- Email: [email protected]
Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules
Waste Management xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules Pablo Dias ⇑, Selene Javimczik, Mariana Benevit, Hugo Veit Programa de Pós-Graduação em Engenharia de Minas, Metalúrgica e de Materiais (PPGE3M), Universidade Federal do Rio Grande do Sul (UFRGS), Av. Bento Gonçalves, 9500, 91509-900 Porto Alegre, RS, Brazil
a r t i c l e
i n f o
Article history: Received 11 April 2016 Revised 15 August 2016 Accepted 30 August 2016 Available online xxxx Keywords: Crystalline silicon Pyrolysis Polymers removal Recycling Solar panel
a b s t r a c t Photovoltaic (PV) modules contain both valuable and hazardous materials, which makes its recycling meaningful economically and environmentally. In general, the recycling of PV modules starts with the removal of the polymeric ethylene-vinyl acetate (EVA) resin using pyrolysis, which assists in the recovery of materials such as silicon, copper and silver. The pyrolysis implementation, however, needs improvement given its importance. In this study, the polymers in the PV modules were characterized by Fourier transform infrared spectroscopy (FTIR) and the removal of the EVA resin using pyrolysis has been studied and optimized. The results revealed that 30 min pyrolysis at 500 °C removes >99% of the polymers present in photovoltaic modules. Moreover, the behavior of different particle size milled modules during the pyrolysis process was evaluated. It is shown that polymeric materials tend to remain at a larger particle size and thus, this fraction has the greatest mass loss during pyrolysis. A thermo gravimetric analysis (TGA) performed in all polymeric matter revealed the optimum pyrolysis temperature is around 500 °C. Temperatures above 500 °C continue to degrade matter, but mass loss rate is 6.25 times smaller. This study demonstrates the use of pyrolysis can remove >99% of the polymeric matter from PV modules, which assists the recycling of this hazardous waste and avoids its disposal. Ó 2016 Published by Elsevier Ltd.
1. Introduction Photovoltaic energy is the third most important renewable energy in terms of installed capacity. Different scenarios expect that PV technologies will provide 2.5–25% of the global electricity demand by 2050 (Silva et al., 2014; Zuser and Rechberger, 2011). Tough photovoltaic energy is a renewable and non-pollutant form of energy (Zhang et al., 2008), the module (or panel) itself has a limited lifespan and eventually becomes electronic waste (e-waste) (Doi et al., 2001). According to (Paiano, 2015), the approximate lifespan of PV modules is 25 years and the first of two waste generation periods has started in 2012. This reflects on why the PV waste issue has only been addressed recently. The lifespan is determined by the nominal power: when it drops below 80%, the module is considered to have reached its end of life. As the PV market continues to grow, so will waste, even if such waste appears with a long time delay (Kazmerski, 2006). E-waste consists in dead electronic and electrical equipment and the current e-waste generation pattern poses one of the world’s greatest pollution problem. The main issues are the lack of appropriate recovery ⇑ Corresponding author. E-mail addresses: [email protected] (P. Dias), [email protected] (H. Veit).
technology (Jujun et al., 2014) and the risk of releasing hazardous substances such as cadmium, rare earths and brominated flame retardants (BFRs) if the waste is not disposed correctly (Widmer et al., 2005; Marwede and Reller, 2012). Most often, the discarded electronic goods end-up in landfills along with other municipal waste or are burnt with no gas emission control, releasing toxic and carcinogenic substances into the atmosphere (Dwivedy et al., 2015). Waste of electric and electronic equipment (WEEE) also consists of numerous precious and scarce metals, which makes its recycling meaningful economically and environmentally (Zeng et al., 2004; Ongondo et al., 2011). PV modules are considered a hazardous waste because of the metals they usually contain (Pb, Cr, Cd, Ni) (Tammaro et al., 2016). They are sorted according to their technology (European Commision, 2011): Crystalline silicon; Thin film; Concentrator photovoltaic (CPV) and emerging technologies; According to Tao and Yu (2015), silicon wafer-based PV modules are the most common type of solar cell manufactured in the world. This is based on the inherent advantages of silicon as a
http://dx.doi.org/10.1016/j.wasman.2016.08.036 0956-053X/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
2
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
semiconductor and on the major industry that has been built to manufacture silicon devices (Bruton, 2002). Crystalline silicon PV modules are made from the following materials, in order of mass: glass, aluminum frame, EVA copolymer transparent encapsulating layer, photovoltaic cells, installation box, TedlarÒ protective foil and assembly bolts (Radziemska, 2014). Modules are encapsulated with various materials to protect the cells and the electrical connectors from the environment (Hansen et al., 2000). The removal of these encapsulating materials (being EVA resin the most common) is usually the first step in recycling PV modules. Several methods to remove the EVA can be employed, such as dissolution using nitric acid and thermal decomposition (usually pyrolysis) (Bruton, 1994). The pyrolysis process consists in heating the material to elevated temperatures, in absence of oxygen, to decompose organic volatile matter (gases, rubbers, etc.) to gas and liquids. The inorganic components (metal, fillers, glasses, etc.) remain almost unaltered during the process, and consequently the valuable components can be recovered and reused. The pyrolysis process is especially appropriate for complex wastes such as WEEE streams, which contain many different plastics mixed with other materials (de Marco et al., 2008). The absence of oxygen during pyrolysis avoids the oxidation of some materials and the formation of dibenzo-p-dioxins and dibenzofurans from the flame retardant found in some devices (Zhang et al., 2016). Most PV recycling methods use thermal processes to remove organics, Kang et al. (2012) use a thermal decomposition to remove the adhesive layer and recover the semiconductor. Wang et al. (2012) use a two-step heating process for the thermal delamination of PV module. Fthenakis (2000) discusses the recovery of metals using a smelter in which EVA and other polymers decompose at high temperatures. Berger et al. (2010) aim to recycle thin film PV modules and use a thermal dismantling process at 470 °C
in order to destroy the EVA layer. Dias et al. (2016a) study the destruction of the polymeric fraction in order to recycle modules and state that the use of pyrolysis could minimize environmental impacts by avoiding the use of organic solvents. Radziemska et al. (2010) recycle silicon solar cells using two main stages: separation and cleansing. The separation stage is a thermal delamination used to remove the EVA resin and separate the other layers. McDonald and Pearce (2010) state that the recycling process for any silicon PV module is identical and involves pyrolysis, which assists in recovering crystalline silicon wafer from the waste PV. The use of an inert gas - such as nitrogen - prevents chemical oxidation of the EVA and the silver contact grid, resulting in a clean, almost residue free cell surface (John and Anisimov, 1997). Previous studies, however, show that silver recovery should occur prior to pyrolysis (Dias et al., 2016b). The EVA pyrolysis in inner atmosphere consists on the removal of the acetic acid from the main chain and the degradation of the remaining polyethylene co-polyethylene (Zeng et al., 2004). All of these studies use pyrolysis in their recycling process but Zeng et al. (2004) published the only study that optimizes the pyrolysis itself. Thus, because of its importance for PV recycling, it is necessary to evaluate other aspects of the PV pyrolysis process. In this study, the polymeric materials used in PV panels were characterized and the use of pyrolysis as a separation process is studied, evaluated and optimized in order to assist recycling of waste PV and enable the recovery the valuable materials present in the modules.
