If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



The Power Of Plastic

Polymer- and organic-molecule-based solar cells show promise as low-cost power generators

by Mitch Jacoby
August 23, 2010 | A version of this story appeared in Volume 88, Issue 34

Like a one-two punch combination, “cheap” and “plastic” are sometimes hurled one after the other at manufactured goods to denote poor quality—as in “cheap plastic toy” or “cheap plastic car parts.” Although the C and P words are used in those cases pejoratively, when it comes to certain types of solar cells, cheap and plastic are precisely what make them so attractive.

Any type of device that converts light to electricity holds promise for tapping the essentially inexhaustible and non-carbon-emitting energy supply that flows from the sun. Devices that mediate that energy conversion by way of light-sensitive organic polymers or other organic molecules may be especially attractive because of the low cost of the materials and manufacturing methods required to produce such cells.

Inexpensive polymer-based solar cells, which are also known as organic photovoltaic cells, already exist; they have been around since the 1990s. But their performance, and in particular the efficiency with which they convert light to electricity—typically on the order of 5%—has remained much lower than the 10–15% conversion efficiencies provided by costly traditional solar cells based on silicon, cadmium-telluride, and other inorganic semiconductors. That single-digit value pales even further when compared with some highly specialized and pricey state-of-the-art inorganic research devices that yield conversion efficiencies topping 40%.

Recently, however, scientists have begun closing the performance gap, albeit in small increments, through an enormous surge in organic solar-cell research. The studies are aimed at designing new types of semiconducting polymers and other novel compounds, improving processing conditions, and developing a more complete understanding of the basic phenomena that govern the performance of organic photovoltaic devices.

Proponents of solar technology often point out that the quantity of solar energy impinging on Earth’s surface in an hour is greater than the total energy consumed by humankind in a single year. With statistics as staggeringly powerful as that one, “we simply cannot ignore the opportunity that is presented by using solar energy,” says Russell A. Gaudiana, vice president for research at Konarka Technologies. Based in Lowell, Mass., Konarka develops and manufactures a line of lightweight and flexible organic photovoltaic panels for portable power use and other applications.

The market opportunities also cannot be ignored. Overall, the photovoltaics industry generated some $37 billion in global revenues in 2008, according to Solarbuzz, an international solar energy research and consulting company. That value rose to $38.5 billion in 2009, the company reports. Nearly all of that revenue, however, comes from sales of inorganic solar-cell products such as rooftop panels.

“For now, we can safely claim that organic photovoltaics has a nearly zero percent share of the market,” says Yang Yang with a note of humor. Yang, an organic electronics specialist and professor of materials science at the University of California, Los Angeles, explains that nearly all organic solar-cell companies are in the research and development phase and not yet selling products.

Although the field has not yet begun to stimulate much sales activity, in terms of research, it’s flourishing. During the past few years, several thousand journal papers and conference proceedings have been published on organic solar cells. For cells based on common inorganic semiconductors, the number of publications during the same period is about an order of magnitude lower.

One reason organic solar cells are such a hot topic, in Yang’s view, “is that the threshold to entering this field of research is very low.” He explains that scientists who want to conduct studies in this area can easily buy relatively inexpensive organic compounds, investigate their properties by using common laboratory tools, and publish the findings. In contrast, working with some types of inorganic semiconductors requires specialized equipment and advanced handling procedures.

But perhaps the biggest reason that research in organic photovoltaics is booming is the magnitude of the potential payoff. “Solar energy only accounts for a small percentage of our energy production today largely because photovoltaic technology is prohibitively expensive,” according to Alex K-Y. Jen, a chemistry and materials science professor at the University of Washington, Seattle. Inorganic photovoltaic products, and in particular high-performance devices, are very costly because of demanding and energy-intensive crystal-growth, device-fabrication, and associated manufacturing processes.

That’s where organic chemistry may be able to offer a key advantage over its inorganic counterpart, Jen says. If device performance can be improved, then organic solar-cell technology may become economically viable by taking advantage of low-cost solution-phase processing and mass production via commercial-scale roll-to-roll printing methods.

In addition, unlike traditional rigid solar panels, lightweight and flexible plastic panels can be incorporated into a wide variety of nontraditional and irregularly shaped products. Examples include backpacks and other types of handbags that can charge laptops, mobile phones, and other portable electronics while the user is on the go. Several of these types of products, which feature Konarka solar panels, are available today from various retailers. Other products based on “smart fabrics” that turn clothing, tents, umbrellas, and other ordinary items into power generators are under development by a number of companies, as are polymer solar panels that are designed to be incorporated into windows, skylights, walls, and other building surfaces.

