Controlled buckling structures in semiconductor interconnects and nanomembranes for stretchable electronics

Abstract

In an aspect, the present invention provides stretchable, and optionally printable, components such as semiconductors and electronic circuits capable of providing good performance when stretched, compressed, flexed or otherwise deformed, and related methods of making or tuning such stretchable components. Stretchable semiconductors and electronic circuits preferred for some applications are flexible, in addition to being stretchable, and thus are capable of significant elongation, flexing, bending or other deformation along one or more axes. Further, stretchable semiconductors and electronic circuits of the present invention are adapted to a wide range of device configurations to provide fully flexible electronic and optoelectronic devices.

Claims

1. A method of making a stretchable and flexible device, the method comprising the steps of: providing a flexible substrate having a receiving surface; bonding a plurality of device components to said receiving surface, wherein at least one device component is connected to another device component by one or more interconnects, wherein at least two interconnects are electrically connected to a first device component; and wherein during said bonding step said at least two interconnects extend along at least two different directions from said first device component in a plane of said receiving surface to form a two-dimensional device array; generating a change in a level of strain to said flexible substrate receiving surface in a first direction and a second direction; wherein said at least one device component is bonded to said flexible substrate at a first level of strain and said change in strain causes a portion of a first interconnect to bend and separate from said flexible substrate in said first direction and a portion of a second interconnect to bend and separate from said flexible substrate in a second direction; and each of said at least two interconnects have a region that is curved and physically separated from said flexible substrate.

2. The method of claim 1, wherein said interconnect is an electrical interconnect and said at least one device component is one or more of a contact pad; a photodiode, a light emitting diode, an electrode, a transistor, an integrated circuit, a biosensor, a chemical sensors, an accelerometer, a pressure sensor, or a transducer.

3. The method of claim 1, wherein at least one of said curved region has a separation distance from said flexible substrate that is between 100 nm and 1 mm.

4. The method of claim 1, wherein each of said at least two interconnect curved regions have a separation distance from said flexible substrate that is between 100 nm and 1 mm.

5. The method of 1, wherein the at least two interconnects each have a first end and a second end, wherein said region that is curved and physically separated from said flexible substrate is a central portion between said first and said second ends; wherein: the first end of each of the at least two interconnects is in electrical communication with said first device component; each central portion of each of the at least two interconnects comprises at least two bent configuration regions and at least one contact point disposed between the at least two bent configuration regions; and each of the at least one contact point is in physical communication with the receiving surface of the flexible substrate.

6. The method of claim 5, wherein the at least one contact point is bonded to said flexible substrate receiving surface.

7. The method of claim 5, wherein each of the at least two interconnects further comprises one or more contact pads in electrical contact with the first end, the second end or both the first end and the second end.

8. The method of claim 7, wherein the first device component is in electrical contact with the one or more contact pads.

9. The method of claim 1, wherein said first device component comprises one or more materials selected from the group consisting of: a metal, a semiconductor, an insulator, a piezoelectric material, a ferroelectric material, a magnetostrictive material, an electrostrictive material, a superconductor, a ferromagnetic material, and a thermoelectric material.

10. The method of claim 1, wherein said first device component is an electronic device, an optical device, an opto-electronic device, a mechanical device, a microelectromechanical device, a nanoelectromechanical device, a microfluidic device or a thermal device.

11. The method of claim 10, wherein said at least two interconnects are tunable device components each having at least one electronic property, optical property or mechanical property that changes selectively with a level of strain of said curved region.

12. The method of claim 1, wherein said at least two interconnects is a plurality of stretchable interconnects, and wherein at least one of said plurality of stretchable interconnects comprises said at least one contact point that is in physical communication with said flexible substrate receiving surface and three or more bent configuration regions extend from said at least one contact point.

13. The method of claim 1, wherein each of the at least two curved regions comprises a folded region, a convex region, a concave region, or any combination thereof.

14. The method of claim 1, wherein the flexible substrate comprises an elastomeric material.

15. The method of claim 1, wherein the at least one device component is a plurality of device components, and wherein the at least two interconnects is a plurality of interconnects.

16. The method of claim 15, wherein the two-dimensional device array has a grid configuration, floral configuration, bridge configuration, or any combination thereof.

17. The method of claim 15, wherein one or more of said plurality of device components is connected to adjacent device components by said plurality of interconnects.

18. The method of claim 15, wherein at least a portion of the two-dimensional device array comprises two or more of the plurality of interconnects aligned in a direction parallel to each other or two or more of the plurality of interconnects oriented in two or more different directions.

19. The method of claim 15, wherein the two-dimensional device array comprises two or more device layers, and wherein each device layer comprises a plurality of the device components and a plurality of the interconnects.

20. The method of claim 15, wherein at least a portion of the receiving surface of the flexible substrate is curved, concave, convex or hemispherical.

21. The method of claim 15, wherein the two-dimensional device array comprises one or more of a photodetector, a photodiode array, a display, a light-emitting device, a photovoltaic device, a sensor array, a sheet scanner, a LED display, a semiconductor laser array, an optical imaging system, a large-area electronic device, a transistor array, a logic gate array, a microprocessor, an integrated circuit, or any combination of thereof.

22. The method of claim 15, wherein the two-dimensional device array has a floral configuration, wherein the at least two interconnects is a plurality of interconnects, and wherein at least one of the plurality of interconnects comprises: at least one contact point that is in physical communication with the receiving surface; and three or more bent configuration regions extending from the at least one contact point.

23. The method of claim 1, wherein said first level of strain elongates said receiving surface, and said step of generating a change in the level of strain is an at least partial relaxation of said elongated receiving surface.

24. The method of claim 1, wherein each of said change in said level of strain in said first direction and said second direction elongates said receiving surface in said first direction and in said second direction, and said step of generating said change in the level of strain is an at least partial relaxation of said elongated receiving surface in said first direction and in said second direction.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1A-1B summarizes one method for making a wavy or buckled stretchable metal interconnect. FIG. 1A is a flow-chart summary and FIG. 1B illustrates the flow-chart steps.

(2) FIG. 2 is a photograph of a stretchable wavy/buckled electrical interconnect, formed by retrieval from a rigid substrate onto a pre-strained, stretchable PDMS rubber substrate followed by the release of the strain to induce buckling.

(3) FIGS. 3A and 3B summarize one method of fabrication of wavy stretchable electrodes by deposition on a wavy-structured elastomer substrate.

(4) FIG. 4A-4B provides details relating to one method for fabricating a smooth wavy elastomer substrate. FIG. 4A is a flow-chart summary and FIG. 4B illustrates the flow-chart steps.

(5) FIG. 5A provides an image of a smoothly wavy PDMS substrate generated by the methods outlined in FIGS. 3A-3B and 4A-4B. The interconnect shown is capable of 22.6% stretchability and has a metal interconnect that is about 900 nm thick (700 nm Al/200 nm Au), a wavelength of about 38 microns and an amplitude (distance from peak to valley) of about 15.6 microns. FIG. 5B shows one end of the interconnect for establishing electrical contact with a device component. The device component may be positioned in a flat portion of the substrate.

(6) FIG. 6A Commercially available lenticular array (from Edmund Optics) with cusps. FIG. 6B Spin-coat photocurable epoxy to make smoothly wavy substrate. FIG. 6C Cast PDMS stamp against substrate from FIG. 6B to generate wavy elastomer stamp with smooth features.

(7) FIG. 7A-7C Stretchable electrodes deposited by evaporation through a shadowmask onto a smoothly wavy elastomer substrate. The electrodes maintain conductivity and connectivity during stretching up to 10% in tension. The scale bar is about 0.1 mm. FIG. 7A is a cross-section of waviness on an elastomeric substrate. FIG. 7B is a top view micrograph of electrode evaporated onto the wavy elastomer substrate. The focal plane is on the peaks of the wavy relief. FIG. 7C is a top view micrograph of an electrode evaporated onto the wavy elastomer substrate. The focal plane is on the valleys of the wavy relief.

(8) FIG. 8 is a schematic illustration of a process for the fabrication of a stretchable passive matrix LED display using stretchable electrodes.

(9) FIG. 9 illustrates the mechanical stretchability of a passive matrix LED display with wavy electrodes.

(10) FIG. 10 illustrates inorganic photodiode arrays distributed on a lens with spherical curvature. Shown: various lens shapes and angles

(11) FIG. 11 illustrates the need for stretchability when a planar sheet is wrapped around a spherical surface.

(12) FIG. 12 summarizes one scheme for fabricating stretchable buckled semiconductor arrays capable of conforming to spherically-curved surfaces.

(13) FIG. 13A-13D Optical microscopic images of buckled stretchable silicon arrays having a single connection grid configuration (FIGS. 13A and 13B), multiple connection (e.g., two) grid configuration (FIG. 13C), and a floral connection configuration (FIG. 13D). The stretchable interconnects are capable of electrically connecting photodiode, light-collecting/detecting devices, and other device components at, for example, the contact pad regions. These systems are capable of conforming to a curved surface. The configurations depicted in FIG. 13A-13D are on a PDMS substrate.

(14) FIG. 14A-14B Electron microscopic images of buckled stretchable silicon arrays in a grid configuration capable of supporting device components and conforming to a curved surface. The scale bar is 200 m in FIG. 14A and 50 m in FIG. 14B.

(15) FIG. 15A-15B Electron microscopic images of buckled stretchable silicon arrays in a grid configuration with adjacent contact pads connected to one another by a plurality (e.g., two) of interconnects and capable of supporting device components and conforming to a curved surface. The scale bar is 200 m in FIG. 15A and 50 m in FIG. 15B.

(16) FIG. 16A-16B Electron microscopic images of buckled stretchable silicon arrays in a floral configuration capable of supporting device components and conforming to a curved surface. The scale bar is 200 m in FIG. 16A and 50 m in FIG. 16B.

(17) FIG. 17A-17B Electron microscopic images of buckled stretchable silicon arrays in a bridge configuration capable of supporting device components and conforming to a curved surface. The scale bar is 200 m in FIG. 17A and 50 m in FIG. 17B.

(18) FIG. 18 Photograph of photodiodes in a grid array configuration on a stretchable buckled silicon array on PDMS.

(19) FIG. 19 demonstrates the reversible behavior of the stretchable interconnects during stretching and relaxation. The system is relaxed in panel 1. The system is stretched as indicated by the stretching arrows in panels 2, 3 and 4. The maximum stretch in panel 4 is about 10% and results in a substantially flat interconnect for the interconnect aligned in the direction of the stretching force. The system is released in panels 5-8, and panel 8 has a geometry and configuration equivalent to that shown in panel 1. The scale bar is 0.2 mm.

(20) FIG. 20 Bubble stamp or Balloon stamp device capable of conformal contact to curved substrates as well as flat substrates.

(21) FIG. 21 Another device capable of conforming to both spherically curved and flat surfaces is a stretchable spherically-molded stamp. The stamp is cast against a curved surface (in this example a concave lens) and removed. The stamp is stretched to substantially flatten its surface and to which interconnects can be transferred.

(22) FIG. 22 Stretchable buckled silicon arrays during a stretching cycle on a bubble or balloon stamp. In this example the interconnect between adjacent contact pads comprises two wavy interconnects (Si 290 nm in thickness). The stretch test uses bubble expansion to provide multi-directional stretching. The right-most panel is under maximum stretch and the bottom two panels show that when the stretching force is removed, the interconnects relax back to their prestretched configuration shown in the top-left panel.

(23) FIG. 23 Silicon printed via balloon stamps onto glass lenses coated with adhesives (PDMS or SU-8).

(24) FIG. 24A-24D summarizes processing steps for engineering 3D buckled shapes in semiconductor nanoribbons. FIG. 24A Fabricating a UVO mask and using it to pattern the surface chemistry on a PDMS substrate. FIG. 24B Forming buckled GaAs ribbons and embedding them in PDMS. FIG. 24C Response of buckled GaAs ribbons to stretching and compressing. FIG. 24D SEM image of a sample formed using the procedures in a and b. The prestrain used for generating this sample was 60%, with W.sub.act=10 m and W.sub.in=400 m.

(25) FIG. 25A-25B Side-view profiles of buckles formed on PDMS substrates using prestrains of 33.7% and with: (FIG. 25A) W.sub.act=10 m and W.sub.in=190 m; and (FIG. 25B) W.sub.act=100 m and W.sub.in=100 m. Both samples exhibit buckles in the inactivated regions due to detachment of ribbons from the PDMS. Sinusoidal waves with small peaks formed only in the activated regions with W.sub.act=100 m. A comparison of these two samples indicates that selecting W.sub.act smaller than a critical value avoids the formation of small wavy structures.

(26) FIG. 26 Side-view image of a buckled GaAs ribbon embedded in PDMS after microtoming. This image shows that the PDMS fully fills the gaps between the ribbons and the underlying substrate. The buckles in this case are formed with a prestrain of 60% and with W.sub.act=10 m and W.sub.in=300 m. The PDMS prepolymer cast on the surface of these buckled ribbons is cured in an oven at 65 C. for 4 hours.

(27) FIG. 27A-27D Optical micrographs of the side-view profiles of buckled (FIG. 27A, 27D) GaAs and (FIG. 27B, 27C) Si ribbons. FIG. 27A GaAs ribbon structures formed on PDMS patterned with W.sub.act=10 m and W.sub.in=190 m, with different prestrains: 11.3%, 25.5%, 33.7%, and 56.0% (from top to bottom). The dotted lines for .sub.pre=33.7% and 56.0% are mathematically predicted interconnect geometrical shape. FIG. 27B Si ribbon structures formed on a PDMS substrate prestrained to 50% and patterned with W.sub.act=15 m and W.sub.in: 350, 300, 250, 250, 300, and 350 m (from left to right). The image was taken by tilting the sample at angle of 45. FIG. 27C Si ribbon structures formed on a PDMS substrate prestrained to 50% and patterned with parallel lines of adhesion sites (W.sub.ad=15 m and W.sub.in=250 m) orientated at angles of 30 with respect to the lengths of the ribbons. The image was taken by tilting the sample at angle of 75. FIG. 27D GaAs ribbon structures formed on PDMS substrates prestrained to 60% with W.sub.act=10 m and different W.sub.in: 100, 200, 300, and 400 m (from top to bottom).

(28) FIG. 28A-28C Stretching and compressing of buckled GaAs ribbons embedded in PDMS. FIG. 28A Images of a single buckled ribbon stretched to different levels of tensile strain (positive %). Fracture occurs near 50%. FIG. 28B Images of a single buckled ribbon compressed to different levels of compressive strain (negative %). Small, short period wavy geometries appear at the peaks of the buckles for compressive strains larger than 15%. FIG. 28C Images of a single buckled ribbon compressed to different levels of compressive strain. The buckles in these cases were formed with prestrain of 60% with W.sub.act=10 m and W.sub.in=400 m (FIG. 28A, 28B) and with W.sub.act=10 m and W.sub.in=300 m (FIG. 28C). The red lines and arrows in each panel indicate the same positions on the same ribbons to highlight the mechanical deformations. The insets provide magnified images of the sections marked with the white boxes, clearly showing the formation of cracks at high compressive strains. The numbers corresponding to stretching or compressing degree were computed according to:

(29) $\left|\frac{L_{\mathrm{projected}}^{\max }-L_{\mathrm{projected}}^0}{L_{\mathrm{projected}}^0}\right|*100\%.$

(30) FIG. 29 Photograph of a sample with two layer of buckled GaAs ribbons arrays. The structure is fabricated in a layer by layer scheme. The first layer of GaAs ribbons (buckled geometry defined with a prestrain of 60% and with W.sub.act=10 m and W.sub.in=400 m) is embedded in PDMS. The second layer of buckled ribbons is formed on the surface of this substrate using a prestrain of 50% and with W.sub.act=10 m and W.sub.in=300 m.

(31) FIG. 30A-30D Bending of buckled ribbons on surfaces and in matrixes of PDMS. FIG. 30A-30D, Optical microscopic images with low magnification (top left frames) and high magnification (right frames) and schematic illustrations (bottom left frames) of buckled GaAs ribbons on PDMS with (FIG. 30A) concave, (FIG. 30B) flat, and (FIG. 30C) convex surfaces. The scale bars in c apply to a and b. (FIG. 30D) Images of buckled ribbons embedded in PDMS (left) before and (right) after bending. The top and bottom frames show the curvatures of the top and bottom surfaces, respectively. The scale bars in the right images apply also to the left images. The buckled ribbons are formed with a pre-strain of 60% and with W.sub.act=10 m and W.sub.in=400 m.

(32) FIG. 31A-31D Characterization of stretchable metal-semiconductor-metal photodetectors (MSM PDs). FIG. 31A Schematic illustrations of the geometry (top), an equivalent circuit (middle), and optical images of a buckled PD before and during stretching (bottom). FIG. 31B Current (I)-voltage (V) curves recorded from a buckled PD that was irradiated by an IR lamp with different output intensities. I-V characteristics of PDs illuminated with constant luminance and (FIG. 31C) stretched or (FIG. 31D) compressed by different degrees.

(33) FIG. 32 A hemispherical elastomeric transfer stamp can liftoff interconnected Si CMOS chiplets from a conventional wafer and then transform their geometry into a hemispherical shape. The pop-up interconnects between the chiplets accommodates the strains associated with this planar to curved surface transformation.

(34) FIG. 33 Transfer of interconnected CMOS chiplets from a hemispherical stamp to a matched hemispherical device substrate. The photocurable adhesive layer bonds to the CMOS to the device substrate and also planarizes the surface.

(35) FIG. 34 Printer apparatus with fixturing, actuators and vision systems compatible with hemispherical stamps.

(36) FIG. 35 Compressible array of single crystal silicon islands electrically connected by pop up ribbon interconnects, on a hemispherical stamp.

(37) FIG. 36 Optical images of an array of interconnected single crystal silicon islands inked onto the surface of a hemispherical stamp with radius of curvature 2 cm.

(38) FIG. 37 Stress/strain curves for various silicone elastomers that can be used for the hemispherical stamps. Linear, purely elastic responses for strains less than 20% are important.

(39) FIG. 38 Finite element modeling of the spherical to planar transformation in a hemispherical stamp with an initially uniform thickness of 0.57 mm.

(40) FIG. 39 Schematic illustration of steps for fabricating two-dimensional, wavy semiconductor nanomembranes on elastomeric supports.

(41) FIG. 40A-40F Optical micrographs of 2D wavy structures in silicon nanomembranes at various stages during their formation. The insets show two-dimensional power spectra. FIG. 40G Image of the fully developed structure, at low magnification. For this sample, the thickness of the silicon is 100 nm with the lateral dimension of ca. 44 mm.sup.2, the substrate is PDMS, and the thermally induced prestrain is 3.8%. FIG. 40H Plot of the short wavelength corresponding to frames (FIG. 40A-40F) and (FIG. 40I) histogram of long wavelength evaluated at various points from frame (FIG. 40G).

(42) FIG. 41A AFM and FIG. 41B-41D SEM images (tilt angle)60) of a 2D wavy Si nanomembrane on PDMS. The thickness of the silicon is 100 nm, and the thermal prestrain is 3.8%. These images highlight the highly periodic nature of the wavy patterns, the good bonding between the Si and the PDMS as evidenced by the intimate contact visible at the edges of the Si and PDMS near the holes etched in the Si, and the lack of correlation between the positions of the wave structures and these holes.

(43) FIG. 42A Optical micrographs of 2D wavy Si nanomembranes with various thickness (55, 100, 260, 320 nm) on PDMS, formed with a thermal prestrain of 3.8%, and FIG. 42B dependence of the short wavelength and amplitude on Si thickness.