2. Materials and methods A schematic diagram has been drawn in order to outline the investigations performed in this study (Fig. 1).
Fig. 1. Schematic diagram illustrating the procedures used in this study.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
Fig. 2. Illustration of silicon module’s components (Dow Corning, 2016)(Dow Corning, 2016).
The experiments were performed with crystalline silicon modules (c-Si modules). The aluminum frames were removed from all modules manually assisted by tools such as screwdriver, plier and small electric saw. The polymeric were also removed manually with the aid of a magnifying glass and a stylet to separate the different layers. A Fourier transform infrared spectroscopy (FTIR) was performed in the encapsulating layer and in the back sheet layer (Fig. 2) using a Spectrum 100 equipment (Perkin Elmer), spectral window of 4000–650 cm 1 and resolution of 4 cm 1. The FTIR was intended to identify the polymers that were used the PV modules. The database used to compare FTIR spectrums using Euclidean distance was Sadtler Inorganic Library. The encapsulating layer is an adhesive material used to bond the glass and the silicon cells. Its removal is the most difficult step in separating the module’s components. Even when the glass layer was removed manually, the adhesive material kept glued to the semiconductor. Thus, the idea of the pyrolysis approach is to remove the adhesive material beforehand so that the other materials can be separated afterwards. The encapsulating layer was scraped manually and the obtained samples were analyzed by thermo gravimetric analysis (TGA) in platinum crucibles using a TGA Q50 (TA Instruments, New Castle, USA). The parameters used in the analysis followed the ASTM E11317 standards (ASTM E1131-08, 2014). The analysis occurred in nitrogen atmosphere, the temperature varied from 20 to
3
500 °C at a 10 °C/min rate. A duplicate sample of the module’s back sheet (Fig. 2) was scraped and analyzed by TGA. Analysis occurred in a nitrogen atmosphere, temperature varied from 20 to 600 °C at a 10 °C/min rate. The modules were milled four times on a SRB 2305 knife mill (Rone, São Paulo, Brasil) based on previous works (Dias et al., 2016b). The mill screen used had a 4 mm opening in the first two times and 2 mm opening in the second two; the output powder weighed approximately 4 kg. As stated by Cui et al. (2003), for maximum separation of materials, WEEE should be shredded to small particles. The powder obtained was separated in three groups, according to the particle size. To separate the powder, a mesh 35 (0.5 mm aperture) and a mesh 18 (1 mm aperture) were used. The sieving occurred in a vibrating sieving device; for every 300 g of material, the equipment vibrated for 15 min with and amplitude of 1 mm. The three obtained fractions (smaller than 0.5 mm, between 1 mm and 0.5 mm and bigger than 1 mm) were stored separately, weighted and named F1, F2 and F3, respectively. The results from the TGA encouraged a study on the procedure parameters for the pyrolysis process. A furnace structure was assembled to control the process, as shown in Fig. 3. The pyrolysis was carried out using 150 30 cm2 alumina boat-shaped crucibles; the crucibles were filled, weighed, placed inside the furnace and removed periodically for weighting (Table 1). The approximate initial weight for each sample was 9.0 g, quartering was used to ensure that the samples were significant. The dwell temperature chosen was 500 °C based on the previous TGA, the heating ramp was 15 °C/min, and 1 L/min nitrogen airflow was maintained throughout the whole process (adapted from Zeng et al., 2004). This pyrolysis was repeated four times (quadruplicate samples) for each particle size fraction (F1, F2 and F3). In the interest of evaluating the influence of the pyrolysis duration and particle size, a variance analysis using the GLM module of SAS software (SAS Inst., Inc. Cary, MC, 1989) was performed. Significance was tested by the F test, and mass loss means compared by LSMean (least squares). The evaluation considered fractions F1, F2 and F3 as well as pyrolysis times 20, 30, 60 and 180 min. In order to verify if pyrolysis was able to remove all organic matter, another TGA was performed in the samples after the pyrolysis. The analysis occurred in nitrogen atmosphere, temperature varied from 20 to 700 °C at a 10 °C/min rate. The parameters used in the analysis followed the ASTM E11317 standards (ASTM E1131-08, 2014).
Fig. 3. Schematic illustration of the furnace setup.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
4
P. Dias et al. / Waste Management xxx (2016) xxx–xxx Table 1 Parameters used in the pyrolysis. Fraction
Initial mass (g)
Temperature (°C)
N2 flow rate (L/min)
Weighing interval (min)
Total pyrolysis time (h)
F1
9
500–550
1
F2
9
500–550
1
F3
9
500–550
1
60 30 20 60 30 20 60 30 20
3 1 1 3 1 1 3 1 1
3. Results Regarding the characterization of the polymers from the modules, visual inspection reveals the back sheet (Fig. 2) is a multilayer structure as illustrated in Fig. 4. The FTIR results show that the encapsulating material is ethylene-vinyl acetate (EVA) (Fig. 5) and that the back sheet layer is composed of polyvinyl fluoride (PVF – commercially known as TedlarÒ) (Fig. 6), Polyethylene terephthalate (PET) (Fig. 7) and EVA (Fig. 5). These results are in accordance with Radziemska (Radziemska, 2014). The adsorption bands at 3000–2800 cm 1 have been attributed to the CAH bond stretching, which presents particularly strong intensity in the
Fig. 4. Illustration of the multilayer back sheet structure.
