Researchers Analyze Behavior of Excitons in Quantum Dots

Extremely fast movement of energy over extremely short distances underlies the performance of solar cells, LEDs, and thermoelectric devices, but different devices demand different, and sometimes opposite, qualities. In the tiny world of nanostructured materials, William A. Tisdale, the Charles and Hilda Roddey Career Development Professor in Chemical Engineering at MIT, has found some surprising phenomena at work.

Using ultrasensitive spectroscopy and other techniques, Tisdale’s research group studies how excitons, which are paired groups of electrons and holes, behave in quantum dots. They measure minute changes that happen as quickly as 2 billionths of a second and cover a distance as short as several nanometers, or several billionths of a meter. Quantum dots are nanoscale crystals made up about of about 1,000 atoms each, and they have unique properties.

Seemingly infinitesimal differences can make a big difference in device performance because kinetics, the rate at which reactions occur, can determine which of several possible outcomes will happen. Chemical engineers use the same mathematical framework to analyze dynamics of excited states in quantum dots as they use to analyze networks of chemical reactions. "The fastest thing that can happen is the most probable thing to happen," Tisdale explains. "Engineering nanoscale materials to do what you want them to do is all about getting the things you want to happen, to happen quickly, and trying to slow down as much as possible the bad things, so that's really the name of the game in all of this work we're doing."

Degrees of freedom

Energy can move within and across materials by the flow of electrons producing electric current, by emission of a photon producing light, by vibrations producing heat, or by excitons, which are important because they can yield either light or direct-energy transfer. "I love working with quantum dots (QDs) because there are so many degrees of freedom that you have with the materials. You have the size, the shape, the elemental composition. You have the ligands on the surface," Tisdale says.

Groups, or ensembles, of quantum dots can be ordered or disordered; they can be all of one size or of different sizes. Their atomic planes can be aligned or random relative to neighboring nanocrystals. Ligands, which are organic molecules bound to the surface of quantum dots, can be polar or non-polar and can differ in both length and functionality.

"You can blend different types of quantum dots together, and so you have all these different degrees of freedom to control a system. What I really enjoy doing is, in a detailed way, finding out how all these different parameters are affecting what's going on. You develop this overall picture of realizing, okay, this is something I can control, this is something I can't control, this is something I need to suppress. Once you have that knowledge, then you can make the material do what you want it to do," Tisdale explains.

"In principle, you could make a material that we hope can transport electric charge very easily but engineer it to simultaneously resist the flow of heat. That would make a very good thermoelectric material, which would be something that would be good for converting waste heat into electricity. Or (whether) we can engineer these quantum dot films to be very, very good at transporting excitons versus very good at splitting apart excitons and transporting free charges. In one case, the free charges, you want that for a solar cell; you want excitons for an LED, a light-emitting device," he says.

"In quantum-dot LEDs, they use quantum dots with long ligands, and in quantum-dot solar cells, they use quantum dots with short ligands in order to tune the exciton diffusion length," says A. Jolene Mork, a fifth-year MIT graduate student in chemistry, and lead author of a Journal of Physical Chemistry paper that analyzed energy transfer in colloidal quantum dots.

Tunable color emission

In a study of energy transfer to the semiconductor molybdenum disulfide (MoS2) from inorganic quantum dots consisting of a cadmium selenide core with a cadmium-zinc-sulfur shell and attached organic ligands, Tisdale’s group showed such a system has the potential to customize the color of light. “By varying the size of quantum dots on the surface of MoS2, you could change the color of light that this hybrid material emits,” Tisdale says.

In the experiment, a laser provided energy to stimulate the quantum dots and researchers measured light emission to determine how much energy was remitted as light and how much was passed from the quantum dots to the molybdenum disulfide as a direct-energy transfer.

“When a semiconductor absorbs light, you are promoting electrons from ground-state energy levels to excited-state energy levels, so that electron has some energy, and then it can release that energy by relaxing back down to the ground state,” Tisdale explains. In relaxing to its ground state, the electron can release a photon that is visible, called fluorescence or luminescence, or it can transfer its energy to an electron in a neighboring quantum dot or other material, raising its neighbor to an excited state. In an LED, for example, an electron and hole combine, and they release light. “You inject current, and you get light out,” Tisdale says.

After an exciton forms and before it emits light, the exciton can move around as an excited state that can be passed from one quantum dot to another, from one molecule to another or from a quantum dot to a two-dimensional material such as MoS2. It can still eventually emit light from another location. “So you could have one thing absorbing and then transfer its energy to another thing, which then ultimately emits that light,” Tisdale says.

In the quantum dot/molybdenum disulfide system, 95 percent of the energy from excitons formed on the quantum dots was transferred to the molybdenum disulfide without light emission from the quantum dots, a phenomenon known as quenching. Just 5 percent of the excitons stayed on the quantum dots and were remitted as light.