(44) FIG. 43A Optical micrographs of 2D wavy Si nanomembranes under different uniaxial strains, applied at three different orientations. These samples consist of Si membranes with thicknesses of 100 nm on PDMS, formed with a thermal prestrain of 3.8%. The images were collected in the relaxed state before stretching (top frames), the relaxed state after stretching (bottom frames), and at uniaxial applied tensile strains of 1.8% (top middle frames) and 3.8% (bottom middle frames). FIG. 43B Dependence of the short wavelength on applied strain in the three different directions.

(45) FIG. 44 AFM images of different regions of a 2D wavy Si nanomembrane, showing the 1D wavy geometry characteristics of a region near the edge of the membrane (top frame), a region slightly removed from this edge area (middle frame), and a region near the center of the membrane (bottom frame). The sample consisted of a Si membrane with thickness of 100 nm on PDMS, formed with a thermal prestrain of 3.8%.

(46) FIG. 45A-45C Optical micrographs of 2D wavy Si nanomembranes with lengths of 1000 m and with widths of 100, 200, 500, and 1000 m. These membranes all have thicknesses of 100 nm and were formed on the same PDMS substrate with thermal prestrains of (FIG. 45A) 2.3% and (FIG. 45B) 4.8%. FIG. 45C Dependence of the edge effect length on prestrain for similar membranes.

(47) FIG. 46A-46D Optical micrographs of 2D wavy Si nanomembranes with different shapes: (FIG. 46A) circle, (FIG. 46B) oval, (FIG. 46C) hexagon, and (FIG. 46D) triangle. These membranes all have thicknesses of 100 nm and were formed on PDMS with a thermal prestrain of 4.8%.

(48) FIG. 47A-47H Optical micrographs of wavy structures of Si nanomembranes with shapes designed to exploit edge effects to provide 2D stretchability in interconnected arrays of flat islands. In both cases illustrated here, the Si is 100 nm thick, the squares are 100100 m and the ribbon connections are 30150 m lines. The prestrain is (FIG. 47A, 47E) 2.3% and (FIG. 47C, 47G) 15%. SEM images (tilt angle of 75) of selected regions that show ribbons and squares of (FIG. 47A, 47C, 47E, 47G) are shown in (FIG. 47B, 47D, 47F, 47H), respectively. The insets of high-magnification SEM images show the raised region of waves in FIG. 47B and FIG. 47D.

(49) FIG. 48 is photograph of the sample of 2D wavy Si nanomembrane (100 nm thick, 45 mm.sup.2, and 3.8% thermal prestrain) on PDMS substrate wave (top frame), and (i) the 1D waves at the edge, (ii) the herringbone waves at the inner region, and (iii) disordered herringbone waves at the center. The scale bar is 50 m.

(50) FIG. 49 Schematic illustration of the characteristic lengths in the herringbone wave structures.

(51) FIG. 50 Si strain as a function of applied thermal prestrain at the herringbone and 1D waves. Si strain was measured experimentally by .sub.Si=(L)/, where L and are the surface and horizontal distance in AFM surface profile.

(52) FIG. 51 Optical microscope images of herringbone waves after the cycles of stretching test (.sub.st=4.0%). The test sample was prepared with 100 nm thick Si membrane and 3.8% biaxial thermal prestrain. The herringbone waves were recovered to have quite similar structures with the original, after the cycles of stretching test up to 15 times, except the some defects originated from the crack of membrane.

(53) FIG. 52 Schematic illustration of the unfolding of herringbone waves with application of uniaxial tensile strain. The compressive strain .sub.cp is due to the Poisson effect with tensile strain .sub.st.

(54) FIG. 53 Optical microscope images of the morphology change of herringbone waves during heating and cooling process as a biaxial stretching test. The test sample was prepared with 100 nm thick Si membrane and 2.9% biaxial thermal prestrain.

(55) FIG. 54 summarizes one method of fabrication of wavy stretchable electrodes by deposition on a structured wavy master, followed by casting a stamp on that master, curing the stamp, and thereby transferring the electrodes to the master upon release.

(56) FIG. 55 provides an image of stretchable metal electrodes (Au, 300 nm thick) on wavy PDMS prepared by the methods in FIG. 4A-4B combined with those in FIG. 54. The bottom panel is a graph of measured electrical resistance data of the stretchable wavy metal electrodes as a function of applied tensile strain (up to 30%).

(57) FIG. 56A-56C is an example of an application of the present method for making flexible, stretchable iLED strip-lights. FIG. 56A is a photomicrograph illustrating the device is capable of large bending, and in this example the bending radius is 0.85 cm. FIG. 56B provides a cross-section (top panel, scale bar 40 m) and a top-view (bottom panel, scale bar 3 mm) of stretchable metal on a wavy PDMS substrate. The metal is capable of stretching about 30% without significant degradation of physical properties. FIG. 56C is a plot of the effects of local strain on the wavelength (squares, left axis) and amplitude (circles, right axis) of sinusoidally-wavy metal interconnects on PDMS (shown in FIG. 56B). As the strain increases, there is a corresponding increase in the wavelength and decrease in the amplitude of the metal.

(58) FIG. 57 Schematic illustration of a printed semiconductor nanomaterials based approach to heterogeneous, three dimensional electronics. The process involves the repetitive transfer printing of collections of nanotubes, nanowires, nanoribbons or other active nanomaterials, separately formed on source substrates, to a common device substrate to generate interconnected electronics in ultrathin, multilayer stack geometries.

(59) FIG. 58A Optical micrograph of a three dimensional multilayer stack of arrays of single crystal silicon metal oxide field effect transistors (MOSFETs) that use printed silicon nanoribbons for the semiconductor. The bottom (labeled 1st), middle (labeled 2nd) and top (labeled 3rd) parts of this image correspond to regions with one, two and three layers of devices, respectively. FIG. 58B Schematic cross sectional (top) and angled (bottom) views. S, D and G refer to source, drain and gate electrodes (all shown in gold), respectively. The light and dark blue regions correspond to doped and undoped regions of the silicon ribbons; the purple layer is the SiO.sub.2 gate dielectric. FIG. 58C Three dimensional images (left frame: top view; right frame: angled view) collected by confocal microscopy on a device substrate like that shown in FIGS. 58A and 58B. The layers are colorized (gold: top layer; red: middle layer; blue: bottom layer; silicon: grey) for ease of viewing. FIG. 58D Current-voltage characteristics of Si MOSFETs in each of the layers, showing excellent performance (mobilities of 47030 cm.sup.2/Vs) and good uniformity in the properties. The channel lengths and widths are 19 and 200 m, respectively.

(60) FIG. 59A Optical micrograph of three dimensional, heterogeneously integrated electronic devices, including GaN nanoribbon HEMTs, Si nanoribbon MOSFETs and SWNT network TFTs, in a three layer stack. FIG. 59B Three dimensional image collected by confocal microscopy. The layers are colorized (gold: top layer, Si MOSFETs; red: middle layer, SWNT TFTs; blue: bottom layer) for ease of viewing. FIG. 59C Electrical characteristics of GaN devices on the first layer (channel lengths, widths and gate widths of 20, 170 and 5 m, respectively), SWNT devices on the second layer (channel lengths and widths of 50 and 200 m, respectively) and Si devices on the third layer (channel lengths and widths of 19 and 200 m, respectively). FIG. 59D Normalized transconductances (g.sub.m/g.sub.0m) of devices in each layer (black squares: Si MOSFETs; red circles: SWNT TFTs; green triangles: GaN HEMTs) as a function of bending radius of the plastic substrate (left). Image of the bent system and probing apparatus (right).

(61) FIG. 60A Image of a printed array of 3D silicon NMOS inverters on a polyimide substrate. The inverters consist of MOSFETs (channel lengths of 4 m, load-to-driver width ratio of 6.7, and a driver width of 200 m) on two different levels, interconnected by electrical via structures. The image on the top right provides a magnified view of the region indicated by the red box in the left frame. The graph on the bottom right shows transfer characteristics of a typical inverter. FIG. 60B Transfer characteristics of a printed complementary inverter that uses a p channel SWNT TFT (channel length and width of 30 and 200 m, respectively) and an n channel Si MOSFET (channel length and width of 75 and 50 m, respectively). The insets provide an optical micrograph of an inverter (left) and a circuit schematic (right). FIG. 60C Current-voltage response of a GaAs MSM (channel length and width of 10 and 100 m, respectively) integrated with a Si MOSFET (channel length and width of 9 and 200 m, respectively) at different levels of illumination from dark to 11 W with an infrared light source at 850 nm. The insets shows and optical image, and a circuit diagram.

(62) FIG. 61 Image of an automated stage for transfer-printing, capable of registration to within 1 m.

(63) FIG. 62A Optical micrographs of three dimensional, heterogeneously integrated arrays of Si MOSFETs and GaN HEMTs on a polyimide substrate. The right inset shows a cross sectional schematic view. The electrodes (gold), SiO.sub.2 (PEO; purple), Si (light blue: undoped; dark blue: doped), GaN (dark green: ohmic contacts; light green: channel), polyimide (PI; brown) and polyurethane (PU; tan) are all shown. FIG. 62B Current-voltage characteristics of a typical Si MOSFET (channel length and width of 19 and 200 m, respectively) and a GaN HEMT with (channel length, widths and gate widths of 20, 170 m and 5 m, respectively). The data for the Si and GaN in the left frames were measured at V.sub.dd=0.1V and V.sub.dd=2V, respectively.

(64) FIG. 63A Optical micrographs of three dimensional, heterogeneously integrated arrays of Si MOSFETs and SWNT TFTs on a polyimide substrate. The right inset shows a cross sectional schematic view. The electrodes (gold), epoxy (cyan), SiO2 (PEO; purple), Si (light blue: undoped; dark blue: doped), SWNTs (grey), polyimide (PI; brown) and cured polyimide (tan) are all shown. FIG. 63B Current-voltage characteristics of a typical SWNT TFT (channel length and width of 75 m and 200 m, respectively) and a typical Si MOSFET (gate length and channel width 19 m and 200 m, respectively). The data for the SWNT and Si in the left frames were measured at V.sub.dd=0.5 V and V.sub.dd=0.1 V, respectively.

(65) FIG. 64A Cross sectional schematic illustration of three dimensional, heterogeneously integrated arrays of Si MOSFETs, SWNT TFTs and GaN HEMTs on a polyimide substrate. FIG. 64B Transfer characteristics, effective mobilities and on/off ratios for several of the Si MOSFETs (channel width=200 m, black line: channel length=9 m, red: 14 m, green:19 m, blue: 24 m), FIG. 64C the SWNT TFTs (channel width=200 m, black line: channel length=25 m, red: 50 m, green:75 m, blue: 100 m) and FIG. 64D transfer characteristics, transconductances and on/off ratios for GaN HEMTs (channel lengths, widths and gate widths of 20 m, 170 m and 5 m, respectively).

(66) FIG. 65A Schematic structure of the cross section of SWNT-Si CMOS inverter built on a silicon wafer substrate. FIG. 65B Transfer and I-V characteristics of n-channel Si MOSFET and p-channel SWNT TFT forming CMOS inverter. FIG. 65C Calculated transfer characteristics of inverter and I-V characteristics of Si and SWNT transistors.

(67) FIG. 66A Schematic structure of the cross section and circuit schematic of GaAs MSM-Si MOSFET IR detector built on a polyimide substrate. FIG. 66B Current-Voltage characteristic of GaAs MSM IR detector (L=10 m, W=100 m) and transfer and I-V characteristics of Si MOSFET (L=9 m, W=200 m) with a 3V supply. FIG. 66C Calculated IV characteristic of GaAs MSM and I-V response of a GaAs MSM integrated with a Si MOSFET with a 3V supply.

(68) FIGS. 67A and 67B schematically illustrates an optical device (waveguide array) produced via the controlled buckling of an optical microstructure partially adhered to a deformable substrate.

(69) FIG. 68 schematically illustrates a mechanical device (e.g., accelerometer/pressure sensor) produced via the controlled buckling of a conductive microstructure partially adhered to a deformable substrate.

(70) FIG. 69 schematically illustrates a thermal device (microbolometer) produced via the controlled buckling of a thermoresistive microstructure partially adhered to a deformable substrate.

DETAILED DESCRIPTION OF THE INVENTION

(71) Stretchable refers to the ability of a material, structure, device or device component to be strained without undergoing fracture. In an exemplary embodiment, a stretchable material, structure, device or device component may undergo strain larger than about 0.5% without fracturing, preferably for some applications strain larger than about 1% without fracturing and more preferably for some applications strain larger than about 3% without fracturing.

(72) A component is used broadly to refer to a material or individual component used in a device. An interconnect is one example of a component and refers to an electrically conducting material capable of establishing an electrical connection with a component or between components. In particular, the interconnect may establish electrical contact between components that are separate and/or can move with respect to each other. Depending on the desired device specifications, operation, and application, the interconnect is made from a suitable material. For applications where a high conductivity is required, typical interconnect metals may be used, including but not limited to copper, silver, gold, aluminum and the like, alloys. Suitable conductive materials may include a semiconductor like silicon, indium tin oxide, or GaAs.

(73) Semiconductor refers to any material that is an insulator at a very low temperature, but which has a appreciable electrical conductivity at a temperatures of about 300 Kelvin. In the present description, use of the term semiconductor is intended to be consistent with use of this term in the art of microelectronics and electronic devices. Semiconductors useful in the present invention may comprise element semiconductors, such as silicon, germanium and diamond, and compound semiconductors, such as group IV compound semiconductors such as SiC and SiGe, group III-V semiconductors such as AISb, AlAs, Aln, AlP, BN, GaSb, GaAs, GaN, GaP, InSb, InAs, InN, and InP, group III-V ternary semiconductors alloys such as Al.sub.xGa.sub.1-xAs, group II-VI semiconductors such as CsSe, CdS, CdTe, ZnO, ZnSe, ZnS, and ZnTe, group I-VII semiconductors CuCl, group IV-VI semiconductors such as PbS, PbTe and SnS, layer semiconductors such as PbI.sub.2, MoS.sub.2 and GaSe, oxide semiconductors such as CuO and Cu.sub.2O. The term semiconductor includes intrinsic semiconductors and extrinsic semiconductors that are doped with one or more selected materials, including semiconductor having p-type doping materials and n-type doping materials, to provide beneficial electronic properties useful for a given application or device. The term semiconductor includes composite materials comprising a mixture of semiconductors and/or dopants. Specific semiconductor materials useful for in some applications of the present invention include, but are not limited to, Si, Ge, SiC, AlP, AlAs, AlSb, GaN, GaP, GaAs, GaSb, InP, InAs, GaSb, InP, InAs, InSb, ZnO, ZnSe, ZnTe, CdS, CdSe, ZnSe, ZnTe, CdS, CdSe, CdTe, HgS, PbS, PbSe, PbTe, AlGaAs, AlInAs, AlInP, GaAsP, GaInAs, GaInP, AlGaAsSb, AlGaInP, and GaInAsP. Porous silicon semiconductor materials are useful for applications of the present invention in the field of sensors and light emitting materials, such as light emitting diodes (LEDs) and solid state lasers. Impurities of semiconductor materials are atoms, elements, ions and/or molecules other than the semiconductor material(s) themselves or any dopants provided to the semiconductor material. Impurities are undesirable materials present in semiconductor materials which may negatively impact the electronic properties of semiconductor materials, and include but are not limited to oxygen, carbon, and metals including heavy metals. Heavy metal impurities include, but are not limited to, the group of elements between copper and lead on the periodic table, calcium, sodium, and all ions, compounds and/or complexes thereof.

(74) Semiconductor element and semiconductor structure are used synonymously in the present description and broadly refer to any semiconductor material, composition or structure, and expressly includes high quality single crystalline and polycrystalline semiconductors, semiconductor materials fabricated via high temperature processing, doped semiconductor materials, organic and inorganic semiconductors and composite semiconductor materials and structures having one or more additional semiconductor components and/or non-semiconductor components, such as dielectric layers or materials and/or conducting layers or materials

(75) An interconnect that is stretchable is used herein to broadly refer to an interconnect capable of undergoing a variety of forces and strains such as stretching, bending and/or compression in one or more directions without adversely impacting electrical connection to, or electrical conduction from, a device component. Accordingly, a stretchable interconnect may be formed of a relatively brittle material, such as GaAs, yet remain capable of continued function even when exposed to a significant deformatory force (e.g., stretching, bending, compression) due to the interconnect's geometrical configuration. In an exemplary embodiment, a stretchable interconnect may undergo strain larger than about 1%, 10% or about 30% without fracturing. In an example, the strain is generated by stretching an underlying elastomeric substrate to which at least a portion of the interconnect is bonded.

(76) A device component is used to broadly refer to an individual component within an electrical, optical, mechanical or thermal device. Component can be one or more of a photodiode, LED, TFT, electrode, semiconductor, other light-collecting/detecting components, transistor, integrated circuit, contact pad capable of receiving a device component, thin film devices, circuit elements, control elements, microprocessors, transducers and combinations thereof. A device component can be connected to one or more contact pads as known in the art, such as metal evaporation, wire bonding, application of solids or conductive pastes, for example. Electrical device generally refers to a device incorporating a plurality of device components, and includes large area electronics, printed wire boards, integrated circuits, device components arrays, biological and/or chemical sensors, physical sensors (e.g., temperature, light, radiation, etc.), solar cell or photovoltaic arrays, display arrays, optical collectors, systems and displays.

(77) Substrate refers to a material having a surface that is capable of supporting a component, including a device component or an interconnect. An interconnect that is bonded to the substrate refers to a portion of the interconnect in physical contact with the substrate and unable to substantially move relative to the substrate surface to which it is bonded. Unbonded portions, in contrast, are capable of substantial movement relative to the substrate. The unbonded portion of the interconnect generally corresponds to that portion having a bent configuration, such as by strain-induced interconnect bending.

(78) A component in conformal contact with a substrate refers to a component that covers a substrate and retains a three-dimensional relief feature whose pattern is governed by the pattern of relief features on the substrate.

(79) In the context of this description, a bent configuration refers to a structure having a curved conformation resulting from the application of a force. Bent structures in the present invention may have one or more folded regions, convex regions, concave regions, and any combinations thereof. Bent structures useful in the present invention, for example, may be provided in a coiled conformation, a wrinkled conformation, a buckled conformation and/or a wavy (i.e., wave-shaped) configuration.

(80) Bent structures, such as stretchable bent interconnects, may be bonded to a flexible substrate, such as a polymer and/or elastic substrate, in a conformation wherein the bent structure is under strain. In some embodiments, the bent structure, such as a bent ribbon structure, is under a strain equal to or less than about 30%, a strain equal to or less than about 10%, a strain equal to or less than about 5% and a strain equal to or less than about 1% in embodiments preferred for some applications. In some embodiments, the bent structure, such as a bent ribbon structure, is under a strain selected from the range of about 0.5% to about 30%, a strain selected from the range of about 0.5% to about 10%, a strain selected from the range of about 0.5% to about 5%. Alternatively, the stretchable bent interconnects may be bonded to a substrate that is a substrate of a device component, including a substrate that is itself not flexible. The substrate itself may be planar, substantially planar, curved, have sharp edges, or any combination thereof. Stretchable bent interconnects are available for transferring to any one or more of these complex substrate surface shapes.

(81) Thermal contact refers to the ability of two materials that are capable of substantial heat transfer from the higher temperature material to the lower temperature material, such as by conduction. Bent structures resting on a substrate are of particular use in providing regions that are in thermal contact (e.g., bond regions) with the substrate and other regions that are not in thermal contact (e.g., regions that are insulated and/or physically separated from the substrate).

(82) Interconnects can have any number of geometries or shape, so long as the geometry or shape facilitates interconnect bending or stretching without breakage. A general interconnect geometry can be described as buckled or wavy. In an aspect, that geometry can be obtained by exerting a force (e.g., a strain) on the interconnect by exerting a force on an underlying deformable substrate, such that a change in a dimension of the underlying substrate generates buckles or waves in the interconnect because portions of the interconnect are bonded to the substrate, and regions between the bound portions are not bonded. Accordingly, an individual interconnect may be defined by ends that are bonded to a substrate, and a curved central portion between the ends that is not substrate-bonded. Curved or buckled refers to relatively complex shapes, such as by an interconnect having one or more additional bond regions in the central portion. Arc-shaped refers to a generally sinusoidal shape having an amplitude, where the amplitude corresponds to the maximum separation distance between the interconnect and the substrate surface.