EVA spectrum due to the long CAH chain (Fig. 5). The EVA also displays the CAO asymmetric stretching vibration band point at 1236 cm 1 and the CH2 rocking vibration at 720 cm 1. The PVF spectrum (Fig. 6) presents the CAH bond stretching as well as the very strong intensity double band CAF bond stretching at 1100–1000 cm 1. Finally, regarding the PET spectrum (Fig. 7), the adsorption bands at 1710 cm 1 has been attributed to the alkanoate ester C@O bond stretching, the 1118 cm 1 to the phthalates asymmetric CAO bond stretching vibration, the 1096 cm 1 to the symmetric CAO bond stretching vibration and the 1243 cm 1 the asymmetric CAOAC bond stretching. The band observed at 723 cm 1 was attributed to the @CAH out-of-plane deformation vibration, which is a result of the para substitution in the aromatic ring found in the adsorption bands 900–675 cm 1. The TGA results from the encapsulating layer are displayed in Fig. 8. The observed behavior is in accordance with Zeng et al. (2004) and Berger et al. (2010) and display the behavior of the EVA polymer when heated. Fig. 8 shows that the entire polymeric layer degrades at 500 °C. The TGA of the milled module sample is displayed in Fig. 9 and shows that approximately 75% (in weight) of the polymeric fraction degrades between 400 °C and 500 °C. Thus, most of the polymeric blends and combinations preset in the PV modules can be degraded reaching 500 °C. A detailed analysis of the substrate’s TGA (Fig. 9) is displayed in Table 2 and shows the average mass loss between 400 °C and 500 °C is 6.25 times greater than the mass loss between 500 °C and 600 °C. These results (Figs. 8, 9 and Table 2) encouraged the use of 500 °C as the temperature dwell for the pyrolysis used in this study.
Fig. 5. FTIR - Peaks found in analysis (above) compared to EVA peaks from database (below).
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
5
Fig. 6. FTIR - Peaks found in analysis (above) compared to PVF peaks from database (below).
Fig. 7. FTIR - Peaks found in analysis (above) compared to PET peaks from database (below).
Fig. 8. TGA results for the encapsulating layer.
Fig. 9. TGA results for the back sheet of the module.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
6
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
Table 2 TGA numerical analysis. Temperature range (°C)
Max mass loss
400–500 500–600
1.48 (at 441 °C) 0.04
%
C
)
Average
%
C
)
0.25 0.04
Table 3 Pyrolysis results for particle size F1, F2 and F3. Pyrolysis total time (h)
F1
F2
F3
Wt%
SD
Wt%
SD
Wt%
SD
t=0 t=1 t=2 t=3
100 93.54 93.74 93.69
– 1.3479 1.8609 4.4227
100 91.544 91.611 91.542
– 3.6177 3.5224 3.6028
100 85.14 84.91 84.91
– 6.378 6.150 6.002
The pyrolysis was carried out for all three particle size fractions (F1, F2, F3). The results show the average weight decrease measured in the experiments (Table 3). It is noticeable that the weight loss increases as particle size increases, this indicates that polymeric fraction tends to concentrate in the larger particle size group: polymer concentration = F3 > F2 > F1. This weight loss behavior is in accordance with Granata et al. (2014). Moreover, the weight decrease from the second hour to the third and forth hour is about 1% or less in all fractions. These results indicate a 1 h dwell at 500 °C results in an efficient pyrolysis for all particle size groups studied and that total pyrolysis time over 1 h does not influence the pyrolysis mass loss. In order to prove this statement, TGAs were performed in samples after pyrolysis for all three fractions. The results are displayed in Figs. 10–12.
Fig. 12. TGA of the pyrolysed sample - fraction F3. Table 4 Particle size and pyrolysis time (duration) influence in mass loss. Fraction
Mass (%)
Error
F1 F2 F3 P
93.55a 90.65b 82.69c <0.001
±0.97 ±0.94 ±0.97
Pyrolysis time (min) 20 30 60 180 P
91.58a 87.5b 88.42b 88.35b 0.08
±1.20 ±1.08 ±1.08 ±1.08
a,b,c Means in the same parameter column with different superscripts are significantly different for given p-value. There were 4 replicates per parameter.
Further investigations were conducted with 30 min and 20 min weighing intervals, the results are displayed in Table 4. The analysis of Table 4 indicates particle size distribution influences significantly in the weight loss (p = 0.001). These results corroborate the hypothesis that the polymers are concentrated in the larger fraction, followed by the intermediate and small fractions where mass loss is smaller. Table 4 also indicates that a 30, 60 and 180 min pyrolysis are not statistically different from each other, thus a pyrolysis lasting 30 min has a mass loss as significant as pyrolysis lasting for 60 or 180 min for p = 0.08. However, pyrolysis 20 min differs statistically from the others. Therefore, based on the previous TGAs (Figs. 10–12) and these statistical results (Table 4), the 30 min pyrolysis is optimum because it is the shortest time that ensures complete pyrolysis regardless of particle size. Fig. 10. TGA of the pyrolysed sample - fraction F1.
4. Conclusions The key conclusions from this study are:
Fig. 11. TGA of the pyrolysed sample - fraction F2.
