RELATED APPLICATIONS

This application claims priority to U.S. provisional application serial no. 61/172,632 filed April 24, 2009 to Baumgart et al, which is hereby incorporated by reference in its entirety.

BACKGROUND

Nanotubes exhibit novel physical properties and play an important role in fundamental research. In addition, nanotubes find many practical applications because of their restricted size and high surface area. See R. Kelsall et al., Nanoscale Science and Technology, Wiley, Chichester, (2006); C. R. Martin, Ace. Chem, Mater. 28, 61 (1995); J. Goldberger et al., Nature, 422 599 (2003); and S. B. Lee et al., Science, 296, 2198 (2002). Nanotubes may be formed from a variety of materials, including metal oxides. In particular, hafnium oxide (hafnia, HfO ₂ ), aluminum oxide ( alumina, Al ₂ O ₃ ), titanium oxide (TiO ₂ ) and zirconium oxide (zirconia, ZrO ₂ ) are important materials widely used in ceramics, chemical sensors, catalysts, opto-electronics, and as high-k dielectrics in microelectronics. The semiconductor, zinc oxide (ZnO), is also used in chemical sensors. See G. D. WiIk, et al., J. Appl. Phys., 89, 10 (2001). However, a need exists for nanotubes formed from metal oxides and other materials that have more complex structures, higher aspect ratios, and higher surface areas. In addition to nanotubes, macroporous and nanoporous materials are also an important aspect of nanotechnology and can be used, in some cases, to build nanostructures such as nanotubes. Macroporous and nanoporous materials can be used in a variety of applications including, for example, electroosmotic (EO) pump membranes. Electroosmotic flows are preferred in microfluidic systems, since they enable fluid pumping and flow control without the need for mechanical pumps or valves, and they also can minimize sample dispersion effects. However, the nanoscale size of the pores can generate difficult challenges in providing useful materials and devices from these materials. Hence, a need also exists to develop better devices such as, for example, electroosmotic pump systems which possess improved flow and electrical properties resulting from nanoscale engineering. Examples of drawbacks of many existing EO pumps have been the need for large operating voltages (e.g., on the order of 1 kV to 10 kV), electrolysis of water, oxidation of electrode surfaces and Joule heating. Especially the need for a high voltage supply can limit the use of EO pumps in lab-on-a-chip (LoC) type portable devices.

SUMMARY

Provided herein are, among other things, devices, novel design of devices, methods of making devices, and methods of using devices. One preferred example of a device is an electroosmotic pump and elements associated with pumps including pump membranes.

In one embodiment, an electroosmotic pump may include a macroporous or nanoporous membrane having one or more micropores or nanopores, a layer of a first material deposited on an inner surface of the micropore or nanopore, a first electrode (e.g., anode) coupled to a first side of the nanoporous substrate, and a second electrode (e.g., cathode) coupled to a second side of the macroporous or nanoporous template. In another embodiment, the electroosmotic pump may further include a layer of a second material deposited on the layer of the first material. A variety of nanoporous substrates and compositions for the first material, the second material, and the electrodes may be used. Specific examples are provided herein. The performance of the disclosed electroosmotic pumps is superior to microscale EO pumps or conventional EO pumps. In some embodiments, the electroosmotic pumps provide a flowrate per unit area and voltage (Q _ma χ/A/V) value of at least 0.3 mL/minVcm or of at least 1 mL/minVcm at low DC voltages of 1-5V. In contrast to prior art electroosmotic pumps, which require voltages exceeding 100 V, the disclosed electroosmostic pump using passive or active Zeta-Potential control operates at very low dc voltages of IV - 5 V and thus enables a mobile battery powered hand-held electroosmotic device for various medical applications.

In particular, one embodiment provides an electroosmotic pump comprising: at least one macroporous substrate or a nanoporous substrate comprising micropores or nanopores, at least one first material disposed on the inner surface of the micropores or nanopores, wherein the first material is electrically conductive, at least one second material disposed on the first material, wherein the second material is an electrical insulator, at least one anode and at least one cathode adapted to generate an electroosmotic flow in the micropores or nanopores.

Another embodiment provides a method of preparing an electroosmotic pump device comprising: a macroporous or nanoporous substrate having one or more micropores or nanopores; a layer of a first material deposited on an inner surface of the micropores or nanopores; a first electrode coupled to a first side of the macroporous or nanoporous substrate; and a second electrode coupled to a second side of the macroporous or nanoporous substrate, the method comprising the step of depositing the first material by atomic layer deposition (ALD) on the nanopores.

Another embodiment provides a method of preparing an electroosmotic pump device comprising: at least one macroporous or nanoporous substrate comprising nanopores, at least one first material disposed on the inner surface of the micropores or nanopores, wherein the first material is electrically conductive, at least one second material disposed on the first material, wherein the second material is an electrical insulator, at least one anode and at least one cathode adapted to generate an electroosmotic flow in the micropores or nanopores, the method comprising the step of depositing by atomic layer deposition (ALD) the first and second materials in the micropores or nanopores.

Another embodiment provides a membrane comprising: at least one macroporous membrane or nanoporous membrane comprising micropores or nanopores, the membrane comprising at least one first material disposed in the micropores or nanopores which is an electrical conductor, and at least one second material disposed on the first material which is an electrical insulator.

Another embodiment provides a method for using an electroosmotic pump comprising: providing an electroosmotic pump comprising: at least one macroporous substrate or nanoporous substrate comprising micropores or nanopores, at least one first material disposed on the inner surface of the micropores or nanopores, wherein the first material is electrically conductive, at least one second material disposed on the first material, wherein the second material is an electrical insulator, at least one anode and at least one cathode adapted to generate an electroosmotic flow in the micropores or nanopores; applying a first voltage across the pump from the anode and cathode to generate electroosmotic flow; applying a second voltage to the first material, independently biased from the first voltage to modify the electroosmotic flow.

Another embodiment provides an electroosmotic pump comprising: a nanoporous substrate having one or more nanopores; a layer of a first material deposited on an inner surface of the nanopore; a first electrode coupled to a first side of the nanoporous substrate; and a second electrode coupled to a second side of the nanoporous substrate.

At least one advantage for at least one of the electroosmotic pump embodiments is high flow rate coupled with lower voltage and power needed to drive the pump. This can enable a battery operated hand held device.

At least one additional advantage for at least one embodiment is ability to control zeta potential, and adapt zeta potential for different applications.

At least one additional advantage for at least one embodiment is a robust, miniaturized EO pump.

In addition, provided herein are coaxial nanostructures, methods for making the coaxial nanostructures, and devices incorporating the coaxial nanostructures. The disclosed coaxial nanostructures can have extremely high aspect ratios and surface areas. Consequently, devices incorporating these coaxial nanostructures exhibit superior properties as compared with conventional devices. These advantages are further discussed below with respect to specific devices incorporating the coaxial nanostructures. The disclosed coaxial nanostructures may be formed using atomic layer deposition (ALD) or other suitable chemical vapor deposition (CVD) techniques to deposit different materials by coating the inner walls of the pores of various nanoporous substrates (also referred to herein as nanoporous templates or nanoporous membranes), one atomic layer at a time. Nanoporous substrates or templates maybe formed from nanoporous alumina, polycarbonate membranes, porous silicon, or any other suitable porous substrate. The ability to achieve coaxial nanostructures with such high aspect ratios is derived, in part, from the inventor's use of long ALD deposition dwell times. The use of long ALD deposition dwell times is contrary to conventional wisdom, since longer ALD deposition times can clog the pores of the underlying porous substrates.

In one aspect, coaxial nano structures are provided. In one embodiment, the coaxial nanostructure may include an inner nanostructure, a first outer nanotube disposed around the inner nanostructure, and a first annular channel between the inner nanostructure and the first outer nanotube. In another embodiment, the coaxial nanostructure may further include a second outer nanotube disposed around the first outer nanotube and a second annular channel between the first outer nanotube and the second outer nanotube. In other embodiments, a third outer nanotube may be disposed around the second outer nanotube, a fourth outer nanotube may be disposed around the third outer nanotube, and so forth, up to an n ^th outer nanotube. The aspect ratio of the coaxial nanostructures may range from about 5 to about 1,200, or about 300 to about 1200, although other aspect ratios are possible.

The materials used to form the inner nanostructure and the outer nanotubes may vary and may include a conductor, an insulator, or a semiconductor. Specific examples of conductors, insulators, and semiconductors are provided herein. A sacrificial material, including Al ₂ O ₃ , may be disposed within the annular channel of any of the coaxial nanostructures.

The coaxial nanostructures may be coupled to other components, including various substrates. In some embodiments, the substrate may be a porous anodic aluminum oxide (AAO) substrate. In other embodiments, a porous silicon substrate or any other suitable porous template may be used. Also provided herein are arrays comprising two or more of any of the disclosed coaxial nanostructures and devices incorporating any of the disclosed coaxial nanostructures.

In another aspect, methods for making the coaxial nanostructures are provided. In one embodiment, the method may include forming a layer of a first material on an inner surface of a nanopore of a nanoporous substrate using atomic layer deposition, forming a first layer of a sacrificial material on the layer of the first material using atomic layer deposition, and forming a layer of a second material on the first layer of the sacrificial material using atomic layer deposition. In another embodiment, the method may further include forming a second layer of a sacrificial material on the layer of the second material and forming a layer of a third material on the second layer of the sacrificial material (until the n ^th layer in the most general case). The aspect ratio of the coaxial nano structures provided by such a method may range from about 5 to about 1,200, or about 300 to about 1200, although other aspect ratios are possible. The methods may further include removing the nanoporous template and/or the layers of the sacrificial material by a variety of methods, including by chemical etching. The nanoporous substrates and the compositions of the first material, the second material, the third material, the n' material and the sacrificial material may vary as described above with respect to the coaxial nano structures.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 illustrates: (a) a general side view of a two terminal or three terminal EO pump, (b) an illustration for the three terminal EO pump, and (c) photograph for an EO pump comprising alumina membranes covered with external gold electrodes.

