Cure state assessment of EVA-copolymers for PV-applications comparing dynamic-mechanical, dielectric and calorimetric properties

https://doi.org/10.1016/j.solmat.2015.07.024Get rights and content

Highlights

  • We compare DSC, DMA and DETA method for EVA cure monitoring.

  • Cure state modeling was conducted using Arrhenius model.

  • The cure monitoring method can be applied in module manufacturing.

Abstract

An ethylene vinyl acetate copolymer is widely used in photovoltaic industry as a pottant for crystalline solar cells. During solar module manufacturing the material undergoes a peroxide curing process which leads to elastomer like material properties. In this work, we applied a dielectric method using comb measuring electrodes for determination of EVA progress of cure. At first, we showed the direct correlation of calorimetric measurements, analyzing the consumption of peroxide cross linking agent, with dynamic viscosimetric measurements at different cure temperatures. In a second step, isothermal dielectric measurements have been chosen as a possible tool for direct cure monitoring in vacuum lamination processes. It was found that all methods show the same correlation and are powerful tools for cure monitoring. As we show in a final practical test, course of dielectric parameters can be monitored during a lamination procedure.

Introduction

Typical solar modules are designed in layer-wise structure where the solar cells are embedded between two foils of encapsulating material [1]. Front glass and back sheet cover the encapsulate stack against environmental stresses. The most common encapsulation material used since decades and still today is the ethylene vinyl acetate (EVA) copolymer [2], [3], [4], [5]. To achieve appropriate material properties, vinyl acetate (VAC) content between 27 and 33% was found to be most suitable for the embedment of brittle crystalline solar cells [6].

Since these thermoplastic EVA-types exhibit melting between 60 and 80 °C and would lose mechanical stability during the use in solar modules, the material needs to be covalently cross-linked during the solar module manufacturing process to ensure reliable cell encapsulation over a lifespan of at least 20 years [7], [8]. After cross linking the material exhibits low creep, better thermal stability, better stability against UV radiation and achieves good and permanent adhesion to glass

Covalently bonded elastomers are formed when linear polymer chains are connected through polymer cross links, leading to a three dimensional polymer network. General properties resulting from cross linking process are determined by the length of the network stands, entanglements, defects and the functionality of the cross links [9]. With increasing degree of cross linking, a decrease in EVA crystallinity can be observed leading to better optical properties [10].

Curing of EVA is achieved by peroxide cross linking agents that are incorporated during the foil manufacturing process. Typical cross linking agents are dicumylperoxide (DCP) [10] or 2,5-Bis(tert-butylperoxy)-2.5-dimethylhexane (TBEC) [1] and have a decomposition temperature starting from approximately 120 °C during solar module lamination, depending on the peroxide composition.

To determine the state of EVA curing, different methods have been developed. Frequently used is the Soxleth extraction method. In this procedure, the mass of a polymer sample before extraction is compared with the remaining mass of the sample after dissolution in a solvent. The test is destructive, meaning samples of laminated EVA films have to be removed from the solar module first. Then, non-bound polymer chains and other free components are extracted by the solvent during the Soxleth test and the mass of remaining cross linked polymer is determined [11].

Other tests are focusing on mechanical [12] or spectroscopic [13] methods, and are often dealing with the consumption of the cross linking agent [14], [6] and the description on the cross linking kinetics. Using these methods ex post after curing, helps to understand shift in glass transition properties or in reduction of heat released by the cross linking agent due to its partial consumption. Instead of controlling the state of cure after EVA processing and curing, in situ detection of the curing process during solar panel manufacturing is not yet solved. An overview on various ex ante, ex post or in situ methods was published recently [13]. The authors showed that parameters such as dynamic-mechanical properties and DSC results are directly correlating with each other during cure reaction. Therefore, we have focused on two main test procedures in this paper, DSC and DMA, and discussed the results with dielectric properties.

Dielectric behavior of EVA-copolymers has already been investigated by some researchers [6]. It was found that resistivity of EVA is decreasing with an increase in temperature. For EVA copolymers containing 33% VAC frequencies between 1 Hz and 100 kHz have been applied, giving good response. Applying these methods to understand cross linking kinetics is very useful for quality control of solar module manufacturing processes, since the consumption of cross linking agent can directly be influenced by the module manufacturer. Knowledge on the curing agent consumption is essential, since remaining peroxides can lead to discoloration of the solar module over time [1].

In this work, we present a dielectric measurement method that can be used for in situ controlling the curing process of EVA-copolymers. Dielectric behavior of polymer changes during chemical or physical reactions or transitions. Using proper electrodes and measurement equipment can be used to implement a permanent cure monitoring during the vacuum lamination process. As comparing analysis differential scanning calorimetry (DSC) and dynamic mechanical viscosimetry (DMA) have been applied to prove the correlation between cure agent consumption and built-up of covalent groups. Investigation of cure kinetics has been done applying Arrhenius modeling from results of viscosimetric and dielectric measurements.

Section snippets

Material

Commercially available EVA film has been used as an encapsulating material. The fast cure material was an EVA Etimex 496.10, manufactured by Etimex Packaging GmbH Germany. The material has been produced as a polymer film with a thickness of 0.48 mm. Specimens for experimental measurements have been taken out of the middle of the rolls, meaning that the first meters have been unrolled before material sampling. This ensures the original material properties because the material had no direct

Dynamic mechanical experiments

When polymers undergo chemical curing reaction, various material properties change during this process. One major property that changes is the mechanical behavior. When a polymer melt is characterized increasing cross linking density results in an increase in viscosity. At the same time, growth in stiffness can be observed. From dynamic-mechanical measurements, complex viscosity η, storage modulus G and loss modulus G have been measured during isothermal cure of EVA as shown in Fig. 2 for a

Conclusions

In this work we applied dynamic viscosimetric and dielectric method using comb electrodes to observe the progress of cure reaction of an EVA based encapsulating material. At first we showed that the consumption of peroxide cross linking agent directly correlates with the progress of cure reaction determined by viscosimetric measurements. In a second step, isothermal dielectric measurements have been carried out and the progress of cure reaction was again observed. While dynamic viscosimetric

Acknowledgments

Gratefully acknowledged is the funding from Fraunhofer society within the innovation cluster initiative SolarPolymers.

References (15)

There are more references available in the full text version of this article.

Cited by (0)

View full text