2. Growth of quasi-mono crystalline ingot by SMART

Growth of quasi-mono crystalline ingot by SMART

  • School of Engineering/Graduate School of Engineering

Noritaka Usami [Professor]


Outline of Seeds

It only makes sense that creating affordable solar electricity requires that we improve the conversion efficiency of the multicrystalline silicon solar cells that represent more than half of solar cells currently in use without increasing manufacturing costs. Multicrystalline silicon has the benefit of being able to be manufactured cheaply through unidirectional solidification of the silicon melt in large-capacity (about a meter square) quartz crucibles, but it is inferior to monocrystalline in terms of conversion efficiency. If we could create high-quality monocrystalline wafers using widespread multicrystalline ingot manufacturing facilities, 100% of the wafers in distribution could be converted to monocrystalline structures to simultaneously achieve high efficiency and low cost. Our research team is proposing an innovative manufacturing method of monocrystalline, large-capacity ingots called SMART (Seed Manipulation for ARtificially-controlled Defect Technique) and is currently developing the basic theories and basic technologies to make it a reality. We expect to be able to apply the SMART method not only to silicon, but to a variety of other materials as well.

Novelty and originality of this research

The SMART method allows for the production of high-quality monocrystalline structures by taking on a variety of functions, including strain relaxation towards intentionally introduced functional defects, trapping impurities, inhibiting polycrystallization, locking in dislocations, and more. We can supply a monocrystalline wafer if we can distribute functional defects to areas that the wafer will not use (such as cut ends from block processing or ingot ends) and make the monocrystalline area wide enough so that it is larger than the size of the actual solar cell. Ideas like these came as a result of collecting knowledge through basic research on the formation mechanisms for silicon multicrystalline structures, including defining the relationship between crystalline grain interfaces and the generation of dislocations, interactions between crystal defects and impurities, and the mechanism by which functional grain interfaces inhibit polycrystallization.

Application and research area for Industry collaboration

The SMART method considers the positional relationship of multiple seed crystals in positioning the bottom of the crucible, and uses defect formation in the growth process to introduce functional defects. This makes it possible to use massively popular multicrystalline ingot manufacturing facilities without modifying them. By utilizing findings from our 3D thermal fluid simulations and a model growth experiment using a compact university device, we have applied the SMART method to manufacturing artificial monocrystalline silicon ingots with the aim of making low-cost production of high-quality monocrystalline silicon wafers possible.

Key Takeaway

We are working to funnel our research outcomes back into society as soon as possible, and by selecting materials based on silicon (a plentiful research resource that is safe and enjoys widespread social acceptance) as our primary topic of study, we are taking steps to ensure that the end goal of what we do is clearly defined, and eagerly doing so from the earliest stages in our industrial-academic joint research as well.


Functional defects, artificial monocrystal, SMART method, silicon, solar cells


  • Crystal growth/processing technologies (from bulk to nano scale)
  • Solar cell fabrication process and characterization techniques
  • Techniques for evaluating the optical and electrical properties of semiconductors
  • Techniques for evaluating microstructures


  • Directional solidification furnace for silicon ingot
  • in situ observation system of crystal growth
  • plasma CVD
  • ALD
  • minority carrier lifetime measurement system
  • minority carrier diffusion length measurement system
  • solar simulator
  • photoluminescence imaging


  • Japanese Unexamined Patent Application Publication No. 2016-172681

Monographs, Papers and Articles

  • Controlling impurity distributions in crystalline Si for solar cells by using artificial designed defects, J. Cryst. Growth (in press)
  • Seed manipulation for artificially controlled defect technique in new growth method for quasi-monocrystalline Si ingot based on casting, Appl. Phys. Express 8, 105501 (2015).
  • Mono-like silicon growth using functional grain boundaries to limit area of multicrystalline grains, IEEE J. Photovolt. 4, 84 (2014)
  • Control of grain boundary propagation in mono-like Si: Utilization of functional grain boundaries, Appl. Phys. Express 6, 025505 (2013).