Available online at www.sciencedirect.

com

Solar Energy 96 (2013) 140–151

www.elsevier.com/locate/solener

Degradations of silicon photovoltaic modules: A literature review

Ababacar Ndiaye a,⇑, Abdérafi Charki b, Abdessamad Kobi b, Cheikh M.F. Kébé a,
Pape A. Ndiaye a, Vincent Sambou a
a
Centre International de Formation et de Recherche en Energie Solaire (CIFRES), Ecole Supérieure Polytechnique-UCAD, BP 5085, Dakar-Fann, Senegal
b
University of Angers-ISTIA-LASQUO, 62 Avenue Notre Dame du Lac, 49000 Angers, France

Received 29 October 2012; received in revised form 5 April 2013; accepted 8 July 2013
Available online 9 August 2013

Communicated by: Associate Editor Elias Stefanakos

Abstract

PV modules are often considered to be the most reliable component of a photovoltaic system. The alleged reliability has led to the
long warranty period for modules up to 25 years. Currently, failures resulting in module degradation are generally not considered
because of the difficulty of measuring the power of a single module in a PV system and the lack of feedback on the various degradation
modes of PV modules. It should be noted that consumers are becoming more and more interested in the reliability and lifetime of their
PV system considering economic issues. Reliability and lifetime of a PV system depend mainly on the energy performance of modules and
their different degradation modes. Accordingly, research must more and more focus on photovoltaic modules degradation. This paper
presents a review of different types of degradation found in literature in recent years. Thus, according to literature, corrosion and dis-
coloration of PV modules encapsulant are predominant degradation modes. Temperature and humidity are factors of PV modules deg-
radation in almost all identified degradation modes. However, despite the identification of PV modules degradation modes, it is still
difficult to study them in real conditions. Indeed, there must be long periods feedback experiences to study the frequency, speed of evo-
lution and impacts of various PV modules degradation modes on energy output. In this paper, models associated with the PV modules
degradation are presented. These models can help to overcome the long-term experiments obstacle in order to study PV modules deg-
radation under real conditions.
Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Photovoltaic module; Degradation; Degradation models

1. Introduction an important position in the renewable energy market.

Indeed, solar energy is used for industry, community as
Energy remains a key element for socio-economic well as individual needs. Over the past decade, the photo-
development of a society. It becomes more problematic voltaic (PV) market has experienced unprecedented
with soaring oil prices on the one hand and the environ- growth. Particularly in the over the past year, the photo-
mental consequences of using fossil fuels on the other voltaic market has reached a cumulative installed capacity
hand. Today different forms of renewable energy in vari- of roughly 40 GW world-wide, with an annual added
ous technologies offer greater flexibility and reliability sat- capacity of 16.6 GW (EPIA, 2011). For a reasonable
isfactory enough to reduce the energy deficit caused by energy need, a PV system can be effective only if installa-
the increased demand. Photovoltaic energy now holds tion is optimized and maintained. For this, it is important
to understand the reliability and lifetime of such a system.
⇑ Corresponding author. Tel.: +221 77 654 63 93; fax: +221 33 823 55 For the analysis of their technical and financial facilities,
74. investors demand increasingly clear answers about reli-
E-mail address: [email protected] (A. Ndiaye). ability and especially PV systems lifetime. Monitoring

0038-092X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.solener.2013.07.005
 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151 141

PV system degradation is important because a high degra- 2. Description of the photovoltaic module
dation rate leads directly to a loss of power output and
therefore to a reduction on the return on investment 2.1. PV Module components
(Short et al., 1995). The lack of accurate information
about degradation rate increases the financial risk (Jor- The photovoltaic module consists of PV cells, an
dan, 2011). The reliability and lifetime of a photovoltaic encapsulant, bypass diodes, connectors, a junction box, a
system depend mainly on the energy performance of the cable, a protective glass on the front face of the module
modules and their different degradation modes (Laronde and a glass or a polymer film (Tedlar generally) on the rear
et al., 2010, Charki et al., 2012). Information about PV side of the module, in Fig. 1. The assembly of these compo-
modules degradation has been available since the early nents can protect cells against different contacts and
1970s (Quintana et al., 2002). According to literature, against environmental conditions such as humidity. A
results from experimental tests are given. Different tech- junction of several cells through interconnection elements
niques and testing methods are used. Most of these results such as ribbons is shown in Fig. 2.
obtained from laboratory tests are not always accurate Materials used for PV modules encapsulating have to be
and consistent on the degradation rates of PV modules resistant to temperature changes and UV rays exposure.
announced. In 2002, the National Renewable Energy Modules must also resist mechanical stress that can occur
Laboratory (NREL) estimated the annual performance during transportation or installation.
degradation of monocrystalline and polycrystalline PV
modules at 0.7% (Osterwald et al., 2002). The LEEE- 2.2. Operating principle
TISO (Laboratory of Energy, Ecology and Economy
Solar-Ticino), test center of photovoltaic modules in Swit- A Solar cell is a device which converts photons in solar
zerland, stated that the power degradation rate of crystal- rays into direct-current (DC) and voltage. The associated
line silicon PV modules could go from 0.7% to 9.8% technology is called solar Photovoltaic (PV). A typical sil-
during the first exposure year and 0.7% to 4.9% during icon PV cell is a thin wafer consisting of a very thin layer of
the second one (LEEE, 2008). Jordan (Jordan and Kurtz, phosphorous-doped (N-type) silicon on top of a thicker
2011) states that in the 2000s, degradation rate measured layer of boron-doped (P-type) silicon. An electrical field
on individual modules was 5% on average. There are only is created near the top surface of the cell where these two
a few long-term studies on the degradation of PV modules materials are in contact (the P-N junction).
published. Skoczek et al. (2009) have measured the per- When the sunlight hits the semiconductor surface, an
formance of 204 field-aged crystalline Si based PV mod- electron springs up and is attracted towards the N-type
ules (53 module types). Exposing started in 1983 at the semiconductor material. This will cause more negatives in
Joint Research Center in north Italy with a moderate sub- the n-type and more positives in the P-type semiconduc-
tropical climate (10 to 35 °C, >90% RH). They find that tors, generating a higher flow of electricity. This is known
applying the performance warranty of 90% initial maxi- as the photovoltaic effect. Fig. 3 below shows the working
mum power after 10 years and 80% after 25 years only mechanism of a silicon solar cell.
17.6% of installed modules failed. These high performance
losses (>20%) are related to losses of the fill factor, caused 2.3. Photovoltaic technologies
by an increased series resistance. The moderate perfor-
mance losses (<20%) can be related to losses of the Many photovoltaic cells technologies are currently pres-
short-circuit current, caused by degradation of the optical ent on the market with different efficiencies (Laronde, 2009;
properties. The long term losses are determined to be Tiwari and Dubey, 2010). The most commonly used tech-
between 0.2% and 1.0% per annum. nologies are (Swanson, 2006; Green, 2006):
However, there is little information on PV modules deg-
radation modes in terms of frequency, speed of evolution
and degree of impact on module lifetime and reliability.
Research on photovoltaic modules is rather focused on
the race to develop new technologies without sufficient
experience feedback on already operational technologies
(Laronde, 2009; Tiwari and Dubey, 2010).
The present article proposes a literature review of PV
modules degradation (Osterwald et al., 2002; Morita
et al., 2003). The description of a photovoltaic module,
its assembly, its working principle and the main photovol-
taic technologies are presented in the first part. The main
types of PV module degradation from recent publications
are studied, and a presentation of some models associated
with the photovoltaic modules degradation is given. Fig. 1. Sectional view of PV module.
142 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151

