DETERMINATION OF BACK CONTACT BARRIER ... - Semantic Scholar
DETERMINATION OF BACK CONTACT BARRIER HEIGHT IN Cu(In,Ga)(Se,S)2 AND CdTe SOLAR CELLS 1
1
2
Galymzhan T. Koishiyev , James R. Sites , Sachin S. Kulkarni , Neelkanth G. Dhere
2
1
Physics Department, Colorado State University, Fort Collins, CO, United States
2
Florida Solar Energy Center, University of Central Florida, Cocoa, FL, United States
ABSTRACT A relatively straightforward technique has been developed to quantify the energy barrier for holes between a Cu(In,Ga)(Se,S)2 (CIGSeS) or CdTe absorber and the back-contact metallization. The input data is the currentvoltage (J-V) curves for the solar cell measured over a range of temperatures. The key parameter is the “turning current” Jt, which is the current at the transition from the positive J-V curvature of a diode to the negative curvature associated with current limitation at a contact barrier. The analytical strategy is to calculate a series of Jt vs. T curves for different values of barrier height and then overlay the experimental values of Jt. Generally the experimental data follow a single barrier-height curve over a wide temperature range. The presentation will describe the turning point technique and apply it to specific solar-cell examples. The range of Jt that can be practically identified 2 extends from approximately 0.1 to 80 mA/cm . Assuming that temperatures between 220 and 340 K are available, the range of barriers that can be determined is between 0.30 and 0.55 eV. This is also the practical range, since lower barriers do not have a measurable effect on the power quadrant and higher ones effectively kill the performance of the cell. Many CIGSeS and CdTe cells, however, do have a back-contact barrier in the 0.30 to 0.55 eV range, and the ability to determine it can assist both cell analysis and process optimization.
Cu:In:Ga:Se:S ratio was 23:25:2:44:5, measured at NREL by EPMA, i.e. only small amounts of Ga and S. The CdTe cell was prepared by close-spaced sublimation. BACKGROUND A metal-semiconductor interface often produces a barrier to carriers. For an ideal contact between a metal and a p-type semiconductor, in the absence of surface states, the contact barrier height for holes is given by [5]:
Φb = χ + Eg − Φ m ,
(1)
where χ is the electron affinity of the semiconductor, Eg is the band gap of the semiconductor, and Φm is the metal work function.
INTRODUCTION The shape of J-V curves for many thin-film polycrystalline solar cells show various deviations from the ideal diode J-V curve due to different types of secondary barriers. One common deviation is current limitation in forward bias, also known as the “rollover” effect. The rollover effect has been documented in several studies [14] and has generally been attributed to the back contact barrier. The back contact barrier can be modeled by an additional diode that has opposite polarity to the primary photodiode, is not light sensitive and shows significant leakage [2]. The back contact barrier height can then be calculated in various ways. The strategy advocated here is used to determine hole barrier heights of Cu(In,Ga)(Se,S)2 (CIGSeS) and CdTe solar cells fabricated at the Florida Solar Energy Center (FSEC) and the National Renewable Energy Laboratory (NREL) respectively. This CIGSeS cell analyzed was prepared by selenization and sulfurization of metallic precursors by rapid thermal annealing. The
978-1-4244-1641-7/08/$25.00 ©2008 IEEE
Fig.1. Band diagram at the interface between a p-type semiconductor and a back contact metal. Figure 1 shows a band diagram of a solar cell at thermodynamic equilibrium at 300K, where only the metalsemiconductor interface is shown. Figure 1 shows Fermi level EF and valence bands EV for three different values of back contact barrier height Φb - 0.42eV, 0.46eV and 0.50eV. The presence of a back contact barrier in a solar cell has the effect of suppressing hole transport. Assuming that the thermionic emission is the driving mechanism at the interface between a p-type absorber and back contact metal, the limiting hole current is given by [1]:
qV J h = J 0 e kT − 1,
2
(2)
mA/cm at 273 K is found from the J-V curve shown in Fig. 2. The saturation current at the back contact, or equivalently Jt, is the turning current is given by [1,5]:
where J0 is the saturation current at the back contact, T is the temperature, k is the Boltzman’s constant, q is the elementary charge, and V is the applied bias.
