Patent Application
20080061222
Quantum confinement structures may be arranged within an optical filter so that the transition or cut—on wavelength—the wavelength above which the optical filter transmits light—is adjustable. The filtered light is then focused through a lens onto a photodetector or plurality of photodetectors to produce a signal or image based on the transmitted wavelengths and excluding the wavelengths blocked by the filter.
For the purposes of this document, the term “optical” refers to visible, ultraviolet, and infrared light which obey the normal rules of optics. By this definition, long-wavelength infrared, microwaves, radio waves, extreme ultraviolet, x-ray, and gamma radiation are not optical radiation.
Through design of the material composition and layering of the optical filter, the cut-on wavelength may be fixed at a particular, desired value for a particular reference temperature (e.g., room temperature), and is then adjustable through changes in the temperature of the optical filter and/or an electric field applied across the optical filter. The filter material may consist of one or more layers of quantum confinement structures (typically a quantum well, but alternatively layers of quantum dots or quantum wires) of precise dimension and composition, surrounded by barrier layers. Optionally, the optical filter may also include a transparent substrate for mechanical stability or other purposes. The optical filter may also include a control system, for example, in the form of heaters, coolers, temperature sensors, thermostats or thermal control circuitry, electrodes, ground planes, and voltage sources or voltage controllers, for control of the temperature of the optical filter and applied electric field across the optical filter, although the control system may alternatively or additionally be external to the optical filter.
The optical response of a semiconductor is a function of its bandgap—a material-specific quantity. For photons with energies below the bandgap, the semiconductor is generally transparent, although material-specific absorption bands may also exist. Photons with energies higher than the bandgap are capable of creating electron-hole pairs within the semiconductor, and thus are generally absorbed or reflected. Thus, a material like gallium arsenide (bandgap 1.424 eV) is transparent to infrared photons with a wavelength of 871 nanometers or greater, and opaque to visible light, whereas SiO2 (bandgap ˜9.0 eV) is transparent to visible and near-ultraviolet light with a wavelength greater than 138 nm. Thus, semiconductor materials are capable of serving as optical, infrared, or ultraviolet long-pass filters.
A semiconductor will also generally show a strong emission or luminescence peak at this cut-on energy, i.e., when stimulated with an electrical current or with absorbed photons of higher energy, the semiconductor material will emit photons at the cut-on energy as a result of electron-hole recombinations within the material. Photoluminescence (i.e., stimulating the material with high-frequency light and measuring the resulting fluorescence or emission spectrum) is therefore useful as a diagnostic tool, to determine the quantum confinement energy of a quantum well and thus predict its optical properties. Strong absorption at and above the cut-on energy is also capable of generating photoelectric effects within the semiconductor as large numbers of electron-hole pairs are created.
However, the energy of an electron confined in a quantum well is not only a function of bandgap but of the quantum confinement energy, which depends on the thickness of the well and the energy height of the surrounding barriers (i.e., the difference in conduction band energy between the well and barrier materials). This “bandgap plus quantum confinement” energy moves the transparency of the material into shorter wavelengths. Thus, while a bulk GaAs sample emits and absorbs photons at approximately 870 nm, a 10 nm GaAs quantum well surrounded by Al0.4Ga0.6As barriers has a 34 meV quantum confinement energy and thus shows the same cut-on at approximately 850 nm. Therefore, for a given set of materials and a given reference temperature, the cut-on energy can be fixed precisely through the fabrication of a quantum well of known thickness. It should be noted, however, that the bandgap is a temperature-dependent quantity. As the temperature of a semiconductor decreases, its bandgap increases slightly. When the semiconductor is heated, the bandgap decreases.
At the time of manufacture, the material for the quantum confinement layer may be selected to have a bandgap near the photon energy of the desired cut-on wavelength. The barrier layers may then be selected to have a higher conduction band energy than the quantum confinement material, which in general means a larger bandgap. This ensures the quantum confinement layer is surrounded by finite (i.e., nonzero) energy barriers capable of confining charge carriers in the desired energy range, and also ensures that the energy barriers are generally transparent to photons at or near the cut-on wavelength (although material-dependent absorption peaks or bands may exist even at lower energies and/or longer wavelengths).
