Solar Energy 177 (2019) 746–761

Contents lists available at ScienceDirect

Solar Energy
journal homepage: www.elsevier.com/locate/solener

Review

Recycling of end of life photovoltaic panels: A chemical prospective on T

process development
⁎
Flavia C.S.M. Padoan , Pietro Altimari, Francesca Pagnanelli
Department of Chemistry, Sapienza University of Rome, Piazzale Aldo Moro 5, 00185 Rome, Italy

A R T I C LE I N FO A B S T R A C T

Keywords: The application of photovoltaics has been rapidly increasing over the past two decades driven by the idea that it
Recycling could provide a fundamental contribution to the transition from traditional fossil fuels to renewable energy
Photovoltaic panel based economies. However, long-term sustainability of photovoltaics will be largely dependent on the effec-
End-of-life panel tiveness of the process solutions that will be adopted to recycle the unprecedented volume of end-of-life panels
Photovoltaic waste
expected to be generated in the near future. Recycling is indispensible to avoid the loss of the valuable materials
employed to produce the photovoltaic panels and, at the same time, prevent that harmful elements, including,
for example, heavy metals, could be dispersed into the environment through improper disposal practices. In this
article, the process solutions proposed over the past two decades to recycle photovoltaic panels are critically
reviewed. Main objective is to provide the basis for the identification of the recycling solutions that can effec-
tively sustain the continuous increase of the photovoltaic market. In order to assess the requirements that should
be satisfied by the recycling processes, the legislation currently in force to regulate the management of end-of-
life photovoltaic panels is reviewed, and the evolution of the PV market over the past two decades is analysed.
Based on this analysis, forecasts are derived for the flux of end-of-life panels that will be generated over the
coming four decades. A technical survey of the previously proposed recycling processes is successively per-
formed by including, in addition to the analysis of the research studies published in scientific articles, a detailed
review of the patented recycling processes. Indications are given to which may be the most promising processes
in terms of their economic sustainability and environmental impact.

1. Introduction
Photovoltaic (PV) configures a competitive technological alter-
native to convert solar radiation into electricity, and it is expected to
provide a fundamental contribution to the transition from traditional
fossil fuels to renewable energy based economies. This technology has
been applied since the 1980s, with the first appreciable photovoltaic
power dating back to the early 1990s (Tao and Yu, 2015).
Over the past decades, the economic sustainability of photovoltaics
has been the subject of numerous studies. Major issue addressed in this
context has been the incidence of the raw material costs on photovoltaic
panel production (Desideri et al., 2012; Gerbinet et al., 2014; Kim et al.,
2014; Peng et al., 2013).
Si-crystalline (mono or poly) panels represent the most consolidated
PV technology and have dominated the PV market over the past dec-
ades. In order to reduce the panel production costs, alternative photo-
voltaic technologies have been developed, the main ones including Si-
amorphous, CdTe and CIGS thin film, organic and hybrid cells.
However, Si-crystalline (mono or poly) panels still represent the tech-
nology that guarantees the highest return on investment (SolarPower
Europe, 2017). Main limitations to the extensive application of PV
technologies, alternatives to Si-crystalline panels, are the use of toxic
elements (Cd in CdTe) and/or rare-critical elements (In, Ga in CIGS and
CIGS). As a consequence, 51% of the PV market share is covered by Si-
polycristalline, 41% by Si-mono, 5% by CdTe and 2% by CIGS (IRENA,
2016).
The average lifetime of a PV panel is, irrespective of the considered
technology, around 25 years (Paiano, 2015). Since the electric power
share from PV installations became relevant starting from the end of
nineties, a dramatic increase in the annual flux of end-of-life PV panels
can be expected around 2025. Motivated by this forecast analysis,
several recycling processes, whose analysis is the main objective of the
present review, have been proposed to recycle PV panels. Main target of
the proposed processes is to meet the material recovery targets set by
the relevant regulation (EU, 2012), ensuring, at the same time, eco-
nomic and environmental sustainability. However, while the proposed

⁎
Corresponding author.
E-mail address: flaviacarla.padoan@uniroma1.it (F.C.S.M. Padoan).

https://doi.org/10.1016/j.solener.2018.12.003
Received 31 July 2018; Received in revised form 15 November 2018; Accepted 1 December 2018
Available online 07 December 2018
0038-092X/ © 2018 Elsevier Ltd. All rights reserved.
F.C.S.M. Padoan et al.
recycling processes appear to satisfactorily meet the material recovery
target fixed by regulation, insufficient attention has been paid to
evaluate whether they fulfil the fundamental pre-requisites of economic
and environmental sustainability.
Economic sustainability is largely dependent on the fraction and the
quality of materials that can be recovered by the proposed recycling
processes. Higher is the value of the recovered materials, closer is the
proposed recycling process to satisfy the requirement of economic
sustainability. From this perspective, definitively assessing the eco-
nomic sustainability of the proposed recycling processes is however a
complex task owing the modifications of the PV panel compositions that
are continuously and rapidly introduced even for the same PV tech-
nology. Variations in the production techniques have led, for example,
to a progressive decrease in the PV panel metal content, which ulti-
mately caused a reduction into the value of recovered materials. A re-
presentative example is Ag, whose content in Si panels was found to be
reduced from 0.14–0.2% on 2003 to 0.07–0.16% on 2023 (Peeters
et al., 2017). Accordingly, estimating the economic feasibility of re-
cycling processes that are based on the recovery of metals is difficult
considering the intrinsic and temporal heterogeneity of the panels.
Similarly, variations in the cost and availability of the primary re-
sources needed to produce PV panels can dramatically influence the
economic feasibility of the proposed recycling processes.
Overproduction of Si resulted, for example, into a reduction in the cost
of Si Watt (3 USD in 2010 to 0.5 USD in 2017 in Europe (IRENA,
2017)), implying that consolidated recycling processes originally de-
veloped to ensure the recovery of the Si cell are now useless.
Many works on PV panel recycling (60% of papers cited in this re-
view) were focused on the treatment of Si-panels (Doni and Dughiero,
2012; Kang et al., 2012; Kim and Lee, 2012; Huang et al., 2017; Shin
et al., 2017), whereas fewer studies presented the development of re-
cycling processes allowing for the treatment of different panels tech-
nologies (Pagnanelli et al., 2016, 2017).
A survey of these recycling processes can be found in several re-
views previously reported in the scientific literature. However, these
reviews were mainly focused on quantifying the environmental impact
(Giacchetta et al., 2016; Latunussa et al., 2016; Malandrino et al., 2017;
Raugei et al., 2012; Rocchetti and Beolchini, 2015; Tammaro et al.,
2015) and detailing the quite general economic-financial aspects
(Corcelli et al., 2017; Müller et al., 2005) of PV panel recycling, without
critically comparing the different proposed recycling processes. Parti-
cularly, insufficient attention has been devoted to separately evaluate
and compare the proposed recycling processes in terms of their eco-
nomic profitability and environmental impact. This imposes to thor-
oughly analyse, for any process, the sequence of physical and chemical
operations carried on, the reagent and the energy consumption, and the
recovery rates of the recycled material. The derivation of these data and
the analysis of the PV market over the past two decades are crucial to
identify the process routes that can provide a sustainable solution to the
treatment of PV panels in the near future.
In this article, we present a review of the processes proposed over
the past two decades to recycle PV panels is presented. In addition to
the analysis of the research studies published in scientific articles, a
detailed analysis of the patented recycling processes is presented. In
order to identify the requirements that recycling processes must satisfy
(material recovery rates, types of panels that should be treated) the
current legislation regulating the management of PV panels and the
evolution of the PV market over the past two decades are preliminarily
analysed. Proposed recycling processes are finally compared in terms of
their economic sustainability and environmental impact.
2. Photovoltaic panels composition
Photovoltaic cells (or solar cells) are devices converting the light
energy from any source into electrical energy. In the photovoltaic
panel, organic and inorganic components are combined. Through the
747
Solar Energy 177 (2019) 746–761
Frame
Glass
EVA
Solar Cell
EVA
Backsheet
Junction box
Fig. 1. General structure of a photovoltaic panel.
sketch presented in Fig. 1, the different components of a photovoltaic
panel can be recognized. Starting from the bottom, we find the plastic
backing (or back-sheet), which is the white surface characteristic of
photovoltaic modules, composed by Tedlar (polyvinyl fluoride) and
polyethylene terephthalate. Going up we find two sheets of ethylene
vinyl acetate (EVA) incorporating the semiconductor, which can vary
depending on the type of panel. Finally, a glass layer is found placed on
top. All these layers are sandwiched within an aluminium frame to
which the junction box is connected. In some cases, an antireflective is
added over the glass surface.
Among the different classifications previously proposed for photo-
voltaic modules, we here follow the classification by generations, which
distinguish three different groups of PV modules: first, second and third
generation. The first generation includes the most popular PV panels,
i.e., the monocrystalline or polycrystalline silicon panels which cover
around 41% and 51% of the market share, respectively (IRENA, 2016).
In the monocrystalline structure, where high homogeneity is ensured by
the laborious production process known as the Czocharalski method
(Bagnall and Boreland, 2008), the atoms are oriented along a single
direction, whereas, in the polycrystalline structure, the atoms are ag-
gregated into small monocrystalline grains arranged randomly. In the
polycrystalline panels, the silicon has a lower degree of purity, which is
responsible for a decrease in the efficiency (19%) as compared to the
monocrystalline (21%) (IRENA, 2016). The most targeted metal in the
recycling process of this panel type is silver, which is present on the
front and back side of the solar cell as electrode (Nieland et al., 2012;
Tao and Yu, 2015; Dias et al., 2016).
Second generation modules are thin films deposited on a substrate
(glass, stainless steel, etc.) and can be made of one or more layers.
Different materials can be used to produce the thin film, including
amorphous silicon (a-Si), cadmium telluride (CdTe) or CIGS (copper
indium gallium selenide).
Because of the low efficiency (10%), a-Si modules, making use of a
non-crystalline form of silicon, have been produced discontinuously
over the past few years, and are currently characterized by a negligible
market share. Therefore, no special waste treatment will be probably
needed in the future for such technology. The two thin-film solar cell
technologies, i.e. CdTe and CIGS, cover 7% of the photovoltaic market
(5% of CdTe panels and 2% of CIGS panels (IRENA, 2016)). In the CdTe
panels, the semiconductor is formed by cadmium and tellurium, and its
efficiency increased from 15.4% in 2015 to 17% in 2016, which is
comparable to that attained with poly-crystalline PV efficiencies. Major
limitation to the extensive application of CdTe panels is however re-
presented by the toxicity of cadmium and the reduced tellurium
availability (IRENA, 2016; Paiano, 2015). Innovative CIGS panels seem
to be very promising since they achieve efficiency of 15%. They have
cells containing Cu(In,Ga)(S,Se)2 deposited in a metal back contact on a
metal substrate, which provides the semiconductor capacity due to the
F.C.S.M. Padoan et al. Solar Energy 177 (2019) 746–761

indium, gallium and selenium metals. On the other hand, this tech-
nology requires higher manufacturing costs due to more complex pro-
cesses.
The third generation includes emerging technologies such as con-
centrator photovoltaics (CPV), dye-sensitized solar cells, organic cells
and hybrid cells. CPV generates electricity from sunlight and relies on
the same concept of the traditional solar modules to generate elec-
tricity, except that they use optics lenses and curved mirrors to con-
centrate the sun onto the solar cells. However, CPV modules must ac-
curately face the sun and are, therefore, used in conjunction with
trackers that follow the sun during the day (Bagher et al., 2015; IEA,
2014). Dye-sensitized solar cells includes a photo-sensitized anode
(fluorine doped tin oxide) based on a semiconductor material (com-
posed mostly of TiO2) whereon the dye molecules (typically a bipyr-
idine metal complex) are covalently bound, an electrolyte (usually an
iodide/triiodide redox-active couple dissolved in an organic solvent)
and a photo-electrochemical cathode (fluorine doped tin oxide coated
with a layer of Pt) (Grätzel, 2000; Wan, 2004). Organic cells are gen-
erally composed of polymeric materials or small organic molecules.
Finally, hybrid cells are a combination of cells that that could be or-
ganic and inorganic cells or crystalline silicon and amorphous silicon
cells (Paiano, 2015; Wu et al., 2005).
It is important to notice that, the structure of the PV panels is always
the same in the various generations and what changes is basically the
chemistry of the photosensitive component, which constitutes, in terms
of panel weight, a percentage around 3.5% (BioIS, 2011; Dubey et al.,
2013).
3. Metal content in photovoltaic panels
A literature analysis was performed to determine the metal content
of PV panels. Data extracted from twelve articles available in the