the market remains below 20%. Taking into account this

level of performance, it is very important to be closely
interested in different possible reasons for its decline over
time. According to literature, only few works are dealing
with the degradation of photovoltaic modules. This study
is essential although photovoltaic technologies are evolving
quickly in order to avoid any source of users dissatisfaction
in the short and medium term.

3. Degradation of a photovoltaic module

3.1. Definition and main PV module degradations

Fig. 2. Interconnection of two cells.
Degradation is the gradual deterioration of the charac-
teristics of a component or of a system which may affect
its ability to operate within the limits of acceptability crite-
ria and which is caused by the operating conditions (Lan-
noy and Procaccia, 2005). A Photovoltaic module
degraded may keep doing its primary function which is to
generate electricity from sunlight, even if its use is no longer
optimal. However, the degraded state of the module can be
more problematic when the degradation exceeds a critical
threshold (Charki et al., 2012). According to Wohlgemuth
et al. (2005), manufacturers consider a PV module degraded
when its power reaches a level below 80% of its initial
Fig. 3. Silicon solar cell and its working mechanism. power. In the following step, the most frequent PV module
degradation types according to literature will be presented.
– Monocrystalline Silicon cells (c-Si). The PV module performance can be degraded due to several
– Polycrystalline Silicon cells (x-Si). factors such as: temperature, humidity, irradiation,
– Amorphous Silicon cells (a-Si). mechanical shock (Quintana et al., 2002; Wohlgemuth
– Cadmium Telluride cells (CdTe). et al., 2005; Osterwald and McMahon, 2009; Munoz
– Copper, Selenium and Indium cells (CIS),- Copper et al., 2011). Each one of these various named factors
Indium, Gallium and Selenium (CIGS). may induce one or more types of module degradation such
as (Vazquez and Ignacio, 2008; Munoz et al., 2011):
Other PV cell types are under development such as
organic PV cells which have the advantage of being recycla- – Corrosion.
ble but their performance is still very low, about 6,8% – Discoloration.
(Green et al., 2013). There are also, but still at the stage – Delamination.
of research, photovoltaic cells using nanotechnology – Breakage and cracking cells.
(nanotubes, nanowires, nanoantennas).
Table 1 gives the energy performance of PV cells for dif- NREL (Bosco, 2010; Wohlgemuth et al., 2010) gives a
ferent technologies. The table shows that, compared to the list (Table 2) of the major identified degradations of the
results determined in the laboratory, PV cells efficiency on crystalline silicon modules.

Table 1
Photovoltaic cells technologies and their yields (Green et al., 2013; Robert, 2013; Makrides et al., 2010; Laronde, 2009).
Technologies Cell Module
Laboratory efficiency Module on market
Monocrystalline silicon 25 ± 0.5 22.7 14–20
Polycrystalline silion 20.4 ± 0.5 16.2 11–15
Amorphous silicon 10.1 ± 0.3 10.4 5–9
CdTe 18.3 ± 0.5 8.3 –
CIS 19.3 13.5 9–11
CIGS 19.6 ± 0.6 10.3
Organic solar cells (thin-film) 10.±0.3 – –
Organic solar cells (submodule) 6.8 ± 0.2 – –
Nanowires 13.8 – –
 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151 143

Table 2 of PV modules edges. Osterwald et al. (2002) argues that

Modes of crystalline silicon modules degradation. the first and faster silicon PV module degradations are
Component Degradations caused by oxygen which is the main factor of the corrosion
Crystalline silicon PV module Interconnections broken of silicon junctions. Kemp (2005) showed that the moisture
Broken cells in the PV module is correlated with the rate of degradation,
Corrosion especially in hot and humid geographic zones such as
Delamination of the encapsulant
Miami in Florida. Due to the relatively fast water diffusion
Discoloration of the encapsulant
Broken glass in the ethylene vinyl acetate (EVA), infiltration of the mois-
Failure of the bypass diode ture into the module remains significant during its lifetime
Failure of the weld ribbons even if the module is constituted by a double glass struc-
ture. According to Kemp (2005), the only way to prevent
moisture penetration is to properly seal by using gaskets
3.2. Corrosion of PV module of low diffusivity containing a large quantity of desiccant.
Therefore, it could be more economical to focus on ways
The moisture that enters the module through the lami- to reduce the corrosion processes accelerated by the
nate edges causes corrosion (Kemp, 2005). The moisture moisture.
retention in the housing of the module increases the electri-
cal conductivity of the material. Indeed, the corrosion 3.3. Delamination of PV module
attacks the metallic connections of PV cells causing a loss
of performance by increasing leakage currents. Corrosion Adhesion loss called delamination occurs between the
also degrades the adhesion between cells and metallic encapsulating polymer and the cells or between cells and
frame. Fig. 4 shows a PV module affected by corrosion at the front glass. It represents a major problem because it
the edge and the junction box (Munoz et al., 2011). causes two effects: the increase of the light reflection and
Wohlgemuth and Kurtz (2011) studied the impact of of water penetration inside the module structure (Munoz
humidity and temperature on PV module degradation. et al., 2011).
They carried out accelerated testing named 85/85 Skoczek et al. (2008) studied the PV modules degrada-
(T = 85 °C/RH = 85%) according to IEC 61215 standard tion related to delamination module from tests based on
(IEC, 1987). They found out that corrosion appeared after IEC 61215 standard. Delamination is most severe when it
1000 h of exposure of PV module under 85 °C and 85% of occurs on the edges of the module because, in addition to
relative humidity. In (Wohlgemuth et al., 2005), tests are the degradation of power, it causes electrical risks for the
carried out on BP Solar modules from the feedback which module and the entire installation.
claim that the most common degradation mode was corro- Delamination is more frequent in hot and humid cli-
sion. This work has exploited the commercial database of mates. It causes moisture penetration in the module and
BP Solar, which collects all information from the technical therefore causes various chemical and physical degrada-
monitoring of their crystalline modules installed since 1994. tions such as metal corrosion of the module structure most
Other studies (Vazquez and Ignacio, 2008; Realini, frequently. A PV module with delamination at the edges is
2003; Quintana et al., 2002) also argue that corrosion shown in Fig. 5.
and discoloration are the predominant modes of photovol- Jansen and Delahoy (2003) claim that delamination may
taic modules degradation. Carlson et al. (2003), in collabo- be caused by salt accumulation and moisture penetration
ration with NREL, after tests on BP Solar modules, into the PV module. Moreover, they argue that interfacial
showed that the sodium contained in the glass which is connection may be etched by hydrofluoric acid formed by
reactive with moisture is a major factor of the corrosion fluorine and tin oxide contained in the PV module.

