STRUCTURALLY CONTROLLED DEPOSITION OF SILICON ONTO NANOWIRES

Abstract

Provided herein are nanostructures for lithium ion battery electrodes and methods of fabrication. In some embodiments, a nanostructure template coated with a silicon coating is provided. The silicon coating may include a non-conformal, more porous layer and a conformal, denser layer on the non-conformal, more porous layer. In some embodiments, two different deposition processes, e.g., a PECVD layer to deposit the non-conformal layer and a thermal CVD process to deposit the conformal layer, are used. Anodes including the nanostructures have longer cycle lifetimes than anodes made using either a PECVD or thermal CVD method alone.

Description

INCORPORATION BY REFERENCE

An Application Data Sheet is filed concurrently with this specification as part of the present application. Each application that the present application claims benefit of or priority to as identified in the concurrently filed Application Data Sheet is incorporated by reference herein in its entirety and for all purposes.

BACKGROUND OF THE INVENTION

Field of the Invention

This invention relates generally to nanostructures, and, more specifically, to multilayered silicon nanowire structures that are useful in battery anodes.

Much work has been done to find a way to use silicon in lithium battery anodes. Silicon holds much promise as it has ten times the lithium capacity as currently-used graphite. But, unfortunately, in absorbing so much lithium, silicon swells by 400%, which usually results in crumbling of the silicon and a short battery lifetime.

SUMMARY

In one aspect an anode for a lithium battery is provided, the anode including a substrate; a nanowire template rooted to the substrate; a first silicon layer substantially coating the nanowire template, the first silicon layer having a first density; and a second silicon layer over the first silicon layer and any exposed nanowire template, the second silicon layer having a density higher than the density of the first silicon layer. According to various embodiments, the first density may be less than 2.2 g/cm³, 2.1 g/cm³, 2.0 g/cm³, 1.9 g/cm³, 1.8 g/cm³, or 1.7 g/cm³. The second silicon layer may have a density greater than 2.25 g/cm³ in some embodiments. In some embodiments, the second silicon layer has a density at least 0.05 g/cm³ greater or 0.15 g/cm³ than that of the first silicon layer.

The first silicon layer may be non-conformal to the nanowire template. In some embodiments, the second silicon layer is conformal to the underlying surface. In some embodiments, the hydrogen content of the first silicon layer is at least 10%. In the same or other embodiments, the hydrogen content of the second silicon layer may be no more than 5%. According to various embodiments, the nanowire template may be conductive and can include silicide nanowires. An example is a nickel silicide nanowire template.

In some embodiments, the first silicon layer is between about 5 and 20 microns thick at its maximum diameter. In some embodiments, the second silicon layer is between about 5 and 500 nanometers thick, e.g., between about 5 and 100 nanometers thick.

Another aspect of the disclosure relates to a lithium battery including an anode as described above, a lithium-containing cathode, and an electrolyte in ionic communication with both the anode and the cathode.

Another aspect of the disclosure relates to a nanostructure including a first silicon layer having a first density; and a second silicon layer over the first silicon layer, the second silicon layer having a density higher than the density of the first silicon layer. According to various embodiments, the first density may be less than 2.2 g/cm³, 2.1 g/cm³, 2.0 g/cm³, 1.9 g/cm³, 1.8 g/cm³, or 1.7 g/cm³. The second silicon layer may have a density greater than 2.25 g/cm³ in some embodiments. In some embodiments, the second silicon layer has a density at least 0.05 g/cm³ greater or 0.15 g/cm³ than that of the first silicon layer. In some embodiments, the nanostructure includes a nanowire within the first silicon wire. The nanowire may be growth-rooted to the substrate.

Yet another aspect of the disclosure relates to method of making an anode for a lithium battery, including providing a substrate; growing nanowires from the substrate, depositing a first silicon layer over the nanowires using a PECVD method; and depositing a second silicon layer over the first silicon layer, the nanowires, and the substrate using a thermal CVD method. In some embodiments, the nanowires are silicide nanowires. In some embodiments, the PECVD method is an expanding thermal plasma (ETP) method.