The possible polymers present in silicon photovoltaic modules are – but are not limited to - ethylene-vinyl acetate (EVA), polyvinyl fluoride (PVF – commercially known as TedlarÒ) and polyethylene terephthalate (PET). Thermo gravimetric analysis reveals 75% of the polymers from PV modules are degraded between the temperatures of 400 and 500 °C. Temperatures above 500 °C continue to degrade matter, but mass loss rate is 6.25 times smaller than before 500 °C. When modules are milled and sieved, polymeric matter concentrates in larger particle size. Polymer size has great influence in mass loss during pyrolysis (p = 0.01). The larger the particle size, the greater the mass loss.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
A pyrolysis performed at 500 °C for 1 h removes >99% of the polymers from PV modules. Statically, pyrolysis performed at 500 °C with a 1 h dwell or with a 30 min dwell do not differ (for a p-value = 0.08). Thus, 30 min dwell at 500 °C are best parameters for the pryrolysis of silicon PV modules. These results are valid for samples of approximately 9 g (bench scale); further investigations should be made to assess the behavior of the process in larger scales. Acknowledgments The authors are grateful to CAPES, CNPq, FINEP and FAPERGS (Brazil) for their financial support. References ASTM E1131-08, 2014. Standard Test Method for Compositional Analysis by Thermogravimetry. ASTM International, West Conshohocken, PA, 2014,. Berger, W. et al., 2010. A novel approach for the recycling of thin film photovoltaic modules. Resour. Conserv. Recycl. 54, 711–718. Bruton, T.M., 1994. Re-cycling of high value, high energy content components of silicon PV modules. In: Proceedings of the 12th EC-PVSEC. Amsterdam, Netherlands: s.n., April 11–15, 1994. pp. 459–463. Bruton, T.M., 2002. General trends about photovoltaics based on crystalline silicon. Sol. Energy Mater. Sol. Cells 72 (1–4), 3–10. Cui, Jirang, Forssberg, Eric, 2003. Mechanical recycling of waste electric and electronic equipment: a review. J. Hazard. Mater. 99 (3), 243–263. de Marco, I. et al., 2008. Pyrolysis of electrical and electronic wastes. J. Anal. Appl. Pyrol. 82 (2), 179–183. Dias, P.R., Benevit, M.G., Veit, H.M., 2016a. Photovoltaic solar panels of crystalline silicon: characterization and separation. Waste Manage. Res. 34 (3), 235–245. Dias, P.R. et al., 2016b. Recycling WEEE: extraction and concentration of silver from waste crystalline silicon photovoltaic modules. Waste Manage. http://dx.doi. org/10.1016/j.wasman.2016.03.016. Doi, Takuya. et al., 2001. Experimental study on PV module recycling with organic solvent method. Sol. Energy Mater. Sol. Cells 67 (1–4), 397–403. Dow Corning, 2016. Solar Energy 101: Introduction to Solar Energy, 2016. Last Accessed: 04-02-2016 . Dwivedy, M., Suchde, P., Mittal, R.K., 2015. Modeling and assessment of e-waste take-back strategies in India. Resour. Conserv. Recycl. 96, 11–18. European Commision, 2011. Study on photovoltaic panles supplementing the impact assessment for a recast of the weee directive. Environment - Waste – WEEE, 2011. (last accessed in: January/2016).
7
Fthenakis, V.M., 2000. End-of-life management and recycling of PV modules. Energy Policy 28, 1051–1058. Granata, G. et al., 2014. Recycling of photovoltaic panels by physical operations. Sol. Energy Mater. Sol. Cells 123, 239–248. Hansen, A.D., et al., 2000. Models for a Stand-Alone PV System. Risø National Laboratory, Roskilde, 2000, pp. Risø-R-1219(EN)/SEC-R-12. John, R.B., Anisimov, I.I., 1997. Possibility of recycling silicon PV modules. In: 26th Photovoltaic Specialists Conference (PVSC), Anaheim, CA, Sept. 30–Oct. 3, 1997, pp. 0-7803-3767-0/97. Jujun, R., Yiminga, Q., Zhenming, X., 2014. Environment-friendly technology for recovering nonferrous metals from e-waste: Eddy current separation. Resour. Conserv. Recycl. 87, 109–116. Kang, Sukmin et al., 2012. Experimental investigations for recycling of silicon and glass from waste photovoltaic modules. Renewable Energy 47 (2012), 152–159. Kazmerski, L.L., 2006. Solar photovoltaics R&D at the tipping point: a 2005 technology overview. J. Electron Spectrosc. Relat. Phen. 150, 105–135. Marwede, M., Reller, A., 2012. Future recycling flows of tellurium from cadmium telluride photovoltaic waste. Resour. Conserv. Recycl. 69, 35–49. McDonald, N.C., Pearce, M.J., 2010. Producer responsibility and recycling solar photovoltaic modules. Energy Policy 38, 7041–7047. Ongondo, F.O., Williams, I.D., Cherrett, T.J., 2011. How are WEEE doing? A global review of the management of electrical and electronic wastes. Waste Manage. 31, 714–730. Paiano, Annarita., 2015. Photovoltaic waste assessment in Italy. Renew. Sustain. Energy Rev. 41, 99–112. Radziemska, E.K., 2014. Recycling of Photovoltaic Solar Cells and Modules - The State-Of-Art. s.l.: Saarbrucken. LAP LAMBERT Academic Publishing, Germany. Radziemska, Ewa Klugmann. et al., 2010. Experimental validation of crystalline silicon solar cells recycling by thermal and chemical methods. Sol. Energy Mater. Sol. Cells 94 (12), 2275–2282. Silva, M.D. et al., 2014. Integration of technologies and alternative sources of water and energy to promote the sustainability of urban landscapes. Resour. Conserv. Recycl. 91, 71–81. Tammaro, Marco. et al., 2016. Experimental investigation to evaluate the potential environmental hazards of photovoltaic panels. J. Hazard. Mater. 306, 395–405. Tao, J., Yu, S., 2015. Review on feasible recycling pathways and technologies of solar photovoltaic modules. Sol. Energy Mater. Sol. Cells 141, 108–124. Wang, T.-Y., Hsiao, J.-C., Du, C.-H., 2012. Recycling of materials from silicon base solar cell module. In: Proceedings of the 38th IEEE Photovoltaic Specialists Conference, pp. 2355–2358. Widmer, R. et al., 2005. Global perspectives on e-waste. Environ. Impact Assess. Rev. 25, 436–458. Zeng, D., Born, M., Wambach, K., 2004. Pyrolysis of EVA and its application in recycling of photovoltaics. J. Environ. Sci. 16 (6), 889–893. Zhang, Lifeng, Ciftja, Arjan., 2008. Recycling of solar cell silicon scraps through filtration, Part I: Experimental investigation. Sol. Energy Mater. Sol. Cells 92 (11), 1450–1461. Zhang, Mengmei, Buekens, Alfons, Li, Xiaodong, 2016. Brominated flame retardants and the formation of dioxins and furans in fires and combustion. J. Hazard. Mater. 304, 26–39. Zuser, A., Rechberger, H., 2011. Considerations of resource availability in technology development strategies: the case study of photovoltaics. Resour. Conserv. Recycl. 56 (1), 56–65.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules Pablo Dias ⇑, Selene Javimczik, Mariana Benevit, Hugo Veit Programa de Pós-Graduação em Engenharia de Minas, Metalúrgica e de Materiais (PPGE3M), Universidade Federal do Rio Grande do Sul (UFRGS), Av. Bento Gonçalves, 9500, 91509-900 Porto Alegre, RS, Brazil