1200 150000
1100 100000
1000
10000
900
Metal content (mg//Kg)

Metal content (mg/kg)

800
8000
700
600 6000
500
400 4000
300
200 2000
100
0 0
Pb Ag Sn Zn Fe Ti Cu Al

5800 110000
5700
5600 100000
90000
1750
80000
Metal content (mg/kg)

Metal contet (mg/kg)

1500
70000
1250 60000
1000 50000

750 40000
30000
500
20000
250
10000
0 0
Pb Cu Zn Cd Te Fe Al

500 7500

400 6000
Metal content (mg/kg)

Metal content (mg/kg)

300 4500

200 3000

100 1500

0 0
Zn Cd Se Mo Pb Ga In Cu

Fig. 2. Metals content (minimum, first quartile, average, median, third quartile and maximum) obtained from available data in the literature for Si (yellow bars),
CdTe (blue bars) and CIGS (pink bars). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

748
F.C.S.M. Padoan et al.
literature are listed in Table1A (Supplementary Material) and are here
shown in the form of a boxplot in Fig. 2 (Fthenakis and Wang, 2006;
Sander et al., 2007; McDonald and Pearce, 2010; Choi and Fthenakis,
2014; Paiano, 2015; Dias et al., 2016; Jung et al., 2016; Latunussa et al.,
2016; Pagnanelli et al., 2016; Steeghs and Water, 2016; Pagnanelli
et al., 2017; Savvilotidou et al., 2017).
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.solener.2018.12.003.
The metal contents reported by the selected studies are compared.
Difficulties were found to evaluate whether data extracted from dif-
ferent works could be compared, mainly because different panel com-
ponents were frequently used to quantify the panel metal content
(Tammaro et al., 2016). Ramos-Ruiz et al. (2017), for example, only
considered the thin film to quantify the metal content of CdTe panels,
while Yi et al. (2014) quantified the metal content of Si-panels by
characterization of the panel wafer alone.
As a matter of fact, by analysing the data in Fig. 2, a great dis-
crepancy between the metal contents reported by the cited works was
found. This can be justified by the fact that each author considers one or
another fraction of the panel, such as aluminium frame, metal contacts
and junction box. Such fractions may or may not be included in the
characterized material. In addition, it is known that the photovoltaic
market is extremely heterogeneous, and the evolution of the PV tech-
nology moves towards decreasing the panel metal content. For ex-
ample, historical data show an exponential drop in the silver content
(IRENA, 2016; Peeters et al., 2017). Accordingly, the quantified metal
content is expected to be higher for older panels, which makes the year
of manufacture a relevant data to be considered. This data is however
missing in most of the articles.
4. Photovoltaic market
The global market for photovoltaic panels has been continuously
increasing over the past decades, reaching recently an important
milestone. According to a report published in 2017 by the SolarPower
Europe, the market for PV grew by around 30% in 2015 and 33% in
2016. During 2016, global solar power capacity exceeded 300 GW,
arriving at 306.5 GW by the end of that year, after it reached 200 GW in
2015 and the 100 GW in 2012 (Fig. 3). In addition, according to the
report, the world’s cumulative solar capacity increased by more than
1500% over just six years – starting from 21.6 GW attained in 2010.
In summary, due to the recent increase in cell efficiency and
800
700
600
500
400
300
200
30%
26%
40%
100 44%
78%
0
2010 2011 2012 2013 2014
Fig. 3. Global installed capacity of photovoltaic solar energy in GW until 2016 (blue bars) and future projections for the next years (gray bars). Data taken from
SolarPower Europe (2017). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
749
Solar Energy 177 (2019) 746–761
decrease in production costs, PV technology became competitive with
that of electricity from fossil fuel, attracting an increased interest in
many countries. The growth in the 2016 production of photovoltaic is
evidenced in Fig. 4. Asia-Pacific (APAC) has become the largest solar-
powered region in the world (147.2 GW), and the European solar pio-
neers are now ranked second (104.3 GW), America third (45.9 GW),
Middle Est (MEA) holds only 4.7 GW and the rest of the world (RoW)
1.4%.
China had a total of 77.9 GW installed PV, representing one quarter
of the entire solar power generation capacity. Japan was the world’s
second largest solar power market in 2016, owing 42.9 GW of capacity,
which corresponds to 14% of the global market. The United States was
the world’s third largest solar market in 2016, reaching 42.4 GW, equal
to 13.8%. Germany ranked fourth, with 41.1 GW and 13.4% share, not
forgetting Italy with 18.9 GW and the UK with 11.5 GW.
5. Photovoltaic panel recycling legislation
Because of the increase in the photovoltaic panels market previously
evidenced, it has become necessary to create the legislation regulating
the management of PV panel waste flow that will be generated in the
near future. For this purpose, several directives have been formulated.
In February 2003, the first Waste Electrical and Electronic
Equipment Directive entered into force. The term electrical and elec-
tronic equipment (EEE) was introduced to define equipment dependent
on electric currents or electromagnetic fields in order to work properly,
and equipment for the generation, transfer and measurement of such
currents and designed for use with a voltage rating not exceeding
1000 V for alternating current and 1500 V for direct current (EU, 2012).
The objective of this legislation was to increase the recycling and re-use
of such products. In 2008, the European Commission proposed a revi-
sion of the Directive, and in August 2012 the new 2012/19/EU Direc-
tive entered into force. In this latter Directive, photovoltaic panels were
defined as EEE, and consequently end-of-life photovoltaic panels were
defined as WEEE.
The European Directive is focused on the collection, reuse, recycling
and other forms of recovery of such wastes. It also aims to regulate end-
of-life equipment treatment and require manufacturers to fulfil the
obligations in order to satisfy waste management policy. Specific goals
of collecting, recovery and recycling that must be achieved in the next
years are described in the Annex V. From 15 August 2015 until 14
August 2018, photovoltaic panels, which are included in category 4 of
33%
2015 2016 2017 2018 2019 2020 2021
F.C.S.M. Padoan et al.
1.5%
15%
34%
Fig. 4. Cumulative installed photovoltaic capacity by regions and their respective countries for the year of 2016. Data taken from SolarPower Europe (2017).
Annex 1, shall be recovered and be prepared for re-use and recycled
according to a percentage of 80% and 70%, respectively. These targets
are subjected to an increase from 15 August 2018, becoming 85% for
recovering and 80% for re-use and recycle.
In addition, there are other relevant references not directly linked to
photovoltaic panels recycling, but they also influence PV panels re-
cycling by legislating about the amount of harmful substances in EEE.
These are, for example, the Directive on the restriction of the use of
certain hazardous substances (RoHS Directive 2002/95/EC lately re-
vised to Directive 2011/65/EU) that requires the substitution of heavy
metals and flame retardants by safer alternatives. China has also im-
plemented a RoHS directive based on the one created by the EU. Fur-
ther, the Waste Framework Directive 2008/98/EC was issued to en-
courage sustainable management and waste prevention, by laying down
some principles as end-of-waste criteria and waste hierarchy.
Lastly, and in line with these regulations, in July 2007, the so-called
PV CYCLE was created by the European PV industry, which aims to
“implement the commitment of the photovoltaic industry to create a
voluntary program for the collection and recycling of end of life mod-
ules and take responsibility for photovoltaic modules throughout their
entire value chain cycle”.
6. Photovoltaic waste projections
Due to the increase in photovoltaic panel installations, it is clear
that a large amount of waste from these modules will be generated.
Thus, an attempt has been made to calculate when and how many tons
of waste will be produced. Two different approaches were implemented
to predict the future streams of end-of-life PV panels. In the first ap-
proach, it was assumed that the amount of end-of-life panels generated
during any prescribed year t is equal to the amount of panels installed
during the year t-tP, tp denoting the average panel lifetime, which gives:
F (t ) = g (t − tp)
where F(t) and g(t) denote the amount (ton/y) of PV modules installed
and dismissed during year t, respectively.
In the second approach, a lifetime distribution was introduced to
describe the possibility that panels are dismissed earlier or later than
the average lifetime. The amount of end-of-life panels generated during
any prescribed year t was thus computed as follows:
t centrator photovoltaics (CPV). The evolution of technology market
F (t ) = ∑a=0 p(a)g(t − a)
where p(t) denotes the fraction of panels with age t reaching the end of
life. This fraction was computed as follows:
Solar Energy 177 (2019) 746–761
1.4%
48%
t+1
p (t ) = ∫t f (t ) dt (3)
f(t) denoting the panel lifetime density probability distribution. The
main issue becomes with this approach to determine a reliable form for
the density probability distribution f(t). Simulations were performed in
the present study by application of the following Weibull distribution:
α
α−1
α t −⎛ t ⎞
f (t ) = ⎜⎛ ⎟⎞ e ⎝β⎠
β ⎝β⎠ (4)
This distribution was proposed by Marwede and Reller (2012) to
describe the panel lifetime probability distribution of cadmium-tell-
uride (CdTe) panels. Different α and β values were selected in the
present study to describe the following scenarios: breakthrough
(α = 12.8 and β = 32.2) and slow progress (α = 8.2 and β = 28.5).
These scenarios correspond to different percentages of the installed PV
modules that are returned before the product warranty time (usually
5 years). Particularly, 0.25% and 4% warranty returns are imposed with
the breakthrough and slow progress scenario, respectively. In any sce-
nario, it is assumed that 99.9% of the modules are dismissed within
40 years.
The two approaches illustrated above (Eqs. (1) and (2)) allow pre-
dicting the evolution of the amount of end-of-life panels if the evolution
of the amount of previously installed panels is known. Data about the
evolution of the European and the Italian PV power (MW/year) yearly
installed over the period 2000–2020 were extracted from reports pub-
lished by the European Photovoltaic Industry Association (EPIA) and
the Italian Energy Management Institution “Gestore Servizi Energetici”
(GSE), respectively (EPIA, 2014; GSE, 2016). For the period
2015–2020, the evolution of the European yearly installed power pre-
dicted by EPIA was adopted (EPIA, 2014). This latter prediction was
exploited to compute the evolution of Italian PV power installed be-
tween 2015 and 2020. For this purpose, it was assumed that growth
rates of the Italian and European PV installed power were identical over
(1) the period 2015–2020. The amount of yearly installed panels (ton/year)
could thus be derived by the following two-step procedure:
– The yearly installed power was computed for any PV technology
multiplying the overall yearly installed power by the technology
market share. The following main PV technologies were considered:
crystalline silicon (c-Si), amorphous silicon (a-Si), cadmium tell-
uride (CdTe), copper indium gallium di-selenide (CIGS) and con-
(2) share in the EU area and in Italy was extracted EPIA (2014) and
Paiano (2015), respectively.
– The amount of yearly installed panels was computed for any PV
750
F.C.S.M. Padoan et al. Solar Energy 177 (2019) 746–761

3000 1200
(a) (b)
2500 c-Si 1000
PV waste*1000 (ton)
a-Si
2000 CdTe 800
CIGS
1500 CVP 600
TOT
1000 400

500 200

0 0
2025 2030 2035 2040 2045 2025 2030 2035 2040 2045
time (year) time (year)
Fig. 5. Evolution of the PV waste streams in Europe (a) and Italy (b) predicted by Eq. (1).

