LBNL Masthead A-Z IndexBerkeley Lab mastheadU.S. Department of Energy logoPhone BookJobsSearch
Mapping the Nanoscale Landscape Print
Wednesday, 27 September 2006 00:00

For the first time, researchers have successfully mapped the chemical structure of conjugated polymer blend films with a spatial resolution of better than 50 nm using scanning transmission x-ray microscopy (STXM). This is not just another application of STXM. It is a breakthrough experiment on several levels. Correlating local composition to electronic/optical device characteristics will pave the way to characterizing a whole new class of materials with STXM—multicomponent organic electronic devices that have intrinsically nanoscale dimensions. Understanding where charge transport and recombination occur in these materials helps explain the efficient performance of polymer-based light-emitting diodes (LEDs) and will lead to a new avenue of research on organic electronic devices, supporting emerging technologies such as molecular computing and promoting increased efficiencies in existing organic technologies (organic LEDs and solar cells).

Go Organic with OSCs

What would we do without electric power? More than a current of electrons converted into heat, light, and motion, it is an essential infrastructure. Much of society's electricity is generated at fossil fuel plants. Fossil fuel is a form of stored solar energy. The energy of the sun is "captured" through photosynthesis, plants and animals are converted over eons to fossil fuel, then we extricate it from geologic deposits. However, fossil fuel use contributes to increases in CO2 emissions, leading to global warming. So why not go directly to the source? No messy extraction, just direct conversion of photons into electricity.

Organic solar cells (OSCs) are the most promising new candidates for low-cost photovoltaics. Presently, OSCs and LEDs based on blends of semiconducting polymer and fullerene derivatives exhibit power conversion efficiencies of 3% under solar conditions and quantum efficiencies of up to 70%. The performance of these devices involves a complex balance between charge generation and charge transport. Studying the active layers of OSCs logically follows upon the nanometer characterization of TFB/F8BT-based LED devices and will be critical to the improvement of performance and energy efficiencies. Although just a third as efficient as fossil fuel, clean green OSCs have the potential for large-area solar collection—on rooftops of buildings and other unused spaces. And every increase in efficiency will reduce the amount of device space needed.

Conjugated polymer blends are good candidates for low-cost optoelectronic devices due to their stability, efficient electroluminescence, and photovoltaic performance. Spin-coating demixes the two polymers, producing a distributed heterojunction structure with interfaces for charge recombination (LEDs) or charge separation (solar cells). These films, however, have a complex nonequilibrium structure, making device function interpretation difficult.

Chemical structures of F8BT and TFB along with their NEXAFS spectra. Although TFB and F8BT exhibit similar NEXAFS spectra above 287 eV, between 283 and 287 eV, the π* absorption peak of F8BT has a broader, weaker structure with a lower absorption onset, consistent with the deeper lowest unoccupied molecular orbital of F8BT compared to TFB. (These spectra analyses formed the basis for the observed chemical contrast in the STXM maps.)

To better understand this structure, an international group of scientists performed STXM measurements at Beamline 5.3.2 at the ALS on blend films of poly(9,9-dioctylfluorene-co-N-(4-butylphenyl)diphenylamine) (TFB) and poly(9,9'-dioctylfluorene-co-benzothiadiazole) (F8BT). STXM, in conjunction with the chemical sensitivity of near-edge x-ray absorption fine-structure (NEXAFS) spectroscopy, allows for quantitative determination of submicron polymer blend composition.

At a lateral resolution of 50 nm, the researchers mapped three thickness layers of subsurface chemical structure (65 nm, 95 nm, and 150 nm). Within the enclosed TFB domains, there is little lateral variation in composition (e,h). In the F8BT region, however, F8BT concentration near the domain interface is high (d,g), measuring 90% (although a TFB wetting layer between blend and substrate indicates this concentration is even higher). F8BT concentration falls to 60% further from the interface and into the domain. The domain interface measures 200 nm, but may be sharper due to additional three-dimensional structure. Such a large decrease in F8BT concentration is not caused by the presence of a TFB surface capping layer, since it would be too thick. These features arise from variations in bulk intermixing between F8BT and TFB.

STXM composition maps (5 µm x 5 µm) of F8BT:TFB blend films (left and center). Comparative atomic-force microscopy (AFM) surface images (right) reveal micrometer-sized domains in blend films deposited from xylene. Both AFM and STXM show similar coarse domain sizes, 1 to 5 µm, confirming domain structures are columnar and round. The left column corresponds to F8BT weight % composition maps; the center to TFB weight % composition maps. (a), (b), and (c) are of a 64-nm-thick film; (d), (e), and (f) are of a 95-nm-thick film; (g), (h), and (i) are of a 150-nm-thick film. The scale bar is 1 µm.

These results will help clarify nanoscale mechanisms of electroluminescence in LEDs based on these blends. Previous studies showed that injection of electrons from a cathode is most efficient into an F8BT-rich domain near a domain interface not covered by a TFB capping layer. The injection of holes in contrast is facilitated by a TFB wetting layer that covers the entire anode, providing efficient injection across the plane of the device. As demonstrated here, however, the F8BT-rich domain is almost pure near the domain boundary, thus the efficient injection of electrons into the near-domain F8BT-rich region is accompanied by efficient charge transport through this almost pure bulk region of the film. Charge capture is then facilitated by charge recombination at either the interface between the micrometer-sized domains or at the horizontal interface between the TFB-rich wetting layer and the nearly pure near-domain F8BT-rich region. As injection of electrons and electron transport are efficient only near the domain interface, both charge capture mechanisms could explain why electroluminescence is only observed near the micrometer-sized domain interface by optical microscopy.

Interfacial enrichment of F8BT, which facilitates efficient charge transport, and interface sharpness, which promotes charge capture and recombination, help explain the efficient performance of TFB/F8BT-based LEDs and demonstrate the exciting potential for this technique to probe the composition, morphology, and electronic processes of polymer films.


Research conducted by B. Watts, L. Thomsen, W.J. Belcher, and P.C. Dastoor (University of Newcastle, Australia); C.R. McNeill and N.C. Greenham (University of Cambridge, U.K.); and A.L.D. Kilcoyne (ALS).

Research funding: U.K. Engineering and Physical Sciences Research Council, Australian Research Council, and Commonwealth of Australia. Operation of the ALS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences.

Publication about this research: C.R. McNeill, B. Watts, L. Thomsen, W.J. Belcher, N.C. Greenham, and P.C. Dastoor, "Nano-scale quantitative chemical mapping of conjugated polymer blends," Nano Letters 6, 1202 (2006).