Figure 2 is a schematic view of the experimental setup for characterization of the performance of the disclosed electroosmotic pumps.

Figure 3 is an illustration of an exemplary structure that may be used to provide a two- terminal passive Zeta Potential controlled electroosmotic pump.

Figure 4 is an illustration of an exemplary structure that may be used to provide a three- terminal active Zeta Potential controlled electroosmotic pump.

Figure 5: (a) Sealed electro osmotic pump used for flow rate measurements, (b) Flow rates of the two terminal electroosmotic pump as a function of the applied voltage under appropriate pressure head.

Figures 6A-B show sheet conductance of an Pt film as a function of growth temperature for 400 ALD cycles.

Figures 7A-B show electrical properties of Pt films. FIG. 7 A shows sheet conductance vs. ALD cycles. FIG. 7B shows resistivity vs. thickness. Figures 8A-E show deposition of Pt by ALD at 300 C for different thickness stage: (a) 50 cycles, (b) 100 cycles, (c) 200 cycles, (d) 400 cycles and (e) 1000 cycles.

Figures 9A-C show cross-sectional SEM images of Pt deposition into an AAO membrane. Fig. 9A show Pt deposition depth with no exposure time. FIG. 9B shows deposition depth after 10 sec exposure time. FIG. 9C shows deposition depth 30 sec exposure time for the Pt precursor.

Figure 10. Platinum tubes fabricated using 30 sec exposure time for the Pt precursor. The insets are TEM images for the Pt tube structure at the AAO surface and deep into the AAO template.

Figures 11 A-B provide an overview of zeta potential measurements of various dielectric film samples. FIG. 1 IA shows a 3-D schematic diagram of the measurement cell containing the sample coated with an ALD high-k porous dielectric thin film where zeta potential measurements can be carried out. FIG 1 IB illustrates particle mobility in nanochannels of the porous films.

Figures 12A-D show zeta potential measurements of thin dielectric films as a function of solution pH.

Figures 13A-B show zeta potential at pH of 7 ALD films of alumina at Fig. 13A, and silica at Fig. 13B

FIGs. 14A and IB show SEM images of a porous anodic aluminum oxide (AAO) substrate. FIG. 14A shows the surface of the substrate after ion milling. FIG. 14B shows a cross- section of a cleaved AAO substrate.

FIG. 15 shows a cross-sectional SEM image of ALD (atomic layer deposited) zirconia coated AAO substrate (A) and a corresponding EDS Zr mapping showing uniform distribution of zirconia throughout the entire thickness of the 60μm AAO substrate (B).

FIG. 16 shows a top-down SEM image of an uncoated AAO substrate (A); the same AAO substrate with a thin film ALD coating of ZrO ₂ (B); and the same coated AAO substrate after the alumina substrate walls have been removed to provide single ZrO ₂ nanotubes (C). FIG. 17 shows a SEM micrograph Of HfO ₂ tube-in-tube coaxial nano structures. A top-down view of the sample surface and a partial side-view from a cleavage site is shown by tilting the sample.

FIG. 18 is a TEM micrograph of a separated HfO ₂ tube-in-tube coaxial nanostructure shown in FIG. 17.

FIG. 19 is a top-down SEM image showing three concentric metal oxide (ZrO2) nanotubes inside large AAO pores following the dissolution of the 2 separating spacer Al ₂ O ₃ layers in order to expose the sidewalls of the coaxial (ZrO2) nanotubes. These five coaxial nanostructures were formed by layering ZrO ₂ /Al ₂ O ₃ /ZrO ₂ /Al ₂ O ₃ /ZrO ₂ and removing the Al ₂ O ₃ layers by chemical etching.

FIG. 20 is a top-down SEM image of hafnia/zirconia coaxial nanostructures disposed in the nanopores of a AAO substrate showing the simultaneous coating of the AAO surface and the inner walls of the nanopores.

FIG. 21 is an illustration of an exemplary coaxial nanostructure having an inner nanotube of ZnO, a first outer nanotube of ZrO ₂ disposed around the inner nanotube, and a first annular channel between the inner nanotube and the outer nanotube.

FIG. 22 is an illustration of an exemplary structure that may be used to provide a chemical sensor.

FIG. 23 illustrates an embodiment for an adaptive microfluidic mirror array using EO pump.

FIG. 24 illustrates other uses of an EO pump.

DETAILED DESCRIPTION

PART I. INTRODUCTION - ELECTROOSMOTIC PUMPS

Additional embodiments and descriptions may be found in co-pending application no. _____ filed on April 23, 2010 (Multiple Walled Nested Coaxial Nanostructures",

Baumgart et al.), and in a conference presentation, Gu, et al, "Zeta Potential of ALD Metal Oxide Films for Microfluidic Applications," presented at American Vacuum Society (AVS) 9 ^th International Conference on Atomic Layer Deposition (ALD 2009), Monterey, CA, July (2009), both of which are hereby incorporated by reference in their entireties.

Electroosmotic pumps and materials which can be used in (and methods for making) an EO pump are known in the art as described in, for example, US Patent No. 7,316,543; 7,540,717; 7,458,783; US Pat. Pub. 2009/0061601 ; Gupta et al, "Localized, low- voltage electro-osmotic pumping across nanoporous membranes", App. Phys. Lett. 91 , 094101-1 (2007) and U.S. Patent No. 7,458,783 to Myers et al.

By way of introduction, Figure 1 is described illustrating several embodiments for which additional description is provided furthermore hereinafter. The left illustration in Figure 1 provides a schematic diagram of an EO pump which can be, as described herein and illustrated more in examples 2 and 3, a 2-electrode or 3-electrode macro- or nano-pore EO pump configuration. An external electric field can be applied through the 2-terminal gold or Pt electrodes deposited on the porous membrane surface without plugging the nano-pores. This membrane can be encased in a channel to form an EO-flow setup. At least one novel aspect of EO-pumps described herein is the ability to control the surface zeta potential by embedded internal Pt electrodes. For the center illustration of Figure 1 , schematics are provided of a nano-pore cross-section. Using ALD, the nanoporous membrane, e.g., Al ₂ O ₃ or silicon membrane, can be covered with a thin conducting layer of Pt (or, for example, TiN or other metals or conducting coatings) that will be subjected to a DC bias to vary the pore- surface zeta potential. This conducting metal layer can be electrically isolated by a thin second ALD insulating oxide layer, which could be silica, hafnia, zirconia, or alumina, or any other electrically insulating material. The porous membrane surface will receive Au or Pt (or any other stable noble metal) anode and cathode contacts for the application of an external electric field parallel to the nano-pores. This novel design can either use the zeta potentials of the ALD oxide layer, or could have an adjustable zeta potential based on the DC bias imposed using the inner electrodes. Hence, one can achieve active control of surface zeta potential using embedded electrode structures within the nano-pores. Data measurements were obtained for Al ₂ O ₃ membranes covered with external gold electrodes. Right — (See also Figure 2 for testing schematic). Other embodiments are described below. Electroosmosis is the motion of ionized liquid relative to a stationary charged surface by an externally applied electric field. Electroosmotic (EO) flows are useful in microfluidic systems, since they enable fluid pumping and flow control without the need for mechanical pumps or valves, and they also minimize the sample dispersion effects. See Karniadakis, G. E., Beskok, A., and Aluru, N., Microflows and Nano flows: Fundamentals and Simulation, Springer, New York, 2005. However, conventional EO pumps suffer from a number of drawbacks, including the need for large operating voltages (on the order of 1 kV to 10 kV), electrolysis of water, oxidation of electrode surfaces, and Joule heating. The need for a high voltage supply limits the use of conventional EO pumps in lab-on-a-chip (LoC) type portable devices, designed for bio-medical, pharmaceutical, environmental monitoring and homeland- security applications.

Electroosmotic flows are generated by interaction of the external electric field with the mobile charges in the electric double layer (EDL). Polarity of the zeta potential on the surface dictates the flow direction. Depending on the ionic concentration, the EDL thickness varies from 3 nm for I x IO ^"2 M electrolyte to 300 nm for deionized water (I xIO ^"6 M). Channel hydraulic diameter and EDL thickness may become comparable in nano-pores, depending on the pore size and the EDL thickness. See Dutta, P., and Beskok, A., "Analytic Solution of Combined Electroosmotic/Pressure Driven Flows in Two- Dimensional Straight Channels: Finite Debye Layer Effects," Analytical Chemistry, 73(9); 1979-1986, 2001. If large ionic concentration buffers that result in EDL thicknesses on the order of a few nanometers are used, as well as nanopore diameters on the order of 200 nm and above, the EDL constitutes a very small portion of the flow domain. This simplification is convenient, since one does not need to consider solution of the Poisson-Boltzmann equation that governs the ion distribution near the surfaces and the corresponding EDL effects on momentum transport. Under such "simplified conditions," fluid velocity in vicinity of the charged surface can be modeled by the Helmholtz-Smoluchowski electroosmotic velocity,

U _HS = - klE*. (Equation 1 ) μ where ζ is the zeta potential, e is permittivity of the liquid, μ is the dynamic viscosity, and E _x is the electric field applied in the stream wise direction. See Karniadakis, G.E., Beskok, A., and Aluru, N., Microflows and N ano flows: Fundamentals and Simulation, Springer, New York, 2005. For steady electric field and constant channel cross-section, this equation also models the EO-flow velocity in the channel/pore, which happens to be a plug type flow. Without considering the EDL effects and/or upstream/downstream pressure head, the flowrate in a pore will be Q=U _HS ^X A, where A is the pore area. This relation makes clear that it is possible to maximize the flow rate within a pore either by increasing the zeta potential or the applied electric field.

