longer time scale and can be due to changes in structure
-100 -80 -60 -40 -20 0 20 40 60 80 100
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
tan ()
T (°C)
Unaged
22±3°C-85% RH, Aged for 3 weeks
22±3°C-85% RH, Aged for 6 weeks
22±3°C-85% RH, Aged for 9 weeks
-1 0 1 2 3 4 5 6 7 8 9 10
-40
-35
-30
-25
-20
-15
-10
-5
0
Time (week)
Tg (°C)
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
119
of EVA (presented in Figure (6.6). Comparing the test results of both damp
heat conditioning conditions, shows the significant impact of moisture
diffusion into the copolymers structure at the early stage of the exposure
which increases the crystallinity but on the other hand the impact of heat on
the polymer ageing and mechanical degradation is noticeable and causes a
decrease in E’ as pointed out in chapter 4 where the storage modules reduced
by thermal ageing as a function of ageing degree. In other words moisture
diffusion and thermal degradation competes in damp heat ageing in particular
in the early stage of ageing. However, the analysis of the results of the damp
heat test at 85°C-85% RH indicate that the effect of heat is more dominant as
the ageing time increases.
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
120
(a)
(b)
(c)
(d)
Figure 6.11: Storage modulus measured at (a) 0˚C, (b) 20˚C (c), 40˚C (d) 60˚C as a function of
ageing time at 85°C-85% RH.
0 1 2 3 4 5 6 7 8 9 10
8
10
12
14
16
18
20
22
E' (MPa)
Time (week)
Aged at 85°C-85%RH
Control sample, 22±3°C-15% RH
0 1 2 3 4 5 6 7 8 9 10
3
4
5
6
7
8
9
10
E' (MPa)
Time (week)
Aged sample at 85°C-85% RH
Control sample, 22±3°C-15% RH
0 1 2 3 4 5 6 7 8 9 10
1
2
3
4
5
E' (MPa)
Time (week)
Aged sample at 85°C-85% RH
Control sample, 22±3°C-15%RH
012345678910
0.5
1
1.5
2
E' (MPa)
Time (week)
Aged sample at 85°C-85% RH
Control sample, 22±3°C-15% RH
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
121
(a)
(b)
(c)
(d)
Figure 6.12: Storage modulus measured at (a) 0˚C, (b) 20˚C (c), 40˚C (d) 60˚C as a function of
ageing time at 22±3°C-85% RH.
In order to have a clear view of the general trend, E’ versus ageing time for
four fixed temperature values (Figures (6.11: a-d) and (6.12: a-d)) were
-1 0 1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
E' (MPa)
Time (week)
Aged sample at 22±3°C-85% RH
Control sample, 22±3°C-15% RH
0 1 2 3 4 5 6 7 8 9 10
5
5.5
6
6.5
7
7.5
E' (MPa)
Time (week)
Aged sample at 22±3°C-85% RH
Control sample, 22±3°C-15% RH
0 1 2 3 4 5 6 7 8 9 10
2
2.5
3
3.5
4
E' (MPa)
Time (week)
Aged sample at 22±3°C-85% RH
Control sample, 22±3°C-15% RH
0 1 2 3 4 5 6 7 8 9 10
0.5
1
1.5
E' (MPa)
Time (week)
Aged sample at 22±3°C-85% RH
Control sample, 22±3°C-15% RH
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
122
combined into one plot as a mean storage modulus versus ageing time. The
consolidated results are shown in Figure (6.13), which shows a decrease in
storage modulus after the third week of ageing at 85°C-85% RH as the
material ages. In the case of 22±3°C-85% RH the storage modulus was
increasing up to week six of ageing and then started to decrease. The impact
of damp heat ageing can be divided into the effect of moisture which
increases the modulus at the early exposure time and ageing effect of heat
which has an opposite and dominant influence as the ageing time increases.
In the case of 85°C-85% RH the samples absorbs and loses moisture faster due
to the higher test temperature which results in a sharp decrease in
crystallinity and E’ as its consequence But when the test temperature is 22°C
the samples moisture desorption is slower which results in a gradual decrease
in crystallinity and E’, therefore, moisture affects the properties of EVA at the
early stage of the test and up to the saturation point where temperature plays
a significant role after this point. Both increasing and decreasing storage
modulus can cause serious problems such as delamination and problems in
mismatches in thermal expansion and cracking of the cell and wiring (Oreski &
Gernot M. Wallner 2010). Thus, the damp heat ageing can result in significant
problems during service lifetime of a PV module.
Figure 6.13: Average storage modulus versus ageing time for EVA aged at 85°C-85% RH and
22±3°C-85% RH.
0 1 2 3 4 5 6 7 8 9
250
300
350
400
450
500
Time (week)
mean E' (MPa)
Aged at 85°C-85% RH
Aged at 22±3°C-85% RH
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
123
Since the moisture diffusion and concentration play an important role in the
first stage of the damp heat ageing it is needed to investigate the relation
between the moisture absorption in EVA and the changes in the mechanical
properties the changes of the storage modulus is plotted versus the moisture
concentration. Figure (6.14) shows the variation of storage modulus at
different moisture concentrations. It can be seen that the storage modulus
increase by increasing absorbed moisture in EVA up to the saturation point
but after this point the storage modulus decreases which can be correlated to
the changes in the crystallinity and the influence of thermal degradation after
the saturation point. Therefore, it can be concluded that after saturation
point, the thermal degradation affects the mechanical properties of the EVA
more significantly.
Figure 6.14: Average storage modulus versus average moisture concentration for EVA aged at
85°C-85% RH.
6.4 Conclusions
The response of the EVA copolymer to damp heat ageing was investigated at
two conditions, with the same RH level and different temperatures (22±3°C-
85% RH and 85°C-85% RH). Various techniques were applied in order to
0 5 10 15
250
300
350
400
450
500
c (kg/m3)
mean E' (MPa)
Investigation of Moisture Diffusion and the Response of Ethylene-vinyl
Acetate Film to Damp-Heat
124
measure the moisture absorption coefficient, viscoelastic properties and
investigate the morphological changes in response to damp heat ageing.
The results of the calorimetry indicated that the glass transition temperature
was unaffected by the ageing, however, the crystallinity increased after three
weeks of ageing, but then decreased on further ageing at 85°C-85% RH. At
22±3°C-85% RH the crystallinity increased up to the sixth weeks and then
decreased with further ageing. The DSC results show significant influence of
temperature on the structure of EVA. These results also suggested that
property changes could be connected to the structural modifications. The
storage modulus was shown to be strongly dependent on the crystallinity and
correlated well with the structural changes in the EVA seen in the DSC. The
results showed the long term damp heat has noticeable influence on the
mechanical properties of EVA and cause decrease in E‘. Comparing the DSC
and DMA results against the variation of moisture concentration indicated the
crystallinity and the storage modulus increased by increasing moisture
concentration up to the saturation point and started to decrease after the
samples reached the saturation. This can be due to the thermal degradation
(described in chapter 4) which plays a dominant role after the saturation
point.
As mentioned in chapter 4 PV modules will be subjected to different
degradation factors in service. This chapter investigated the combination of
humidity and heat which showed different results than the heat only and UV
only ageing. The next chapter reviews and discusses the results with a
comparative view in order to achieve a better understanding of the impact of
the different ageing factors on the mechanical properties of EVA.
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
125
Chapter 7
Comparative study of ageing impact on the
structure and mechanical properties EVA under
different ageing conditions
7.1 Introduction
PV modules need to have a lifetime of 20-25 years to be economically
reasonable. In this regard the encapsulation material should be stable at high
UV exposure, elevated temperatures and humid environments. However, EVA
undergoes chemical and physical degradation on exposure to harsh
environmental conditions.
EVA undergoes complex degradation mechanisms owing to the interaction
between the degradation factors. To tackle and simplify this interaction the
logical approach was first to establish the properties and response of EVA in
isolation before considering the complex interaction between the
environmental degradation factors in the field. In this chapter, the response
of Ethylene-vinyl Acetate (EVA) film particularly the changes in crystallinity,
storage modulus and the dependence of the material properties on different
ageing conditions were compared. In the next sections the ageing impact of
degradation factors is compared in order to identify the dominant ageing
factor.