technology multiplying the computed installed power by the tech-
nology weight-to-power ratio (ton/MW). The values of the weight-
to-power ratio were extracted for the different PV technologies from
Sander et al. (2007). We notice that these latter values are re-
presentative of panels produced earlier than 2007, whereas weight-
to-power ratios have been progressively decreasing over the fol-
lowing period owing to the achievement of increased panel effi-
ciencies, which is confirmed by the report presented by IRENA and
IEA agencies (IRENA (2016)). Taking into account that most of the
panels to be treated over the period that was selected to elaborate
the present analysis have been produced before 2010 and that the
variation in the weight-to-power ratio over the past ten years was
lower than about 30% (IRENA, 2016), the weight-to-power ratios
reported by Sanders (2007) were here adopted.
Fig. 5 displays the evolution of the PV waste streams predicted by
Eq. (1) in the assumption of constant panel lifetime equal to 25 years.
The evolutions of the PV waste generated in Italy and in the EU area
exhibit a sharp peak around the year 2036. To understand the origin of
such peak, it must be remarked that, in accordance with Eq. (1), the
amount of end-of-life panels dismissed during any prescribed year t is
equal to the amount of panel installed during the year t-25. The sharply
increasing amount of PV waste around the year 2036 can thus be at-
tributed to the rapid growth experienced by the PV market between
2008 and 2011. This growth was determined by the implementation of
policies supporting the application of photovoltaic technology in few
leading countries as, for example, Italy and Germany. The progressive
arrest in the implementation of such supporting policies caused the
amount of yearly installed panels to decrease following 2011. This
latter effect is responsible for the decreasing PV waste stream predicted
following 2036 in the EU area and in Italy (Fig. 6). The reduction in the
PV market size experienced following 2011 was however more pro-
nounced in Italy compared to other EU member states including, for
example, Germany. Particularly, while a progressive recovery in the PV

1800 300
1600 (a) (c)
c-Si 250
PV waste*1000 (ton)

1400
a-Si
1200 CdTe 200
1000 CIGS
150
800 CVP
TOT 100
600
400 50
200
0
0

1600 250
1400 (b) (d)
200
PV waste*1000 (ton)

1200
1000 150
800
100
600
400 50
200
0
0
2030 2040 2050 2060 2030 2040 2050 2060
time (year) time (year)
Fig. 6. Evolution of the PV waste streams in Europe (a, b) and Italy (c, d) predicted by Eq. (1) in the breakthrough (a, c) and slow-progress scenario (b, d).

751
F.C.S.M. Padoan et al. Solar Energy 177 (2019) 746–761

Table 1
Literature recycling processes specified by type of treatment and panel treated and expected obtained products.
Treatment Panel type Authors Year Products

Physical Si Doi et al. 2001 Si cell

Kim and Lee 2012 Si cell
Thin film Berger et al. 2010 Glass and enriched In and Te products for new modules
All types Granata et al. 2014 Glass, Al frame and fine fraction (to be chemically treated)

Thermal Si Frisson et al. 2000 Si cell

Doni and Dughiero 2012 Glass and Si cell
Wang et al. 2012 Glass, cell chips and ribbons (to be chemically treated)

Chemical Si Klugmann-Radziemska and Ostrowski 2010 Si cell

Palitzsch and Loser 2011 Silver and poly-aluminum-hydroxide-chloride
Kuroiwa et al. 2014 Germanium solution
Yi et al. 2014 Si 99.98% pure
Dias et al. 2016 AgCl (94%)
Jung et al. 2016 Glass, Al frame, AgCl (90%), Cu (79%), Pb(OH)2 (93%)
Shin et al. 2017 Si cell
CdTe Fthenakis and Wang 2006 Cd (99.99%), Te
CIGS SENSE 2008 Na2SO3, gallium oxide and In and Mo precipitate mixture
Gustafsson et al. 2014 Selenium oxide (99.99% pure)
Gustafsson et al. 2015 Ga (97%), In (93%)

More than one type combined CdTe Sasala et al. 1996 Glass, Te sludge and Cd
CIGS Kushiya 2003 Al frame, Glass
Si Kang et al. 2012 Si cell
Huang et al. 2017 Glass, Al frame, junction box, Si (90%), Ag (74%), Cu (83%)
All types Pagnanelli et al. 2017 Al frame, glass, fine metal fraction

market size was observed in these latter member states, a barely dis-
tinguishing growth was found in PV Italian market following 2011. This
explains why a progressive growth in the PV waste streams is expected
in the EU area (Fig. 5a) following 2036 while no appreciable growth is
found in Italy (Fig. 5b).
The analysis illustrated in Fig. 5 reveals that the PV waste streams
generated up to 2045 will be mainly constituted by c-Si panels. This
effect is particularly pronounced in Italy with the curves describing the
evolutions of total PV and c-Si waste streams that can be barely dis-
tinguished (Fig. 5b). This element suggests that the fundamental pre-
requisite of any recycling process should be in the medium term the
ability to process c-Si panels.
Fig. 6 displays the evolution of the PV waste streams predicted by
implementation of Eq. (2) in the assumption of breakthrough (Fig. 6a,
c) and slow progress scenario (Fig. 6b, d).
It is apparent that the introduction of the lifetime probability dis-
tribution Eq. (4) significantly modifies the evolution of the PV waste
streams. A first remarkable effect is that, compared to Fig. 5, the
maximum PV waste stream is considerably reduced (by more than 50%)
and shifted, irrespective of the selected scenario (Table 1), to larger
time values. Particularly, the maximum waste stream is predicted to
occur between 2040 (slow progress scenario) and 2045 (breakthrough
scenario), which correspond to a delay ranging between about 5 and
10 years compared to Fig. 6. No significant difference was found be-
tween the maximum waste streams generated in the slow progress and
breakthrough scenario.
7. Literature recycling processes
Several studies have appeared over the past two decades reporting
the development of processes to recycle photovoltaic panels. In what
follows, a review of these studies is presented. A list of the relevant
articles that have been examined to complete the review is reported in
Table 1. Three different process approaches to PV panel recycling are
distinguished and detailed in the remainder of the section: physical
treatment and EVA dissolution with organic solvents, thermal treat-
ment, and chemical processes. Processes relying on the combined ap-
plication of these process approaches are separately discussed.
7.1. Physical treatment
Physical processes were considered in this work as mechanical
treatments like crushing and attrition, and EVA dissolution by the
contact with organic solvents. Physical treatment of silicon panels was
proposed by Doi et al. (2001). In this study, silicon recovery by using
trichloroethylene was proposed to dissolve EVA. With this process, si-
licon cell without any damage could be recovered by using the solvent
at 80 °C for 10 days. Kim and Lee (2012) analysed the dissolution of
EVA in several organic solvents combined with ultrasonic irradiation.
All the employed solvents except O-dichlorobenzene caused the swel-
ling of EVA during the process causing some cracks on the surface of the
PV cells. Berger et al. (2010) investigated the recycling of thin film PV
panels (CdTe and CIS) by using wet mechanical processes including
attrition and flotation, as well as dry mechanical methods like vacuum
blasting. Granata et al. (2014) investigated the recycling of poly-
crystalline silicon panels, amorphous silicon and CdTe photovoltaic
panels by two alternative sequences of physical operations: two blade
rotors crushing followed by thermal treatment and two blade rotors
crushing followed by hammer crushing. Size distribution, X-ray dif-
fraction and X-ray fluorescence analyses of the obtained products were
carried out to characterize the products. Results showed that, for all
kinds of the investigated photovoltaic modules, the two blade rotors
crushing followed by hammer crushing and then by thermal treatment
was the best option to recover glass.
According to Zhang et al. (2013), in China, end of life PV modules
are separated and recycled by abrasive machining performed under
cryogenic conditions and by electrostatic separation. It is reported that
the silicon recovered by such process cannot be employed to produce
new wafers because characterized by insufficient purity. The author’s
conclusion is that, as compared to the advanced PV recycling tech-
nology adopted in Europe, PV recycling technology in China should be
improved to handle the PV waste flow that will be generated in the
future.
Physical treatments relying on crushing and attrition have the ad-
vantage of being inexpensive and allowing for direct recovery of glass,
but they cannot attain the recovery of high value materials (e.g. high
value metals), which requires more elaborate treatments. In addition,
crushing and attrition produce a large amount of dust and create noise
pollution (Xu et al., 2018). The physical processes that involve the

752
F.C.S.M. Padoan et al.
dissolution of EVA in organic solvents typically require long processing
times, though faster dissolution of EVA can be attained by optimized
processes including, for example, the application of ultrasonic probes
(Doi et al., 2001; Kim and Lee, 2012). In these processes, if the solvent
is not reused, a large amount of volatile organic liquid waste is gen-
erated, which is difficult to treat (Xu et al., 2018). In addition, solvent
treatment, in some cases, makes it impossible to attain the integral
recovery of the Si wafer because the swelling of EVA causes some cracks
on the wafer surface (Kim and Lee, 2012).
7.2. Thermal treatment
An effective thermal treatment is represented by delamination. This
uses high temperatures to modify the characteristics of the solar mod-
ules and decompose EVA, which allows for the subsequent mechanical
separation of clean glass and silicon solar cells (Doni and Dughiero,
2012). An alternative thermal treatment was proposed by Frisson et al.
(2000), where the whole panel is introduced in a conveyer belt furnace
and EVA is decomposed under nitrogen atmosphere at temperatures
around 450 °C. The solar cells recovered by the process lose their effi-
ciency after the treatment and it is necessary to perform a cleaning step
before their reutilization. Wang et al. (2012) used a thermal treatment
to recycle the materials from silicon based solar modules. Two heating
steps were performed: a first step at 330 °C to separate Tedlar from the
module, and second step at 400 °C to burn the EVA and thus recover the
glass plate, the cell chips and the ribbons. Doni and Dughiero (2012)
proposed an electro-thermal process that heats the core of Si panel
samples at temperatures lower than the decomposition temperature of
EVA, which allows easily removing the glass fragments. The latter
fragments can be directly supplied to glass recycling facilities, whereas
the inner part of the samples can be further treated for Si recovery.
Thermal treatments are very effective in removing EVA, which
allow recycling up to 90% of PV module components (Nieland et al.,
2012). However, thermal processes are characterized by high energy
consumption, which makes questionable the achievement of the eco-
nomic and environmental sustainability (Tao and Yu, 2015; Xu et al.,
2018). A possible approach to enhance the economic and environ-
mental sustainability of processes relying on the thermal degradation of
the polymeric PV panel fraction would be the energy recovery by
combustion of thermal degradation products. Remarkably, this process
route has been almost ignored by previous studies analysing the sus-
tainability of PV panel recycling processes. Marwede et al. (2013), for
example, conclude that PV panel pyrolysis is a highly energy de-
manding process and imposes the recourse to expensive gas treatment
technologies. This is confirmed not only by the available scientific lit-
erature but also by the analysis of commercial processes. In the
Deutsche Solar process, for example, pyrolysis is performed but the
produced gases are condensed and no energy recovery is foreseen
(Müller et al., 2005). Only one article on PV panel recycling could be
found considering energy recovery by combustion of the pyrolysis
products (Frisson et al., 2000). The scarce attention paid to the latter
energy recovery strategy may be explained by the formation of fluorine
gas during thermal treatment from the back-sheet, which introduces
severe technological difficulties to design and operate the post-com-
bustion section. In this sense, processes that consume low energy, for
instance, during the heat recovery step are required (IEA, 2018).
7.3. Chemical processes
Chemical processes are mainly aimed at the recovery of the module
metal fraction. These processes require, as compared to physical
treatments, larger costs, becoming economically feasible only if high
value metals can be recovered. For example, the treatment for thin film
modules is made cost-efficient by the value of the recovered rare ele-
ments (Nieland et al., 2012).
In this framework, considerable efforts were devoted to the
753
Solar Energy 177 (2019) 746–761
development of chemical processes to recover Ag, Al and the Si wafer in
Si type panels. Klugmann-Radziemska and Ostrowski (2010) proposed a
process including a thermal and a chemical treatment to recover Si
wafer. The process includes the application of HNO3 to extract Ag, KOH
to remove Al metal coatings, and the successive combined application
of HNO3, HF, ethanoic acid and Br to remove the antireflective layer.
Shin et al. (2017) fabricated Pb-free panels by using the wafers re-
covered from Si-panels. The entire panel was placed in a furnace and
the panel layers were separated by heat treatment at 480 °C. Then, the
Si wafer was immersed into HNO3 and then KOH solutions to extract Ag
and Al, respectively. The antireflective layer was removed by the ap-
plication of an etching paste containing H3PO4 and by heating. The
recycled wafers were then used to fabricate new, Pb-free solar panels.
Dias et al. (2016) demonstrated a process for the extraction and
concentration of Silver from Si-type panels. After removing the frame,
the panels were crushed and sieved to obtain fractions > 0.5 mm
and < 0.5 mm. Leaching of the finer fraction was then performed with
64% nitric acid and subsequent addition of NaCl, allowing obtaining an
AgCl precipitate. This process made it possible to concentrate 94% of
the silver contained in the modules. Palitzsch and Loser (2011a,b) de-
veloped a de-metallization method for Al and Ag recovery from broken
and end of life Si panels. In this process, the treatment with an alumi-
nium chloride solution leads to poly-aluminium-hydroxide-chloride,
which is used in the paper industry and in wastewater treatment ap-
plications. Ag can then be recovered by the treatment with HNO3.
Yi et al. (2014) demonstrated the technical feasibility of Si recovery
from PV modules by a chemical process where Si cells are first leached
in HNO3 3 M to remove Ag, and then in sodium hydroxide 3 M for 2 h to
remove Al and N. Following these process stages, smelting at 1520 °C
with CaO–CaF2–SiO2 was performed to scavenge the remaining im-
purities and thus produce Si with purity larger than 99.998%. In the
process proposed by Jung et al. (2016), thermal treatment of Si-type
panels at temperature values up to 480 °C is performed to separate
glass, EVA and solar cell layers. Ag, Al, Cu, and Pb from the solar cell
and the Cu wires are then dissolved in nitric acid at room temperature
for one hour. Cu is recovered from the nitric acid solution through
LIX84-I solvent extraction Hydrochloric acid is added to the solution in
order to precipitate AgCl, which is further purified. The remaining so-
lution containing Pb is treated with sodium hydroxide to form Pb(OH)2,
which is filtered and purified. The Si wafer free of metals is treated with
solutions of phosphoric acid and potassium hydroxide for the removal
of SiNx and Al impurities.
In the work by Kuroiwa et al. (2014), germanium was recovered
from a powder referred to by the authors as waste of solar panels. The
powder, mainly composed of SiO2 and GeO2, is suspended in an alka-
line solution and, when catechol is added, a complex of germanium
with catechol is formed. The complex is concentrated using a mem-
brane system and eluted with an acid solution. Then the catechol
forming the complex with germanium is removed by solvent extraction
using trioctylphosphine oxide. The ratio germanium/silicate dissolved
in the alkaline solution was 0.023 increased to 0.89 at the end the
process, meaning that the concentration of germanium was increased
by a factor 39.
Cadmium and tellurium are the main targets of CdTe panels re-
cycling. Fthenakis and Wang (2006) optimized the recovery of Cd and
Te from CdTe panels reporting that the cost of the optimized process is
2 cents per W for a 10 MW/year recycling plant. The proposed process
is based on leaching with sulphuric acid and hydrogen peroxide at room
temperature and the subsequent separation of Cd and Te by cation-
exchange resin. In a different study by the same authors, batch tests
with resin allowed removing approximately 91% of Cd and 21% of Te,
while experiments with two columns in series attained a removal of
cadmium higher than 99.99%, with 90–96% of the tellurium remaining
in the solution (Wang and Fthenakis, 2005).
Different authors proposed processes for the extraction of rare me-
tals like In, Ga and Se from the CIGS panels after acid leaching. In the
F.C.S.M. Padoan et al. Solar Energy 177 (2019) 746–761