The dimensions of the quantum wells, wires, or dots in the quantum confinement layer of the optical sensor may then be selected such that the quantum confinement energy of the confined carriers, added to the bandgap energy of the well material, yields a cut-on frequency at exactly the desired value for the reference temperature. Depending on the materials and desired energies/wavelengths, this dimension may be anywhere from less than a nanometer to several tens of nanometers, or even several microns for devices intended to operate at cryogenic temperatures. In the more typical case, for room-temperature optical sensor devices made of common semiconductor materials and operating in the visible, near-infrared, and near-ultraviolet wavelengths, thicknesses between 2 nm and 20 nm may be the most common.
The tunable filter may then be operated by adjusting its temperature with a control system. At higher temperatures, the bandgap and the cut-off energy (i.e., the bandgap+quantum confinement energy) both decrease, resulting in a larger cut-off wavelength. In other words, the filter becomes opaque to certain frequencies where it had previously been transparent. When the temperature of the tunable filter is reduced, the opposite happens: the bandgap and cut-off energy increase, the cut-off wavelength becomes smaller, and the tunable filter becomes transparent at wavelengths where it had previously been opaque. In a similar way, an electrostatic tunable filter in an optical sensor may be operated by adjusting the electric field applied across it.
A photodetector array may produce a voltage, signal, or stored charge in response to photons within a desired range of energies and wavelengths. This range is a function of the composition and design of the photodetector device. However, in general the photodetector will respond to any photon within this range, whether it represents a desirable signal or not. The purpose of the optical filter is to restrict the transmission of certain wavelengths to the detector. For example, color CCD cameras typically include a short-pass filter, which allows visible light to pass through, but blocks the passage of near infrared light that would upset the color balance of the image. The optical filter described herein may be a solid-state tunable long-pass filter whose cut-on wavelength can be adjusted dynamically.
It should be realized that the quantum well layer 102 may be formed as a layer of quantum wires or quantum dots in order to increase the quantum confinement energy of the quantum well layer 102 without increasing the thickness, or for other reasons relating to the specific application for which the thermochromic filter device 100 is to be used. In either case, unfiltered light enters through the top of the thermochromatic filter device 100 and is modified by the thermochromic filter 110, so that filtered light exits through the lower surface of the thermochromatic filter device 100.
In addition, a control system, for example, a heating device 105, a temperature sensor 106, and a cooling device 107 may be attached to the thermochromic filter 110. In one embodiment these devices may be annular in shape and/or may be located around the periphery of the thermochromic filter 110, so as not to block the light passing through the center of the thermochromic filter 110. In addition, although mechanical heat pumps and thermometers may be used as part of a control system, the heater 105, the cooler 107, and the temperature sensor 106 may all be solid-state devices with no moving parts, other than electrons. For example, a control system with a thermocouple acting as the temperature sensor 106, a resistive heating element (e.g., a power resistor) acting as the heater 105, and a thermoelectric cooler ((TEC), e.g., a Peltier junction, a Peltier-seebeck junction, or a Thompson effect device) acting as a cooling device 107, the thermochromic filter 110 may be adjustable over a broad range of temperatures.
The heating device 105, cooling device 107, and temperature-sensing device 106 may be connected by wires 108 to a temperature-regulating device 109, which reads the temperature of the thermochromic filter 110 and adjusts the output of the heating device 105 or the cooling device 107 appropriately in order to keep the thermochromic filter 110 at a particular desired temperature, and thus a particular cut-on wavelength. The temperature-regulating device 109 may be a solid-state thermostat or thermal control circuit.
It should be understood that the optical sensor devices disclosed herein may include any necessary or convenient supporting hardware, such as solder, adhesives, or mechanical attachments to hold the hardware in place on the thermochromic filter 110; thermal gaskets, seals, connectors or heat sink compounds to improve heat flow between the heating devices 105 and the cooling devices and the thermochromic filter material; and/or a heat sink or heat pump attached to the thermoelectric cooler to maximize its ability to transport heat away from the thermochromic filter 110. It should also be understood that for some applications, adequate thermal control may be achieved with some of these components absent, non-operational, or external to the thermochromic filter device 100.