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
126
7.2 Comparative study of UV, thermal and damp heat
ageing impact on the properties of EVA
For further investigation of the effect of ageing on the structure and
mechanical properties of EVA it is worth studying the impact of ageing factors
comparatively. Figure (7.1) shows the changes in crystallinity versus ageing
duration for different ageing conditions (see section (3.3)). The results show
that the impact of UV on reduction in crystallinity is greater than other ageing
factors considering that UV ageing was performed at 50°C while thermal
ageing was at 85°C. This is because the product of thermochemical
degradation at the in-service temperature of PV modules is acetic acid (as
shown in Figure (2.4)) which is not significant where EVA undergoes a complex
degradation process via Norrish reactions which yield to formation of acetic
acid, carboxylic acid, ketone, lactone, vinyl and etc. However, it was also
noticed that thermal only ageing has more deteriorating impact than damp
heat ageing. The results showed that the initial effect of damp heat was
moisture absorption and significant increase in crystallinity and storage
modulus up to the saturation point which can cause some severe problems
during service lifetime of a PV module and can lead to cracking of the solar
cells and their wiring connections (Oreski & Gernot M Wallner 2010) and
should be considered by PV module manufacturers.
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
127
Figure 7.1: Crystallinity versus ageing duration for different ageing conditions.
Figures (7.2: a-c) show storage modulus (E’) at three fixed temperatures.
These results show a decreasing storage modulus as the material ages for UV
and thermal ageing but in the case of damp heat ageing there is an initial
increase due to the moisture diffusion into the copolymer’s structure and
increase in crystallinity of EVA as a result of creation of double bonded water
(Iwamoto & Matsuda 2005; Iwamoto et al. 2003) which is followed by a
decrease in E’ as ageing time increases. As shown in Figures (7.2) the
reduction in E’ is at a greater rate in the case of UV ageing which is due to the
influence of UV in crystallinity reduction (Figure (7.1)), chemical degradation
of EVA (Figures (5.7)-(5.9)) and also presence of antioxidants (S Isarankura Na
Ayutthaya & Wootthikanokkhan 2008) therefore it can be concluded that UV
is the dominant ageing factor among all degradation factors. These findings
are important in terms of the influence of degradation factors on the
mechanical properties of EVA and the criteria in selection of suitable modules
for different climatic conditions (Oreski & Gernot M Wallner 2010).
In the next section the results regarding the chemical changes in EVA after
ageing and the influence of ageing on the chemical structure of EVA will be
discussed for the sake comparison.
-10 0 10 20 30 40 50 60 70 80
1
2
3
4
5
6
7
8
9
10
11
Time (day)
Xc (%)
Thermal ageing
UV ageing
DH 85°C-85% ageing
DH Troom-85% ageing
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
128
(a)
(b)
(c)
Figure 7.2: Comparitive effect of degradation factors on the storage modulus of EVA at (a)
0°C, (s) 20°C, (c) 40°C versus ageing duration for control sample and aged EVA at different
conditions.
-10 0 10 20 30 40 50 60 70 80
0
5
10
15
20
25
Time (day)
E' (MPa)
Thermal ageing
UV ageing
DH 85°C-85% ageing
control sample
-10 0 10 20 30 40 50 60 70 80
0
1
2
3
4
5
6
7
8
9
10
Time (day)
E' (MPa)
Thermal ageing
UV ageing
DH 85°C-85% ageing
control sample
-10 0 10 20 30 40 50 60 70 80
-1
0
1
2
3
4
5
Time (day)
E' (MPa)
Thermal ageing
UV ageing
DH 85°C-85% ageing
control sample
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
129
7.3 Comparative study of influence of the UV thermal
and damp-heat ageing factors on chemical changes in
EVA
In order to understand the influence of different ageing factors the produced
results are reviewed with a comparative view. The TGA and FTIR-ATR results
showed that acetic acid is the common degradation product in both thermal
degradation and photdegradation of EVA however, in the case thermal
degradation acetic acid is formed through the so-called deacetylation process
in the early stage of the decomposition and is the main degradation product
(Figure (4.3)) where quantities of other products (CO, CO2 and CH4) are
negligible (Marcilla & Sempere 2004). There is also a second stage of
degradation which occurs above 400°C which is not considered in this study.
The degradation behaviour of EVA is significantly dependant on vinyl acetate
concentration (VAc) and in this study EVA with 33% VAc is studied to focus on
the PV industry grade EVA. On the other hand photodegradation of EVA is a
more complex process and occurs by different processes than thermal
degradation where Norrish reactions are the dominant photodegradation
processes despite the effect of elevated temperature presented in chapter 4.
In photodegradation of EVA acetic acid is formed via ester elimination
(Copuroglu & Sen 2005) and through decomposition of vinyl acetate segment
represented by Norrish I and Norrish II as shown in Figure (5.3). Apart from
acetic acid formation photodegradation of EVA has other products which have
significant influence in the durability and performance of EVA. These products
include carboxylic acid, vinylene and vinyl which cause yellowing and affect
the optical transmission and deteriorate the performance of EVA as the
encapsulant (Morlat-Therias et al. 2007). The absorbance peaks related to
other products including carboxylic acid, lactone and vinyl sharply increased
on the irradiated surface after the UV irradiation where no significant changes
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
130
were observed in other samples which show the dominant effect of UV in the
chemical degradation of EVA. The differences in the FTIR-ATR spectra on the
UV irradiated and non-irradiated side of the UV aged samples show that the
intensity is depth dependant. In this study photodegradation on the UV
exposed surface of EVA is predicted by a numerically developed model which
shows in case of unprotected UV exposure the degradation process completes
in around 0.5 year where It is reported in the literature that laboratory
environmental chamber conditions may be different from in-service
conditions and for instance the reduction in optical transmission after 6
month in an environmental chamber is much less than desired PV lifetime (M
D Kempe 2008) and also it is reported that discolouration of EVA can develop
in around 5 years (Czanderna & Pern 1996). The differences in results come
from different sample types where in the previous studies laminated samples
with low iron top glass have been used where most of the harmful UV is
absorbed by glass but in this study free film is used and the exposure was not
protected. The developed method is validated with FTIR-ATR results and can
be used in wide range of application including polymer degradation and solar
energy capture.
Damp heat had a different effect than UV and elevated temperature where
the main difference in IR spectra before and after damp heat ageing was
observed at between 3400 and 3700 cm-1. The observed changes were related
to moisture diffusion into the EVA’s structure and bonded water formation
which also caused an increase in the crystallinity of EVA in the early stage of
damp heat ageing and up to the saturation point however, the increase in
crystallinity has not been observed in EVA with VAc lower than 30% (X.-M. Shi
et al. 2009) which shows the significant impact of VAc on the ageing
behaviour of EVA. It should be mentioned that the PV degree EVA has VAc of
33% which is also used in this study.
Comparative study of ageing impact on the structure and mechanical
properties EVA under different ageing conditions
131
7.4 Conclusion
In this chapter the influence of different ageing factors on the crystallinity,
mechanical properties and chemical ageing behaviour of EVA was discussed
with a comparative view. The results showed all degradation factors cause a
reduction in crystallinity and mechanical properties of EVA in a long term
exposure. However, damp heat causes an initial increase in crystallinity and
storage modulus which was followed by a decrease in them as ageing time
increases. ATR-FTIR, DSC and DMA results indicated that UV has a greater
degrading influence comparing to other degradation factors which is due to
the photodegradation process EVA undergoes which yields to formation of
different chemical components where the thermochemical degradation of
EVA results in formation of mainly acetic acid under working temperature of
PV modules. It should also be mentioned that the damage caused by the
environmental factors and changes in mechanical properties can cause
mismatches in thermal expansion, delamination and damage to solar cells
which eventually results in failure of a PV modules in the field.
The next chapter presents achievements and conclusions of the research. The
recommendation for future works will be also included in the next chapter.
Conclusion and Recommendations for Future Works
132
Chapter 8
Conclusion and Recommendations for Future Work
8.1 Conclusions
This work aimed to fully understand and predict the degradation of the
encapsulant material, EVA, under different environmental conditions using
artificial laboratory ageing and associate changes in mechanical properties by
examining the link between the chemistry, the structure and the mechanical
behaviour. This leaded to better understanding of module degradation and
lead to production of more durable PV modules. Based on the experimental
results and corresponding numerical analysis the following conclusions can be
made:
The thermal degradation of EVA was investigated using techniques that
enabled the viscoelastic properties and thermal stability to be measured.
The findings showed
The activation energy of deacetylation was unaffected by the ageing
process.
The chemical changes due to thermal activation did not significantly
influence the lifetime of the EVA.
Ageing reduced the crystallinity and storage modules as a function of
ageing degree.
DSC results suggested that property changes could be connected to
structural modifications.
The photodegradation of EVA and its influence on the structure and
mechanical properties were investigated using ATR-FTIR, DSC and DMA.
Conclusion and Recommendations for Future Works
133
The IR spectra showed notable increases in the absorbance peaks
related to carboxylic acid, ketone and vinyl on UV ageing.
The most significant changes were observed on the UV exposed side of
the sample, where the UV non-exposed side of the samples were not
affected.
The differences in the ATR-FTIR spectra on the UV affected and back
side of the UV aged samples shows that the intensity is depth
dependant.