These and other aspects are described further below with reference to the figures.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing aspects and others will be readily appreciated by the skilled artisan from the following description of illustrative embodiments when read in conjunction with the accompanying drawings.

FIG. 1 is a schematic illustration of nanowires over which a silicon layer has been deposited using thermal CVD (chemical vapor deposition).

FIG. 2 is a schematic illustration of nanowires over which a silicon layer has been deposited using PECVD (plasma enhanced chemical vapor deposition).

FIG. 3 is a schematic illustration of nanowires over which a first silicon layer has been deposited using PECVD and then a second silicon layer has been deposited using thermal CVD, according to an embodiment of the invention.

FIG. 4 is a SEM (scanning electron microscope) top-view image of nanowires over which a silicon layer has been deposited using PECVD.

FIG. 5 is a SEM top-view image of nanowires over which a first silicon layer has been deposited using PECVD and then a second silicon layer has been deposited using thermal CVD, according to an embodiment of the invention.

FIG. 6 is a SEM cross-section view image of a nanostructure that has an inner core of PECVD silicon and an outer layer of thermal CVD silicon, according to an embodiment of the invention.

FIG. 7 shows a schematic depiction of a representation of a non-conformal silicon coating on a template nanowire.

FIGS. 8A and 8B are graphs shown capacity vs. cycle number and capacity retention vs. cycle number for electrodes having templated PECVD+thermal CVD (TCVD) Si layers, templated PECVD-only Si layers, and templated TCVD-only layers.

FIG. 9A is schematic representation of a plan view of a partially-assembled electrochemical cell that uses electrodes described herein, according to certain embodiments.

FIG. 9B is schematic representation of a cross-sectional view of an electrode stack of a partially-assembled electrochemical cell that uses electrodes described herein, according to certain embodiments.

FIGS. 10A-10C are schematic representations of various views of electrodes wound together with two sheets of separator to form a cell according to certain embodiments.

FIGS. 11A and 11B are schematic representations of cross-sectional and perspective views of a stacked cell that includes a plurality of cells according to certain embodiments.

FIG. 12 is schematic representation of cross-sectional view of a wound cylindrical cell, in accordance with certain embodiments.

DETAILED DESCRIPTION

Certain embodiments are illustrated in the context of silicon deposition onto silicide nanowires. The skilled artisan will readily appreciate, however, that the materials and methods disclosed herein will have application in a number of other contexts where it is useful to adjust deposition to produce layers with specific characteristics. For example, various embodiments are described herein with reference to nanowires. It should be understood, however, that unless otherwise stated, the references herein to nanowires include other types of nanostructures as described in U.S. Pat. No. 8,257,866, incorporated by reference herein, such as nanotubes, nanoparticles, nanospheres, nanorods, nanowhiskers, and the like.

Generally, the term “nanostructures” refers to structures having at least one dimension that is less than about 1 micron. This dimension could be, for example, a diameter of the nanostructure (e.g., a silicide template nanowire), a thickness of the shell formed over a template (e.g., a thickness of the amorphous silicon layers), or some other nanostructure dimension. It should be understood that any of the overall dimensions (length and diameter) of the final coated structure do not have to be at a nanoscale. For example, a final structure may include a layer that is about 10 microns in thickness at its greatest diameter and coated over a template that is about 100 nanometers in diameter and 20 microns in length. While this overall structure is about 10.1 microns at its greatest diameter and 20 microns in length, it could be generally referred to as a “nanostructure” because of the dimensions of the template. In specific embodiments, the term “nanowire” refers to structures with nano-scaled shells positioned over elongated template structures.