a r t i c l e
i n f o
Article history: Received 11 April 2016 Revised 15 August 2016 Accepted 30 August 2016 Available online xxxx Keywords: Crystalline silicon Pyrolysis Polymers removal Recycling Solar panel
a b s t r a c t Photovoltaic (PV) modules contain both valuable and hazardous materials, which makes its recycling meaningful economically and environmentally. In general, the recycling of PV modules starts with the removal of the polymeric ethylene-vinyl acetate (EVA) resin using pyrolysis, which assists in the recovery of materials such as silicon, copper and silver. The pyrolysis implementation, however, needs improvement given its importance. In this study, the polymers in the PV modules were characterized by Fourier transform infrared spectroscopy (FTIR) and the removal of the EVA resin using pyrolysis has been studied and optimized. The results revealed that 30 min pyrolysis at 500 °C removes >99% of the polymers present in photovoltaic modules. Moreover, the behavior of different particle size milled modules during the pyrolysis process was evaluated. It is shown that polymeric materials tend to remain at a larger particle size and thus, this fraction has the greatest mass loss during pyrolysis. A thermo gravimetric analysis (TGA) performed in all polymeric matter revealed the optimum pyrolysis temperature is around 500 °C. Temperatures above 500 °C continue to degrade matter, but mass loss rate is 6.25 times smaller. This study demonstrates the use of pyrolysis can remove >99% of the polymeric matter from PV modules, which assists the recycling of this hazardous waste and avoids its disposal. Ó 2016 Published by Elsevier Ltd.
1. Introduction Photovoltaic energy is the third most important renewable energy in terms of installed capacity. Different scenarios expect that PV technologies will provide 2.5–25% of the global electricity demand by 2050 (Silva et al., 2014; Zuser and Rechberger, 2011). Tough photovoltaic energy is a renewable and non-pollutant form of energy (Zhang et al., 2008), the module (or panel) itself has a limited lifespan and eventually becomes electronic waste (e-waste) (Doi et al., 2001). According to (Paiano, 2015), the approximate lifespan of PV modules is 25 years and the first of two waste generation periods has started in 2012. This reflects on why the PV waste issue has only been addressed recently. The lifespan is determined by the nominal power: when it drops below 80%, the module is considered to have reached its end of life. As the PV market continues to grow, so will waste, even if such waste appears with a long time delay (Kazmerski, 2006). E-waste consists in dead electronic and electrical equipment and the current e-waste generation pattern poses one of the world’s greatest pollution problem. The main issues are the lack of appropriate recovery ⇑ Corresponding author. E-mail addresses: [email protected] (P. Dias), [email protected] (H. Veit).
technology (Jujun et al., 2014) and the risk of releasing hazardous substances such as cadmium, rare earths and brominated flame retardants (BFRs) if the waste is not disposed correctly (Widmer et al., 2005; Marwede and Reller, 2012). Most often, the discarded electronic goods end-up in landfills along with other municipal waste or are burnt with no gas emission control, releasing toxic and carcinogenic substances into the atmosphere (Dwivedy et al., 2015). Waste of electric and electronic equipment (WEEE) also consists of numerous precious and scarce metals, which makes its recycling meaningful economically and environmentally (Zeng et al., 2004; Ongondo et al., 2011). PV modules are considered a hazardous waste because of the metals they usually contain (Pb, Cr, Cd, Ni) (Tammaro et al., 2016). They are sorted according to their technology (European Commision, 2011): Crystalline silicon; Thin film; Concentrator photovoltaic (CPV) and emerging technologies; According to Tao and Yu (2015), silicon wafer-based PV modules are the most common type of solar cell manufactured in the world. This is based on the inherent advantages of silicon as a
http://dx.doi.org/10.1016/j.wasman.2016.08.036 0956-053X/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
2
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
semiconductor and on the major industry that has been built to manufacture silicon devices (Bruton, 2002). Crystalline silicon PV modules are made from the following materials, in order of mass: glass, aluminum frame, EVA copolymer transparent encapsulating layer, photovoltaic cells, installation box, TedlarÒ protective foil and assembly bolts (Radziemska, 2014). Modules are encapsulated with various materials to protect the cells and the electrical connectors from the environment (Hansen et al., 2000). The removal of these encapsulating materials (being EVA resin the most common) is usually the first step in recycling PV modules. Several methods to remove the EVA can be employed, such as dissolution using nitric acid and thermal decomposition (usually pyrolysis) (Bruton, 1994). The pyrolysis process consists in heating the material to elevated temperatures, in absence of oxygen, to decompose organic volatile matter (gases, rubbers, etc.) to gas and liquids. The inorganic components (metal, fillers, glasses, etc.) remain almost unaltered during the process, and consequently the valuable components can be recovered and reused. The pyrolysis process is especially appropriate for complex wastes such as WEEE streams, which contain many different plastics mixed with other materials (de Marco et al., 2008). The absence of oxygen during pyrolysis avoids the oxidation of some materials and the formation of dibenzo-p-dioxins and dibenzofurans from the flame retardant found in some devices (Zhang et al., 2016). Most PV recycling methods use thermal processes to remove organics, Kang et al. (2012) use a thermal decomposition to remove the adhesive layer and recover the semiconductor. Wang et al. (2012) use a two-step heating process for the thermal delamination of PV module. Fthenakis (2000) discusses the recovery of metals using a smelter in which EVA and other polymers decompose at high temperatures. Berger et al. (2010) aim to recycle thin film PV modules and use a thermal dismantling process at 470 °C