(83) The interconnect can have any cross-sectional shape. One shape interconnect is a ribbon-shaped interconnect. Ribbon refers to a substantially rectangular-shaped cross-section having a thickness and a width. Specific dimensions depend on the desired conductivity through the interconnect, the composition of the interconnect and the number of interconnects electrically connecting adjacent device components. For example, an interconnect in a bridge configuration connecting adjacent components may have different dimensions than a single interconnect connecting adjacent components. Accordingly, the dimensions may be of any suitable values, so long as a suitable electrical conductivity is generated, such as widths that are between about 10 m and 1 cm and thickness between about 50 nm to 1, or a width to thickness ratio ranging from between about 0.001 and 0.1, or a ratio that is about 0.01.

(84) Elastomeric refers to a polymeric material which can be stretched or deformed and return, at least partially, to its original shape without substantial permanent deformation. Elastomeric substrates commonly undergo substantially elastic deformations. Exemplary elastomeric substrates useful in the present include, but are not limited to, elastomers and composite materials or mixtures of elastomers, and polymers and copolymers exhibiting elasticity. In some methods, the elastomeric substrate is prestrained via a mechanism providing for expansion of the elastic substrate along one or more principle axes. For example, prestraining may be provided by expanding the elastic substrate along a first axes, including expansion in a radial direction to transform a hemispherical surface to a flat surface. Alternatively, the elastic substrate may be expanded along a plurality of axes, for example via expansion along first and second axis orthogonally positioned relative to each other. Means of prestraining elastic substrates via mechanisms providing expansion of the elastic substrate include bending, rolling, flexing, flattening, expanding or otherwise deforming the elastic substrate. The prestraining means also includes prestraining provided by raising the temperature of the elastic substrate, thereby providing for thermal expansion of the elastic substrate. Elastomers useful in the present invention may include, but are not limited to, thermoplastic elastomers, styrenic materials, olefenic materials, polyolefin, polyurethane thermoplastic elastomers, polyamides, synthetic rubbers, PDMS, polybutadiene, polyisobutylene, poly(styrene-butadiene-styrene), polyurethanes, polychloroprene and silicones.

(85) Strain is defined as: =L/L for lengths changed from L (at rest) to L+L (under an applied force), where L is the displacement distance from resting. Axial strain refers to a force applied to an axis of the substrate to generate the displacement L. Strain is also generated by forces applied in other directions, such as a bending force, a compressive force, a shearing force, and any combination thereof. Strain or compression may also be generated by stretching a curved surface to a flat surface, or vice versa. Level of strain refers to the magnitude of the strain and can range from negative (corresponding to compression) to zero (relaxed state) to positive (corresponding to elongation or stretching).

(86) Young's modulus is a mechanical property of a material, device or layer which refers to the ratio of stress to strain for a given substance. Young's modulus may be provided by the expression;

(87) $\begin{array}{cc}E=\frac{\left(\mathrm{stress}\right)}{\left(\mathrm{strain}\right)}=\left(\frac{L_0}{L}\frac{F}{A}\right);& \left(\mathrm{II}\right)\end{array}$
wherein E is Young's modulus, L.sub.0 is the equilibrium length, L is the length change under the applied stress, F is the force applied and A is the area over which the force is applied. Young's modulus may also be expressed in terms of Lame constants via the equation:

(88) $\begin{array}{cc}E=\frac{\left(3+2\right)}{+};& \left(\mathrm{III}\right)\end{array}$
wherein and are Lame constants. High Young's modulus (or high modulus) and low Young's modulus (or low modulus) are relative descriptors of the magnitude of Young's modulus in a give material, layer or device. In the present invention, a high Young's modulus is larger than a low Young's modulus, preferably about 10 times larger for some applications, more preferably about 100 times larger for other applications and even more preferably about 1000 times larger for yet other applications. Complex surface shapes are obtained by polymerizing an elastomer having a spatially-varying Young's modulus and/or by layering an elastomer with multiple layers in various locations having different elasticity.

(89) Compression is used herein in a manner similar to the strain, but specifically refers to a force that acts to decrease a characteristic length, or a volume, of a substrate, such that L<0.

(90) Fracturing or fracture refers to a physical break in the interconnect, such that the interconnect is not capable of substantial electrical conductivity.

(91) A pattern of bond sites refers to spatial application of bonding means to a supporting substrate surface and/or to the interconnects so that a supported interconnect has bond regions and non-bond regions with the substrate. For example, an interconnect that is bonded to the substrate at its ends and not bonded in a central portion. Further shape control is possible by providing an additional bond site within a central portion, so that the not-bonded region is divided into two distinct central portions. Bonding means can include adhesives, adhesive precursors, welds, photolithography, photocurable polymer. In general, bond sites can be patterned by a variety of techniques, and may be described in terms of surface-activated (W.sub.act) areas capable of providing strong adhesive forces between substrate and feature (e.g., interconnect) and surface-inactive (W.sub.in) where the adhesive forces are relatively weak. A substrate that is adhesively patterned in lines may be described in terms of W.sub.act and W.sub.in dimensions. Those variables, along with the magnitude of prestrain, .sub.pre affect interconnect geometry.

(92) Spatial variation refers to a parameter that has magnitude that varies over a surface, and is particularly useful for providing two-dimensional control of component relief features, thereby providing spatial control over the bendability of a device or device component.

(93) Carbon nanomaterial refers to a class of structures comprising carbon atoms and having at least one dimension that is between one nanometer and one micron. In an embodiment, at least one dimension of the carbon nanomaterial is between 2 nm and 1000 nm. Carbon nanomaterials include allotropes of carbon such as single walled nanotubes (SWNTs), multiwalled nanotubes (MWNTs), nanorods, single walled and/or multiwalled fullerenes, graphite, graphene, carbon fibers, carbon films, carbon whiskers, and diamond, and all derivatives thereof.

(94) Spatial aligned refers to positions and/or orientations of two or more structures that are defined with respect to each other. Spatially aligned structures may have positions and/or orientations that are preselected with respect to each other, for example, preselected to within 1 micron, preferably for some applications to within 500 nanometers, and more preferably for some applications to within 50 nanometers.

(95) Heterogeneous semiconductor elements are multicomponent structures comprising a semiconductor in combination with one or more other materials or structures. Other materials and structures in the context of this description may comprise elements, molecules and complexes, aggregates and particles thereof, that are different from the semiconductor in which they are combined, such as materials and/or structures having a different chemical compositions and/or physical states (e.g. crystalline, semicrystalline or amorphous states). Useful heterogeneous semiconductor elements in this aspect of the invention include an inorganic semiconductor structure in combination with other semiconductor materials, including doped semiconductors (e.g., N-type and P-type dopants) and carbon nanomaterials or films thereof, dielectric materials and/or structures, and conducting materials and/or structures. Heterogeneous semiconductor elements of the present invention include structures having spatial homogeneous compositions, such as uniformly doped semiconductor structures, and include structures having spatial inhomogeneous compositions, such as semiconductor structures having dopants with concentrations that vary spatially in one, two or three dimensions (i.e. a spatially inhomogeneous dopant distribution in the semiconductor element).

(96) The invention may be further understood by the following non-limiting examples. All references cited herein are hereby incorporated by reference to the extent not inconsistent with the disclosure herewith. Although the description herein contains many specificities, these should not be construed as limiting the scope of the invention but as merely providing illustrations of some of the presently preferred embodiments of the invention. The scope of the invention should, therefore, be determined by the appended claims and their equivalents, rather than by the examples given.

(97) One method for making buckled or wavy interconnects is generally summarized in FIG. 1. A metal feature 10 (such as a metal feature that will be an interconnect) is provided on a substrate 20. The contacting metal feature and/or substrate surfaces are optionally treated for reduced adhesion such as by photolithography or with a shadowmask. A separation (crack) 25 is introduced between the feature 10 and the substrate 20 such as by micromachining, etching and/or mechanical scribing. The metal feature 10 is retrieved with a compliant elastomer stamp 30. Subsequent deformation of the stamp 30 generates in the metal feature 10 a wavy or buckled geometry 40. Generation of the buckles is provided by stamp 30 that is under strain when the metal feature 10 is retrieved and subsequently releasing the applied tension, or by compressing stamp 30 after the metal feature is retrieved.

(98) One example of a buckled or wavy metal feature generated by the method summarized in FIG. 1 is shown in FIG. 2. FIG. 2 is a photograph of a stretchable wavy/buckled electrical interconnect 40, formed by retrieval from a rigid substrate onto a pre-strained, stretchable PDMS rubber substrate 30, followed by release of the strain, thereby inducing buckling.

(99) A method for generating wavy stretchable electrodes and/or interconnects is provided in FIG. 3. As shown in FIG. 3A, wavy features 22 are prepared on a substrate 20, such as by micromachining processes, for example. The substrate 20 with a surface having wavy features 22 serves as a master for molding elastomer stamps 30 with a corresponding wavy surface 32. Metal features 10 are deposited on the wavy surface 32, such as by evaporation through a shadow mask and/or electrodeposition.

(100) FIG. 4 provides one method for fabricating a smooth wavy elastomer substrate. Anisotropic Si (1 0 0) etching provides a substrate 20 having sharp-edges 24 (FIG. 4Btop panel). Spin PR smooths the sharp-edged valleys by depositing PR 26 in the sharp-edged valleys 24 of substrate 20. An elastomeric stamp 34 is cast against substrate 20. Stamp 34 has sharp-edged recess features. A second elastomeric stamp 36 is cast on stamp 34 to generate a stamp having sharp-edged peaks. Stamp 36 is embossed with Su-8 50 and cured as appropriate. Spin PR 26 smooths the sharp-edged valleys of 50. Elastomeric substrate 30 is cast against the 50 having smooth valleys. Substrate 30 is removed to reveal a wavy and smooth surface 32.

(101) FIG. 54 summarizes one method of fabrication of wavy stretchable electrodes by deposition on a wavy master, followed by casting a stamp on that master, curing the stamp, and thereby transferring the electrodes to the master upon release. FIG. 55 shows images of stretchable metal electrodes (Au, 300 nm thick) on wavy PDMS prepared by the methods in FIG. 4 combined with those in FIG. 54. Interface 112 is shown between metal feature 10 and substrate 20. Interface 112 may comprise material that facilitates removal of metal features 10 by stamp 30 illustrated in the bottom panel. Briefly, one method uses: on pre-cleaned 23 glass slides, spin on a thin coating of SU-8 10 so that glass surface is completely covered. Bring slide/SU-8 into contact with PDMS stamp having the desired wavy surface features (smooth valleys and sharp peaks) and gently apply pressure so that all air pockets are removed. Flash cure the stamp/mold structure under a UV lamp for 30 seconds on the front side, flip, and cure for an additional 40 seconds on the reverse side. After cure, bake on hot plate at 65 C. for 5 minutes. After bake, allow sample to cool to room temperature and peel apart SU-8 mold from PDMS master. SU-8 will now have wavy surface relief with sharp edged valleys. To smooth out these valleys, mix one part SU-8 2 with one part SU-8 thinner, and spin on at high RPM for 90 seconds. Expose to UV light for 20 seconds to cure and post bake for 3 minutes at 65 C. Once cool, metal lines or contacts are deposited via electrodeposition, photolithograph and etching/lift-off, and/or evaporation through a shadowmask. Treat the metal on SU-8 with MPTMS for 1 hour and then cast elastomeric substrate against it. When removed, the PDMS has wavy surface relief with smoothed peaks and valleys along with transferred metal structures. FIG. 55 is a photograph of the a wavy stretchable electrode made by the process summarized in FIG. 54, and also provides measured electrical resistance data of the stretchable wavy metal electrodes as a function of applied tensile strain (up to 30%).

(102) An example of a smoothly wavy PDMS substrate 30 made by the method summarized in FIG. 4 is provided in FIG. 5. A device component 60 can be supported to wavy substrate 30 in a non-wavy region (e.g., substantially flat portion) and connected to an interconnect 10 as desired.

(103) An example of spin coating of a smoothing layer into a sharp-edged valley or recess feature is shown in FIG. 6. A sharp-edged substrate 34 (FIG. 6A) is smoothed by spin-coating a photcurable epoxy 26 to generate a smoothly wavy substrate. An elastomer (e.g., PDMS) stamp 30 having a smoothly wavy surface 32 is obtained by casting a PDMS stamp against the substrate of FIG. 6B and subsequently removing the stamp 30 from the substrate 34.

(104) FIG. 7 are photographs of a stretchable electrode. FIG. 7A is a photograph of a cross-section of an elastomer substrate 30 having a wavy surface 32. FIG. 7B is a top view micrograph of an electrode made by evaporating metal 10 on the wavy elastomer substrate surface 32. The image's focal plane is on the peaks of the wavy relief. In FIG. 7C, the focal plane is on the valleys of the wavy relief and the metal interconnect 10 is in electrical contact with the electrode 250. The stretchable electrode is deposited by evaporation through a shadowmask onto a smoothly wavy elastomer substrate. In this example, the electrode 250 maintains conductivity and connectivity via interconnects 10 during stretching up to about 10% in tension.

(105) The methods and devices disclosed herein may be used to fabricate a variety of electronic devices, including for example, a stretchable passive matrix LED display (see FIG. 8). Wavy electrodes (e.g., interconnects 10 and contact pad 70) are patterned on two elastomeric substrates 30. A device component 60 (in this case ILED pixels) are patterned on the wavy electrodes at contact pad 70 by transfer printing. The two substrates 30 are accordingly assembled such that the interconnects 10 run in different orientations (perpendicular, in this example). The 2-D mechanical stretchability of such a passive matrix LED display is illustrated in FIG. 9. In addition to being able to stretch uniaxially and biaxially, the display is capable of substantial bending without breaking. Such multi-axial bending provides the capability of molding electronic devices to curved surfaces to produce curved electronic devices and for incorporation into smart electronic fabrics or displays.

(106) One such example of a curved electronic device is provided in FIG. 10. FIG. 10 illustrates an artificial eye comprising an inorganic photodiode array distributed on a spherically curved lens. Four different views of the artificial array are shown. The requirement for stretchable planar electronic devices is schematically illustrated in FIG. 11. In order to wrap a planar sheet around a spherical surface, the sheet must stretch in more than one direction.

(107) FIG. 12 is a fabrication scheme for making a stretchable buckled semiconductor array capable of conforming to curved surfaces. Thin Si elements are fabricated with selective Au or Ti/Au deposition on a substrate, such as the illustrated mother wafer in panel (i). Si is bonded to prestrained (indicated as L+L) and UVO treated PDMS (panel (ii)). Prestrain is provided in two directions, as illustrated. The bonding is by any means known in the art such as an adhesive, for example, applied to the Si elements, the substrate, or both. The bonding means is applied in a selected pattern so that the Si has bonded regions that will remain in physical contact with the substrate (after deformation) and other regions in a bent configuration that are not in physical contact with the substrate (e.g., regions that are not bonded or are weakly bonded relative to the adhesive force in the bond regions). The prestrained substrate is removed from the wafer substrate to reveal a flat grid of semiconductor arrays (panel (iii)). Upon relaxation of the substrate from L+L to L, the interconnects 10 buckle in the weakly-bonded regions (see panel (iv)) to a bent configuration whereas the device component 60 (e.g., semiconductor Si contact pad) remains bonded to the substrate 30. Accordingly, buckled interconnects 10 impart stretchability to the entire array, and specifically the capability for motion of component 60 relative to other components 60. without breaking electrical contact between components 60, thereby providing conformal capability to a curved surface or a bendable surface.

(108) FIG. 13 provides an optical microscopic image of a buckled stretchable silicon array in a single grid configuration 140 (top two panels), grid configuration having a plurality of connected interconnects 160 (bottom left panel), and a floral configuration 150 (bottom right panel). In each of these examples, interconnect 10 is buckled in a central portion, with interconnect ends attached to a contact pad 70. The interconnects and contact pad 70 are supported on a PDMS substrate 30. Close-up views of a number of different interconnect geometries are further provided in FIGS. 14-17. FIG. 14 provides electron microscopic images to show a basic buckled or wavy interconnect 10 having a central portion 90 with a first end 100 and second end 110. The central portion is in a bent configuration. Ends 100 and 110 are connected to a device component, in this case a contact pad 70 capable of establishing electrical contact with a device component. The interconnect 10 and contact pad 70 are supported on a substrate 30, such as an elastomeric PDMS substrate.

(109) FIG. 15 is an electron microscopic image of adjacent device components (e.g., contact pad 70) connected to each other by a plurality (two) of interconnects 160. Comparing FIG. 15 to FIG. 14 demonstrates that adjacent device components 70 can be connected to one another by one or more interconnects 10 to provide additional flexibility to the electronic device. For example, a device component or contact pad 70 having a relatively large footprint is optionally connected to another device component by multiple interconnects.

(110) FIG. 16 is an electron microscopic image of interconnects in a floral configuration 150. A floral configuration, in contrast to a grid configuration, has interconnects oriented in more than two longitudinal directions. In this example, there are four distinct orientations, so that a device component such as contact pad 70 is capable of contacting diagonally-adjacent device components. In this example, the interconnect 10 has an optional bond region 102 in between interconnect ends 100 and 110 that are electrically connected to a device component (not shown), thereby dividing central portion 90 into two non-bonded regions 92, each having a bent configuration.

(111) FIG. 17 is an electron microscopic image of interconnects arranged in a bridge configuration 130. In a bridge configuration, a bridge central portion peak 120 from which three or more interconnect ends extend therefrom. For example, two interconnects that intersect in a non-bonded region results in a peak 120 having four interconnect ends extending therefrom. For the situation where the device components are in a staggered arrangement, the peak 120 may have three ends extending therefrom. In the case of multiple interconnect connections between device components, more than four ends may extend from peak 120.

(112) Although many of the drawings provided herein show a device component that is a contact pad 70, the methods and devices claimed herein are capable of connecting to a vast population of device components to provide stretchable and therefore shape-conforming, electronic devices. For example, FIG. 18 shows a device component 60 that is a photodiode connected to other photodiodes in an array configuration by buckled interconnects 10 supported on an elastomeric substrate 30.

(113) FIG. 19 depicts one-dimensional stretching behavior of a buckled silicon array. Panel (i) is a picture of a buckled silicon array without any straining force applied. A stretching force is applied (as indicated by the arrows above panel (i)) to stretch the array in one direction. As shown in panels (2)-(4), the buckled interconnect flattens. When the stretching force is released in panel (5), the array reverses to its buckled configuration (see panels (6)-(8)). A comparison between panels (1) and (8) shows that the buckle configuration pre and post-stretch are identical, indicating the process is reversible.

(114) Buckled arrays of device components may be readily transferred to curved surfaces, including rigid or inelastic curved surfaces. An example of one device and process for facilitating conformal contact to curved surfaces is provided by the bubble or balloon stamp 400 of FIG. 20. An elastomeric substrate 30, in this example an about 20 m thick PDMS membrane is fixed in a housing chamber 300 to provide a chamber volume 310 defined by the interior-facing substrate wall and housing chamber. Applying a positive pressure (e.g., pressure in chamber 300 greater than exterior pressure) generates a convex 200 substrate surface capable of conformal contact with a concave-shaped receiving substrate. A negative pressure, in contrast, generates a concave surface 210 capable of conformal contact with a convex-shaped receiving substrate. Spatial manipulation of local elasticity (e.g., Young's modulus) of the substrate permits generation of complex curved geometry. The bottom left panel of FIG. 20 illustrates one means for controlling the pressure in the housing volume 310 by a syringe that introduces or removes gas to and from chamber 310. The images on the right side of the figure are different curvatures of a PDMS membrane in response to increasing level of positive pressure. Any of the methods and devices for providing buckled interconnects on an elastomeric substrate may be used with such devices for transfer printing to a curved substrate.

