HOLOGRAPHIC MICROFABRICATION AND CHARACTERIZATION SYSTEM FOR SOFT MATTERAND

BIOLOGICAL SYSTEMS

CROSS-REFERENCE TO RELATED PATENT APPLICATIONS

[0001] This application claims the benefit under 35 USE 119(3) of U.S. Application No. 60/857,348, filed November 7, 2006, incorporated herein by reference in its entirety.

[0002] The United States Government has certain rights in this invention pursuant to grants from the National Science Foundation by Grant Numbers DMR-0451589 and DBI-0629584.

[0003] This invention is directed to a fully integrated holographic microfabrication system and method. More particularly, the invention is directed to an optical system for assembling and dynamically controlling complex three dimensional structures from objects suspended in a fluid, the objects being, for example, colloidal particles, macromolecules, nanoclusters, nanowires and biological materials, such as cells. Such objects can be of any size and shape which are readily assembled and manipulated for a selected commercial purpose.

BACKGROUND OF THE INVENTION

[0004] The use of optical traps or tweezers has undergone substantial development over recent years. This technique can manipulate matter, including very small objects and small portions of larger objects, with great precision. Recent progress has resulted in the ability to create large arrangements of optical traps to perform

simultaneously many tasks at various spatial locations. These traps can also be individually specified as to trapping strength, optical character and size, given the needs of the situation. In view of all these degrees of freedom, however, little has been accomplished in terms of complex commercial applications.

SUMMARY OF THE INVENTION

[0005] One Three dimensional assembly, micromanipulation and dynamic configuring of objects is accomplished by use of computer generated holograms which can trap objects, exert precision force at selected system locations and assemble complex arrangements of objects in any selected three dimensional configuration, including extensive stacks of objects. Collectively the assembly of optical traps can execute processing and manufacturing protocols for a wide variety of commercial purposes. This system can carry out such manufacturing steps as assemblying three dimensional functional structures from various building blocks, such as microscopic fluid-borne objects (colloidal particles, e.g.), macromolecules, nanoclusters, nanowires and various biological media, such as biological cells. The system can carry out assembly, processing, testing and inspection of the assembled particle array or an object, execute chemical processing steps, as well as perform mechanical and optical processing using a selectable range of light wavelengths, including white light to perform these functions. Further, the system can be used as a sensor or probe for optical, electrical, chemical, biological and force gradient properties. In addition the system employs a holographically focused microscope with each image itself being a hologram, incorporating volumetric data and in effect is three dimensional versus conventional holographic microscopy where images are two dimensional.

[0006] Various aspects of the invention are described hereinafter, and these and other improvements are described in detail hereinafter, including the drawings described in the following section.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIGURE 1 illustrates a schematic diagram of a holographic optical trapping system;

[0008] FIGURE 2A illustrates a computer generated hologram (CGH); FIG. 2B illustrates a light array at an intermediate focal plane; and FIG. 2C illustrates a resulting optical trap array;

[0009] FIGURE 3 A illustrates an assembly of 173 colloidal silica spheres in a quasicrystal layer array; FIG. 3B illustrates the trapped particles translated into a desired three dimensional configuration; FIG. 3C shows reduction of the scale to create an optically dense material which can then be gelled; and FIG. 3D shows a laser diffraction pattern at 633 nm wavelength which shows 10 fold diffraction peaks; [0010] FIGURE 4A illustrates a helical phase mask transformation of a TEMoo mode to a helical mode with winding number, I; FIG. 4B illustrates the helical mode

focused to a ring of light of radius R _e ^oc I with I = 30; and FIG. 4C shows a multiply

exposed photograph of a single colloidal silica sphere dispersed in water and circulating around the optical vortex of FIG. 4B;

[0011] FIGURE 5 A illustrates a phase hologram for encoding a micro fluidic gear pump using counter-rotating optical vortices with topological charge £ _± = ±30; FIG. 5B illustrates the projected light pattern and processing with the hologram before

adaptive optimization; FIG. 5C shows after adaptive optimization; and FIG. 5D shows an operating microfluidic pump filled with 700 nm diameter silica spheres and pumping water through a central channel at 5 //m/s;