SENSE project, a liquid-liquid extraction with D2EHPA was performed
to separate In and Mo from the leaching solution. The metals In and Mo
were then extracted from the organic phase with hydrochloric acid,
precipitated and filtered. Selenium was recovered from the inorganic
solution by reduction and precipitation with Na2SO3. Gallium was re-
covered from the inorganic solution by adding sodium hydroxide. The
resulting hydroxide can be further purified by means of electrolytic
processes, crystallization, etc. (SENSE, 2008; Marwede et al., 2013). A
different order of the extraction sequence was proposed by Dattilo
(2011): first the compound is deselenized, then Cu is removed and then
In and Ga are separated (Marwede et al., 2013).
Gustafsson et al. (2014) proposed the recovery of selenium from
CIGS panel by 1 h oxidation at 800 °C. With the proposed process, the
gaseous selenium dioxide generated during oxidation forms upon
cooling crystals that can be collected and then reduced to high purity
selenium. The residue contains oxides of copper, indium, and gallium. A
different article by the same research group describes a process for the
separation of these metals by high-temperature chlorination with am-
monium chloride (Gustafsson et al., 2015). The high-temperature
chlorination is a separation method relying on the treatment of a ma-
terial rich in metals with a chlorination agent at high temperatures to
form volatile metal chlorides. In the proposed chlorination process, the
residue obtained after the separation of selenium is introduced into a
furnace at the required reaction temperatures, which are 260 and
340 °C for gallium and indium, respectively. During the experiments
nitrogen gas transports volatile products to the furnace outlet, where
the product is sublimed. The authors report that by this approach a
good separation between Ga and Cu can be attained, but the content of
In in the recovered Ga and Cu products needs to be decreased
(Gustafsson et al., 2015).
Chemical treatment is typically performed after the physical or
thermal pre-treatment of the panel and relies on the application of
harmful reagents. Further, we notice that while the employed reagents
could result not expensive, which is the case of sulphuric acid and
hydrogen peroxide, their application determines the generation of large
volumes of liquid wastes. This contributes to decrease the economic and
environmental sustainability of a process including chemical treat-
ments, as compared to physical and thermal processes. In practice, the
additional costs and the environmental disadvantages must be out-
weighed by the economic and environmental benefits resulting from the
recovery of the metals. To this purpose, the possibility to recover rare
and precious metals can make the chemical treatments cost-efficient,
especially if thin film technologies are concerned. Recycling processes
targeting the recovery of Ag from Si crystalline panels can also be very
interesting. The economic feasibility of this route is largely influenced
by the year of manufacture of the processed panels, which significantly
influence the Ag content.
7.4. Combined physical and chemical treatment
Several studies have investigated the optimization of the PV panels
recycling by combination of different types of treatments. Following
this approach, Pagnanelli et al. (2017) treated different types of pho-
tovoltaic panels by a process route including two main steps: a physical
treatment (triple crushing and thermal treatment) and a chemical
treatment. According to the authors, three different fractions were ob-
tained by triple crushing: an intermediate fraction directly recovered as
glass, a coarse fraction mainly composed of Si cells fragments and glass
particles glued to EVA, and a finer fraction composed of a glass powder
and metals. The coarse fraction was thermally treated for 1 h at 650 °C
to detach EVA from glass, while 3 h chemical treatment of the finer
fraction with H2SO4 and H2O2 at 60 °C allowed dissolving metals and
thus obtain an additional recoverable glass fraction. The overall process
allowed recovering 91% of the treated panels. Savvilotidou et al. (2017)
contributed to characterize the composition of the different PV panel
components, such as Al frame, panel edge and junction box. Following

Table 2
Patented recycling processes specified by title, year of publication and expected obtained products, according to the PV technology.
Panel type Treatment Patent name Publication number Year Products

Si Thermal Recycling silicon photovoltaic modules US6063995 A 2000 Glass, lead and solar cell

CdTe Chemical Recycling of CdTe photovoltaic waste US5997718 A 1999 Glass, CdO and TeO or Te metallic
Reclaiming metallic material from an article comprising a non- US 6391165 B1 2002 Glass, CdCO3 and elemental Te
metallic friable substrate
System and method for separating tellurium from cadmium waste US7731920 B2 2010 Glass, metallic Cd or CdSO4
System and process for recovery of cadmium telluride (CdTe) US 8048194 B2 2011 CdTe
Recovery processing method of cadmium telluride thin-film solar CN 103199147 A 2013 Solid mixture containing Cd and Te
cell
Combined Process for recycling CdTe/CdS thin film solar cell modules US6572782 B2 2003 CdCl2 and TeCl4
Process to recycle end of life CdTe modules and manufacturing US2012325676 A1 2012 Glass and Metallic Te
scrap

CIS/CIGS Chemical Recycling of CIS photovoltaic waste US5779877 A 1998 Glass, metallic Cd, CuO, SeO and ZnO/InO mixture
Method for recovery of copper, indium, gallium and selenium US0329970 A1 2010 Elemental Se, metallic Cu and metallic In
Combined CIS based thin film solar cell module, method for producing the EP 1830411 A1 2007 glass substrate, solar cell device, EVA resin, cover
same, and method for separating solar cell module glass, back sheet, and non-adhesive sheet
Method of recovering constituent member of CIS type thin film WO2008102457A1 2008 Glass and panel constituents separated and recovered
solar cell module
Recycling of copper indium gallium diselenide WO2013089630A1 2013 Elemental Se
Treatment of indium gallium alloys and recovery of indium and CA 2721518C 2013 Ga and In salt separated solutions
gallium
Recovery method of copper indium gallium selenide thin-film CN103184338A 2013 Se elemental, CuOH, In and Ga solutions
solar panel
Recovery method of thin-film solar cells CN 104201248 A 2014 Reusable thin-film solar cell
Optical Recycling method for thin layer solar cell modules US 20090308535 A1 2009 Reusable substrate layer
Method for the separating detachment of layers of a composite WO2015051977A1 2015 Glass
component formed from at least two layers

Generic Thermal Recycling insulating and/or laminated glass panels DE 4418573 C1 1996 Glass and remaining metal parts
Method for separating the components of a laminated glazing EP0893250 B1 2003 Solar module components separated
Combined Film removal US0147230 A1 2011 Reusable substrate layer
Process for treating spent photovoltaic panels EP 2997169 A1 2016 Aluminium frame, glass, metallic Te or TeO and Fe,
Cd and Al precipitate separately

