This page covers the degradation of photovoltaic cells and modules in the field: the long-term power loss rate and its warranty risk, light-induced degradation of crystalline silicon (boron-oxygen LID and light- and elevated-temperature-induced degradation, LeTID) in PERC cells, and potential-induced degradation (PID) of crystalline silicon and thin-film modules by shunting, polarization, delamination and corrosion. The cell physics is on the solar cell pages of this vault (Optoelectronics, Solar Panels), moisture and corrosion models on the Environmental and Chemical page and the life statistics on the Failure Rate page; inverters and system wiring are outside the scope of the vault.

1. Why It Happens

Trigger: Years of outdoor exposure to light, heat, humidity and system voltage (Jordan et al.); illumination or current injection at elevated temperature for LID and LeTID (Fokuhl et al.); a high negative or positive voltage between the cells and the grounded frame or glass, in humidity (Hacke; Hacke et al.).
Mechanism, LID: Boron-oxygen complexes form under carrier injection and degrade the bulk lifetime; they regenerate at 65 $^{\circ} C$ or more through hydrogen passivation (Fokuhl et al.).
Mechanism, LeTID: A further, slower bulk degradation at elevated temperature, mainly above about 50 $^{\circ} C$ , with temporary recovery; hydrogen is believed to play a key role (Fokuhl et al.).
Mechanism, PID-shunting: Sodium ions driven by the voltage from the glass drift to the cells and diffuse along stacking faults, which shunts the junction (Hacke; Hacke et al.).
Mechanism, other PID modes: Charge in the passivation layers repels or attracts carriers (polarization PID, recovered by light within 5 $h$ ); delamination with sodium migration and electrochemical products; TCO corrosion and carrier loss in thin-film modules (Hacke; Hacke et al.).
What sets the rate: Climate, with more fill-factor loss in hot and humid climates (Jordan et al.); module temperature and injection (Fokuhl et al.); humidity, voltage and mounting, which change the PID acceleration by up to about 50 (Hacke).
Failure criterion: Power below the warranted level at 25 years (Jordan et al.); a power loss beyond 5 per cent in a qualification test (IEC practice, background knowledge).

2. Where to Look

Location: The bulk of the silicon wafer for LID and LeTID (Fokuhl et al.); the cells near the frame and the encapsulant-glass and encapsulant-cell interfaces for PID and delamination (Hacke et al.; Hacke); the TCO near edge clamps of thin-film modules (Hacke).
Signature: A loss mostly in the short-circuit current for most climates, and in the fill factor in hot and humid climates and after 20 years (Jordan et al.); a fast first loss within days of installation followed by slower LeTID and partial recovery in winter (Fokuhl et al.); shunted, dark cells in electroluminescence for PID-shunting (Hacke; background knowledge).
Confirm with: Indoor current injection with BO stabilization before the LeTID test (Fokuhl et al.); PID chamber tests at 60 or 85 $^{\circ} C$ and 85 per cent RH at system voltage, with leakage-current monitoring (Hacke; Hacke et al.); dark I-V for the saturation current and shunt resistance (Hacke).
Look-alikes: Polarization PID, which recovers under light, against permanent shunting (Hacke); BO LID counted again in a LeTID test without stabilization (Fokuhl et al.); finger detachment, which added up to 1.3 per cent of loss in an insulated module (Fokuhl et al.).

3. Prevention

Design: Use the measured uncertainty of the degradation rate, not only its mean, for warranty planning (Jordan et al.); limit the negative voltage to ground of the string or use positive grounding where PID-shunting dominates (background knowledge).
Material and process: Adjust firing profiles and use thinner wafers against LeTID (Fokuhl et al.); low-sodium or barrier glass and encapsulants with low ionic conductivity against PID (Hacke et al.; background knowledge).
Operation: Keep modules ventilated, since insulated modules ran 4.8 to 5.9 $K$ warmer on average and lost more (Fokuhl et al.).
Screening and test: PID tests at 60 $^{\circ} C$ , 85 per cent RH, system voltage and 96 $h$ for conventional c-Si, or 85 $^{\circ} C$ for higher severity (Hacke); LeTID tests at 75 $^{\circ} C$ after BO stabilization (Fokuhl et al.); include light in future PID tests (Hacke).