Nanowires (as a specific case of nanostructures) have an aspect ratio of greater than one, typically at least about two and more frequently at least about four. In specific embodiments, nanowires have an aspect ratio of at least about 10 and even at least about 100 or 500. Nanowires may make use of their one larger dimension to connect to other electrode components (e.g., a conductive substrate, other active material structures, or conductive additives). For example, nanowires may be substrate rooted such that one end (or some other part) of the majority of the nanowires is in contact with the substrate. Because the two other dimensions are small and there is an adjacent void volume available for expansion, the internal stress built up in the nanowires during lithiation (e.g., expansion of the nano-shells positioned over the silicide templates) is also small and does not break apart the nanowires (as happens with larger structures). Certain dimensions of the nanowires (e.g., an overall diameter and/or a shell thickness) are kept below the corresponding fracture levels of the active material used. Nanowires also permit a relatively high capacity per unit area of the electrode surface due to their elongated structure, which corresponds to the height of the template structure. This results from their relatively high aspect ratio and terminal connection to the substrate.

Silicon nanostructures can be made by first growing a nanowire template structure that is not pure silicon and then coating the template with silicon. Thermal CVD (chemical vapor deposition), HWCVD (hot-wire CVD), and/or PECVD (plasma enhanced chemical vapor deposition) may be used to deposit the silicon.

Various deposition processes produce different profiles when depositing silicon onto nanowires. For example, thermal CVD creates a conformal amorphous Si coating. HWCVD (also known as catalytic CVD) makes a high density, non-conformal amorphous Si coating that is thicker at the tips of the nanowires and thinner at the roots of the nanowires near the substrate. PECVD also produces non-conformal amorphous Si coating that is thicker at the ends of the nanowires and thinner at the roots of the nanowires near the substrate, but the coating has a low density with many small voids.

FIG. 1 is a schematic illustration of nanowires over which a conformal silicon layer has been deposited using thermal CVD. A nanowire template 110 is grown from a substrate 120. A silicon layer 130 is deposited onto the nanowire template 110. Note that the silicon layer 130 coats both the nanowire template 110 and the substrate 120, and that the coating has approximately the same thickness everywhere.

One advantage for the structure shown in FIG. 1 is that its structure is very dense. When an anode made using such a structure is cycled in a battery, the solid electrolyte interface (SEI) layer that forms on the silicon is very thin as the silicon formed by thermal CVD is very dense. One of the disadvantages for the structure shown in FIG. 1 is the thick layer of silicon at the roots of the nanowires. When an anode made using such a structure is cycled in a battery, there is great expansion and contraction in the silicon layer. Expansion in the root area can induce wrinkling of the Si layer and delamination of the nanowires, causing battery failure. Another disadvantage occurs because the nanowires in the template may not be exactly parallel as shown in FIG. 1. Instead, they grow at different angles and very often grow in clusters. The nanowires do not all have the same length. When coated by thermal CVD, the root area of the clusters, together with some short nanowires, can form a continuous silicon layer, which also can lead to delamination.

FIG. 2 is a schematic illustration of nanowires over which a silicon layer has been deposited using PECVD. Initially, PECVD may deposit a very thin layer (less than 1 micron and typically much less than 1 micron, such as 0.1-0.4 microns) of silicon on the substrate and at the roots of the nanowires. However, these areas are quickly shadowed by subsequent deposition at the tips of the nanowires. Depending on nanowire uniformity across the substrate surface, the very thin layer on the substrate may or may not be continuous. Under ideal surface and deposition conditions, a uniform thin layer may be expected but in practical cases, at the substrate-nanowire interface, there may be a discontinuity due to non-uniform distribution of nanowires on the surface.

In FIG. 2, a nanowire template 210 is grown from a substrate 220. A silicon layer 240 is deposited onto the nanowire template 210. Note that the silicon layer 240 is thickest at the tips of the nanowires of the nanowire template 210 and tapers off until there is essentially no silicon at the roots of nanowires. There is essentially no silicon on the substrate 220 either.