While molybdenum disulfide is a good electrical conductor, quantum dots are not always good electrical conductors. The high-energy transfer efficiency shown in the study could be useful in applications such as photodetectors, using quantum dots to enhance the absorption of highly conductive molybdenum disulfide. However, Tisdale is most excited by the potential to tune the color of light emitted by an MoS2/QD system, because the color of light emitted from quantum dots is directly related to their size. “You would inject electric current into MoS2, form an exciton in MoS2 that would then transfer to the quantum dots and emit light, so in that way you could use the quantum dots to tune the color of your semiconductor,” Tisdale says.

Surprising findings

The quantum dot/molybdenum work also showed the unexpected result that the thinnest layer of molybdenum disulfide yielded the highest energy transfer from the quantum dots, transferring energy at triple the rate compared with bulk MoS2. "It was a very surprising result, and we sat on our data for months because we couldn't explain it. We weren't sure if it was because it was an artifact, or some sort of anomaly, or something real. We eventually grew more and more convinced that we were measuring something real. We still didn't have an understanding of why it was behaving that way," Tisdale explains.

"Normally the way energy transfer works is if I'm a donor and I have four or five people sitting around me, I could transfer my energy to any of them, and the more people around me, the faster I'd be giving up my energy, because each of them would independently contribute an additional pathway of energy transfer. The prediction from all of the existing energy-transfer theory is that if I had more layers of molybdenum disulfide, I would get faster energy transfer because if I have just one layer I can only transfer to that layer, but if there is another layer behind it, I could transfer to the first layer or the second layer. If I've got three layers, I'd go first, second, third. But what we saw was the opposite. Actually the fewer layers we had, the faster the energy transfer was. And that was a very scientifically surprising result," Tisdale explains.

They found a theoretical explanation in a paper by J. M. Gordon and Yu N. Gartstein, from the University of Texas at Dallas Department of Physics, which was published September 2013 in the Journal of Physics: Condensed Matter. "All of the quantitative predictions laid out in that theory matched our experimental results exactly," Tisdale says. "It turned out that the theory applied very appropriately." The explanation for the fastest transfer happening in the thinnest molybdenum disulfide layer has to do with exotic dielectric screening behavior that happens in very thin semiconductor materials, he says. The phenomenon was predicted in the Gordon-Gartstein paper, but not specifically for an MoS2 system, Tisdale says.

Understanding exciton diffusion

The process of excitons moving, or "hopping," is known as diffusion, and both the distance excitons travel and their lifetime affect potential applications. A collaborative study among professors Tisdale, Vladimir Bulovic, and Adam Willard of diffusion in quantum dot solids measured exciton lifetimes and modeled exciton diffusion lengths.

"What we learned is you want to make the center-to-center distance as small as possible to have a longer diffusion length, to maximize your diffusion length," says Mark C. Weidman, a fourth-year MIT chemical engineering graduate student and co-author of the study.

Faculty mentorship

A former postdoctoral associate in Vladimir Bulovic's group in the MIT Research Laboratory of Electronics, Tisdale and Bulovic have continued to collaborate. With Bulovic, Tisdale was part of the group of MIT researchers who reported the first observation of excitons in action in April 2014. Techniques such as time-resolved photoluminescence spectroscopy developed for that research have continued to inform ongoing projects in the Tisdale Lab. "I think it's a little bit less common for a lot of these spectroscopy techniques that we're using to be used in chemical engineering departments, and that is a role that I am excited to play in helping to bring a lot of these advanced spectroscopy techniques to bear on problems that chemical engineers care about," he says.

Some of Tisdale's projects have used quantum dots provided by QD Vision, a company spun out of Bulovic's lab, but Tisdale says QD Vision does not pay consulting fees or sponsor or fund research, so there is no conflict of interest. "There's not a particular result they are looking for from us. They're just giving us materials and we use them for whatever experiments we want to use them for. Sometimes we see some interesting stuff that gets them excited; sometimes they don't care at all about what we're doing with them."

Tisdale says he has frequent conversations with Seth Coe-Sullivan PhD '05, co-founder and chief technology officer of QD Vision, who co-invented the technology at MIT. "It's really nice to have that contact with someone actually doing this in an industrial setting, thinking about commercializing these technologies, thinking about where the opportunities are and what the scientific challenges are and how we as an MIT research group can play a role in advancing the field," Tisdale says.

Tisdale has also worked with the Materials Processing Center / Center for Materials Science and Engineering Summer Scholars program. In the past two years, high-school seniors Megan Beck and Sarah Arveson worked as interns in the Tisdale Lab during the summers of 2013 and 2014, respectively. Beck was co-author of a paper with Weidman on controlled synthesis of lead sulfide quantum dots, and Arveson worked with postdoctoral associate Pooja Tyagi on excitonic properties of organometal halide perovskites.