Field Degradation Rate

Name	Formula	Related Formulae	Variables/Constants
Field degradation rate of PV modules	$P_{ma x} (t) = P_{ma x, 0} (1 - R_{d} t)$	$P_{ma x} (t) \approx P_{ma x, 0} (1 - R_{d})^{t} (compound form)$	$P_{ma x} (t) : ma x im u m p o w er a f t er t y e a r s$ $P_{ma x, 0} : ini t ia l ma x im u m p o w er$ $R_{d} : d e g r a d a t i o n r a t e p er y e a r; J or d an e t a l . (2128 r a t es f r o m o v er 100 p u b l i c a t i o n s in 32 co u n t r i es) : m e d ian 0.5 %/ yr, a v er a g e 0.8 %/ yr, ab o u t 80 p er ce n t o f t h e r a t es b e l o w 1 %/ yr, m os t m o d u l es d e g r a d e b y ab o u t 0.5 %/ yr, m or e t han ha l f o f t h e r a t es co m e f r o m I - V m e a s u r e m e n t s an d P_{ma x} i s t h e m os t co mm o n p a r am e t er, t h er e i s n o s t an d a r d m e t h o d, P_{ma x} l oss f o l l o w e d I_{sc} l oss in 9 o f 12 N R E L m o d u l es an d F F l oss in 3, h o t an d h u mi d c l ima t es s h o w m or e F F l oss, F F l oss i s s ma l l in t h e f i r s t tw o d ec a d es b u t m u c h l a r g er a f t er 20 y e a r s, t hin - f i l m r a t es ha v e a pp r o a c h e d c - S i, an d F i r s t S o l a r r eco mm e n d s 0.7 %/ yr f or h o t an d 0.5 %/ yr f or o t h er c l ima t es$ $(co m p u t e d : a f t er 25 y e a r s, 87.5 p er ce n t a t 0.5 %/ yr, 82.5 p er ce n t a t 0.7 %/ yr an d 80.0 p er ce n t a t 0.8 %/ yr w i t h t h e l in e a r f or m, an d 88.2, 83.9 an d 81.8 p er ce n t w i t h t h e co m p o u n d f or m)$
Warranty risk from an uncertain degradation rate	$P_{d e f} = Pr [P_{ma x} (25) < P_{w a r r}], R_{d} \sim N (\overset{ˉ}{R}_{d}, σ_{R}^{2})$	$P_{ma x} (25) = P_{ma x, 0} (1 - 25 R_{d})$	$P_{d e f} : p r o babi l i t y o f f a l l in g b e l o w t h e w a r r an t e d p o w er P_{w a r r} a t y e a r 25$ $\overset{ˉ}{R}_{d} : m e an d e g r a d a t i o n r a t e$ $σ_{R} : i t s u n cer t ain t y$ $N : n or ma l d i s t r ib u t i o n (M o n t e C a r l o); J or d an e t a l . : a g ain s t a 2010 ma x im u m w a r r an t y, R_{d} = 0.8 \pm 0.2 %/ yr g i v es a 24 p er ce n t an d 0.8 \pm 1.0 %/ yr a 57 p er ce n t p r o babi l i t y o f d e f a u l t, so t h e u n cer t ain t y o f t h e r a t e ma tt er s a s m u c h a s i t s m e an$ $(co m p u t e d : n o n e; t h e w a r r an t y c u r v e u se d w a s n o t in t h e s l i d es r e a d, so t h e tw o p r o babi l i t i es c ann o t b e r e p r o d u ce d w i t h a s in g l e l in e a r t h r es h o l d (m y c h ec k))$