754
F.C.S.M. Padoan et al.
this work, they also tested acid baths to recover the pure components
for further use. They found that the application of sulphuric acid for 1 h
and of lactic acid for 4 days were sufficient to attain the delamination of
Si-a and CIGS panels, respectively. The authors also noted that tem-
perature and stirring did not play a significant role for panel's delami-
nation. Sasala et al. (1996) studied the recycling of CdTe modules by
both physical and chemical operations. They proposed a pre-treatment
by water blasting and chemical operations such as leaching, precipita-
tion and electroplating of the semiconductors materials. Kang et al.
(2012) proposed a solvent treatment integrated to thermal and che-
mical treatments. They first recovered glass of silicon modules by dis-
solving EVA with toluene, decomposed the residual EVA by a thermal
step at 600 °C, and finally removed metal impurities by an etching
treatment with acids in order to recover pure silicon. Kushiya et al.
(2003) proposed a process to recycle CIGS panels whereby the panel
without the aluminium frame and the junction box (manually removed)
was heated to a temperature below 250 °C to soften the EVA and thus
facilitate the separation of glass. They successively immersed the device
in an acetic acid solution to remove the remaining EVA, collected the
CIGS-based absorber as metal powder by mechanical scribing tech-
nique, and finally dissolved the Mo layer with nitric acid to recover the
glass. Huang et al. (2017) treated Si panels by a process whereby the
mechanical removal of the Al frame and of the junction box was fol-
lowed by the burning of polymers to recover glass. The obtained se-
parated cells were then leached in nitric acid to dissolve Ag, Pb, Cu and
Sn, which were recovered one by one by electrowinning. Si was re-
covered by an etching process. Authors highlight that chemicals used in
the process can be neutralized forming NaNO3 which can be used as a
fertilizer, while the polymers can be used as a heat source in a furnace,
though the application of proper filters is needed in this latter appli-
cation to prevent harmful gas emissions.
Following a partially different approach, Fernandez et al. (2011)
examined the possibility of recycling silicon modules by incorporating
them into cement matrices, but follow this pathway means the loss of
conventional and rare resources (Palitzsch and Loser, 2011a,b).
8. Patented recycling processes
In the commercial field, many patented recycling processes can be
found, most of which are listed in Table 2. Large part of these patented
processes has been developed by large companies of the photovoltaic
sector aiming at the recovery of wafer and secondary raw materials. In
addition, patents developed through the cooperation of academic re-
searchers and companies can be distinguished.
8.1. Si processes
Bohland and Anisimov (2000) (Patent US6063995 A) developed a
method for the First Solar company to recycle crystalline silicon pho-
tovoltaic modules. In this process, the crystalline silicon cells glued to
EVA and TEDLAR are slowly heated and the coating film is removed
manually. Then, the pyrolysis of EVA is performed under an inert at-
mosphere at about 500 °C to recover solar cells. The recovered cells
retain between about 80–100% of their delaminated cell efficiency.
Lead that felt on the floor of the chamber during the heating process can
be swept and recovered.
8.2. CdTe processes
Many processes can be found involving recycling or reclaiming of
components from thin film photovoltaic panels as compared to different
technologies. This is probably explained by the larger content of high
value materials found in thin film panels, which can ensure the eco-
nomic viability of the recycling process.
In this framework, Goozner et al. (1999) patented a process
(US5997718 A) for Drinkard Metalox, Inc. based on the extraction and
755
Solar Energy 177 (2019) 746–761
reclaim of metals and glass from CdTe photovoltaic cells and photo-
voltaic manufacturing waste. The process includes the following steps:
• Metal extraction by leaching with nitric acid to obtain glass, which
is reused to produce new modules.
• Skimming the plastic laminate from the leaching solution.
• Neutralization of the leaching solution by CaO to separate Cd and Te
by precipitation.
• Roasting of the solid precipitate and cooling by washing to recover
CdO and TeO.
An alternative embodiment is the replacement of the neutralization
with the electrolysis of the leachate to separate Cd from Te, wherein the
Te is deposited on a cathode while the Cd remains in solution.
Bohland et al. (2002) in a partnership with First Solar developed a
process (US 6391165 B1) which comprises leaching of crushed panels
with sulphuric acid and hydrogen peroxide. Following leaching, pre-
cipitation is performed by using sodium carbonate to obtain CdCO3 and
TeO2. Another leaching is carried out to bring Te in solution, which is
conveyed to an electrolytic cell, where separation of elemental Te is
attained. Campo et al. and ANTEC Solar GmbH (2003) (US6572782 B2)
patented a recycling process for CdTe and CdS thin film modules which
achieves the size reduction of the modules by mechanical disintegration
or crushing, heat treatment at 300 °C under an oxygen atmosphere
(causing pyrolysis of the organic adhesive material contained in the
module) and final treatment of residual fragments with chlorine gas at
400 °C, generating CdCl2 and TeCl4 and subsequent condensation and
precipitation by cooling.
A patent by Fthenakis and Wang and the Brookhaven Science As-
sociates (2010) (US7731920 B2) in a partnership with First Solar pro-
poses a method to extract and separate Cd and Te from a waste in-
cluding solar modules (Fig. 7). The process is basically composed of the
following steps:
Crushed PV waste
LEACHING TREATMENT
H2SO4 + H2 O2
CATION EXCHANGE RESIN
Amberlyst ® + Dowex®
Cd elution Te
H2SO4
Cd
eletrolysis Evaporation
Cd metal CdSO4
Fig. 7. Block diagram of Brookhaven Science Associates with First Solar part-
nership process.
F.C.S.M. Padoan et al.
• Leaching in sulphuric acid and hydrogen peroxide at room tem-
perature.
• Ion-exchange resins at room temperature for the separation of Cd
and Te.
• Acid elution of Cd from the resins.
• Electroplating or evaporation in order to obtain metallic Cd and
cadmium sulphate, respectively.
In 2011, Rathweg (2011) patented a process (US 8048194 B2) for
the recovery of cadmium telluride plated onto components used in the
manufacture of photovoltaic modules. The system includes a vacuum
oven configured for being heated from a temperature about 650 to
about 850 °C. This allows sublimating CdTe off of components placed in
the oven. A collection member, who includes a base and a tray, is
disposed so that sublimated CdTe generated in the oven diffuses to it.
The collection member is subsequently processed to collect the plated
CdTe, by shaking, banging, or bending the tray in a controlled en-
vironment to cause the plated CdTe to flake off of the collection
member. In an alternative embodiment, chemical treatment of the
collection members is foreseen to remove the CdTe.
Taylor et al. (2012) in a patent from 2012 (US2012325676 A1)
proposed a method for the recycling of cadmium telluride end-of-life
photovoltaic modules and manufacturing scraps. The process involves
delamination in an oven under N2 atmosphere at about 400 °C, blasting
to remove some impurities or glass pieces, roasting under an oxygen
containing atmosphere at about 700 °C, resulting in an oxidized coating
based in CdO and TeO2. Alkaline leaching is successively performed,
taking advantage of the different solubility of CdO and TeO2 in alkaline
solutions, in order to bring in solution Te, which is recovered by elec-
trolysis. The authors also suggest that Cd can be recovered from the
sludge by extraction with acids followed by electrolysis. Still in the field
of CdTe panels recovery, the invention CN 103199147 A (2013) dis-
closes a process to recycle CdTe thin-film solar cells relying on the
leaching with sulphuric acid and hydrogen peroxide of the crushed and,
consequently, totally exposed material. Filtration is performed fol-
lowing leaching to separate the liquid from the solid material con-
taining mostly glass and EVA. A flocculation process is then promoted
by adding an alkaline and a polymeric aluminium solution that adsorb
Cd and Te ions. Finally, an activated carbon powder is added in order to
remove some impurities and permit the last step of filtration of the
solids containing Cd and Te with a DF film.
8.3. CIS and CIGS processes
Another branch of thin film panel that has been gaining market are
the panels whose composition is based on copper, indium, selenium
(CIS) and gallium (CIGS) due to their high photoelectric conversion
efficiency. Consequently, this expansion and the necessity to reduce the
cost and meet the environmental requirements leveraged the develop-
ment of many patents with the aim of recovering such panels, which are
listed below.
The company Drinkard et al. (1998) are the owners of a patent
(US5779877 A) describing a method for extracting and reclaiming
metals from scrap CIS photovoltaic cells and associated photovoltaic
manufacturing waste. Initially, leaching is performed with nitric acid,
generating a leaching solution and a glass substrate that can be reused
in the manufacture of PV cells. EVA floats to the top of leaching solution
and is removed by skimming. Subsequently, copper and selenium are
deposited on a first cathode by electrolysis and then re-electrolyzing the
leaching solution allows to deposit cadmium, obtaining a residual so-
lution containing indium and zinc. Selenium and copper are further
separated by oxidation and distillation while the solution containing
indium and zinc is subjected to evaporation and decomposition,
yielding a precipitate based on zinc and indium oxides.
Kushiya and Tanaka (2007), in part of their patent (EP 1830411
A1), describe five steps for separating CIS based thin film solar cell
756
Solar Energy 177 (2019) 746–761
modules. The photovoltaic module is subjected to a separation process
comprising a vacuum step, sandblasting, a wet chemical method, which
can be boiling, high-temperature steam blowing, or immersion in an
acid and mechanical stripping. The authors claim that by this process,
the photovoltaic module can be separated into individual constituent
materials, such as glass substrate, solar cell device, EVA resin, cover
glass, back sheet, and non-adhesive sheet, and these constituent mate-
rials can be separately recovered as resources.
The invention of Kushiya and Okazawa (2008) (WO2008102457A1)
provides the method of recovering constituent members of a CIS thin-
film solar cell module at low cost by physical operations, wherein
peripheral members are separated and recovered by a pre-treatment.
Thereafter, the structure is heated in a closed heating oven at a tem-
perature in the range of 400–500 °C to burn EVA resin, whereby the
cover glass can be separated and recovered. In this manner and with the
aid of other techniques, like scraping followed by sandblasting or belt
sander, several panel constituents may be sequentially separated and
recovered.
Solar Applied Materials Technology Corporation with Lian et al.
(2010) (US0329970 A1) have developed a method for the recovery of
copper, indium, gallium and selenium including leaching with a mixed
solution containing hydrochloric acid and hydrogen peroxide to dis-
solve the copper, indium, gallium and selenium. Then, hydrazine is
added to the solution to separate selenium out by reducing it to ele-
mental selenium. After that, copper is reduced by cementation with
indium metal. Later, a combination of a supported liquid membrane
containing D2EHPA and a strip dispersion solution with hydrochloric
acid separates the gallium from the indium. The remaining solution
is then subjected to an electrolytic process for obtaining the indium
(Fig. 8).
A patent of Midsummer Ab with Foreman et al. (2013)
(WO2013089630A1) relates a recovery method of valuable elements
from a compound having the chemical formula
CuInxGa(1−x)Se2, wherein x has a value from 0.01 to 0.99 and may be
useful for recycling in the field of solar cell technology. In this process,
crushed solar cell in the form of a powder is subjected to heating to at
least 500 °C and contacting the material with a gas flow comprising
oxygen, forming selenium dioxide and remaining CIG residue. The
authors explain that elemental selenium of high purity may be obtained
when the SeO2 is subjected to Riley reaction conditions or reduced with
Crushed PV waste
Elemental
Se REDUCED PRECIPITATION Se
Hydrazine
Recovered
CEMENTATION
Indium
Cu
SOLVENT EXTRACTION Recovered
D2EHPA membrane
Ga HCl
ELECTROLYSIS
Recovered In
Fig. 8. Block diagram of Solar Applied Materials Technology Corporation
process.
F.C.S.M. Padoan et al.
sulphur dioxide. Cesare Ferron et al. in a patent of 2013 (CA 2721518C)
propose a hydrometallurgical process of recovering indium and gallium
from mixtures with other metal values for example recycled and scrap