Fig. 4. PV module affected by corrosion of the edge (a) and the junction box (b).
144 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151

EVA films that causes a loss of the photovoltaic module

power (Berman and Faiman, 2007). However, for a
1000 Wm2 radiation, no change occurred in the wave-
lengths range between 280 nm and 380 nm after 500 h
exposure.
Wohlgemuth and Kurtz (2011) carried out UV tests on
PV modules under a 60 °C temperature; it was found that
the discoloration of the encapsulant only appears when
global UV irradiation reaches 15 kW h m2 in a wave-
length range included between 280 nm and 385 nm without
exceeding a 250 Wm2 exposure.
Fig. 5. PV Module with a delamination (Kaplanis and Kaplani, 2011). Osterwald et al. (2002) argue that the slow module
degradations in the long term are linearly correlated to
exposure PV modules under UV radiation. These last
3.4. Discoloration of PV module
years, most papers on degradation of silicon crystalline
PV modules were focused on the EVA degradation
Discoloration usually results in a degradation of the
(Kempe, 2006; Kempe et al., 2007; Kempe, 2010). Realini
encapsulant module, EVA (Ethylene Vinyl Acetate) or
(2003) carried out an experiment between 1982 and 2003,
adhesive material between the glass and the PV cells. Mod-
which allowed him to correlate the electrical characteristics
ule discoloration is a change in color of material which
of the module and discoloration of its encapsulant. Discol-
turns yellow and sometimes brown. It modifies transmit-
oration degrades the short-circuit current (Isc) of PV mod-
tance of light reaching PV cells and therefore the power
ule; this degradation of Isc may vary from 6% to 8% below
generated by the module is reduced.
the nominal value for a partial discoloration of the PV
Oreski and Wallner (2009) claim that the main causes of
module surface and from 10% to 13% for complete discol-
EVA degradation are UV rays combined with water under
oration. The Maximum power (Pmax) of the PV module is
temperatures higher than 50 °C. Discoloration may appear
also degraded by module discoloration as shown in Fig. 7.
in different and not neighboring zones of a PV module; this
may be due to the polymers used with different characteris-
tics. This could mean that discoloration comes from the 3.5. Breakages and cracks of PV modules
encapsulating polymer instead of the adherent element
such as EVA in general. But also that EVA is not spread Glass breakage is an important degradation factor of
in a similar way in all the areas of the modules using the PV modules. They occur in most of the cases during instal-
same polymers. Discolored PV cells are shown in Fig. 6. lation, maintenance, and especially during the transporta-
Kojima and Yanagisawa (2004) were interested in the tion of modules on their installation sites (Wohlgemuth
yellowing of the EVA used in PV modules. For this exper- and Kurtz, 2011). The broken modules or with cracks
iment, they exposed PV modules under artificial radiation. may keep functioning correctly. In Fig. 8, a cracked poly-
They are only interested in the UV rays contribution crystalline PV module which operated for five years with-
(wavelengths range is from 280 nm to 380 nm). When a out any noticeable power degradation is shown.
radiation of 4000 Wm2 is applied, a rapid PV cells degra- However, the risk of an electrical shock and of a moisture
dation appears with an increase in photosensitivity after infiltration increases.
400 h exposure and an increase in transmissivity between Breakages and cracks are usually followed by other deg-
280 nm and 380 nm. This is mainly due to the depression radation types such as corrosion, discoloration and delam-
of the UV absorber which must protect PV cells of photo- ination (Quintana et al., 2002). In order to save silicon and
degradation. Besides, there appears a weak yellowing at the reduce the manufacturing costs of PV cells, producers have

Fig. 6. Solar cells discolored (LEEE-TISO, 2008; Kaplanis and Kaplani, 2011).
 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151 145