in order to destroy the EVA layer. Dias et al. (2016a) study the destruction of the polymeric fraction in order to recycle modules and state that the use of pyrolysis could minimize environmental impacts by avoiding the use of organic solvents. Radziemska et al. (2010) recycle silicon solar cells using two main stages: separation and cleansing. The separation stage is a thermal delamination used to remove the EVA resin and separate the other layers. McDonald and Pearce (2010) state that the recycling process for any silicon PV module is identical and involves pyrolysis, which assists in recovering crystalline silicon wafer from the waste PV. The use of an inert gas - such as nitrogen - prevents chemical oxidation of the EVA and the silver contact grid, resulting in a clean, almost residue free cell surface (John and Anisimov, 1997). Previous studies, however, show that silver recovery should occur prior to pyrolysis (Dias et al., 2016b). The EVA pyrolysis in inner atmosphere consists on the removal of the acetic acid from the main chain and the degradation of the remaining polyethylene co-polyethylene (Zeng et al., 2004). All of these studies use pyrolysis in their recycling process but Zeng et al. (2004) published the only study that optimizes the pyrolysis itself. Thus, because of its importance for PV recycling, it is necessary to evaluate other aspects of the PV pyrolysis process. In this study, the polymeric materials used in PV panels were characterized and the use of pyrolysis as a separation process is studied, evaluated and optimized in order to assist recycling of waste PV and enable the recovery the valuable materials present in the modules.
2. Materials and methods A schematic diagram has been drawn in order to outline the investigations performed in this study (Fig. 1).
Fig. 1. Schematic diagram illustrating the procedures used in this study.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
Fig. 2. Illustration of silicon module’s components (Dow Corning, 2016)(Dow Corning, 2016).
The experiments were performed with crystalline silicon modules (c-Si modules). The aluminum frames were removed from all modules manually assisted by tools such as screwdriver, plier and small electric saw. The polymeric were also removed manually with the aid of a magnifying glass and a stylet to separate the different layers. A Fourier transform infrared spectroscopy (FTIR) was performed in the encapsulating layer and in the back sheet layer (Fig. 2) using a Spectrum 100 equipment (Perkin Elmer), spectral window of 4000–650 cm 1 and resolution of 4 cm 1. The FTIR was intended to identify the polymers that were used the PV modules. The database used to compare FTIR spectrums using Euclidean distance was Sadtler Inorganic Library. The encapsulating layer is an adhesive material used to bond the glass and the silicon cells. Its removal is the most difficult step in separating the module’s components. Even when the glass layer was removed manually, the adhesive material kept glued to the semiconductor. Thus, the idea of the pyrolysis approach is to remove the adhesive material beforehand so that the other materials can be separated afterwards. The encapsulating layer was scraped manually and the obtained samples were analyzed by thermo gravimetric analysis (TGA) in platinum crucibles using a TGA Q50 (TA Instruments, New Castle, USA). The parameters used in the analysis followed the ASTM E11317 standards (ASTM E1131-08, 2014). The analysis occurred in nitrogen atmosphere, the temperature varied from 20 to
3
500 °C at a 10 °C/min rate. A duplicate sample of the module’s back sheet (Fig. 2) was scraped and analyzed by TGA. Analysis occurred in a nitrogen atmosphere, temperature varied from 20 to 600 °C at a 10 °C/min rate. The modules were milled four times on a SRB 2305 knife mill (Rone, São Paulo, Brasil) based on previous works (Dias et al., 2016b). The mill screen used had a 4 mm opening in the first two times and 2 mm opening in the second two; the output powder weighed approximately 4 kg. As stated by Cui et al. (2003), for maximum separation of materials, WEEE should be shredded to small particles. The powder obtained was separated in three groups, according to the particle size. To separate the powder, a mesh 35 (0.5 mm aperture) and a mesh 18 (1 mm aperture) were used. The sieving occurred in a vibrating sieving device; for every 300 g of material, the equipment vibrated for 15 min with and amplitude of 1 mm. The three obtained fractions (smaller than 0.5 mm, between 1 mm and 0.5 mm and bigger than 1 mm) were stored separately, weighted and named F1, F2 and F3, respectively. The results from the TGA encouraged a study on the procedure parameters for the pyrolysis process. A furnace structure was assembled to control the process, as shown in Fig. 3. The pyrolysis was carried out using 150 30 cm2 alumina boat-shaped crucibles; the crucibles were filled, weighed, placed inside the furnace and removed periodically for weighting (Table 1). The approximate initial weight for each sample was 9.0 g, quartering was used to ensure that the samples were significant. The dwell temperature chosen was 500 °C based on the previous TGA, the heating ramp was 15 °C/min, and 1 L/min nitrogen airflow was maintained throughout the whole process (adapted from Zeng et al., 2004). This pyrolysis was repeated four times (quadruplicate samples) for each particle size fraction (F1, F2 and F3). In the interest of evaluating the influence of the pyrolysis duration and particle size, a variance analysis using the GLM module of SAS software (SAS Inst., Inc. Cary, MC, 1989) was performed. Significance was tested by the F test, and mass loss means compared by LSMean (least squares). The evaluation considered fractions F1, F2 and F3 as well as pyrolysis times 20, 30, 60 and 180 min. In order to verify if pyrolysis was able to remove all organic matter, another TGA was performed in the samples after the pyrolysis. The analysis occurred in nitrogen atmosphere, temperature varied from 20 to 700 °C at a 10 °C/min rate. The parameters used in the analysis followed the ASTM E11317 standards (ASTM E1131-08, 2014).
