photovoltaic solar modules electrical properties evolution under ...

1ère Conférence Franco-Syrienne sur les énergies renouvelables

Damas, 24-28 Octobre 2010

PHOTOVOLTAIC SOLAR MODULES ELECTRICAL PROPERTIES EVOLUTION UNDER EXTREME STRESS Jihad SIDAWIa, Nadine ABBOUDb, Georges JELIANb, Roland HABCHIb, Mario ELTAHCHI and Chafic SALAMEb,c a

Faculty of Engineering III, Lebanese University, Hadath, Lebanon LPA, Faculty of Sciences II, Lebanese University, B.P 90656 Jdeidet El Mten, Lebanon c CNRSL, National Council for Scientific Research, Beirut, Lebanon b

[email protected]

The purpose of this paper is to discuss the effect of electric reverse stress currents on the performance of photovoltaic solar modules. The effect of a reverse introduced current as a function of time is studied on the I-V and C-V characteristics and parameters which were extracted and analyzed using numerical analysis based on a reliable double exponential model. The effect of an introduced reverse current for different periods simulated the effect of accumulated extreme reverse currents which may arise in solar cells and modules due to different reasons, causing dramatic changes in the shunt resistance as well as other characteristics, mainly when the time of the current application exceeded a certain limit. The paper contributes to the research on the damaging effects of reverse currents on the normal operation of the solar cells and modules. Keywords Solar cell/module, Recombination current shunt

Photovoltaic,

Current–voltage

characteristics,

Capacitance–voltage

characteristics,

1. Introduction Photovoltaic solar cell research has accelerated in the past years to comply with the increasing demand and need of a better cell / module quality. Solar cell manufacturing introduced a variety of materials and processing methods to enhance the performance of the solar cell. The fracture mechanics approach to material science and microstructures assumes that any product already contains a kind of defect such as material, fabrication or even design defect. These defects play a major role in determining the life time and performance of such a product. A solar cell is treated as a diode of larger area silicon p-n junction forward bias with a photovoltage. This photovoltage is created from the dislocation of the electrons as a result of incident photons within the junction or diode. Any disturbance of this electron flow, mainly through the silicon crystal /crystals or the cell junction, is due to what is considered as material defects, such as grain boundaries, dislocations, or any other inhomogeneity in the microstructure, will have a large impact on a part or on the over whole performance of the solar cell. Extensive research has proved by experimental evidence that the existence of defects of any type and anywhere in the solar cell will play a degrading factor and influence the dark current voltage (I-V) characteristics of the p-n junction (O. Breitenstein et al., 2009). Shading or partial shading on a solar module or array has demonstrated to influence the I-V characteristics and reduce the total output of the solar cell. This problem may become more serious when the shaded cells get reverse biased, leading to a high resistance diode, which will get overheated when the difference in illumination is high enough. The overheating of the diode will eventually lead to serious damage in the photovoltaic cell (Ramabadran and Mathur, 2009).

