Energy Procedia 27 ( 2012 ) 214 – 218

1876-6102 © 2012 Published by Elsevier Ltd. Selection and peer-review under responsibility of the scientifi c committee of the SiliconPV 2012 conference.doi: 10.1016/j.egypro.2012.07.054

SiliconPV 2012, 03-05 April 2012, Leuven, Belgium

Characterization and Analysis of Local Series Resistance on Quasi-Monocrystalline Solar Cells

Y. Yang*, Z. Li, Z. Bian, G. Xu, P. Verlinden, Z. Feng, Q. Huang State Key Lab of PV Science and Technology, Trina Solar Limited Company, No.2 Trina Road, Trina PV Park, Xinbei District,

Changzhou, 213031, China

Abstract

This work evaluates the application of Quasi-Monocrystalline silicon (QM-Si) wafers on the conventional structure of silicon solar cells textured by an alkaline solution to generate random pyramids. Although most areas of QM-Si wafers are <100> orientated by using a seeded-grown and directional-solidification process, there are still some none-<100> orientated areas randomly distributed on the wafers. These areas stay more planar, featuring worse reflectance and light trapping. Furthermore, series resistance (Rs) mapping on the final cells shows that these areas also have a higher local Rs. Our experiments and analysis show that the higher local Rs is mainly due to the poor Ag/Si contacts according to the results of our TLM measurement. Such higher local Rs values have led to only 76.3% fill factor and 17.1% efficiency performance of 500 cells on average. Further detailed investigations show that the local Rs, or local contact resistance, depends greatly on the surface roughness for screen printed cells, which implies that to take advantage of better quality of the QM-Si wafers, silver paste needs to be improved to form good contacts on both textured and planar silicon surface, when these wafers are textured with a random-pyramid alkaline process.

© 2012 Published by Elsevier Ltd. Selection and peer-review under responsibility of the scientific committee of the SiliconPV 2012 conference Keywords: quasi-monocrystalline; contact resistance; series resistance; solar cells

1. Introduction

Quasi-monocrystalline (QM) silicon wafers have entered the PV industry recently and become a current topic for many research institutes [1-2]. In addition to fewer grain boundaries and higher carrier lifetime, another major advantage for QM-Si wafers is the application of the random pyramid texture

* Corresponding author, E-mail address: yang.yang02@trinasolar.com

Vailable online at www.sciencedirect.com

© 2012 Published by Elsevier Ltd. Selection and peer-review under responsibility of the scientifi c committee of the SiliconPV 2012 conference. Open access under CC BY-NC-ND license.

Open access under CC BY-NC-ND license.

Y. Yang et al. / Energy Procedia 27 ( 2012 ) 214 – 218 215

using alkaline solution, which greatly improves the light trapping compared to traditional isotropic texturing. However, due to the side growth effect close to the crucible walls [3] or between the two seed areas, there are still some multicrystalline areas that don’t have <100> orientations. After alkaline texturing, such areas may become “polished” planar or variably tilted pyramids with <111> orientations facing upward, which lead to higher reflectivity.

While the optical properties [4] and the dislocation properties from bottom to top of QM-Si wafers (ingots) have been studied [3], few investigations have been reported on the local electrical properties of QM-Si cells. In this work, special attention is paid to the characterization of the local series resistance of QM-Si cells. The high local contact resistance has resulted in low fill factor and overall cell performance. Finally, the influence of surface roughness on local series resistance has been characterized on several samples.

2. Experiment

The 500 pcs QM-Si wafers which have a resistivity of 1.5 Ohm.cm, an initial thickness of 180 μm and a dimension of 15.6 15.6 cm2were used in the experiment. The as-cut wafers received a standard KOH/IPA alkaline texturization process for 25 mins at 80°C, and then a POCl3 tube diffusion to form an averaged 68 Ohm/sq sheet resistance emitter, followed by wet chemical etching for edge isolation, PECVD SiNx at the front, screen printing on both sides and co-firing. Note that all these wafers came from the same ingot and share almost the same percentage of none <100> orientation areas. The percentage of such wafers in the whole ingot is about 40%. Another 500 pcs cells were fabricated by the same cell process except that they were made on normal multicrystalline silicon (mc-Si) wafers and an isotropic etching replaced the standard alkaline texturing step.

3. Results and discussions

3.1 cells’ performance

Table 1. Averaged cell performance based on QM-Si wafers and mc-Si wafers

Cell type Voc [mV]

Jsc [mA/cm2]

Rs [Ohm.cm2]

FF [%]

[%]

QM-Si wafers 619 36.2 1.04 76.3 17.1

mc-Si wafers 617 34.6 0.735 77.9 16.6

As shown in Table I, the main improvement for QM-Si cells is Jsc, which increased from 34.6 to 36.2 mA/cm2. However, the fill factor drops from 77.9% to 76.3%, due to the increased Rs from 0.735 Ohm.cm2 to 1.04 Ohm.cm2. In average, the final efficiency for QM-Si cells only presents a 0.5% absolute gain and is still much lower than the mono-crystalline solar cell level. To find out the cause for higher Rs, reflectivity and series resistance mapping obtained from luminescence imaging [5] were measured on a typical QM-Si cell, as shown in Fig.1.