The zeta potential varies as a function of the buffer, its ionic strength and pH, as well as the surface characteristics. The magnitude of zeta potential for aluminum oxide in contact with 1 mM KCl pH=7 is 37 mV. See Simonnet, C. and Groisman, A., "Chaotic Mixing in a Steady Flow in a MicroChannel, " Physical Review Letters, 94(13); 134501-1-134501-4, 2005. The zeta potential for silica, zinc oxide, and zirconia is |f|=80 mV; 45 mV; and 90 mV, respectively. See Degen, A. and Kosec, M., "Effects of pH and Impurities on the Surface Charge of Zinc Oxide in Aqueous Solution," Journal of European Ceramic Society, 20(6); 667-673, 2000; and De Bellaistre, M.C., Renaud, L., Kleimann, P., Morin, P., Randon, J., and Rocca, J-L., "Streaming Current Measurements in Zirconia-Coated Capillaries," Electrophoresis, 25(18-19); 3086-3091, 2004. Thus, it is possible to double the flowrate by selecting an appropriate surface coating material such as silica. In addition to such "passive zeta potential control," it is also possible to "actively control the zeta potential" by embedding thin electrode layers behind an insulating surface oxide layer.

As noted above with respect to conventional EO pumps, large electric fields are used to maximize the EO flow, which results in electrolysis of water, electrode oxidation and Joule heating. The inventors have discovered that these adverse effects may be avoided or eliminated by locally applying the electric fields across a thin membrane, and by altering the surface zeta potential both by surface coatings and the "active zeta potential control."

POROUS SUBSTRATES

Porous substrates described herein can be macroporous, microporous, or nanoporous. The pores can be micropores or nanopores. The average pore size can be, for example, less than about 10 microns, or less than about 5 microns, or less than about one micron. The average pore size can be, for example, about 200 nm to about 10 microns, or about 200 nm to about 5 microns, or about 200 nm to about three microns. Substrates can be made of inorganic materials such as, for example, silicon or alumina.

TWO TERMINAL EMBODIMENTS

In one embodiment, a two-terminal electroosmotic pump comprises a macroporous or nanoporous substrate having one or more nanopores and a layer of a first material deposited on an inner surface of the micropore or nanopore. The layer of the first material provides a nanotube disposed within the micropore or nanopore of the macroporous or nanoporous substrate. Electrodes may be coupled to both sides of the macroporous or nanoporous substrate. Any of the macroporous or nanoporous substrates described herein may be used.

In some embodiments, the aspect ratio of the nanopores of the macroporous or nanoporous substrate ranges from about 5 to about 1,200, or about 300 to about 1200. This includes embodiments in which the aspect ratio is about, for example, 50, 100, 200, 300, 400, 500, 600, 700, 800, 900, or 1000. The composition of the first material may vary. In some embodiments, the first material comprises an oxide or metal oxide or a metal nitride. The first material can be an insulator. Any of the oxides or metal oxides or nitrides disclosed above may be used, particularly if an insulator, including, but not limited to HfO ₂ , ZrO ₂ , Al ₂ O ₃ , or SiO ₂ .

Similarly, the composition of the electrodes may vary. In some embodiments, the electrodes comprise a metal or metal nitride. Examples of useful metals, include, but are not limited to, Au, Pt, and W (and in general, any stable noble metal).

In FIG. 3, a layer of a dielectric material is deposited on the inner surfaces of the nanopores of an AAO substrate. A layer of metal may be further deposited on the top surface and the bottom surface of the AAO substrate. Such metal layers may serve as cathode and anode electrodes of the EO pump. The structure shown in FIG. 3 may be used as a two- terminal electroosmotic pump. THREE TERMINAL EMBODIMENTS

In another embodiment, a three-terminal electroosmotic pump comprises a macroporous or nanoporous substrate having one or more nanopores, a layer of a first material deposited on an inner surface of the micropore or nanopore, and a layer of a second material deposited on the layer of the first material. The layer of the first material can provide an outer tube, including a nanotube, and the second material provides an inner tube, including a nanotube, resulting in a coaxial structure, including a nano structure, disposed within the nanopore of the nanoporous substrate (or nanopores of macroporous substrate). Electrodes may be coupled to both sides of the macroporous or nanoporous substrate. Any of the macroporous or nanoporous substrates described herein may be used. In some embodiments, the aspect ratio of the nanopores of the macroporous or nanoporous substrate ranges from about 5 to about 1,200, or about 300 to about 1200. This includes embodiments in which the aspect ratio is about 400, 500, 600, 700, 800, 900, or 1000.

In FIG. 4, a layer of a metal is deposited on the inner surfaces of the nanopores of an AAO substrate. Next, a layer of a dielectric material is deposited on the layer of metal or conducting metal nitrides or any other electrically conducting coating. Finally, another layer of metal may be further deposited on the top surface and the bottom surface of the AAO substrate to serve as cathode and anode. The structure shown in FIG. 4 may be used as a three-terminal electroosmotic pump.

FIRST MATERIAL, SECOND MATERIAL, AND ELECTRODES

The composition of the first material, the second material, and the electrodes may vary. In one embodiment, the first material and/or the second material are not materials prepared by sol-gel methods. For example, silica can be used which is not prepared by sol- gel methods.

In some embodiments, the first material comprises a metal, a conducting metal nitride, or a semiconductor, or a conducting metal oxide. Non-limiting examples of metals and metal nitrides include Ti, Au, Pt, Al, Cu, Ag, and metal nitrides, such as TiN, thereof or any other stable noble metal. A non-limiting example of a semiconductor includes ZnO.

In some embodiments, the second insulating material comprises a semiconductor oxide or an oxide or insulating metal oxide. Non-limiting examples of semiconductor oxides or metal oxides include HfO ₂ , ZrO ₂ , Al ₂ O ₃ , TiO ₂ , and SiO ₂ or any other electricially insulating material.

Other possible metals, metal nitrides, semiconductors, semiconductor oxides, or metal oxides include, but are not limited to, those described above.

In some embodiments, the electrodes comprise a metal. Examples of useful metals, include, but are not limited to, Au, Pt, and W, and any other stable noble metal.

METHODS OF FABRICATION

The methods for forming these and other electroosmotic pumps is similar to the methods described herein, involving the use of atomic layer deposition to deposit the desired number of layers of materials in macroporous or nanoporous substrates.

Methods for depositing electrodes and patterning contacts on the electroosmotic pumps using photolithography or wire bonding techniques are known.

For the three terminal embodiment, process steps for fabrication can comprise: First, coat the inside walls of the nanoporous template (e.g., nanopore silicon membranes, or AAO) with an electrically conducting ALD film (e.g., Pt film) of desired thickness, e.g., about 10 nm. Optimize the ALD process window to ensure complete penetration and saturation of the AAO template with the chemical ALD precursors so that the nanochannels are coated from top to bottom throughout the thickness of the porous alumina membrane or porous Silicon template.

Second, fabricate an external third contact to conducting ALD Pt film lining the pores by photolithography and wire bonding techniques to achieve dielectric insulation from the cathode and anode contacts. Third, use ALD technology to grow selected dielectric films on top of the metal layer in a conformal nested coaxial composite configuration for electrical insulation and additional passive zeta control.

Fourth, define by photolithographic patterning the area of external cathode and anode metal contacts on the nanoporous template surface excluding the dielectrically isolated 3 ^rd contact to Pt coating of the inner pore walls.

Finally, an E-beam evaporation or alternative sputter process can be carried out for initially disposing or depositing metal (e.g. Au, Pt, or any noble metal) electrodes as anode and cathode on both sides of the template (e.g., AAO or macroporous silicon) and in the second phase Pt electrodes. The metal cathode and anode electrodes can be adapted to only cover the surface of the thin AAO template walls and not obstruct the openings of the nano- pores.

Other materials can be used in these steps as described herein.

PERFORMANCE; FLOW RATES

The performance of the disclosed two-terminal electroosmotic pump can exceed that of conventional electroosmotic pumps. A measure of pump performance is flowrate per unit area and voltage (Q _max /A/V). As further discussed below, it is possible to both theoretically calculate and experimentally determine the Q _ma χ/A/V value for a particular pump. In some embodiments, the disclosed electroosmotic pump provides a Q _max /A/V value of at least 0.3 mL/minVcm ² . This includes embodiments in which the Q _max /A/V value is at least 0.4 mL/minVcm ² , 0.5 mL/minVcm ² , 0.6 mL/minVcm ² , or 0.7 mL/minVcm ² .

Similar to the two-terminal electroosmotic pumps described above, the performance of the three-terminal electroosmotic pumps also can exceed that of conventional electroosmotic pumps. In some embodiments, the disclosed electroosmotic pump provides a Q _ma χ/A/V value of at least 0.8 mL/minVcm ² . This includes embodiments in which the Q _maX /A/V value is at least 0.9 mL/minVcm ² , 1 mL/minVcm ² , 1.1 mL/minVcm ² , or 1.2 mL/minVcnT.

The pump can be operated at voltage of five volts or less, including IV to 5V. A description of methods and calculations for evaluating the performance of the disclosed electroosmotic pumps is provided below. In providing an EO pump, materials can be selected to be biocompatible and satisfy other commercialization requirements. Several embodiments can be described further for which other materials can be used and adapted for a particular application. Flow chambers may be built by sandwiching a single as well as multiple nanoporous substrates within a PDMS microchannel and a millimeter-scale plexiglass (PMMA) channel. The PDMS microchannel may be cured while inserting a spacer (e.g., razor-blade) in place of the nanoporous substrate. After curing the PDMS, the spacer (e.g., razor blade) may be removed and the nanoporous substrate installed. The integrated system may be sealed using an adhesive. For the meso-scale PMMA channel, all structural parts may be machined using conventional CNC machining (e.g., Roland PNC-300 CNC Milling Machine), which allows 1 μm machining tolerance. In one design, space will be created for each nanoporous substrate within the two pieces of the PMMA channels. Thin PDMS spacers or plastic o-rings may be used as a bushing material for bolting the nanoporous substrates between the two sides of the PMMA channels.