Light-Induced Degradation

Name	Formula	Related Formulae	Variables/Constants
Boron-oxygen and light- and elevated-temperature-induced degradation	$Δ P_{L I D} = Δ P_{B O} + Δ P_{L e T I D}$	$k_{d e g}, k_{r e g} \propto e^{- E_{a} / (k_{B} T)} (Arrhenius type)$	$Δ P_{L I D} : t o t a l l i g h t - in d u ce d p o w er l oss$ $Δ P_{B O} : l oss f r o m b or o n - o x y g e n d e f ec t s$ $Δ P_{L e T I D} : l oss f r o m l i g h t - an d e l e v a t e d - t e m p er a t u r e - in d u ce d d e g r a d a t i o n$ $k_{d e g} an d k_{r e g} : d e g r a d a t i o n an d r e g e n er a t i o n r a t es$ $E_{a} : a c t i v a t i o n e n er g y (n o v a l u e g i v e n)$ $k_{B} : B o l t z mann co n s t an t (8.617 \times 1 0^{- 5} eV / K)$ $T : ab so l u t e t e m p er a t u r e; F o k u h l e t a l . (P E R C m o d u l es) : B O L I D 1.4 p er ce n t (1.1 t o 1.5) a f t er 96 h o f c u r r e n t inj ec t i o n a t 25^{\circ} C, s t abi l i z e d w i t hin 24 t o 48 h, L e T I D a t 75^{\circ} C an d M P P 1.2 t o 1.5 p er ce n t m or e a f t er B O s t abi l i z a t i o n b u t 2.7 t o 2.9 p er ce n t w i t h o u t i t, 1.5 an d 1.9 p er ce n t a t 85^{\circ} C a f t er 262 an d 360 h, - 3.8 an d - 4.2 p er ce n t f or m u l t i - cr y s t a l l in e m o d u l es a t 85^{\circ} C, L e T I D main l y ab o v e ab o u t 50^{\circ} C an d t h er ma l l y p oss ib l e w i t h o u t inj ec t i o n ab o v e 130^{\circ} C, c a r r i er inj ec t i o n s p ee d s u p t h e t e m p or a r y r eco v er y b y ab o u t f i v e t im es, 35 t o 54 p er ce n t o f B O d e f ec t s r e g e n er a t e d in f i v e w ee k s a t 75^{\circ} C, an d o u t d oor m o n o - P E R C l osses o f 2.3 an d 1.7 p er ce n t a f t er o n e m o n t h an d 2.2 an d 1.8 p er ce n t a f t er 15 m o n t h s f or n o n - in s u l a t e d an d in s u l a t e d m o d u l es; h y d r o g e n i s b e l i e v e d t o pl a y a k ey r o l e in L e T I D$ $(co m p u t e d : t h e 1.5 p er ce n t d i f f er e n ce b e tw ee n s t abi l i z e d an d u n s t abi l i z e d L e T I D t es t s e q u a l s t h e B O l oss o f 1.4 p er ce n t w i t hin t h e s p r e a d, so an u n s t abi l i z e d t es t co u n t s B O L I D tw i ce (m y r e a d in g))$