DSC and DMA analysis showed that the structure and the storage
modulus of the samples were extremely influenced by UV exposure
respectively and a sharp decrease was observed in the crystallinity and
accordingly in storage modulus.
A numerical model was developed to enable the prediction of the
photodegradation in EVA. The model showed a good agreement on the UV
exposed surface but there were significant discrepancies on the non-
irradiated surface which could be due to the presence of UV absorber in
the EVA which explains why no effect of UV was observed on the non-
irradiated side.
The response of EVA to damp heat environment was investigated at two
conditions with same RH level and different temperatures.
The measurement of moisture absorption coefficient was carried out
by gravimetry and WVTR technique (Mocon device). The measured
values were in a same range and in agreement with the literature.
However, there was a slight difference in the measured values which
could be because of the different test conditions and different nature
of the tests.
The distribution of the moisture concentration in depth was modelled
and the impact of moisture concentration on the crystallinity and
storage modulus of EVA were investigated.
Conclusion and Recommendations for Future Works
134
The calorimetry results indicated that the crystallinity increases in the
initial stage of damp heat ageing due to incorporation of moisture into
the polymer’s structure however it decreases as ageing time increases.
The rate of the decrease in crystallinity was shown to have
dependency on the temperature as the higher temperature resulted in
greater rate of decrease in crystallinity after the initial increase.
DMA results showed that the storage modulus has strong dependency
on the crystallinity and follows the structural changes of EVA.
A comparative analysis to include the effect of all ageing factors was
applied to give a better understanding of the degradation factors.
All degradation factors cause a reduction in crystallinity and
mechanical properties of EVA in a long term ageing.
ATR-FTIR, DSC and DMA results indicated that UV has a greater
degrading influence comparing to other degradation factors which can
be summarised as UV > T > DH.
8.2 Recommendations for future work
This work focused on investigating the chemical degradation of encapsulant
to understand the degradation mechanisms to be able to develop a predictive
model for lifetime behaviour prediction and link the chemical degradation to
the changes in the structure and mechanical properties of the EVA. Although
extensive work has been conducted, more work is still required to enhance
the modelling part and use other techniques to be able to improve the
understanding of the degradation of EVA to include other aspects of
degradation such as delamination and extend the knowledge by incorporating
field conditions. This should include:
Considering the combining effect of environmental factors and the
interaction between them on the degradation of EVA: In this work the
impact of degradation factors on chemical degradation, structure and
mechanical properties of EVA was investigated in isolation. In the field PV
Modules are subjected to the combination of elevated temperatures, light
Conclusion and Recommendations for Future Works
135
and humidity. It is important that the interaction between the effects of
these environmental factors and effect of daily cycling including moisture
absorption and desorption to be investigated in order to enhance the
understanding regarding the field conditions.
Investigating the impact of the discolouration on optical properties of EVA:
UV exposure causes discolouration in EVA which affects its optical
properties and the efficiency of the modules. It is essential to find out the
relation between the degree of discolouration, light transmission through
EVA and electrical efficiency in order investigate the rate of loss in
modules’ efficiency versus degree of discolouration.
Enhance the photodegradation model by including the effect of UV
absorbers to be able predict the photodegradation in depth: UV exposure
causes serious damages to EVA which starts from the surface and
develops in depth. In order to be able to predict the photodegradation in
depth it is essential to use different experimental techniques to have
access to data in depth of EVA and develop a predictive model which
includes the effect of additives such as UV absorber to achieve a valid
accurate model. In this regards the experimental results and the initial
model developed in this research can pave the way to improve the model.
Investigating the influence of UV ageing on moisture diffusion coefficient:
Moisture diffusion coefficient was measured and moisture concentration
was modelled accordingly in this research at single condition however, in
the field modules are exposed to cyclic conditions and the moisture
diffusion coefficient might be affected by degradation factors due to the
structural changes caused by these factors. Therefore, it is important to
investigate the impact of degradation factors on the moisture diffusion in
EVA.
Appendix B
135
References
A. L. Rosenthal, M. G. Thomas, and S.J.D., 1994. A ten year review of
performance of photovoltaic systems. In NREL Photovoltaic Performance
and Reliability Workshop, NREL/CP-411-7414. pp. 279–285. Available at:
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=346934.
A.K. Plessing, E. von S., 2003. No Title. In R. W. L. G.M. Wallner, ed.
Proceedings of Polymeric Solar Materials. Leoben, pp. XII1–XII8.
Abdelkader, A.F. & White, J.R., 2005. Water absorption in epoxy resins: The
effects of the crosslinking agent and curing temperature. Journal of
Applied Polymer Science, 98, pp.2544–2549.
Abrusci, C. et al., 2012. Photodegradation and biodegradation by bacteria of
mulching films based on ethylene-vinyl acetate copolymer: Effect of pro-
oxidant additives. Journal of Applied Polymer Science, 126(5), pp.1664–
1675. Available at: http://doi.wiley.com/10.1002/app.36989 [Accessed
December 1, 2014].
Agroui, K. et al., 2007. Quality control of EVA encapsulant in photovoltaic
module process and outdoor exposure. Desalination, 209(1-3), pp.1–9.
Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0011916407000926
[Accessed October 28, 2014].
Akpalu, Y.A. et al., 1999. Structure Development during Crystallization of
Homogeneous Copolymers of Ethene and 1-Octene: Time-Resolved
Synchrotron X-ray and SALS Measurements. Macromolecules, 32,
pp.765–770. Available at:
http://pubs.acs.org/doi/abs/10.1021/ma9810114.
Alizadeh, a. et al., 1999. Influence of Structural and Topological Constraints
on the Crystallization and Melting Behavior of Polymers. 1. Ethylene/1-
Octene Copolymers †. Macromolecules, 32(19), pp.6221–6235. Available
at: http://pubs.acs.org/doi/abs/10.1021/ma990669u.
Allen, N.S. et al., 2001. Aspects of the thermal oxidation , yellowing and
stabilisation of ethylene vinyl acetate copolymer. , 71, pp.1–14.
Allen, N.S. et al., 1994. Physicochemical aspects of the environmental
degradation of poly(ethylene terephthalate). Polymer Degradation and
Stability, 43(2), pp.229–237. Available at:
http://linkinghub.elsevier.com/retrieve/pii/0141391094900744
[Accessed December 1, 2014].
Allen, R.D. et al., 2005. Design of experiments for the qualification of EVA
expansion characteristics. Robotics and Computer-Integrated
Manufacturing, 21(4-5), pp.412–420. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0736584504001231
Appendix B
136
[Accessed December 1, 2014].
Androsch, R., 1999. Melting and crystallization of poly(ethylene-co-octene)
measured by modulated d.s.c. and temperature-resolved X-ray
diffraction. Polymer, 40(10), pp.2805–2812. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0032386198004704
[Accessed December 10, 2014].
Anon, 2014. Renewable Energy Association. REA Publications. Available at:
http://www.r-e-a.net/resources/rea-publications.
Anon, 2013. UK Solar PV Strategy Part 1: Roadmap to a Brighter Future,
London. Available at:
https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/249277/UK_Solar_PV_Strategy_Part_1_Roadmap_to_a_Brighte
r_Future_08.10.pdf.
Ashcroft, I.A. et al., 2012. Environmental Ageing of Epoxy-Based
Stereolithography Parts Part 2: Effect of absorbed moisture on
Mechanical Properties. Plastic, Rubber and Composites, 41(3), pp.129–
136. Available at: https://dspace.lboro.ac.uk/2134/8687.
Ashcroft, I.A. et al., 2001. The effect of environment on the fatigue of bonded
composite joints. Part 1: testing and fractography. Composites Part A:
Applied Science and Manufacturing, 32(1), pp.45–58. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1359835X00001317.
Ayutthaya, S.I.N. & Wootthikanokkhan, J., 2008. Investigation of the
photodegradation behaviors of an ethylene/vinyl acetate copolymer
solar cell encapsulant and effects of antioxidants on the photostability of
the material. Journal of Applied Polymer Science, 107(6), pp.3853–3863.
Available at: http://doi.wiley.com/10.1002/app.27428 [Accessed
December 1, 2014].
Ayutthaya, S.I.N. & Wootthikanokkhan, J., 2008. Investigation of the
photodegradation behaviors of an ethylene/vinyl acetate copolymer
solar cell encapsulant and effects of antioxidants on the photostability of
the material. Journal of Applied Polymer Science, 107(6), pp.3853–3863.
Available at: http://doi.wiley.com/10.1002/app.27428 [Accessed
November 3, 2014].
Badiee, A., Wildman, R. & Ashcroft, I., 2014. Effect of UV aging on degradation
of ethylene-vinyl acetate (EVA) as encapsulant in photovoltaic (PV)
modules N. G. Dhere, ed. , 9179, p.91790O. Available at:
http://proceedings.spiedigitallibrary.org/proceeding.aspx?doi=10.1117/1
2.2062007 [Accessed October 22, 2014].