[0012] FIGURE 6 shows a general system for optical processing;

[0013] FIGURE 7 shows a holographic optical trapping module;

[0014] FIGURE 8 shows a holographic 3D imaging module;

[0015] FIGURE 9 shows a white light based optical processing system; and

[0016] FIGURE 10 shows a multipoint force monitoring/processing module.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0017] A system constructed in accordance with an embodiment of the invention is shown generally at 100 in FIG. 1. The system 100 uses a computer-designed diffractive optical element (DOE) 110 to split a single collimated laser beam 120 into multiple independent beams, each of which subsequently is focused into an optical trap or tweezers 130 (see FIG. 2C) preferably by a strongly converging objective lens 140 in FIG. 1. The DOE 110 preferably takes the form of a spatial light modulator (SLM) to create a computer-generated hologram (CGH) as shown in FIG. 2A. This CGH creates a plurality of light beams 125 in FIG. 2B specifically designed to create a particular pattern of the optical tweezers 130 (see FIG. 2C). Projecting a sequence of computer-designed holograms with an SLM reconfigures the projected traps 130, thereby translating trapped particles or manipulating objects along selected independent paths. The optical tweezer 130 is created in a conventional manner by bringing an intense beam of light, such as the laser beam 120, to a diffraction- limited

focus. The light's electric field polarizes nearby dielectric objects; and the induced dipoles are drawn up intensity gradients toward the focus. Radiation pressure due to absorption and surface scattering competes with this optical gradient force and tends to repel an optically trapped particle. Stable trapping is only possible if the gradient force dominates. For this reason, optical tweezer systems often are built around the microscope objective lens 140 (see FIG. 1) whose large numerical aperture and well- corrected aberrations optimize axial intensity gradients.

[0018] The optical trap 130 can be placed anywhere within the objective lens' focal volume by appropriately selecting the input beam's direction of propagation and degree of collimation. For example, a collimated beam 135 passing straight into an infinity-corrected form of the objective lens 140 comes to a focus in the center of the lens' focal plane, while another beam (not shown) entering at an angle comes to a focus proportionately off-center. A diverging beam focuses downstream of the focal plane and a converging beam focuses upstream. By the same token, multiple beams entering the lens' input pupil simultaneously form multiple ones of the optical trap 130 in the focal volume, each at a location determined by its degree of collimation angle of incidence.

[0019] Using the system 100 of FIG. 1, FIG. 3 A shows 173 colloidal silica spheres arranged in a single plane within a three-dimensional sample volume. Comparable planar rearrangements also can be implemented with a conventional single rapidly scanned optical tweezer in a time-shared configuration. Unlike these other conventional techniques, however, holographic trapping also can create three- dimensional structures. The images in FIG. 3B and 3C show the same spheres being reorganized into the third dimension, their images changing appearance as they move

along an optical axis 145. Other three-dimensional optical trapping techniques, such as the generalized phase contrast (GPC) method require two-sided optical access and cannot stack objects along the optical axis 145. By contrast, the system 100 can stack micrometer-scale objects at least 12-14 deep along the optical axis 145. [0020] In addition to arbitrary three-dimensional control, the holographic traps 130 offer other advantages over conventional scanned tweezers. HOT patterns have extensive degrees of freedom than time-shared arrays which must periodically release and retrieve each trapped object. Additionally, the lower peak intensities required for continuously illuminated traps are less damaging to photosensitive samples. [0021] A laboratory-scale implementation of a dynamic form of the holographic optical tweezers 130 preferably used a Hamamatsu X8267 parallel-aligned nematic SLM for the DOE 110 to reshape the laser beam 120 from a frequency-doubled diode- pumped Nd: YVO ₄ laser (Coherent Verdi) into a designated pattern of beams. Each is transferred to the entrance pupil of a 10Ox NA 1.4 oil immersion objective lens 140 mounted in a Nikon TE2000U inverted optical microscope 150 and then focused into the optical 130 trap. A dichroic mirror 160 reflects the laser beam 120 into the objective lens 140 while allowing images of the trapped particles to pass through to a video camera (not shown). When combined with a 0.63x widefield video eyepiece, this optical train offers a 86 x 65 μm ² field of view, with a magnification of 135 nm per pixel at the video camera.