solar panel cells. Their method comprises the following steps:
• Crushing the starting material.
• Leaching under stirring the obtained mixture with sulphuric acid
and hydrogen peroxide so as to form indium and gallium salt spe-
cies.
• Adjusting the pH of the leaching solution to a range of about 0.5 and
0.6.
• Indium solvent extraction by with D2EHPA.
• Adjusting the pH of the leaching solution to a range of about 1.5 and
2.5.
• Gallium solvent extraction by with D2EHPA.
• Stripping the indium-loaded organic compound with hydrochloric
acid so as to recover pure indium salt.
• Stripping the gallium-loaded organic compound with sodium hy-
droxide so as to recover substantially pure gallium salt.
The authors suggest that additional techniques may be employed for
purification, if required. For example, indium cementation with Al or
Zn plates, direct electrowinning of the gallium hydroxide solution fol-
lowed by mineral acid cleaning. They also cite a possible recovery of
other metals values present in the mixture, such as high temperature
oxidative removal of selenium dioxide by roasting and reductive pre-
cipitation of selenium by adding sulphur. Guangdong Xiandao Rare
Material Co. Ltd. (2013) processed a hydrometallurgical treatment that
after grinding CIGS panels in pieces < 0.5 mm, leaches them with sul-
phuric acid and hydrogen peroxide. Then, they recover In by solvent
extraction, reduced Se with sodium sulphide and precipitate Cu as
hydroxides by adding sodium hydroxide. The invention CN
104201248A (2014) relates a physical recovery method of thin-film
solar cells including the steps of knocking a thin-film solar cell to form a
crack mesh in a tempered glass layer; freezing the broken thin-film solar
cell with liquid nitrogen, which is enough to at least submerge the thin-
film solar cell; and knocking to separate the frozen thin-film solar cell to
obtain a thin-film solar cell electromagnetic board. The authors report
that the recovery method obtains a thin-film solar cell electromagnetic
board with a clean surface free of impurities, and the thin-film solar cell
electromagnetic board that can be reusable after being treated.
Wagner and Schmieder (2009) (US 20090308535 A1) described a
method for recycling thin film solar cell modules using laser beam in
order to perform complete edge deletion by guiding the laser beam,
which is absorbed by the layers to be removed. The authors exploit the
property of the substrate layer to be laser beam proof, and the first
functional layer, an electrode layer to be able to absorb this laser beam.
The free surface of the substrate layer is scanned by the laser beam so
that layer is partially vaporized due to having absorbed the laser beam,
and is thus detached from the substrate layer. The substrate layer, se-
parately from the functional layers that are attached to the plastic layer
and the cover layer, is subsequently available for further processing. By
following an analogous optical approach, Loser Chemie with Leson
et al. (2015) (WO2015051977A1) developed a technology with two
treatment options: the use of laser and flash lamp for the separation of
the glass structures. After this separation, the layers are chemically
treated and then products are sent to metal refineries to be purified
(IEA, 2018).
8.4. Generic processes
In this section are presented processes to treat more than one type of
PV panel. Weinfurtner (1996) (DE 4418573 C1) patented a method in
which insulating or laminated glass panels are loaded into a basket and
charged into a fluidized bed furnace at a temperature between 450 and
500 °C for about one hour. In this stage, all organic substances, such as
757
Solar Energy 177 (2019) 746–761
phosphors, adhesives and films are combusted. Then, the basket is
transported to a further station wherein the containing water sprays
which quench the basket and contents. Glass fragments and other
particles fall through the grid to a conveyor, while remaining metal
parts are separated for further recycling.
Deutsche Solar AG by Stötzel and Wambach (2003) (Patent
EP0893250 B1) developed a solution for the treatment of crystalline
silicon, CIS and CdTe modules in order to separate their components.
The laminated glass of the solar modules are placed in a heatable frame
oven and heated in the presence of some oxidant agents, reaching the
final temperature of at least 300 °C, which provides the decrease of the
adhesion of the plastic to the glass. The maximum temperature is
maintained until the plastic layer is substantially decomposed or eva-
porated (at least 95% of the initial amount). Thereafter, cooling of the
remaining components of the solar module is performed. In a patent by
First Solar, Inc. with Reed and Wang (2011) (US0147230 A1) it is de-
scribed a method for recycling active elements in photovoltaic modules.
The photovoltaic modules of the patent are those having a semi-
conductor material deposited on a substrate such as cadmium sulphide
or cadmium telluride or those characterized by the presence of a cov-
ering of transparent conductive oxide which can be deposited before
the conductor layer, for example before the cadmium stannate. The
layers may be removed using chemical, mechanical, electrochemical
methods, and a combination thereof. The pathway described in this
patent is that of the chemical method by treatment with HCl made more
effective by means of mechanical action, and an electrochemical
method which consists of the use of cadmium stannate as the cathode. A
substrate treated using the methods and apparatuses described herein
may be reused, washed, or recycled. The invention of Toro et al. (2016)
(EP 2997169 A1) relates to a process for the treatment of photovoltaic
end-of-life panels, such as those made of CdTe and crystalline and
amorphous silicon all together, without any kind of preliminary selec-
tion. The process involves combined automated physical and chemical
operations that allow recovering glass in the first place. This process
initiates with disassembling of the outer frame from which aluminum is
directly recoverable followed by crushing, grinding and screening op-
erations. The coarse fraction is sent to a subsequent washing step with
cyclohexane and acetone in order to solubilize the adhesive, resulting in
separation of glass-silicon non-reflecting material. The intermediate
fraction is directly recoverable as glass. The fine fraction is subjected to
an alkaline leaching with sodium hydroxide at pH 11 in order to so-
lubilize and obtain a Te solution that is subjected to electrolysis re-
covering Te in a metallic form or as an oxide by precipitation with nitric
acid. Second alkaline leaching step at pH 12–14 provides solubilisation
and obtainment of a Zn and Al solution. Then Al is precipitate by adding
sulphuric acid until reach pH 3–6. The solid residue from the second
alkaline leaching is treated with sulphuric acid for the solubilize Fe and
Cd, the solution is filtered, brought to pH 4 precipitating Fe and then to
pH 11 to precipitate Cd. The process guarantees a recovery of glass of
80–85% by weight and the metallic materials are recovered with dif-
ferent degrees of purity and can be used as secondary raw materials in
production processes conventionally fed with primary raw materials,
for example from mining.
9. Current end of life PV panels recycling processes
In Europe, since 2012 end of life photovoltaic panels are classified
as WEEE and they must therefore be recycled. In order to comply with
the current EU regulation, the following processes are currently oper-
ated:
• First Solar: it is the treatment developed by the US company pro-
ducer of panels based on Cadmium-Telluride. The recovery phases
are distinguished in mechanical and chemical processes. The first
phase comprises the crushing of the photovoltaic device. Then the
film removal is carried out inside a steel drum where, for slow
F.C.S.M. Padoan et al.
rotation and addition of acid and peroxide, the film is removed. The
film removed by acid treatment, is separated from the various pieces
of panel. A phase of valorization of the liquid fraction is carried out
by means of precipitation of dissolved metals present in the liquid
fraction. Precipitation reactions occur in several steps increasing pH
and concentrating the precipitate in thickening tanks. The solid
fraction is separated into glass and laminated materials with vi-
brating screens followed by washing with water to remove the re-
sidual films on this fraction, while the laminate is subjected to dis-
posal according to appropriate procedures (First Solar, 2018).
The chemical approach would not be as efficient if there was no
mechanical pre-treatment of grinding, which increased the surface
contact of the panel with the chemical agent. However, it is known that
such treatment is susceptible to toxic gas and liquid emissions, re-
quiring environmentally friendly measures that control such emissions
(IEA, 2018).
• Deutsche Solar: This is a company affiliated to the SolarWord
group, where the recovery operations aim at the silicon as raw
material also from damaged panels. The process is composed by a
first heat treatment followed by a chemical treatment. The heat
treatment is carried out in an oven at 600 °C, incinerating EVA and
tedlar. The produced gases are treated according to a specific
abatement process. The solid residue (glass and aluminum) is
manually disassembled and sent to the chemical treatment.
Photovoltaic wafer is removed through chemical attacks on the
surface, given by acids (nitric acid, sulphuric acid, acetic acid) with
a sequential and selective approach (Bombach et al., 2005).
The energy consumption of the Deutsche Solar recycling process is
substantial. However, according to the LCA analysis done by Müller
et al. (2005), recycling can be justified by the environmental benefits
attained with the reuse of the recovered wafers. Nevertheless, results of
the LCA analysis showed the importance of reducing the energy con-
sumption of the thermal treatment.
• Nike*: The end of life photovoltaic panels are subjected to a pre-
treatment phase from which the cables and the junction box are
removed, to be subsequently shredded. A double grinding is applied
to obtain a size of about 20 mm, followed by the separation of the
ferrous and non-ferrous (alluminum) metal components. The frac-
tion free of magnetic and aluminum material is screened to separate
silicon, glass, plastics and other metals. Glass is refined for separa-
tion from impurities with a shake table (Nike*, 2018).
All of the above described processes start with a mechanical treat-
ment step, which basically consists of the dimensional reduction of the
panel through grinding. The Nike* process, for example, is a series of
only mechanical treatments. This type of approach is interesting from
an economic point of view and recovers the amount provided by leg-
islation. However, the purity and consequently the economic value of
the product obtained is inferior to that obtained by chemical treatment.
In addition, legislation is expected to be tightened on toxic compounds
in WEEE. Thus, the mechanical treatment will be insufficient and the
processes should be upgraded (IEA, 2018).
10. Economic efficiency of a recycling process
The reuse of material derived from waste to manufacture new
products represents one of the most important sustainability challenges
(Cucchiella et al., 2015). It is known that PV recycling will provide
secondary raw materials that can be used in the production of other
photovoltaic systems, which can reduce the consumption of primary
raw materials and thus simultaneously generate an environmental and
an economic benefit.
758
Solar Energy 177 (2019) 746–761
Table 3
Materials market price (MetalBulletin).
Material Al Cd Te Cu In Ga Se Glass
€/ton 1780 1740 40,000 5725 230,500 144,367 35,530 50
In this section, we aimed at selecting the recycling processes with
the highest economic potential. The applied methodology included the
following preliminary steps:
(1) Selecting the patented processes whose recovery rates are available
based on the previous patents analysis (Table 2).
(2) Quantifying the amount of each target material in the modules
based on characterization data available in the literature (Average
data used to prepare Fig. 2).
(3) Identifying the current market price of each target material
(Table 3).
Following the previous steps, the economic potential (EP) of a re-
cycling process (€/ton of treated panel) can be calculated as follows:
n
EP = ∑i =1 (ri ∗ai ∗ pi ) (5)
where ri, ai and pi denote the recovery rate (%), the content (g/ton) and
the price (€/g) of a target material, respectively.
Table 3 shows the market prices for each target material of the re-
cycling processes studied in this section. It should be noted that such
data are very volatile. This is the case of the tellurium price, which has
fluctuated significantly since 2010. Similarly, the price of indium de-
creased significantly in 2015 as compared to previous years. The ana-
lysis here presented was based on reference data reported for the first
half of 2018.