*
2
J0 = Jt = A T e
−
qΦ b kT
,
(3)
*
where A is the Richardson constant, Φb is the back contact barrier height. The strategy of finding the barrier Φb experimentally is to determine turning currents Jt using the technique described above for a range of operating temperatures and then overlay the experimental values with analytical curves given by equation (3). The procedure is relatively insensitive to the value of A* used.
J-V-T MEASUREMENTS ON CIGSeS AND CdTe SOLAR CELLS AND DETERMINATION OF THEIR BARRIER HEIGHTS
Fig. 2. J-V curve at 273 K with linear fits to determine Jt. Figure 2 shows an example of an experimental J-V curve of a CIGSeS solar cell with a measurable value of back contact barrier. The current density at which a J-V curve at any given temperature bends over, the turning current Jt is approximately the saturation current, can be used as the key parameter describing the rollover effect. To find the turning current, one can make two linear fits to the data points above and below the onset of the rollover. The intersection of these two straight lines defines the turning current. For example, a turning current of 6
The basic room-temperature J-V parameters of the two cells used as examples are VOC = 0.54 V, JSC = 35.0 2 mA/cm , FF = 58%, and η= 11.0% for the CIGSeS solar 2 cell, and VOC = 0.82 V, JSC = 23.4 mA/cm , FF = 70%, and η= 12.9% for the CdTe device. Figure 3 shows experimental light J-V curves under illumination for (a) the CIGS device from 238 K to 303 K and (b) the CdTe device from 223 K to 293 K, with intervals of 5 K in both cases. The open-circuit voltage increases as temperature is reduced, but the short-circuit current is essentially unchanged for either device. From Fig. 3, one can clearly see the effect of the contact barrier in both devices. The turning current for each curve is plotted in Fig. 4.
Fig. 3. Experimental light J-V curves of (a) CIGSeS device from 238 K to 303 K and (b) CdTe device from 223K to 293K with intervals of 5 K in both cases.
978-1-4244-1641-7/08/$25.00 ©2008 IEEE
[2] G. Stollwerck and J.R. Sites, “Analysis of CdTe back th contact barriers,” 13 European Photovoltaic Solar Energy Conference, Nice, France, 1995, pp 2020-2022 [3] Alex Niemegeers and Marc Brgelman, “Effects of Au/CdTe Back Contact on I-V and C-V Characteristics of Au/CdTe/CdS/TCO Solar Cells,” J. Appl. Phys. 81, 28812886 (1997). [4] B.E. McCandless, J.E. Phillips, and J.Titus, “Characterizing Contacts to p-type CdTe in CdS/CdTe nd Solar Cells,” 2 World Conference and Exhibition on Photovoltaic Solar Energy Conversion, Vienna, 1998, pp 448-452. [5] S.M. Sze, Physics of Semiconductor Devices, 2 1981, pp 259-262. Fig. 4. Solid lines are calculated curves from eq. (3) with
Φb ranging from 0.36 to 0.50 eV. Solid and open dots are the experimental CIGSeS and CdTe Jt data points. The solid lines in figure 4 are calculated curves of the turning current for different values of the back contact barrier height ranging from 0.36 to 0.50 eV with 0.02-eV increments following Eqn. (3). The dots are experimental data points from J-V curves at different temperatures. Fig. 4 clearly illustrates that the back contact barrier heights of these two CIGSeS and CdTe devices are approximately 0.44 eV and 0.40 eV respectively. The rollover effect can be very pronounced at low temperatures, but it generally disappears at high temperatures. As can be seen in Fig. 3, it becomes increasingly difficult to determine the turning point at higher temperatures, because the negative curvature becomes very small. At very low temperatures, on the other hand, the rollover effect is quite pronounced, but the small value of the turning current makes it difficult to deduce with confidence. CONCLUSIONS The proposed strategy for determining the back contact barrier height has yielded reasonable values for barrier heights of CIGS and CdTe solar cells. In the CIGSeS example given, the barrier height was determined to be 0.44 eV, and in the CdTe example, 0.40 eV. In both cases, the fit over a broad temperature range gives additional confidence in the results. ACKNOWLEDGMENT The research reported here was supported by the National Renewable Energy Laboratory. REFERENCES [1] S. Demtsu, “Impact of Back Contact Materials on Performance and Stability of CdS/CdTe Solar Cells,” Ph.D. Thesis, Colorado State University, Spring 2006.
