Thin Film

Tin Sulfide (SnS):

SnS-grains

The ideal absorber material for photovoltaic power generation should combine strong optical absorption (favoring a direct band-gap, as in GaAs) and long minority carrier lifetimes (favoring an indirect band-gap, as in Si).  Existing photovoltaic materials excel in only one of these areas.  We have identified Tin Sulfide (SnS) as a material that could optimize both properties simultaneously, thanks to its staggered band gap structure.  SnS has a direct band gap of 1.3 eV and an indirect band gap of 1.1 eV [1].  This combination leads carriers excited over the direct band-gap to thermalize down to energy levels where they recombine over the indirect one, increasing carrier lifetimes.

 

Not only is SnS uniquely suited for photovoltaic energy conversion, it is also abundant enough to be used for terawatts of power generation.  In addition, good quality SnS films can be fabricated at relatively low temperatures, supporting low-cost manufacturing.  Photovoltaic modules using the chalcogenide family of materials, which includes SnS, Cadmium Telluride (CdTe), and copper (indium, gallium) (diselenide, disulfide)—CIGS—represent an increasing share of the commercial market (~14% 2009 [2]).  This existing scientific and manufacturing infrastructure allows us to leverage insights from CIGS and CdTe research to improve the efficiency SnS-based devices and will reduce the capital expenditures associated to begin producing SnS-based devices industrially.

 

Our work on SnS focuses in three areas:

 

  • Developing annealing procedures to improve the structural quality and electronic properties of SnS films.  For these experiments, we employ a custom-built sulfur and hydrogen sulfide annealing furnace.
  • Developing fast vapor deposition techniques for the fabrication of high quality SnS films [3].
  • Together with Prof. Roy Gordon’s group at Harvard [4], we are investigating the interfaces of SnS with various n-type semiconductors to identify the optimal junction structure.

 

[1] M. Parenteau and C. Carlone, Phys. Rev. B 41, 5227 (1990).
[2] C.A. Wolden et al, J. Vac. Sci. Technol. A 29, 030801 (2011).
[3] K. Hartman et al, Thin Solid Films 519, 7421 (2011).
[4] P. Sinsermsuksakul et al, Adv. Energy Mat., DOI:10.1002/aenm.201100330.

Cuprous Oxide (Cu2O):

Sputtering

Cuprous or copper oxide (Cu2O) was one of the first semiconductors studied [1], and one of the first to be used in photovoltaic devices [2]. Cu2O has several properties that make it an attractive absorbing layer in a thin-film solar cell devices:

 

  • Its constituent elements are non-toxic and abundant enough to enable terrawatts of power generation.  Furthermore, as an oxide, it is chemically robust.
  • It absorbs strongly above 1.9-2.1eV, which is appropriate for combination with a lower bandgap material in a tandem cell structure similar to a-Si/μc-Si cells.
  • The ionic nature of its bonds suggest the potential for long minority carrier lifetimes.  This supposition is further borne out by the observation of long exciton lifetimes [3].

 

We recently developed a method of depositing high-quality Cu2O material using reactive magnetron direct-current sputtering [4], a relatively cost-effective process that can be used for large-area device fabrication.  While the film is polycrystalline with micron-size grains, the mobilities achieved at temperatures above 250 K are comparable to monocrystalline Cu2O , reaching 62 cm2/V·s at room temperature. Phase purity has been confirmed by x-ray diffraction measurements, and spectrophotometry indicates an optical bandgap in the range of 1.9-2.0 eV. We have successfully increased carrier density up to 2×1017 cm-3 and reduced resistivity down to 9.5 Ω-cm by nitrogen-doping without sacrificing optical properties.  We have also deposited very low resistivity contacts (~10-6 Ω·cm2) on the nitrogen-doped Cu2O films.  Our ongoing work focuses on optimizing the junction properties using n-type oxides as a window layer and a transparent top conductor—currently, ZnO and ITO—to maximize the efficiency of this all-oxide device.

 

[1] C. Starr, Physics 7, 15 (1936).
[2] E.D. Fabricius, J. Appl. Phys. 33, 1597 (1962).
[3] A. Mysyrowicz et al, Phys. Rev. Lett. 43, 1123 (1979).
[4] Y.S. Lee et al, Appl. Phys. Lett. 98, 192115 (2011).

 

 

Back to What We Do