Bauer, F. et al., 1999. TG-FTIR and isotopic studies on coke formation during
the MTG process. Microporous and Mesoporous Materials, 29(1-2),
pp.109–115. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1387181198003242.
Berman, D., Biryukov, S. & Faiman, D., 1995. EVA laminate browning after 5
Appendix B
137
years in a grid-connected, mirror-assisted, photovoltaic system in the
Negev desert: effect on module efficiency. Solar Energy Materials and
Solar Cells, 36(4), pp.421–432. Available at:
http://linkinghub.elsevier.com/retrieve/pii/0927024894001987
[Accessed December 2, 2014].
Berman, D. & Faiman, D., 1997. EVA browning and the time-dependence of
I−V curve parameters on PV modules with and without mirror-
enhancement in a desert environment. Solar Energy Materials and Solar
Cells, 45(4), pp.401–412. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0927024896000876
[Accessed December 2, 2014].
Bianchi, O. et al., 2011. Changes in activation energy and kinetic mechanism
during EVA crosslinking. Polymer Testing, 30(6), pp.616–624. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0142941811000742
[Accessed August 7, 2014].
Brogly, M., Nardin, M. & Schultz, J., 1997a. Effect of vinylacetate content on
crystallinity and second-order transitions in ethylene vinyl acetate
copolymers. Journal of Applied Polymer Science, 64(10), pp.1903–1912.
Available at: http://doi.wiley.com/10.1002/(SICI)1097-
4628(19970606)64:10<1903::AID-APP4>3.0.CO;2-M [Accessed December
1, 2014].
Brogly, M., Nardin, M. & Schultz, J., 1997b. Effect of vinylacetate content on
crystallinity and second-order transitions in ethylene?vinylacetate
copolymers. Journal of Applied Polymer Science, 64(10), pp.1903–1912.
Available at: http://doi.wiley.com/10.1002/%28SICI%291097-
4628%2819970606%2964%3A10%3C1903%3A%3AAID-
APP4%3E3.0.CO%3B2-M.
Bruker Optics, 2011. Application Note AN # 79 Attenuated Total Reflection (
ATR ) – a versatile tool for FT-IR spectroscopy Refractive index, Available
at: https://www.bruker.com/fileadmin/user_upload/8-PDF-
Docs/OpticalSpectrospcopy/FT-IR/ALPHA/AN/AN79_ATR-Basics_EN.pdf.
Camino, G. et al., 2000. Investigation of flame retardancy in EVA. Fire and
Materials, 24(2), pp.85–90. Available at:
http://doi.wiley.com/10.1002/1099-
1018%28200003/04%2924%3A2%3C85%3A%3AAID-
FAM724%3E3.0.CO%3B2-T [Accessed November 27, 2014].
Carlsson, T. et al., 2006. Absorption and desorption of water in glass/ethylene-
vinyl-acetate/glass laminates. Polymer Testing, 25(5), pp.615–622.
Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0142941806000821
[Accessed October 23, 2014].
Chen, S., Zhang, J. & Su, J., 2009a. Effect of Damp-Heat Aging on the
Properties of Ethylene-Vinyl Acetate Copolymer and Ethylene- Acrylic
Acid Copolymer Blends.
Appendix B
138
Chen, S., Zhang, J. & Su, J., 2009b. Effect of Hot Air Aging on the Properties of
Ethylene- Vinyl Acetate Copolymer and Ethylene-Acrylic Acid Copolymer
Blends. , 112, pp.1166–1174.
Chiu, M.-C. & Wang, M.-J.J., 2007. Professional footwear evaluation for clinical
nurses. Applied ergonomics, 38(2), pp.133–41. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/16765904 [Accessed December 1,
2014].
Claassen, R. S.; Butler, B.L., 1980. Introduction to solar materials science. Solar
materials science, pp.3–51. Available at:
http://adsabs.harvard.edu/abs/1980sms..book....3C.
Copuroglu, M. & Sen, M., 2005. A comparative study of UV aging
characteristics of poly(ethylene-co-vinyl acetate) and poly(ethylene-co-
vinyl acetate)/carbon black mixture. Polymers for Advanced
Technologies, 16(1), pp.61–66. Available at:
http://doi.wiley.com/10.1002/pat.538 [Accessed December 1, 2014].
Çopuroğlu, M. & Şen, M., 2004. A comparative study of thermal ageing
characteristics of poly(ethylene-co-vinyl acetate) and poly(ethylene-co-
vinyl acetate)/carbon black mixture. Polymers for Advanced
Technologies, 15(7), pp.393–399. Available at:
http://doi.wiley.com/10.1002/pat.485 [Accessed October 21, 2014].
Costache, M.C., Jiang, D.D. & Wilkie, C.A., 2005. Thermal degradation of
ethylene–vinyl acetate coplymer nanocomposites. Polymer, 46(18),
pp.6947–6958. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0032386105007329
[Accessed November 27, 2014].
Crank, J., 1968. Diffusion in Polymers G. S. P. John Crank, ed.,
Crank, J., 1975. Mathematics of Diffusion, Oxford: Oxford University Press.
Czanderna, a. W. & Jorgensen, G.J., 1997. Service lifetime prediction for
encapsulated photovoltaic cells/minimodules. AIP Conference
Proceedings, 394, pp.295–312. Available at:
http://scitation.aip.org/content/aip/proceeding/aipcp/10.1063/1.52899.
Czanderna, A.W. & Pern, F.J., 1996. Encapsulation of PV modules using
ethylene vinyl acetate copolymer as a pottant: A critical review. Solar
Energy Materials and Solar Cells, 43(2), pp.101–181. Available at:
http://linkinghub.elsevier.com/retrieve/pii/0927024895001506.
Department of Energy & Climate Change, 2013. UK renewable energy
roadmap: 2013 update, London. Available at:
https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/255182/UK_Renewable_Energy_Roadmap_-_5_November_-
_FINAL_DOCUMENT_FOR_PUBLICATIO___.pdf.
Dhere, Neelkanth G. and Pandit, Mandar, B., 2001. Study of Delamination in
Acceleration Tested PV Modules Authors. In 17th European Photovoltaic
Solar Energy Conference. Munich.
Appendix B
139
Dhere, N.G., 2000. PV module Durability in Hot and Dry Climate. In 16th
European Photovoltaic Solar Energy Conference.
Dolores Fernández, M. & Jesús Fernández, M., 2007. Thermal decomposition
of copolymers from ethylene with some vinyl derivatives. Journal of
Thermal Analysis and Calorimetry, 91(2), pp.447–454. Available at:
http://link.springer.com/10.1007/s10973-007-8552-3.
E. U. Reisner, G. Stollwerck, H. Peerlings, F.S., 2006. Humidity in a solar
module - Horror vision or negligible? In European Photovoltaic Solar
Energy. Dresden.
Elmahdy, A.E. et al., 2010. Stress measurement in East Asian lacquer thin films
owing to changes in relative humidity using phase-shifting
interferometry. Proceedings of the Royal Society A: Mathematical,
Physical and Engineering Sciences, 467(2129), pp.1329–1347. Available
at:
http://rspa.royalsocietypublishing.org/cgi/doi/10.1098/rspa.2010.0414
[Accessed October 30, 2014].
Fan, X.J., Lee, S.W.R. & Han, Q., 2009. Experimental investigations and model
study of moisture behaviors in polymeric materials. Microelectronics
Reliability, 49, pp.861–871.
Feng, L. & Kamal, M.R., 2005. Crystallization and melting behavior of
homogeneous and heterogeneous linear low-density polyethylene resins.
Polymer Engineering & Science, 45(8), pp.1140–1151. Available at:
http://doi.wiley.com/10.1002/pen.20389.
Flynn, J.H., 1995. A critique of lifetime prediction of polymers by thermal
analysis. Journal of Thermal Analysis and Calorimetry, 44(2), pp.499–512.
Available at: http://link.springer.com/10.1007/BF02636139.
Gilbert, J. et al., 1962. Carbonization of polymers I—Thermogravimetric
analysis. Polymer, 3, pp.1–10. Available at:
http://linkinghub.elsevier.com/retrieve/pii/0032386162900605
[Accessed November 27, 2014].
Glikman, J.-F. et al., 1986. Photolysis and photo-oxidation of ethylene-ethyl
acrylate copolymers. Polymer Degradation and Stability, 16(4), pp.325–
335.
Globus, A. et al., 2004. Teleoperated modular robots for lunar operations. In
American Institute of Aeronautics and Astronautics Inc. Chicago, pp. 788–
807.
Gu, H. et al., 2012. Thermal degradation kinetics of semi-aromatic polyamide
containing benzoxazole unit. Journal of Thermal Analysis and
Calorimetry, 107, pp.1251–1257.