(115) Another means for generating buckled or pop-up interconnects on a curved surface is summarized in FIG. 21. A thin elastomeric film is cast against a shaped surface to generate an elastomeric substrate having at least a portion that is curved. The substrate is capable of being stretched to flatten the surface so that the substrate is capable of conforming to both curved and flat surfaces. An interconnect is applied to the flat stamp, and upon release of the stretching force, the substrate surface relaxes back to a curved geometry, generating a strain in the interconnect that is accommodated by a pop-up of the interconnect central portion.

(116) An example of two-dimensional stretching of a buckled silicon array by the device shown in FIG. 20, is provided in FIG. 22. In this example, the interconnect comprises a plurality of buckled interconnect connections in a grid configuration, with the interconnects made of 290 nm thick Si. The initially flat buckled silicon array (top left image) is placed into the housing, and a positive pressure exerted to expand the array into a bubble or balloon configuration (e.g., a curved surface). Maximum expansion is shown in the right-most image, and subsequently the positive pressure removed. Similar to the results for uniaxial stretching of a flat substrate, this bending stretching is reversible. At any stage of expansion, that maximizes conformal contact with a curved surface, the array may be transferred to the curved surface by any means known in the art. An example of silicon printing by balloon stamps onto glass lenses coated with adhesives (elastomeric substrate or SU-8) is shown in FIG. 23. The lens may be either concave or convex. In this example R=19.62 mm and 9.33 mm, respectively.

(117) Example 1

(118) Controlled Buckling Structures in Semiconductor Nanoribbons with Application Examples in Stretchable Electronics

(119) Control over the compositions, shapes, spatial locations and/or geometrical configurations of semiconductor nanostructures is important for nearly all applications of these materials. Although methods exist for defining the material compositions, diameters, lengths, and positions of nanowires and nanoribbons, there are relatively few approaches for controlling their two- and three-dimensional (2D and 3D) configurations. Provided herein is a mechanical strategy for creating certain classes of 3D shapes in nanoribbons that are otherwise difficult to generate. This example involves the combined use of lithographically patterned surface chemistry to provide spatial control over adhesion sites and elastic deformations of a supporting substrate to induce well-controlled local displacements. Precisely engineered buckling geometries are created in nanoribbons of GaAs and Si in this manner and these configurations can be described quantitatively with analytical models of the mechanics. As one applications example, particular structures provide a route to electronics (and optoelectronics) with extremely high levels of stretchability (up to 100%), compressibility (up to 25%) and bendability (with curvature radius down to 5 mm).

(120) The 2D and 3D configurations of nanoribbons and wires are controlled during their growth to yield certain geometries, such as coils, rings, and branched layouts, or after their growth to produce, as examples, sinusoidal wave-like structures by coupling these elements to strained elastomeric supports or tube-like (or helical) structures by using built-in residual stresses in layered systems. Semiconductor nanoribbons with wavy geometries are of interest in part because they enable high performance, stretchable electronic systems for potential applications such as spherically curved focal plane arrays, intelligent rubber surgical gloves and conformable structural health monitors. This approach, in which the electronic devices themselves are stretchable, is different than and perhaps complementary to an alternative route to these same applications that use rigid device islands with stretchable metal interconnects. The previously described wavy nanoribbons have two main disadvantages: (i) they form spontaneously, with fixed periods and amplitudes defined by the moduli of the materials and the thicknesses of the ribbons, in a way that offers little control over the geometries or the phases of the waves and (ii) the maximum strains that they can accommodate are in the range of 20-30%, limited by the non-optimal wavy geometries that result from this process. The procedures introduced here use lithographically defined surface adhesion sites together with elastic deformations of a supporting substrate to achieve buckling configurations with deterministic control over their geometries. Periodic or aperiodic designs are possible, for any selected set of individual nanoribbons in large scale, organized arrays of such structures. Specialized geometries designed for stretchable electronics enable strain ranges of up to nearly 150%, even in brittle materials such as GaAs, consistent with analytical modeling of the mechanics, and as much as ten times larger than previously reported results.

(121) FIG. 24 shows the steps in this procedure. The fabrication starts with the preparation of a mask for patterning surface chemical adhesion sites on an elastomeric substrate of poly(dimethylsiloxane) (PDMS). This process involves passing deep ultraviolet (UV) light (240-260 nm) through an unusual type of amplitude photomask (fabricated via step i), referred to as a UVO mask, while it is in conformal contact with the PDMS. The UVO mask possesses recessed features of relief in the transparent regions, such that exposure to UV creates patterned areas of ozone in proximity to the surface of the PDMS. The ozone converts the unmodified hydrophobic surface, dominated by CH.sub.3 and H terminal groups, to a highly polar and reactive surface (i.e. activated surface), terminated with OH and OSiO functionalities. The unexposed areas retain the unmodified surface chemistry (i.e. inactivated surface). The procedures introduced here involve exposures on PDMS substrates (thickness 4 mm) under large, uniaxial prestrains (.sub.pre=L/L for lengths changed from L to L+L) (step ii). For masks with simple, periodic line patterns, we denote the widths of the activated (indicated as lines labeled activated surface) in step (iii) of FIG. 24A and inactivated stripes (e.g., the distance between adjacent activated stripes) as W.sub.act and W.sub.in in step (i). The activated areas can bond strongly and irreversibly to other materials that have exposed OH or SiO groups on their surfaces. These patterned adhesion sites are exploited to create well defined 3D geometries in nanoribbons, as outlined below. Alternatively, a similar adhesive bond site pattern is provided by similarly patterning the interconnects prior to contact with the substrate.

(122) In this example, nanoribbons consisted of both single crystal Si and GaAs. The silicon ribbons are prepared from silicon-on-insulator (SOI) wafers using procedures described previously (see Khang et al. Science 311, 208-212 (2006)). The GaAs ribbons involved multilayers of Si-doped n-type GaAs (120 nm; carrier concentration of 410.sup.17 cm.sup.3), semi-insulating GaAs (SI-GaAs; 150 nm) and AlAs (200 nm) formed on a (100) SI-GaAs wafer by molecular-beam epitaxy (MBE). Chemically etching the epilayers in an aqueous etchant of H.sub.3PO.sub.4 and H.sub.2O.sub.2, using lines of photoresist patterned along the (0 1 1) crystalline orientation as etch masks, define the ribbons. Removing the photoresist and then soaking the wafer in an ethanol solution of HF (2:1 in volume between ethanol and 49% aqueous HF) removes the AlAs layer, thereby releasing ribbons of GaAs (n-GaAs/SI-GaAs) with widths determined by the photoresist (100 m for the examples in FIG. 24D). The addition of ethanol to the HF solution reduces the probability of cracking of the fragile ribbons due to the action of capillary forces during drying. The low surface tension (compared to water) also minimizes drying-induced disorder in the spatial layout of the GaAs ribbons. In the final step, a thin layer of SiO.sub.2 (30 nm) is deposited to provide the necessary SiOH surface chemistry for bonding to the activated regions of the PDMS.

(123) Laminating the processed SOI or GaAs wafers against a UVO treated, pre-stretched PDMS substrate (ribbons oriented parallel to the direction of prestrain), baking in an oven at 90 C. for minutes, and removing the wafer transferred all of the ribbons to the surface of the PDMS (step iv). Heating facilitates conformal contact and the formation of strong siloxane bonds (i.e., OSiO) between the native SiO.sub.2 layer on the Si ribbons or the deposited SiO.sub.2 layer on the GaAs ribbons and the activated areas of the PDMS. Relatively weak van der Waals forces bond the ribbons to the inactivated surface regions of the PDMS. Relaxing the strain in the PDMS generates buckles through the physical separation of the ribbons from the inactivated regions of the PDMS (step v). The ribbons remain tethered to the PDMS in the activated regions due to the strong chemical bonding. The resulting 3D ribbon geometries (i.e. the spatially varying pattern of buckles) depend on the magnitude of prestrain and the patterns of surface activation (e.g., shape and dimensions of W.sub.in and W.sub.act). (Similar results can be achieved through patterned bonding sites on the ribbons). For the case of the simple line pattern, W.sub.in and the prestrain determine the width and amplitude of the buckles. Sinusoidal waves with wavelengths and amplitudes much smaller than the buckles also formed in the same ribbons when W.sub.act was >100 m, due to mechanical instabilities of the type that generate wavy silicon. (see FIG. 25, images of samples formed with different W.sub.act). As a final step in the fabrication, the 3D ribbon structures can be encapsulated in PDMS by casting and curing a liquid prepolymer (see FIG. 24 step vi). Due to its low viscosity and low surface energy, the liquid flows and fills the gaps formed between the ribbons and the substrate (see FIG. 26).

(124) FIG. 24D shows a tilted-view scanning electron microscope (SEM) image of buckled GaAs ribbons on PDMS, in which .sub.pre=60% and with W.sub.act=10 m and W.sub.in=400 m. The image reveals uniform, periodic buckles with common geometries and spatially coherent phases for all ribbons in the array. The anchoring points are well registered to the lithographically defined adhesion sites. The inset shows an SEM image of a bonded region; the width is 10 m, consistent with W.sub.act. The images also reveal that the surface of the PDMS is flat, even at the bonding sites. This behavior, which is much different than the strongly coupled wavy structures reported previously, suggests that the PDMS induces the displacements but is not intimately involved in the buckling process (i.e. its modulus does not affect the geometries of the ribbons), for the cases described here. In this sense, the PDMS represents a soft, nondestructive tool for manipulating the ribbons through forces applied at the adhesion sites.

(125) FIG. 27A shows side-view optical micrographs of buckled ribbons formed on PDMS with different .sub.pre (W.sub.act=10 m and W.sub.in=190 m). The heights of the buckles (e.g., amplitude) increase with .sub.pre. The ribbons in the inactivated regions do not fully separate at low .sub.pre (see the samples formed with .sub.pre=11.3% and 25.5%). At higher .sub.pre, the ribbons (thickness h) separate from the PDMS to form buckles with vertical displacement profiles characterized by:

(126) $y=\frac{1}{2}{A}_1^0\left[1+\cos \left(\frac{}{L_1}x\right)\right]$
Where:

(127) $A_1^0=\frac{4}{}\sqrt{L_1{L}_2\left({\mathrm{.Math.}}_{\mathrm{pre}}-\frac{h^2{}^2}{12L_1^2}\right)}$$L_1=\frac{W_{\mathrm{in}}}{2*\left(1+{\mathrm{.Math.}}_{\mathrm{pre}}\right)}$$L_2=L_1+\frac{W_{\mathrm{act}}}{2}$

(128) as determined by nonlinear analysis of buckles formed in a uniform, thin layer. The maximum tensile strain in the ribbons is, approximately,

(129) ${\mathrm{.Math.}}_{\mathrm{peak}}={|}_{\max }\frac{h}{2}=y^n{|}_{\max }\frac{h}{2}=\frac{h}{4}{A_1^0\left(\frac{}{L_1}\right)}^2$

(130) The width of the buckles is 2L.sub.1 and the periodicity is 2L.sub.2. Because h.sup.2.sup.2/(12L.sub.1.sup.2) is much smaller than .sub.pre (i.e., >10% in the report) for h<1 m, the amplitude is independent of the mechanical properties of ribbons (e.g., thickness, chemical composition, Young's modulus, etc.) and is mainly determined by the layout of adhesion sites and the prestrain. This conclusion suggests a general applicability of this approach: ribbons made of any material will form into similar buckled geometries. This prediction is consistent with the results obtained with Si and GaAs ribbons used here. The calculated profiles, plotted as dotted lines in FIG. 27A for pre-strains of 33.7% and 56.0%, agree well with the observations in GaAs ribbons. Further, the parameters (including periodicity, width, and amplitude) of the buckles shown in FIG. 27A are consistent with analytical calculations, except at low .sub.pre (Tables 1 and 2). An interesting result of this study is that the maximum tensile strains in the ribbons are small (e.g. 1.2%), even for large .sub.pre (e.g. 56.0%). This scaling enables stretchability, even with brittle materials such as GaAs, as discussed subsequently.

(131) The lithographically defined adhesion sites can have more complex geometries than the simple grating or grid patterns associated with the structures in FIG. 24. For example, buckles with different widths and amplitudes can be formed in individual ribbons. FIG. 27B shows, as an example, an SEM image of buckled Si ribbon (widths and thicknesses of 50 m and 290 nm, respectively), formed with a prestrain of 50% and adhesion sites characterized by W.sub.act=15 m and W.sub.in=350, 300, 250, 250, 300, and 350 m along the lengths of the ribbons. The image clearly shows the variation of widths and amplitudes of adjacent buckles in each of the ribbons. Buckled ribbons can also be formed with different phases for different ribbons. FIG. 27C presents an example of a Si system designed with phases in the buckles that vary linearly with distance perpendicular to the lengths of the ribbons. The UVO mask used for this sample has W.sub.act and W.sub.in of 15 and 250 m, respectively. The angle between the activated stripes on PDMS stamp and Si ribbons is 30. Many other possibilities can easily be achieved, due to the simple lithographic control of the adhesion sites, and some are shown in FIGS. 13-17, for example.

(132) The simple case of buckled GaAs ribbons on PDMS with .sub.pre=60%, W.sub.act=10 m and different W.sub.in, as shown in FIG. 27D, illustrates an aspect that is important for applications in stretchable electronics. The profiles, which agree well with analytical solutions to the mechanics, show failure due to cracking in the GaAs when W.sub.in=100 m (and smaller). The failure results from tensile strains (2.5% in this case) that exceed the yield point of the GaAs (2%). An optimized configuration for robustness to stretching and compressing can, therefore, be achieved by selecting W.sub.in (>>W.sub.act) proportional to .sub.pre. In this situation, prestrains up to and greater than 100% can be accommodated. We demonstrated this type of stretchability directly by applying forces to the PDMS supports. Changes in the end-to-end distances (L.sub.projected) of segments of the ribbons provided a means to quantify the stretchability and compressibility, according to:

(133) $\left|\frac{L_{\mathrm{projected}}^{\max }-L_{\mathrm{projected}}^0}{L_{\mathrm{projected}}^0}\right|*100\%,$

(134) Where L.sub.projected.sup.max represents the maximum/minimum length before fracture and L.sub.projected.sup.0 is the length in the relaxed state. Stretching and compressing correspond to L.sub.projected.sup.max greater and less than L.sub.projected.sup.0, respectively. Buckled ribbons on PDMS with W.sub.act=10 m and Win=400 m and .sub.pre=60%, exhibit stretchability of 60% (i.e., .sub.pre and compressibility up to 30%. Embedding the ribbons in PDMS mechanically protects the structures, and also produces a continuous, reversible response, but with slight changes in the mechanics. In particular, the stretchability and compressibility decreased to 51.4% (FIG. 28A) and 18.7% (FIG. 28B), respectively. The PDMS matrix on top of the ribbons flattens the peaks of the buckles slightly, due partly to curing induced shrinkage of the overlying PDMS. Small period waves form in these regions under large compressive strains, due to the spontaneous mechanics of the type that generated the wavy ribbon structures described previously. Mechanical failure tended to initiate in these areas, as illustrated in FIG. 28B, thereby reducing the compressibility. Buckled structures with W.sub.act=10 m and W.sub.in=300 m avoided this type of behavior. Although such samples exhibited slightly lower stretchability than the one shown in FIG. 28A, the absence of the short period waves increased the compressibility to 26%. Overall, single-crystalline GaAs nanoribbons with buckles formed on the pre-stretched PDMS substrates with patterned surface chemical adhesion sites exhibit stretchability higher than 50% and compressibility larger than 25%, corresponding to full scale strain ranges approaching 100%. These numbers are further improved by increasing .sub.pre and W.sub.in and by using a substrate material capable of higher elongation than PDMS. For even more elaborate systems, these fabrication procedures can be repeated to generate samples with multiple layers of buckled ribbons (see FIG. 29).

(135) A direct consequence of this large stretchability/compressibility is extreme levels of mechanical bendability. FIG. 30A-C present optical micrographs of bent configurations that illustrate this feature. The PDMS substrate (thickness 4 mm) is bent into concave (radius of 5.7 mm), flat, and convex (radius of 6.1 mm) curvatures, respectively. The images illustrate how the profiles changed to accommodate the bending induced surface strains (20-25% for these cases). The shapes are similar, in fact, to those obtained in compression (by 20%) and tension (by 20%). The embedded systems exhibit even higher levels of bendability due to neutral mechanical plane effects. When the top and bottom layers of PDMS had similar thicknesses, there was no change in the buckling shapes during bending (FIG. 30D).

(136) To demonstrate these mechanical properties in functional electronic devices, we build metal semiconductor-metal photodetectors (MSM PDs) using buckled GaAs ribbons with profiles similar to those shown in FIG. 30, by depositing thin gold electrodes onto the SI-GaAs sides of the ribbons for Schottky contacts. FIG. 31A shows the geometry and equivalent circuit, and top-view optical micrographs of an MSM PD before and after stretching by 50%. In the absence of light, little current flowed through the PD; the current increased with increasing illumination with an infrared beam (wavelength 850 nm) (FIG. 31B). The asymmetry in the current/voltage (I-V) characteristics can be attributed to differences in the electrical properties of the contacts. FIG. 31C (stretching) and FIG. 31D (compressing) show I-Vs measured at different degrees of stretching and compressing. The current increased when the PD was stretched by up to 44.4% and then decreased with further stretching. Because the intensity per unit area of the light source is constant, the increases in current with stretching can be attributed to increases in the projected area (referred to as effective area, S.sub.eff) of the buckled GaAs ribbon as it flattens. Further stretching the PD might induce the formation of defects on the surface and/or in the lattice of the GaAs ribbon, resulting in the decrease of current and eventually, at fracture, an open circuit. Similarly, compression decreased S.sub.eff and thus decreased the current (FIG. 31D). These results indicate that buckled GaAs ribbons embedded in PDMS matrix provide a fully stretchable/compressible type of photosensor useful for various applications, such as wearable monitors, curved imaging arrays and other devices.

(137) In conclusion, this example indicates that soft elastomers with lithographically defined adhesion sites are useful as tools for creating certain classes of 3D configurations in semiconductor nanoribbons. Stretchable electronics provide one example of the many possible application areas for these types of structures. Simple PD devices demonstrate some capabilities. The high level of control over the structures and the ability to separate high temperature processing steps (e.g. formation of ohmic contacts) from the buckling process and the PDMS indicate that more complex devices (e.g. transistors, and small circuit sheets) are possible. The well controlled phases of buckles in adjacent ribbons provide an opportunity for electrically interconnecting multiple elements. Also, although the experiments reported here used GaAs and Si nanoribbons, other materials (e.g. GaN, InP, and other semiconductors) and other structures (e.g. nanowires, nanomembranes) are compatible with this approach.

(138) FABRICATION OF GaAs RIBBONS: GaAs wafers with customer-designed epitaxial layers (details described in the text) were purchased from IQE Inc., Bethlehem, Pa. Photolithography and wet chemical etching generated the GaAs ribbons. AZ photoresist (e.g., AZ 5214) was spin cast on the GaAs wafers at speed of 5000 rpm for 30 seconds and then soft baked at 100 C. for 1 minute. Exposure through a photomask with patterned lines oriented along the (0 1 1) crystallographic direction of GaAs, followed by development generated line patterns in the photoresist. Mild O.sub.2 plasma (i.e., descum process) removed the residual photoresist. The GaAs wafers were then anisotropically etched for 1 minute in the etchant (4 mL H.sub.3PO.sub.4 (85 wt %), 52 mL H.sub.2O.sub.2 (30 wt %), and 48 mL deionized water), cooled in the ice-water bath. The AlAs layers were dissolved with an HF solution (Fisher Chemicals) diluted in ethanol (1:2 in volume). The samples with released ribbons on mother wafers were dried in a fume hood. The dried samples were coated with 30 nm SiO.sub.2 deposited by electron beam evaporation.

