User login

You are here

Journal Club January 2011: Mechanics and Materials for Solar Energy

Harley T. Johnson's picture

Brian C. McGuigan and Harley T. Johnson, UIUC

The photovoltaic (PV) energy conversion industry continues to make progress toward grid parity; within 3-5 years solar energy may be competitive with fossil fuel energy sources on a cost per watt basis.  New forms of lightweight, flexible, off-grid solar energy conversion are also in development.  Many of the barriers to further PV development are associated with materials and mechanics issues.  For example, in silicon PV technology, which currently makes up more than 80% of the commercial PV market, over 60% of the cost of a PV module is due to the silicon in the PV wafers.  This journal club issue presents an overview of PV energy conversion challenges from a mechanics and materials perspective, highlighting a few key issues that may be of interest to iMechanica readers.  The key classes of PV energy conversion technologies are briefly described, including silicon PV; other thin film PV technologies; and then multi-junction designs and other third generation technologies such as organic PV.  Several cross-cutting issues that apply to multiple PV technologies are also noted.  Some of the references below may be useful in understanding PV topics on a broad level, while several citations are somewhat focused and may be more relevant to specific mechanics issues.

Silicon dominates the commercial PV market.  Silicon solar cell energy conversion efficiencies are generally only around 20%, but the production of silicon PV modules is relatively cheap, and modules are available at under $2/watt.  Silicon solar cells contain either crystalline or amorphous silicon as the active material that converts incident photons into electron-hole pairs for photocurrent.  Amorphous cells generally have the lowest efficiency, but they are good in low light intensity applications and have been widely used in low power consumer applications for many years.  Crystalline cells are more efficient and are more widely used in grid-tied applications.  Generally, single crystalline cells have higher efficiencies, up to 25%, while polycrystalline cells have efficiencies up to 20%.1   Many of the mechanics issues in silicon PV are associated with the processing of the silicon wafer material, prior to assembly of the electrical interconnects, application of antireflective coatings, and other necessary fabrication steps. There are many processes used to manufacture solar grade silicon; the most widely used is the Czochralski process.2 3   This process, also commonly used in silicon microelectronics, involves melting high-grade silicon, introducing a seed crystal connected to a rod, and then lowering the rod into the molten silicon a few inches below the surface. The rod is slowly extracted and rotated while crystallization of the silicon occurs. The result is an ingot of nearly single crystalline silicon, typically about 150mm in diameter.  This ingot is later sawed into silicon wafers that are used as the light absorbing material in the solar cells described here.  However, there are several key mechanics and materials problems that may arise during these processing steps.

For example, during preparation of the ingots, boron-oxygen atomic scale defect complexes are grown into the material due to the oxygen present in the process and the boron needed for electrical doping.  A boron-oxygen defect complex acts as an electron-hole recombination center when the cell is in operation. 4 5 6 This reduces the photocurrent of the cell and, therefore, its efficiency when measured using the microwave-photoconductance-decay (μW-PCD) technique.7 Germanium doping is one means of modifying the diffusion barrier for the oxygen present, thus suppressing any B-O defects.8   Other inclusion-type defects arise from impurity materials such as carbides and nitrides; these inclusions can cause cascading microvoid defects that propagate through the crystal growth or pulling process.  Such defects can directly hinder the performance of the silicon in the solar cell.9 10  However, at low levels, the inclusions can promote the precipitation of oxygen impurities, resulting in the immobilization of dislocations in the silicon.11 

Wire sawing of the silicon ingots into wafers of about 250 microns in thickness can also cause significant damage to the material. 11 12 The abrasive cutting wires often leave surface imperfections in the wafers.  These imperfections serve as stress concentrations, which degrade the device performance and may lead to other mechanical issues such as subsurface cracking.13 14  15 Also, the sawing process often leaves a residual film that requires washing of the wafers in a scrubber.  The brushes of the scrubber can scratch the surface, which affects the fine finish and optical quality required of the silicon wafers.16 Handling of the silicon wafers after wire sawing may also be problematic; separation of the wafers from each other can be relatively difficult because of surface energetic effects between them.  Lubricants used in the sawing process to lessen vibration and heat generation, reducing damage to the wafer surfaces, also allow for easier separation.17 18 

Silicon casting is also used to produce PV grade silicon wafers.  Casting techniques used to produce polycrystalline silicon ingots require the cooling of molten silicon in a mold. 19 However, the iron present in the crucible walls can lead to formation of FeB complex defects that reduce the lifetime of the solar cell once it is assembled.20 21 22  Other crucible materials have been considered in order to eliminate iron diffusion; for example, water cooled copper crucibles have been studied.23 Wafer handling issues mentioned above also contribute to cell degradation in silicon wafers made through casting. 

