CO2 absorption into a polymer within a multilayer structure: The case of poly(ethylene-co-vinyl acetate) in photovoltaic modules
Graphical Abstract
Introduction
Since the early 2000s, increased worldwide demand and the transition to renewable energy have driven a sharp increase in solar photovoltaic (PV) energy production. This growth is expected to continue, with an estimated PV capacity of between 7.6 and 22 TWp in 2050 [1]. First generation PV panels are now reaching the end of their lifetimes. The cumulative mass of end-of-life PV panels worldwide is expected to reach between 2 and 8 million tons by 2030 [2]. In this context of growing demand for solar-generated electricity and with the aim of recovering valuable materials, there is considerable demand for cost-effective and eco-friendly recycling processes. Existing PV panels mostly contain crystalline silicon PV cells: until 2016, more than 80% of installed PV panels were based on the Si-Al-BSF technology [3]. However, since cell technologies have and will continue to evolve rapidly, recycling techniques must be adaptable to new developments.
A crystalline silicon-based PV panel consists of a PV module stiffened by an aluminum frame and electrically controlled by a junction box. The PV module itself is a set of electrically connected crystalline silicon cells sandwiched between two layers of encapsulating polymer, a glass front face and a polymer “backsheet” (Fig. 1). The encapsulating polymer used in more than 90% of PV panels as of 2019 was poly(ethylene-co-vinyl acetate) (EVA) [1]. The “backsheet” is a multilayer assembly of polymers, most often a polyethylene terephthalate (PET) layer sandwiched between two layers of a fluoropolymer such as poly(vinyl fluoride) (PVF).
Recycling these modules involves several stages [4]. First a pre-treatment to remove the aluminum frame and the junction box; second, thermal [5], [6], chemical [7], [8], [9] and/or mechanical [10] treatments to delaminate the multilayer structure (silicon cell/glass/backsheet/EVA); and finally, a specific chemical treatment to extract the valuable metals (Ag, In, Si…) from the silicon cells [11], [12], [13], [14] or to recycle the silicon [15].
The chemical processes used for the delamination step typically involve organic solvents to dissolve the EVA (mainly by heating or with ultrasound or mechanical stirring). The reaction is slow however (> 1 h for 3 cm thick modules), and the solvents themselves are difficult to recycle [7]. Thermal treatments at temperatures above 500 °C to decompose the EVA and separate the glass from the cell are also lengthy (> 3 h for 5 × 5 cm2 pieces) [6]. The polymers also release toxic gases (HF from the backsheet and CO) as they decompose, requiring complex post-treatment of the gaseous effluents [5], substantially increasing the environmental burden of the recycling process [4]. Finally, existing mechanical treatments are generally poorly selective and lead to the loss of valuable materials such as silver and indium [10]. To integrate PV modules into a circular economy and reduce their environmental impact, we have therefore been investigating the use of supercritical CO2 (SC-CO2) as a greener means to recycle end-of-life PV panels.
In SC-CO2, the different layers of a PV module are separated, layer by layer, without breaking the glass and contaminating any of the valuable materials. The critical stage of PV delamination in a SC-CO2 medium is the separation of the EVA from the various components it encapsulates. The polymer absorbs SC-CO2 before rapid depressurization (at 1–300 bar·s−1) leads to foaming, which as observed for other polymers [16], [17], [18], can lead to loss of adhesion at the interfaces [19] and even delamination [20].
This paper focuses on the first stage of SC-CO2-induced delamination: the absorption of CO2 into EVA in PV modules. The absorption of CO2 into polymers has been well studied [21], [22], [23], [24]. In particular, Shieh and Lin [25] have investigated how different grades (i.e. with different weight percentages of vinyl acetate: EVA-16, EVA-18, EVA-25 and EVA-28) absorb CO2 at temperatures ranging from 25 °C to 52 °C and pressures of up to 340 bar, and Jacobs et al. [26] have studied CO2 absorption by EVA-40 at between 50 °C and 75 °C from atmospheric pressure up to 250 bar. However, the absorption of CO2 into a polymer in a multilayer structure has never been described. The absorption of a gas into a polymer depends mainly on two properties of the system: the phase equilibrium of the gas and the polymer (the solubility of the former in the latter) and the absorption kinetics (diffusion coefficient) of the gas in the polymer. In this study, we also consider how the absorption of CO2 into EVA is affected by the interfaces in the PV module.