(139) FABRICATION OF Si RIBBONS: The silicon ribbons are fabricated from an silicon-on-insulator (SOI) wafer (Soitect, Inc., top silicon 290 nm, buried oxide 400 nm, p-type). The wafer is patterned by conventional photolithography using AZ 5214 photoresist and etched with SF6 plasma (PlasmaTherm RIE, SF6 40 sccm, 50 mTorr, 100 W). After the photoresist is washed away with acetone, the buried oxide layer is then etched in HF (49%).

(140) FABRICATION OF UVO MASKS: Fused quartz slides are cleaned in piranha solution (at 60 C.) for 15 minutes and thoroughly rinsed with plenty of water. The cleaned slides are dried with nitrogen blowing and placed in the chamber of electron-beam evaporator to be coated with sequential layers of 5-nm Ti (as adhesive layer) and 100-nm Au (mask layer for UV light). Negative photoresist, i.e., SU8 5, is spin cast on the slides at speed of 3000 rpm for 30 seconds to yield 5 m thick films. Soft baking, exposing to UV light, post baking, and developing generated patterns in the photoresist. Mild O.sub.2 plasma (i.e., descum process) removes the residual photoresist. The photoresist serves as mask to etch Au and Ti using gold etchant (i.e., aqueous solution of I2 and KI) and titanium etchant (i.e., diluted solution of HCl), respectively.

(141) PREPARATION OF PDMS STAMPS: PDMS substrates with thickness of 4 mm were prepared by pouring the prepolymer (A:B=1:10, Sylgard 184, Dow Corning) into a Petri dish, followed by baking at 65 C. for 4 hours. Slabs with suitable sizes and rectangular shapes were cut from the resulting cured piece and the rinsed with isopropyl alcohol and dried with nitrogen blowing. A specially designed stage was used to mechanically stretch the PDMS to desired levels of strain. Illuminating these stretched substrates to short-wavelength UV light (low-pressure mercury lamp, BHK, 173 W/cm2 from 240 to 260 nm) for 5 min through a UVO mask placed in contact with the PDMS generated the patterned surface chemistries.

(142) FORMATION AND EMBEDMENT OF BUCKLED GaAs RIBBONS: GaAs wafers with released ribbons coated with SiO.sub.2 were laminated against the stretched PDMS with patterned surface chemistry. Baking in an oven at 90 C. for 5 minutes, cooling to room temperature in air, and then slowly relaxing the strain in the PDMS generated buckles along each ribbon. Embedding the buckled ribbons, involved flood exposing to UV light for 5 minutes and then casting of liquid PDMS prepolymers to a thickness of 4 mm. Curing the sample either in an oven at 65 C. for 4 hours or at room temperature for 36 hours cured the prepolymer, to leave the buckled ribbons embedded in a solid matrix of PDMS.

(143) CHARACTERIZATION OF BUCKLED RIBBONS: The ribbons were imaged with an optical microscope by tilting the sample by 90 (for nonembedded samples) or 30 (for embedded samples). The SEM images were recorded on a Philips XL30 field-emission scanning electron microscope after the sample was coated with a thin layer of gold (5 nm in thickness). The same stage used for pre-stretching the PDMS stamps was used to stretch and compress the resulting samples.

(144) FABRICATION AND CHARACTERIZATION OF SMS PDs: Fabrication of PDs started with samples in the configuration shown in the bottom frame of FIG. 24B. A 0.8 mm wide strip of poly(ethylene terephthalate) (PET) sheet was gently placed on the PDMS with its longitudinal axis perpendicular to the longitudinal axes of the ribbons. This strip served as a shadow mask for electron beam evaporation of a 30-nm thick gold film (to form Schottky electrodes). Removing the PET stripe, and relaxing the pre-stretched PDMS stamp formed SMS PDs built with buckled GaAs ribbons. Liquid PDMS prepolymer was cast onto the regions of the ribbons without electrodes and then cured in an oven. The gold electrodes extended beyond the top PDMS to enable probing with a semiconductor parameter analyzer. (Agilent 4155C). In measurements of the photoresponse, the PDs were manipulated using a mechanical stage for stretching and compressing. An IR LED source (with wavelength of 850 nm) provided the illumination.

(145) Example 2

(146) Transfer Printing

(147) Our technical approach uses certain ideas embodied in the planar stamp based printing methods previously described. Although these basic techniques provide a promising starting point, many fundamentally new features must be introduced to meet the challenges of the HARDI (Hemispherical Array Detector for Imaging) system, as described in the following.

(148) FIGS. 32 and 33 illustrate a general strategy related to transfer printing to curved surfaces. The first set of steps (FIG. 32) involves the fabrication and manipulation of a thin, spherically curved elastomeric stamp designed to lift off interconnected Si CMOS chiplets from the planar surface of a wafer and then to transform the geometry into a hemispherical shape. A stamp for this process is formed by casting and curing a liquid prepolymer to obtain an elastomer such as poly(dimethylsiloxane) (PMDS) against high quality optical elements (i.e. a matched pair of convex and concave lenses) selected with the required radii of curvature. The stamp has a molded circular rim. Radially stretching this element by mating the molded groove (dashed circle in FIG. 32) on this rim onto an appropriately sized rigid, circular retaining ring transforms this spherical stamp into a stretched, planar sheet. Contacting this stretched stamp to a mother wafer that supports preformed and undercut etched Si CMOS chiplets with thin interconnects and then peeling the stamp away inks this element with these interconnected chiplets. Van der Waals interactions between the chiplets and the soft, elastomeric element provides sufficient adhesion for this process.

(149) Removing the retaining ring causes the PDMS to relax back to its initial hemispherical shape, thereby accomplishing a planar to spherical transformation of the chiplet array. This transformation induces compressive strains at the surface of the stamp. These strains are accommodated in the CMOS chiplet array by local delamination and lifting up of the interconnects (bottom left of FIG. 32). These pop-up interconnects absorb the strains in a manner that avoids damage to the chiplets or detrimental strain-induced changes in their electrical properties. Maintaining the strains in the chiplets below 0.1% accomplishes these two goals. The space required for the interconnects limits the maximum fill factor of the CMOS chiplets. The photodectectors, however, consumes nearly the full pixel areas, thereby providing a straightforward path to an 80% fill factor target.

(150) In the second set of steps (FIG. 33), the inked hemispherical stamp is used to transfer print these elements onto the final device substrate having a matching shaped cavity (e.g., in this example a glass substrate with a matching hemispherically-shaped cavity). This transfer process uses an ultraviolet (UV) curable photopolymer, such as photocurable BCB (Dow Chemical) or polyurethane (Norland Optical Adhesive) as an adhesive. These materials are applied to the device substrate in the form of a thin (tens of microns thick) liquid film. Upon contact with the stamp, this liquid layer flows to conform to the relief structures associated with the chiplets and the pop-up interconnects. UV light passed through the transparent substrate cures the photopolymer and transforms it into a solid form to yield a smooth, planarized top surface upon removal of the stamp. Final integration to form a functional system involves the deposition and patterning of electrodes and photodetector materials, and lithographic definition of the bus lines to external control circuits.

(151) The approach of FIGS. 32 and 33 has several notable features. First, it exploits state-of-the-art planar electronics technology to enable reliable, cost-effective and high performance operation on the hemispherical substrates. In particular, the chiplets consist of collections of silicon transistors processed at the 0.13 m design rule to yield the local, pixel-level processing capabilities for the HARDI system. Conventional processing is used with silicon-on-insulator wafers to form these devices. The buried oxide provides the sacrificial layer (undercut etching with HF) to prepare the chiplets for printing. The interconnects consist of narrow and thin (100 nm) metal lines.

(152) A second feature is that the approach uses elastomeric elements and mechanical designs to enable a well-controlled planar to hemispherical transformation. Reversible, linear mechanics in the transfer stamps and comprehensive mechanical modeling accomplishes this control, as outlined subsequently. A third attractive aspect is that certain basic components of the transfer processes and strategies to control adhesion have been demonstrated in planar applications. In fact, the stages that have been engineered for those planar printing applications can be adapted for the process of FIGS. 32 and 33. FIG. 34 shows a home-built printer with integrated vision systems and air pressure actuators suitable for use in this process.

(153) These types of printer systems are used to demonstrate several aspects of the process of FIGS. 32 and 33. FIG. 35 shows scanning electron micrograph images of the surface of a hemispherical stamp that is inked with an array of single crystal silicon islands interconnected in a square array with heavily doped silicon ribbons. FIG. 36 shows optical images. During the planar to spherical transformation, these ribbon interconnects pop up in the manner depicted in FIG. 32. A key aspect of these types of interconnects is that, when combined with the transfer of fully formed chiplets, they reduce the need for high resolution, curved surface lithography or other forms of processing directly on the hemisphere.

(154) In addition to materials and general processing strategies, full computational modeling of the elastic mechanical response of the hemispherical stamps, the pop-up interconnects and the interactions with the rigid device islands is performed. These calculations reveal the physics of the process at a level that facilitates engineering control and optimization. Simple estimates based on linear elastic plate theory suggest that the strain levels associated with the processes of FIG. 32 can reach 10% or higher for a 2 mm-thick stamp and a sphere with 1 cm radius. For reliable engineering control, it is necessary, therefore, that the stamp is operated in the linear elastic regime for strains up to twice this valuei.e. 20%. FIG. 37 shows experimental stress/strain curves of several variants of PDMS with which we have experience at the level of bulk, planar stamp based printing. The 184-PDMS appears to provide a good initial material because it provides highly linear and elastic response up to strains of 40%.

(155) Mechanical measurements such as these, coupled with literature values for the moduli and geometries of the chiplets and the ribbon pop-up interconnects, provide information necessary for the modeling. Two approaches to the calculations are adopted. The first is full-scale finite element modeling (FEM), in which the details of device and interconnect geometry (e.g., size, spacing, multi-layers) on the planar substrate are analyzed. Different materials (e.g., stamp, silicon, interconnects) are accounted for directly in the analysis. A lateral pressure is imposed to deform the stamp and circuits onto the desired spherical shape. The finite element analysis gives the strain distribution, particularly the maximum strain in devices and interconnects, and non-uniform spacing between transformed devices. The advantage of such an approach is that it captures all the details of device geometry and materials, and therefore can be used to explore the effects of different designs of transfer printing process in order to reduce the maximum strain and non-uniformity. This approach, however, is computationally intensive and therefore, time consuming since it involves a wide range of length scales, and the modeling of large numbers of structures devices on the stamp.

(156) The second approach is a unit-cell model for devices (chiplets) that analyzes their mechanical performance upon loading. Each device is represented by a unit cell, and its response upon mechanical loading (e.g., bending and tension) is studied thoroughly via the finite element method. Each device is then replaced by the unit cell linked by interconnects. This unit-cell model is then incorporated into the finite element analysis to replace the detailed modeling of devices and interconnects. Furthermore, away from the edge of the sphere, the strains are relatively uniform such that the many unit-cells can be integrated and their performance can be represented by a coarse-level model. Near the sphere edge the strains are highly non-uniform such that the detailed modeling of devices are still necessary. The advantage of such an approach is that it significantly reduces the computational effort. The full-scale finite element analysis in the first approach is used to validate this unit-cell model. Once validated, the unit-cell model provides a powerful design tool since it is suitable for the quick exploration of different designs of devices, interconnects, and their spacing.

(157) FIG. 38 presents preliminary FEM results for stretching a hemispherical stamp into a planar geometry (and relaxing it back to its hemispherical shape), as outlined in FIG. 32. The top frame shows a cross sectional view of a hemispherical stamp with a geometry like the one schematically illustrated in FIG. 32. These results show slight spatial non-uniformities in the strains of the stretched membrane, as evidenced by its non-uniform thickness. Engineering the thickness profiles of the stamps, through proper selection of the structures against which they are formed by casting and curing, can eliminate these non-uniformities. It is important to note, however, that some non-uniform strains are acceptable because (i) the pop-up interconnects are inherently distortion tolerant, and (ii) the chiplets do not have to be centered perfectly at each pixel location; the larger photodetectors will fill the pixel areas with a uniform back electrode that can establish electrical contact to the chiplets independent of their positions within the pixel areas.

(158) The modeling can also determine the levels of strain in the Si CMOS chiplets. The systems should be designed to keep these chiplet strains below 0.1-0.2% to avoid changes in the electrical properties and, possibly, mechanical failures due to fracture or delamination. This modeling facilitates the design of stamps and processing conditions to avoid exposure of the chiplets to strains above this range.

(159) Example 3

(160) Biaxially Stretchable Wavy Silicon Nanomembranes

(161) This example introduces a biaxially stretchable form of single crystalline silicon that consists of two dimensionally buckled, or wavy, silicon nanomembranes on elastomeric supports. Fabrication procedures for these structures are described, and various aspects of their geometries and responses to uniaxial and biaxial strains along various directions are presented. Analytical models of the mechanics of these systems provide a framework for quantitatively understanding their behavior. These classes of materials provides a route to high-performance electronics with full, two-dimensional stretchability.

(162) Electronics that offer mechanically bendability are of interest for applications in information display, X-ray imaging, photovoltaic devices, and other systems. Reversible stretchability is a different and much more technically challenging mechanical characteristic that would enable device possibilities that cannot be realized with electronics that are only bendable, such as smart surgical gloves, electronic eye cameras, and personal health monitors. In one approach to electronics of this type, stretchable wires interconnect rigid device islands to provide circuit level stretchability with device components that are not stretchable. In an alternative strategy, certain structural forms of thin single-crystal semiconductor and other electronic materials allow stretchability in the devices themselves. Recent demonstrations involved the use of buckled, one-dimensional wavy geometries in nanoribbons (thicknesses between tens and hundreds of nanometers and widths in the micrometer range) of silicon and gallium arsenide to achieve uniaxial stretchability in metal oxide semiconductor field effect transistors (MOSFETs), metal semiconductor field effect transistors (MESFETs), pn junction diodes, and Schottky diodes. This example shows that nanomembranes of similar materials can be formed into two-dimensional (2D) wavy geometries to provide full 2D stretchability. The fabrication procedures for such systems are described, together with detailed experimental characterization and analytical modeling of their mechanical response.

(163) FIG. 39 schematically illustrates the steps for forming two-dimensionally stretchable Si nanomembranes on elastomeric supports. For this example, these membranes are fabricated from silicon-on-insulator (SOI) wafers (Soitec, Inc., p-type) starting with the formation of a square array of holes in the top silicon (2.5 m diameter, and 25 m pitch), by defining suitable patterns of photoresist by photolithography and then removing the exposed silicon by reactive ion etching (PlasmaTherm RIE, SF.sub.6 40 sccm, 50 mTorr, 100 W). This same step defines the overall lateral dimensions of the membranes which, for the samples reported here, are in the range of 3-5 mm square. The thicknesses are between 55 and 320 nm. Immersing the etched samples in concentrated hydrofluoric acid (HF 49%) removes the buried SiO.sub.2 layer (1451000 nm thick); washing in acetone removed the photoresist. Casting and curing prepolymers of poly(dimethylsiloxane) (PDMS) against polished silicon wafers generated flat, elastomeric substrates (4 mm thick). Exposure to an ozone environment created by intense ultraviolet light (240-260 nm) for 5 min converted the hydrophobic PDMS surface (CH.sub.3 and H end groups) to a hydrophilic state (OH and OSiO end groups). Heating such an activated PDMS substrate briefly at 70-180 C. in a convection oven induced a controlled degree of isotropic thermal expansion. Contacting this element to the processed SOI wafer and then peeling it off again transferred the entire nanomembrane to the PDMS. Continued heating in the convection oven for a few minutes facilitated the formation of strong adhesive bonds between the membrane and the PDMS. In the final step, the nanomembrane/PDMS structure i cooled to room temperature (ca. 25 C.) to release the thermally induced prestrain (L/L). This process led to the spontaneous formation of two-dimensional (2D) wavy structures of relief in the Si nanomembrane and the near surface region of the PDMS. These structures exhibit different behaviors near the edges, where one-dimensional periodic waves predominate, at the inner regions, where two-dimensional herringbone layouts are typically observed, and near the centers, where disordered herringbone structures often occur. The herringbone region is characterized by the distance between adjacent peaks in the waves, which we refer to as the short wavelength , the amplitude of wave A.sub.1 (not shown in FIG. 1), and a longer distance 2/k.sub.2 (along the x2 direction), associated with the separation between adjacent jogs in the herringbone structure, which we refer to as the long wavelength. Other characteristic length are the jogs wavelength 2/k.sub.1 (along the x.sub.1 direction, normal to the long wavelength direction x.sub.2), the amplitude A.sub.2 of the jogs, the jog angle . The bottom frames of FIG. 39 illustrate these features schematically.

(164) Parts a-f of FIG. 40 show optical micrographs collected at different stages during the formation of herringbone waves, for the case of a nanomembrane with 100 nm thickness (lateral dimension of ca. 44 mm.sup.2) and thermal prestrain (defined by heating to 150 C.) of 3.8%. These images indicate structure formation in two stages, the first of which involves predominantly one-dimensional waves over large areas followed by bending of these wave structures to form, ultimately, compact herringbone layouts at full cooling (FIG. 40 d-f). FIG. 40h shows the time evolution of both characteristic wavelengths. The short wavelength tends to decrease as the cooling leads to progressively larger compressive strain on the silicon due to the relatively larger thermal contraction of the PDMS. In particular, this value decreases from 17-18 m in the initial stages to 14.7 m when the herringbone structure becomes prominent and finally to 12.7 m in the fully cooled state. This wavelength is uniform (5% variation) over large areas. By contrast, the long wavelength associated with the herringbone layout exhibits a broad range of values, as is evident from the image in FIG. 40g. Measurements at 100 spots across this sample yield a distribution of values, summarized in the histogram of FIG. 40g. The herringbone structure can be represented by an out-of-plane displacement w=A.sub.1 cos [k.sub.1x.sub.1+k.sub.1A.sub.2 cos(k.sub.2x.sub.2)] (FIG. 49). Here the coefficients, amplitude of wave A.sub.1, long wavelength 2/k.sub.2, jogs wavelength 2/k.sub.1 and the amplitude A.sub.2 of the jogs are determined by analysis for a particular membrane thickness, mechanical properties of the film, and substrate. The short wavelength is (2/k.sub.1)sin(/2). The modeling uses the Si strain, as determined from measured contour lengths and periods of the wavy structures, as the applied prestrain, instead of the thermal prestain (FIG. 50). The actual strain that deforms the Si is typically somewhat smaller than the estimated thermal prestain, due possibly to a loading effect of the Si on the PDMS. The Si strain, for example, is 2.4% at the thermal prestrain of 3.8%. For such a displacement w, the stress, strain, and displacement fields in the Si film can be obtained in terms of A.sub.1, k.sub.1, A.sub.2, and k.sub.2 from the Von Karman plate theory. The fields in the PDMS substrate are obtained from 3D elasticity theory. Minimizing the total energy, which consists of the membrane energy and bending energy in the Si film and the elastic energy in the PDMS substrate gives A.sub.1, k.sub.1, A.sub.2, and k.sub.2. The Young's modulus and Poisson's ratio of Si and PDMS are E.sub.Si=130 GPa, v.sub.Si=0.27, E.sub.PDMS=1.8 MPa, and V.sub.PDMS=0.5. Both the experiments and model give the jog angle to be about 90. The short wavelength given by the theory is 12.4 m at 2.4% biaxial prestain, which agrees well with the experimental results above. The large variation in the long wavelength 2/k.sub.2 is also predicted by the theoretical calculation, from 30 to 60 m.