Thin film solar cells made from materials other than silicon make up most of the remaining 20% of the commercial market.  The dominant thin film material in this class is CdTe. Amorphous silicon (a-Si) and Cu(In,Ga)Se2 (CIGS) have also been commercially produced.22 These systems have lower efficiencies but include easier handling, increased flexibility, low materials usage, and comparable manufacturing cost relative to silicon.23 There are several key materials issues that can compromise performance of a thin film solar cell.  For example, grain boundaries in the thin semiconductor absorbing layer can be potential sources of recombination in the cells, inhibiting the movement of photogenerated electrons and limiting the operating voltage of the cell.24 Therefore, the growth of larger grains is usually desirable.25 Also, the front contacts of these devices are susceptible to moisture exposure, which results in accelerated deterioration and reduced cell performance.  Methods of moisture prevention include the use of a transparent Al2O3 layer deposited by atomic layer deposition (ALD).26Other reliability issues occur in the primary junction of the cells where pinhole defects in the thin semiconductor layer, for example, lead to electrical shunting and reduced efficiency of the device.27

Multi-junction solar cells, while much more expensive per unit energy produced, have been shown to exceed 40% efficiency in the laboratory, and are the subject of extensive current research.28 These solar cells use two or more semiconductor materials with different bandgap energies to absorb different wavelengths of light for increased efficiency.26  The semiconductors can be mechanically stacked or fabricated monolithically, whereby each layer is epitaxially grown upon the layer beneath it.  However, in any multi-junction design, mismatch in lattice coefficients, coefficients of thermal expansion, doping levels, and the quality of the materials all may contribute to degradation of the solar cell performance.29 30 31 32 Strain arising from lattice mismatch between adjacent semiconductor layers is especially important because it strongly affects the bandgap energy of the material; bandgap energy is directly related to the wavelength of light most efficiently absorbed.33 The quality of the semiconductor material in a multijunction cell is also critical.  For example, in InGaP/InGaAs/Ge solar cells, rapid thermal annealing has been used in order to remove phosphorus-vacancy-related defect complexes, thereby improving device performance.34

In addition to materials considerations for the active region of a PV cell, electrical contacts and interconnects are potential sources of mechanics and materials issues in most of the solar cells described here.  These device structures are necessary both for the collection of the photo-generated electrons in the semiconductor as well as for the physical integrity of the device.  Adhesion of these contacts to the semiconductors is one significant concern.  Suitable adhesives should not lead to significant stress in the semiconductor material, and often must be electrically conductive.35 36The adhesion of these contacts tends to make these junctions sites for electron-hole recombination.  For this reason, doping may be increased near the contact, or a separate material may be introduced as a buffer to reduce recombination effects.37 

Several other emerging solar cell technologies are the focus of significant ongoing research.   For example, organic solar cells -- including dye-sensitized solar cells -- and intermediate band solar cells share some of the mechanics and materials issues of the technologies described above.38 39 In some cases there are additional issues; for example, organic solar cells are prone to air and light degradation, so proper sealing and containment of the polymer materials is often important.40 41 Intermediate band solar cells, in which quantum dots or other nanoparticles provide additional energy levels for photoabsorption, contain more heterostructured material interfaces than even typical multi-junction designs.  Thus, strain effects and the presence of dislocations and other atomic scale defects must be considered in developing new solar cells in this class.