[0022] The collimated laser beam 120 initially has planar wavefronts with a uniform phase profile φ(p) = 0 . The DOE 110 imprints them with the phase profile φ(p) at

each 40 μm wide pixel in a 768 x 768 array. The DOE's calibrated phase transfer

function imparts one hundred fifty distinct phase shifts ranging from O to 2π at the operating wavelength of λ = 532 nm. The phase shift imposed at each pixel is specified through a computer interface with an effective refresh rate of 2 Hz for the entire array. Despite the DOE's inherently limited spatial bandwidth, it can project such sophisticated trapping patterns as illustrated by considering FIGS. 2A-2C and

3A-3D.

[0023] Three-dimensional control is attained by integrating the objective lens' phase profile, into the phase hologram, φ(p), encoding an array of the optical traps 130. This translates the optical trap 130 by z along the optical axis 145. Here/is the focal length of the objective lens 140. The algorithms we have developed for computing trap-forming phase holograms can separately specify φ(p) for each trap-forming beam, thereby allowing each of the optical traps 130 to move independently with respect to the focal plane.

[0024] Engineering the individual traps' wavefronts imparts additional functionality. For example, the helical phase profile φ, (p) = /θ mod 2π (2) converts an ordinary Gaussian laser beam into a helical mode, and its corresponding optical tweezer into a so-called optical vortex. Here θ is the polar coordinate in the DOE plane, and the integer I describes the helical wavefront pitch and is known as the topological charge. Destructive interference along the helical screw dislocation cancels the beam's intensity along its axis all the way to the focus. As shown in

FIG. 4B, optical vortices 170 thus focus to bright rings of light, rather than points. The image in FIG. 4B was obtained by placing a mirror (not shown) in the microscope's focal plane and collecting the reflected light with the conventional imaging train. These dark form of the optical tweezers 130 have proved useful for trapping objects that are not compatible with conventional optical tweezers 130, including ultra-high-index particles for photonic applications (such as assembling the icosahedral quasicrystal of FIGs. 3A-3C). These optical tweezers 130 also can trap reflecting, absorbing and low-index particles that are useful for biomedical applications. [0025] Combining (p _e (p) with a phase hologram encoding an array of the optical

tweezers 130 yields an array of optical vortices 170 (see FIGS. 5A-5D). Our algorithms for computing phase-only holograms also can imprint vortex-forming helical phase fronts onto individual traps ones of the optical 130 in an array, creating mixed patterns with different winding numbers, such as the array shown in FIG. 5B- 5D.

[0026] Because of their helical phase profile, the optical vortices 170 carry orbital angular momentum, independent of polarization, that they can transfer to trapped materials. The resulting torque can drive trapped objects around the trap's circumference, as shown in FIG. 5D. The optical vortices 170 are useful, therefore, for creating motion at small length scales, for example in rotating semiconductor nanowires during assembly. Arrays of the optical vortices 170 can act as pumps and mixers, which also are useful for preparing and processing materials for

characterization and assembly. A typical pump is shown operating in FIG. 5D. The optical vortices 170 can further act as conveyor belts for a manufacturing operation. [0027] Holographic wavefront engineering lends itself to other advantageous applications, with virtually any mode of light having potential applications. For example, the axicon phase profile φ _γ (p) = γp creates an approximation of a Bessel