It may further be recognized that the thermochromic filter 110 may also function with the quantum confinement layer 102 deleted. In this case the cut-on wavelength of the semiconductor can still be adjusted through changes in its temperature, but the nominal value at the reference temperature cannot be selected at the time of manufacture, but rather will be determined exclusively by the bandgap of the material forming the thermochromic filter 100. Numerous other variations exist which may alter the design or appearance of the device without affecting its basic operation or underlying principles. The configuration shown in
It should be realized that it may be desirable for these multiple quantum well layers 202, 202′, 202″ to have different thicknesses, or possibly even different compositions from one another, in order to set certain parameters of the thermochromic filter 203, such as the slope of its transmission spectrum near the cut-on wavelength. Thus, the filtering properties of the tunable thermochromatic filter device 200 may be enhanced, reducing its transmissivity to unwanted wavelengths, with no effect on its nominal cut-on energy or cut-on wavelength, or its transparency to desirable wavelengths. It may also be desirable for some or all of these well layers to be replaced with layers of quantum dots or quantum wires, not necessarily identical to one another in dimension or composition. However, use of quantum dots or quantum wires or quantum wires does not fundamentally change the thermochromic filtering device 200.
It should be recognized that multiple quantum confinement layers can be arranged to form an interference filter, i.e., an optical filter that reflects a particular spectral band or group of spectral lines and transmits all others (except possibly harmonics). This has been accomplished in the past by alternating layers of material that have thicknesses that are ½ the wavelength of the light to be reflected, with materials properties such that constructive interference occurs between the light scattered by each layer. Such filters are reflective due to the interference effects that occur between incident and reflective waves at the boundaries.
By alternating layers of quantum wells and barriers as described herein, interference effects on electromagnetic radiation are demonstrable, and the semiconductor bandgaps of such a layered structure may be changed by varying the temperature of the layers, and/or by varying a voltage placed across the layers (for example, via the Stark effect). Changes in bandgap provide tunable wavelength selection around the nominal wavelength of the system. Similarly, it is possible to tune the absorptive and transmissive properties of the system using temperature and voltage effects. Additionally, varying the temperature of the quantum wells may directly affect the lattice spacing and index of refraction, affecting the reflectance, adsorption, and transmission properties of the material and thus the spectral band of the interference filter. The reflection band of such a tunable system may be Gaussian in shape, allowing the filter to serve as a notch or bandblock filter in a transmissive mode. In addition, the light reflected from the interference filter may behave as though it has passed through a bandpass filter.
If the openings 328 are smaller than or comparable to the de Broglie wavelength of the confined charge carriers, then lateral quantum confinement effects will be observed within the quantum well 302 when the surface electrode 314 is charged through the application of a voltage 326 between the electrode 314 and a transparent ground plane 308 by electrode leads 312, 318. Specifically, a respective quantum dot QD may be formed in the quantum well transport layer 302 between the barrier layers 304 and 310 beneath each opening 328 in the grid electrode 314, by the same principles discussed above. Thus, a plurality of artificial atoms may be created in the tunable filter device 300 corresponding to each opening 328 in the grid electrode 314.
Such lateral confinement of the charge carriers in the quantum well 302 will increase the quantum confinement energy of the charges carriers, increasing the bandgap-plus-quantum-confinement energy total, and thus altering the cut-on wavelength of the tunable filter device 300. In addition, it should be understood that if the grid openings 328 are deleted from the design, an applied electric field will still alter the quantum confinement energy in the quantum wells 302 via the Stark effect. Thus, regulation of the tunable filter device 300 may be achieved by a control system varying the electric field instead of the temperature. This effect can also be achieved by using a layer of quantum dot particles or devices attached to fibers or wires.
