Feb. 26, 2008 – Rohm and Haas Electronic Materials and IBM have agreed to jointly develop patterning materials and processes to enable implant at and below the 32nm node.

Implant technology will play an increasingly critical role at the 32nm node and below — the companies claim in a statement that ~40% of photo levels in advanced logic will be implant — but controlling implants in junction formation also is becoming increasingly challenging as designs shrink. “Finding the right solutions to difficult technical challenges for the next-generation nodes depends not only on strong engineering and design but also close collaboration with leaders in the semiconductor industry,” stated James Fahey, president of microelectronic technologies for Rohm and Haas Electronic Materials. “Partnering with IBM will accelerate the development of new materials and ensure that we are on track to meet the needs for 32nm and 22 nm nodes.”

Work will take place at IBM’s facilities in East Fishkill, Yorktown, and Albany, NY, as well as Rohm and Haas Electronic Materials’ Advanced Technology Center in Marlborough, MA, where a new ultrahigh-NA 193nm immersion cluster is expected to be ready by June 2008.

by Katherine Derbyshire, Contributing Editor, Solid State Technology

Though the vast majority of solar cells depend on p-n junctions in inorganic semiconductors, the limitations of those cells have inspired researchers to investigate a variety of alternate means for harnessing the sun’s energy. Thermophotovoltaic systems drive turbines with heat that would otherwise be lost. Organic semiconductors raise the intriguing possibility of liquid photovoltaics, as easy to apply as house paint. And dye-sensitized solar cells (DSSCs), one of the most successful alternative designs, depend on a mechanism that, according to inventor Michael Grätzel, is more akin to photosynthesis than to conventional photovoltaics.

Light absorption is a substantial challenge for any solar cell. Junction-based cells are most efficient when incoming light matches the material’s band gap. Light with less energy fails to excite free carriers; when light has more energy, the excess dissipates as heat. The DSSC, in contrast, separates light absorption from charge generation and transport. A dye layer, usually based on ruthenium-polypyridine compounds, absorbs light across a broad section of the solar spectrum. Excited electrons are then injected into a semiconductor. Grätzel’s innovation was the use of a semiconducting nanoporous TiO₂ layer. This structure’s large surface area accommodates more dye-infiltration than a dense film, and thus more carrier injection. The electrons make their way through the TiO₂ to one of the terminals of the cell.

As researchers at the U. of California/Berkeley, pointed out, electron transport through nanoparticle clusters is slow. One proposed scheme substitutes an array of crystalline ZnO nanowires, allowing much faster electron transport. ¹ Though interesting, the Berkeley structures achieved about one-fifth the surface area of nanoparticle films, and only 1.5% efficiency.

Another approach, investigated by Athanassios Kontos and colleagues at the Athens Institute of Physical Chemistry, combines titania powder with a titanium alkoxide sol in an attempt to produce a more homogenous mixture. Their cells achieved 1.9% efficiency, a substantial improvement over cells prepared with titania powder alone. ²

Injecting electrons into the semiconductor leaves the dye with a surplus of holes, and a net positive charge. An electrolyte solution, usually based on the iodide/tri-iodide (I-/I3) redox pair, replenishes the dye with electrons, becoming positively charged. The electrolyte transports the excess holes from the dye to the counter electrode, returning itself to the electron rich state.

Dye-sensitized cells have achieved efficiencies as high as 11%, comparable to many thin-film inorganic technologies. However, the best cell designs are poorly suited to the rigors of long-term outdoor installations. For example, the iodine-based electrolyte is a volatile liquid. Without adequate sealing, it leaks, decomposes, and evaporates.

Yet achieving high efficiency in commercially interesting cell designs has been challenging. Other materials struggle to match the performance of liquid electrolyte. The ionic conductivity of the electrolyte defines the rate at which the dye can replenish its supply of carriers — if the ionic conductivity is too low, current generation is transport-limited and the thickness of the cell must be reduced to compensate. Yet thinner cells absorb less light and yield less photocurrent. Similarly, the interface conductivity between the dye-impregnated TiO₂ structure and the electrolyte limits the performance of the cell. Liquid electrolytes can conform to the complex nanostructure.

Several groups have attempted to solidify the liquid electrolyte. For example, researchers at the Chinese Academy of Sciences (Beijing) mixed imidazole polymers with iodine and lithium iodide to form a freestanding gel. Clamping the material between the dye-sensitized TiO₂ electrode and a counter electrode allowed the gel to conform to the pore structure. Unfortunately, the diffusion constant of tri-iodide decreased rapidly as gel viscosity increased. Ions move by diffusion, so this behavior degraded the ionic conductivity as well. Still, the group’s results were promising, with reported efficiencies as high as 7.6%. ³

To summarize, DSSCs that deliver high efficiency in a practical, manufacturable package remain elusive. Still, optimized semiconductor structures and gel electrolytes suggest promising directions for further study. If these challenges can be met, these cells may offer a cost-effective alternative to traditional junction-based photovoltaics. K.D.

¹ Matt Law, et al., “Nanowire dye-sensitized solar cells,” Nature Materials, vol. 4 pp 455-459 (2005).

² Athanassios I. Kontos, et al., “Nanostructured TiO₂ films for DSSCS prepared by combining doctor-blade and sol