An example of an EO pump is shown in Figure 2. In order to characterize the performance of the EO-pumps, the experimental setup 800 schematically shown in FIG. 2 may be used. Two reservoirs 804 are connected upstream and downstream of the pump 808 for storing and collecting the liquid, respectively. An electric field is supplied by a DC power supply 812, and the resultant electrical current is measured using a multimeter 816. Flow rate is determined by monitoring the weight of inlet reservoir using a precision scale (ACCULAB ATILON, ATL- 124-1). Pressure variations at the inlet and outlet of the pump are measured using two pressure transducers 818 (OMEGA, PX303015G5V). The maximum pressure build-up across the pump is determined during the operation by closing the outlet valve 822. Using the flowrate predictions that are detailed below, the deflection of the free surface in the reservoirs was calculated, as well as the hydrostatic pressure difference created by this. Calculations have shown that change of the fluid height in reservoirs due to running EO flow for one hour will be limited to 0.4 mm when cylindrical reservoirs of 10 cm in diameter are used. This will induce a pressure driven back- flow on the system on the order of 0.4 μl/min, which is three orders of magnitude smaller than the predicted maximum EO flowrate in the system. Therefore, the pressure driven back- flow in the maximum flowrate experiments will be negligible. Experiments may utilize various buffered aqueous electrolytes prepared by using deionized (DI) water or DI water itself. A Delsa-Nano system may be used to measure the zeta potential of alumina, silica, hafnia, zirconia, and zinc oxide in various buffers as a function of the buffer ionic concentration and pH.

The EO-pump performance may be evaluated based on the flow rate and thermodynamic efficiency, as outlined in Chen, et al., "Low-Voltage Electroosmotic Pumping Using Porous Anodic Alumina Membranes," Microfluid Nanofluid, 5(2); 235-244, 2008. The thermodynamic efficiency (ϊjeff) is given by

l]eff= η^y (Equation 2)

where Fis the applied voltage, and /is the electric current, AP is the pressure change across the membrane and Q is the volumetric flowrate. The numerator of this equation is the power delivered by the pump, while the denominator is the electrical power input to drive the EO- Pump. For a fixed pressure load, it is desirable to have the maximum flowrate per given electrical power.

In order to obtain the total mass flow rate from the nanoporous substrate, the formula derived by Yao, et al., "Porous Glass Electroosmotic Pumps: Design and Experiments," Journal of Colloid and Interface Science, 268(1); 143-153, 2003 and Yao, S. and Santiago, J. G., "Porous Glass Electroosmotic Pumps: Theory," Journal of Colloid and Interface Science, 268(1); 133-142, 2003 may be used:

Qmax = — — = ψAδx U _HS , (Equation 3 ) μ

where is the porosity, A is the membrane area, and δ is a correction parameter to account for the EDL displacement thickness. See Dutta, P., and Beskok, A., "Analytic Solution of Combined Electroosmotic/Pressure Driven Flows in Two- Dimensional Straight Channels: Finite Debye Layer Effects," Analytical Chemistry, 73(9); 1979-1986, 2001. Using Equation 1, the maximum flowrate is related to the Helmholtz-Smoluchowski EO velocity UHS and ψ*A is the net flow area, while δ is the correction for finite EDL effects on the flowrate,

-2 which is expected to be δ =0.9 for a 1 x 10 M electrolyte. The axial electric field Ex is predicted by dividing the potential drop across the membrane ( Veff) with the membrane thickness, L. Using a porosity value typical value), zeta potential of 25 mV, nanoporous substrate area of 0.78 c ) and a target potential difference of 1 V across the L=50 μm thick nanoporous substrate, a maximum flowrate of 50 μL/min is calculated. Using Equation 3, the maximum flowrate increases linearly with membrane porosity, zeta potential and the applied electric field. Porosities of about and zeta potential of f= 80 mV using ALD covered silica nanoporous substrates will provide a 10 fold increase in the flowrate, resulting in Qιnax=0.5 ml/min. The maximum value oϊQmaxlAJV

2 reported for conventional EO pumps was about 0.15 ml/(minV cm ). See Tripp, J. A., Svec, F., Fre'chet, J.M.J., Zeng, S., Mikkelsen, J.C., and Santiago, J.G., "High-Ressure Electroosmotic Pumps based on Porous Polymer Monoliths," Sensors and Actuators B Chemical, 99(1); 66-73, 2004; Vajandar, S.K., Xu, D., Markov, D.A., Wikswo, J.P., Hofmeister, W., and Li, D., "SiO2-Coated Porous Anodic Alumina Membranes for High Flow Rate Electroosmotic Pumping," Nanotechnology, 18(27); 275705.1-275705.8, 2007, and Chen, Y.F., Li, M.C., Hu, Y.H., Chang, W.J., and Wang, CC, "Low-Voltage Electroosmotic Pumping Using Porous Anodic Alumina Membranes," Microfluid Nanofluid, 5(2); 235-244, 2008. Performance of the prototype, 2-terminal EO-pump shown in the

2

Examples below is Qmax/A/V= 0.3 ml/(min V cm ), twice the maximum value for the conventional pumps.

2

The calculations above predict a Qmax/A/V value of 0.63 ml/min/V/cm for a two- terminal EO pump based on silica coated nanoporous substrates — a four-fold increase in the pump performance over conventional EO pumps. Calculations show that the performance of the three-terminal EO pumps will provide an order of magnitude increase in the maximum Qmax/A/V value achievable with conventional EO pumps. At the same time, the disclosed EO pumps use a minimal applied electric potential to reduce the electrolysis, electrode oxidation and Joule heating effects, and increase the thermodynamic efficiency. If a nanoporous silica membrane is used, then some advantages can include (a) a convenient short thermal oxidation process (by rapid thermal annealing, RTA) to oxidize the inner pore walls of the porous silicon template into SiO ₂ (silica) as a high zeta potential insulating material in and/or coating the pores, (b) relatively structurally reliable, (c) availability of relatively large membrane surface area, e.g., 4-6 inch diameter, providing more surface for more flowrate. One can pack this to even larger scale membranes.

Flow can be modified. In some embodiments, the modification of electroosmotic flow is an increase in flow rate, a decrease in flow rate, or a reversal of flow.

APPLICATIONS

There are a variety of applications for EO pumps. The core structure for the membrane and electrodes can be adapted to function with other pump components such as, for example, fluid chambers, inlet port(s), and outlet port(s), as known in the art.

These applications include, for example, lab-on-a-chip devices and applications, inkjet printing, ink delivery, drug delivery, liquid drug delivery, chemical analysis, chemical synthesis, proteomics, healthcare related applications, defense and public safety applications; medical applications, pharmaceutical or biotech research applications, environmental monitoring and defense applications, in vitro diagnostic and point-of-care applications, and medical devices. EO pumps of the embodiments disclosed herein may also be incorporated into an inkjet printing device. Other application include PCR (DNA amplification, including real time PCR on a chip), electronic cooling (e.g., for microelectronics), using EO pumps as valves by opposing pressure driven flow, using EO pumps to fill and empty flexible reservoirs to induce functionality via shape change. Still other applications include, for example, pumping ionized fluids and colloidal particles, heat transfer and electronic cooling, adaptive microfluidic mirror arrays, EO-actuators, and EO valves. In another application, pores can be coated with certain chemicals, enabling chemical reactions and synthesis of new materials. A benefit for at least one of the embodiments is high throughput screening and compound profiling.

Multiple EO pumps can be used together in series or parallel. The EO pumps can be also integrated within micro-meter and millimeter scale fluidic systems, by, for example, stacking them together, for example, to increase the pressure build-up, or to maintain flowrate to overcome the viscous losses and pressure loads in long channels.

The devices described herein can be run on small watch batteries, and can thus enable a variety of hand held devices.

Electroosmotic pumps and their applications are known and can be adapted for the EO pumps provided herein. See, for example, US Patent Nos. 7,667,319;7, 645, 368; 7,274,106; 7,231 ,839, 7,185,697, 7,149,085, 7,134,486, 7,131 ,486, 7,105,382, 7,086,839, 7,084,495, 7,037,416, 6,992,381, 6,991 ,024, 6,942,018, 6,934,154, 6,861 ,274, 6,805,841, 6,747,285, 6,726,920, 6,639,712, 6,613,211, 6,595,208, 6,541 ,021 , 6,395,106, 6,323,042, 6,315,940, 5,573,651.

Figure 23 provides an embodiment for adaptive microfluidic mirror arrays. This embodiment is based on displacement of fluid confined in a chamber composed of two reservoirs which are connected by an EO pump. As the polarity of the EO pump changes, upper and lower reservoirs in each chamber experience either pressure or suction modes. An elastic reflective surface tethered to the top of the upper reservoirs work as the adaptive mirror. One can have, for example, about six or seven micron deflection in a polydimethylsiloxane (PDMS) membrane. The EO pump can be prepared with use of porous alumina. Compared to current electrostatic actuation technology, this design can (1) eliminate large voltage requirements, (2) avoid snap-in from electrostatics, (3) allow better resolution than the current systems with 61 actuators in a 5 cm diameter mirror.