CFSER 2010

Page 1



Moreover, recent research used to identify and locate material defects in monocrystalline and multicrystalline solar cells has shown that, when an external current is fed into the solar cell in the direction of the forward or reverse bias, the solar cell begins to emit light all around except on the grain boundaries contained in the multicrystalline. Defects in the material such as grain boundaries become immediately recognizable as dark areas since the light-emitting efficiency is lower in these areas than in the areas free from defects. Solar cells in the defective areas cannot produce a photo-current when illuminated, and consequently cannot generate any electricity. The analysis of the reverse bias region of solar cells also has an outstanding technological significance. Under normal use, solar cells operate with forward bias. Yet if a shadow falls across a solar module, for instance falling leaves or shadows cast by a tree, the solar cells located in the shadow are suddenly under reverse bias. If they cannot resist the voltage, then a strong current flows due to the effect of the electric breakthrough. In this event, charge carriers typically are accelerated in a strong electrical field produced inside the diode within the solar cell. That generates new pairs of charge carriers and can thus lead to a high, and in the worst case uncontrollable current that can destroy the solar cell and the entire module. This breakthrough current arises only on certain defects and very locally. Moreover, its principal occurrence does not depend on the surface preparation of the solar cells with acidic or alkaline treatment which improves the solar cell's absorption of light (anti-reflection coating) (Lausch et al., 2009). The reverse current is also referred to certain fault currents and environmental conditions in photovoltaic arrays with several strings connected in parallel to form a PV array with a directcurrent output equal to the sum of the PV string outputs. The panel circuitry can be then referred to as the PV generator. Standards, such as the Australian Standard AS/NZ5033 “Installation of PV Arrays”, recommend fuses to protect both cabling and PV modules in case of the occurrence of these fault conditions. The relationship between acceptable reverse current levels and exposure durations of reverse currents on PV modules are compared with trip current of fuses and typical time delays experienced with fuse tripping (Calais et al., 2008). To prevent the damage which may arise as a result of a reverse current, different national electric codes requires a disconnect switch to be provided on the DC side of the PV-inverter. Additional switches may be used between the PV panels and the AC grid since there exist a number of components, including the inverter, the interconnecting cables, wires, over-current protection, surge protection, grounding equipment and means for switching and disconnecting different parts of the circuit. To investigate more about the effect of a reverse current on the performance and efficiency of a PV solar cell/ module, an extensive analysis of the dark I-V characteristic of solar cells is needed. This work focuses on the experimental quantitative measurement of the effect of an introduced reverse current, due to the above mentioned reasons, on I-V dark current characteristics due to material damage, by applying a high level of electric stress in the opposite electron flow direction of the solar cell/ module to simulate a damaging reverse bias affecting the module over a certain period of time. The results and discussion offered aimed to contribute to material, performance and efficiency of photovoltaic solar cell/ module research. 2. Experimental A commercial photovoltaic modules 6x6 (36 cells) 3.8 v and 85 mA were collected from the local consumer market. These modules are often used in the industry to power small devices such as lamps, mobile phones and small calculators. In this experimental set up, the PV modules were placed in dark in order to study their behavior without illumination. To simulate the reverse current that occur in solar modules when partially shaded, the modules underwent several amounts of laboratory induced reverse current through the pn junction. For instance, a reverse current of 40 mA was used to collect the data presented in this paper. Our results

CFSER 2010

Page 2



contains both experimental electrical characterization along with some mathematical calculation used to determine some of the junction parameters through an equivalent circuit model of the module. The general mathematical description of the output characteristics for a PV cell has been studied for over the past four decades. The equivalent circuit of the general model consists of a photo current, a diode, a parallel resistor expressing a leakage current, and a series resistor describing an internal resistance to the current flow (Tsai et al., 2008). An even more precise mathematical description of a solar cell, which is called the double exponential model, as shown in Figure 1, is derived from the physical behavior of solar cells constructed from polycrystalline silicon. This model is composed of a two ideal diodes, a series resistance Rs and a parallel shunt resistance Rsh. It considers the calculation of both series and shunt resistances along with the junction ideality factor A and the components of the diode diffusion I01 and recombination I02 saturation currents.

Figure 1 Idealized equivalent circuit of a double exponential model (Salame and Habchi, 2008)

Experimentally collected I-V curves were introduced into specially designed software that performs numerical calculations based on the double exponential model of a p-n junction formulated by the following equation (Salame and Habchi, 2008):

Where: I: Total cell current (A) V: Cell voltage (V) I01: Diffusion reverse saturation current (A) I02: Recombination reverse saturation current (A) Rs: Series resistance (Ω) Rsh: Shunt resistance (Ω) A: Ideality factor (1-2) VT: Thermal voltage VT = kT/q T: Cell working temperature (K) Q: Elementary charge = 1.602177x10-19 (C) k: Boltzmann's constant = 1.380662x10-23 (J/K)

3. Results and Discussion The reverse current of 40 mA was applied for several periods of time. At each step, the current was interrupted to perform an I-V characteristic. For a module or array of PV cells, the shape of the I-V curve does not change from that of a cell. However, it is scaled based on the number of cells connected in series and in parallel.

