Making light of the small Photonics and nanotechnology have a synergy rarely seen. But nano-optical devices are not prevalent in the marketplace, even though the academic labs churn out one breakthrough after another. So where’s the light? By Richard Gaughan Photonics and nanotechnology seem the perfect match. In fact, photons are inherently nanoscale entities, generally interacting with a single electron of a single atom. But even though there seems a natural marriage between photons and nanomaterials, nano-optical devices are not yet prevalent. But the delay is not due to any fundamental scientific misunderstanding or technological failure, but is rather a reflection of the nature of technological development. In fact, a look at some of the nano-optical devices that are at or near commercial reality shows a predictable pattern of needs: tight integration, an existing market, a clear cost benefit and a scaleable solution. Some examples of innovative photonic technology on the cusp of market acceptance are quantum dots, membrane deformable mirrors, and photonic bandgap fibers. Each of these technologies has distinct engineering challenges, and each has a distinct market, but together they provide insight into the types of challenges faced by most nano-optical technologies. Quantum dots shine The unique absorption and emission characteristics of quantum dots (QDs) were first demonstrated in the 1980s. A quantum dot is a semiconductor particle just a handful of nanometers in diameter. The QD creates a potential well that constrains the electrons within the semiconductor to specific energy levels dependent upon the material and the particle size. The specific energy band structure determines the wavelength of photons that can be absorbed or emitted by the QD. In general, the absorption band is relatively broad, but the emission wavelength of a specific QD is narrow. In practice, this means different diameter quantum dots can be excited by the same illumination source, but each will emit at its unique wavelength. By layering the proper blend of quantum dot diameters on top, these UV LEDs emit visible light with high conversion efficiency. Here Lauren Rowher of Sandia National Laboratory showcases a couple of different laboratory devices designed to pave the way to commercial development. Photo courtesy of Sandia National LabClick here to enlarge image For example, QD labels treated with binding molecules will attach to specific molecular targets, so when a solution is illuminated with a single source the different labels emit light of a different color. The fluorescing quantum dots serve to make the otherwise invisible target molecules visible. According to Steve Talbot, chief marketing officer at Evident Technologies, a Troy, N.Y., company that makes a variety of products based on quantum dots, QDs have rapidly infiltrated life science applications at least partly “because they are easily integrated with the existing technologies” – such as the surface binding methods and fluorescence readers prevalent in the marketplace. The next target application for Evident Technologies is solid state lighting. Light emitting diodes (LEDs) are expected to be efficient replacements for current lighting technologies, for applications from decorative accent lighting to aircraft and automobile lighting – and eventually the general illumination marketplace. Different colors can be realized by designing devices of unique materials and customized semiconductor structures. But a more efficient solution may be to use bright UV LEDs to excite a phosphor layer which will absorb the UV and emit in its characteristic color. Sounds like a perfect match for quantum dots, which absorb in a broad range in the UV and emit at a precise wavelength. QDs of different diameters can be integrated into a single phosphor layer, with the emitted light being a summation of all the different colors – including blends that can create white emission. To produce LEDs with this range of color would usually require different semiconductor materials and different structures. But by coating a UV LED with quantum dots of different sizes, identical components emit different visible spectra. Photo courtesy of Evident TechnologiesClick here to enlarge image Conceptually that’s easy to understand, but to implement the LED QD nanophosphor requires success over a number of steps. Mike Locascio, Evident’s chief technical officer, identified a host of issues for LEDs. “For an LED to be successful,” he said, “it needs not only an exact color match and good color uniformity, but also a high color rendering index [a measure of white light quality], longevity, and high brightness, all at a competitive price point.” Although the fundamentals of QD manufacturing are understood, to make them application-specific requires more than just a grasp of how large to make a QD core. The surface layer of the QD modifies the color, then an encapsulant provides both an interface between the LED and QDs and a matrix for deposition of the nanophosphor. And the entire assembly must survive a high temperature cure that will not degrade its environmental or performance capabilities. The application initiates a cascade of development steps. Complex, yes, but the challenges can be overcome: Evident is now shipping sample LEDs with integrated QD nanophosphors. Adaptive optics for the masses Adaptive optics refers to the capability to measure and control the shape of a propagating wavefront. Sensors provide input into a control system that generates signals to change the optical path length of a small part of the cross-section of an optical beam. A deformable mirror introduces wavefront changes by tilting and positioning small areas of the mirror surface. One problem with traditional deformable mirrors is that they’re expensive. In the mid-1990s a MEMS deformable mirror was first demonstrated, constructed by assembling an electrode pattern surface parallel to a very thin reflective and conductive membrane. With