Figure 24 illustrates other uses for embodiments for porous alumina structure and EO pumps. The left drawing shows schematics for a MEMS valve compared to an EO valve. The valve can close the fluid flow in the channel. The right drawing shows a water purification system with flow of steam into a membrane and then flow of liquid from the membrane.

PART II - NANOSTRUCTURES In addition, provided herein are coaxial nanostructures, methods for making the coaxial nanostructures, and devices incorporating the coaxial nanostructures. At least some of these methods can be used to prepare electroosmotic pumps and membranes.

COAXIAL NANOSTRUCTURES

The coaxial nanostructures include an inner nanostructure and at least one outer nanotube disposed around the inner nanostructure. The coaxial nanostructures may include multiple outer nanotubes (up to n outer nanotubes) arranged concentrically around the inner nanostructure. This includes embodiments in which the coaxial nanostructure includes a first outer nanotube disposed around an inner nanostructure, a second outer nanotube disposed around the first outer nanotube, a third outer nanotube disposed around the second outer nanotube, and so forth. In any of these embodiments, the inner nanostructure may also be a nanotube. However, the innermost nanostructure may also be a nanorod.

The coaxial nanostructures may also include an annular channel between the inner nanostructure and the at least one outer nanotube. For those embodiments having more than one outer nanotube, the coaxial nanostructure may include additional annular channels between the additional outer nanotubes. By way of example only, a coaxial nanostructure may include a first outer nanotube disposed around an inner nanostructure, a first annular channel between the inner nanostructure and the first outer nanotube, a second outer nanotube disposed around the first outer nanotube, a second annular channel between the first outer nanotube and the second outer nanotube, and so forth. In some embodiments, the annular channel comprises air. In other embodiments, the annular channel may comprise a sacrificial material. Sacrificial materials are further described below.

The materials used to form the coaxial nanostructures may vary. By way of example only, the inner nanostructure and any of the outer nanotubes may comprise a conductor, an insulator, or a semiconductor. A variety of conductors may be used, including metals or nitrides of metals. Non-limiting examples of metals include Ti, Au, Pt, Al, Cu, Ag, and W. Non-limiting examples of metal nitrides include TiN, TaN, AIn, GaN, Zr ₃ N _4, Si ₃ N ₄ and Hf ₃ N ₄ . Similarly, a variety of insulators may be used, including metal oxides. Non-limiting examples of metal oxides include SiO ₂ , HfO ₂ , ZrO ₂ , Al ₂ O ₃ , TiO ₂ , RuO ₂ , Ta ₂ O ₅ , WO ₂ , WO ₃ , La ₂ O ₃ , PtO, Y ₂ O ₃ , MoO ₂ , In ₂ O ₃ , V ₂ O ₅ , NiO and ITO. A variety of semiconductors may also be used, including, but not limited to ZnO, GaAs, GaP, GaN, InP, InAs, AlAs, and Ge. In some embodiments, the inner nanostructure and any of the outer nanotubes are substantially free of carbon. By "substantially free of carbon," it is meant that the nanostructures do not include, and are not formed of, carbon. However, such nanostructure may include trace amounts of carbon that may be unavoidable due to the methods used to form the nanostructures. The structures can be different from and not comprise carbon nanotubes including multi-walled carbon nanotubes, single walled carbon nanotubes, and other types of carbon nanotubes. In still other embodiments, the inner nanostructure and any of the outer nanotubes are completely free of carbon. The inner nanostructure and each of the outer nanotubes may be formed of the same material. Alternatively, the inner nanostructure and each of the outer nanotubes may be each formed of different materials. Finally, some of outer nanotubes and the inner nanostructure may be formed of the same material while others are formed of different materials.

The dimensions of the coaxial nanostructures may also vary. The diameter of the coaxial nanostructures may range from about 50 nm to about 300 nm for alumina templates and at the upper range pore diameters may range as large as several micrometers for porous silicon templates. The pore diameter range that is achievable depends on the material parameters of the membrane material and the electro-chemical parameters of the fabrication method used. This includes embodiments in which the diameter is about 60 nm, 75 nm, 90 nm, 125 nm, 150 nm, 175 nm, 200 nm, or 250 nm. The length of the coaxial nanostructures may range from about 15 μm to about 75 μm. This includes embodiments in which the length is about 20 μm, 30 μm, 40 μm, 50 μm, 60 μm, or 70 μm. The aspect ratio (the ratio of the length of the coaxial nanostructure to the diameter of the coaxial nanostructure) may also vary. In some embodiments, the aspect ratio ranges from about 5 to about 1 ,200, or about 300 to about 1200. This includes embodiments in which the aspect ratio is about 400, 500, 600, 700, 800, 900, or 1000. Finally, the cross-sectional shape of the coaxial nanostructures may vary. In some embodiments, the cross-sectional shape is a polyhedron, such as an octahedron. In other embodiments, the cross-sectional shape is substantially circular. By "substantially circular," it is meant that shape is circular-elliptical, but not necessarily perfectly circular. Similarly, the dimensions of the outer nanotubes and the inner nanorod or nanotube forming the coaxial nano structures may vary, depending upon number of such structures present in the coaxial nanostructure and the overall dimensions of the coaxial nanostructure itself. The width of the walls of the nanotubes and the width of the annular channel (if present) may also vary. In some embodiments, the width ranges from about 5 nm to about 30 nm. This includes embodiments in which the width is about 10 nm, 15 nm, 20 nm, 25 nm and 35nm.

Coaxial nanostructures may be coupled to other elements. In some embodiments, the coaxial nanostructure is coupled to a substrate. A variety of substrates may be used, including any of the metals described above. In some embodiments, the substrate is an Al substrate. In such embodiments, the coaxial nanostructure may be attached to the substrate at one of the ends of the coaxial nanostructure. In other embodiments, the substrate may be a nanoporous substrate and the coaxial nanostructure may be disposed within a pore of the nanoporous substrate. A variety of nanoporous substrates may be used, including, but not limited to, porous anodic aluminum oxide (AAO) substrates, polycarbonate nanoporous membranes, and porous silicon. Such nanoporous substrates are known and AAO is commercially available. In still other embodiments, the coaxial nanostructure may be coupled to both a metal substrate, such as an Al substrate, and a nanoporous substrate, such as an AAO substrate. In such an embodiment, the coaxial nanostructure may be disposed within a pore of the nanoporous substrate and attached to the metal substrate at one of the ends of the coaxial nanostructure.

Regarding AAO substrates, anodic aluminum oxide is formed by electrochemical oxidation of aluminum in acidic solutions to form regular porous channels, which are parallel to each other. See H. Masuda and K. Fukuda, Science, 268, 1466 (1995); V. P. Menon and CR. Martin, Anal. Chem., 67, 1920 (1995); and M. A. Cameron, LP. Gartland, J.A. Smith, S.F. Diaz and S. M. George, Langmuir, 16, 7435 (2000). The individual pore diameters inside the porous alumina membrane are mainly defined by the anodization voltage. The diameter of the pore depends on the electrolyte nature, its temperature and concentration, the current density and other parameters of the anodization process. Aside from the modulation of the pore diameters by variation of the electrolyte composition and anodization conditions, it is possible to further enlarge the pore diameters by another subsequent selective etching of the porous template walls. See H. Masuda and K. Fukuda, Science, 268, 1466 (1995); V. P. Menon and CR. Martin, Anal. Chem., 67, 1920 (1995); and M. A. Cameron, LP. Gartland, J.A. Smith, S.F. Diaz and S. M. George, Langmuir, 16, 7435 (2000). The Examples below provide an exemplary method for making a suitable AAO substrate.

Also provided herein are arrays of two or more of any of the coaxial nano structures described above. The arrays of coaxial nano structures may be coupled to any of the substrates described above.

A non-limiting exemplary coaxial nanostructure is illustrated in FIG. 21. In FIG. 21, a first layer of a metal oxide (e.g., ZrO ₂ ) is deposited on the inner surfaces of the nanopores of an AAO substrate. Next, a second layer of a metal oxide (e.g., Al ₂ O ₃ ) is deposited on the first layer of the metal oxide. Next, a third layer of a metal oxide (e.g., ZnO) may be deposited on the second layer of the metal oxide. Finally, both the AAO substrate and the second layer of the metal oxide may be removed by etching to provide a coaxial nanostructure comprising an inner nanotube of ZnO, a first outer nanotube Of ZrO ₂ disposed around the inner nanotube, and a first annular channel between the inner nanotube and the outer nanotube.

METHODS

The coaxial nanostructures described above may be prepared according to the following methods. The methods can use atomic layer deposition or other suitable chemical vapor deposition (CVD) techniques to deposit layers (also referred to as films herein) of the types of materials described above on the inner surface of the nanopores of a nanoporous substrate. ALD is a known technique. Briefly, ALD technology deposits thin films using pulses of chemical precursor gases to adsorb at the target surface one atomic layer at a time. ALD is based on the sequential deposition of individual monolayers or fractions of a monolayer in a controlled fashion. More specifically, in ALD the growth substrate surface is alternately exposed to the vapors of one of two chemical reactants (complementary chemical precursors), which are supplied to the reaction chamber one at a time. The exposure steps are separated by inert gas purge or pump-down steps in order to remove any residual chemical precursor or its by-product before the next chemical precursor can be introduced into the reaction chamber. Thus, ALD involves a repetition of individual growth cycles. See also Ritala, M., "Atomic Layer Deposition", p. 17 - 64, in Institute of Physics Series in Materials Science and Engineering " High-k Gate Dielectrics" edited by Michel Houssa, Institute of Physics Publishing, Bristol and Philadelphia 2003.; Leskala, M., and Ritala, M., "ALD Precursor Chemistry: Evolution and Future Challenges," J. Phys.. IV 9, p. 837-852, 1999.