Gulmine, J.. et al., 2002. Polyethylene characterization by FTIR. Polymer
Testing, 21(5), pp.557–563. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0142941801001246.
Hamed Gholami, Gholam Reza Razavi, M.S., 2011. Examination Synthesis of
Appendix B
140
Fluorapatite with Using Free Model (TG) and Kissingeer Method. In
Internation Conference on Advanced MaterialsEngineering. Singapore:
IACSIT Press, pp. 41–44.
Harris, K. & Annut, A., 2013. Digest of United Kingdom Energy Statistics
Production team : Iain MacLeay,
Häußler, L. , Pompe, G., Albrecht, V., Voigt, D., 1998. Bestimmung des
Vinylacetat-Gehaltes in eva copolymeren. Möglichkeiten und Grenzen
der Anwendung MS-gekoppelter Analysenmethoden. Thermal Analysis
and Calorimetry, 52(1), pp.131–143.
Heng-Yu Li, Laura-Emmanuelle Perret-Aebi, Ricardo Theron, Christophe Ballif,
Yun Luo, R.F., 2010. Towards in-line determination of EVA Gel Content
during PV modules Lamination Processes. In Proceedings of the 25th
PVSC Conference.
Herrick, C.E., 1966. Solution of the Partial Differential Equations Describing
Photodecomposition in a Light-absorbing Matrix having Light-absorbing
Photoproducts. IBM Journal of Research and Development, 10(1), pp.2–5.
Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5392
099.
Hülsmann, P. et al., 2010. Measuring and simulation of water vapour
permeation into PV modules under different climatic conditions. In N. G.
Dhere, J. H. Wohlgemuth, & K. Lynn, eds. SPIE. pp. 777307–777307–6.
Available at:
http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=7538
90 [Accessed December 5, 2014].
Hülsmann, P., Heck, M. & Köhl, M., 2013. Simulation of Water Vapor Ingress
into PV-Modules under Different Climatic Conditions. Journal of
Materials, 2013, pp.1–7. Available at:
http://www.hindawi.com/journals/jma/2013/102691/.
Hülsmann, P., Philipp, D. & Köhl, M., 2009. Measuring temperature-
dependent water vapor and gas permeation through high barrier films.
The Review of scientific instruments, 80(11), p.113901. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/19947736 [Accessed September
12, 2014].
Iwamoto, R. et al., 2003. Basic Interactions of Water with Organic Compounds.
The Journal of Physical Chemistry B, 107(31), pp.7976–7980. Available at:
http://dx.doi.org/10.1021/jp030561n\nhttp://pubs.acs.org/doi/pdf/10.1
021/jp030561n.
Iwamoto, R. & Matsuda, T., 2005. Interaction of water in polymers:
Poly(ethylene-co-vinyl acetate) and poly(vinyl acetate). Journal of
Polymer Science Part B-Polymer Physics, 43(7), pp.777–785. Available at:
file:///U:/Documents/Literature/Water
IR/iwamoto.water_interact_poly.jpolymersci.05.pdf.
Appendix B
141
J. Brandrup, E.H.I., 1999. Polymer handbook 4th ed. E. A. Grulke, ed., New
York: Wiley-Interscience.
J. C. McVeigh, 1983. Sun Power: An Introduction to the Applications of Solar
Energy 2nd (2013) ed., Oxford: Pergamon Press.
Jin, J., Chen, S. & Zhang, J., 2009. Investigation of UV aging influences on the
crystallization of ethylene-vinyl acetate copolymer via successive self-
nucleation and annealing treatment. Journal of Polymer Research, 17(6),
pp.827–836. Available at: http://link.springer.com/10.1007/s10965-009-
9374-8 [Accessed October 23, 2014].
Jin, J., Chen, S. & Zhang, J., 2010a. UV aging behaviour of ethylene-vinyl
acetate copolymers (EVA) with different vinyl acetate contents. Polymer
Degradation and Stability, 95(5), pp.725–732. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391010000911
[Accessed October 23, 2014].
Jin, J., Chen, S. & Zhang, J., 2010b. UV aging behaviour of ethylene-vinyl
acetate copolymers (EVA) with different vinyl acetate contents. Polymer
Degradation and Stability, 95(5), pp.725–732. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391010000911
[Accessed August 7, 2014].
Jorgensen, G.J. & McMahon, T.J., 2008. Accelerated and outdoor aging effects
on photovoltaic module interfacial adhesion properties. Progress in
Photovoltaics: Research and Applications, 16(6), pp.519–527. Available
at: http://doi.wiley.com/10.1002/pip.826 [Accessed November 20,
2014].
Kaczaj, J. & Trickey, R., 1969. Multiple isothermal degradation method for
determination of combined vinyl acetate in vinyl acetate-ethylene
copolymers. Analytical Chemistry, 41(11), pp.1511–1512. Available at:
http://pubs.acs.org/doi/abs/10.1021/ac60280a017 [Accessed November
27, 2014].
Kamath, P.M. & Wakefield, R.W., 1965. Crystallinity of ethylene–vinyl acetate
copolymers. Journal of Applied Polymer Science, 9(9), pp.3153–3160.
Available at: http://doi.wiley.com/10.1002/app.1965.070090919
[Accessed December 9, 2014].
Kapur, J., Proost, K. & Smith, C.A., 2009. Determination of moisture ingress
through various encapsulants in glass/glass laminates. In 2009 34th IEEE
Photovoltaic Specialists Conference (PVSC). IEEE, pp. 001210–001214.
Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5411
235.
Karimi, M., 2006. Diffusion in Polymer Solids and Solutions.
Kempe, M., 2006. Modeling of rates of moisture ingress into photovoltaic
modules. Solar Energy Materials and Solar Cells, 90(16), pp.2720–2738.
Available at:
Appendix B
142
http://linkinghub.elsevier.com/retrieve/pii/S0927024806001632
[Accessed November 3, 2014].
Kempe, M.D., 2008. Accelerated UV test methods and selection criteria for
encapsulants of photovoltaic modules. In 2008 33rd IEEE Photovolatic
Specialists Conference. IEEE, pp. 1–6. Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=4922
771 [Accessed November 21, 2014].
Kempe, M.D., 2005. Control of moisture ingress into photovoltaic modules.
Conference Record of the Thirty-first IEEE Photovoltaic Specialists
Conference, 2005., pp.503–506. Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=1488
180.
Kempe, M.D., 2008. Replacement for the Back-Sheet and Encapsulant Layers
Preprint.
Kempe, M.D., 2005. Rheological and Mechanical Considerations for
Photovoltaic Encapsulants. In DOE Solar Energy Technologies Program
Review Meeting. Denver.
Kempe, M.D., 2010. Ultraviolet light test and evaluation methods for
encapsulants of photovoltaic modules. Solar Energy Materials and Solar
Cells, 94, pp.246–253.
Kempe, M.D., Moricone, T.J. & Kilkenny, M., 2011. Accelerated Stress Testing
of Encapsulants for Medium- Concentration CPV Applications. In
Photovoltaic Specialists Conference (PVSC), 2009 34th IEEE. Pennsylvania:
NREL/CP-5200-46085. Available at:
http://www.nrel.gov/docs/fy11osti/46085.pdf.
Ken, Z., 1990. Harnessing Solar Power the Phpotovoltaics Challenge, New
York: Plenum Press.
Kim, N. & Han, C., 2013. Experimental characterization and simulation of
water vapor diffusion through various encapsulants used in PV modules.
Solar Energy Materials and Solar Cells, 116, pp.68–75. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0927024813001724
[Accessed July 15, 2014].
King, D.L. et al., 2000. Photovoltaic module performance and durability
following long-term field exposure. Progress in Photovoltaics: Research
and Applications, 8(2), pp.241–256. Available at:
http://doi.wiley.com/10.1002/%28SICI%291099-
159X%28200003/04%298%3A2%3C241%3A%3AAID-
PIP290%3E3.0.CO%3B2-D [Accessed November 20, 2014].
Klemchuk, P. et al., 1997. Investigation of the degradation and stabilization of
EVA-based encapsulant in field-aged solar energy modules. Polymer
Degradation and Stability, 55(3), pp.347–365. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391096001620
[Accessed November 3, 2014].
Appendix B
143
Kojima, T., 2004. Ultraviolet-ray irradiation and degradation evaluation of the
sealing agent EVA film for solar cells under high temperature and
humidity. Solar Energy Materials and Solar Cells, 85, pp.63–72. Available
at: http://linkinghub.elsevier.com/retrieve/pii/S0927024804002041
[Accessed November 3, 2014].
Kojima, T. & Yanagisawa, T., 2004. The evaluation of accelerated test for
degradation a stacked a-Si solar cell and EVA films. Solar Energy
Materials and Solar Cells, 81(1), pp.119–123. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0927024803002356
[Accessed December 1, 2014].