a voltage pattern introduced on the electrodes, electrostatic attraction pulls the membrane into a desired shape, changing the wavefront of a beam reflected off its surface. Because of the advantages of scale offered by MEMS manufacturing, membrane deformable mirrors are much less expensive than traditionally-manufactured deformable mirrors, which brings the cost into range for mainstream projects. Although the cost of continuous-membrane MEMS deformable mirrors is attractive, they had an operational restriction that was a bit cumbersome. Under certain conditions the membrane gets so close to the electrode that the electric field strength rapidly rises, forcing the membrane to come in contact with the electrode, leading to electrical discharge through the membrane, and catastrophic membrane failure. AgilOptics, of Albuquerque, N.M., avoided this “snapdown” problem by restricting the usable voltage range, but that also limited the utility of the deformable mirror. The solution was acceptable, but not ideal; so development continued, and AgilOptics’ commercially-available membrane mirrors now have an insulating coating that retains its flexibility, but eliminates snapdown entirely. Guiding the unguidable In 1998 Yoel Fink and others at MIT reported on a class of reflective coatings that offered characteristics no other reflective coatings could match: angle-independent reflectivity over a wide range of wavelengths. By depositing alternating layers, a photonic crystal structure was created, with a bandgap that prohibited propagation for a range of wavelengths determined by the index of refraction of the two materials and their layer thickness. Fink realized this principle could be applied to reflective surfaces along a waveguide of arbitrary shape to control the propagation of wavelengths that traditionally are difficult to guide. For example, a hollow core surrounded by alternating layers of materials of high index of refraction would be able to guide the 10.6 μm wavelength of CO2 lasers. Fink and his colleagues created a company called OmniGuide, in Cambridge, Mass., to commercialize applications of the photonic bandgap (PBG) fiber. MEMS deformable mirrors are affordable enough to bring wavefront control to a variety of new applications. For example, this membrane mirror system stores up to 100 frames that can be played back continuously to simulate changes in atmospheric conditions. Photo courtesy of AgilOpticsClick here to enlarge image But fabricating a laboratory scale device for academic research is quite different from manufacturing commercially significant quantities, and Fink was presented with the challenge of scaling the manufacturing. He needed a method that would control the thickness of each of the layers surrounding the core, yet still be able to produce large quantities of the PBG fiber. He was drawn to the drawdown process traditionally used to produce optical fiber: a macroscopic preform is fabricated, then heated and pulled into a long, thin strand. “Conceptually,” said Fink, “the difficult and tedious process of reducing feature size becomes straightforward with fiber drawdown, and the length can be kilometers.” But several challenges stood in the way of translating that concept into reality. First, the feature sizes of the PBG fiber are one or two orders of magnitude smaller than those in traditional fiber – layers 100 nm thick instead of tens of microns. Second, rather than using the homogeneous glasses of traditional fiber, high-index semiconductors were needed. Finally, each of the multiple layers of the waveguide must be precisely controlled at a level well beyond that required for traditional fiber manufacturing. Those three challenges changed the project into a two-year, market-driven research effort. The ideal market Identifying the ideal market is like a “cutest baby” competition: it all depends on your perspective. Different technologies for different applications also have different criteria for what constitutes the ideal market opportunity. For the application of quantum dots for LED wavelength conversion, the ideal market has the potential for extremely high volume. Other technical solutions exist for generating a desired spectrum from solid state devices, but none is firmly entrenched. QD manufacturing technology is efficient enough to allow market entry at a competitive price point, and improvements in process control promise future cost reductions. And, although the market for solid-state lighting is fair-sized already, the general illumination market holds huge potential. Evident Technologies’ Locascio noted that “market areas and subsegments within each area have their own set of challenges. We look at the price, performance, and packaging requirements to determine if quantum dots can provide an effective solution.” For MEMS deformable mirror applications, the ideal application is either one in which a conventional optical instrument provides acceptable, but not optimum performance, or an application where wavefront control is being performed in much more expensive ways. Dennis Mansell, president of AgilOptics, described the new Aeri atmospheric simulator the company has developed that can loop 100 frames to emulate rapid changes in optical transmission. “The system is highly capable, and several large customers are interested. But it’s a bit frustrating waiting for them to see the value.” The ideal customer for Omniguide’s PBG fiber is one that has an important problem the technology can solve, and they’re willing to pay a premium for the solution. Whether the market is a relatively small number of customers willing to pay top dollar, a huge opportunity with smaller margins, or somewhere in between, each of these companies emphasizes the need to understand the customer’s requirements. The fundamental technology is already understood and the issue becomes one of tailoring the characteristics and the manufacturing process to meet the customer’s performance and price requirements.