Since a film deposited by ALD is grown in a layer-by-layer fashion and the total film thickness is given by the sum of the number of ALD cycles, it is possible to calculate the number of cycles necessary to reach a desired final film thickness. Conversely the thickness of a film can be set digitally by counting the number of reaction cycles. In general, ALD achieves deposition rates on the order of 0.1 - 1.0 A per cycle, with cycle times ranging from one to ten seconds. Due to the self-limiting nature of the surface reactions, accidental overdosing with precursors does not result in increased film deposition. Thus, ALD is able to achieve very precise across-wafer film thickness uniformity, unmatched step coverage and exceptional conformality. Because of the nature of ALD, film thickness is immune to variations caused by non-uniform distribution of reactant vapor or temperature in the reaction chamber. See Niinisto, L., Paivasaari, J., Niinisto, J., Putkonen, M., and Mieminen, M., "Advance electronic and optoelectronic materials by Atomic Layer Deposition: An overview with special emphasis on recent progress in processing high-k dielectrics and other oxide materials", Phys. Stat. Solid, (a) 201, p. 1443 -1452, (2004); and Ritala, M., "Atomic layer deposition," Editors Michel Houssa, High-k Gate Dielectrics, p. 17 - 64, Publisher Institute of Physics Publishing, Bristol, UK, 2004.

A variety of chemical precursors may be used with ALD, depending upon the desired film. The general requirements and properties of useful chemical precursors are known. See Sneh, O., Clark-Phelps, R.B., Londergan, A.R., Winkler J., and Seidel, T., "Thin film atomic layer deposition equipment for semiconductor processing," Thin Solid Films, Vol.402, Issues 1-2, p. 248 - 261, 2002 and Leskela, M., and Ritala, M., "Atomic Layer Deposition (ALD): from precursor to thin film structures, "Thin Solid Films, 409, p. 138 - 146, 2002. Specific chemical precursors are provided in the Examples below. In one embodiment of the disclosed methods, the method comprises forming a layer of a first material on an inner surface of a nanopore of a nanoporous substrate using atomic layer deposition and forming a layer of a second material on the layer of the first material using atomic layer deposition. In another embodiment, a layer of a third material may be formed on the layer of the second material, a layer of a fourth material may be formed on the layer of the third material, and so forth. In each of these embodiments, the layer of the first material corresponds to an outer nanotube of the coaxial nanostructures described above. The layer of the second material provides either an additional outer nanotube, or an inner nanostructure, depending upon the number of layers of materials deposited. The first material, second material, and third material may include any of the conductors, insulators, and semiconductors described above. Similarly, any of the nanoporous substrates described above may be used with the disclosed method.

The method may further comprise removing the nanoporous substrate after the coaxial nanostructure is formed. A variety of methods may be used to remove the nanoporous substrate, including, but not limited to chemical etching. A variety of chemical etchants may be used, depending upon the composition of the nanoporous substrate. By way of example only, when the nanoporous substrate is AAO, NaOH may be used to remove the substrate.

In another embodiment of the disclosed methods, the method comprises forming a layer of a first material on an inner surface of a nanopore of a nanoporous substrate using atomic layer deposition, forming a first layer of a sacrificial material on the layer of the first material using atomic layer deposition, and forming a layer of a second material on the first layer of the sacrificial material using atomic layer deposition. Other sacrificial layers and layers of additional materials may be deposited. For example, a second layer of a sacrificial material may be formed on the layer of the second material, a layer of a third material may be formed on the second layer of the sacrificial material, and so forth. By "sacrificial material," it is meant a material that is capable of being substantially removed (i.e., removed, but not necessarily completely removed) by a chemical etchant. A non-limiting example of a sacrificial material is Al ₂ O ₃ , which is capable of being substantially removed by a variety of chemical etchants, including NaOH. As discussed above, the first material, second material, and third material may include any of the conductors, insulators, and semiconductors described above. Similarly, any of the nanoporous substrates described above may be used with the disclosed method. In an EO pump application, sacrificial layers can be avoided.

The description of the coaxial nanostructures, AAO substrates, and ALD process make clear that the dimensions of the coaxial nanostructures are both a function of the pore sizes of the AAO substrates as well as the number of cycles and length of each cycle of the ALD process. In order to make the highest aspect ratio coaxial nanostructures for a given AAO substrate, the length of the cycle may be maximized to ensure deposition along the entire length of the nanopore. Long cycle times, however, are contrary to the conventional wisdom that cycle times should be minimized to prevent clogging the pores of the AAO substrates.

DEVICES AND APPLICATIONS

The coaxial nanostructures described above may be incorporated into a variety of devices for use in a variety of applications. By way of example only, the coaxial nanostructures may be used in electroosmotic pumps, chemical sensors, photovoltaic devices, and photonic crystals. The coaxial nanostructures may also find use as extremely hard and highly durable nanometer-sized pipette tips for various medical applications. Although many of these devices are known, devices incorporating the disclosed coaxial nanostructures are expected to exhibit superior properties over conventional devices due to the high aspect ratio and high surface area of the coaxial nanostructures. These devices are further described below.

Non-limiting exemplary devices are further shown in FIG. 22. In FIG. 22, a layer of a metal oxide (e.g., ZrO ₂ ) is deposited on the inner surfaces of the nanopores of an AAO substrate. Next, a layer of a metal (e.g., Pt) is deposited on the top surface and the bottom surface of the AAO substrate. Finally, the AAO substrate may be dissolved by chemical etching. The structure shown in FIG. 22 may be used as, for example, an oxygen sensor.

Additional description is provided with use of the following non-limiting examples. EXAMPLES

The following examples were carried out with an ALD reactor from Cambridge Nanotech, Model Savannah 100.

Example 1 : Two-terminal Passive Zeta-Potential (Q Controlled Electroosmotic Pump by coating the inside of the nanopore channels with suitable ALD metal oxides

A 2-terminal electroosmotic pump with Au or Pt metallization as cathode and anode on the front and backside of the device was fabricated. A dozen such devices were fabricated, each which exhibited excellent pumping capacity of DI water at low voltages of 1 — 3 V. The average diameter of the nanopore channels of the alumina membrane was 250 nm to 300 nm.

For passive Zeta-Potential control, very thin coatings of insulating metal oxides in the range of 10 nm to 25 nm Of HfO ₂ , Al ₂ O ₃ , and ZrO ₂ were applied to the inside walls of the nanopores by Atomic Layer Deposition (ALD).

Initially, a 15 μL DI water droplet was placed with a pipette on the surface of the electroosmotic pump. Gravity could not move the DI water through the nanopores during a 10 min waiting period. As soon as a DC electric field of 3 V was applied, the device pumped the 15 μL DI water within 5 seconds to the backside of the substrate, wetting the tissue paper underneath. It was also possible to pump the 15 μL of DI water back to the surface against gravity by reversing the electrodes.

The experimental conditions were as follows: Applied voltage = 3 V, nanoporous substrate thickness = 60 μm, corresponding electrical field = 3V/60 μm = 0.05 V/μm. The performance of the pump was as follows: Amount of DI water: 1 droplet (15 μL); Water

2 2 contact area: ~ 0.2 cm ; Time: 5 seconds; Flowrate per unit area = 15 μL/5 second/0.2 cm =

2

0.9 mL/min/cm at 3 V or electrical field of 0.05 V/μm. Example 2: Three- Terminal Electrode Active Zeta-Potential (Q Controlled

Electroosmotic Pump in Nano-Porous Membranes using Atomic Layer Deposition (ALD)

A novel 3 terminal device for an electroosmotic (EO) pump with "Active Zeta Potential Control" by a third electrode in nano-porous membranes is provided. This novel device architecture is achieved by applying the concept of active control of the Zeta Potential which requires biasing with a low dc voltage of the interior nano-pore walls that have been coated with an electrically conducting thin film followed by an insulating layer. This novel electroosmotic device architecture with 3 terminals enables the active modulation of the zeta potential over a much wider range in order control the flow rate of the electro-osmotic device. The front and backside surface of the porous alumina membrane is metalized with Au or Pt to serve as cathode and anode in such a way to leave the pore opening on each surface free for fluid transport. In order to independently bias the pore walls, a coating with a conducting film is applied. To avoid an electrical short, the conducting film is electrically isolated from the cathode and anode on the device surfaces. Electrical isolation is achieved by use of a double film coating inside the nano-channels in the alumina or porous silicon membrane. Film deposition is accomplished by Atomic Layer Deposition (ALD) for the aspect ratios of, for example, 5-1,200 or 300-1200.

The integrated process flow for the manufacturing of the electro-osmotic nano-device involves several important techniques: the template guided approach for the fabrication of the nano-porous alumina membrane or porous silicon membrane; atomic layer deposition (ALD) technology to coat the interior regions of the cylindrical pores of the porous (AAO) or porous silicon membrane with suitable films for active and passive modulation of the zeta potential; micro photolithography and lift-off techniques for patterning the 3-terminal device; backend metallization techniques for direct gold or platinum contacts to use as cathode and anode on the front and backside of the porous membrane; and external contact leads to the metalized front and backside in the form of wire bonding or copper tapes. For the independent active control of the Zeta potential via a third contact, a fabrication process is provided employing ALD coatings of the interior walls of the nano-pores with three classes of electrically conducting thin films: elemental metal films; electrically conductive metal nitride films such as TiN; and semiconducting thin films such as ZnO. Since this electrically conductive ALD thin film could get shorted with the anode and cathode metallization on the free surfaces of the alumina membrane, this inner nano-pore lining may be encapsulated with a suitable electrically insulating ALD layers such as SiO ₂ or any of the high-k insulating layers such as HfO ₂ , ZrO ₂ , Al ₂ O ₃ , TiO ₂ and SiO ₂ .