It should also be understood that the methods for forming a grid-shaped electrode are similar to those for forming an electrode of any other shape, and need not be described here. However, a partial list of techniques may include electron beam lithography and anodic oxidation lithography using the probe tip of a scanning probe microscope. It should also be noted that certain lithographic processes are particularly well suited for the nano-patterning of macroscopic areas. These include photolithography (particularly extreme ultraviolet or EUV photolithography), atom holography, and nanoimprint lithography, whether directly depositional or relying on the contamination and later developing and stripping of a “resist” layer, lend themselves to the rapid production of large and relatively uniform grids. Other methods, for example, X-ray crystallography, are capable of producing extremely fine interference patterns that may be used to expose a resist and produce grid-like patterns in a metal layer, which may be used to divide a quantum well or heterojunction into quantum dot regions.
In addition there are molecular self-assembly processes such as the anodization of aluminum into alumina, or the glassification of carefully designed diblock copolymers, which create a thin membrane or “mask” of material pierced by a regularly spaced array of vertical pores, typically arranged in a hexagonal symmetry. A milling process such as reactive ion etching (RIE) may then be used to remove the metal directly beneath a pore, while leaving the metal beneath the solid mask intact. These methods for producing the electrode or electrodes of the tunable filter device are also embodiments of the present invention, although this should not be construed as limiting the scope of the invention. A tunable filter device 300 of the type shown in
Notably, placing the quantum dots QD close together as shown in
In either case, whether constructive interference is required or merely incidental, the close packing of quantum dots QD increases the density of artificial dopants in the quantum confinement layer 302, and therefore alters the optical cut-on energy and thus the properties as an optical filter material. However, if the quantum dots QD are packed too closely, the surface electrode 314 may be easily disrupted by small cracks, impurities, or other flaws in its conductive material, and the device will not function. It should also be noted that there is a maximum and minimum value for the size of the openings 328, as well as the spacing, in order for the electric fields to assume the desired shape for quantum confinement. Thus, the exact behavior of the device under specific environmental conditions is a function of these various dimensions.
In one exemplary form of the invention, the surface electrode 314 may be a 10 nanometer thick layer of gold with a 3 nanometer adhesion layer of titanium beneath it. The barrier layer 304 may be composed of aluminum gallium arsenide approximately 5 nm thick, and the insulator 320 may be the native oxide of that material, which is normally approximately 2 nm thick. The quantum well transport layer 302 may be composed of gallium arsenide and may be 6-12 nm thick. The ground plane 308 may be composed of n-doped gallium arsenide with very low resistivity.
The surface electrode 314 may be patterned by first spin-coating it with a surface treatment consisting of a random copolymer of styrene (S), 4-vinyl benzocyclobutene (BCB), and methyl methacrylate (MMA) with proportions S/BCB/MMA equal to 56/2/42, with an average molecular weight of approximately 35,000, dissolved in the solvent toluene. The device may then be heated in a nitrogen atmosphere and then a diblock copolymer consisting of approximately 70% styrene and 30% MMA, with a molecular weight of approximately 95,000, may be applied by the same spin coating method. The device 300 may then be heated in vacuum beyond the glass cutoff temperature of the polymers, cooled to room temperature, exposed to ultraviolet light, and then rinsed in acetic acid. The resulting polymer membrane has a hexagonal array of pores whose size and spacing is proportional to the molecular weight of the diblock copolymer—in this case approximately 30 nm diameter and 52.5 nm center-to-center spacing.
The device 300 may be placed in a reactive ion etcher to remove the metal beneath the pores, and then the polymer may be stripped off. A mask may then be applied so that the metal electrode 314, and possibly the insulator 320 and upper barrier 304 may be etched away in selected regions with the reactive ion etcher. The input and output electrodes 312, 318 may then be attached to the electrode 314, the confinement layer 302, the upper barrier 304, or the insulator 320, and the ground plane 308 leaving a finished device 300. This method may be used to pattern wafer surfaces from 0.5 cm to 20 cm in diameter with approximately equal difficulty, and may also be used to pattern larger or smaller areas. This example is included for illustrative purposes only and should in no way be construed as limiting in scope.
