Dielectric properties of poly(ethyelene-co-butyl acrylate) filled with Alumina nanoparticles

87 Pages·2011·4.58 MB·English

co
Dielectric properties of poly(ethyelene- -butyl
acrylate) filled with Alumina nanoparticles
NADEJDA JÄVERBERG
Licentiate Thesis
Stockholm, Sweden 2011
TRITA-EE 2011:017 KTH School of Electrical Engineering
ISSN 1653-5146 SE-100 44 Stockholm
ISBN 978-91-7415-900-4 SWEDEN
Akademisk avhandling som med tillstånd av Kungl Tekniska högskolan framlägges
till offentlig granskning för avläggande av teknologie licentiatexamen fredagen den
25 mars 2011 klockan 10.15 i Seminarierummet, Teknikringen 33, 1 tr, Kungl
Tekniska högskolan, Stockholm.
© Nadejda Jäverberg, mars 2011
Tryck: Universitetsservice US AB
iii
Abstract
In this work dielectric properties of the poly(ethylene-co-butyl acrylate)
filled with alumina nanoparticles are evaluated. These nanocomposite mate-
rialsweremanufacturedatthedepartmentofFibreandPolymerTechnology,
KTH.
This study is limited to the properties of general importance for the AC ap-
plications. The dielectric permittivity of the nanocomposite materials was
studiedasafunctionoffillersize,fillercontent,coating,temperatureandair
humidity used for conditioning of the samples. The ultimate goal with this
projectistodescribetheinfluenceofmaterialcomposition,temperatureand
air humidity on the dielectric properties and model these dependencies.
In this thesis the experimental setup for voltage endurance testing of the
nanocomposites, namely studying applied voltage frequency dependence of
partial discharges in electrical trees, with a possibility of following electrical
treeing optically, was developed and described.
Thedielectricspectroscopymeasurementswereperformedonthoroughlydried
nanocomposites - so-called dry DS study. It was shown that the experimen-
tal data can be fitted with Havriliak-Negami approximation, which justifies
the correctness of the measurement results. It has been shown that adding
nanoparticlestotheEBAmatrixchangesthelowfrequencydispersionsignif-
icantly for the dried samples. It was also indicated that the particle coating
usedhasverylowimpactontheresultingpermittivityofthethoroughlydried
samples. From the dry DS studies it was suggested that the main cause of
the scattering in data between the dry samples is most likely the influence
of the material inhomogeneity and possibly the moisture absorption. This
leadstoapossibilityofusingdielectricspectroscopyasatoolforprobingthe
dispersion of nanoparticles in the polymer matrix.
Thedielectricspectroscopymeasurementswerealsocarriedoutonthenanocom-
positesconditionedintheenvironmentswithdifferenthumiditylevelsofairin
order to study the influence of absorbed water on the dielectric permittivity
- so-called wet DS study. From the wet study it was shown that for the wet
samples the amplitude of the loss peak is defined by the filler size, filler con-
tent and coating used; while its position in frequency domain is determined
by the coating and the humidity level used for conditioning.
Keywords: nanocomposites, poly(ethylene-co-butyl acrylate), EBA, alu-
mina, dielectric permittivity, absorbed water, humidity, temperature, dielec-
tric spectroscopy, partial discharge, electrical treeing
iv
Sammanfattning
Inomdettaarbeteevaluerasdedielektriskaegenskapernahosensampoly-
meravetenochbutylakrylatfylldmedAl2O3nanopartiklar.Dessananokom-
posita material framställdes på avdelningen för Fiber och Polymerteknologi,
KTH.
DennastudieavgränsastillegenskapernaavgenerelltintresseförACtillämp-
ningar. Den dielektriska permittiviteten av nanokomposita material studer-
adessomfunktionavpartiklarnasstorlek,fyllhalt,ytbehandling,temperatur
samt relativa luftfuktigheten under provbitarnas konditionering. Det ultima-
ta syftet med det här projektet är att beskriva inverkan av materialens sam-
mansättning,omgivningenstemperaturochluftfuktighetpådielektriskaegen-
skaper samt modellera dessa beroenden.
Ennymetodförundersökningavnanokompositernasspänningensuthållighet
harutvecklats,nämligenundersökningenavdenpålagdaspänningensfrekvens-
beroendeavpartiellaurladdningarielektriskaträd,medenmöjlighetattfölja
tillväxten av elektriska träd optiskt.
De dielektriska spektroskopimätningarna utfördes på torra nanokompositer
- så kallad torr DS studie. Det visades att experimentella data kan an-
passas med Havriliak-Negamis ekvation, vilket stärker mätdatas korrekthet.