The process flow for nano-manufacturing of a three terminal electroosmotic pump is provided as follows. First, coat the whole AAO template with an electrically conducting ALD film of total thickness sufficient for good electrical conductivity, which can be in some cases of about less than 10 nm, which requires about 100 ALD film deposition cycles in some embodiments. Ensure that the pores are coated from top to bottom throughout the thickness of the porous alumina membrane. Second, fabricate an external third contact to conducting ALD film lining the pores by photolithography and copper tapes or wire bonding techniques. Third, deposit selected dielectric material on top of the metal layer by ALD for electrical insulation and additional passive zeta potential control. Fourth, conduct photolithography patterning of cathode and anode contacts. Fifth, sputter metal electrodes as anode and cathode on both sides of the AAO template. The sputtered electrodes only cover the AAO surface and do not obstruct the open nano-pores.

ADDITIONAL EXAMPLES

Platinum deposition is particularly important for a three terminal EO pump and forming an inner third electrode inside the nanopores extending from side to side. Deposition studies on flat surfaces were used, in some cases, to determine process parameters for deposition into pores.

Example 3 : Atomic Layer Deposition (ALD) of nanoscaled Pt thin films on a Si substrate

Platinum thin films were successfully deposited by atomic layer deposition (ALD) using (methylcyclopentadienyl) trimethylplatinum (MeCρPtMe3) and oxygen (O ₂ ) as the precursors. Sheet conductance, G _S h = t/p (where t is the film thickness and p is the resistivity) is proportional to film thickness, which has a linear relationship to the ALD cycles. Thus, when the resistivity approaches a constant value over a deposition temperature rage, an ALD process window may be established. As shown in Fig. 6, the resistivity was assumed to be constant over the temperature range between 270-320 C, where stable ALD Pt film growth was achieved. The increase in growth rate after 320 C is indicative of Pt precursor decomposing. Based on other properties such as surface roughness and impurity species in the film, and the data of Fig. 6, the deposition temperature of 300 C was used for conducting the additional experiments from which the data in Figs. 7A-B was derived.

Fig. 7A shows good linearity between sheet conductance and ALD cycles for Pt films deposited at 300 ^° C, specifically after 300 cycles. This reveals a constant growth rate for Pt thin films by ALD. This effect was therefore utilized to establish a new metrology method for Pt film thickness determination by calibrating the sheet conductance plot with accurate thickness measurement by Transmission Electron Microscopy (TEM).

The Pt thin films were also annealed in forming gas (95% N ₂ and 5% H ₂ ) at 450 C for 30 min. The forming gas annealing (FGA) was effective to improve the adhesion and conductivity of the Pt thin films by passivating dangling bonds in the grain boundaries with hydrogen. As also shown in Fig. 7A, a higher sheet conductance was observed after FGA.

However, as shown in the figure, there is a drop in the sheet conductance for both as- deposited and annealed Pt films with ALD cycles less than 300. While not limited to a particular theory, it is believed that the decrease in the sheet conductance or the increase in the sheet resistance is either due to the surface scattering or the structure of Pt film when the film is thinner than a certain thickness. For example, as shown in Fig. 7A, for the first 150 cycles, no sheet conductance was observed. This may be because no continuous Pt film was formed for the first 150 cycles.

The resistivities of the Pt thin films, as shown in Fig. 7B, were calculated using the film thickness as determined by cross-sectional TEM. The resistivity was shown not to be strongly thickness dependent with Pt film thicker than 15 nm. The 20 nm thick Pt film after FGA showed a resistivity of 12 μΩ-cm, which is very close to the bulk resistivity of Pt (10.8 μΩ-cm). Accordingly, a Pt film of 15 nm is thick enough to achieve good electrical conductivity. The detailed structures of the ALD Pt thin films were analyzed by cross-sectional TEM (XTEM). Figure 8 A shows the very early stage of Pt deposition on an Si substrate with native oxide. For about 50 ALD cycles, no continuous Pt thin film was formed. Instead, as shown in the figure, only isolated Pt nano-particles (size less than 5 nm) were randomly distributed on the surface.

With more deposition cycles the Pt nano-particles continued to grow and in certain areas, the Pt islands coalesced into larger ones but were still isolated from each other, as shown in Figure 8B. Figure 8C shows the detailed structure of the ALD grown Pt films at 200 ALD cycles when the Pt islands coalesced and connected to each other, thereby forming a continuous Pt network. However, at this stage, the Pt film may not have covered the whole surface, leaving a porous surface with low sheet conductance, corresponding to that at 200 ALD cycles as in Figure 7A. As the Pt film became thicker and thicker, a continuous Pt formed with a smooth surface, as shown in Figures 8 D and 8E.

From the detailed physical analysis by XTEM, for example as shown in Figs. 8A-E, a growth model for Pt thin film was elucidated as follows: the early stage of ALD Pt film growth occurs by random heterogeneous nucleation forming initially discontinuous Pt islands until a critical thickness is attained where the Pt islands coalesce into a continuous Pt film. Once the substrate surface is completely covered with the coalesced Pt islands, the ALD growth follows the classical model of monolayer by monolayer growth per ALD deposition cycle. A stable growth rate 0.5 A/ALD cycle was achieved for the precursor used in this example. From resistivity measurements, it is determined that 15 nm is an important thickness to achieve continuous Pt film, which corresponds to 300 ALD cycles.

Example 4; Atomic Layer Deposition (ALD) of nanoscaled Pt thin films on an AAO substrate

The conformal deposition of Pt film on porous anodic aluminum oxide (AAO) membranes was investigated. The pore size and thickness of the AAO membranes were 250- 300 nm and 60 μm, respectively, which correspond to an aspect ratio of more than 200. Figure 9A shows that the Pt can be deposited 10 μm deep into the AAO pore without exposure time, with possible thickness gradient from the surface. The bright area has Pt deposition which was confirmed by energy dispersive spectroscopy (EDS) mapping. The penetration of Pt is deeper with 10 sec exposure for the Pt precursor (Figure 9B). However, it was found from Figure 9C that the depth penetration into the nanoporous template saturates at around 20 μm from the surface even with exposure times of 30 sec. This is attributed to the combined effects of AAO pore size and the Pt precursor diffusion rate, which is inversely proportional to the square root of the molecular weight for this Pt precursor.

Platinum tubes were made to further analyze the quality of the thickness conformity by etching away the AAO membrane in NaOH solution. Figure 10 shows the Pt tube structure, wherein the tubes are separated from each other after removal of the AAO template. The SEM image reveals that the Pt tubes are about 15 μm in length. This length is less than the Pt penetration depth shown in Figure 9C with 30 sec exposure time, which can be explained by the TEM inset images. These TEM images were taken after the separation of the Pt tubes. They are similar to the TEM planar view images, showing the structures of the tube ends. At the end close to the surface, the Pt film is continuous and forms dense tubes.

Example 5.

Additional testing for zeta potential can be carried out as illustrated in Figures 11-13.

Example 6: Formation of a Nanoporous AAO Substrate

The nanoporous AAO substrate was prepared by a two-step anodization procedure. High purity aluminum sheets (0.5 mm thick) were degreased in acetone. The Al sheets were then electropolished in a solution of HClθ4and ethanol (1 :4, v/v) at 20 V for 5-10 min or until a mirror like surface was achieved. The first anodization step was carried out in a 0.3 M oxalic acid solution electrolyte under a constant direct current (DC) voltage of 80 V at 17 ⁰ C for 24 h. The porous alumina layer was then stripped away from the Al substrate by etching the sample in a solution containing 6 wt% phosphoric acid and 1.8 wt% chromic acid at 60 ⁰ C for 12 h. The second anodization step was carried out in a 0.3 M oxalic acid solution under a constant direct current (DC) voltage of 80 V at 17 ⁰ C for 24 h. The AAO substrates with highly ordered arrays of nanopores were then obtained by selectively etching away the unreacted Al in a saturated HgCb solution. FIG. 14A shows the SEM image of the pore structure of the AAO after the surface was planarized by ion milling. The pore size is in the range of 200-300 nm and the wall width between pores is around 50 nm. Some of the pores were connected through thinning of the wall. The cross-sectional SEM image shown in FIG. 14B reveals that the pores are all parallel to each other and across the whole substrate of 60 μm thickness. The inset to FIG. 14B shows the formation of branches in some of the pores. These branches may be eliminated with shorter anodization times, which results in a shorter pore length. A closer view of tube opening showed that the side connected to the cathode has smaller pore size, to a depth of a few micrometers. This thin layer can be removed by etching to achieve uniform pore diameter across the entire substrate depth. High magnification FE-SEM of a cleavage sample highlights the microstructure of partially split open nanopores of AAO. The smooth morphology of the inside walls of the AAO nanopores can be clearly seen. Excellent surface finish of the inner pore walls of the template is useful for obtaining highly ordered tube-in- tube nanostructures, since the ALD thin film coating technique replicates the surface finish on an Angstrom scale.

Example 7: Formation of HfQ?, ZrO^ ₁ and ZnO Nanotubes

The AAO substrates were subsequently transferred to the ALD chamber for Zrθ2, Hfθ2 and ZnO coating of the inside surfaces of the nanopores. The Zrθ2 and Hfθ2 deposition was performed at 250 ⁰ C using water vapor as the oxidant and tetrakis (dimethylamido) hafnium (IV) and tetrakis (dimethylamido) zirconium (IV) as the precursor, respectively. The deposition rate is about 1 A/cycle at this temperature. ZnO was grown with diethyl zinc (DEZ) as precursor and water vapor as oxidation source. The optimum ALD process window for ZnO was determined to be in the temperature range between 110 ⁰ C and 160 ⁰ C.