mode which focuses to an axial line trap whose length is controlled by γ. Arrays of Bessel beam line optical traps can pass through quite deep microfluidic channels and thus improve optical fractionation of fluid-borne samples. [0028] All of these trapping capabilities are controlled by the phase profile imprinted on the input laser beam 120 by the DOE 110 (preferably the spatial light modulator). Thus, they can be selected and changed in real time under computer control, without requiring any hardware modifications. A single computer-designed phase hologram can project distinct patterns of the optical traps 130 in multiple wavelengths of laser light. Multi-color trapping and photochemical processing is very useful for assembling multi-component tissues, materials and devices. [0029] FIG. 6 illustrates a system 200 which includes modular subsystems that comprise the proposed integrated fabrication and characterization system. Included in the system 200 can be holographic trapping subsystem 210 and holographic microscopy subsystem 220. The primary assembly system 200 and the subsystems 210, 220 can be supported by various spectroscopy components in both transmission and reflection, and an advantageous form of force spectroscopy. [0030] Rapid and precise fabrication of photosensitive materials requires appropriate choice of wavelength. Earlier holographic trapping systems have been

based on high-powered visible lasers, both for their ease of use, and also for their comparatively high trapping efficiency for micrometer-scale objects. Longer wavelength lasers are less likely to damage biological materials, however, and can be more effective for trapping semiconductor nano wires and related nanomaterials. [0031] In the subsystem 210 of FIG. 6, recent advances in commercial fiber lasers make these a particularly attractive class of infrared sources for trapping applications. The subsystem 210 preferably uses an IPG YLD-IO-LP linearly polarized single- mode continuous wave fiber laser 240 operating at 1075 nm. This laser 240 provides up to 10 W, which can power up to 2000 independent traps. The laser beam 250 can be expanded to 5 mm diameter using a standard fiber-coupled beam expander (IPG) to illuminate the DOE 110 (preferably a spatial light modulator). [0032] A preferred embodiment of the holographic optical trapping subsystem 210 projects the optical traps 130 using a Holoeye HEO- 1080p liquid crystal on silicon (LCoS) SLM tuned to provide 2π phase shift at the proposed laser wavelength. This SLM (a preferred form of the DOE 110) comprises an array of 1920 x 1080 phase pixels, each 8 μm across. In selecting a suitable SLM for a given application, several optimization criteria may be considered. For example, larger numbers of phase pixels, higher rates of phase pattern changes and finer resolution of phase modulation al improves the performance of the holographic optical trapping subsystem. Smaller phase pixels allow for more compact design of the holographic optical trapping subsysterm 210.

[0033] In a most preferred embodiment, trapping and imaging will be based on a choice of Nikon CFI Plan Apo 10Ox and 6Ox oil immersion IR objective lenses 140.

These lenses 140 have proved excellent for optical trapping both because of their high numerical aperture (NA 1.4), and also because of their excellent aberration correction. The IR variants of these lenses 140 are designed for operation at infrared and visible wavelengths.

[0034] In a preferred embodiment shown in FIG. 7, laser power will be fine tuned with a half-wave plate 300 in a precision rotation mount and a polarizing beam splitter 310. This also will improve diffraction efficiency by improving the beam's polarization. The beam splitter 310 also provides a port for a second laser 320 at a different wavelength, useful for multicolor holographic trapping. [0035] The role of the relay optics is to project an appropriately scaled image of the DOE 110 face onto the input pupil of the objective lens 140. The field and depth of view for the trapping system then is set by the number of pixels spanning the projected aperture. With the 6Ox objective lens 140, the proposed system 200 will be able to place traps at arbitrary locations over a 120 μm x 120 μm area and at ±50 μm with respect to the focal plane. Three-dimensional placement resolution is roughly 30 nm, and deliberately exploiting high-order diffraction will allow us to extend this range by nearly a factor of two, with limitations set by reduced diffraction efficiency. [0036] As shown in FIG. 8, the optical train 400 that projects the holographic traps 130 is readily modified to work as an imaging system 410. As shown schematically in FIG. 8, light gathered by the objective lens 140 is projected onto a video camera 420 by a relay lens 430 whose transfer function is controlled by a second phase-only spatial light modulator 440. The two physical relay lenses 430, 450 are arranged as a standard 4/ telescope, transferring the wavefronts of light emerging from the

objective lens' pupil to the face of the SLM 440. A virtual lens φ _z (p), encoded on the SLM 440 then selects the desired focal plane.