216 Y. Yang et al. / Energy Procedia 27 ( 2012 ) 214 – 218

(a) (b) (c) Fig. 1. (a) Measured surface morphology of a planar (A) and pyramidal (B) area; (b) reflectivity (R) mapping at 400 nm wavelength, where red colour means lower R and blue colour means higher R; (c) Rs mapping on the same cell.

The bright areas in Rs mapping fit quite well the planar areas in reflectivity mapping, which means that the “planar” surface may suffer from contact problems for screen printed cells. To find a more direct support for this, a simplified TLM measurement was made on the laser-cut regions A and B in Fig.1 (b), with the results shown in Fig.2 (b). In this case the measured contact resistance at the planar surface (region A) is about 4 times larger than that at the pyramidal surface (region B), while the apparent sheet resistance is lower for the “planar” areas than for the well-textured areas.

(a) (b)

Fig. 2. (a) Illustration of the simplified TLM measurement, Ri=Vi/Ii; (b) measured resistances between 1 to 4 fingers for region A and B, respectively. The intercepts at the Y axis represent the contact resistances.

3.2 Doping profiles

To investigate the reasons for poor contact at the planar surface, we first check the difference of doping profiles at rough and planar surface, since variations in the surface doping concentration (Ns) may be a reason for high contact resistance as well. For Ns > 1e19 cm-3, the specific contact resistance is strongly dependent on the surface doping concentration and given by [6],

4exp s Bn

cs

mkqTA h N

(1) Two wafers have been randomly selected for Electro-Chemical Voltage profile (ECV) measurement

after POCl3 diffusion and HF dip to remove the PSG. There are 3 measured points with different roughness for each wafer. The actual doping profiles were recalculated according to the 4pp measured

Reflectivity Rs

A B B

A

Y. Yang et al. / Energy Procedia 27 ( 2012 ) 214 – 218 217

sheet resistance and the mobility model of Klaassen [7]. As shown in Fig.3, although the planar surface regions have a slightly higher sheet resistance than that on pyramidal surface regions, they share nearly the same profile and same Ns at around 6e20 cm-3, which implies that the high local contact resistances are not caused by a variation of Ns.

(a) (b)

Fig. 3. Dopant profiles measured by ECV on two wafers, (a) and (b). Three regions with different surface roughness have been chosen for each wafer

3.3 Surface roughness

To quantify the dependence of local series resistance on surface roughness, several multi-crystalline wafers have been textured by KOH/IPA solution and followed by the processes outlined above. The surface area roughness (Sa) was measured using 3D microscope, while the corresponding local Rs was extracted from Rs mapping. As shown in Fig.4, the local Rs increases when the surface roughness decreases.

Fig. 4. The dependence of local series resistance on surface area roughness. (a) Rs mapping; (b) 3D microscope pictures; (c) Sa vs Local Rs

A B

CE

F

Series resistance mapping A

D

E

F

D

(a) (b) A B C D E F

(c) D

Zone

218 Y. Yang et al. / Energy Procedia 27 ( 2012 ) 214 – 218

4. Conclusion

The high local series resistance on planar surface areas of quasi-monocrystalline silicon solar cells have been characterized and analyzed. It is found that the high local series resistance was due to an increase of the contact resistance for the planar areas compared to the well-textured areas. Such local high Rs effect has led to a very modest increase in efficiency when compared to normal mc-Si solar cells, from 16.6% to17.1%, due to an actual reduction in fill factor from 77.9% to 76.3%. The contact resistance depends greatly on the surface roughness for screen printed cells, which implies that to take the advantage of the better quality of the QM-Si wafers, silver paste needs to be improved to form good contacts on both textured and planar silicon surface.

Acknowledgement This work was supported by the National Basic Research Program of China, 2012CB724407. Reference

[1] K.Fujiwara, W. Pan, N. Usami, K. Sawada, M. Tokairin, Y. Nose, A. Nomura, T. Shishido, K. Nkajima, Acta Materialia, Vol. 54(12), pp3191-3197, 2006

[2]N. Stoddard, B. Wu, I. Witting, M. Wagener, Y. Park, G. Rozgonyi, R. Clark, “ Casting single crystal silicon: Novel defect profiles from BP Solar’s Mono2 wafers”, Solid State Phenomena, Vol. 131-133, pp1-8,2008

[3] B. Marie, S. Bailly, A. Jouini, D. Ponthenier, N. Plassat, L. Dubost, “ Seeded grown mono-like Si ingots: effect on recombination activity of dislocations”, proceeding of 26th PVSEC, pp1910-1914,2011

[4] V. Projapati, E. cornagliotti, R. Russell, JM. Fernandez, RF. Clark, N. Stoddard, P. Choulat, J. John, “ HIGH EFFICIENCY INDUSTRIAL SILICON SOLARCELLS ON SILICON MONO2 TM CAST MATERIAL USING DIELECTRIC PASSIVATION AND LOCAL BSF”, Proceeding of 24th PVSEC, pp1171-1174, 2009

[5] T. Trupke, E. Pink, RA. Bardos, MD. Abbott, “ Spatially resolved series resistance of silicon solar cells obtained from luminescence imaging”. APL, 90, 093506 (2007)

[6] Mohamed M. Hilali, et al. “Understanding the Formation and Temperature Dependence of Thick-Film Ag Contacts on High-Sheet-Resistance Si Emitters for Solar Cells”, Journal of The Electrochemical Society, 152 (10) G742-G749 (2005).

[7] D. B. M. Klaassen, Solid-State Electron. 35, 953 (1992)