978-1-4244-1641-7/08/$25.00 ©2008 IEEE
nd
ed.,
1
2
Galymzhan T. Koishiyev , James R. Sites , Sachin S. Kulkarni , Neelkanth G. Dhere
2
1
Physics Department, Colorado State University, Fort Collins, CO, United States
2
Florida Solar Energy Center, University of Central Florida, Cocoa, FL, United States
ABSTRACT A relatively straightforward technique has been developed to quantify the energy barrier for holes between a Cu(In,Ga)(Se,S)2 (CIGSeS) or CdTe absorber and the back-contact metallization. The input data is the currentvoltage (J-V) curves for the solar cell measured over a range of temperatures. The key parameter is the “turning current” Jt, which is the current at the transition from the positive J-V curvature of a diode to the negative curvature associated with current limitation at a contact barrier. The analytical strategy is to calculate a series of Jt vs. T curves for different values of barrier height and then overlay the experimental values of Jt. Generally the experimental data follow a single barrier-height curve over a wide temperature range. The presentation will describe the turning point technique and apply it to specific solar-cell examples. The range of Jt that can be practically identified 2 extends from approximately 0.1 to 80 mA/cm . Assuming that temperatures between 220 and 340 K are available, the range of barriers that can be determined is between 0.30 and 0.55 eV. This is also the practical range, since lower barriers do not have a measurable effect on the power quadrant and higher ones effectively kill the performance of the cell. Many CIGSeS and CdTe cells, however, do have a back-contact barrier in the 0.30 to 0.55 eV range, and the ability to determine it can assist both cell analysis and process optimization.
Cu:In:Ga:Se:S ratio was 23:25:2:44:5, measured at NREL by EPMA, i.e. only small amounts of Ga and S. The CdTe cell was prepared by close-spaced sublimation. BACKGROUND A metal-semiconductor interface often produces a barrier to carriers. For an ideal contact between a metal and a p-type semiconductor, in the absence of surface states, the contact barrier height for holes is given by [5]:
Φb = χ + Eg − Φ m ,
(1)
where χ is the electron affinity of the semiconductor, Eg is the band gap of the semiconductor, and Φm is the metal work function.
INTRODUCTION The shape of J-V curves for many thin-film polycrystalline solar cells show various deviations from the ideal diode J-V curve due to different types of secondary barriers. One common deviation is current limitation in forward bias, also known as the “rollover” effect. The rollover effect has been documented in several studies [14] and has generally been attributed to the back contact barrier. The back contact barrier can be modeled by an additional diode that has opposite polarity to the primary photodiode, is not light sensitive and shows significant leakage [2]. The back contact barrier height can then be calculated in various ways. The strategy advocated here is used to determine hole barrier heights of Cu(In,Ga)(Se,S)2 (CIGSeS) and CdTe solar cells fabricated at the Florida Solar Energy Center (FSEC) and the National Renewable Energy Laboratory (NREL) respectively. This CIGSeS cell analyzed was prepared by selenization and sulfurization of metallic precursors by rapid thermal annealing. The
978-1-4244-1641-7/08/$25.00 ©2008 IEEE
Fig.1. Band diagram at the interface between a p-type semiconductor and a back contact metal. Figure 1 shows a band diagram of a solar cell at thermodynamic equilibrium at 300K, where only the metalsemiconductor interface is shown. Figure 1 shows Fermi level EF and valence bands EV for three different values of back contact barrier height Φb - 0.42eV, 0.46eV and 0.50eV. The presence of a back contact barrier in a solar cell has the effect of suppressing hole transport. Assuming that the thermionic emission is the driving mechanism at the interface between a p-type absorber and back contact metal, the limiting hole current is given by [1]:
qV J h = J 0 e kT − 1,
2
(2)
mA/cm at 273 K is found from the J-V curve shown in Fig. 2. The saturation current at the back contact, or equivalently Jt, is the turning current is given by [1,5]:
where J0 is the saturation current at the back contact, T is the temperature, k is the Boltzman’s constant, q is the elementary charge, and V is the applied bias.
*
2
J0 = Jt = A T e
−
qΦ b kT
,
(3)
*
where A is the Richardson constant, Φb is the back contact barrier height. The strategy of finding the barrier Φb experimentally is to determine turning currents Jt using the technique described above for a range of operating temperatures and then overlay the experimental values with analytical curves given by equation (3). The procedure is relatively insensitive to the value of A* used.