[0037] In the simplest implementation, the imaging train is focused to a plane at height z above the objective's focal plane by imprinting the same phase pattern, (see Eq. (1) used to displace the optical trap 130). In this case, the imaging plane can be effectively scanned through a three-dimensional sample without physically moving the sample relative to the objective lens 140. The focusing hologram also can be used to adaptively correct for geometric aberrations in the imaging train. [0038] Employing the same class of the SLM 440 for imaging and trapping ensures that the field of view and depth of focus for imaging will cover the same range as the trapping system. This SLM's updates will be synchronized to the video camera 420 to gather volumetric data at video rates.

[0039] Images are most preferably acquired with a Roper Cascade 512B electron- multiplied charge-coupled device (EMCCD) camera. This camera 420 incorporates field-effect amplifiers at each pixel so that it can offer both low-light level imaging and also extremely low-noise bright-field imaging. It also offers flexible triggering and electronic shuttering, as well as adjustable resolution and frame rate. Low-noise operation is particularly useful for particle tracking and deconvolution microscopy, whose accuracy degrades rapidly with decreasing signal-to-noise ratio. [0040] Unwanted diffraction orders due to the SLM's pixellated structure can be eliminated with an appropriate set of pupils mounted with the ocular lens (not shown). Additionally, a small beam block 425 (shown in phantom in FIG. 8) introduced into the relay lens' intermediate focal plane can be used to convert the optical train to

dark-field operation. Other amplitude- and phase-modifying elements introduced in this plane will provide additional modes of operation, including variants of phase- contrast microscopy.

[0041] Commercially available SLMs 440 impose different degrees of phase modulation on light of different wavelengths. This means that the holographically focused system 410 would suffer from chromatic aberration were it used with white light illumination. This is a principal reason for replacing standard microscope illumination with a monochromatic source for bright-field and dark-field imaging. The system 440 preferably is based on a 200 mW fiber-coupled laser diode 460, operating at 860 nm by SDL, Inc. Bending fiber 470 to scramble the wavefront yields uniform speckle-free illumination. The laser diode 460 also can be rapidly gated for short exposures and multiple stroboscopic exposures.

[0042] The fiber-coupled laser 460 can be collimated with a commercial fiber collimator 480 and focused onto sample 465 with a second 6Ox objective lens mounted as a condenser. The beam will be reflected into the condenser 490 preferably using a dielectric multilayer mirror (Chroma Technology) mounted at 45°, thereby providing additional optical access to the condenser 490 at other wavelengths. In particular, this arrangement will provide access for a standard white light illuminator, which can be convenient for some applications. In this case, the SLM 440 will likely not play an active role, and the imaging system 410 will produce standard two-dimensional images.

[0043] Volumetric image reconstruction can best be performed using standard algorithms of deconvolution microscopy implemented in the IDL programming language. Simple deblurring operations with the nearest-neighbor algorithm can

proceed in near-real time on a standard workstation. This will be ideal for on-line inspection of systems being assembled through holographic trapping. More accurate reconstructions will proceed either with myopic deconvolution or with constrained maximum entropy algorithms, depending on the nature of the sample. These more computationally intensive algorithms are useful for quantitative structural measurements on finished objects.

[0044] The condenser and illumination system 410 can be mounted on a spring- loaded rack-and-pinion translation stage both to facilitate Kόhler alignment and also to provide access to samples.

[0045] The samples 465 can be mounted on an integrated translation stage with three-axis control. Coarse focusing is performed with a precision spring-loaded rack- and-pinion drive to minimize drift. Precise computer-controlled focusing and lateral translation can be performed with a conventional Mad City Labs Nano-View LP200, which combines stepping-motor two-axis coarse translators with piezoelectric three- axis controllers. The fine controllers offer 200 μm range in each axis with 0.4 nm resolution and better than 1 nm repeatability.