Table 4 lists the patented process whose recovery rates are available
as well as the economic potential generated from the sale of recovered
products, as calculated from equation (5). It is worth to notice that the
EP was calculated by substituting in Eq. (5) the market prices pi of the
recovered materials as if they were pure. Accordingly, the estimated EP
value can be thought to provide the maximum economic value that can
be obtained by the sale of recovered products. However, in most of the
proposed recycling processes, the final products do not present high
purity, which can decrease the profit. Quantifying the effect of product
purities on the economic feasibility of the considered patented pro-
cesses is frequently impossible mainly because insufficient data are
available about product purities. Nonetheless, we notice that, by an
economic potential assessment of several types of WEEE, Cucchiella
et al. (2015) reported that market price of a recovered material can be
confidently reduced by a value equal to the its standard deviation.
It should be then remarked that factors other than the EP value can
influence economic feasibility, which include, for example, energy and
reactants consumption, the ratio between the mass of generated waste
and that of treated PV panels, and the costs due for the treatment of
generated wastes. Extending such an analysis to any considered process
goes far beyond the scope of the present article. However, data reported
in Table 4 can provide an important indication about the possibility to
attain economic feasibility with the proposed processes.
In accordance with the EP values reported in Table 4, we notice that
the economic potential of the recycling processes proposed for CIGS and
CdTe are similar, even though the prices of In and Ga (found in CIGS)
are much larger than those found for Cd and Te (found in CdTe). This is
due to the very low concentrations of In and Ga in the CIGS panels. It
should be remarked that EP values reported for CIGS panels have been
computed in the assumption of In and Ga recovery rates larger than
97%. Such effective recovery rates is the main element that can sustain
the recycling of CIGS panels, which is in agreement with the analysis
presented by Cucchiella et al. (2015) and Paiano (2015).
F.C.S.M. Padoan et al.
Table 4
Material recovery rates and total revenue of patented processes.
Reference Treated panel Recovery products and rates
US5997718 A CdTe Glass (100%) Cd (99% as CdO) Te (100% as TeO)
US 6391165 B1 CdTe Glass (100%) Cd (not reported) Te (67% as Te elemental)
US2012325676 A1 CdTe Glass (100%) Te (37.8% as Te metallic)
US5779877 A CIGS Glass (100%) Cd (not reported) Cu (not reported)
Se (59.4% as SeO) Zn and In (not reported)
US0329970 A1 CIGS Se (97.6% as Se elemental) Cu (97.7% as Cu metallic) and In (97.7% as In metallic)
WO2013089630A1 CIGS Se (88.4%)
CA 2721518C CIGS Ga (68.8%) In (99.4%)
CN103184338A CIGS In (97.46%) Se (97.94%) Cu (98.79%)
11. Risk and environmental impacts of the PV recycling processes
Recycling of photovoltaic panels is still a young area. Therefore,
there may be several advantages and disadvantages for each of the
techniques used. In general, recycling of PV cells is complicated, either
due to the fact that the materials to be recovered are in low con-
centration in the modules, or due to lack of logistics in the collection of
panels since they are dispersed geographically (Masoumian and
Kopacek, 2015). It is possible to compare such processes following the
classification of the techniques used in the previous paragraph.
It is generally agreed that the extensive application of PV panels
does not introduce any environmental risk and can, rather, generate an
environmental benefit allowing for the application of renewable solar
energy in place of the conventional energy resources from fossil fuels,
which can ultimately lead to the reduction of greenhouse gases and
toxic gases including NOx, CO and SO2. Yet, a more rigorous evaluation
of the environmental impact resulting from the application of the PV
technology imposes to thoroughly analyse the end of life phase of solar
modules.
The recycling of PV panels must be carried out, according to the
legislation in force, to prevent the leaching of hazardous substances, the
loss of conventional resources and of rare metals (BioIS, 2011). In this
framework, some issues concerning the end of life photovoltaic panels
must be taken into account to definitively assess the environmental
impact of PV technology, including the consumption of energy and
reagents, and the emissions of pollutants that can be generated by the
recycling and recovery processes (Tammaro et al., 2015). The man-
agement of end of life photovoltaic panels proves to be an important
phase because it develops materials that would be dispersed if in-
cinerated or dismantled (Malandrino et al., 2017). In this context, a
study by Tammaro et al. (2015) investigated the amount of metals that
release to the gas phase and ashes in a thermal treatment of photo-
voltaic panels at 600 °C. The authors state that it is not possible to
generalize the obtained information, due to the variable composition of
the panels employed in the study, even these were manufactured in the
same year or by the same producer. However, they found the presence
of the toxic metals Pb and Cd in the gaseous phase and raised the im-
portance to prevent the dispersion of ashes by a flue gas treatment
section including, for example, an electrostatic precipitator and/or a
fabric filter.
Many reactants are used in the chemical treatment of photovoltaic
panels, especially when the objective is the recovery of metals or the
reuse of the wafers. In this context, Klugmann-Radziemska and
Ostrowski (2010) concluded that the etching mixtures used in the
chemical recovery of pure silicon from recycled PV contain high
amounts of toxic nitrogen oxides, fluorides and different silicon species,
which require costly disposal measures.
For what concerns the energy consumptions, Müller et al. (2005)
performed a life cycle assessment for a Deutsche Solar photovoltaic
recycling process, which recovers silicon wafers. They concluded that
the energy needed to recover a reusable wafer is two thirds lower as
compared to the energy needed to produce a new wafer. By reducing
759
Solar Energy 177 (2019) 746–761
Total revenues (€/ton of treated panel)
76
66
58
47
62
8
48
63
the production of new wafers and ensuring the simultaneous recycle of
glass and metals, the proposed process was found to reduce the en-
vironmental burden for different impact categories.
The results obtained by Corcelli et al. (2017) for a thermal recycling
process show that, adding the recovery costs of decommissioned mod-
ules to the cost of production and operation of PV modules, the final
energy demand is still much lower as compared to a conventional fossil
powered plant.
12. Conclusions
The rapid increase in the photovoltaic power installed worldwide
will cause over the next few decades a dramatic increase in the volume
of end-of-life photovoltaic panels. In accordance with the analysis
presented in this article, peaks for dismissal of PV panels are expected
to occur around the years 2036 and 2045. The improper disposal of
these waste fluxes could cause harmful effects to human health and to
economy of the manufacture sector by the dispersion of toxic elements
and loss of valuable material resources including rare metals, respec-
tively.
Several countries, especially European nations, have legislated to
recycle end-of-life PV panels, encouraging the development of re-
claiming methods. In the commercial field, there are many patented
recycling processes, most of them (43%) involving the hydro-
metallurgical treatment of CIGS panels. Several process solutions have
been also patented including the chemical treatment for Si and CIGS
end-of-life panels. It has also been proven that processes that consider
the recycling of thin film panels tend to be economically more attractive
due to the content of high value materials recovered.
In addition to patented processes, numerous recycling routes have
proposed by scientific articles including mechanical, thermal and che-
mical treatments. Mechanical processes have the advantage of being
inexpensive, but they cannot attain the recovery of high value mate-
rials, which requires more elaborate treatments. In addition, these
processes will be insufficient due to the need to treat and/or dispose
toxic elements, which will be required by the future legislation.
Proposed thermal processes typically rely on the pyrolysis of plastic
panel fraction and it is generally agreed that they are characterized by
high energy consumption. Remarkably, only one article could be found
considering the possibility to recover energy by the combustion of
pyrolysis products. The scarce attention to energy recovery can be ex-
plained by the technological difficulties that would be imposed by the
treatment of the generated gaseous emissions. Thermal processes are
very interesting because they do not generate further waste but produce
toxic gas emissions, like fluorine gas. Investigating this issue is there-
fore deemed fundamental to attain the environmental and economic
sustainability of thermal recycling processes. Chemical processes are
mainly aimed at the recovery of the module metal fraction and use
solvents and other reagents. These processes require larger costs but the
purity of the final output materials, especially high value metals, are
higher than those obtained from mechanical processes. Major issue to
be investigated is therefore the development of processes allowing for
F.C.S.M. Padoan et al.
the regeneration of the employed reactants, which could considerably
reduce the environmental impact and the processing costs.
It is worth remarking that, even considering the negative aspects
generated by the recycling of photovoltaic panels, like the expenditure
reagents and gas emissions, such processes are still convenient and can
be remedied with gas emission filters, reuse of reagents and wastewater
treatment. The energy expenditure in a recycling process is less than
that spent in the manufacture of a new panel. Thus, the research on PVP
recovery techniques is in constant development, emphasizing that an
ideal process is one that results in the treatment of all typologies, since
other technologies tend to emerge and share the market with Si tech-
nology.
Acknowledgements
This work was supported by CNPq (Process 202624/2015-3).
References
Bagher, A.M., Vahid, M.M.A., Mohsen, M., 2015. Types of solar cells and application. Am.
J. Opt. Phot. 3 (5), 94–113.
Bagnall, D.M., Boreland, M., 2008. Photovoltaic technologies. Energ. Pol. 36, 4390–4396.
Berger, W., Simon, F., Weimann, K., Alsema, E.A., 2010. A novel approach for the re-
cycling of thin film photovoltaic modules. Resour. Conserv. Recycl. 54, 711–718.
Bio Intelligence Service (BioIS), 2011. Study on photovoltaic panels supplementing the
impact assessment for a recast of the WEEE directive.
Bohland, J.R., Anisimov, I.I., 2000. Recycling silicon photovoltaic modules. Patent
US6063995 A.
Bohland, J.R., Anisimov, I.I., Dapkus, T.J., Sasala, R.S., Smigielski, K.A., Kamm, K.D.,
2002. Reclaiming metallic material from an article comprising a non-metallic friable
substrate. Patent US 6391165 B1.
Bombach, E., Wambach, K., Müller, A., Röver, I., 2005. Recycling of solar cells and
modules – recent improvements. In: 20th European Photovoltaic Solar Energy
Conference, Barcelona.
Campo, M.D., Bonnet, D., Gegenwart, R., Beier, J., 2003. Process for recycling CdTe/CdS
thin film solar cell modules. Patent US6572782 B2.
Choi, J., Fthenakis, V., 2014. Crystalline silicon photovoltaic recycling planning: macro
and micro perspectives. J. Clean. Prod. 66, 443–449.
CN 103199147 A, 2013. Recovery processing method of cadmium telluride thin-film solar
cell.
CN 104201248 A, 2014. Recovery method of thin-film solar cells.
Corcelli, F., Ripa, M., Ulgiati, S., 2017. End-of-life treatment of crystalline silicon pho-
tovoltaic panels. An emergy-based case study. J. Clean. Prod. 161, 1129–1142.
Cucchiella, F., D’Adamo, I., Lenny Koh, S.C., Rosa, P., 2015. Recycling of WEEEs: an
economic assessment of present and future e-waste streams. Renew. Sust. Energ. Rev.
51, 263–272.
Dattilo, M., 2011. CI(G)S PV modules: recycling technology status. In: 2nd International
Conference on PV Module Recycling, Madrid, Spain.
Desideri, U., Proietti, S., Zepparelli, F., Sdringola, P., Bini, S., 2012. Life Cycle Assessment
of a ground-mounted 1778 kWp photovoltaic plant and comparison with traditional
energy production systems. Appl. Energ. 97, 930–943.
Dias, P., Javimczik, S., Benevit, M., Veit, H., Bernardes, A.M., 2016. Recycling WEEE:
Extraction and concentration of silver from waste crystalline silicon photovoltaic
modules. Waste Manage. 57, 220–225.
Directive 2008/98/EC of the European Parliament and of the council on waste and re-
pealing certain Directives.
Directive 2011/65/EU of the European Parliament and of the council on the restriction of
the use of certain hazardous substances in electrical and electronic equipment.