Potential-Induced Degradation

Name	Formula	Related Formulae	Variables/Constants
Chamber-to-field scaling of potential-induced degradation	$t_{F} = \frac{Q _{F, c hamb er}}{( d Q / d t ) _{f i e l d}}$	$J_{l e ak} = \frac{I _{l e ak}}{A _{m o d u l e}}$	$t_{F} : f i e l d t im e t o t h e s am e d e g r a d a t i o n$ $Q_{F, c hamb er} : c ha r g e t r an s f er r e d f r o m t h e f r am e or g l a ss t o t h e ce l l s in t h e c hamb er t es t, (d Q / d t)_{f i e l d} : c ha r g e - t r an s f er r a t e in t h e f i e l d$ $J_{l e ak} : l e ak a g e c u r r e n t d e n s i t y$ $I_{l e ak} : l e ak a g e c u r r e n t a t t h e sy s t e m v o l t a g e$ $A_{m o d u l e} : m o d u l e a r e a; H a c k e (P V Q A T) : c - S i m o d u l es l os t 4.3 p er ce n t a f t er - 600 V, 60^{\circ} C, 85 p er ce n t R H an d 96 h, F l or i d a r e pl i c a s s h o w e d n o l oss a f t er 5 y e a r s an d l oss a f t er 6, 85^{\circ} C an d 95 p er ce n t R H i s ab o u t f o u r t im es m or e s t r ess f u l t han 60^{\circ} C, m o d u l es a r e g r a d e d f r o m t y p e A (f i e l d f ai l u r e) t o F, t hin - f i l m sc a l in g o f 46 f i e l d d a y s p er c hamb er d a y, m o u n t in g c han g es t h e f a c t or b y u p t o ab o u t 50, an d t h e m e t h o d f ai l s w i t h m o i s t u r e in g r ess or r eco v er y; H a c k e e t a l . (f o u r 60 - ce l l m o d u l es, - 1000 V a t 72^{\circ} C an d 95 p er ce n t R H u p t o 292 h) : c u r r e n t s o f 4.9, 1.8, 16 an d 0.027 nA/c m^{2}, d e l amina t i o n o n l y in t h e m o d u l e w i t h t h e hi g h es t c u r r e n t, a f t er 156 h, l e ak a g e in E V A r i ses b y ha l f a d ec a d e f r o m 30 t o 95 p er ce n t R H, an d c hamb er c u r r e n t t r an s f er ab o u t t e n t im es t h e f i e l d ma x ima in F l or i d a an d G r an C ana r ia$ $(co m p u t e d : 1000 h a t 85/85 t im es 46 i s 5.2 y e a r s an d 96 h i s 0.50 y e a r s; ha l f a d ec a d e o f l e ak a g e o v er a 3.2 t im es R H in cr e a se i s an R H e x p o n e n t o f ab o u t 1.0; 4.4 \times 1 0^{- 4} A a g ain s t mi d 1 \times 1 0^{- 5} A i s ab o u t 9 t im es)$
Saturation current and voltage loss in CIGS PID	$Δ V_{oc} \approx \frac{n k _{B} T}{q} ln \frac{J _{0, 2}}{J _{0, 1}}$	$R_{s h} ↓ in the second stage$	$Δ V_{oc} : l oss o f o p e n - c i r c u i t v o l t a g e$ $n : i d e a l i t y f a c t or$ $k_{B} : B o l t z mann co n s t an t (8.617 \times 1 0^{- 5} eV / K)$ $T : ab so l u t e t e m p er a t u r e$ $q : e l e m e n t a r y c ha r g e$ $J_{0, 1} an d J_{0, 2} : s a t u r a t i o n c u r r e n t d e n s i t i es b e f or e an d a f t er s t r ess$ $R_{s h} : s h u n t r es i s t an ce (d i o d e r e l a t i o n, ba c k g r o u n d k n o w l e d g e); H a c k e (P V Q A T) : C I GS P I D f i r s t l o w er s t h e i o ni z e d c a r r i er d e n s i t y, t h e b u i l t - in p o t e n t ia l an d r ai ses J_{0}, t h e n l o w er s t h e s h u n t r es i s t an ce, w i t h J_{0} = 7.59 \times 1 0^{- 8}, 1.61 \times 1 0^{- 7} an d 6.56 \times 1 0^{- 3} A/c m^{2}, R_{s h} = 3164, 2290 an d 187 Ω c m^{2} an d N_{A} / N_{A 0} = 1.00, 0.23 an d 1.28 \times 1 0^{- 5} o v er t h r ee r o u n d s; P E R C u n d er + 1000 V w i t h U V A a t 60^{\circ} C an d 85 p er ce n t R H : P_{ma x} / P_{ma x, 0} = 0.915 a t 0.15 C/c m^{2}; p o l a r i z a t i o n P I D r eco v er s f u l l y u n d er l i g h t w i t hin 5 h$ $(co m p u t e d : J_{0} g r e w 8.6 \times 1 0^{4} t im es, w hi c h a l o n e cos t s ab o u t 0.29 V o f V_{oc} f or n = 1 a t 25^{\circ} C)$