Fig. 7. Current–voltage characteristics of PV module healthy (white) and discolored (yellow), (Realini, 2003). (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)

via the grounded framework when the insulation between

structure and active layers is not perfect and creating then
leakage currents (Schütze et al., 2011). From this phenom-
enon subsists a polarization that may degrade the electrical
characteristics of the photovoltaic cells. This phenomenon
known as PID (Potential Induced Degradation) is charac-
terized by the progressive performance deterioration of
crystalline silicon photovoltaic modules, due to the pres-
ence of an electrical current induced in the module (Pingel
et al., 2010; Berghold et al., 2010). Hacke et al. (2011)
showed that PID is more favorable in humid climates than
in hot and dry ones. Shütz et al. (2011) confirm this idea by
showing that leakage current increases with humidity. In
his study, a voltage is applied between the metallic struc-
Fig. 8. PV module with broken glass. ture and the PV module contact of 60 cells. The applied
voltage is a ramp from 600 V at sunrise to 0 V at sunset.
varied the thickness and the surface of cells in recent years.
The thickness of silicon PV cells decreased from 300 lm to 3.6.2. Hot spots
less than 200 lm and sometimes to less than 100 lm. In A hot spot is an area of a PV module that has a very
addition to the reduction in thickness of PV cells, the cell high temperature that could damage a cell or any other ele-
surface increased to 210 mm  210 mm (Dallas et al., ment of the module. The hot spot cause could be a variety
2007). This makes PV cells more fragile and more suscepti- of cell failures, including partial shadowing, cells mismatch
ble to breakage during handling (rolling and storage). It is or failures in the interconnection between cells (Molenb-
generally impossible to detect cracks on the already opera- roek et al., 1991). The short-circuit current and the open-
tional PV module to the naked eye. Detection can be done circuit voltage are imposed by the PV cell showing the low-
by using optical methods (Rueland et al., 2005). est electrical performance respectively in series and parallel
montage. In short circuit conditions, when a PV cell is
3.6. Other factors of PV modules degradation defective, its voltage is reversed and becomes equal and
opposite to the voltage of the other cells in series. This
3.6.1. Potential Induced Degradation (PID) defective cell becomes both a load for other cells and a
Individual modules in PV systems are often connected in place of a relatively high thermal dissipation constituting
series to increase the voltage of the system. The potential thus a hot spot (Orgeret, 1985; Cox III et al., 1982; Raus-
difference of this so formed chain can sometimes reach sev- chenbach and Mauder, 1972).
eral hundred volts (Schütze et al., 2011). In order to protect
people against electrical shocks, all metallic structures of 3.6.3. Bubbles
modules are often grounded. Because of this electrical volt- This type of degradation is similar to delamination but
age between PV modules and their structure, it is possible in this case, the loss of EVA adhesion only affects a small
that electrons in materials used for PV modules escape area and is combined with the surface swelling whose
146 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151

adhesion was degraded. The bubbles are generally due to

chemical reactions that emit gases trapped in the PV mod-
ule. When this happens on the back side of module, conges-
tion appears either in the encapsulating polymer or on the
back side of the module thus forming the bubbles. They
make it more difficult for cells heat dissipation, increasing
their overheating and reduce then their lifetime (Stephan
et al., 2004). In Fig. 9, a PV module with a large number
of bubbles on the back side is shown. They usually appear
in the center of the cell and may be due to poor adhesion of
the cell caused by the high temperature (Munoz et al.,
2011). Bubbles located on the module front side can pro- Fig. 10. Silicon PV modules: representativeness of degradation modes
duce a reduction of the radiation reaching the module; they (Changwoon et al., 2012).
cause a decoupling of the light and increase reflection
(Munoz et al., 2011).
where a (acorrosion and adiscoloration) and b (bcorrosion and bdis-
In Fig. 10 is given a review of the representativeness of
coloration) are the parameters of the degradation model.
the main silicon PV modules degradation modes presented
The overall degradation of the PV module is estimated
in this paper.
as:
4. The main degradation models of a PV module Y
n
DPVmodule ðtÞ ¼ 1  ð1  Di ðtÞÞ ð4Þ
i¼1
In general, the degradation of photovoltaic module is
assessed by measuring the power, and therefore the power where n is the number of degradation modes.
loss during its lifetime compared to its initial power. Cur- Parameters of degradation model a and b are deter-
rently, the degradation models of PV modules are still mined by accelerated testing. For corrosion, the test
few and developments are still to be done. parameters are temperature and humidity that are the main
factors of corrosion.
4.1. The model of Pan At least two damp heat tests such as 85/85 (T = 85 °C/
HR = 85%) should be performed to determine acorrosion
In (Pan et al., 2011), Pan (Jet Propulsion Laboratory) and bcorrosion (Oreski and Wallner, 2005). Temperatures
proposes a degradation model of the PV module output taken during tests shall not exceed the temperature of the
power given by: PV modules technological limit equal to 120 °C according
to Kern (Kern, 1999). In (Wohlgemuth et al., 2005) have
DðtÞ ¼ 1  expðb  ta Þ ð1Þ been studied degradation of a polycrystalline PV module
where a and b are parameters of the degradation model. from model of Pan (Pan et al., 2011); they have determined
They are determined from accelerated testing (Charki a and b parameters from damp heat tests 85/85. They find
et al., 2013; Laronde, 2012; Laronde, 2009). a = 3.0868 and b = 5762.1012 that led to trace the evolu-
The constants a and b depend on the degradation mode tion of degradation D (t) in Fig. 11.
considered. Thus, Eqs. (2) and (3) give respectively the deg- This degradation might be caused by corrosion whose
radation model due to corrosion and discoloration: factors are temperature and humidity which are the expo-
sure parameters in these tests. The model of Pan is very
Dcorrosion ðtÞ ¼ 1  expðbcorrosion  tacorrosion Þ ð2Þ
dependent on the tests determining the degradation param-
adiscoloration
Ddiscoloration ðtÞ ¼ 1  expðbdiscoloration  t Þ ð3Þ eters a and b. Thus, it will be constrained by the accuracy
and testing time.
This approach presents some problems because it never
occurs in real conditions. Indeed, when PV module has
high humidity, temperature is low. Failure modes occur-
ring after long term 85/85 testing are not being observed
in the field. Depending on acceleration factors better per-
formance in 85/85 may not be an indication of better long
term performance in the field.

4.2. The exponential model

Vazquez and Ignacio (2008) have developed an analyti-

cal model for estimating degradation of PV module power.
Fig. 9. Bubbles on the back side of a PV module (Munoz et al., 2011). They posed the following assumptions:
 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151 147

This model gives an indication of the PV module degra-

dation during its lifetime. However, many assumptions are
used and therefore the results obtained do not always
reflect the reality relating to the real operating conditions.