Det visades att tillsättningen av nanopartiklar till EBA matrisen ändrar
den lågfrekventa dispersionen signifikant för torra provbitar. Det indikeras
även att partiklarnas ytbehandlingen har väldigt liten betydelse för de torra
provbitarnaspermittivitet.DetantydesidentorraDSstudienattdenstörsta
anledningen för spridningen i de torra provbitarnas mätdata är högst san-
nolikt inverkan av materialens inhomogenitet och möjligen absorberad fukt.
Detta leder till en möjlighet att använda dielektrisk spektroskopi som ett
verktyg för utforskningen av nanopartiklarnas dispersion i polymermatrisen.
De dielektriska mätningarna utfördes även på nanokompositer konditioner-
ade i luft med olika relativ luftfuktighet med avsikt att studera inverkan av
absorberat vatten på dielektrisk permittivitet - så kallad våt DS studie. Det
visades i den våta studien att förlusttoppens amplitud för de våta provbitar-
na definieras av partiklarnas storlek, fyllhalt samt ytbehandling; medan dess
position i frekvens bestäms av partiklarnas ytbehandling och den relativa
luftfuktigheten under konditioneringen.
v
Acknowledgements
First and foremost I would like to thank my supervisor Hans Edin for all the sup-
port, help and encouragement during the course of this project.
Mydeepthanksgotoallthemembersofthenano-projectgroup: PatriciaNordell,
Sohail Nawaz, Henrik Hillborg, Bruska Azhdar and Ulf W. Gedde for providing
feedback and sharing valuable knowledge that helped me greatly along the way.
My special thanks go to Patricia Nordell for putting up with my endless questions
and fruitful discussions.
I am very grateful to Nathaniel Taylor, Cecilia Forséen and Valentinas Dubikas for
makingmefeelwelcomeinthehigh-voltageresearchgroupwhenIstartedwiththis
project, and to Xiaolei Wang, Mohamad Ghaffarian Niasar, Respicius Clemence,
Håkan Westerlund and Roya Nikjoo for creating a wonderful and friendly atmo-
sphere to work in. My special thanks go to Nathaniel Taylor for introducing me
to the mysterious world of the IDA and all the discussions, sharing ideas and also
for posting the latex cheat sheet on my notice board (even though it was done in a
very sneaky way in the night) that helped me greatly with writing this thesis.
Iwouldalsoliketothankallthemembersofthereferencegroupinthenano-project
-ThomasLiljenberg(ABB)andClairPitois(ABB)forhelpingmetobecomemore
goal-orientedandfocusedinmywork;SvenJansson(Elforsk)foralwaysbeingkind
and supportive; Eva Malmström (KTH), Christian Andersson (Ericsson), Niclas
Schönborg (Svenska Kraftnät) and Erik Perzon (Borealis) for providing feedback
and guidance.
The project was funded by the Swedish Energy Agency, Elforsk AB, ABB AB,
Swedish Railway Company, Ericsson Network Technologies and Borealis via the
ELEKTRA program, grant 36028, and is gratefully acknowledged.
Last but not the least I would like to thank my family for all the encouragement
and support they’ve given me.
Nadja Jäverberg
Stockholm, March 2011
vi
List of publications:
This thesis is based on the following papers:
I "Applied Voltage Frequency Dependence of Partial Discharges in Electrical
Trees", N. Jäverberg and H. Edin, Proceedings of the 21st Nordic Insulation
Symposium, Göteborg, Sweden 2009
II "DielectricPropertiesofAlumina-filledpoly(ethylene-co-butylacrylate)Nanocom-
posites", N. Jäverberg, H. Edin, P. Nordell, H. Hillborg, B. Azhdar and U.W.
Gedde,2010AnnualReportConferenceonElectricalInsulationandDielectric
Phenomena, West Lafayette, USA, 2010
III "DielectricPropertiesofAlumina-filledpoly(ethylene-co-butylacrylate)Nanocom-
posites, part I - dry studies", N. Jäverberg, H. Edin, P. Nordell, S. Nawaz, H.
Hillborg, B. Azhdar and U.W. Gedde, submitted to IEEE Trans. Dielectr.
Electr. Insul.
IV "DielectricPropertiesofAlumina-filledpoly(ethylene-co-butylacrylate)Nanocom-
posites, part II - wet studies", N. Jäverberg, H. Edin, P. Nordell, S. Nawaz,
H. Hillborg, B. Azhdar and U.W. Gedde, submitted to IEEE Trans. Dielectr.
Electr. Insul.