Fig. 3. Schematic illustration of the furnace setup.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
4
P. Dias et al. / Waste Management xxx (2016) xxx–xxx Table 1 Parameters used in the pyrolysis. Fraction
Initial mass (g)
Temperature (°C)
N2 flow rate (L/min)
Weighing interval (min)
Total pyrolysis time (h)
F1
9
500–550
1
F2
9
500–550
1
F3
9
500–550
1
60 30 20 60 30 20 60 30 20
3 1 1 3 1 1 3 1 1
3. Results Regarding the characterization of the polymers from the modules, visual inspection reveals the back sheet (Fig. 2) is a multilayer structure as illustrated in Fig. 4. The FTIR results show that the encapsulating material is ethylene-vinyl acetate (EVA) (Fig. 5) and that the back sheet layer is composed of polyvinyl fluoride (PVF – commercially known as TedlarÒ) (Fig. 6), Polyethylene terephthalate (PET) (Fig. 7) and EVA (Fig. 5). These results are in accordance with Radziemska (Radziemska, 2014). The adsorption bands at 3000–2800 cm 1 have been attributed to the CAH bond stretching, which presents particularly strong intensity in the
Fig. 4. Illustration of the multilayer back sheet structure.
EVA spectrum due to the long CAH chain (Fig. 5). The EVA also displays the CAO asymmetric stretching vibration band point at 1236 cm 1 and the CH2 rocking vibration at 720 cm 1. The PVF spectrum (Fig. 6) presents the CAH bond stretching as well as the very strong intensity double band CAF bond stretching at 1100–1000 cm 1. Finally, regarding the PET spectrum (Fig. 7), the adsorption bands at 1710 cm 1 has been attributed to the alkanoate ester C@O bond stretching, the 1118 cm 1 to the phthalates asymmetric CAO bond stretching vibration, the 1096 cm 1 to the symmetric CAO bond stretching vibration and the 1243 cm 1 the asymmetric CAOAC bond stretching. The band observed at 723 cm 1 was attributed to the @CAH out-of-plane deformation vibration, which is a result of the para substitution in the aromatic ring found in the adsorption bands 900–675 cm 1. The TGA results from the encapsulating layer are displayed in Fig. 8. The observed behavior is in accordance with Zeng et al. (2004) and Berger et al. (2010) and display the behavior of the EVA polymer when heated. Fig. 8 shows that the entire polymeric layer degrades at 500 °C. The TGA of the milled module sample is displayed in Fig. 9 and shows that approximately 75% (in weight) of the polymeric fraction degrades between 400 °C and 500 °C. Thus, most of the polymeric blends and combinations preset in the PV modules can be degraded reaching 500 °C. A detailed analysis of the substrate’s TGA (Fig. 9) is displayed in Table 2 and shows the average mass loss between 400 °C and 500 °C is 6.25 times greater than the mass loss between 500 °C and 600 °C. These results (Figs. 8, 9 and Table 2) encouraged the use of 500 °C as the temperature dwell for the pyrolysis used in this study.
Fig. 5. FTIR - Peaks found in analysis (above) compared to EVA peaks from database (below).
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
5
Fig. 6. FTIR - Peaks found in analysis (above) compared to PVF peaks from database (below).
Fig. 7. FTIR - Peaks found in analysis (above) compared to PET peaks from database (below).
Fig. 8. TGA results for the encapsulating layer.
Fig. 9. TGA results for the back sheet of the module.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
6
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
Table 2 TGA numerical analysis. Temperature range (°C)
Max mass loss
400–500 500–600
1.48 (at 441 °C) 0.04
%
C
)
Average
%
C
)
0.25 0.04
Table 3 Pyrolysis results for particle size F1, F2 and F3. Pyrolysis total time (h)
F1
F2
F3
Wt%
SD
Wt%
SD
Wt%
SD
t=0 t=1 t=2 t=3
100 93.54 93.74 93.69
– 1.3479 1.8609 4.4227
100 91.544 91.611 91.542
– 3.6177 3.5224 3.6028
100 85.14 84.91 84.91
– 6.378 6.150 6.002
The pyrolysis was carried out for all three particle size fractions (F1, F2, F3). The results show the average weight decrease measured in the experiments (Table 3). It is noticeable that the weight loss increases as particle size increases, this indicates that polymeric fraction tends to concentrate in the larger particle size group: polymer concentration = F3 > F2 > F1. This weight loss behavior is in accordance with Granata et al. (2014). Moreover, the weight decrease from the second hour to the third and forth hour is about 1% or less in all fractions. These results indicate a 1 h dwell at 500 °C results in an efficient pyrolysis for all particle size groups studied and that total pyrolysis time over 1 h does not influence the pyrolysis mass loss. In order to prove this statement, TGAs were performed in samples after pyrolysis for all three fractions. The results are displayed in Figs. 10–12.
Fig. 12. TGA of the pyrolysed sample - fraction F3. Table 4 Particle size and pyrolysis time (duration) influence in mass loss. Fraction
Mass (%)
Error
F1 F2 F3 P
93.55a 90.65b 82.69c <0.001
±0.97 ±0.94 ±0.97
Pyrolysis time (min) 20 30 60 180 P
91.58a 87.5b 88.42b 88.35b 0.08
±1.20 ±1.08 ±1.08 ±1.08
a,b,c Means in the same parameter column with different superscripts are significantly different for given p-value. There were 4 replicates per parameter.
Further investigations were conducted with 30 min and 20 min weighing intervals, the results are displayed in Table 4. The analysis of Table 4 indicates particle size distribution influences significantly in the weight loss (p = 0.001). These results corroborate the hypothesis that the polymers are concentrated in the larger fraction, followed by the intermediate and small fractions where mass loss is smaller. Table 4 also indicates that a 30, 60 and 180 min pyrolysis are not statistically different from each other, thus a pyrolysis lasting 30 min has a mass loss as significant as pyrolysis lasting for 60 or 180 min for p = 0.08. However, pyrolysis 20 min differs statistically from the others. Therefore, based on the previous TGAs (Figs. 10–12) and these statistical results (Table 4), the 30 min pyrolysis is optimum because it is the shortest time that ensures complete pyrolysis regardless of particle size. Fig. 10. TGA of the pyrolysed sample - fraction F1.