4.3. Model degradation by UV stress

Photodegradation caused by UV radiation is the main

degradation reason of materials exposed to direct sunlight.
It is clear that UV radiation is a major degradation factor
for photovoltaic modules especially in their discoloration
(Kojima and Yanagisawa, 2004; Oreski and Wallner,
Fig. 11. Degradation of polycrystalline silicon PV module under test 85/ 2009; Wohlgemuth and Kurtz, 2011). The total dose of
85 (Laronde, 2011; Charki et al., 2013). UV radiation can be considered as the number of photons
absorbed by the material degrading and that will cause its
chemical change. For PV module, this degradation is
– The module power (P) is considered as reference for reflected by the change of the encapsulating module trans-
evaluating the performance. mittance that induces a decrease in the PV module current–
– Degradation of PV module is assessed relatively to its voltage characteristic. This decrease is quantified by the rel-
initial power (P0). ative variation of module short-circuit current representing
the degradation factor given by (Zimmerman, 2008):
The PV module power measured at a given time follows I sc ðEÞ
a Gaussian distribution. According to (Reis et al., 2002; DðEÞ ¼ ð9Þ
I sc ðE ¼ 0Þ
Sakamoto and Oshiro, 2003), the probability density is
given by: where Isc (E) is the short-circuit current of the PV module
  and E the dose of ultra-violet (UV).
1 1 p  l2 Zimmerman (2008) proposes the degradation expression
PðpÞ ¼ pffiffiffiffiffiffiffiffi exp  ð5Þ
2pr 2 r factor given by the following equation:
where P is the power module, l is its mean value and r its R kmax
SRðkÞT cmx ðkÞP ðkÞT ðk; EÞdk
standard deviation. DðEÞ ¼ Rkmin
kmax
ð10Þ
Average power of PV module decreases linearly in time. kmin
SRðkÞT cmx ðkÞP ðkÞT 0 ðkÞdk
Thus, it can be calculated by (Osterwald et al., 2003; Mar-
where SR(k) is the spectral response of PV cell, T(k, E) and
ion and Adelstein, 2003; Raghuraman et al., 2006):
Tcmx(k) respectively represent the transmittances of the
lðtÞ ¼ P 0  At ð6Þ encapsulant and the glass slide of PV module, T0(k) is the
where P0 is the average power at t = 0 (i.e. the nominal transmittance of not irradiated PV cell. P(k) is the spectral
module power), A is a parameter which reflects annual de- power density of the sun. [kmin, kmax] represents the integra-
crease of module power and t is time in years. Of course, tion interval for the wavelength range in which the spectral
validity of Eq. (6) is restricted by time (t) lower than P0/ response of the PV cell is not zero.
A. Another limitation comes from the assumption accord- For a Photodegradation T(k, E)/T0(k) in the encapsu-
ing to which A is constant in time. lant less than 70% in the range [kmin, kmax], the transmit-
From Eq. (6), it is simple to demonstrate, for two suc- tance can be written:
cessive years, the power module is degraded at a fixed rate T ðk; EÞ ¼ T 0 ðkÞ½1  bcmx ðkÞ lnð1 þ acmx EÞ ð11Þ
relative to its initial power:
lðnÞ  lðn  1Þ A
¼ ð7Þ
P0 P0
A/P0 (year1) is the annual degradation rate. This linear
degradation model is limited by the insufficiency of experi-
ences confirming this assumption in literature. From stud-
ies on electronic components degradation, we can consider
the assumption according to which degradation rate is
exponential (Chuang et al., 1997; Xie and Pecht, 2003):
lðtÞ ¼ P 0 eat ð8Þ
Fig. 12 shows PV module degradation as a function of Fig. 12. Exponential degradation of PV module power (Vazquez and
time from the exponential model. Ignacio, 2008).
148 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151

where acmx and bcmx are parameters of material used for PV – The rate constant varies linearly with irradiance in the
cell (Zimmerman, 2008). range of irradiances considered. This may not be the
The relationship between UV dose E and exposure time case for all polymers or at high irradiances such as sev-
t in the solar spectrum P (k) is: eral times the maximum daytime irradiance near the
E ¼ ct ð12Þ equator. In such case, a power law may apply to irradi-
ance (Pickett et al., 2008a).
where – The activation energy Ea of the overall temperature
Z 400 dependent process (in the presence or absence of light)
C¼ T cmx ðkÞP ðkÞdk ð13Þ is constant over the considered temperature range.
0
– Activation energies for weathering under accelerated
The integral extends up to 400 nm that represents a conditions are only known for a few materials. When
practical limit to UV photodegradation. the activation energy of a particular material is not
Combining Eqs. (10), (12), and (13) and using the mean available, it needs to be estimated from published activa-
value theorem, Eq. (14) obtained to estimate the UV degra- tion energies for similar materials, or determined from
dation of PV module is given by: in-house experiments. For instance, although the activa-
DðtÞ ¼ 1  n lnð1 þ acmx ctÞ ð14Þ tion energy of the photochemical yellowing of EVA has
not been published, it can be estimated from the Ea of
where n = bcmx(k), k e [kmin, kmax] and [kmin, kmax] repre- the photochemical yellowing established for a range of
sents the integration interval for the wavelength range aromatic polymers (Pickett et al., 2008b).
where the spectral response of PV cell is not zero.
This model presents a major constraint. In fact, its use The Arrhenius-based model presents some limits.
requires knowledge of the intrinsic characteristics of the Indeed, the Arrhenius equation can be used to quantify
materials used for the production of PV cells such as acmx, the effect of varying temperature and irradiance on the rate
bcmx, SR(k), T(k,E), Tcmx(k) and T0(k). The measurement of a property change. However it cannot provide a com-
of these characteristics requires a rigorous instrumentation plete picture of the long-term degradation of PV modules,
without which the model accuracy is compromised. as other stress factors or combination of stresses are
involved. These include moisture (inducing physical and
4.4. Model degradation by temperature stress chemical processes and generating mechanical stress in
combination with temperature), temperature cycling (gen-
The Arrhenius equation is one of the most commonly erating thermomechanical stress), electricity production
reported models for the temperature dependence of degra- (inducing electrical and electrochemical stresses), and other
dation processes. For temperature dependent processes, the externally applied stresses (wind, hail, airborne pollutants,
Arrhenius law in Eq. (15) may be applied to predict the marine air, air blown sand, etc.).
increase in rate resulting from an increase in temperature,
as given by (Cocca et al., 2011). 4.5. Model degradation by temperature and humidity stress
Ea
RT
K ¼ Ae ð15Þ
The acceleration model with the capacity to take into
K1 Ea 1
ð 1Þ account temperature (T) and relative humidity (HR) is
¼ e R T 2 T 1 ¼ AF T ð16Þ
K2 the Peck model (Escobar and Meeker, 2006), which is
where K is the rate constant of the process, A is an Arrhe- defined by the following relation:
nius pre-exponential factor, Ea is the apparent activation  
n Ea
energy, R is the gas constant, T1 and T2 are the sample tem- s ¼ A  ðHRÞ  exp  ð17Þ
k:T
peratures, AFT is the acceleration factor for thermal degra-
dation (ratio of rate constants). where Ea is the effective activation energy of the degrada-
To obtain these equations, several assumptions and tion process; k is the Boltzmann’s constant
approximations are necessary: (=8.617.105 eV/°K), and A and n are two constants
dependent on the failure mode.
– The rate constant K applies to a single change in prop- According to Charki et al. (2013), the Wiener process
erty or performance. For instance, if a material degrades can be estimated as follows:
through color change and loss in mechanical properties, a0
ðxi ; Z ij Þ ¼ 1  expðb0  ðrðxi ; cÞtij Þ Þ ð18Þ
both degradation processes will most likely exhibit dis-
tinct rate constants. In this case, x1 and x2 correspond respectively to relative
– The overall temperature dependent process (either in the humidity (RH) and module temperature (T); x01 and x02
presence or absence of light) leading to a change in per- respectively correspond to known relative humidity (RH0)
formance follows an Arrhenius dependency. and temperature (T0) in the reference conditions.
 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151 149