CFSER 2010

Page 3



Log I-V results are presented in Figure 2 before and after applying the 40 mA stress for 10, 20 and 30 minutes respectively. The reverse current is maintained through a programmable current source, this specific current was chosen because it plays as a threshold current below which similar preliminary results, carried out on the same type of cells in our laboratories, occurred in a slower manner and above which sudden failure occurred. This technique is used to prevent sudden breakdown of the module by an avalanche phenomenon. The linear part of the I-V characteristics is seen to shift up as the time of stress is increased. This led to a decrease in the diffusion voltage illustrated in Figure 3. The diffusion voltage decrease is time dependant. The declination of the curve is steep up to the 10-15 minutes interval, after which the curve demonstrates some stability for 5 minutes up to the 20 minutes time interval and again it shows steep declination till the end of the experimental procedure when damage of the module occurred. 40 mA Without stress 10 min 20 min 30 min

0.1 0.01 1E-3 1E-4 1E-5 1E-6 1E-7 1

2

3

4

5

6

7

8

Voltage (V)

Figure 2 Log I vs V by collapsed time

5.5

40 mA 5.0

4.5

4.0

3.5

3.0 0

5

10

15

20

25

30

Stress time (min)

CFSER 2010

Page 4



Figure 3 Diffusion voltage Vd vs time As the inverse of the slope of the linear part of the I-V curve in the reverse bias is a recognized method to measure the shunt resistance of a PV cell calculated as (Keithley Instruments, 2007): Rsh = ∆Vreverse bias/ ∆ Ireverse bias, Figure 5 shows the extensive increase in the slope of the linear part of the curve, thus inversely affecting the shunt resistance values, mainly after 20 minutes of stress application. These results are in full conformity with the above discussed results of shunt resistance. Checking the validity of the double exponential model with experimental data and simple slope calculation proved the accuracy and validity of both methods. 40 mA

Without stress 10 min 20 min 30 min

1E-3

1E-4 0

10

20

30

40

50

Reverse voltage (V)

Figure 4 Log reverse I vs reverse V

The calculated shunt resistance in Figure 5, which represents the loss due to surface leakage along the edge of the cell or due to crystal defects, was dramatically affected by the reverse current applied during the time collapse of 10 -15 minutes by dropping from 3x105 to a few ohms, after which this drop seems to stabilize about a minimum value till the end of the experiment. Ideally, the shunt parallel resistance should be infinite (Rsh = ∞) and the series resistance 0 ohms (Rs = 0) under normal operating conditions and ambient temperatures. The consequences of the shunt resistance Rsh drop caused serious damage, clearly observed on the surface of the PV cell at the end of the experiment, as black burned flakes was a result of extensive overheating of the solar cell.

CFSER 2010

Page 5



300000

40 mA

250000

200000

150000

100000

50000

0

5

10

15

20

25

30

Stress time (min)

Figure 5 Shunt resistance Rsh vs stress time The overheating took place because of charges, having a certain energy level, get excited and by gaining extra energy pass through the pn junction. Neutral traps are originally present at the structure surface and in the semiconductor bulk and are facing phonons and lattice vibrations caused by the reverse current and temperature elevation (Salame and Habchi, 2008). The overheating effect to damage took only 10 -15 minutes to occur, which in turn is so short to be noticed by a PV user if not having any type of protection to the PV system against reverse currents before the DC connection to the inverter. Taking into consideration the ideality factor A which indicated a stabilized value of about 1, achieved in ideal cells, till just before the 15 minutes time interval where it started to increase to a value of 2.5, which is considered relatively high, providing additional information about the permanent damage in the PV solar cell and module due to the degradation in the junction internal structure The diffusion current is caused by the diffusion of carriers across the junction of the PV cell. In ideal operating conditions of equilibrium, the net current from the PV cell is zero, because of the balancing effect of the diffusion, drift and the recombination currents. Upon applying a current in the reverse direction of electron flow, this balance is disturbed and due to overheating of the p-n junction and expanding the junction width, diffusion as well as recombination currents will after a certain time start to increase. The behavior of the recombination and diffusion currents as a function of time are documented in figures 6 and 7 respectively. Stable linearity at the zero level up to the previously mentioned 15-20 minutes after which steep increase occurred. Above the 30 minutes time interval there seems that the double exponential model is not valid anymore. Similar conclusions have been observed when MOSFET devices have been exposed to an elevated temperature of above 250 o C, due to device improper functionality and accelerating degradation, resulting in irreversible and induced permanent structural damage (Salame and Habchi, 2008).