Krauter, S. et al., 2011. PV MODULE LAMINATION DURABILITY. In ISES Solar
Word Congress. Kassel. Available at: http://www.pi-
berlin.com/images/pdf/publication/ISES-2011-Krauter_et_al-
Module_Lamination_Durability.pdf.
Křižanovský, L. & Mentlík, V., 1978. The use of thermal analysis to predict the
thermal life of organic electrical insulating materials. Journal of Thermal
Analysis, 13(3), pp.571–580. Available at:
http://link.springer.com/10.1007/BF01912396.
Lacoste, J. et al., 1991. Polyethylene hydroperoxide decomposition products.
Polymer Degradation and Stability, 34(1-3), pp.309–323.
Lange, R.F.M. et al., 2011. The lamination of (multi)crystalline and thin film
based photovoltaic modules. Progress in Photovoltaics: Research and
Applications, 19(2), pp.127–133. Available at:
http://doi.wiley.com/10.1002/pip.993 [Accessed November 3, 2014].
Lee, K.-Y. & Kim, K.-Y., 2008. 60Co γ-ray irradiation effect and degradation
behaviors of a carbon nanotube and poly(ethylene-co-vinyl acetate)
nanocomposites. Polymer Degradation and Stability, 93(7), pp.1290–
1299. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391008001213
[Accessed December 1, 2014].
Li, C. et al., 2004. Crystallization of partially miscible linear low-density
polyethylene/poly(ethylene-co-vinylacetate) blends. Materials Letters,
58(27-28), pp.3613–3617. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0167577X04005129
[Accessed December 1, 2014].
Li, H. et al., 2013. Optical transmission as a fast and non-destructive tool for
determination of ethylene-co-vinyl acetate curing state in photovoltaic
modules. , (October 2011), pp.187–194.
Liu, M. et al., 2003. Study on diffusion behavior of water in epoxy resins cured
by active ester. Physical Chemistry Chemical Physics, 5, pp.1848–1852.
Liu, X., Wildman, R.D., Ashcroft, I. a., et al., 2012a. Modelling the mechanical
response of urushi lacquer subject to a change in relative humidity.
Proceedings of the Royal Society A: Mathematical, Physical and
Appendix B
144
Engineering Sciences, 468(2147), pp.3533–3551. Available at:
http://rspa.royalsocietypublishing.org/cgi/doi/10.1098/rspa.2012.0008
[Accessed August 7, 2014].
Liu, X., Wildman, R.D., Ashcroft, I. a., et al., 2012b. Modelling the mechanical
response of urushi lacquer subject to a change in relative humidity.
Proceedings of the Royal Society A: Mathematical, Physical and
Engineering Sciences, 468(2147), pp.3533–3551. Available at:
http://rspa.royalsocietypublishing.org/cgi/doi/10.1098/rspa.2012.0008
[Accessed October 23, 2014].
Liu, X., Wildman, R.D. & Ashcroft, I.A., 2012. Experimental investigation and
numerical modelling of the effect of the environment on the mechanical
properties of polyurethane lacquer films. Journal of Materials Science,
47(13), pp.5222–5231. Available at:
http://link.springer.com/10.1007/s10853-012-6406-2 [Accessed October
30, 2014].
Loo, Y.-L. et al., 2005. Thin crystal melting produces the low-temperature
endotherm in ethylene/methacrylic acid ionomers. Polymer, 46(14),
pp.5118–5124. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0032386105004805
[Accessed December 10, 2014].
Malmström, J., Wennerberg, J. & Stolt, L., 2003. A study of the influence of
the Ga content on the long-term stability of Cu(In,Ga)Se2 thin film solar
cells. Thin Solid Films, 431-432, pp.436–442. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0040609003001858
[Accessed December 5, 2014].
Marcilla, a., Gómez, a. & Menargues, S., 2005. TG/FTIR study of the thermal
pyrolysis of EVA copolymers. Journal of Analytical and Applied Pyrolysis,
74(1-2), pp.224–230. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0165237005000331
[Accessed October 31, 2014].
Marcilla, A. & Sempere, F.J., 2004. Differential scanning calorimetry of
mixtures of EVA and PE . Kinetic modeling. , 45, pp.4977–4985.
Marín, M.L. et al., 1996. Thermal degradation of ethylene (vinyl acetate).
Journal of Thermal Analysis, 47(1), pp.247–258. Available at:
http://link.springer.com/10.1007/BF01982703 [Accessed November 28,
2014].
Markus Bregulla, Michael Köhl, Benjamin Lampe, Gernot Oreski, Daniel
Philipp, Gernot Wallner, K.-A.W., 2007. DEGRADATION MECHANISMS OF
ETHYLENE-VINYL-ACETATE COPOLYMER – NEW STUDIES INCLUDING
ULTRA FAST CURE FOILS. In The compiled state-of-the-art of PV solar
technology and deployment. 22nd European Photovoltaic Solar Energy
Conference. Milan, pp. 5–8.
McGrattan, B.J., 1994. Examining the Decomposition of Ethylene-Vinyl
Acetate Copolymers Using TG/GC/IR. Applied Spectroscopy, 48, pp.1472–
Appendix B
145
1476.
Mcintosh, K.R. et al., 2009. THE EFFECT OF ACCELERATED AGING TESTS ON
THE OPTICAL PROPERTIES OF SILICONE AND EVA ENCAPSULANTS. In
Proceedings of the 24th European PVSEC (2009). Michigan. Available at:
http://www.dowcorning.com/content/publishedlit/06-1042.pdf.
McMahon, T.J., 2004. Accelerated testing and failure of thin-film PV modules.
Progress in Photovoltaics: Research and Applications, 12(23), pp.235–
248. Available at: http://doi.wiley.com/10.1002/pip.526 [Accessed
November 20, 2014].
McNeill, I.C. et al., 1976. The thermal degradation of copolymers of vinyl
acetate with methyl methacrylate and other monomers. European
Polymer Journal, 12(5), pp.305–312. Available at:
http://linkinghub.elsevier.com/retrieve/pii/0014305776901567
[Accessed November 27, 2014].
Moly, K.A. et al., 2005. Nonisothermal crystallisation, melting behavior and
wide angle X-ray scattering investigations on linear low density
polyethylene (LLDPE)/ethylene vinyl acetate (EVA) blends: effects of
compatibilisation and dynamic crosslinking. European Polymer Journal,
41(6), pp.1410–1419. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S001430570400374X
[Accessed December 1, 2014].
Moly, K.A. et al., 2005. Nonisothermal crystallisation, melting behavior and
wide angle X-ray scattering investigations on linear low density
polyethylene (LLDPE)/ethylene vinyl acetate (EVA) blends: Effects of
compatibilisation and dynamic crosslinking. European Polymer Journal,
41(6), pp.1410–1419.
Morlat-Therias, S. et al., 2007. Polymer/carbon nanotube nanocomposites:
Influence of carbon nanotubes on EVA photodegradation. Polymer
Degradation and Stability, 92(10), pp.1873–1882. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391007002054
[Accessed August 7, 2014].
Morshedian, J., 2009. The Accuracy of Approximation Equations in the Study
of Thermal Decomposition Behaviour of Some Synthesized Optically
Active Polyamides. , 18(11), pp.103–128.
Motta, C., 1997. The effect of copolymerization on transition temperatures of
polymeric materials. Journal of thermal analysis, 49(1), pp.461–464.
Available at: http://link.springer.com/10.1007/BF01987471 [Accessed
October 30, 2014].
Mubashar, a. et al., 2009. Moisture absorption–desorption effects in adhesive
joints. International Journal of Adhesion and Adhesives, 29(8), pp.751–
760. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0143749609000499
[Accessed October 23, 2014].
Appendix B
146
Mubashar, A. et al., 2009. Modelling Cyclic Moisture Uptake in an Epoxy
Adhesive. The Journal of Adhesion, 85, pp.711–735.
Muralidhara, K.S. & Sreenivasan, S., 2010. Thermal Degradation Kinetic Data
of Polyester , Cotton and Polyester-Cotton Blended Textile Material. ,
11(2), pp.184–189.
Myer Ezrin, Gary Lavigne, Petet Klemchuk, W. Holley, S. Agro, J. Galica, L.
Thomas, R.Y., 1995. Discolouration of EVA Encapsulant in Photovoltaic
Cells. ANTEC, pp.3957 – 3960. Available at:
http://syringeless.com/ANTEC1995.pdf.
Nogueira, P. et al., 2001. Effect of water sorption on the structure and
mechanical properties of an epoxy resin system. Journal of Applied
Polymer Science, 80, pp.71–80.