Important Questions

Ques: Do PV modules degrade by 0.5 or 0.8 per cent per year?

Both numbers come from the same NREL survey of 2128 degradation rates. Jordan et al. give a median of 0.5 per cent per year and an average of 0.8 per cent per year, with about 80 per cent of the rates below 1 per cent per year. The difference comes from a tail of fast-degrading systems, which pulls the mean above the median, so 0.5 per cent per year is the more likely rate for a typical module, while 0.8 per cent per year is the more conservative planning value (my reading). Over 25 years the linear form gives 87.5 and 80.0 per cent of the initial power (computed). Climate matters: First Solar recommends 0.7 per cent per year for hot climates, and hot and humid sites show more fill-factor loss, especially after 20 years. Reference: D. Jordan, J. Wohlgemuth and S. Kurtz, ‘Technology and Climate Trends in PV Module Degradation’, 27th European PV Solar Energy Conference, Frankfurt (2012), NREL/PR-5200-56690 (slides, fetch-tool summary).

Ques: Why does the uncertainty of the degradation rate matter for a warranty?

Because the warranty fails on the tail of the distribution, not on the average module. Jordan et al. ran a Monte Carlo analysis for year 25 against a 2010 maximum warranty and found that a rate of 0.8 per cent per year with an uncertainty of 0.2 gives a 24 per cent probability of default, while the same mean with an uncertainty of 1.0 gives 57 per cent. Two products with the same mean rate can therefore carry very different financial risk, and an accurate measurement of the rate is worth as much as a lower mean. The slides also note that there is no standard method for determining the rate, and most rates come from I-V measurements of the maximum power, so the measurement method itself adds uncertainty. Reference: D. Jordan, J. Wohlgemuth and S. Kurtz, ‘Technology and Climate Trends in PV Module Degradation’, 27th European PV Solar Energy Conference, Frankfurt (2012), NREL/PR-5200-56690 (slides, fetch-tool summary); Failure Rate and the Bathtub Curve page of this vault.

Ques: How do boron-oxygen LID and LeTID differ, and why must a LeTID test start with a stabilized module?

Boron-oxygen LID is a fast loss under light or current injection, 1.4 per cent of power in the PERC modules of Fokuhl et al. after 96 $h$ at 25 $^{\circ} C$ , which stabilizes within one to two days and regenerates at 65 $^{\circ} C$ or more. LeTID is slower, appears mainly above about 50 $^{\circ} C$ and recovers partly over time; at 75 $^{\circ} C$ and maximum power point it added 1.2 to 1.5 per cent after BO stabilization. Without stabilization the same test showed 2.7 to 2.9 per cent, 1.5 per cent more, about the BO loss itself (computed), so the BO loss is counted as LeTID. Hydrogen is believed to drive LeTID, and firing profiles and thinner wafers reduce it. Outdoors, mono-PERC modules lost about 2 per cent within 15 months. Reference: E. Fokuhl, D. Philipp, G. Muelhoefer and P. Gebhardt, ‘LID and LETID evolution of PV modules during outdoor operation and indoor tests’, EPJ Photovoltaics 12, 9 (2021) (fetch-tool summary).

Ques: What causes PID-shunting in crystalline silicon modules, and how is it tested?