One advantage for the structure shown in FIG. 2 is that the root areas of the nanowires are coated only with only this very thin silicon layer. When an anode made using such a structure is cycled in a battery, the possibility of failure caused by expansion at the nanowire root area is greatly reduced. Another advantage is that PECVD coatings are not as dense as thermal CVD coatings; they can contain a large volume of voids and pores. Such defects can be very helpful in providing space into which the silicon can expand as it absorbs lithium.

One of the disadvantages of the structure shown in FIG. 2 is that the root area of the nanowires is not coated effectively. When an anode made using such a structure is cycled in a battery, the regions around the roots cannot participate electrochemically, thus causing a reduction in volumetric energy density of the battery. If more silicon were deposited onto the nanowires, the silicon layer at the tips of the nanowires would be too thick, causing fracturing of the silicon during cycling and subsequent anode failure. Another disadvantage is that the presence of voids can lead to large surface area of the coated nanowires and water absorption within the layer or at the surface. Very thick SEI layers can form as the battery is cycled, thus reducing the columbic efficiency and shortening the cycle life of the battery.

In one embodiment of the invention, the deposition methods discussed above are combined to provide optimum Si coatings as shown in FIG. 3 which is a schematic illustration of nanowires over which two silicon layers have been deposited. A nanowire template 310 is grown from a substrate 320. A first silicon layer 340 with a tapered profile is deposited onto the nanowire template 310 using PECVD. The first silicon layer 340 may be between about, for example, 0.5 and 10 microns thick at the top of the nanowire. In some embodiments, the first silicon layer may be thicker, e.g., between about 10 to 50 microns, or 10-20 microns. A second silicon layer 330 with a conformal profile is deposited onto the top of first silicon layer 340 using thermal CVD. The second silicon layer 330 may be, for example, between about 10 and 500 nm thick. The resulting structure has much more silicon at the tips of the nanowires than at the root ends. In more particular embodiments, the second silicon layer may be between 5 and 200 nm thick, or 10 to 90 nm thick. The second silicon layer 330 is conformal to the underlying surface, which includes the first silicon layer 340, the substrate 320, and any exposed portions of the nanowire template 310. As noted above, it has approximately uniform thickness.

According to various embodiments, the layer on the substrate surface may or may not be continuous. High performance may be achieved unless the layer on the substrate is too thick (e.g., greater than 2 microns) and continuous.

FIG. 4 is a SEM (scanning electron microscope) top-view image of nanowires over which a silicon layer has been deposited using PECVD.

FIG. 5 is a SEM top-view image of nanowires over which a first silicon layer has been deposited using PECVD and then a second silicon layer has been deposited using thermal CVD.

The novel structures described herein have many advantages. In some embodiments, there is more silicon near the tip of the nanowire than at the root, but there is still some silicon at the root. Having such a thin silicon layer at the root ensures that delamination does not occur during cycling. In some embodiments, there may also be electrochemical participation throughout the entire length of the nanowire. Further, SEI layer formation may be stabilized.

In some embodiments, the first silicon layer deposited is amorphous and has a low density, such as a density of about 1.70 g/cm³ or less, or 2.10 g/cm³ or less, or 2.2 g/cm³ or less, or less than 2.25 g/cm³ and may include many small voids. The second silicon layer deposited is amorphous and has a high density, such as a density of about 2.25 g/cm³ or more. The density of each layer is less than that of crystalline silicon.

According to various embodiments, the nanostructures may be characterized by a second silicon layer over a first silicon layer, the second silicon layer having a density higher than the density of the first silicon layer. As discussed above, in some embodiments, the first silicon layer provides space into which the silicon can expand as it absorbs lithium, while the second silicon layer reduces SEI layer formation. As such, the densities of each layer may be adjusted depending on the electrolyte, capacity of the battery, nanowire template density, etc. Accordingly, in some embodiments, the densities may be characterized in terms of a difference between densities of the layers, rather than or in addition to by absolute density. In some embodiments, the second silicon layer may have a density of at least 0.05 g/cm³ greater than the first silicon layer, or at least 0.1 g/cm³ greater than the first silicon layer, or at least 0.2 g/cm³ than the first silicon layer, or at least 0.3 g/cm³ greater than the first silicon layer.