J-V-T MEASUREMENTS ON CIGSeS AND CdTe SOLAR CELLS AND DETERMINATION OF THEIR BARRIER HEIGHTS
Fig. 2. J-V curve at 273 K with linear fits to determine Jt. Figure 2 shows an example of an experimental J-V curve of a CIGSeS solar cell with a measurable value of back contact barrier. The current density at which a J-V curve at any given temperature bends over, the turning current Jt is approximately the saturation current, can be used as the key parameter describing the rollover effect. To find the turning current, one can make two linear fits to the data points above and below the onset of the rollover. The intersection of these two straight lines defines the turning current. For example, a turning current of 6
The basic room-temperature J-V parameters of the two cells used as examples are VOC = 0.54 V, JSC = 35.0 2 mA/cm , FF = 58%, and η= 11.0% for the CIGSeS solar 2 cell, and VOC = 0.82 V, JSC = 23.4 mA/cm , FF = 70%, and η= 12.9% for the CdTe device. Figure 3 shows experimental light J-V curves under illumination for (a) the CIGS device from 238 K to 303 K and (b) the CdTe device from 223 K to 293 K, with intervals of 5 K in both cases. The open-circuit voltage increases as temperature is reduced, but the short-circuit current is essentially unchanged for either device. From Fig. 3, one can clearly see the effect of the contact barrier in both devices. The turning current for each curve is plotted in Fig. 4.
Fig. 3. Experimental light J-V curves of (a) CIGSeS device from 238 K to 303 K and (b) CdTe device from 223K to 293K with intervals of 5 K in both cases.
978-1-4244-1641-7/08/$25.00 ©2008 IEEE
[2] G. Stollwerck and J.R. Sites, “Analysis of CdTe back th contact barriers,” 13 European Photovoltaic Solar Energy Conference, Nice, France, 1995, pp 2020-2022 [3] Alex Niemegeers and Marc Brgelman, “Effects of Au/CdTe Back Contact on I-V and C-V Characteristics of Au/CdTe/CdS/TCO Solar Cells,” J. Appl. Phys. 81, 28812886 (1997). [4] B.E. McCandless, J.E. Phillips, and J.Titus, “Characterizing Contacts to p-type CdTe in CdS/CdTe nd Solar Cells,” 2 World Conference and Exhibition on Photovoltaic Solar Energy Conversion, Vienna, 1998, pp 448-452. [5] S.M. Sze, Physics of Semiconductor Devices, 2 1981, pp 259-262. Fig. 4. Solid lines are calculated curves from eq. (3) with
Φb ranging from 0.36 to 0.50 eV. Solid and open dots are the experimental CIGSeS and CdTe Jt data points. The solid lines in figure 4 are calculated curves of the turning current for different values of the back contact barrier height ranging from 0.36 to 0.50 eV with 0.02-eV increments following Eqn. (3). The dots are experimental data points from J-V curves at different temperatures. Fig. 4 clearly illustrates that the back contact barrier heights of these two CIGSeS and CdTe devices are approximately 0.44 eV and 0.40 eV respectively. The rollover effect can be very pronounced at low temperatures, but it generally disappears at high temperatures. As can be seen in Fig. 3, it becomes increasingly difficult to determine the turning point at higher temperatures, because the negative curvature becomes very small. At very low temperatures, on the other hand, the rollover effect is quite pronounced, but the small value of the turning current makes it difficult to deduce with confidence. CONCLUSIONS The proposed strategy for determining the back contact barrier height has yielded reasonable values for barrier heights of CIGS and CdTe solar cells. In the CIGSeS example given, the barrier height was determined to be 0.44 eV, and in the CdTe example, 0.40 eV. In both cases, the fit over a broad temperature range gives additional confidence in the results. ACKNOWLEDGMENT The research reported here was supported by the National Renewable Energy Laboratory. REFERENCES [1] S. Demtsu, “Impact of Back Contact Materials on Performance and Stability of CdS/CdTe Solar Cells,” Ph.D. Thesis, Colorado State University, Spring 2006.
978-1-4244-1641-7/08/$25.00 ©2008 IEEE
nd
ed.,