Oreski, G. & Wallner, G., 2005. Delamination behaviour of multi-layer films for
PV encapsulation. Solar Energy Materials and Solar Cells, 89(2-3),
pp.139–151. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0927024805000668
[Accessed November 3, 2014].
Oreski, G. & Wallner, G.M., 2010. Damp heat induced physical aging of PV
encapsulation materials. 2010 12th IEEE Intersociety Conference on
Thermal and Thermomechanical Phenomena in Electronic Systems, pp.1–
6. Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5501
297.
Oreski, G. & Wallner, G.M., 2010. DAMP HEAT INDUCED PHYSICAL AGING OF
PV ENCAPSULATION MATERIALS.
Órfão, J.J.M. & Martins, F.G., 2002. Kinetic analysis of thermogravimetric data
obtained under linear temperature programming - A method based on
calculations of the temperature integral by interpolation. Thermochimica
Acta, 390, pp.195–211.
Osterwald, C.R. et al., 2002. Degradation Analysis of Weathered Crystalline-
Silicon PV Modules Preprint. In IEEE PV Specialists Conference. Louisiana.
Available at: http://www.nrel.gov/docs/fy02osti/31455.pdf.
Osterwald, C.R. & McMahon, T.J., 2009. History of accelerated and
qualification testing of terrestrial photovoltaic modules: A literature
review. Progress in Photovoltaics: Research and Applications, 17(1),
pp.11–33. Available at: http://doi.wiley.com/10.1002/pip.861 [Accessed
November 21, 2014].
Osterwald, C.R., McMahon, T.J. & del Cueto, J.A., 2003. Electrochemical
corrosion of SnO2:F transparent conducting layers in thin-film
photovoltaic modules. Solar Energy Materials and Solar Cells, 79(1),
pp.21–33. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S092702480200363X
[Accessed November 20, 2014].
Appendix B
147
Palacios, J. et al., 2012. Thermal degradation kinetics of PP/OMMT
nanocomposites with mPE and EVA. Polymer Degradation and Stability,
97(5), pp.729–737. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391012000663
[Accessed October 23, 2014].
Peike, C. et al., 2012. Impact of Permeation Properties and Backsheet-
Encapsulant Interactions on the Reliability of PV Modules. ISRN
Renewable Energy, 2012, pp.1–5. Available at:
http://www.hindawi.com/journals/isrn.renewable.energy/2012/459731/
[Accessed August 7, 2014].
Pern, F., 1996. Factors that affect the EVA encapsulant discoloration rate upon
accelerated exposure. Solar Energy Materials and Solar Cells, 41-42,
pp.587–615. Available at:
http://linkinghub.elsevier.com/retrieve/pii/092702489500128X.
Pern, F.., Glick, S.. & Czanderna, A.., 1996. EVA encapsulants for pv modules:
Reliability issues and current R&D status at NREL. Renewable Energy, 8(1-
4), pp.367–370. Available at:
http://linkinghub.elsevier.com/retrieve/pii/0960148196888793
[Accessed December 3, 2014].
Pern, F.J., 1997. Ethylene-Vinyl Acetate (EVA) encapsulants for photovoltaic
modules: degradation and discoloration mechanisms and formulation
modifications for improved piiotostability. Angewandte
Makromolekulare Chemie, 252, pp.192–216.
Pern, F.J., 1993. Luminescence and absorption characterization of ethylene-
vinyl acetate encapsulant for PV modules before and after weathering
degradation. Polymer Degradation and Stability, 41(2), pp.125–139.
Available at:
http://linkinghub.elsevier.com/retrieve/pii/014139109390035H
[Accessed November 3, 2014].
Pern, F.J. et al., 2000. Photothermal Stability of Various Module Encapsulants
and The Effects of Superstrate and Substrate Materials Studied for
PVMaT Sources. In Denver, pp. 67–68.
Pern, F.J. & Glick, S.H., 2003. Adhesion Strength Study of EVA Encapsulants on
Glass Substrates, Denver.
Pern, F.J. & Glick, S.H., 1997. Improved Photostability of NREL- Developed EVA
Pottant Formulations for PV Module Encapsulation. , (September).
Pielichowski, K. & Njuguna, J., 2005. Thermal Degradation of Polymeric
Materials, Smithers Rapra Technology. Available at:
http://app.knovel.com/web/toc.v/cid:kpTDPM000C/viewerType:toc/root
_slug:thermal-degradation-polymeric/b-off-set:0/b-cat-name:Plastics
%26 Rubber/b-cat-slug:plastics-rubber/b-cat-id:214/b-sub-cat-id:8/b-
topic-name:Properties %26amp%3B Testing/b-topic-slug:properties-
amp-testing/b-off-set:0/b-order-by:copyright_year/b-sort-
by:descending/b-filter-by:all-content [Accessed February 4, 2015].
Appendix B
148
Post, E. et al., 1995. Study of recyclable polymer automobile undercoatings
containing PVC using TG / FTIR. , 263.
Price, D.M. & Church, S.P., 1997. FTIR evolved gas analysis of the
decomposition products of cellulose diacetate. Thermochimica Acta,
294(1), pp.107–112. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0040603196031504.
Quintana, M.A. et al., 2002. Commonly observed degradation in field-aged
photovoltaic modules. In Conference Record of the Twenty-Ninth IEEE
Photovoltaic Specialists Conference, 2002. IEEE, pp. 1436–1439. Available
at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=1190
879 [Accessed November 3, 2014].
Rashtchi, S. et al., 2012. Measurement of moisture content in photovoltaic
panel encapsulants using spectroscopic optical coherence tomography: a
feasibility study N. G. Dhere & J. H. Wohlgemuth, eds. , p.84720O.
Available at:
http://proceedings.spiedigitallibrary.org/proceeding.aspx?doi=10.1117/1
2.928959.
Rimez, B., Rahier, H., Van Assche, G., Artoos, T., Biesemans, M., et al., 2008.
The thermal degradation of poly(vinyl acetate) and poly(ethylene-co-
vinyl acetate), Part I: Experimental study of the degradation mechanism.
Polymer Degradation and Stability, 93(4), pp.800–810. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391008000207
[Accessed November 20, 2014].
Rimez, B., Rahier, H., Van Assche, G., Artoos, T. & Van Mele, B., 2008. The
thermal degradation of poly(vinyl acetate) and poly(ethylene-co-vinyl
acetate), Part II: Modelling the degradation kinetics. Polymer
Degradation and Stability, 93(6), pp.1222–1230. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391008000190
[Accessed November 20, 2014].
Riva, a. et al., 2003. Fire retardant mechanism in intumescent ethylene vinyl
acetate compositions. Polymer Degradation and Stability, 82(2), pp.341–
346. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391003001915
[Accessed November 27, 2014].
Rodríguez-Vázquez, M. et al., 2006. Degradation and stabilisation of
poly(ethylene-stat-vinyl acetate): 1 – Spectroscopic and rheological
examination of thermal and thermo-oxidative degradation mechanisms.
Polymer Degradation and Stability, 91(1), pp.154–164. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0141391005002065
[Accessed December 9, 2014].
S. Dietrich, M. Sander, M. Ebert, J.B., 2008. Mechanical Assessment of Large
Photovoltaic Modules by Test and Finite Element Analysis. In 23rd
European Photovoltaic Solar Energy Conference. Valencia, Spain, pp.
Appendix B
149
2889 – 2892. Available at: http://www.eupvsec-
proceedings.com/proceedings?paper=3604.
S.-H. Schulze, S. Dietrich, M. Ebert, J.B., 2008. Development of Test Methods
for Polymer Material Characterization in View of Long-Term Durability of
PV-Modules. In 23rd European Photovoltaic Solar Energy Conference.
Valencia, Spain, pp. 2903 – 2907. Available at: http://www.eupvsec-
proceedings.com/proceedings?paper=3650.
Salin, I.M. & Seferis, J.C., 1993. Kinetic analysis of high-resolution TGA variable
heating rate data. Journal of Applied Polymer Science, 47, pp.847–856.
Sánchez-Jiménez, P.E., Criado, J.M. & Pérez-Maqueda, L. a., 2008. Kissinger
kinetic analysis of data obtained under different heating schedules.
Journal of Thermal Analysis and Calorimetry, 94(2), pp.427–432.
Available at: http://link.springer.com/10.1007/s10973-008-9200-2.
Sascha Dietrich, Matthias Pander, M.E., 2009. Mechanical challenges of PV-
modules and its embedded cells experiments and finite element analysis.
In 24th European Photovoltaic Solar Energy Conference. Hamburg, pp.
3427–3431.
Schulze, S.-H. et al., 2009. Influence of Vacuum Lamination Process on
Laminate Properties – Simulation and Test Results. In 24th European
Photovoltaic Solar Energy Conference. Hamburg, pp. 3367–3372.
Sheats, J.R. & Diamond, J.J., 1988. Photochemistry In Strongly Absorblng
Media. , pp.4922–4938.