A high negative voltage of the cells against the grounded frame drives sodium ions from the glass through the encapsulant to the cells, where they diffuse along stacking faults and shunt the junction. Hacke reports a 4.3 per cent power loss after 96 $h$ at -600 $V$ , 60 $^{\circ} C$ and 85 per cent RH, and judges this level adequate to qualify conventional modules for the continental USA, while 85 $^{\circ} C$ is about four times more stressful. Florida replicas showed no loss after five years but loss after six. Humidity raises the leakage: in EVA it grows by half a decade from 30 to 95 per cent RH, an exponent of about 1 (computed). Hacke et al. also found that the module with the highest current, 16 $nA/c m^{2}$ , delaminated. Reference: P. Hacke (NREL), ‘Testing and analysis for multiple PID mechanisms and stresses’, PVQAT TG3 presentation (NIST-hosted slides; year not stated) (fetch-tool summary); P. Hacke, M. Kempe, J. Wohlgemuth, J. Li and Y.-C. Shen, ‘Potential-Induced Degradation-Delamination Mode in Crystalline Silicon Modules’, NREL/CP-5J00-67256 (2016 workshop, report dated March 2018) (fetch-tool summary).

Ques: How many years in the field does a 1000 h PID chamber test represent?

Hacke scales chamber tests to the field by the transferred charge, $t_{F} = Q_{F, c hamb er} / (d Q / d t)_{f i e l d}$ . For thin-film modules his slides give 46 field days per chamber day, so 1000 $h$ at 85 $^{\circ} C$ and 85 per cent RH equals 5.2 years and 96 $h$ only 0.50 years (computed), in line with his statements of 5.3 years and under half a year. The factor depends strongly on the mounting, by up to about 50, because the field leakage depends on the insulation and wetting of the module, and the scaling fails when moisture enters the module or when the damage recovers. Hacke et al. likewise measured chamber leakage about ten times the field maxima in Florida and Gran Canaria. The scaling is therefore a guide, not a lifetime calculation. Reference: P. Hacke (NREL), ‘Testing and analysis for multiple PID mechanisms and stresses’, PVQAT TG3 presentation (NIST-hosted slides; year not stated) (fetch-tool summary); P. Hacke, M. Kempe, J. Wohlgemuth, J. Li and Y.-C. Shen, ‘Potential-Induced Degradation-Delamination Mode in Crystalline Silicon Modules’, NREL/CP-5J00-67256 (2016 workshop, report dated March 2018) (fetch-tool summary).

Ques: How does PID reduce the voltage of a CIGS module?

Hacke describes two stages. First, the ionized carrier concentration falls, which lowers the built-in potential and raises the saturation current density; in his data $N_{A} / N_{A 0}$ fell to 0.23 and then to $1.28 \times 1 0^{- 5}$ , while $J_{0}$ rose from $7.59 \times 1 0^{- 8}$ to $1.61 \times 1 0^{- 7}$ and then to $6.56 \times 1 0^{- 3}$ $A/c m^{2}$ . Second, the shunt resistance falls, from 3164 to 187 $Ω c m^{2}$ . The rise of $J_{0}$ by $8.6 \times 1 0^{4}$ times alone costs about 0.29 $V$ of open-circuit voltage for an ideality factor of 1 at 25 $^{\circ} C$ (computed), from $Δ V_{oc} \approx (n k_{B} T / q) ln (J_{0, 2} / J_{0, 1})$ . In thin-film modules, TCO corrosion near edge clamps is a further PID mode, with 7 to 14 chamber days at -1000 $V$ and 85/85 equal to 1.1 years in the field. Reference: P. Hacke (NREL), ‘Testing and analysis for multiple PID mechanisms and stresses’, PVQAT TG3 presentation (NIST-hosted slides; year not stated) (fetch-tool summary); solar cell pages of this vault (Optoelectronics, Solar Panels).

Quartz 5

Explorer

Solar Cells and Modules (Crystalline Silicon and Thin Film)