(165) FIG. 41 presents atomic force microscope (AFM) and scanning electron microscope (SEM) images of structures similar to those illustrated in the fully cooled state of FIG. 40. These images clearly show that the herringbone patterns are characterized by zigzag structures that define two characteristic directions, even though the compressive strain is completely isotropic. The herringbone structures represent a minimum elastic energy configuration that reduces the overall in-plane stress in the system and relieves biaxial compression in both directions. This geometry is, therefore, preferred over large areas, compared to the checkerboard and 1D wave layouts because the herringbone mode is the only one of these three modes that relaxes the in-plane stress in all directions without incurring significant stretch energy. Only in the immediate vicinity of the jogs is significant stretch induced. The 1D mode lowers the prestress only in one direction. The checkerboard mode lowers the stress in all directions, but it produces significant stretch energy accompanying the bending.

(166) The two linecuts extracted from the AFM images indicate periodic, although only approximately sinusoidal, relief profiles along the jogs direction (profile i) and perpendicular to waves (profile ii). The and A.sub.1 of waves, determined from profile ii, are 12.8 and 0.66 m, respectively. The given by theoretical analysis, 12.4 m, is similar with the experimental data; however, the A.sub.1 from theoretical analysis is 0.90 m, somewhat higher value than the experimental results. The SEM images show clearly the intimate bonding between the membrane and the PDMS, as evidenced by the behavior of the sample near the small holes in the silicon in both the raised and recessed regions of the waves. These images also indicate that the wave structures are completely uncorrelated to the position of these holes, since the hole size of 2.5 m is much smaller than the characteristic wavelengths of the deformation modes in our experiments. Studies of the dependence of the geometry of the wavy structures on the thickness of the silicon can provide additional insights into the physics and further validate the mechanics models. FIG. 42 shows some results, including optical micrographs and wavelengths and amplitudes of wave structures formed in membranes with different thicknesses for similar thermal strains. For 100 nm thickness, the and A.sub.1 of waves are 12.6 (0.37) and 0.64 (0.07) m, respectively, and for 320 nm thick, they are 45.1 (1.06) and 1.95 (0.18) m. These values correspond reasonably well with theoretical calculations, which yield the and A.sub.1 are 12.4 and 0.90 m for the 100 nm case and 45.1 and 3.29 m respectively, for the 320 nm case.

(167) These wavy membranes provide true stretchability for strains in various in-plane directions, as opposed to the one-dimensional stretchability provided by previously described ribbon geometries. To investigate this aspect, we perform uniaxial tensile stretching tests along different directions using a calibrated mechanical stage and a 2D stretchable membrane prepared with a thermally induced prestrain of 3.8%. FIG. 43 provides some images. In case i, tensile stain (.sub.st) applied along the direction of the long waves caused the herringbone structure to unfold (.sub.st) 1.8%), gradually leading to a 1D wavy geometry at a fully stretched state (.sub.st) 3.8%). This stretching induces, by the Poisson effect, a compressive strain in the orthogonal direction with an amplitude roughly equal to half of the tensile strain. This compressive strain can be accommodated by compression of the wavy structures in this direction. Upon release of the applied tensile strain, the original herringbone waves recovered to exhibit structures quite similar to the original. (FIG. 51 shows optical micrographs collected after 5, 10, and 15 stretching cycles).

(168) Tensile strains applied in a diagonal direction (case ii), showed similar structural changes although at full stretching the 1D wave structures aligned along a direction defined by the applied strain, rather than the initial geometry. For the perpendicular case iii, at small strain .sub.st 1.8%) certain portions of the sample lose completely the herringbone layout to yield new 1D waves along the stretching direction. With increasing strain, more regions undergo this transformation until the entire area consists of these oriented 1D waves. These newly formed 1D waves are perpendicular to the orientation of the original waves; upon release, they simply bend to create a disordered herringbone-like geometry. For all cases shown in FIG. 43B, the wavelength increases with tensile strain and recovers to its initial value upon the release, even though compressive stresses are induced in the orthogonal direction by the Poisson effect. This behavior arises from an increase of induced by the unfolding of the herringbone waves that is larger than the decrease in this wavelength caused by the Poisson effect. (FIG. 52) For case i, the jogs wavelength, 2/k.sub.1 (FIG. 52A) decreases to 2/k.sub.1 (FIG. 52B), i.e., k.sub.1>k.sub.1, under the applied tensile strain, .sub.st, due to the Poisson effect. However, the corresponding jog angle is larger than the angle due to the unfolding of the herringbone structure. The short wavelength =(2/k.sub.1) sin(/2) becomes =(2/k.sub.1)sin(/2), which may be larger than when the effect of angle change overcomes the Poisson effect. Our theoretical model gives =12.4, 14.6, and 17.2 m for .sub.st=0, 1.8, and 3.8%, which confirms that the short wavelength increases with the applied strain, as observed in experiments. For case iii, both and 2/k.sub.1 increased with applied stretching strain, since waves were relaxed along the direction of stretching strain, and the jogs angle () was not changed significantly by the Poisson effect. The biaxial stretchability of the buckled membranes was also investigated by thermally induced tensile strains (FIG. 53). The herringbone waves generated by thermal strain slowly disappeared as the sample was heated; they recovered completely upon cooling.

(169) These observations apply only to the central regions of the membranes. As indicated in the bottom frames of FIG. 39, the edges of the membranes show 1D wave structures with wavevectors oriented along the edges. AFM images and linecut profiles of the edge region, the central region, and the transition area between them are shown in FIG. 44. The 1D waves that originate near the edge of the Si (top frame) gradually become bent (middle frame) until they transform into the herringbone geometries in the central regions (bottom frame). The values in these regions are 16.6, 13.7, and 12.7 m, respectively (from top frame), with A.sub.1 of 0.52, 0.55, and 0.67 m. Compared to the 1D waves at the edges, the 2D herringbone waves have smaller and A.sub.1, suggesting that the inner area of Si is affected more strongly by the compressive strain than the edges. The stress state near the edge is approximately uniaxial compression within some distance range because of the traction-free edge of the membrane. This uniaxial compression is parallel to this free edge and therefore leads to 1D waves along the edge. The stress state, however, becomes equi-biaxial compressive in the central region where herringbone structures result. For the transition area between the 1D wavy edge and the herringbone waves, the unbalanced biaxial compression causes a semi-herringbone wave with the large jog angle. Our model yields and A.sub.1 of 16.9 and 0.83 m, respectively, for the 1D waves and 12.4 and 0.90 m for the herringbone structure. These results agree reasonably well with the experimentally observed values.

(170) To investigate further these edge effects, we fabricated rectangular membranes with lengths of 1000 m and with widths of 100, 200, 500, and 1000 m, all on the same PDMS substrate. FIG. 45 shows optical micrographs of these structures, for two different levels of thermal prestrain. At low thermal prestrain (ca. 2.3%, FIG. 45A), the 100 and 200 m wide membranes exhibit perfect 1D waves from one side to the other, with flat, undeformed regions at the ends. The 500 m wide membrane shows similar 1D waves and flat regions, but the waves have slightly bent geometries in the middle of the structures, with overall ordering and uniformity in orientation that are substantially less than the 100 and 200 m cases. For the 1000 m square, 1D waves are present in the central regions of the edges, with flat areas in the corners. The center part of the membrane shows fully developed herringbone geometries. For the corner flat region, there is an approximately stress-free state due to the two free edges. No waves form near such corners. With increasing prestrain (4.8%, FIG. 45B), the flat regions in all cases decrease in size. 1D wavy behavior persists in the 100 and 200 m ribbons, but pronounced herringbone morphologies appear in the central region of the 500 m case. At the higher prestrain, equi-biaxial compressive strains are present in the inner region of 500 m wide membrane. For the 1000 m square membrane, the herringbone behavior extends to regions close to the edges. The characteristic length scales that define the spatial extent of the flat regions, which we refer to as the edge effect length, Ledge, can be evaluated as a function of membrane size and prestrain. FIG. 45C shows results that indicate a linear scaling of this length with prestrain, in a manner that is independent of the size of the membrane, for the cases investigated here. As the prestrain becomes higher, the length of uniaxial strain region becomes smaller. Therefore, shorter range 1D waves form and similar behavior can be observed in the stress-free regions near the two free edges.

(171) FIG. 46 shows optical micrographs of wavy structures that form in other membrane geometries, including circles, ovals, hexagons, and triangles. The results are qualitatively consistent with observations in the ribbons and squares of FIG. 45. In particular, the edge regions show 1D waves oriented parallel to the edges. Waves with the orthogonal orientation only appear at distances larger than the L.sub.edge from the edge. For the circle, 1D waves appear near the edges, with an overall radial orientation due to the shape of the membrane. Herringbone waves appear in the center. The ovals exhibit similar behavior, although with flat regions at the edges of the major axis, due to the small radius of curvature in these regions. For the hexagon and triangle shapes, the sharp corners (angles of 120 and 60, respectively), lead to flat regions. Herringbone geometries appear in the center of the hexagons. The centers of the triangles show the merging of 1D waves, for the level of prestrain shown here. For shapes with clear corners (e.g., hexagon, triangle, and tip of ellipse), there is no wave near the corner because the two intersecting free edges (not necessarily perpendicular) give a stress-free state. For the triangle shape, there is not enough space to generate the herringbone structure even in the central region.

(172) The membranes themselves provide a path to biaxially stretchable electronic devices. The edge effects outlined above can be exploited to realize a particular outcome that might be useful for certain classes of such devices. In particular, in an imaging system, there might be value in maintaining flat, undeformed regions at the locations of the photodetectors to avoid nonideal behavior that occurs when these devices have wavy shapes. FIG. 47 presents some representative examples of stretchable membranes that achieve this outcome. These structures consist of 100100 m square islands connected by 30 m150 m ribbons (30 m210 m for orthogonal ribbons) in the vertical and horizontal directions (FIG. 47A, C), and in the vertical, horizontal and diagonal directions (FIG. 47E, G). Changes in the amplitudes and wavelengths of the waves in the ribbons provide a means to accommodate applied strains, in a manner that largely avoids deformations in the regions of the square islands. We examined the behavior of these structures at several different applied strains. Parts a and e of FIG. 47 show representative cases in the low strain (ca. 2.3%) regime, applied by heating the samples in an oven. Parts c and g of FIG. 47 show the same structures at relatively high biaxial strains (ca. 15%), applied using a mechanical stage. As is evident, in the low-strain regime the islands remain flat; at sufficiently high strains, wave structures begin to form in these regions. Good adhesion between PDMS and Si was maintained at all strains, as shown in the tilted angle SEM images (FIG. 47B, D, F, H). The insets of high-magnification SEM image in parts b and d of FIG. 47 also confirm the strong bond of Si with PDMS.

(173) In summary, nanomembranes of silicon can be integrated with prestrained elastomeric substrates to create 2D wavy structures with a range of geometries. Many aspects of the mechanical behavior of these systems are in good agreement to theoretically predicted behaviors. These results are useful for applications of electronics in systems where full stretchability is required during use or during installation.

REFERENCES

(174) 1. Duan, X. & Lieber, C. M. General synthesis of compound semiconductor nanowires. Adv. Mater. 12, 298-302 (2000). 2. Xiang, J., Lu, W., Hu, Y., Wu, Y., Yan, H. & Lieber, C. M. Ge/Si nanowire heterostructures as high-performance field-effect transistors. Nature 441, 489-493 (2006). 3. Wu, Y., Yan, H., Huang, M., Messer, B., Song, J. H. & Yang, P. Inorganic semiconductor nanowires: rational growth, assembly, and novel properties. Chem. Eur. J. 8, 1261-1268 (2002). 4. Pan, Z. W., Dai, Z. R. & Wang, Z. L. Nanobelts of semiconducting oxides. Science 291, 1947-1949 (2001). 5. Peng, X., Manna, L., Yang, W., Wickham, J., Scher, E., Kadavanich, A. & Alivisatos, A. P. Shape control of CdSe nanocrystals. Nature 404, 59-61 (2000). 6. Wang, D., Chang, Y.-L., Lu, Z. & Dai, H. Oxidation resistant germanium nanowires: bulk synthesis, long chain alkanethiol functionalization, and Langmuir-Blodgett assembly. J. Am. Chem. Soc. 127, 11871-11875 (2005). 7. Huang, M. H., Wu, Y., Feick, H., Tran, N., Weber, E. & Yang, P. Catalytic growth of zinc oxide nanowires by vapor transport. Adv. Mater. 13, 113-116 (2001). 8. Gudiksen, M. S., Wang, J. & Lieber, C. M. Synthetic control of the diameter and length of single crystal semiconductor nanowires. J. Phys. Chem. B 105, 4062-4064 (2001). 9. Yu, H., Li, J., Loomis, R. A., Wang, L.-W. & Buhro, W. E. Two- versus three-dimensional quantum confinement in indium phosphide wires and dots. Nat. Mater. 2, 517-520 (2003). 10. Sun, Y. & Rogers, J. A. Fabricating semiconductor nano/microwires and transfer printing ordered arrays of them onto plastic substrates. Nano Lett. 4, 1953-1959 (2004). 11. Yin, Y., Gates, B. & Xia, Y. A soft lithography approach to the fabrication of nanostructures of single crystalline silicon with well-defined dimensions and shapes. 12, 1426-1430 (2000). 12. Kodambaka, S., Hannon, J. B., Tromp, R. M. & Ross, F. M. Control of Si nanowire growth by oxygen. Nano Lett. 6, 1292-1296 (2006). 13. Shan, Y., Kalkan, A. K., Peng, C.-Y. & Fonash, S. J. From Si source gas directly to positioned, electrically contacted Si nanowires: the self-assembling grow-in-place approach. Nano Lett. 4, 2085-2089 (2004). 14. He, R., Cao, D., Fan, R., Hochbaum, A. I., Carraro, C., Maboudian, R. & Yang, P. Si nanowire bridges in microtrenches: integration of growth into device fabrication. Adv. Mater. 17, 2098-2102 (2005). 15. Lee, K. J., Motala, M. J., Meitl, M. A., Childs, W. R., Menard, E., Shim, A. K., Rogers, J. A. & Nuzzo, R. G. Large-area, selective transfer of microstructured silicon: a printing-based approach to high-performance thin-film transistors supported on flexible substrates. Adv. Mater. 17, 2332-2336 (2005). 16. Gao, P. X., Ding, Y., Mai, W., Hughes, W. L., Lao, C. & Wang, Z. L. Conversion of zinc oxide nanobelts into superlattice-structured nanohelices. Science 309, 1700-1704 (2005). 17. Kong, X. Y., Ding, Y., Yang, R. & Wang, Z. L. Single-crystal nanorings formed by epitaxial self-coiling of polar nanobelts. Science 303, 1348-1351 (2004). 18. Chen, P., Chua, S. J., Wang, Y. D., Sander, M. D. & Fonstad, C. G. InGaN nanorings and nanodots by selective area epitaxy. Appl. Phys. Lett. 87, 143111 (2005). 19. Manna, L., Milliron, D. J., Meisel, A., Scher, E. C. & Alivisatos, A. P. Controlled growth of tetrapod-branched inorganic nanocrystals. Nat. Mater. 2, 382-385 (2003). 20. Dick, K. A., Deppert, K., Larsson, M. W., Mrtensson, T., Seifert, W., Wallenberg, L. R. & Samuelson, L. Synthesis of branched nanotrees by controlled seeding of multiple branching events. Nat. Mater. 3, 380-384 (2004). 21. Khang, D.-Y., Jiang, H., Huang, Y. & Rogers, J. A. A stretchable form of single-crystal silicon for high-performance electronics on rubber substrates. Science 311, 208-212 (2006). 22. Schmidt, O. G. & Eberl, K. Thin solid films roll up into nanotubes. Nature 410, 168-168 (2001). 23. Zhang, L., Ruh, E., Grtzmacher, D., Dong, L., Bell, D. J., Nelson, B. J. & Schnenberger, C. Anomalous coiling of SiGe/Si and SiGe/Si/Cr helical nanobelts. Nano Lett. 6, 1311-1317 (2006). 24. Jin, H.-C., Abelson, J. R., Erhardt, M. K. & Nuzzo, R. G. Soft lithographic fabrication of an image sensor array on a curved substrate. J. Vac. Sci. Technol. B 22, 2548-2551 (2004). 25. Someya, T., Sekitani, T., Iba, S., Kato, Y., Kawaguchi, H. & Sakurai, T. A large-area, flexible pressure sensor matrix with organic field-effect transistors for artificial skin applications. Proc. Natl. Acad. Sic. U.S.A. 101, 9966-9970 (2004). 26. Nathan, A., Park, B., Sazonov, A., Tao, S., Chan, I., Servati, P., Karim, K., Charania, T., Striakhilev, D., Ma, Q. & Murthy, R. V. R. Amorphous silicon detector and thin film transistor technology for large-area imaging of X-rays. Microelectronics J. 31, 883-891 (2000). 27. Lacour, S. P., Jones, J., Wagner, S., Li, T. & Suo, Z. Stretchable interconnects for elastic electronic surfaces. Proc. IEEE 93, 1459-1467 (2005). 28. Childs, W. R., Motala, M. J., Lee, K. J. & Nuzzo, R. G. Masterless soft lithography: patterning UV/Ozone-induced adhesion on poly(dimethylsiloxane) surfaces. Langmuir 21, 10096-10105 (2005). 29. Sun, Y., Kumar, V., Adesida, I. & Rogers, J. A. Buckled and wavy ribbons of GaAs for high-performance electronics on elastomeric substrates. Adv. Mater. in press. 30. Sun, Y., Khang, D.-Y., Hua, F., Hurley, K. Nuzzo, R. G. & Rogers, J. A. Photolithographic route to the fabrication of micro/nanowires of III-V semiconductors. Adv. Funct. Mater. 15, 30-40 (2005). 32. Loo, Y.-L.; Someya, T., Baldwin, K. W., Bao, Z., Ho, P., Dodabalapur, A., Katz, H. E. & Rogers, J. A. Soft, conformable electrical contacts for organic semiconductors: high-resolution plastic circuits by lamination. Proc. Natl. Acad. Sci. U.S.A. 99, 10252-10256 (2002). 33. Suo, Z., Ma, E. Y., Gleskova, H., Wagner, S. Mechanics of rollable and foldable film-on-foil electronics. Appl. Phys. Lett. 74, 1177-1179 (1999). P. Mandlik, S. P. Lacour, J. W. Li, S. Y. Chou, and S. Wagner, Ieee Electron Device Letters 27, 650-652 (2006). D. S. Gray, J. Tien, and C. S. Chen, Advanced Materials 16, 393+(2004). S. P. Lacour, S. Wagner, Z. Y. Huang, and Z. Suo, Applied Physics Letters 82, 2404-2406 (2003). S. P. Lacour, J. Jones, S. Wagner, T. Li, and Z. G. Suo, Proceedings of the Ieee 93, 1459-1467 (2005). J. Jones, S. P. Lacour, S. Wagner, and Z. G. Suo, Journal of Vacuum Science & Technology A 22, 1723-1725 (2004). S. P. Lacour, J. Jones, Z. Suo, and S. Wagner, Ieee Electron Device Letters 25, 179-181 (2004). W. T. S. Huck, N. Bowden, P. Onck, T. Pardoen, J. W. Hutchinson, and G. M. Whitesides, Langmuir 16, 3497-3501 (2000). N. Bowden, S. Brittain, A. G. Evans, J. W. Hutchinson, and G. M. Whitesides, Nature 393, 146-149 (1998). S. Wagner, S. P. Lacour, J. Jones, P. H. I. Hsu, J. C. Sturm, T. Li, and Z. G. Suo, Physica E-Low-Dimensional Systems & Nanostructures 25, 326-334 (2004). H. Kudo, T. Sawada, E. Kazawa, H. Yoshida, Y. Iwasaki, and K. Mitsubayashi, Biosensors & Bioelectronics 22, 558-562 (2006). T. Li, Z. G. Suo, S. P. Lacour, and S. Wagner, Journal of Materials Research 20, 3274-3277 (2005). S. P. Lacour, D. Chan, S. Wagner, T. Li, and Z. G. Suo, Applied Physics Letters 88 (2006). S. P. Lacour, C. Tsay, and S. Wagner, Ieee Electron Device Letters 25, 792-794 (2004). S. P. Lacour, S. Wagner, R. J. Narayan, T. Li, and Z. G. Suo, Journal of Applied Physics 100 (2006). Reuss, R. H et al. Proc. IEEE 2005, 93, 1239. Jain, K. et al. Proc. IEEE 2005, 93, 1500. Nathan, A. et al. Microelectron. Reliab. 2002, 42, 735. Someya, T et al. T. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9966. Hsu, P. H. I. et al. IEEE Trans. Electron. DeV. 2004, 51, 371. Jin, H. C. et al. Vac. Sci. Technol., B: Microelectron. Nanometer Struct.-Process., Meas., Phenom. 2004, 22, 2548. Nathan, A.; et al. Microelectron. J. 2000, 31, 883. Someya, T. et al. Proc. Natl. Acad. Sci. U.S.A. 2005, 103, 12321. Lacour, S. P. et al. Proc. IEEE 2005, 93, 1459. (c) Lacour, S. P. et al. Appl. Phys. Lett. 2003, 82, 2404. Khang, D.-Y. et al. Science 2006, 311, 208. Sun, Y. et al. Adv. Mater. 2006, 18, 2857. Sun, Y. et al. Nat. Nanotechnol. 2007, 1, 201. Ouyang, M. et al. Chem. Mater. 2000, 12, 1591. Childs, W. R.; Nuzzo, R. G. J. Am. Chem. Soc. 2002, 124, 13583. Efimenko, K. et al. J. Colloid Interface Sci. 2002, 254, 306. Hillborg, H. et al. Langmuir 2004, 20, 785. Buma, T. et al. Appl. Phys. Lett. 2001, 79, 548. Properties of Silicon; INSPEC: New York, 1998. The coefficients of thermal expansion are RPDMS) 3.110.sup.4 K.sup.1 and .sub.Si) 2.610.sup.6 K.sup.1 for PDMS substrate and Si nanomembrane, respectively. The thermal prestrain for the samples prepared at 150 C. was calculated by T=(3.110.sup.42.610.sup.6)(15025)=3.8%. Timoshenko, S. Theory of Plates and Shells; McGraw-Hill: New York, 1940. Timoshenko, S.; Goodier, J. N. Theory of Elasticity, 3rd ed.; McGraw-Hill: New York, 1969. Chen, X.; Hutchinson, J. W. J. Appl. Mech. Trans. ASME 2004, 71, 597. Chen, X.; Hutchinson, J. W. Scr. Mater. 2004, 50, 797. Huang, Z. Y. et al. J. Mech. Phys. Solids 2005, 53, 2101. Bietsch, A.; Michel, B. J. Appl. Phys. 2000, 88, 4310. Ohzono, T.; Shimomura, M. Phys. Rev. B 2004, 69, 132202. Ohzono, T.; Shimomura, M. Langmuir 2005, 21, 7230.

