Straight Dope


    For decades, silicon has been the material of choice for solar cell manufacturers, just as it has been for consumer electronics makers. Silicon crystals are available in high purity and lend themselves easily to traditional solar cell fabrication techniques. But many other semiconductors exist that are not so amenable to conventional methods. Alternative fabrication techniques for these materials might give manufacturers more flexibility and choices and could lead to more efficient cells. Recently, researchers in the physics department at UC Berkeley and at Lawrence Berkeley National Laboratory developed a new way to fabricate a solar cell from just about any semiconductor. Their method greatly expands the choices for efficient and easier-to-fabricate photovoltaic devices.

    Under normal circumstances, a semiconductor crystal behaves like an insulator and does not conduct electricity. But if sunlight is absorbed in the crystal, it becomes conductive, unleashing energized charge carriers that are free to roam. This charge is most easily collected if the semiconductor is chemically “doped,” meaning that impurities are intentionally added that donate additional charge carriers. A clever doping scheme positions positive and negative charge carriers in such a way that a useful current flows out of the solar cell in response to absorbed light.

    Silicon is the most commonly used semiconductor for photovoltaic devices. However, silicon is by no means the only show in town. Several other types of semiconductors, including metal oxides, show great promise for use in solar cells because they can efficiently absorb sunlight and can be easily obtained. So far, these contenders have mostly sat on the sidelines because they do not lend themselves easily to traditional chemical doping. If a convenient way is found to dope more types of semiconductors, then a wide variety of efficient solar cell materials could enter the market, with a dramatic effect on solar cell costs.

    This is where the technological advances at Berkeley come into play. Researchers in the groups of Professors Alex Zettl and Feng Wang came up with new techniques for forming charge-collecting electrodes on semiconductors to make photocells. The configuration relies on a voltage applied near the surface of the semiconductor, establishing an electric field in the device that causes mobile charge carriers to shift around, which results in local doping of the material. This eliminates the need for traditional doping. For this to work, the electrodes that conduct useful current from the cell must not be so thick or bulky that they interfere much with the applied electric field that emanates from a region directly above them. The contacts must only slightly screen, but not fully block, the applied electric field.

    To form the charge-collecting electrodes, the Berkeley team turned to two concepts on the forefront of modern materials science: nanowires and graphene. Nanowires have widths on a nanometer scale and can form contacts narrow enough to partially screen, but not fully block, the electric field applied to the solar cell. Graphene, on the other hand, is a one-atom-thick sheet—thin enough that an electric field can effectively pass through it. Either of these methods is sufficient to produce the required partial screening. The development of these nanowires and graphene layers enabled this technological advance, and the resulting devices are called screening-engineered field-effect solar cells.

    A solar photovoltaic power plant at Nellis Air Force Base, Nevada. Advances in photovoltaic technology may significantly lower the cost of building solar power plants. Credit: Nadine Y. Barclay

    A solar photovoltaic power plant at Nellis Air Force Base, Nevada. Advances in photovoltaic technology may significantly lower the cost of building solar power plants.
    Credit: Nadine Y. Barclay

    Will Regan and Steven Byrnes, recent PhD graduates in the Zettl and Wang groups, were the first authors on the paper describing the new method, published in Nano Letters. According to Regan, the development of nanowire and thin graphene contacts to achieve sufficiently reduced screening were the tough technological problems that needed to be solved for this idea to work. Now that these problems have been overcome, “virtually any semiconductor can be made into a solar cell using this technique,” Regan explains. “The single deposition machine required could take the place of several separate pieces of equipment currently needed in a solar cell fabrication line.”

    With these savings, efficient solar cells can be more profitable to industry and cheaper for customers. Berkeley physicists are thus applying cutting-edge technology to solving one of the toughest problems of modern society: how to provide energy for a growing number of people in a sustainable way.