Fig. 13. Equivalent temperature versus Ea (a) and equivalent relative humidity versus n (b) for the four cities studied.

The acceleration factor is expressed as (Charki et al., modules degradation identified in the literature are corro-
2013): sion, discoloration, delamination and breakage. However,
    according to literature, corrosion and discoloration are
HR Ea 1 1
rðx1 ; x2 ; cÞ ¼ expðn ln   ð19Þ the predominant modes of PV module degradation. Envi-
HR0 k T T0
ronmental parameters such as temperature, humidity and
where c = (n, Ea). UV radiation are the main factors of PV module degrada-
We consider that a0 = a for each stress level. The rela- tion. It should be noted that the modeling of different deg-
tion (18) becomes: radation types is still poorly studied in literature.
Degradation models would make it possible to study the
ðxi ; Z ij Þ ¼ 1  expðbðxi ; cÞtaij Þ ð20Þ
     PV module behavior according to different degradation
where bðxi ; cÞ ¼ b0 exp a  n  ln RH RH
0  aEk a T1  T10 factors. The lack of experience feedback over long periods
could be reduced by degradation models of modules. Iden-
If the acceleration factor is defined by relation (19), it is tified models do not target one particular degradation
possible to estimate an equivalent temperature, Teq, which mode but rather the whole degradation of the module
represents the degradation that would have occurred if the power. Thus, it is difficult to study the impact (speed, fre-
module had been aged for the same length of time but at a con- quency) of a particular degradation mode. The work pre-
stant temperature. The equivalent temperature Teq can be cal- sented in this paper has underlined the degradation
culated using the following relation (Kurtz et al., 2011): modes and factors to consider in experiments over long
  Z t2  
Ea 1 Ea periods. Development of models associated with different
exp ¼ exp dt ð21Þ degradation modes of PV modules is an interesting
k:T eq t2  t1 t1 k  T module ðtÞ
research field to improve the knowledge of photovoltaic
where t is time, Tmodule(t) is the time-dependent module modules behavior during their life cycle.
temperature, and t1 and t2 are the integration start and
end times.
References
Using the same methodology below, the equivalent of
relative humidity Heq can be estimated as Berghold, J., et al., 2010. Potential Induced Degradation of solar cells and
Z t2 panels. In: 25th EUPVSEC, pp. 3753–3759.
n 1 n
ðH eq Þ ¼ ðH ðtÞÞ dt ð22Þ Berman, D., Faiman, D., 2007. EVA browning and the time-dependence
t2  t1 t1 of I–V curve parameters on PV modules with and without mirror-
enhancement in a desert environment. Solar Energy Materials & Solar
where t is time, H(t) is the time-dependent environmental Cells 45 (4), 401–412.
relative humidity, and t1 and t2 are the integration start Bosco, N., 2010. Reliability Concerns Associated with PV Technologies.
and end times. <http://www.nrel.gov/pv/performance_reliability/pdfs/
The module temperature Tmodule(t) and the relative failure_references.pdf>.
humidity H(t) are simulated hour by hour for a period of Carlson, D.E., Romerol, R., Willing, F., Meakin, D., Gonzalez, L.,
Murphyl, R., Moutinho, H.R., Al-Jassim, M., 2003. Corrosion effects
50 years, for at four cities: Paris (France), Oslo (Norway), in thin-film photovoltaic modules. Progress Photovoltaics: Research
Madrid (Spain) and Tamanrasset (Algeria). The equivalent and Applications 11, 377–386. http://dx.doi.org/10.1002/pip.500.
temperature Teq and the equivalent relative humidity Heq Changwoon, H., Nochang, P., Jaeseong, J., 2012. Lifetime prediction of
obtained versus Ea and n respectively using relations (21) silicon PV module ribbon wire in three local weathers. Photovoltaic
and (22) are plotted in Fig. 13. Module Reliability Workshop.
Charki, A., Laronde, R., Bigaud, D., 2012. The Time-Variant Degrada-
tion of a Photovoltaic System. Journal of Solar Energy Engineering
5. Conclusion
135 (2). http://dx.doi.org/10.1115/1.4007771.
Charki, A., Laronde, R., Bigaud, D., 2013. Accelerated degradation
In this paper, a review of the different modes of PV mod- testing of a photovoltaic module. J. Photon. Energy 3 (1), 033099.
ule degradation has been presented. The main modes of PV http://dx.doi.org/10.1117/1.JPE.3.033099.
150 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151