In another exemplary form, the composition and arrangement of the metal electrode and semiconductor layers is the same, but the surface of the device may be spin-coated with the lithographic resist polymethyl methacrylate (PMMA) and then patterned with an array of holes via anodic oxidation lithography using the probe tip of a scanning probe microscope. The tip is held a few nanometers from the surface and then biased so that an electron beam passes between the tip and the surface, exposing the PMMA resist. The device may then be rinsed with a developer solution that removes the exposed PMMA, leaving behind a polymer mask with hole size and spacing depending on the bias voltage and programmed motion of the probe tip. The device may then be etched and cleaned and the control electrodes attached as in the previous example. A hole diameter of approximately 70 nm, with center-to-center spacing of 74 nm, has been found to work well. An electron microscope may be used in place of a scanning probe microscope for the lithography step, although the “proximity effect” makes it more difficult to place features close together. This description is included for explanatory purposes only and should in no way be construed as limiting in scope.
The multifunctional sensor 400 may be operated by first adjusting the cut-on wavelength of the tunable filter 401 by controlling its temperature or electric field through a control system as described above. In the most general case, the tunable filter 401 may combine the functions exemplified in
In the simplest case, the multifunctional sensor 400 may then be operated by capturing a single signal from the photodetector 403, based on the filtered light passing through the tunable filter 401 and lens 402. The resulting detection signal will be a response only to the wavelengths passed through by the tunable filter 401, not the ones blocked by it.
In a more general case, the sensor 403 may be operated by capturing successive signals from the photodetector array 403 using different settings of the tunable filter 401. Where the light or scene being imaged is constant, but the amount of light reaching the detector 403 varies as a function of the cut-on wavelength of the filter 401, the user can then extract considerable information about the spectrum of the incoming light, i.e., its varying intensity as a function of wavelength. This in turn provides information about the composition of the objects or light sources being imaged, which can be processed by computing hardware external to the sensor 400.
In the prior art this spectral analysis is generally accomplished with a filter wheel that rotates a succession of different static filters in front of an imaging sensor. Occasionally, for narrow-field-of-view applications, the same effect is accomplished with a Fabry-Perot interferometer or etalon. In both cases, moving parts are required, with all the attendant reliability problems. In contrast the present invention permits such hyperspectral analysis with a completely solid-state device.
Where the photodetector 403 is an imaging sensor such as a CCD camera, successive images taken with different filter settings may provide detailed spectral information for every pixel of the image, enabling detailed and sophisticated analysis of a scene. This is useful, for example, in robotic space probes attempting to analyze the composition of various landscape features. This principle can also be applied to create, for example, a single sensor 400 that, by changing its wavelength response at different times, combines the functions of a spacecraft's star sensor, sun sensor, Earth horizon sensor, and general-purpose imaging sensor, thus reducing the number of sensors required by a single spacecraft. Once again, this may require computational or analytic hardware external to the sensor 400.
In the interests of concision, the term “optical” has been used throughout this document, even though it excludes other electromagnetic spectra including long-wavelength infrared, microwaves, radio waves, extreme ultraviolet, x-ray, and gamma radiation. However, it will be apparent to a person skilled in the art that if the filter materials, quantum confinement dimensions, lens materials and photodetector hardware are selected such that the resulting sensor has both a cut-on wavelength and a detection capability in one of these electromagnetic bands, the multifunctional sensor will still function as described above, even though the radiation is not “optical” per se. It should also be noted that some materials (e.g., mercury, vanadium dioxide) behave as conductors at higher temperatures and as semiconductors at lower temperatures. Since conductors generally reflect light rather than absorb it, a quantum well layer composed of such a material would, above the conductivity threshold temperature of the material, suddenly become reflective. This is a thermochromic optical property and, where incorporated into the filter of a multifunctional sensor, is explicitly considered an alternate embodiment.
From the description above, a multifunctional sensor may be understood to provide a number of new capabilities. Specifically, the multifunctional sensor incorporates a tunable electromagnetic filter that can regulate the flow of light based on the temperature of the filter material and/or an applied electric field, within a range specified by the composition and internal structure of the tunable filter, in a completely solid-state package. Unlike tunable filters which rely on piezoelectric actuators for mechanical rotation or deformation, the tunable filter described herein contains no moving parts other than photons and electrons and is therefore more compact and more robust than prior known tunable filters. Thus, the multifunctional sensor may be smaller, more capable, and more reliable than any prior known, comparable device.