[0046] A major application area for the proposed fabrication instrument is in holographic assembly of photonic materials and devices. Another involves organization and monitoring of living biological samples. Consequently, a preferred system 500 shown in FIG. 9 incorporates a fiber spectrometer 510 and a pair of white light sources 520 to provide transmission and reflection spectra in real time. Spectroscopic information then can be used to fine tune structures before they are permanently set in place. It also can be used to assess trapped cells' viability and

their response to external stimuli. For optoelectronic applications, the spectroscopy subsystem 530 can be used to analyze samples' fluorescence when exposed to the trapping laser. Some semiconductor nanowires are observed to fluoresce brightly when trapped, and this fluorescence can be useful for selecting nanowires for assembly.

[0047] The spectroscopy subsystem 500 most preferably includes an Ocean Optics USB4000 fiber optic spectrometer, which offers better than 4 nm wavelength resolution over the range 300 to 1000 nm. Some regions in this wavelength range will be suppressed by the transmission characteristics of the dielectric multilayer mirrors used in the holographic trapping and imaging train. Mounting the spectrometer's input coupler below the microscope permits simultaneous holographic trapping, three- dimensional imaging and real-time spectroscopy, however. This type of coordination is essential for creating precise photonic structures under interactive control and also will be useful for characterizing biological systems during optical micromanipulation. The loss of sensitivity in selected wavelength ranges, therefore, is compensated by the additional functionality. Two Ocean Optics R-LS-I-LL rack-mounted halogen light sources will provide broad-band illumination for transmission and reflection spectroscopy. The entire system can be calibrated with standard samples. [0048] In another form of the invention, multi-point force spectroscopy and manipulation can be performed using a calibrated arrangement of optical tweezers' potential energy wells. An object's instantaneous displacement from the trap's equilibrium point can be used to measure the instantaneous force making it move. Most effectively, the potential energy well of a single one of the optical tweezers 130 can be calibrated by tracking the thermally driven motions of a trapped particle. This

general approach avoids the necessity of characterizing and calibrating an externally applied reference force. Statistically optimal methods can be used for analyzing trajectories of optically trapped particles to obtain time-resolved measurements of the forces on multiple holographically trapped particles simultaneously. These methods can be applied also to video microscopy data obtained in the proposed system. Imaging measurements of forces, however, only work when the trapped particles' displacements are large enough and slow enough to track with a video camera. They also require accurate calibrations for each one of the optical traps 130. [0049] Light scattered out of the optical trap or tweezer 130 by a trapped particle interferes with the unscattered portion of the beam to yield an interference pattern in the far-field forward-scattering direction. This interference transforms small particle motions into large intensity variations. Measuring these variations with a quadrant photodiode records the particle's displacement with sub-nanometer resolution over a bandwidth of tens of kilohertz. Once translated into equivalent forces, this technique can attain attonewton force resolution and can measure forces as large as several piconewtons. These specifications greatly exceed what is possible through imaging- based measurements.

[0050] It is noted that an alternative technique has been developed that relaxes the requirement to calibrate the optical tweezer 130 for force measurements and also is amenable to parallelization. In a further embodiment of the invention, the trapped particle deflects the trapping beam by an amount that depends on its displacement from the center of the trap. The beam's mean deflection corresponds to an average change in momentum imparted to the trap's photons by the particle, and thus to a force when normalized by the flux of photons in the beam. The effective force

deflecting the beam equals the force displacing the particle by Newton's third law. A single calibration of the trapping beam's intensity and the imaging train's magnification therefore calibrates the force transducer, independent of the optical tweezer' s trapping characteristics.

[0051] An individual trapping beam's displacement can therefore be measured by imaging the far-field scattering pattern through the condenser in a plane intermediate between the objective's front and back focal planes. Thus, optical deflection force spectroscopy can be applied to multiple optical traps 130 simultaneously, provided that the traps' images are resolved in the intermediate plane. The individual traps' deflections can be measured with sufficient precision with a conventional video camera 600 (see FIG. 10) to provide sub-femtonewton resolution over a ten piconewton range at a bandwidth limited by the camera's frame rate. The benefit over previously described approaches is that holographic trapping force spectroscopy provides information from a large number of traps simultaneously. The implementation then uses the video camera 600, protected with neutral density filters to detect the holographic traps' forward scattered beams 610. The camera 600 is focused with a tube lens 620 to optimize the trade-off between force resolution and spatial resolution. Output from the camera 600 will be digitized and analyzed with a conventional analysis software of the Applicant.