Regardless of the shape and intended application, all photovoltaic cells have common features. At the heart of a solar cell is a light-sensitive semiconducting material in which the first steps of the power generation process occur. In traditional solar cells, that material is typically silicon. In an organic photovoltaic device, the photoactive region generally consists of two materials. One serves as a light-harvesting electron donor and the other as an electron acceptor.

Photons impinging on that region can cause electronic excitations in the donor material leading to formation of excitons. Excitons are high-energy couples in which an energetic electron is bound to a positively charged electron vacancy or hole. To produce electrical current, the electron-hole pair must migrate to the interface between the electron donor and acceptor materials, where it splits into separate mobile charges. The charges then diffuse to their respective electrodes—electrons are transported by way of the electron acceptor material to the cathode, and holes travel via the electron donor to the anode. Overall, the charge-transfer and charge-transport processes generate a flow of electrical current that can be used as a power source.

Credit: Konarka
Made from light-sensitive polymers, lightweight flexible photovoltaic sheets (such as the Konarka product shown here) are converting ordinary surfaces into inexpensive power generators.

Researchers have experimented with a variety of electron donor and acceptor materials as well as with ways of pairing them. In the 1990s, a team led by chemistry Nobel Laureate Alan J. Heeger, a professor at UC Santa Barbara, discovered that photoinduced electron transfer in composites of conducting polymers (electron donors) and C60-type compounds (electron acceptors) proceeds much more efficiently than in the absence of the fullerenes. The early studies focused on phenylene-vinylene polymers and functionalized fullerenes, including a phenyl-butyric acid-substituted C61 compound known as PCBM.

One way to physically bring the acceptor and donor materials together is to deposit a layer of one material on the other. In semiconductor parlance, the planar interface between the electronically dissimilar materials in that arrangement is known as a planar heterojunction. That geometry is sometimes used to fashion solar cells from nonpolymeric organic compounds. The planar design is simple. But because the interfacial area is small and fixed, it’s difficult to tweak the cell’s performance—in particular the conversion efficiency.

To greatly increase the interfacial area, Heeger’s team blended the polymer and fullerene components in a way that formed an interpenetrating bicontinuous network of donor-acceptor junctions. In that arrangement, which is now referred to as a bulk heterojunction (BHJ), the polymer and fullerene phases are intermingled on the nanometer scale.

Gaudiana likens the morphology of a BHJ active layer to a sponge. The solid part represents the nanosized interconnected bits of fullerene. The polymer is represented by the holes, which are intimately connected to other holes throughout the sponge and never far from a solid region. Blending the phases on that scale, in effect, distributes small regions of interface throughout the photoactive layer. As a result, excitons need to diffuse only a short distance before quickly reaching a donor-acceptor interface where they can dissociate into separate charges.

Once they separate, electrons and holes follow tortuous paths—hopping repeatedly from one nanosized domain of fullerene (or polymer) to the next until they reach their respective electrodes. It’s a complicated charge-transfer and charge-transport mechanism, but thus far nearly all of the top-performing organic solar cells—the ones providing the greatest conversion efficiencies—have been based on BHJs.

In the 1990s, by applying the BHJ strategy to the phenylene-vinylene-PCBM system, the Santa Barbara group produced solar cells that yielded conversion efficiencies of some 3%, which was an outstanding value at that time for organic photovoltaics. Since then, functionalized fullerenes including PCBM have remained the electron acceptor materials of choice. In contrast, many types of polymers have been studied since then but poly(alkyl-thiophenes)—and in particular poly(3-hexylthiophene) (P3HT)—paired with PCBM have emerged during the past five years as some of the most important and best-studied systems. Various groups have reported attaining 5% conversion efficiencies from cells made from P3HT-PCBM.

In addition to thiophenes, various types of fluorene-, carbazole-, and cyclopentadithiophene-based copolymers have been studied widely by researchers in academia and industry. Although companies tend to keep quiet about the specifics of the solar-cell materials they study, Konarka researchers acknowledge that in addition to other promising materials, they have examined the P3HT-PCBM system and the cyclopentadithiophene class of polymers in detail. A team of the company’s leading scientists surveyed results from those organic solar cells and others in a review published in Advanced Materials (2009, 21, 19).

Heavy lifting has been required to raise the performance of these lightweight solar cells above the best results reported just one or two years ago. One research thrust focuses on improving the cells’ operating characteristics by customizing the electron energy levels of the polymers with respect to those of the fullerene. Tuning the levels appropriately could enhance exciton dissociation kinetics and raise the value of a solar-cell parameter known as the open-circuit voltage.