Due to the extremely high (60 μm) depth of the nanopores and the diffusivity of the chemical precursors, the entire nanopores may not be coated unifoπnly unless an extended ALD cycle time is used. For AAO pores coated with 20 nm Hfθ2, cross sectional energy dispersive spectroscopy (EDS) mapping demonstrated that Hf signal was detected up to a depth of about 15 μm from the sample surface without any added ALD exposure time. For AAO pores coated with 20 nm Zrθ2, the surface pore diameter was reduced after Zrθ2 deposition, indicating that Zrθ2 was also deposited on AAO template. Increased ALD exposure times were used for the Zr precursor to reach saturation of precursor species inside walls of the pores and ensure uniform coating along the length of the pores.

FIGs. 15A and 15B show the cross-sectional SEM image and EDS mapping of the AAO substrate coated with 20 nm ZrCh using 30 s additional ALD exposure time. It can be observed that there is still a gradient in the Zr signal following the length of the metal oxide the nanotubes. This is because the AAO substrate was placed in the ALD chamber flat on one side so that access of the Zr precursor to the backside opening was blocked. The uniformity of coating can be improved by lifting the AAO substrate so that the precursor can access both sizes of the pore opening during ALD deposition.

FIGs. 16A and 16B show an AAO substrate before being coated with ZrO ₂ (A) and after being coated with 20 nm Zrθ2 (B). A comparison of the figures shows that the pore size has been reduced because the wall thickness has been increased by growing a ZrOi film. In order to fabricate free-standing Zrθ2 nanotubes (i.e., nanotubes unsupported along their lengths by the AAO substrate), the alumina walls between the pores were dissolved by a 6M NaOH solution. The porous AAO surface was first cleared of its ZrOa films by ion milling to expose the AAO wall to the etchant. FIG. 3C shows the free-standing ZxOi nanotubes after ion milling and chemical dissolution of alumina walls. The SEM image clearly shows the empty trenches in place of the former alumina side walls. The dimensions of the nanotubes are dependent upon the thickness and pore diameter of the AAO substrate and the ALD deposition time. Smaller tubes or even rods can be fabricated using this method by using AAO substrates with smaller pores. Different materials can still be deposited inside of the nanotubes depending on the application.

Example 8: Formation of a HfO? Tube-in-Tube Coaxial Nanostructure

In this example, a second nanotube having a smaller dimension was deposited inside of the aforementioned HfCbnanotubes. To fabricate this tube-in-tube structure, two layers of 10 nm Hfθ2 films were deposited inside of the AAO pores and separated by 25 run of a layer of AkCb, which was deposited by ALD at 300 ⁰ C using [A1(CH3)3] (TMA) and water vapor as the aluminum and oxygen source, respectively. AkCb is the same material as the AAO substrate and can be easily etched away. Following the three layer coating, the sample surface was again polished by ion milling and then dipped into NaOH solution to etch both the AAO substrate and AI2O3 layer between Hfθ2 layers. FIG. 17 shows a double-walled HfOa tube-in-tube structure after wet etching in NaOH solution. The Hfθ2 tube thicknesses are very uniform from both the top and cross section. The expected wall thicknesses for both tubes are 10 nm, as determined from the number of ALD cycles. However, the Hfθ2 tubes look much thicker from the SEM image due to the gold coating for charging release.

Transmission electron microscopy (TEM) was used to examine the Hfθ2 tube-in-tube structure and tube wall thickness using the following processing sequence. After NaOH etching the Hfθ2 nanotubes were suspended in isopropanol solution and separated by sonicating. The HfO ₂ nanotubes in isopropanol were subsequently poured onto the TEM copper grid. FIG. 18 shows TEM high magnification micrographs of double-walled HfO ₂ tube-in-tube structure. The tube-in-tube structure shown in FIG. 18 was achieved even from AAO pores with branches or dead ends. FIG. 18 also reveals that upon release of the nanotubes from the AAO substrates, the coaxial nano structures have undergone a shape transformation from an irregular octahedral shaped cross-section (compare FIGS. 16 and 17) to a circular cross-section. Since this shape transformation takes place at room temperature during the chemical release of the coaxial nanostructures, temperature activated diffusion processes are ruled out. Coaxial nanostructures having circular cross-sections are desirable for pipette tips for various medical applications.

Example 9: Formation of a ZrO? Tube-in-Tube-in-Tube Coaxial Nanostructure

The method of Example 8 was modified to provide two nanotubes having a smaller dimension deposited inside of the aforementioned Zrθ2 nanotubes. Three layers of Zrθ2 films were deposited inside of AAO pores, separated by a layer of AI2O3. The Al ₂ O ₃ layers were removed as described above. The resulting tube-in-tube-in-tube coaxial nanostructure is shown in FIG. 19.

Example 10: Formation of HfCVZrO? Coaxial Nanostructure

ALD was used to deposit a layer OfHfO ₂ inside the nanopore of an AAO substrate followed by a layer of ZrO ₂ on the layer of HfO ₂ to provide a double-walled coaxial nanostructure. FIG. 20 shows the surface morphology and tube size after the two layer coating.

The following references can be used to further carry out the claimed inventions, particularly with respect to zeta potential measurement.

References

Hunter, R. J. Foundations of Colloid Science; Oxford University Press: New York, 1995; C. Wemer, U. Konig, A. Augsburg, C. Amhold, H. Korber, R. Zimmermann, HJ. Jacobash," Electrokinetic surface characterization of biomedical polymers - a survey," Colloids and Surfaces A, vol.l, pp. 159 519, 1999.

C. Ye, D. Sinton, D. Erickson, and D. Li, "Electrophoretic Motion of a Circular Cylindrical Particle in a Circular Cylindrical MicroChannel," Langmuir, Vol.18, pp. 9095-9101, 2002 H.J. Keh, and J. L. Anderson, "Boundary effects on electrophoretic motion of colloidal spheres." Journal oj ^' Fluid Mechanics, vol. 153 pp. 417-439, 1985.

Martinez, R.A., Pokrovsky, O. S., Schott, J., and Oelkers, E. H., "Surface charge and zeta- potential of metabolically active and dead cyanobacteria," Journal of Colloid and Interface Science, vol. 323, pp. 317-325, 2008.

Wilson, W. W., Wade, M.M., Holman, S.C., and Champlin, F.R., "Status of methods for assessing bacterial cell surface charge properties based on zeta potential measurements,", Journal of Microbiological Methods , vol. 43, pp 153-164, 2001.

Huisman, LH. , Tragardh,G., Tragardh., C, and Arto Pihlajamaki,A., "Determining the zeta- potential of ceramic microfiltration membranes using the electroviscous effect,", Journal of Membrane Science, vol. 147, pp. 187-194, 1998.

Teixeira, M. R., Rosa, MJ. , and Nystrom, M., "The role of membrane charge on nanofiltration performance," Journal of Membrane Science, vol. 265, pp. 160-166, 2005.

Al-Amoudi, A., Williams, P., Mandale, S., and Lovitt R. W., "Cleaning results of new and fouled nanofiltration membrane characterized by zeta potential and permeability," Separation and Purification Technology, vol. 54, pp. 234-240, 2007. Lee, G. B., Fu, L. M., Lin, C. H., Lee CY. , and Yang, R. J., ''Dispersion control in microfluidic chips by localized zeta potential variation using the field effect," Electrophoresis, vol. 25, pp.

1879-1887, 2004.

Tandon, V., Bhagavatula, S.K., Nelson, W,C, and Kirby BJ., "Zeta potential and electroosmotic mobility in microfluidic devices fabricated from hydrophobic polymers: 1.

The origins of charge,", Electrophoresis, vol. 29. pp. 1092-1 101 , 2008.

D. Gu, K. Tapily, P. Shrestha, M. Zhu, G. Celler and H. Baumgart, "Experimental Study of ALD HfCh Deposited on Strained Silicon-on-Insulator and Standard SOI", Journal of the Electrochemical Society, vol. 155 no. 6, G129 - G133 ( 2008)

K. Tapily, J.E. Jakes, D.S. Stone, P. Shrestha, D. Gu, H. Baumgart and A.A. Elmustafa, "Nanoindentation Investigation of HfCh and AbCb Films Grown by Atomic Layer Deposition", Journal of the Electrochemical Society, vol. 155 no.7, H545-H551 (2008)

Diefeng Gu, Helmut Baumgart, Tarek Abdel-Fattah, and Namkoong Gon, "Atomic Layer Deposition of ZrO ₂ and HfO ₂ Nanotubes by Template Replication," Electrochem. and Solid- State Lett., 12 (4), K25 (2009)

Xiaoxiao. Zhu, D. Gu, Qiliang Li', D. E. Ioannou, H. Baumgart, J. S. Suehle, C. A. Richter, "Silicon Nanowire Non-Volatile Memory with High-A: Gate Dielectric Stack," Microelectron. Eng. (Elsevier), 86, p. 1957 (2009)

As will be understood by one skilled in the art, for any and all purposes, particularly in terms of providing a written description, all ranges disclosed herein also encompass any and all possible subranges and combinations of subranges thereof. Any listed range can be easily recognized as sufficiently describing and enabling the same range being broken down into at least equal halves, thirds, quarters, fifths, tenths, etc. As a non-limiting example, each range discussed herein can be readily broken down into a lower third, middle third and upper third, etc. As will also be understood by one skilled in the art all language such as "up to," "at least," "greater than," "less than," and the like include the number recited and refer to ranges which can be subsequently broken down into subranges as discussed above.

All publications, patent applications, issued patents, and other documents referred to in this specification are herein incorporated by reference as if each individual publication, patent application, issued patent, or other document were specifically and individually indicated to be incorporated by reference in its entirety. Definitions that are contained in text incorporated by reference are excluded to the extent that they contradict definitions in this disclosure.

For the purposes of this disclosure and unless otherwise specified, "a" or "an" means "one or more."