(175) Example 4

(176) Heterogeneously Integrated, Three Dimensional Electronics by Use of Printed Semiconductor Nanomaterials

(177) We have developed a simple approach to combine broad classes of dissimilar materials into heterogeneously integrated (HGI) electronic systems with two or three dimensional (3D) layouts. The process begins with the synthesis of different semiconductor nanomaterials (e.g. single walled carbon nanotubes and single crystal nanowires/ribbons of gallium nitride, silicon and gallium arsenide) on separate substrates. Repeated application of an additive, transfer printing process using soft stamps and these substrates as donors, followed by device and interconnect formation, yields high performance 3D-HGI electronics that incorporate any combination of these (or other) semiconductor nanomaterials on rigid or flexible device substrates. This versatile methodology can produce a wide range of unusual electronic systems that are difficult or impossible to achieve using other techniques.

(178) Many existing and emerging electronic devices benefit from the monolithic, heterogeneous integration (HGI) of dissimilar classes of semiconductors into single systems, in either two or three dimensional (2D or 3D) layouts. Examples include multifunctional radio frequency communication devices, infrared (IR) imaging cameras, addressable sensor arrays and hybrid CMOS/nanowire/nanodevice circuits (3-7). In some representative systems, compound semiconductors or other materials provide high speed operation, efficient photodetection or sensing capabilities while silicon CMOS provides digital readout and signal processing, in circuits that often involve stacked 3D configurations. Wafer bonding (8) and epitaxial growth (9,10) represent the two most widely used methods for achieving these types of 3D-HGI systems. The former process involves physical bonding, by use of adhesives or thermally initiated interface chemistries, of integrated circuits, photodiodes or sensors formed separately on different semiconductor wafers. This approach works well in many cases, but it has important drawbacks, including (i) limited ability to scale to large areas or to more than a few layers in the third (i.e. stacking) dimension, (ii) incompatibility with unusual (e.g. nanostructured materials) or low temperature materials and substrates, (iii) challenging fabrication and alignment for the through-wafer electrical interconnects, (iv) demanding requirements for flat, planar bonding surfaces and (v) bowing and cracking that can occur from mechanical strains generated by differential thermal expansion/contraction of disparate materials. Epitaxial growth provides a different approach that involves the direct formation, by molecular beam epitaxy or other means, of thin layers of semiconductor materials on the surfaces of wafers of other materials. Although this method avoids some of the aforementioned problems, the requirements for epitaxy place severe restrictions on the quality and type of materials that can be grown, even when buffer layers and other advanced techniques are used. By contrast, emerging classes of semiconductor nanomaterials, such as nanoscale wires, ribbons, membranes or particles of inorganic materials, or carbon based systems such as single walled carbon nanotubes (SWNTs) or graphene sheets (11-14), can be grown and then suspended in solvents or transferred onto substrates in a manner that bypasses the need for epitaxial growth or wafer bonding. Recent work shows, for example, the integration, in 2D layouts, of crossed nanowire diodes formed by solution casting (15). The results presented here illustrate how dissimilar single crystal inorganic semiconductors (e.g., nanowires/ribbons of GaN, Si and GaAs) can be combined with one another and also with other classes of nanomaterials (e.g. SWNTs) using a scalable and deterministic printing method to yield complex, HGI electronic systems in 2D or 3D layouts. In particular, ultrathin multilayer stacks of high performance metal-oxide-semiconductor field-effect transistors (MOSFETs), metal-semiconductor field-effect transistors (MESFETs), thin film transistors (TFTs), photodiodes and other components, integrated into device arrays, logic gates and actively addressable photodetectors on rigid inorganic and flexible plastic substrates demonstrate some of the capabilities.

(179) FIG. 57 illustrates representative steps for producing these 3D-HGI systems. The process begins with synthesis of the semiconductor nanomaterials, each on their own source substrate. The devices presented here integrate nanowires and nanoribbons of single crystal Si, GaN and GaAs, formed using wafer based source materials and lithographic etching procedures (16-21), and networks of SWNTs grown by chemical vapor deposition (13,21). Scanning electron micrographs at the top of FIG. 57 show these semiconductor nanomaterials, after their removal from the source substrates. For circuit fabrication, these elements remain in the configurations defined on the wafers during the fabrication or growth stage: aligned arrays in the case of the Si, GaN and GaAs nanowires/ribbons and sub-monolayer random networks for the SWNTs. High temperature doping and annealing procedures for ohmic contacts to the Si, GaN and GaAs can be performed on the source substrates. The next step involves transferring these processed elements, using an elastomeric stamp-based printing technique described previously, from the source substrates to a device substrate, such as a sheet of polyimide (PI) as illustrated in FIG. 57. In particular, laminating a stamp of polydimethysiloxane (PDMS) against the source substrate establishes soft, van der Waals adhesion contacts to the semiconductor nanomaterial elements. Contacting the inked stamp to a device substrate (e.g. PI sheet) with a thin, spin-cast layer of a liquid prepolymer (e.g. polyamic acid) on its surface and then curing the polymer leaves these semiconductor materials embedded on and well adhered to this layer (16-20) when the stamp is removed. Similar procedures work well with a range of substrates (i.e. rigid or flexible; organic or inorganic) and semiconductor nanomaterials [A slightly modified version of this process is used for the SWNTs (21).]. The thickness of the interlayer (PI in this case) can be as small as 500 nm and is typically 1-1.5 m, for the systems described here. After some additional processing, including formation of gate dielectrics, electrodes and interconnects, the transfer printing and device fabrication steps can be repeated, beginning with spin-coating a new prepolymer interlayer on top of the previously completed circuit level. Automated stages specially designed for transfer printing or conventional mask aligners enable overlay registration accuracy of 1 m over several square centimeters. (22) (FIG. 61). Layer to layer interconnects (23) are formed simply by evaporating metal lines over and into openings in the interlayers defined by photopatterning and/or dry etching. This unusual approach to 3D-HGI electronics has several important features. First, all of the processing on the device substrate occurs at low temperatures, thereby avoiding differential thermal expansion/shrinkage effects that can result in unwanted deformations in multilayer stacked systems. This operation also enables the use of low temperature plastic substrates and interlayer materials, and it helps to ensure that underlying circuit layers are not thermally degraded by processing of overlying devices. Second, the method is applicable to broad classes of semiconductor nanomaterials, including emerging materials such as thin films of SWNT. Third, the soft stamps enable non-destructive contacts with underlying device layers; these stamps, together with the ultrathin semiconductor materials, can also tolerate surfaces that have some topography. Fourth, the ultrathin device geometries (<1 m) and interlayers (<1.5 m) allow easy formation of layer to layer electrical interconnects. These features, which overcome many of the disadvantages of conventional approaches, are illustrated in the several circuit demonstrations described in the following.

(180) FIG. 58 presents a three layer, 3D stack arrayed Si MOSFETs fabricated using the general process flow illustrated in FIG. 57, using single crystal silicon nanoribbons, with doped contacts (formed on the source wafer), plasma enhanced chemical vapor deposited SiO2 dielectrics, and Cr/Au metallization for source, drain and gate (24). Each device uses three aligned nanoribbons, with widths, thicknesses and lengths of 87 m, 290 nm and 250 m, respectively. FIG. 2A shows a top view optical micrograph of an edge of the system with a layout designed to reveal separately the parts of the substrate that support one, two and three layers of MOSFETs. A ninety degree rotation of the device geometry for the second layer, relative to the first and third, helps to clarify the layout of the system. Schematic cross-sectional and angled views of the stacked structure appear in FIG. 58B. The sample can be viewed in 3D using confocal optical microscopy. FIG. 58C shows top and angled views of such images, colorized for ease of viewing. (The image quality degrades somewhat with depth, due to scattering and absorption from the upper layers). FIG. 58D presents electrical measurements of representative devices [top gate MOSFETs with channel lengths (L.sub.c) of 19 m, channel overlap distances (Lo), defined by distance that the gate electrode extends over the doped source/drain regions, of 5.5 m, and channel widths (W) of 200 m] in each layer. Devices on each of the three layers, which are formed on a PI substrate, show excellent properties (linear mobilities of 47030 cm.sup.2/Vs, on/off ratios >104 and threshold voltages of 0.10.2V) and no systematic differences between devices in different layers. Additional layers can be added to this system, by repeating the same procedures. In addition to 3D circuits with a single semiconductor, as illustrated in FIG. 59, various semiconductors can be used in multiple layers to form full 3D-HGI systems. To illustrate this capability, we fabricated arrays of MESFETS (in particular, high electron mobility transistors, HEMTs), MOSFETs and TFTs using GaN and Si nanoribbons, and SWNT films, respectively, on PI substrates. FIGS. 59A and 59B show high magnification optical and confocal images, respectively, of the resulting devices. The GaN HEMTs on the first layer use ohmic contacts (Ti/Al/Mo/Au, annealed on the source wafer) for source and drain, and a Schottky (Ni/Au) contacts for the gates. The channel lengths and widths, and the gate widths are 20, 170, and 5 m respectively. Each device uses GaN ribbons (composed of multilayer stacks of AlGaN/GaN/AlN) with thicknesses, widths and lengths of 1.2, 10 and 150 m, respectively, interconnected electrically by processing on the device substrate. The SWNT TFTs on the second layer use SiO.sub.2/Epoxy for the gate dielectric and Cr/Au for source, drain and gate, with channel lengths and widths of 50 and 200 m, respectively. The Si MOSFETs use the same design as those shown in FIG. 58. Various other 3D-HGI devices can be constructed using different combinations of Si, SWNT and GaN (FIGS. 61 and 62). FIG. 59C presents the current-voltage characteristics of typical devices in the systems of FIGS. 59A and 59B. In all cases, the properties are similar to those fabricated on the source wafers: the GaN HEMTs have threshold voltages (V.sub.th) of 2.40.2 V, on/off ratios >10.sub.6, and transconductances of 0.60.5 mS; the SWNT TFTs have V.sub.th=5.31.5 V, on/off ratios >10.sub.5 and linear mobilities of 5.92.0 cm.sup.2/Vs; the Si MOSFETs have V.sub.th=0.20.3 V, on/off ratios >10.sup.4 and linear mobilities of 50030 cm.sup.2/Vs. An interesting aspect of these devices, which follows from the use of thin PI substrates (25 m), devices (2.4 m) and PI/PU interlayers (5 m), is their mechanical bendability, which is important for applications in flexible electronics. We evaluated the effective transconductance (g.sub.eff) for the Si, SWNT, and GaN devices in the 3D-HGI system of FIG. 59A as a function of bend radius. FIG. 59D, which shows these data, as normalized to the transconductance in the unbent state (g.sub.0eff), illustrates the stable performance for bend radii down to 3.7 mm.

(181) Electrical interconnections formed between different levels in these 3D-HGI devices can create interesting circuit capabilities. The thin polymer interlayers enable these interconnects to be formed easily by evaporating metal lines over and into lithographically defined openings. FIG. 60 presents some examples. The first, shown in FIG. 60A, is a 3D NMOS inverter (logic gate) in which the drive (L=4 m, W=200 m) and load (L=4 m, W=30 m) Si MOSFETs are on different levels. With a supply voltage of 5V, this double-layer inverter exhibits well-defined transfer characteristics with gains of 2, comparable to the performance of conventional planar inverters that use the similar transistors (25). FIG. 60B shows an inverter with a complementary design (CMOS) by use of integrated n-channel Si MOSFETs and p-channel SWNT TFTs, designed in order to equalize the current-driving capability in both pull-up and pull-down directions (FIG. 65). Transfer curves collected with a 5 V bias to the VDD terminal and gate voltage (input) swept from 0 V to 5 V appear in FIG. 60A. The curve shapes and gains (as high as 7) are qualitatively consistent with numerical circuit simulations (FIG. 65). As a third example, we built GaAs metal-semiconductor-metal (MSM) infrared (IR) detectors (26) integrated with Si MOSFETs on flexible PI substrates, to demonstrate a capability for fabricating unit cells that could be used in active IR imagers. In this case, printed nanoribbons of GaAs (thicknesses, widths and lengths of 270 nm, 100 m and 400 m, respectively) transferred onto a substrate with a printed array of Si nanoribbon MOSFETs form the basis of the MSMs. Electrodes (Ti/Au=5/70 nm) deposited on the ends of these GaAs nanoribbons form back-to-back Schottky diodes with separations of 10 m. The resulting detector cells exhibit current enhancement as the intensity of IR illumination increases (FIG. 60C), consistent with circuit simulation (FIG. 66). A responsivity of about 0.30 A/W at the 850 nm wavelength is observed from 1 to 5 V, without taking into account the light reflected from the surface of the semiconductor. The system also exhibits bendability with radii of curvature below 1 cm, which could be useful for advanced systems such as curved focal plane arrays for wide angle IR night vision imagers.

(182) Printed semiconductor nanomaterials provide new approaches to 3D-HGI systems and could have important applications in various fields of application, not only those suggested by the systems reported here, but also others including microfluidic devices with integrated readout and sensing electronics, chem/bio sensor systems that incorporate unusual sensing materials with conventional silicon based electronics and photonic/optoelectronic systems that combine light emitters of compound semiconductor with silicon drive electronics or microelectromechanical structures. Further, the compatibility of this approach with thin, lightweight plastic substrates may create additional opportunities for devices that have unusual form factors or mechanical flexibility as key features.

(183) MATERIALS AND METHODS: Device fabrication: Silicon devices: The fabrication begins with definition of contact doped thin ribbons of single crystal silicon, by processing silicon on insulator wafer (SOI; Soitec unibond with a 290 nm top Si layer with doping level of 6.09.410.sup.14/cm.sup.3). The first step involved phosphorous doping, using a solid source and spin-on-dopant (Filmtronic, P509), and a photolithographically defined layer of plasma enhanced chemical vapor (PECVD) deposited SiO.sub.2 (Plasmatherm, 300 nm, 900 mTorr, 350 sccm, 2% SiH.sub.4/He, 795 sccm NO.sub.2, 250 C.) as a mask to control where dopant diffuses into the silicon. After doping, SF.sub.6 plasma etching through a patterned layer of photoresist defined the ribbons. Undercut etching of the buried oxide with concentrated HF solution (Fisher Chemicals) released the ribbons from the wafer. This procedure completed the fabrication of contact doped ribbons of single crystal silicon. In the next step, contacting a flat elastomeric stamp of polydimethylsiloxane (PDMS, A:B=1:10, Sylgard 184, Dow Corning) with the photoresistcoated ribbons and then peeling back the stamp removed the ribbons from the wafer and left them adhered, by van der Waals forces between the hydrophobic PDMS and the photoresist, to the surface of the stamp. The stamp thus inked with s-Si ribbons from wafer was laminated against a polyimide (PI) sheet of 25 m (Dupont, Kapton100E) spincoated with a thin layer (1.5 m) of liquid PI precursor, polyamic acid (Sigma_Aldrich Inc.). Curing the precursor, peeling off the PDMS stamp, and stripping the photoresist left the ribbons embedded on and well adhered to the surface of the PI substrate. The gate dielectric layer consisted of a layer of SiO.sub.2 (thickness 100 nm) deposited by PECVD at relatively low temperature, 250 C. Photolithography and CF.sub.4 plasma etching defined openings to the doped source/drain regions of the silicon. Source, drain and gate electrodes of Cr/Au (5/100 nm, from bottom to top by electron beam evaporation, Temescal FC-1800) were defined in a single step by photolithography and wet etching.