Directive 2012/19/EU of the European Parliament and of the council on waste electrical
and electronic equipment (WEEE).
Doi, T., Tsuda, I., Unagida, H., Murata, A., Sakuta, K., Kurokawa, K., 2001. Experimental
study on PV module recycling with organic solvent method. Sol. Energ. Mat. Sol.
C.67, 397–403.
Doni, A., Dughiero, F., 2012. Electrothermal heating process applied to c-Si PV recycling.
In: 38th IEEE Photovoltaic Specialists Conference, pp. 757–762.
Drinkard Jr., W.F., Long, M.O., Goozner, R.E., 1998. Recycling of CIS photovoltaic waste.
Patent US5779877 A.
Dubey, S., Jadhav, N.Y., Zakirova, B., 2013. Socioeconomic and environmental impacts of
silicon based photovoltaic (PV) technologies. Energy Proced. 33, 322–334.
European Photovoltaic Industry Association (EPIA), 2014. Global Market Outlook for
Photovoltaics 2014–2018.
Fernandez, L.J., Ferrer, R., Aponte, D.F., Fernandez, P., 2011. Recycling silicon solar cell
waste in cement-based systems. Sol. Energ. Mat. Sol. C. 95, 1701–1706.
Ferron, C.G., 2013. Treatment of indium gallium alloys and recovery of indium and
gallium. Patent CA 2721518 C.
First Solar, 2018. First Solar Recycling Brochure. < www.firstsolar.com > (accessed in
02/10/2018).
Foreman, M.R.S., Gustafsson, A.M., Ekberg, C., 2013. Recycling of copper indium gallium
diselenide. Patent WO2013089630A1.
Frisson, L., Lieten, K., Bruton, T., Declercq, K., Szlufcik, J., De Moor, H., Goris, M., Benali,
760
Solar Energy 177 (2019) 746–761
A., Aceves O., 2000. Recent improvements in industrial PV module recycling. In: 16th
European Photovoltaic Solar Energy Conference, Glasgow, UK.
Fthenakis, V.M., Wang, W., 2006. Extraction and separation of Cd and Te from cadmium
telluride photovoltaic manufacturing scrap. Prog. Photovolt: Res. Appl. 4, 363–371.
Fthenakis, V., Wang, W., 2010. System and method for separating tellurium from cad-
mium waste. Patent US7731920 B2.
Gerbinet, S., Belboom, S., Leonard, A., 2014. Life Cycle Analysis (LCA) of photovoltaic
panels: a review. Renew. Sust. Energ. Rev. 38, 747–753.
Gestione Servizi Energetici (GSE), 2016. Rapporto Statistico Solare Fotovoltaico.
<http://www.gse.it/it/Statistiche/RapportiStatistici/Pagine/default.aspx>.
Giacchetta, G., Leporini, M., Marchetti, B., 2016. Evaluation of the environmental ben-
efits of new high value process for the management of the end of life of thin film
photovoltaic modules. J. Clean. Prod. 51, 214–224.
Goozner, R.E., Long, M.O., Drinkard, W.F., 1999. Recycling of CdTe photovoltaic waste.
Patent US5997718 A.
Granata, G., Pagnanelli, F., Moscardini, E., Havlik, T., Toro, L., 2014. Recycling of pho-
tovoltaic panels by physical operations. Sol. Energy Mat. Sol. Cells 123, 239–248.
Grätzel, M., 2000. Perspectives for dye-sensitized nanocrystalline solar cells. Prog.
Photovolt.: Res. Appl. 8, 171–185.
Guangdong Xiandao Rare Material Co. Ltd., 2013. Recovery method of copper indium
gallium selenide thin-film solar panel, Patent CN103184338A.
Gustafsson, A.M.K., Foreman, M.R.St.J., Ekberg, C., 2014. Recycling of high purity sele-
nium from CIGS solar cell waste materials. Waste Manage. 34, 1775–1782.
Gustafsson, A.M.K., Steenari, B., Ekberg, C., 2015. Recycling of CIGS solar cell waste
materials: separation of copper, indium, and gallium by high-temperature chlorina-
tion reaction with ammonium chloride. Sep. Sci, Technol. 50, 2415–2425.
Huang, W., Shin, W.J., Wang, L., Sun, W., Tao, M., 2017. Strategy and technology to
recycle wafer-silicon solar modules. Sol. Energy 144, 22–31.
International Energy Agency (IEA), 2014. Technology Roadmap: Solar Photovoltaic
Energy.
International Energy Agency (IEA), 2018. End-of-Life Management of Photovoltaic
Panels: Trends in PV Module Recycling Technologies.
International Renewable Energy Agency (IRENA) and International Energy Agency (IEA),
2016. End-of-life management: solar photovoltaic panels.
International Renewable Energy Agency (IRENA), 2017. Renewable Power Generation
Costs in 2017.
Jung, B., Park, J., Seo, D., Park, N., 2016. Sustainable system for raw-metal recovery from
crystalline silicon solar panels: from noble-metal extraction to lead removal. ACS
Sustain. Chem. Eng. 4, 4079–4083.
Kang, S., Yoo, S., Lee, J., Boo, B., Ryu, H., 2012. Experimental investigations for recycling
of silicon and glass from waste photovoltaic modules. Renew. Energ. 47, 152–159.
Kim, Y., Lee, J., 2012. Dissolution of ethylene vinyl acetate in crystalline silicon PV
modules using ultrasonic irradiation and organic solvent. Sol. Energy Mat. Sol. Cells
98, 317–322.
Kim, H., Cha, K., Fthenakis, V.M., Sinha, P., Hur, T., 2014. Life cycle assessment of
cadmium telluride photovoltaic (CdTe PV) systems. Sol. Energy 103, 78–88.
Klugmann-Radziemska, E., Ostrowski, P., 2010. Chemical treatment of crystalline silicon
solar cells as a method of recovering pure silicon from photovoltaic modules. Renew.
Energ. 35, 1751–1759.
Kuroiwa, K., Ohura, S., Morisada, S., Ohto, K., Kawakita, H., Matsuo, Y., Fukuda, D.,
2014. Recovery of germanium from waste solar panels using ion-exchange membrane
and solvent extraction. Miner. Eng. 55, 181–185.
Kushiya, K., Ohshita, M., Tanaka, M., 2003. Development of recycling and reuse tech-
nologies for large-area Cu(1nGa)Se2-based thin-film modules. In: 3rd World
Conference on Photovoltaic Energy Conversion, pp. 1892–1895.
Kushiya, K., Tanaka, M., 2007. CIS based thin film solar cell module, method for pro-
ducing the same, and method for separating solar cell module. Patent EP
1830411 A1.
Kushiya, K., Okazawa, T., 2008. Method of recovering constituent member of CIS type
thin film solar cell module. Patent WO2008102457A1.
Latunussa, C.E.L., Ardente, F., Blengini, G.A., Mancini, L., 2016. Life Cycle Assessment of
an innovative recycling process for crystalline silicon photovoltaic panels. Sol. Energy
Mat. Sol. Cells 156, 101–111.
Lesson, A., Klotzbach, U., Kuntze, T., Loser, U., Palitzsch, W., 2015. Method for the se-
parating detachment of layers of a composite component formed from at least two
layers. Patent WO2015051977A1.
Lian, J., Huang, I., Lee, C., Chen, H., 2010. Method for recovery of copper, indium, gal-
lium and selenium. Patent US0329970 A1.
Malandrino, O., Sica, D., Testa, M., Supino, S., 2017. Policies and measures for sustain-
able management of solar panel end-of-life in Italy. Sustainability 9, 481.
Marwede, M., Berger, W., Schlummer, M., Mäurer, A., Reller, A., 2013. Recycling paths
for thin-film chalcogenide photovoltaic waste – current feasible processes. Renew.
Energ. 55, 220.
Marwede, M., Reller, A., 2012. Future recycling flows of tellurium from cadmium tell-
uride photovoltaic waste. Resources Conserv. Recy. 69, 35–49.
Masoumian, M., Kopacek, P., 2015. End-of-life management of photovoltaic modules.
IFAC-PapersOnLine 48–24, 162–167.
McDonald, N.C., Pearce, J.M., 2010. Producer responsibility and recycling solar photo-
voltaic modules. Energ. Pol. 38, 7041–7047.
Müller, A., Wambach, K., Alsema, E., 2005. Life cycle analysis of a solar module recycling
process. In: 20th European Photovoltaic Solar Energy Conference, Barcelona, Spain.
Nieland, S., Neuhaus, U., Pfaff, T., 2012. New approaches for component recycling of
crystalline solar modules. In: Electronics Goes Green, Berlin, Germany.
Nike*, 2018. Recupero apparecchiature elettriche ed elettroniche. <www.nikesrl.com>.
Pagnanelli, F., Moscardini, E., Abo, Atia T., Toro, L., 2016. Photovoltaic panel recycling:
from type-selective processes to flexible apparatus for simultaneous treatment of
F.C.S.M. Padoan et al.
different types. Min. Proc. Ext. Met. 125 (4), 221–227.
Pagnanelli, F., Moscardini, E., Granata, G., Abo Atia, T., Altimari, P., Havlik, T., Toro, L.,
2017. Physical and chemical treatment of end of life panels: an integrated automatic
approach viable for different photovoltaic technologies. Waste Manege. 59, 422–431.
Paiano, A., 2015. Photovoltaic waste assessment in Italy. Renew. Sust. Energ. Rev. 41,
99–112.
Palitzsch, W., Loser, U., 2011a. A new and intelligent de-metalization step of broken
silicon cells and silicon cell production waste in the recycling procedure of crystalline
Si modules. 37th IEEE Photovoltaic Specialists Conference 3269–3270.
Palitzsch, W., Loser, U., 2011b. Economic PV waste recycling solutions – results from R&D
and practice. 37th IEEE Photovoltaic Specialists Conference 628–631.
Peeters, J.R., Altamirano, D., Dewulf, W., Duflou, J.R., 2017. Forecasting the composition
of emerging waste streams with sensitivity analysis: a case study for photovoltaic
(PV) panels in Flanders. Resources Conserv. Recy. 120, 14–26.
Peng, J., Lu, L., Yang, H., 2013. Review on life cycle assessment of energy payback and
greenhouse gas emission of solar photovoltaic systems. Renew. Sust. Energ. Rev. 19,
255–274.
Ramos-Ruiz, A., Wilkening, J.V., Field, J.A., Sierra-Alvarez, R., 2017. Leaching of cad-
mium and tellurium from cadmium telluride (CdTe) thin-film solar panels under si-
mulated landfill conditions. J. Hazard. Mater. 336, 57–64.
Rathweg, C., 2011. System and process for recovery of cadmium telluride (CdTe). Patent
US 8048194 B2.
Raugei, M., Fullana-i-Palmer, P., Fthenakis, V., 2012. The energy return on energy in-
vestment (EROI) of photovoltaics: methodology and comparisons with fossil fuel life
cycles. Energ. Pol. 45, 576–582.
Reed, J.D., Wang, W., 2011. Film removal. Patent US0147230 A1.
Rocchetti, L., Beolchini, F., 2015. Recovery of valuable materials from end-of-life thin-
film photovoltaic panels: environmental impact assessment of different management
options. J. Clean. Prod. 89, 59–64.
Sander, K., Schilling S., Reinschmidt, J., Wambach, K., Schlenker, S., Müller, A., Springer,
J., Fouquet, D., Jelitte, A., Stryi-Hipp, G., Chrometzka, T., 2007. Study on the de-
velopment of a take back and recovery system for photovoltaic products, European
Photovoltaic Industry Association, German Solar Association.
Sasala, R.A., Bohland, J., Smigielski, K., 1996. Physical and chemical pathways for eco-
nomic recycling of cadmium telluride thin-film photovoltaic modules. Photovoltaic
Specialists Conference, Washington, USA, 865–868.
Savvilotidou, V., Antoniou, A., Gidarakos, E., 2017. Toxicity assessment and feasible
recycling process for amorphous silicon and CIS waste photovoltaic panels. Waste
Manage. 59, 394–402.
SENSE, 2008. LCA Analysis: University of Stuttgart, Wuerth Solar GmbH, Free Energy
Europe SA, Zentrum für Sonnenenergie- und Wasserstoffforschung, Fraunhofer
Gesellschaft zur Förderung der angewandten Forschung e.V., Ambiente Italia srl,
761
Solar Energy 177 (2019) 746–761
Fundacion Gaiker.
Shin, J., Park, J., Park, N., 2017. A method to recycle silicon wafer from end-of-life
photovoltaic module and solar panels by using recycled silicon wafers. Sol. Energy
Mat. Sol. Cells 162, 1–6.
SolarPower Europe, 2017. Global Market Outlook for Solar Power 2017-2021.
Steeghs, W., Water, S., 2016. Recovery of indium and gallium from flat panel displays and
photovoltaic (CIGS) modules. In: RECLAIM Workshop on Reclamation of key metals
from Energy Efficient Lighting, Flat Panel Displays and Photovoltaic modules,
Mechelen, Belgium.
Stötzel, E. Dr., Wambach, K. Dr., 2003. Method for separating the components of a la-
minated glazing. Patent EP0893250 B1.
Tammaro, M., Rimauro, J., Fiandra, V., Salluzzo, A., 2015. Thermal treatment of waste
photovoltaic module for recovery and recycling: experimental assessment of the
presence of metals in the gas emissions and in the ashes. Renew. Energ. 81, 103–112.
Tammaro, M., Salluzzo, A., Rimauro, J., Schiavo, S., Manzo, S., 2016. Experimental in-
vestigation 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. Energ. Mat. Sol. C. 141, 108–124.
Taylor, P., DeFilippo, M., 2012. Process to recycle end of life CdTe modules and manu-
facturing scrap. Patent US2012325676 A1.
Toro, L., Pagnanelli, F., Granata, G., Moscardini, E., 2016. Process for treating spent
photovoltaic panels. Patent EP 2997169 A1.
Wagner, U., Schmieder, F., 2009. Recycling method for thin layer solar cell modules.
Patent US 20090308535 A1.
Wan, H., 2004. Dye Sensitized Solar Cells. University of Alabama, Department of
Chemistry.
Wang, T., Hsiao, J., Du, C., 2012. Recycling of materials from silicon base solar cell
module. 38th IEEE Photovoltaic Specialists Conference 2355–2358.
Wang, W., Fthenakis, V., 2005. Kinetics study on separation of cadmium from tellurium in
acidic solution media using ion-exchange resins. J. Hazard. Mater. B125, 80–88.
Weinfurtner, J., 1996. Recycling insulating and/or laminated glass panels. Patent DE
4418573 C1.
Wu, L., Tian, W., Jiang, Z., 2005. Silicon based solar cells system with a hybrid PV
module. Sol. Energ. Mat. Sol. C. 87, 637–645.
Xu, Y., Li, J., Tan, Q., Peters, A.L., Yang, C., 2018. Global status of recycling waste solar
panels: a review. Waste Manage. 75, 450–458.
Yi, Y.K., Kim, H.S., Tran, T., Hong, S.K., Kim, M.J., 2014. Recovering valuable metals
from recycled photovoltaic modules. J. Air Waste Manage. Assoc. 64, 797–807.
Zhang, J., Lv, Fang, Ma, L., Yang, L., 2013. The status and trends of crystalline silicon
PVmodule recycling treatment methods in Europe and China. Adv. Mat. Res.
724–725, 200–204.