With those goals in mind, University of Chicago chemistry professor Luping Yu teamed up with UCLA’s Yang to design and test a series of novel copolymers made by reacting a benzodithiophene compound with various thienothiophenes. The team aimed to lower the polymers’ highest occupied molecular orbital (HOMO) levels by attaching successively stronger electron-withdrawing groups to the polymer backbone.

The strategy worked. By replacing an alkoxy group that was adjacent to a carbon­yl group with an alkyl chain at the same position, the group lowered the HOMO level by roughly 0.1 eV. They lowered the level by another 0.1 eV by adding a fluorine atom. Then the group paired the novel polymers with PCBM to prepare solar cells. Consistent with the trend in the customized electron energy levels, the solar cell containing the fluoropolymer yielded the best results—and a record-breaking conversion efficiency of roughly 6.8%, as certified by the National Renewable Energy Laboratory, in Golden, Colo. (Nat. Photonics 2009, 3, 649). In follow-up work on this family of polymers, Yu’s group recently reported slightly higher conversion efficiencies (just over 7%) (Acc. Chem. Res., DOI: 10.1021/ar1000296).

Banking on a belief that academic success can spawn commercial success, organic photovoltaics start-up Solarmer Energy licensed a portfolio of technology developed in the UCLA and Chicago laboratories and recently set up shop in El Monte, Calif. Vishal Shrotriya, a research director at Solarmer, notes that in 2009 the company set three certified world records and just beat the best of the trio last month with a solar cell that has an efficiency of 8.13%. “We are definitely on target to reach 10% by the end of 2011 or even sooner,” Shrotriya says. The company expects to begin selling products for powering portable electronics and solar cells incorporated into smart fabrics next year.

Credit: Adapted from J. Phys. Chem. Lett.
Credit: Adapted from J. Phys. Chem. Lett.

In addition to tuning electron energy levels, scientists are trying to improve organic solar cells by stabilizing BHJs. University of Massachusetts, Amherst, chemistry professor Dhandapani Venkataraman explains that the BHJ structure is controlled by subtleties in molecular architecture, intermolecular interactions between polymers and fullerenes, and the packing propensities of the molecules’ π-conjugated moieties.


Regarding packing, for example, the hexyl chains in annealed (heat treated) P3HT films form lamellar structures that pack via π-π interactions with a spacing of about 3.5 Å, which is roughly half the packing spacing in PCBM, he says. As a result of the packing mismatch, there is very little order in unannealed films of the blend. Overannealing the films leads to unwanted phase segregation (too much order) and forms highly crystalline regions. Only gently annealed blends adopt the BHJ structure with the molecular order suitable for solar cells (J. Phys. Chem. Lett. 2010, 1, 947).

Controlling molecular order in solar-cell films is one of the key objectives in research led by Scott E. Watkins and Gerard J. Wilson of the Commonwealth Scientific & Industrial Research Organization (CSIRO), Australia’s national lab agency. Just recently, the team used a free-radical polymerization method designed for making well-defined block copolymers to prepare a series of novel benzothiadiazole-containing pendant polymers. Among other findings, the team observed on the basis of atomic force microscopy that films of the block copolymers are far smoother and more ordered than films made from blends of the two small-molecule pendants or blends of the two homopolymers (Macromolecules DOI: 10.1021/ma1008572).

In related work, David J. Jones and Wallace W. H. Wong of Australia’s University of Melbourne, together with CSIRO scientists, showed that using compounds such as hexabenzocoronenes for harvesting light and transporting holes in solar-cell films is advantageous because these small molecules self-assemble, which enhances the rates of those processes (Adv. Funct. Mater. 2010, 20, 927).

Although much of the publicity surrounding solar-cell research focuses on experimental findings, it’s no surprise that computational experts are using their computers to search for promising candidate molecules. It may be surprising, however, to learn that they are doing much of the job with other people’s computers. At Harvard University, Johannes Hachmann, a postdoc working with chemistry professor Alán Aspuru-Guzik, and coworkers carry out these calculations via a large distributed computing network. By signing up, participants around the world allow their computers to be used when they would otherwise be idle. Hachmann notes that the second phase of this effort, called the Clean Energy Project, began this summer. “During the first month we already screened nearly 200,000 molecules in 2 million quantum chemistry calculations,” he says.

The process of designing new organic compounds for photovoltaics, calculating their properties, synthesizing and purifying the molecules, analyzing the films formed from them, and testing the solar cells in which they are incorporated is lengthy indeed. “It takes several weeks to put a new polymer through the battery of tests,” Konarka’s Gaudiana says. “It’s a slow process and requires a lot of work,” he admits, “but it’s the only way to understand what’s going on in detail. Without that level of understanding, there’s little chance for improvement.”


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.