Section snippets
Materials
The EVA used in this study was EVA-28 manufactured by STRE (Llanera, Spain). This copolymer is produced by radical polymerization and shaped into films (600 µm thick for the present study) containing cross-linking agents (peroxides) for hot lamination (> 120 °C) in PV modules. At ambient temperature (above the glass transition temperature), EVA is a semi-crystalline rubbery polymer. The constituent units derived from ethylene tend to form crystalline phases, while the vinyl acetate components
Melting behavior of EVA as a function of the CO2 pressure
As described in Section 2.1, the EVA used for solar applications is cross-linked and semi-crystalline. The presence of a crystalline phase, assumed to be impermeable to CO2, modifies the absorption of CO2 within the polymer. The temperature range at which PV modules are treated with SC-CO2 (60–90 °C) includes the melting range of EVA-28 at atmospheric pressure (70–80 °C), and since the presence of CO2 can lower the melting temperature of polymers [28], the variation of the melting range of EVA
Conclusion
This paper describes a holistic method to study SC-CO2 absorption into a polymer inside a multilayer structure, in this particular case EVA in PV modules. The characterization of EVA-28 by HP-DSC produced the phase diagram of the polymer in CO2 at pressures ranging from atmospheric to 150 bar, and showed that the melting range of EVA varies little as a function of pressure.
The swelling of EVA at equilibrium after stepped pressure increases was well modeled by the SL EoS, modified to account for
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (39)
- et al.
Life Cycle Assessment of an innovative recycling process for crystalline silicon photovoltaic panels
Life Cycle Environ. Ecol. Impact Anal. Sol. Technol.
(2016) - et al.
End-of-life of silicon PV panels: a sustainable materials recovery process
Waste Manag
(2019) - et al.
Dissolution of ethylene vinyl acetate in crystalline silicon PV modules using ultrasonic irradiation and organic solvent
Sol. Energy Mater. Sol. Cells
(2012) - et al.
Experimental investigations for recycling of silicon and glass from waste photovoltaic modules
Renew. Energy
(2012) - et al.
Pre-concentration and recovery of silver and indium from crystalline silicon and copper indium selenide photovoltaic panels
J. Clean. Prod.
(2020) Silver recovery aqueous techniques from diverse sources: hydrometallurgy in recycling
Waste Manag.
(2016)- et al.
Recovery of silver from leachate of silicon solar cells by solvent extraction with TOPO
Sep. Purif. Technol.
(2019) - et al.
Resource efficient recovery of critical and precious metals from waste silicon PV panel recycling
Waste Manag.
(2019) - et al.
A method to recycle silicon wafer from end-of-life photovoltaic module and solar panels by using recycled silicon wafers
Sol. Energy Mater. Sol. Cells
(2017) - et al.
Production of controlled polymeric foams by supercritical CO2
J. Supercrit. Fluids
(2007)
Foaming of polymers with supercritical fluids and perspectives on the current knowledge gaps and challenges
J. Supercrit. Fluids
Foaming of polystyrene and poly(methyl methacrylate) multilayered thin films with supercritical carbon dioxide
J. Supercrit. Fluids
Understanding and optimizing delamination/recycling of printed circuit boards using a supercritical carbon dioxide process
J. Clean. Prod.
Modeling solubilities of subcritical and supercritical fluids in polymers with cubic and non-cubic equations of state
J. Supercrit. Fluids
Towards a predictive thermodynamic description of sorption processes in polymers: the synergy between theoretical EoS models and vibrational spectroscopy
Mater. Sci. Eng.: R: Rep.
Equilibrium solubility of CO2 in rubbery EVA over a wide pressure range: effects of carbonyl group content and crystallinity
Polymer
Foam processing of poly(ethylene-co-vinyl acetate) rubber using supercritical carbon dioxide
Polymer
Swelling of network structures
Polymer
An equation of state (EoS) based model for the fluid solubility in semicrystalline polymers
Fluid Phase Equilibria
Cited by (5)
Facile preparation of high-performance fluorescent aramid using supercritical CO<inf>2</inf>
2023, Journal of Supercritical FluidsRecent progress in silicon photovoltaic module recycling processes
2022, Resources, Conservation and RecyclingCitation Excerpt :Combining supercritical CO2 and toluene solvent can delaminate the PV sandwich within 1 h (Lovato et al., 2021).( Briand et al., 2022b, 2022a; Cyril and Cedric, 2015) Pagnanelli et al. demonstrated that cyclohexane can be used to recover high-purity glass from pre-crushed fractions (a few mm in diameter) within 1 h. The authors found that reducing the size of the sample (i.e., crushing the intact module into smaller fractions) decreases the time needed for the solvent to penetrate through the polymer and therefore reduces the reaction time (Pagnanelli et al., 2019).
Deformation-induced delamination of photovoltaic modules by foaming ethylene-vinyl acetate with supercritical CO<inf>2</inf>
2022, Journal of CO2 UtilizationCitation Excerpt :After peeling the three reference specimens in each of the four samples (one per interface studied), these were placed in a 12 L autoclave connected to a complete SC-CO2 pilot assembly providing semi-automatic control of the experiments. The samples were contacted with SC-CO2 for 6 h before depressurization, about three times longer than the thermal equilibration time of CO2/EVA (about 2 h) [10]. The free volume in the autoclave was filled with cylinders made of high-density polyethylene (a material not sensitive to CO2 for the operating parameters used here) to increase the depressurization rate.