IEEE Spectrum July, 2014 - 17

tOP: University Of OxfOrd; bOttOm: erik vrielink

Piles of Perovskites: The photovoltaic
wunderkind's Achilles' heel is its lead content, but
laboratories have found some success using tin instead.

sity, in Evanston, Ill. "The field has grown
by leaps and bounds."
Solar cell manufacturers face a tricky tradeoff between performance and cost. Most commercial solar cells rely on slabs of crystalline
silicon that are more than 150 micrometers
thick and take a lot of energy to produce.
Thin-film solar cells-those containing just
a few micrometers of such semiconductors
as copper indium gallium selenide (CIGS)-
have lower material costs, but they are also
less efficient. Cells using crystalline gallium
arsenide, on the other hand, can reach
30 percent, but the materials involved are
too costly for utility-scale solar power.
Perovskites could resolve this quandary
by matching the output of silicon cells at a
lower price than that of thin-film CIGS: Their
ingredients are cheap bulk chemicals, and
the cells can be built using simple, low-cost
processing techniques.
Perovskites made their debut in dyesensitized solar cells. When light is absorbed
by a dye in the cell, it injects excited electrons into a semiconductor such as titanium
dioxide nanoparticles, which carry the
charge away and ultimately generate current. Perovskites made attractive replacements for dyes because they absorb light
efficiently over a broad spectrum, but researchers soon realized that they are also
excellent charge carriers in their own right.
In 2013, Snaith unveiled a cell using
a layer of perovskite without titanium

dioxide nanoparticles, simplifying the cell's
architecture and pushing its efficiency above
15 percent.
Since then, Sang Il Seok at the Korean Research Institute of Chemical Technology, in
Daejeon, has refined the chemical reaction
that forms the perovskite layer to smooth defects out of its structure, which pushed his
cells' efficiency to 17.9 percent. And although
Yang declined to discuss the technical details
of his 19.3-percent cell, pending publication,
other researchers say that removing defects
was also crucial to his success. Further gains
in performance could come from tweaking
perovskites' chemical compositions to broaden the material's absorption spectrum. "Progress has been so fast," says Seok.

aBout faCe: Perovskites have a distinct structure,
although their chemical compositions vary. In one
photovoltaic crystal, lead [blue] lies at the center
of the structure, with ammonia ions [purple] at the
vertices and iodines at the center of each cubic face.

However, there are lingering concerns that
a widespread use of lead-based cells could
pose environmental problems, making investors somewhat wary of the technology.
Snaith points out that annual lead emissions
from coal combustion are 10 times as much
as the amount of lead that would be needed
for a 1-terawatt array of perovskite solar cells,
but he acknowledges that "it would be better to have completely nontoxic materials."
In May, Snaith and Kanatzidis independently reported a possible solution: tin perovskites
that managed 6 percent efficiencies. Although
the tin perovskites are much less stable in air
than their lead counterparts, both scientists
are convinced that they have potential.
As competition in the lab grows ever more
intense, a parallel commercial race is heating up. Oxford Photovoltaics is in the lead
for now, but "a lot of other companies are
starting to look at this," says Snaith.
Oxford is working with glass manufacturers
to create photovoltaic glazing products. Windows coated with the company's perovskite
have a gray tint and generate electricity with
6 to 8 percent efficiency. "That's more than
adequate to meet builders' expectations,"
says Chris Case, Oxford's chief technical officer. It costs about 10 percent more than normal glass, and the electricity should pay back
the cost of the windows within 10 years. The
company expects the perovskite windows to
be produced "in volume" by the end of 2017.
Oxford is also working on utility-scale
products and investigating whether
perovskites could be teamed with conventional silicon solar cells. The materials absorb different wavelengths of light,
and perovskites generate a higher voltage
than silicon, so using them in tandem would
boost the power output of conventional cells.
The company has a development team of
24 people, backed by £7 million in financing.
With perovskite efficiencies rising by the
month, Oxford is now aiming to raise more
money and expand. "It's not a cottage industry," says CEO Kevin Arthur. "We are going gangbusters on this." -mark peplow

nEwS

SPECTRUM.IEEE.ORG

nORTh aMERICan

JUL 2014

http://SPECTRUM.IEEE.ORG

Table of Contents for the Digital Edition of IEEE Spectrum July, 2014