The Renaissance of Iron Pyrite Photovoltaics: Progress, Challenges, and Perspectives - Low-cost Nanomaterials: Toward Greener and More Efficient Energy Applications - page 136

1.2 Pyrite Crystal Structure and Properties
Pyrite (FeS 2 , Fool's Gold) crystal structure is one of the best examples of cubic
AB 2 structures. This structure is mainly characteristic of AB 2 compounds of
pnictides, such as P, As, Sb, and chalcogenides, such as S, Se, Te. Since FeS 2 is the
most prominent example of this structure, pyrite has been used to name this
family. In this section, we focus mainly on FeS 2, but this structure can be used for
the other members of this family. The pyrite structure can be represented by
imagining the NaCl cubic crystal structure and replacing the sodium with the iron
atom and the chlorine atom with dumbbells of the S 2 dimer compound. The iron
atom is in a distorted octahedral coordination site surrounded by six sulfur atoms,
while the sulfur atoms sit in a distorted tetrahedral coordination surrounded by
three other sulfur atoms and one iron atom. The distortion of the pure NaCl
structure reduces the symmetry of the pyrite structure, putting in the Pa3 point
group. Lattice constants of pyrite is found to be 5.418 A with Fe-S distances being
2.26 A and S-S distances being 2.14 Å [ 4 ]. It is important to note that FeS 2 can
also crystallize into an orthorhombic structure named marcasite. While the pyrite
structure has corner linked coordination octahedral, marcasite exhibits edge linked
octahedral. Marcasite offers different properties, which are usually detrimental to
pyrite's promising characteristics for solar application.
From crystal field theory, it is known that transition metal's d orbitals are
nondegenerate in an octahedral environment. In pyrite, the t 2g set made of the d xy ,
d yz , d xz orbitals controls the valence band and the e g set made with the remaining
d 2 and the d s 2 -y 2 orbitals control the conduction band. Since iron is in an oxidation
state of 2+ in pyrite, this leaves six electrons remaining to fill up the three t 2g
orbitals, making it a diamagnetic low-spin semiconductor. The splitting between
these two orbital sets controls the band gap of the semiconductor and for pyrite it
has been found that this gap is 0.95 eV. While this band gap has mostly been
accepted, solar devices made with this material exhibit low open circuit voltage,
which is mostly attributed to sulfur vacancies in the crystal that pin the Fermi
levels. Recently, there have been computational studies that have started debates in
the literature on whether pyrite is a purely stoichiometric compound [ 19 - 22 ].
While this is beyond the scope of this chapter, it is important to know that this is
due to its affect on creating working devices.
It has been shown both in computational studies and shape control studies of
nanocrystal pyrite that the equilibrium faces that can be obtained are {100} and
{111} [ 21 , 23 - 25 ]. Other facets, such as {110} and {210}, can be obtained in
macroscopic crystals, but focus is given to the nanocrystal formations in this work
[ 21 ]. It is necessary to first note that pyrite crystal surface energies are an enigma
compared to normal crystal theory. It has been shown that the {100} face is lower
in surface energy than the {111} energy, which is the opposite of normal theory
[ 26 ]. This leads to many different observables in synthesis, which will be discussed
later. Here, focus is given on the difference between the two facets. Figure 1 shows
models of both (100) and (111) surfaces. In all cases the {111} surface is sulfur