1. M. Green,"Solar Cell Efficiency Tables," Progress in Photovoltaics: Research and Applications, 18,  346-352, 2010

2. D. Yang, "Oxygen in Czochralski silicon used for solar cells,"  Solar Energy Materials and Solar Cells,  72, 133-138, 2002

3.  J. Turley, " The Essential Guide to Semiconductors," p 63, 2003, New Jersey, Pearson Education Inc

4.  J. Schmidt, K. Bothe, R. Hezel, "Oxygen-related minority-carrier trapping centers in p-type Czochralski silicon," Applied Physics Letters, 80, 4395, 2002

5. T. K. Vu, Y. Ohshita, K. Araki, M. Yamaguchi, "Generation and annihilation of boron-oxygen related defects in boron-doped Czochralski-grown Si solar cells," Journal of Applied Physics, 91, 4853, 2002

6.Q. Wang, H. Ihsiu, "High-temperature interstitial oxygen diffusion retardation in epitaxial-layered heavily arsenic- or boron-doped Czochralski silicon wafers," Applied Physics Letters,  88, 15, 2006

7. Y. Xuegong, W. Peng, C. Peng, "Suppression of boron-oxygen defects in p-type Czochralski silicon by germanium doping," Applied Physics Letters, 97, 2010

8.Y. Xuegong, W. Peng, C. Peng, "Suppression of boron-oxygen defects in p-type Czochralski silicon by germanium doping," Applied Physics Letters, 97, 2010

9. B. O. Kolbesen, "CARBON IN SILICON: PROPERTIES AND IMPACT ON DEVICES," Solid-State Electronics, 25, 759-775, 1982.

10.  F. Shimura, R. S. Hockett , "NITROGEN EFFECT ON OXYGEN PRECIPITATION IN CZOCHRALSKI SILICON," Applied Physics Letters, 48, 224, 1986

11. I. Yonenaga, "Nitrogen effects on generation and velocity of dislocations in Czochralski-grown silicon," Journal of Applied Physics,  98, 1-6, 2005

12. H. Möller, "Basic mechanisms and models of multi-wire sawing,"  Advanced Engineering Materials, 6, 501-513, 2004

13. K. Jung, K. Young, "The saw-damage-induced structural defects on the surface of silicon crystals," Proceedings -Electrochemical Society, 5, 27-35, 2004

14. L. Seong Min," Prevention of dicing-lnduced damage in semiconductor wafers," Key Engineering Materials, 345, 485-488, 2007

15. G. Yufei, "Investigation of subsurface damage depth of single crystal silicon in electroplated wire saw slicing," Key Engineering Materials, 416, 306-310, 2009

16.  B. Albrecht, L. Fritz, "Reducing defect density using an optimized wafer scrubber," MICRO,  15, 39-42, 1997

17. G. Allardyce, R. Barr, R. Chan, "Interaction between post wire saw cleaning and the subsequent cell fabrication saw damage etch and texturing process," Record of the IEEE Photovoltaic Specialists Conference, 3494-3497, 2010

18. T. Kuan, K. Shih, V. Vechten, "EFFECT OF LUBRICANT ENVIRONMENTS ON SAW DAMAGE IN Si WAFERS,"  Journal of the Electrochemical Society, 127, 1387-1394, 1980

19. T. Nobuyuki, A. Masahito, "DEVELOPMENT OF HIGH EFFICIENCY POLYCRYSTALLINE SILICON SOLAR CELLS USING SOLAR GRADE CAST WAFERS," Japanese Journal of Applied Physics, 25, 958-960, 1986

20.  J. Schmidt, "Effect of dissociation of iron-boron Pairs in crystalline silicon on solar cell properties," Progress in Photovoltaics: Research and Applications, 13, 325-331, 2005

21.G. Coletti, "Effect of iron in silicon feedstock on p - And n -type multicrystalline silicon solar cells," Journal of Applied Physics, v 104, 2008

22. K. Wolfram, "Diode breakdown related to recombination active defects in block-cast multicrystalline silicon solar cells," Journal of Applied Physics, v 106, 2009

23. I. Takashi, "Refining of silicon for solar cells," First International Conference on Processing Materials for Properties, 441-444, 1993

24. J. Sites, "Thin-film photovoltaics: What are the reliability issues and where do they occur?," IEEE International Reliability Physics Symposium Proceedings, p 494-498, 2010