Chuang, S., Ishibashi, A., Kijima, S., Nakayama, N., Ukita, M., Laronde, R., 2011. Fiabilité et durabilité d’un système complexe dédié aux
Taniguchi, S., 1997. Kinetic model for degradation of lightemitting énergies renouvelables: Application à un système photovoltaı̈que.
diodes. IEEE Journal of Quantum electronics 33, 970–979. PhD, University of Angers (France).
Cocca, M., D’Arienzo, L., D’Orazio, L., 2011. Effects of Different Laronde, R., Charki, A., Bigaud, D., 2010. Reliability of photovoltaic
Artificial Agings on Structure and Properties of Whatman Paper modules based on climatic measurement data. International Journal of
Samples. Materials Science 2011, ID 863083, 7 pages. doi:10.5402/ Metrology and Quality Engineering 1 (01), 45–49. http://dx.doi.org/
2011/863083.. 10.1051/ijmqe/2010012.
Cox III, C.H., Silversmith, D.J., Mountain, R.W., 1982. Reduction of Laronde, R., Charki, A., Bigaud, D., 2012. Lifetime Estimation of a
photovoltaic cell reverse breakdown by a peripheral by pass diode. In: Photovoltaic Module Subjected to Corrosion Due to Damp Heat
6th IEEE photovoltaic specialist’s conference. Testing. Journal of Solar Energy Engineering 135 (2). http://
Dallas, W., Polupan, O., Ostapenko, S., 2007. Resonance ultrasonic dx.doi.org/10.1115/1.4023101.
vibrations for crack detection in photovoltaic silicon wafers. Measure- LEEE-TISO, 2008. <http://leee.dct.supsi.ch/PV/Results/
ment Science and Technology 18, 852–858. Tested_modules.htm>.
EPIA: European Photovoltaic Industry Association, 2011. Global Market Makrides, G., Zinsser, B., Norton, M., Georghiou, G.E., Schubert, M.,
Outlook for Photovoltaics until 2015. Werner, J.H., 2010. Potential of photovoltaic systems in countries with
Escobar, L.A., Meeker, W.Q., 2006. A review of accelerated test models. high solar irradiation. Renewable and Sustainable Energy Reviews 14
Statistical Science 21 (4), 552–577. (2), 754–762.
Green, M.A., Emery, K., Hishikawa, Y., Warta, W., Dunlop, E.D., 2013. Marion, B., Adelstein, J., 2003. Long-term performance of the SERF PV
Solar cell efficiency tables (Version 41). Progress in Photovoltaics: systems. NCPV and Solar Program Review Meeting.
Research and Applications 21, 1–11. Molenbroek, E., Waddington, D.W., Emery, K.A., 1991. Hot spot
Green, M.A., 2006. Recent developments and future prospects for third susceptibility and testing of PV modules. In: Conference Record of the
generation and other advanced cells. IEEE Photovoltaic Energy 22th IEEE, vol. 1. Las Vegas, 1991, pp. 547–552.
Conversion, Vol. 1, May 2006, pp. 15–19. Morita, K., Inoue T., Kato T., Tsuda I., Hishikawa Y., 2003. Degradation
Hacke, and al., 2011. System Voltage Potential-Induced Degradation factor analysis of crystalline-Si PV modules through long-term field
Mechanisms in PV Modules and Methods for Test. In: 37th IEEE exposure test. In: 3rd World Conference on Photovoltaic Energy
PVSC. Conversion, Osaka, Japan, pp. 1948–1951.
IEC: International Electrotechnical Commission, 1987. Standard Munoz, M.A., Alonso-Garcia, M.C., Nieves, V., Chenlo, F., 2011. Early
IEC61215: Crystalline Silicon Terrestrial Photovoltaic (PV) Modules. degradation of silicon PV modules and guaranty conditions. Solar
Design Qualification and Type Approval IEC Central Office, Geneva, Energy 85, 2264–2274.
Switzerland. Oreski, G., Wallner, G.M., 2009. Evaluation of the aging behavior of
Jansen, K.W., Delahoy, A.E., 2003. A laboratory technique for the ethylene copolymer films for solar applications under accelerated
evaluation of electrochemical transparent conductive oxide delamina- weathering conditions. Solar Energy 83, 1040–1047.
tion from glass substrates. Thin Solid Films 423, 153–160. Oreski, G., Wallner, G.M., 2005. Aging mechanisms of polymeric films for
Jordan, D.C., 2011. Methods for analysis of outdoor performance data. PV encapsulation. Solar Energy 79, 612–617.
NREL PV Module reliability Workshop, Golden CO, USA. <http:// Orgeret, M., 1985. Les piles solaires. Paris.
www.nrel.gov/pv/pvmrw.html>. Osterwald, C.R., McMahon, T.J., 2009. History of accelerated and
Jordan, D.C., Kurtz, S.R., 2011. Photovoltaic Degradation Rates-an qualification testing of terrestrial photovoltaic modules: a literature
Analytical Review. Progress in Photovoltaics: Research and Applica- review. Progress in Photovoltaics: Research and Applications 17 (1),
tions, doi:10.1002/pip.1182. 11–33.
Kaplanis, S., Kaplani, E., 2011. Energy performance and degradation Osterwald, C.R., Benner, J.P., Pruett, J., Anderberg, A., Rummeland, S.,
over 20 years performance of BP c-Si PV modules. Simulation Ottoson, L., 2003. Degradation in weathered crystalline-silicon PV
Modeling Practice and Theory 19, 1201–1211. modules apparently caused by UV radiation. In: 3rd World Confer-
Kemp, M.D., 2005. Control of moisture ingress into photovoltaic ence on Photovoltaic Energy Conversion, Osaka, Japan, pp. 2911–
modules. In: 31st IEEE Photovoltaic Specialists Conference and 2915.
Exhibition, Lake Buena Vista, Florida. Osterwald, C.R., Anderberg, A., Rummel, S., Ottoson, L., 2002. Degra-
Kempe, M.D., 2006. Modelling the moisture ingress into photovoltaic dation Analysis of Weathered Crystalline-Silicon PV Modules. In: 29th
modules. Solar Energy Materials & Solar Cells 90, 2720–2738. IEEE Photovoltaic Specialists Conference, New Orleans, Louisiana.
Kempe, M.D., Jorgensen, G.J., Terwilliger, G.J., McMahon, T.J., Pan, R., Kuitche, J., Tamizhmani, G., 2011. Degradation analysis of solar
Kennedy, C.E., Borek, T.T., 2007. Acetic acid production and glass photovoltaic modules: Influence of environmental factor. In: Proc.
transition concerns with ethylene-vinyl acetate used in photovoltaic Annual Reliability and Maintainability Symposium.
devices. Solar Energy Materials & Solar Cells 91, 315–329. Pickett, J.E. et al., 2008a. Polymer Degradation and Stability 93 (2008),
Kempe, M.D., 2010. Ultraviolet test and evaluation methods for 1597–1606.
encapsulants of photovoltaic modules. Solar Energy Materials & Pickett, J.E. et al., 2008b. Polymer Degradation and Stability 93 (2008),
Solar Cells 94, 246–253. 684–691.
Kern, G., 1999. SunSineTM300: Manufacture of an AC Photovoltaic Pingel, S., et al., 2010. Potential Induced Degradation of solar cells and
Module. Final Report, Phases I & II NREL/SR-520-26085. panels. In: 35th IEEE PVSC.
Kojima, T., Yanagisawa, T., 2004. The evaluation of accelerated test for Quintana, M.A., King, D.L., McMahon, T.J., Osterwald, C.R., 2002.
degradation a stacked a-Si solar cell and EVA films. Solar Energy Commonly observed degradation in field-aged photovoltaic modules.
Materials & Solar Cells 81 (1), 119–123. In: Proc. 29th IEEE Photovoltaic Specialists Conference, pp. 1436–
Kurtz, S., Whitfield, K., Tamizhmani, G., Koehl, M., Miller, D., Joyce, J., 1439.
Wohlgemuth, J., Bosco, N., Kempe, M., Zgonena, T., 2011. Evalu- Raghuraman, B., Laksman, V., Kuitche, J., Shisler, W., Tamizhani, G.,
ation of high-temperature exposure of photovoltaic modules. Progress Kapoor, H., 2006. An overview of SMUDs outdoor photovoltaic test
in Photovoltaics: Research and Applications. program at Arizona State University. In: 4th IEEE World Conference
Lannoy, A., Procaccia, H., 2005. Evaluation et maı̂trise du vieillissement on Photovoltaic Energy Conversion, Hawaii, USA.
industriel. Edition Lavoisier. Rauschenbach, H., Mauder, E.E., 1972. Breakdown phenomena in reverse
Laronde, R., 2009. Solutions Photovoltaı̈ques dans le Bâtiment. GINGER biased silicon solar cells. In: Proceeding 9th photovoltaic specialist’s
CATED. conference.
 A. Ndiaye et al. / Solar Energy 96 (2013) 140–151 151