Contents
Contents vii
1 Introduction 1
1.1 Need of new materials . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Nanocomposites: historical perspective . . . . . . . . . . . . . . . . . 2
1.3 Scope and aim of the project . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Thesis outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Author’s contributions . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Material description 7
2.1 Materials and manufacture data . . . . . . . . . . . . . . . . . . . . 7
2.2 Manufacturing process . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 Summary of material characterization . . . . . . . . . . . . . . . . . 9
2.4 Experimental matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3 Experimental evaluation methods 11
3.1 Dielectric spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 Voltage endurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4 Experimental 17
4.1 Dielectric spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Voltage endurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5 Summary of publications 35
6 General conclusions and future work 37
A Dry DS - Loss Tan δ 39
B Wet DS - Loss Tan δ 45
Bibliography 47
vii

Chapter 1
Introduction
1.1 Need of new materials
Electric insulation is of major importance in any electrical installation. Its main
function is to isolate two different electric potentials. A good electric insulation
material should be able to withstand
1. electrical stress (i.e. high breakdown strength, voltage endurance)
2. thermal stress (i.e. high fire point, low thermal expansion coefficient, etc.)
3. mechanical stress (i.e. supporting mechanical load)
4. ambient conditions (i.e. insulation material cannot be a health hazard, has
to be resistant to weather conditions, etc.)
Also, a low value of dielectric loss is desirable. Normally insulation material is
exposed to all of the above mentioned stresses which, in general, depend on one
another, e.g. both electrical and mechanical properties depend on temperature. A
lot of research directed at improving material properties was done over the years.
In the case of electrical insulation the particular interest was directed at the pu-
rification of the materials for enhancing electrical properties. On the other hand
such material properties as mechanical and thermal can be improved by filling
the material with small particles. The classical approach was filling materials with
micro-sizedparticleswhichgenerallyresultedinenhancingmechanicalandthermal
properties, but regretfully had negative impact on insulating properties. There are
theoretical models available for treating of micro-composite materials that agree
very well with experimental results, such as for example Maxwell-Garnett mixing
rule [1]. These mixing rules are based on a homogeneous mixture of inclusions of
different shapes and orientation in the host matrix.
However, the next generation of fillers, namely nano-sized inorganic fillers, show
promise for dielectric and insulation material properties. One interesting fact is
1
2 CHAPTER 1. INTRODUCTION
that the dielectric permittivity of a nano-composite sometimes decreases in com-
parisontothebasematrix, whiletheresultingpermittivity ofamicro-composite is
expected to be higher than that of the host polymer. This fact indicates that the
nanocomposites do not behave as classical filled heterogeneous materials.
There exist a large number of examples that illustrate enhancement of the di-
electric properties for the materials filled with nano-particles, for instance [2], [3],
[4]. The positive outcome is attributed to the role of the interface between the
nano-fillers and the host polymer matrix [5], [6]. Achieving good dispersion of
nanofillers in the host polymer matrix is one of the great challenges in manufac-
turing of nanocomposites; another issue is understanding the influence of surface
modificationofnanofillersonmaterialproperties. Theultimategoalwithnanocom-
posites is tailoring material properties for engineering applications. In this work a
modifiedPEhasbeenselectedasthebasematrix,namelyEBA-anEthylene-Butyl
Acrylate copolymer with 13 wt% butyl acrylate content, with the expectation that
using polar butyl acrylate groups will improve the particle dispersion. Alumina
nanoparticles are used as a filler in this study, as aluminium oxide is chemically
inert, non-toxic and widely used in industrial applications. The dielectric constant
of alumina is approimately 10 at 50 Hz with electrical resistivity higher than 1014
Ohm·cm.
1.2 Nanocomposites: historical perspective
The word "nanocomposite" is a composition of two words: "nano", which indicates
that the filler size has to be below 100 nm (at least in one dimension), and the
word "composite", which suggests that the material in question has at least two
distinct constituents. A lot of research is carried out on polymer nanocomposites
nowadays [7], [5], [8], [9], but the concept is by no means new. One example from
ancient times is the famous Lycurgus cup, illustrating that nanocomposites have
been successfully manufactured already in the 4th century AD. The uniqueness
of the cup is due to the dichroic effect exhibited by the glass of the cup: the
glass has greenish colour in transmitted light and reddish colour in reflected light.
This effect is believed to originate from the finely dispersed silver-gold alloy (7:3)
nanoparticles with diameters ranging between 50 to 100 nm; gold nanoparticles
causing reddish colour in transmission while silver nanoparticles being responsible
for greenish reflection [10].
AnotherexampleisthelustretechniquepopularduringtheRenaissanceperiod.
Lustreisametallicdecorationwhichcontainssuchsubstancesassilver,copperand
iron oxide applied on a ceramic. It is concluded that adding nanoparticles with
diameters ranging between 30 to 100 nm is responsible for the rich colors and the
iridescent effects typical to the lustre decorations produced by Italian masters. It
is suggested that adding silver and copper nanoparticles together produces gold
colour, while adding only copper nanoparticles results in ruby-red colour [11].
Both of the abovementioned examples are colloids. A colloidal system is de-