25. J. Major, "Control of grain size in sublimation-grown CdTe, and the improvement in performance of devices with systematically increased grain size," Solar Energy Materials and Solar Cells, 94, 1107-1112, 2010

26. S. Hegedus, "Encapsulation of Cu(InGa)Se2 solar cells with ALD Al 2O3 flexible thin-film moisture barrier: Stability under 1000 hour damp heat and UV exposure,"  Conference Record of the IEEE Photovoltaic Specialists Conference, p 1178-1183, 2010

27. A.R. Davies, A.E. Enzenroth, W.S. Sampath, "All-CSS processing of CdS/CdTe thin-film solar cells with thin CdS layers," Materials Research Society Symposium Proceedings, 1012, 157-162, 2007

28. G. Kinsey, P. Pien, P. Hebert, R. Sherif, "Operating characteristics of multijunction solar cells ," Solar Energy Materials and Solar Cells, 93, 950-951, 2009

29. J. Gee, "MECHANICALLY-STACKED MULTIJUNCTION SOLAR CELLS," Commission of the European Communities, p 245-253, 1985

30. G.F. Virshup, "Temperature coefficients of multijunction solar cells," Conference Record of the IEEE Photovoltaic Specialists Conference, 1, 336-338, 1990

31. J. Hutchby, "MATERIAL ASPECTS OF THE FABRICATION OF MULTIJUNCTION SOLAR CELLS," Proceedings of SPIE - The International Society for Optical Engineering, 543, 40-61, 1985

32. I. Bhattacharya, "Effects of gallium-phosphide and indium-gallium-antimonide semiconductor materials on photon absorption of multijunction solar cells,"  Conference Proceedings - IEEE SOUTHEASTCON, 316-319, 2010

33. H.F. MacMillan, H.C. Hamaker, G.F. Virshup, J.G Werthen, "Multijunction III-V solar cells: Recent and projected results ," Conference Record of the IEEE Photovoltaic Specialists Conference, 1, 48-54, 1988

34. Y. Min-De, L. Yu-Kai, "Improvement of material quality of multijunction solar cells by rapid thermal annealing,"  Japanese Journal of Applied Physics, 47, 4499-4501, 2008

35. D.W.K. Eikelboom, "Conductive adhesives for low-stress interconnection of thin back-contact solar cells," Conference Record of the IEEE Photovoltaic Specialists Conference, p 403-406, 2002

36. J. Luboš, "Reliability of solar cell's solder joints,"  Nano Technologies for Electronics Packaging, Conference Proceedings, p 188-192, 2006

37. F. Einsele, P. Rostan, M. Schubert, "Recombination and resistive losses at ZnOa-Si:Hc-Si interfaces in heterojunction back contacts for Si solar cells,"  Journal of Applied Physics, v102, 2007

38.  S. Shaheen, "Mechanisms of operation and degradation in solution-processable organic photovoltaics," Annual Proceedings - Reliability Physics (Symposium), p 248-252, 2007

39. G. Jolley, "Electron-hole recombination properties of In0.5 Ga0.5 As/GaAs quantum dot solar cells and the influence on the open circuit voltage," Applied Physics Letters, 97, 12, 2010

40. A. Rivaton, "Light-induced degradation of the active layer of polymer-based solar cells,"  Polymer Degradation and Stability, 95, 278-284, 2010

41. K. Kawano, "Degradation of organic solar cells due to air exposure,"  Solar Energy Materials and Solar Cells, 90, 3520-3530, 2006


Cai Shengqiang's picture

Thank you,McGuigan and Johnson. This review is really good for us to appreciate the mechanics in the development of solar energy. The collection of literatures is also inspiring as I read.

Harley T. Johnson's picture

Thanks for the feedback, Shengqiang.  We think the area is very interesting and promising, and we hope this serves as a helpful brief introduction.  We welcome questions and comments, and will do our best to answer them.

Oleksandr Glushko's picture

Thanks for a nice review with useful references!

Could you give an idea how the cost of solar power is usually calculated? You mentioned the price of 2$/watt for silicon solar cells.

Subscribe to Comments for "Journal Club January 2011: Mechanics and Materials for Solar Energy"

Recent comments

More comments


Subscribe to Syndicate