4. Conclusions The key conclusions from this study are:
Fig. 11. TGA of the pyrolysed sample - fraction F2.
The possible polymers present in silicon photovoltaic modules are – but are not limited to - ethylene-vinyl acetate (EVA), polyvinyl fluoride (PVF – commercially known as TedlarÒ) and polyethylene terephthalate (PET). Thermo gravimetric analysis reveals 75% of the polymers from PV modules are degraded between the temperatures of 400 and 500 °C. Temperatures above 500 °C continue to degrade matter, but mass loss rate is 6.25 times smaller than before 500 °C. When modules are milled and sieved, polymeric matter concentrates in larger particle size. Polymer size has great influence in mass loss during pyrolysis (p = 0.01). The larger the particle size, the greater the mass loss.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036
P. Dias et al. / Waste Management xxx (2016) xxx–xxx
A pyrolysis performed at 500 °C for 1 h removes >99% of the polymers from PV modules. Statically, pyrolysis performed at 500 °C with a 1 h dwell or with a 30 min dwell do not differ (for a p-value = 0.08). Thus, 30 min dwell at 500 °C are best parameters for the pryrolysis of silicon PV modules. These results are valid for samples of approximately 9 g (bench scale); further investigations should be made to assess the behavior of the process in larger scales. Acknowledgments The authors are grateful to CAPES, CNPq, FINEP and FAPERGS (Brazil) for their financial support. References ASTM E1131-08, 2014. Standard Test Method for Compositional Analysis by Thermogravimetry. ASTM International, West Conshohocken, PA, 2014,
7
Fthenakis, V.M., 2000. End-of-life management and recycling of PV modules. Energy Policy 28, 1051–1058. Granata, G. et al., 2014. Recycling of photovoltaic panels by physical operations. Sol. Energy Mater. Sol. Cells 123, 239–248. Hansen, A.D., et al., 2000. Models for a Stand-Alone PV System. Risø National Laboratory, Roskilde, 2000, pp. Risø-R-1219(EN)/SEC-R-12. John, R.B., Anisimov, I.I., 1997. Possibility of recycling silicon PV modules. In: 26th Photovoltaic Specialists Conference (PVSC), Anaheim, CA, Sept. 30–Oct. 3, 1997, pp. 0-7803-3767-0/97. Jujun, R., Yiminga, Q., Zhenming, X., 2014. Environment-friendly technology for recovering nonferrous metals from e-waste: Eddy current separation. Resour. Conserv. Recycl. 87, 109–116. Kang, Sukmin et al., 2012. Experimental investigations for recycling of silicon and glass from waste photovoltaic modules. Renewable Energy 47 (2012), 152–159. Kazmerski, L.L., 2006. Solar photovoltaics R&D at the tipping point: a 2005 technology overview. J. Electron Spectrosc. Relat. Phen. 150, 105–135. Marwede, M., Reller, A., 2012. Future recycling flows of tellurium from cadmium telluride photovoltaic waste. Resour. Conserv. Recycl. 69, 35–49. McDonald, N.C., Pearce, M.J., 2010. Producer responsibility and recycling solar photovoltaic modules. Energy Policy 38, 7041–7047. Ongondo, F.O., Williams, I.D., Cherrett, T.J., 2011. How are WEEE doing? A global review of the management of electrical and electronic wastes. Waste Manage. 31, 714–730. Paiano, Annarita., 2015. Photovoltaic waste assessment in Italy. Renew. Sustain. Energy Rev. 41, 99–112. Radziemska, E.K., 2014. Recycling of Photovoltaic Solar Cells and Modules - The State-Of-Art. s.l.: Saarbrucken. LAP LAMBERT Academic Publishing, Germany. Radziemska, Ewa Klugmann. et al., 2010. Experimental validation of crystalline silicon solar cells recycling by thermal and chemical methods. Sol. Energy Mater. Sol. Cells 94 (12), 2275–2282. Silva, M.D. et al., 2014. Integration of technologies and alternative sources of water and energy to promote the sustainability of urban landscapes. Resour. Conserv. Recycl. 91, 71–81. Tammaro, Marco. et al., 2016. Experimental investigation to evaluate the potential environmental hazards of photovoltaic panels. J. Hazard. Mater. 306, 395–405. Tao, J., Yu, S., 2015. Review on feasible recycling pathways and technologies of solar photovoltaic modules. Sol. Energy Mater. Sol. Cells 141, 108–124. Wang, T.-Y., Hsiao, J.-C., Du, C.-H., 2012. Recycling of materials from silicon base solar cell module. In: Proceedings of the 38th IEEE Photovoltaic Specialists Conference, pp. 2355–2358. Widmer, R. et al., 2005. Global perspectives on e-waste. Environ. Impact Assess. Rev. 25, 436–458. Zeng, D., Born, M., Wambach, K., 2004. Pyrolysis of EVA and its application in recycling of photovoltaics. J. Environ. Sci. 16 (6), 889–893. Zhang, Lifeng, Ciftja, Arjan., 2008. Recycling of solar cell silicon scraps through filtration, Part I: Experimental investigation. Sol. Energy Mater. Sol. Cells 92 (11), 1450–1461. Zhang, Mengmei, Buekens, Alfons, Li, Xiaodong, 2016. Brominated flame retardants and the formation of dioxins and furans in fires and combustion. J. Hazard. Mater. 304, 26–39. Zuser, A., Rechberger, H., 2011. Considerations of resource availability in technology development strategies: the case study of photovoltaics. Resour. Conserv. Recycl. 56 (1), 56–65.
Please cite this article in press as: Dias, P., et al. Recycling WEEE: Polymer characterization and pyrolysis study for waste of crystalline silicon photovoltaic modules. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.08.036