Shi, X. et al., 2009. Effect of Damp-Heat Aging on the Structures and
Properties of Ethylene-vinyl Acetate Copolymers with Different Vinyl
Acetate Contents.
Shi, X.M. et al., 2009. Effect of damp-heat aging on the structures and
properties of ethylene-vinyl acetate copolymers with different vinyl
acetate contents. Journal of Applied Polymer Science, 112(4), pp.2358–
2365.
Shi, X.M., 2008. Non-isothermal crystallization and melting of ethylene-vinyl
acetate copolymers with different vinyl acetate contents. eXPRESS
Polymer Letters, 2(9), pp.623–629. Available at:
http://www.expresspolymlett.com/articles/EPL-0000692_article.pdf
[Accessed October 31, 2014].
Shi, X.-M. et al., 2009. Effect of damp-heat aging on the structures and
properties of ethylene-vinyl acetate copolymers with different vinyl
acetate contents. Journal of Applied Polymer Science, 112(4), pp.2358–
2365. Available at: http://doi.wiley.com/10.1002/app.29659.
Shirrell, C.D., 1978. Diffusion of Water Vapor in Graphite/Epoxy Composites,
Advanced Composite Materials-Environmental Effects, ASTM STP 658 J. R.
Vinson, ed., American Society for Testing and Materials.
Siah, L.F., 2010. Moisture Sensitivity of Plastic Packages of IC Devices. ,
pp.503–522. Available at:
Appendix B
150
http://www.springerlink.com/index/10.1007/978-1-4419-5719-1.
Skowronski, T., Rabek, J.F. & Y, B.N.B., 1984. Photodegradation of Some
Poly(Viny1 Chloride) (PVC) Blends: PVC/Ethylene-Vinyl Acetate (EVA)
Copolymers and PVC/Butadiene-Acrylonitrile (NBR) Copolymers. , 24(4).
Smil, V., 2006. Energy: A Beginner’s Guide., Oxford: Oneworld.
Starink, M.., 2003a. The determination of activation energy from linear
heating rate experiments: a comparison of the accuracy of isoconversion
methods. Thermochimica Acta, 404(1-2), pp.163–176. Available at:
http://eprints.soton.ac.uk/18822/1/The_Determination_of_Activation_..
_by_MJ_Starink.pdf.
Starink, M.., 2003b. The determination of activation energy from linear
heating rate experiments: a comparison of the accuracy of isoconversion
methods. Thermochimica Acta, 404(1-2), pp.163–176. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0040603103001448
[Accessed August 7, 2014].
Stark, W. & Jaunich, M., 2011. Investigation of Ethylene/Vinyl Acetate
Copolymer (EVA) by thermal analysis DSC and DMA. Polymer Testing,
30(2), pp.236–242. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0142941810002126
[Accessed October 23, 2014].
Stone, J.L., 1993a. Photovoltaics: Unlimited Electrical Energy from the Sun.
Physics Today, 46(9), p.22. Available at:
http://scitation.aip.org/content/aip/magazine/physicstoday/article/46/9
/10.1063/1.881362 [Accessed December 28, 2014].
Stone, J.L., 1993b. Photovoltaics: Unlimited Electrical Energy from the Sun.
Physics Today, 46(9), p.22. Available at:
http://scitation.aip.org/content/aip/magazine/physicstoday/article/46/9
/10.1063/1.881362 [Accessed November 21, 2014].
Sultan, B.-Å. & Sörvik, E., 1991a. Thermal degradation of EVA and EBA—A
comparison. I. Volatile decomposition products. Journal of Applied
Polymer Science, 43(9), pp.1737–1745. Available at:
http://doi.wiley.com/10.1002/app.1991.070430917 [Accessed
November 27, 2014].
Sultan, B.-Å. & Sörvik, E., 1991b. Thermal degradation of EVA and EBA—A
comparison. II. Changes in unsaturation and side group structure. Journal
of Applied Polymer Science, 43(9), pp.1747–1759. Available at:
http://doi.wiley.com/10.1002/app.1991.070430918 [Accessed
November 27, 2014].
Sultan, B.-Å. & Sörvik, E., 1991c. Thermal degradation of EVA and EBA—A
comparison. III. Molecular weight changes. Journal of Applied Polymer
Science, 43(9), pp.1761–1771. Available at:
http://doi.wiley.com/10.1002/app.1991.070430919 [Accessed
November 27, 2014].
Appendix B
151
Tambe, S.P. et al., 2008. Ethylene vinyl acetate and ethylene vinyl alcohol
copolymer for thermal spray coating application. Progress in Organic
Coatings, 62(4), pp.382–386. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0300944008000489
[Accessed December 1, 2014].
Trombetta, M. et al., 2000. An FT-IR study of the internal and external
surfaces of HZSM5 zeolite. Applied Catalysis A: General, 192(1), pp.125–
136. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0926860X99003385.
Varghese, H. et al., 2002. Dynamic mechanical behavior of acrylonitrile
butadiene rubber/poly(ethylene-co-vinyl acetate) blends. Journal of
Polymer Science Part B: Polymer Physics, 40(15), pp.1556–1570. Available
at: http://doi.wiley.com/10.1002/polb.10204.
Vázquez, M. & Rey‐Stolle, I., 2008. Photovoltaic module reliability model
based on field degradation studies. Progress in Photovoltaics: Research
and Applications, 16(5), pp.419–433. Available at:
http://doi.wiley.com/10.1002/pip.825 [Accessed November 21, 2014].
Vieth, W.R., 1991. Diffusion in and Through Polymers: Principles and
Applications, Oxford University Press,.
Visoly-Fisher, I. et al., 2003. Factors Affecting the Stability of CdTe/CdS Solar
Cells Deduced from Stress Tests at Elevated Temperature. Advanced
Functional Materials, 13(4), pp.289–299. Available at:
http://doi.wiley.com/10.1002/adfm.200304259 [Accessed December 5,
2014].
Vyazovkin, S., 2006. MODEL-FREE KINETICS Staying free of multiplying entities
without necessity *. , 83, pp.45–51.
W. Herrmann, N. Bogdanski, F.R. et al., 2010. PV module degradation caused
by thermomechanical stress: real impacts of outdoor weathering versus
accelerated testing in the laboratory. In The Special Patrol
Insertion/Extraction Optics and Photonics (SPIE’10).
W.H. Holley, S.C. Agro, J.P.G. and R.S. & Yorgensen, 1996. UV Stability and
module testing of non-browning experimental PV encapsulants. In 25th
IEEE PVSC. Washington, pp. 1259–1262.
Wang, H., Tao, X. & Newton, E., 2004. Thermal degradation kinetics and
lifetime prediction of a luminescent conducting polymer. Polymer
International, 53, pp.20–26.
Wenger, H.J. et al., 1991. Decline of the Carrisa Plains PV power plant: the
impact of concentrating sunlight on flat plates. In The Conference Record
of the Twenty-Second IEEE Photovoltaic Specialists Conference - 1991.
IEEE, pp. 586–592. Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=1692
80 [Accessed December 2, 2014].
Wohlgemuth, J.H. & Petersen, R.C., 1991. Solarex experience with ethylene
Appendix B
152
vinyl acetate encapsulation. Solar Cells, 30(1-4), pp.383–387. Available
at: http://linkinghub.elsevier.com/retrieve/pii/037967879190071V
[Accessed November 3, 2014].
X. E. Cai, H.S., 1999. Apparent activation energies of the non-isothermal
degradation of EVA copolymer. Thermal analysis and calorimetry, 55,
pp.67–76.
Zanetti, M. et al., 2001. Synthesis and thermal behaviour of layered silicate ±
EVA nanocomposites. , 42, pp.4501–4507.
Appendix B
153
Appendix A
Water Vapour Transmission Rate Results
Appendix B
154
Appendix B
155
Appendix B
Original DSC curves
Appendix C
158
Appendix C
Publications
Appendix C
159
Technical papers in peer-reviewed journals
Status: Available online
Title: "The thermo-mechanical degradation of ethylene vinyl acetate used
as a solar panel adhesive and encapsulant"
doi:10.1016/j.ijadhadh.2016.03.008
Conferences
PVSAT-10, Loughborough, UK-Poster presentation 2014
Title: “Measurement of Moisture Diffusion Coefficient and Effect of Damp-
Heat Aging on the Structure and Properties of Ethylene-vinyl Acetate
(EVA) Copolymer as Encapsulant in Photovoltaic (PV) Modules"
SPIE, San Diego, USA-Oral presentation 2014
Title: "Effect of UV Aging on Degradation of Ethylene-vinyl Acetate (EVA)
as Encapsulant in Photovoltaic (PV) Modules"
Proc. SPIE 9179; doi: 10.1117/12.2062007