Realini, A., 2003. Mean Time before Failure of Photovoltaic Modules. modules to the IEC61215 test sequence. Solar Energy Materials &
Final Report (MTBF Project), Federal Office for Education and Solar Cells 92, 1593–1604.
Science Tech. Rep., BBW 99.0579. Stephan, M., Thomas, K., Wolfgang, J., Hurbert, F., 2004. Quality testing
Reis, A.M., Coleman, N.T., Marshall, M.W., Lehman, P.A., Chamber- for PV modules according to standards and performance control for
lain, C.E., 2002. Comparison of PV module performance before and supporting manufacturing. In: Proceedings of the 19th EU PVSEC,
after 11-years of field exposure. In: Proceedings of the 29th IEEE Paris.
Photovoltaic Specialists Conference, New Orleans, Louisiana, USA. Swanson, R.M., 2006. A vision for crystalline silicon photovoltaics.
Robert F., 2013. Solar Energy Performance of Nanowire Solar Cells on Progress in Photovoltaics: Research and Applications 14, 443–453.
the Rise. Science 18 January 2013: 263. Tiwari, G., Dubey, S., 2010. Fundamentals of photovoltaic modules and
Rueland, E., Herguth, A., Trummer, A., Wansleben, S., Fath, P., 2005. their applications. Royal Society of Chemistry.
Optical l-crack detection in combination with stability testing for Vazquez, M., Ignacio, R.S., 2008. Photovoltaic module reliability model
inline inspection of wafers and cells. In: Proceedings of 20th EU based on field degradation studies. Progress in Photovoltaics: Research
PVSEC, Barcelona, pp. 3242–3245. and Applications 16, 419–433.
Sakamoto, S., Oshiro, T., 2003. Field test results on the stability of crystalline Wohlgemuth, J.H., Kurtz, S., 2011. Reliability Testing Beyond Qualifi-
silicon photovoltaic modules manufactured in the 1990s. In: 3rd World cation as a Key Component in Photovoltaic’s Progress Toward Grid
Conference on Photovoltaic Energy Conversion, Osaka, pp. 1888–1891. Parity. In: IEEE International Reliability Physics Symposium Mon-
Schütze, M., Junghänel, M., Friedrichs, O., Wichtendahl, R., Scherff, M., terey, California, April 10–14.
Müller, J., Wawer, P., 2011. Investigations Of Potential Induced Wohlgemuth, J.H., Cunningham, D.W., Nguyen, A.M., Kelly, G., Amin,
Degradation Of Silicon Photovoltaic Modules. In: 26th European D., 2010. Failure Modes of Crystalline Silicon Modules. In: Proceed-
Photovoltaic Solar Energy Conference, Hamburg, Germany. ings of PV Module Reliability Workshop. <http://www1.eere.energy.-
Schütze, M., and al., 2011. Laboratory Study of Potential Induced gov/solar/pv_module_reliability_workshop_2010.html>.
Degradation of Silicon Photovoltaic Modules. In: 37th IEEE PVSC. Wohlgemuth, J.H., Cunningham, D.W., Nguyen, A.M., Miller, J., 2005.
Short, W., Packey, D.J., Holt, T., 1995. A manual for the economic Long Term Reliability of PV Modules. In: Proc. 20th European
valuation of energy efficiency and renewable energy technologies. Photovoltaic Solar Energy Conference, pp. 1942–1946.
Report NREL/TP-462-5173. Xie, J., Pecht, M., 2003. Reliability prediction modelling of semiconductor
Skoczek, A., Sample, T., Dunlop, E.D., 2009. The results of performance light emitting device. IEEE Transactions on Device and Materials
measurements of field-aged crystalline silicon photovoltaic modules. Reliability 3, 218–222.
Progress in Photovoltaics: Research and Applications 17, 227–240. Zimmerman, C.G., 2008. Time dependent degradation of photovoltaic
Skoczek, A., Sample, T., Dunlop, E.D., Ossenbrink, H.A., 2008. Electrical modules by ultraviolet light. Applied Physics Letter 92, 241110.
performance results from physical stress testing of commercial PV