SOLAR TODAY - June 2011

innovators | Arthur J. Nozik, Ph.D.

20 June 2011 SOLAR TODAY solartoday.org

for women in defense plants, Nozik’s widowed mother joined the workforce. He was not yet old enough for kindergarten, so she parked him each day in Hebrew school. When he started public school a couple of years later, young Nozik continued studying the Torah and Talmud after hours. This scholarly discipline served him well when he fell in love with science. He was nine years old and already passionate about scientific inquiry when the world learned of nuclear fission in 1945.

While attending Classical High School in Springfield, Nozik worked a variety of after-school jobs to save money for college. He picked tobacco in the summer and after turning 16 drove an ice cream truck. With several scholarships to bolster his savings, he was able to enter Cornell in 1953. Strong in math, physics and chemistry, he settled on a five-year bachelor’s degree program in chemical engineering as a financially secure career path. Nozik took a break to work as a junior scientist at Monsan-to, returned to school and got married.

He graduated in 1959.

Nozik landed an engineering job with Douglas Aircraft working on rocket propellants (yes, that makes him a real rocket scientist.) Within a year he decided he preferred basic science, and joined the Ph.D. program at Yale in physical chemistry. Money was still an issue: When his first child was born in 1961, he went to work for American Cyanamid in Stamford, Conn. In 1964, Cyanamid awarded him a scholarship to complete the doctorate. Degree in hand, in 1967 he returned to Cyanamid.

Among other products, the company produced pigments, especially the prevalent white pigment, titanium dioxide. Nozik investigated some of its characteristics as a strong photo-oxi-dizer. He worked on magnetic semiconductors and coatings for information storage. He discovered a transparent semiconductor, cadmium stanate, which still is studied today as the best transparent conducting substrate in cadmium telluride solar cells.

In 1972, Akira Fujishima and Kenji Honda published a letter claiming that titanium dioxide was capable of photolysis (using light to split water into hydrogen and oxygen). The timing was great: The first OPEC oil embargo ensued a year later and research attention turned to non-fossil energy sources. Nozik joined Allied Chemical in 1974 to work on photoelectrochemistry. By

1975, he had built a working photolysis cell, using sunlight alone to split water at an efficiency of 0.5 percent. The cell used a semiconductor anode and cathode of titanium oxide and gallium phosphide, respectively. That led Nozik to invent the photochemical diode, a semiconductor structure that can split water and drive other specific chemical reactions when immersed in appropriate solutions.

With the launch of the Solar Energy Research Institute (SERI, now known as the National Renewable Energy Lab, NREL) in 1977, Nozik’s attention turned to other photon energy-conver-sion issues and especially to the flow of hot electrons. He joined SERI in 1978, and was appointed chief of the photoconversion branch in 1980.

system would be a nanocrystal. They proposed this solution in 1984. With Olga Mićić of the Boris Kidric Institute of Belgrade, Serbia, Nozik explored quantum effects in colloidal nanocrystal semiconductors, now also known as quantum dots. When the electron and the hole it leaves behind are created by light in such a small region of space, the electron and hole are attracted to each other, because they are oppositely charged. The coupled pair is called an exciton. After 2001, Nozik and Mićić focused on measuring hot-elec-tron cooling rates in these tiny crystals.

Along the way, Nozik said, the team “learned another way to utilize hot electrons.” Beginning in 1995, he theorized that within a quantum dot, a high-energy photon might excite two or three

Nozik theorized that if some of those hot electrons could be captured and conducted away before they recombined, photovoltaic cell efficiency might be roughly doubled, to about 66 percent.

In 1961, William Shockley and Hans Queisser described a natural limit for photovoltaic efficiency. They showed two forms of thermal losses. When you get enough hot electrons flowing in silicon, most of them will either move the wrong way (away from the collecting conductor) or will encounter holes previously occupied by other electrons and fall in — they recombine. Shockley and Queisser showed that the highest possible thermodynamic ceiling for power production in a simple one-layer silicon cell is roughly 32 percent.

Nozik theorized that if some of those hot electrons could be captured and conducted away before they recombined, photovoltaic cell efficiency might be roughly doubled, to about 66 percent. The goal would be to slow the cooling of the hot electrons before their extra energy was lost back into the lattice as heat. “You would need as many as 30 collisions with lattice vibrations sequentially in typical photovoltaic semiconductors in a few picoseconds or less to cool it,” he said.

Nozik and his colleague Ferd Williams real-
ized at SERI that if the electron could be con-
fined in a small enough space — about 1 nano-
meter ( 10-9 meters) — the time available for
cooling would increase to a point where the yield
of captured hot electrons would improve. The
electrons at a time, not just one. In effect, the
high-energy blue end of the solar spectrum can
double the number of electrons created. This pro-
cess in quantum dots is termed multiple exciton
generation (MEG) and was confirmed in 2004,
first at Los Alamos National Laboratory, then at
NREL and in other labs around the world. The
effect was initially controversial but Nozik says
that labs have shown MEG is a valid phenom-
enon. It has been demonstrated in quantum dots
formed from silicon, lead sulfide, lead selenium,
lead telluride, cadmium selenium, cadmium tel-
luride and indium phosphide, and various alloys
of some of these semiconductors.
Most of these materials lend themselves to
low-cost manufacturing processes, because they
self-assemble into thin colloidal arrays using
liquid-phase chemistry at relatively low tem-
peratures. “You make an ink, print it and dry it,”
Nozik said. His lab has tested an early lead sulfide
thin-film MEG ink that yields a solar photovoltaic
cell with 4. 4 percent efficiency. The long-term
target is to get 50 percent efficiency on a printed
module costing $150 per square meter. That’s
less than 30 cents per peak watt.

Arthur J. Nozik, Ph.D., was awarded the Gustavus John Esselen Award for Chemistry in the Public Interest, by the American Chemical Society, in April. ST