(184) GaN devices: GaN microstructures were fabricated on a bulk wafer of GaN with heteorostructure [AlGaN(18 nm)/GaN(0.6 m)/AlN(0.6 m)/Si]. An ohmic contact area defined by AZ 5214 photoresist and then cleaned with SiCl.sub.4 plasma in a RIE system. A Ti/Al/Mo/Au (15/60/35/50 nm) metal layer was then deposited by e-beam evaporation (Ti/Al/Mo) and thermal evaporation (Au). Washing away the resist completed left metal contacts on the GaN. Thermal annealed at 850 C. for 30 sec in N.sub.2 ambient formed the ohmics. SiO2 (Plasmatherm, 300 nm, 900 mTorr, 350 sccm, 2% SiH4/He, 795 sccm NO.sub.2, 250 C.) and Cr metal (e-beam evaporator, 150 nm) layers were deposited as the mask materials for subsequent inductively coupled plasma (ICP) etching. Photolithography, wet etching, and RIE processing (50 mTorr, 40 sccm CF4, 100 W, 14 min) defined the ribbon geometries of the GaN. After removing the photoresist with acetone, ICP dry etching (3.2 mTorr, 15 sccm Cl2, 5 sccm Ar, 100V Bias, 14 min) was used to remove the exposed GaN and to etch slightly into the Si (1.5 m) to facilitate the subsequent anisotropic etching. The Si was then etched away from underneath the GaN using a tetramethyl ammonium hydroxide (Aldrich, 150 C. for 4 min 30 sec). The sample was dipped in BOE (6:1, NH.sub.4F: HF) for 30 sec to remove the PECVD SiO.sub.2 and a new 50 nm e-beam evaporated SiO2 layer was deposited on top of the GaN ribbons. A PDMS slab inked with the GaN ribbons from mother wafer was then laminated against a PI sheet coated with 2 m polyurethane (PU, Norland optical adhesive, No. 73). The sample was exposed to UV light (173 Wcm.sup.2) for 15 min to cure the PU. Peeling back the PDMS and removing the e-beam SiO2 by immersion in BOE for 20 sec resulted in the transfer of the GaN elements to the plastic substrate. A negative photoresist (AZ nLoF2020) was used to pattern Schottky contacts of Ni/Au (80/180 nm). The photoresist was removed with an AZ stripper (KWIK for 30 min).

(185) SWNT devices: Chemical vapor deposition (CVD) was used to grow random networks of individual single walled carbon nanotubes on SiO2/Si wafers. Ferritin (Sigma Aldrich) deposited on the substrate with a methanol was used as a catalyst. The feeding gas was methane (1900 sccm CH.sub.4 with a 300 sccm H.sub.2). The quartz tube in the furnace was flushed with a high flow of Ar gas for cleaning before growth. During the growth, the temperature was held at 900 C. for 20 minutes. The transfer involved either procedures similar to the printing like processes described previously, or a slightly different method in which a thick Au layer and a PI precursor were coated on the SiO.sub.2/Si substrate with the tubes. After curing the PI, the Au/PI was peeled back. Laminating this layer against a prepatterned device substrate coated with a thin epoxy layer (SU8, 150 nm) and then removing the PI and Au layer by oxygen reactive ion etching and wet etching, respectively, completed the transfer. In the case of bottom gate devices, the substrate supported prepatterned gate electrodes and dielectrics. In particular, gate electrodes of Cr/Au/Cr (2/10/10 nm) were patterned by photolithography and then, 300 nm SiO.sub.2 was deposited on the substrate using PECVD. The source and drain electrodes of Cr/Au (2/20 nm) were defined directly on top of the tubes.

(186) 3D Circuit: 3D Si NMOS inverter: Multilayer devices were constructed by repetitively applying the same fabrication procedures. In particular, to the PI precursor was spin-cast on the top of an existing layer of devices, and silicon ribbons were transfer-printed on top. The same processes were then used to fabricate devices. For vertical metal interconnects, an electrode area was defined by photo-patterning openings in a layer of AZ4620 photoresist, and then etching away the SiO.sub.2 and PI in this exposed area using CF.sub.4 and O.sub.2 plasma in a RIE system. Depositing 300 nm Al into this area established contacts at the bottom, and provided an electrically continuous connection over the step edge formed by the etched SiO.sub.2 and PI.

(187) SWNT and Si CMOS inverter: The SWNT devices consisted of source/drain contacts of Au (20 nm) defined by photolithography on the tube networks. The SiO.sub.2 (100 nm)/Si wafer substrate provided the gate dielectric and gate. Epoxy (SU8, 500 nm) was then spin-coated onto this substrate after the SWNT transistors were selectively coated with photoresist (AZ5214). After UV exposure for curing of epoxy, a PDMS slab inked with undoped Si ribbons was laminated against the substrate and subsequently removed by slow manual peeling to complete the transfer-printing process. Cr/Au (5/100 nm) were used as Schottky contacts for source and drain electrodes in the silicon devices. Al (100 nm) was used to connect the SWNT and Si transistor.

(188) GaAs MSM IR detector integrated with Si TFT: GaAs wafers (IQE Inc., Bethlehem, Pa.) were used to generate back-to-back schottky diodes. The ribbons were generated from a high-quality bulk wafer of GaAs with multiple epitaxial layers [Si-doped n-type GaAs(120 nm)/semi-insulating(SI)-GaAs(150 nm)/AlAs(200 nm)/SI-GaAs]. The carrier concentration of n-type GaAs is 41017 cm.sup.3. GaAs wafers with photoresist mask patterns were anisotropically etched in the etchant (4 mL H3PO4 (85 wt %), 52 mL H2O2 (30 wt %), and 48 mL deionized water). The AlAs layers were etched away with a diluted HF solution in ethanol (1:2 in volume). Layers of 2 nm Ti and 28 nm SiO2 were the deposited by e-beam evaporator. A PDMS stamp inked with the GaAs ribbons was then contacted to a layer of Si transistors coated with PI (thickness 1.5 m). Peeling back the PDMS and removing Ti and SiO.sub.2 by BOE etchant completed the transfer of GaAs to the device substrate. Metals (Ti/Au=5/70 nm) for the Schottky contacts were deposited by e-beam evaporation. Electrical interconnects between the GaAs back-to-back Schottky diodes and the Si MOSFET were defined by first patterning a layer of AZ4620 photoresist, then etching through the openings using CF.sub.4 and O.sub.2 plasma in a RIE system and then depositing a 300 nm of Al.

(189) Device characterization: A semiconductor parameter analyzer (Agilent, 4155C) and a conventional probing station were used for the electrical characterization of the diodes and transistors. The IR response was measured under IR LED source with wavelength of 850 nM.

(190) Circuit Simulation: To compare the measured transfer curve of the CMOS inverter with a simulation, level 2 PSPICE models for the n-channel Si MOSFET and the p-channel SWNT TFT were generated empirically. These PSPICE models were created based on the default PSPICE MOSFET model (MbreakN and MbreakP) with extracted parameters to fit the measured IV curves of both Si NMOS and SWNT PMOS shown in FIG. 65B. The PSPICE model for GaAs MSM photo-detector was created empirically using back-to-back schottky diodes connected in series with Si MOSFET.

(191) References for Example 4

(192) 1. K. Banerjee, S. J. Souri, P. Kapur, K. C. Saraswat, Proc. IEEE, 89, 602 (2001).

(193) 2. S. F. Al-Sarawi, D. Abbott, P. D. Franzon, IEEE Trans. Components, Packaging, and Manufacturing Technology, Part B, 21, 2 (1998).

(194) 3. A. S. Brown, W. A. Doolittle, N. M. Jokerst, S. Kang, S. Huang, S. W. Seo Materials Science and Engineering B 87, 317 (2001).

(195) 4. Y.-C. Tseng, P. Xuan, A. Javey, R. Malloy, Q. Wang, J. Bokor, H. Dai, Nano letters 4, 123 (2004).

(196) 5. C. Joachim, J. K. Gimzewski, A. Aviram, Nature 408, 541 (2000).

(197) 6. G. Roelkens et al. Optics Express 13, 10102 (2005).

(198) 7. D. B. Strukov, K. K. Likharev, Nanotechnology 16, 888 (2005).

(199) 8. K. Vanhollebeke, I. Moerman, P. Van Daele, P. Demeester, Prog. Cryst. Growth Charact. Mater. 41, 1 (2000).

(200) 9. H. Amano, N. Sawaki, I. Akasaki, Y. Toyoda, Appl. Phys. Lett. 48, 353 (1986).

(201) 10. T. Kuykendall, P. J. Pauzauskie, Y. Zhang, J. Goldberger, D. Sirbuly, J. Denlinger, P. Yang, Nature Materials 3, 524, (2004).

(202) 11. A. M. Morales, C. M. Lieber, Science 279, 208 (1998).

(203) 12. M. Law, D. J. Sirbuly, J. C. Johnson, J. Goldberger, R. J. Saykally, P. Yang, Science 305, 1269 (2004).

(204) 13. J. Kong, H. T. Soh, A. M. Cassell, C. F. Quate and H. Dai, Nature 395, 878 (1998)

(205) 14. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Science 306, 666 (2004).

(206) 15. Y. Huang, X. Duan, C. M. Lieber, Small 1, 1 (2005).

(207) 16. M. A. Meitl, Z. Zhu, V. Kumar, K. Lee, X. Feng, Y. Huang, R. G. Nuzzo, J. A. Rogers, Nature Materials 5, 33 (2006).

(208) 17. E. Menard, K. J. Lee, D. Y. Khang, R. G. Nuzzo, J. A. Rogers, Appl. Phys. Lett. 84, 5398 (2004).

(209) 18. Y. Sun, S. Kim, I. Adesida, J. A. Rogers, Appl. Phys. Lett. 87, 083501 (2005).

(210) 19. K. Lee, M. A. Meitl, V. Kumar, J.-H. Ahn, I. Adesida, J. A. Rogers, R. G. Nuzzo, Appl. Phys. Lett. accepted.

(211) 20. S.-H. Hur, D.-Y. Khang, C. Kocabas, J. A. Rogers, Appl. Phys. Lett. 85, 5730 (2004).

(212) 21. Materials and Methods are available as supporting material on Science Online.

(213) 22. J. Dong, M. A. Meitl, E. Menard, P. Ferreira and J. A. Rogers, unpublished.

(214) 23. S. Linder, H. Baltes, F. Gnaedinger, and E. Doering: Proc. IEEE Micro Eletro Mech. Systems 349, (1994).

(215) 24. J.-H. Ahn, H.-S. Kim, K. Lee, Z.-T. Zhu, E. Menard, R. G. Nuzzo, J. A. Rogers, IEEE Electron Devices Lett. 27, 460 (2006).

(216) 25. J.-H. Ahn, H.-S. Kim, K. Lee, Z.-T. Zhu, E. Menard, R. G. Nuzzo, J. A. Rogers, unpublished.

(217) 26. J. B. D. Soole, H. Schumacher, IEEE J. Quantum Electron. 27, 737 (1991).

(218) The pop up architecture is one that enables a range of device architectures and structures integrating structures that embed useful but difficult to achieve features. It is an architecture enabling important competencies devices that express electronic, optical, mechanical, and thermal forms of functionality. In many cases, the system designs exploit a hierarchy of such effects to enable explicit device level performance outcomes, although for simplicity we discuss specific embodiments below in terms of a dominant mode of functioning.

(219) Electronic Systems. The most direct form of utility in this sphere is the provisions the described architectures make for the design of complex mechanically compliant electronic devices that directly embed high performance electronic circuitsdisplays, sensing elements, RF-ID tags comprising some challenging forms of application that benefit from the integration of high performance electronic circuits within a flexible system level architecture. The designs disclosed herein significantly extend the full range of mechanical compliances that can be realized. It does so by enabling the provision, at the system design level, of specific architectural details that can extend range of mechanical deformations that can be toleratedwell beyond the general limit of 1% strain that is typical for a device based on a planar integration of components. The examples show a specific architecture for the simplest system elements, the interconnects, that can be used to withstand formal system high level strains (>30% in the form factor appropriate for the construction of bus lines and interconnects in a display) as well as providing for other more demanding forms of mechanical compliance (stretchability). These benefits can be extended as well to more complex device level components as illustrated by the form factor of the exemplary devices shown in FIG. 31, a GaAs MSM IR photodetector as described. Essentially every functional component of a complex electronic system can be integrated in a design-specific, mechanically compliant form using the methods taught herein.

(220) Optical Components and Systems. Optical components, e.g., a waveguide can respond with extreme sensitivity to flexure. The methods and systems provide new architectures for such devices that can both tolerate and, more importantly, exploit mechanical flexure to benefit functional performance. Examples of technologies that can directly exploit the methods disclosed herein include advanced forms of photonic components including, but not limited to, waveguide optical couplers and related forms of optical switches and limiters. Mechanical flexure at the system level of the integrated structure (via compression or extension) provides a direct means to effect these functionalities. The loss in a channel as well directly relates to the flexure of the waveguidehigh bending radii promoting leakage in a controllable way from core to sheath modes. Such effects can be directly exploited in a variety of devices. For example, FIG. 67 schematically illustrates a waveguide array produced via the controlled buckling of an optical microstructure partially adhered to a deformable substrate. FIG. 67A shows the optical device is produced by attaching a component 330 (e.g., waveguide such as an optical fiber or other elongated microstructure) to a substrate 30 by contact printing, for example. The attachment includes strongly bound contacting regions 310 and weakly bound regions corresponding to raised regions 320. Upon deformation the second electrode buckles and the weakly-bound region of the waveguide separates physically from the substrate, thereby producing the raised region. The device may operate simply as a waveguide capable of significant (5 to 50%) stretchability (see FIG. 67B). Alternatively, the indices of refraction of the waveguide and substrate as well as the buckling geometry may be chosen such that the device operates as an optical switch, allowing light to pass in the elongated state (FIG. 67B) but not in the shortened state (FIG. 67A), due to the high curvature in the buckled waveguides.

(221) Mechanically Functional Systems. The intersection between mechanics and electronics is fundamental for several critical classes of technologiesinertial and other forms of force sensors comprise specific examples that are both of current interest and finding broad use. The methods and systems disclosed herein provide a route to generating new forms of such devices. FIG. 68 is a representative example of a mechanical system, specifically an entwined multilayer architecture for capacitively coupled sensing. This exemplary architecture directly enables important forms of force related sensinginertial and pressure measurements most notably. In each case the methods and systems disclosed herein provide a relatively direct means to control many systems level aspects of the performance of these devicesdynamic range and region of optimum sensitivity most notablywhile enabling their integration into compact, novel form-factor systems (e.g. by allowing the integration of the electronic systems in new ways). These structures compliment established MEMS based approaches to devices of this type. Referring to FIG. 68, a mechanical device 400 (e.g., accelerometer/pressure sensor) is produced via the controlled buckling of a conductive microstructure partially adhered to a deformable substrate 30. This device architecture operates by monitoring changes in the capacitance between the bottom electrode 450 and the other electrode 440 that occur when the raised region 320 of the electrode 440 is displaced relative to the substrate via acceleration or pressure in the z-direction. The device 400 is produced by preparing an electrode (bottom electrode 450) on the substrate 30, then by attaching another electrode 440 by contact printing. The attachment includes strongly bound contacting regions 310 and weakly bound regions (e.g., in the region below 320). Upon deformation, the second electrode 440 buckles and the weakly-bound region separates physically from the substrate, thereby producing the raised region 320.

(222) Thermally Functional Devices. The pop up structures afforded by the present invention engender new capacities to provide for the thermal isolation of complex electronic components. An explicit device class provides a general design for the pixel elements of a long wavelength imaging system that requires the integration of high performance electronic components that provide control, read out, data handling and other capabilities for the system while providing direct integration and precise thermal isolation of thermally responsive (and for this example) two terminal devices. This demanding architecture is readily accessed using the methods taught by the current invention. In the present case it is possible to place functional electronic componentssuch as the AD converter needed to read a pixelin close proximity to the IR responsive elements (appropriate examples include but are not limited to Si and thin film multilayers of photoresistive metal oxides supported on Si.sub.3N.sub.4 membrane), a feature that makes it possible to both simplify design and enhance performance. Most notably, the systems and devices presented herein provides a capability of integrating of such device elements in a non-planar focal array. FIG. 69 shows a thermal device 500 (microbolometer) produced via the controlled buckling of a thermoresistive microstructure partially adhered to a deformable substrate. The device 500 is produced by attaching to the substrate 30 an electrode 550 that contains a thermoresistive material 560 by contact printing. The attachment includes strongly bound contacting regions 310 and weakly bound regions corresponding to raised regions 320. Upon deformation, the electrode 550 buckles and the weakly-bound region separates physically from the substrate, thereby producing the raised region 320 that is to a large extent thermally isolated from the substrate, thereby providing accurate and localized temperature sensing.

(223) U.S. patent application Ser. Nos. 11/115,954, 11/145,574, 11/145,542, 60/863,248, 11/465,317, 11/423,287, 11/423,192, and 11/421,654 are hereby incorporated by reference to the extent not inconsistent with the present description.

(224) All references throughout this application, for example patent documents including issued or granted patents or equivalents; patent application publications; and non-patent literature documents or other source material; are hereby incorporated by reference herein in their entireties, as though individually incorporated by reference, to the extent each reference is at least partially not inconsistent with the disclosure in this application (for example, a reference that is partially inconsistent is incorporated by reference except for the partially inconsistent portion of the reference).

(225) The terms and expressions which have been employed herein are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments, exemplary embodiments and optional features, modification and variation of the concepts herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention as defined by the appended claims. The specific embodiments provided herein are examples of useful embodiments of the present invention and it will be apparent to one skilled in the art that the present invention may be carried out using a large number of variations of the devices, device components, methods steps set forth in the present description. As will be obvious to one of skill in the art, methods and devices useful for the present methods can include a large number of optional composition and processing elements and steps.

(226) Every formulation or combination of components described or exemplified herein can be used to practice the invention, unless otherwise stated.

(227) Whenever a range is given in the specification, for example, a temperature range, a time range, or a composition or concentration range, all intermediate ranges and subranges, as well as all individual values included in the ranges given are intended to be included in the disclosure. It will be understood that any subranges or individual values in a range or subrange that are included in the description herein can be excluded from the claims herein.

(228) All patents and publications mentioned in the specification are indicative of the levels of skill of those skilled in the art to which the invention pertains. References cited herein are incorporated by reference herein in their entirety to indicate the state of the art as of their publication or filing date and it is intended that this information can be employed herein, if needed, to exclude specific embodiments that are in the prior art. For example, when composition of matter are claimed, it should be understood that compounds known and available in the art prior to Applicant's invention, including compounds for which an enabling disclosure is provided in the references cited herein, are not intended to be included in the composition of matter claims herein.

(229) As used herein, comprising is synonymous with including, containing, or characterized by, and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps. As used herein, consisting of excludes any element, step, or ingredient not specified in the claim element. As used herein, consisting essentially of does not exclude materials or steps that do not materially affect the basic and novel characteristics of the claim. In each instance herein any of the terms comprising, consisting essentially of and consisting of may be replaced with either of the other two terms. The invention illustratively described herein suitably may be practiced in the absence of any element or elements, limitation or limitations which is not specifically disclosed herein.

(230) One of ordinary skill in the art will appreciate that starting materials, biological materials, reagents, synthetic methods, purification methods, analytical methods, assay methods, and biological methods other than those specifically exemplified can be employed in the practice of the invention without resort to undue experimentation. All art-known functional equivalents, of any such materials and methods are intended to be included in this invention. The terms and expressions which have been employed are used as terms of description and not of limitation, and there is no intention that in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments and optional features, modification and variation of the concepts herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention as defined by the appended claims.

(231) TABLE-US-00001 TABLE 1 Parameters extracted (from experiments and calculations) from the buckles as shown in FIG. 31A. The calculations assume that the widths (i.e., 10 m for the samples shown in the figure) of the activated regions are the same before and after stretching. measured calculated calculated measured calculated amplitude amplitude peak strain pre- width width A.sub.m A.sub.cal .sub.peak strain (m) (m) (m) (m) (%) 11.3% 136.6 170.7 37.5 37.6 0.38 25.5% 139.6 151.4 51.5 50.3 0.65 33.7% 140.1 142.1 56.4 54.3 0.80 56.0% 124.3 121.8 63.6 60.4 1.2

(232) TABLE-US-00002 TABLE 2 Parameters extracted (from experiments and calculations) from the buckles as shown in FIG. 31D measured calculated measured calculated calculated wavelength wavelength amplitude amplitude peak strain W.sub.in .sub.m .sub.cal A.sub.m A.sub.cal .sub.peak (m) (m) (m) (m) (m) (%) 100 N/A 69 N/A 33.2 2.5 200 123 131 66.3 64.1 1.2 300 199 194 100.6 94.9 0.80 400 253 256 129.3 128.8 0.61