Given the ability to reconstruct the J-V curves, what is the impact on the efficiency of the associated solar cells ?

What do your studies suggest with respect to finding an even better combination of fullerene or polymer as a replacement for these tried-and-true p-n components?

Our results imply that the p-n material morphology determines the electrical properties more than anything else (and quite strongly so), which suggests that research effort should be directed toward understanding and controlling the nanoscale structure of these polymer:fullerene blends. Our work also suggests that for this system the lowest unoccupied molecular orbital (LUMO) offset between the p and n components is nearly optimal — less of an offset would seem to promote more electron back transfer. Understanding why this is is of fundamental importance, because the LUMO-LUMO offset required for splitting excitons is an energy loss process that should be minimized.

How common is the observation of non-Langevian recombination in organic solar cells?

A Langevin “reduction factor” that can range from 1 (pure Langevin) to 10^-3 (“reduced” Langevin) is often used to equate the mobility and recombination constants. However, until now, no one has reported anti-Langevin recombination like we do above. We find that the mobility increases while the recombination simultaneously decreases with extra PCBM content. Typically, people find that the recombination and mobility scale together, but the “reduction factor” is needed to marry the two.

Thank you, Steven. Your results are truly unique and keep up the good work!

Thank you!

Since nanoscale morphology is so critical, how do es one control it? Interfacial free energy would be a critical parameter that one has little or no control over without altering the composition.

Experimentally researchers alter the nanoscale blend morphology by adding in small amounts (<3% v/v) of a solvent additive to the solution before casting and/or using post-production treatments like annealing or exposure to e.g., a solvent rich atmosphere.

Really, the bane of this field is measuring, controlling, and understanding what determines the film morphology. In the end it is extraordinarily difficult just to “look at” the nanoscale (<10 nm) structure because it is practically all amorphous conjugated carbon. Furthermore, it is difficult to single-out the effects of one particular material property (e.g., interfacial free energy) on device performance since so many things determine device performance. There is presently a significant research effort being directed toward measuring/imaging the morphology so that a detailed understanding of the structure-processing-properties relationships can be achieved.

Could you explain the terms geminate and nongeminate recombination and why one type might be preferred for your application?

Indeed, apologies for the jargon. Geminate recombination refers to recombination between electron-hole pairs that were created from the same photon — i.e., e-h pairs that never dissociated in the first place. Non-geminate recombination refers to everything else, which includes free-to-free, free-to-trapped, and surface (wrong contact) recombination. However, in the context of my data, we are referring to the former two.

Preferably one only wants nongeminate recombination since geminate recombination is a unique and additional loss process associated with excitonic materials; that is, materials that don’t create “free” carriers immediately upon photoexcitation. Almost all inorganic semiconductors create free charges upon photoexcitation, whereas all organic semiconductors first create excitons. Our data suggests that geminate recombination plays a minor role in determining the overall power-conversion efficiency of our devices. Instead, excessive free-to-free and/or free-to-trapped recombination are the primary loss processes that limit our (already relatively high) cell efficiencies. Moreover, it appears that the morphology is the dominating factor that governs the two aforementioned processes as well as the equally important charge carrier mobilities.

This is the type of understanding necessary to make OPV viable. Good job.

I think solar is great, and hopefully your work will make it a more viable energy source. I take issue with the idea that these can transform “unused land” into energy sources, and you show a photo of an awesome sagebrush-type habitat! I hate to get on my soapbox, but unused land is not often wasted! Why not stress that putting these in urban and suburban areas will reduce infrastructure requirements and save money! Less travel time for the electricity! There are lots of rooftops and parking lots that solar panels can be put on.

Robert: Thanks!

Maggi: I’m afraid I have to stand by my comment and respectfully disagree with some of your points. First, I agree that less travel-time for the electricity is good and there are lots of prime rooftops, etc. that would be great substrates for solar energy harvesting.

However, I think maybe my image choice was poor, because there is plenty of uninhabited land that sees lots of sun. Also, solar farms are usually quite small when compared to the surrounding territory. I think in the end one has to weigh the pros/cons of generating megawatts of clean renewable energy for society vs. destroying a very small portion of local habitat (that is preferably hardly inhabited). It seems shortsighted to me to disallow a solar farm in favor of, for example, mining/burning coal for energy.

Steven-
Looks like you’ve gained an important insight that can serve as the basis for optimizing device composition and morphology to improve efficiency.

RLG

Hi Steven, i interesting work! I’m curious if you have done these experiments with other materials or if you have an idea of how much these recombination losses are dependent on the specific donor/acceptor pair. Do you think using a different acceptor (or polymer) could reduce these types of recombinations and if so, is there an approach to specifically design/select them? Thanks!