[0052] In yet another aspect of the invention, raw materials can be introduced to the optical fabrication system and finished products removed using microfluidic sample handling. Pioneered with conventional microlithography, microfluidics systems have since been implemented in polymeric materials with soft lithographic techniques that permit rapid prototyping at extremely low costs. Polymer-based microfluidic systems

also permit integration of microscopic pumps and valves. Pulsating flows from such pumps can be compensated by phase-locked modulation in laser intensity to maintain optimal conditions for optical fractionation. Integrated microfluidic systems also are compatible with electrokinetic driving technologies developed for capillary electrophoresis.

[0053] In still another embodiment of the invention, a single optical processing instrument can be combined with a large number of distinct microfluidic chips to generate a range of different optical fabrication and fractionation applications, such as manufacturing microfluidic chips through soft lithography in polydimethysiloxane (PDMS).

[0054] Another embodiment of the invention enables manufacture and quality assurance in fabricating three-dimensional structures out of dielectric building blocks. One application is to assemble and characterize three-dimensional photonic bandgap materials. These "semiconductors for light" have been demonstrated for radio and microwave wavelengths using macroscopic assembly techniques, and have been demonstrated for visible wavelengths in one and two dimensional lithographically defined microstructures. Processing high-index materials into three-dimensional photonic bandgap microstructures for optical applications has proved challenging. Creating appropriate small-scale structures with the proposed optical fabrication instrument therefore would open up new avenues for research and development in photonics as well as in biomolecular spectroscopy.

[0055] Very recently, icosahedral quasicrystals have been identified as the best candidate structure for achieving three-dimensional photonic bandgaps. These pose even greater challenges to conventional fabrication techniques than periodic

structures. These structures can be holographically assembled and typical results appear in FIG. 2C. These colloidal quasicrystals were assembled from silica spheres and then gelled into solid structures by photopolymerizing the surrounding fluid medium.

[0056] We also can use the optical fabrication and characterization instrument to organize ultra-high-index titania spheres into comparable permanent structures, and to measure their transmission and reflection spectra at optical wavelengths. Three- dimensional holographic microscopy will be particularly important for guiding and assessing the assembly process. Bulk photopolymerization of prototype structures can take advantage of ultraviolet light-emitting diodes (LED's) arranged as a ring illuminator around the condenser lens. Properties of the gel can be assessed with force spectroscopy on the spheres themselves.

[0057] In yet another application of the invention the optical tweezers 130 can manipulate and process semiconductor nanowires into three-dimensional structures to enable creating electronic and optoelectronic devices from chemically nanostructured materials. Heretofore, nanowire devices were created by randomly depositing the wires onto substrates and then defining functional structures through painstaking lithographic techniques. Now, devices can be assembled to order and use the systems described herein to build functional devices out of silicon nanowires, with a particular emphasis on sensor applications for biological and environmental monitoring. [0058] In yet another embodiment, infrared holographic trapping is useful for manipulating living biological cells. Holographic trapping, in particular, is useful for arranging multiple disparate cells into specific three-dimensional configurations. This kind of structuring is crucial for the proper growth and development of cells in living

tissues. Optically organized cellular assemblies have been demonstrated in model systems, including hepatocytes as a liver progenitor and islet cells for creating pancreatic implants. The three-dimensional cellular assemblies are transformed into artificial tissues by synthesizing a biodegradable gel around them. [0059] Holographic trapping coupled with holographic microscopy within microfluidic environments will greatly facilitate and accelerate the optical assembly of artificial tissues. For example, the systems herein can be used to organize chondrocytes and osteoblasts into three-dimensional models for developing teeth, with the intention of creating transplantable artificial dentin. [0060] The foregoing description of embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the present invention to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from practice of the present invention. The embodiments were chosen and described in order to explain the principles of the present invention and its practical application to enable one skilled in the art to utilize the present invention in various embodiments, and with various modifications, as are suited to the particular use contemplated.