CFSER 2010

Page 6



6.0x10 -55

5.0x10

40 mA

-55

4.0x10

-55

3.0x10

-55

2.0x10

-55

1.0x10

0.0 0

5

10

15

20

25

30

Stress time (min)

Figure 6 Diffusion current vs time -23

2.5x10

40 mA

-23

2.0x10

-23

1.5x10

-23

1.0x10

-24

5.0x10

0.0 0

5

10

15

20

25

30

Stress time (min)

Figure 7 Recombination current vs time

4. Conclusions The effect of reversely induced electric stress currents has demonstrated to be a very important factor in determining not only the performance and efficiency of PV solar cells and modules, but it has also proven to be a dominant factor in determining the operating lifetime of the cells. This fact is shown by the consequences of the induced currents on shunt resistance, ideality factor, I-V characteristics as well as diffusion and recombination currents as a function of time. The importance of the achieved results stands for a new approach in determining the sum magnitude of the reverse current in a PV module, which may occur as a result of many factors

CFSER 2010

Page 7



discussed in the paper, accompanied with the time limit to total and permanent destruction of the module. The results are applicable for solar cells, modules, arrays, or even a farm of arrays creating a photovoltaic solar generator. It is recommended to take these results into consideration in future research projects, for the sake of producing and offering safe, friendly and sound photovoltaic solar systems operating under the effect of severe conditions.

References Breitenstein, O., Bauer, J., Lotnyk, A. and Wagner, J.M. (2009), “Defect induced non-ideal dark I-V characteristics of solar cells”, Superlattices and Microstrutures, ELSEVIER, Vol. 45, pp. 182-189. Calais, M., Wilmot, N., Ruscoe, A. Arteaga, O. and Sharma H. (2008), “Over-Current protection in PV array installations”, ISES-AP - 3rd International Solar Energy Society Conference - Asia Pacific Region (ISES-AP-08) Incorporating the 46th ANZSES Conference, 25-28 November 2008, Sydney, Austarlia. Keithley Instruments, Inc., (2007), “Making I-V and C-V measurements on solar/ photovoltaic cells using the model 4200-SCS semiconductor characterization system”, Keithley Instruments, Application note series number 2876, www.keithley.com, USA. Report downloaded Nov. 20, 2009. Lausch, D., Petter, K., Wenckstern, H., and Grundmann, M., (2009), “Correlation of prebreakdown sites and bulk defects in multicrystalline silicon solar cells”, Physica Status Solidi RRL, Vol. 3, No. 70, http://dx.doi.org/10.1002/pssr.200802264. Ramabadran, R. and Mathur, B. (2009), “Effect of shading on series and parallel connected solar PV modules”, Modern Applied Science, Vol. 3, No. 10, pp. 32- 42. Salame, C and Habchi. R., (2008), “Silicon MOSFET devices electrical parameters evolution at high temperatures”, Microelectronics International, Vol. 25, No. 1, pp. 21-24. Tsai, H., Tu, C. and Su, Y. (2008), “Development of generalized photovoltaic model using MATLAB/SIMULINK”, Proceedings of the World Congress on Engineering and Computer Science 2008, WCECS 2008, October 22 - 24, San Francisco, USA.

CFSER 2010

Page 8