The multifunctional electromagnetic sensor may also be fully programmable, i.e., its light-detection and analysis capabilities properties may be controlled externally, through the application of electrical energy to the heaters, coolers, and electrodes, or through command signals to temperature controller or voltage controller circuits.
The electromagnetic sensor may include a simple photodetector, but may also be a complex imaging sensor (for example, a CCD digital camera). The addition of a tunable filter to an imaging sensor makes it a multispectral or hyperspectral sensor, capable of measuring the emission or reflection spectrum of the objects or materials it is imaging, and thus gaining significant information about their composition, temperature, or (in the case of manmade devices) their design and operation. Also, unlike an etalon filter, the solid-state, tunable filter does not restrict the field of view of the multifunctional sensor, but operates over the entire field of view of the instrument, as defined by its apertures and lenses.
The electromagnetic sensor is also capable of achieving specific cut-on energies that do not correspond with the bandgap of any known material, and thus may be difficult to achieve through any other means. In addition, as a side effect of its composition and structure, the multifunctional sensor which may also be capable of generating light (for example, when fluorescing in response to short-wavelength light), or generating electricity from incident light (e.g., via the photoelectric effect), in addition to its normal function as a tunable optical filter.
The electromagnetic sensor also provides a multifunctional, tunable optical sensor that may be constructed entirely out of solid state components. This allows it to have a smaller mass and volume than a comparable system incorporating electromechanical or piezoelectrically actuated filters, and offers greatly improved resistance to shock, vibration, wear, fatigue, particulate interference, gumming of lubricants, and other forms of spontaneous mechanical failure.
Although the description above contains many specificities, these should not be construed as limiting in scope, but rather construed as merely providing illustrations of certain exemplary embodiments. There are various possibilities for making a tunable filter of different materials (including insulators, semiconductors, conductors, or superconductors) and in different configurations. There are particular advantages to using higher bandgap materials, as they not only allow for energetically “deeper” quantum confinement devices, but in many cases also allow the well and barriers to be transparent to light of higher energies and shorter wavelengths.
Numerous other variations exist which do not affect the core principles of the invention's operation. For example, the shape of the tunable filter need not be planar as shown in the figures, but could be in the form of flexible sheets, ribbons, or fibers with quantum dot devices on one or both surfaces, or formed into or around solid shapes including, but not limited to, cylinders, spheres, cones, prisms, and polyhedrons, both regular and irregular, asymmetric forms, and other two- and three-dimensional structures. The quantum well layers could be replaced with layers of quantum dots or quantum wires. It is also conceivable to grow the tunable filter on the inside surface of a complex, porous, or “spongy” material/structure such as an acrogel. The filter could include multiple input pathways (to serve as, for example, a mixer or signal combiner) or multiple output pathways (to serve as, for example, a signal splitter or diverter), or both.
Although various embodiments of this invention have been described above with a certain degree of particularity, or with reference to one or more individual embodiments, those skilled in the art could make numerous alterations to the disclosed embodiments without departing from the spirit or scope of this invention. All directional references e.g., proximal, distal, upper, lower, upward, downward, left, right, lateral, front, back, top, bottom, above, below, vertical, horizontal, clockwise, and counterclockwise are only used for identification purposes to aid the reader's understanding of the present invention, and do not create limitations, particularly as to the position, orientation, or use of the invention. Connection references, e.g., attached, coupled, connected, and joined are to be construed broadly and may include intermediate members between a collection of elements and relative movement between elements unless otherwise indicated. As such, connection references do not necessarily imply that two elements are directly connected and in fixed relation to each other. It is intended that all matter contained in the above description or shown in the accompanying drawings shall be interpreted as illustrative only and not limiting. Changes in detail or structure may be made without departing from the basic elements of the invention as defined in the following claims.
This application claims priority pursuant to 35 U.S.C. 119(e) to U.S. provisional application Nos. 60/825,385 and 60/825,405, both filed 12 Sep. 2006, each of which is hereby incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 60825385 | Sep 2006 | US | |
| 60825405 | Sep 2006 | US |