FIG. 6 is a SEM cross-section view image of a nanostructure that has an inner core of PECVD silicon 640 and an outer layer of thermal CVD silicon 630, according to an embodiment of the invention. The porous morphology of the PECVD silicon 640 can be seen clearly as can the non-porous morphology of the thermal CVD silicon 630.

Aspects of the methods and structures disclosed herein may be implemented with other high capacity active materials, in addition to or instead of silicon. An electrochemically active material may have a theoretical lithiation capacity of at least about 500 mAh/g or, more specifically, of at least about 1000 mAh/g. Active materials with such capacities may be referred to as “high capacity active materials.” In addition to amorphous silicon, examples of high capacity active materials include silicon-containing compounds, tin and tin-containing compounds, germanium and germanium-containing compounds. For example, in some embodiments, an electrode may include a dual density germanium layer including a first inner layer having a lower density and second outer layer having a higher density.

In certain embodiments, high capacity active materials or templates are formed as substrate rooted nanostructures. These nanostructures may be physically and conductively attached to a conductive substrate, which may serve as a current collector for this electrode. The physical attachment may be more than a simple mechanical contact, which might result, for example, from coating a binder with discrete nanostructures onto the substrate. In some embodiments, the physical attachment results from fusion of the nanostructures to the substrate or deposition of the nanostructures or some portion of the nanostructures directly onto the substrate, for example, using CVD techniques or, even more, specifically using vapor-liquid-solid CVD growth. In yet another example, physical attachment results from ballistic impalement of the nanostructures onto the substrate. In certain embodiments, physical attachment includes certain forms of metallurgical bonds, such as a formation of alloys of two bonded materials (e.g., silicides).

In many embodiments, the nanowire template is a conductive material, with examples of conductive templates including metal templates and metal silicide templates. In some embodiments, a conductive template may include an oxide. As used herein, the term “conductive” refers broadly to electrical conductors, as distinct from semiconductors and insulators. The nanowire template may have conductivity of at least about 10³ S/m, or more specifically at least about 10⁶ S/m or even at least about 10³ S/m. Conductive templates can be useful to provide a transport path from the silicon electrochemically active material to a current collector as well as mechanically supporting the silicon layers. In some embodiments, however, the nanowire template may be a semiconductor or insulator (e.g., an oxide) that provides mechanical support to the silicon. Silicon nanowires may also be used as a template, with one or more coatings of a-Si deposited over the silicon nanowires as described herein. Silicon nanowire anodes are described in U.S. Pat. No. 7,816,031, incorporated by reference herein.

Nanowires of the nanowire template may be non-branched linear nanowires or branched nanowires. An electrode may include a combination of non-branched and branched nanowires, or only include only one of these types. While the templates may generally be elongated template structures, as described above, nanospheres, nanorods, nanowhiskers, nanoparticles, and the like may be employed. A nanowire template may be part of or include a multidimensional structure. One example of such a structure is a central core to which multiple nanowires are attached, forming “fuzzy ball-like” or “snowball-like” structures. An example of such a structure is shown in U.S. patent application Ser. No. 13/277,821, incorporated by reference herein.

In certain embodiments, nanowires in a nanowire template are between about 10 nanometers and 100 nanometers in diameter and between about 10 microns and 100 microns in length. In one example, a structure may include nanowires ranging from 5 to 40 microns. However, nanowires having other dimensions may be used.

Nanowire density can depend on the length of the nanowires, the desired capacity, the expansion ratio of the active material, and the particular application. If spacing between template structures is less than the coating thickness, it can cause significant interconnections of the active material layer. Interconnections near the roots can create agglomerated or continuous film like structures, which impede good cycle performance. The nanowires may be randomly distributed, with a variety of lengths and randomly-oriented. In some implementations, however, templated or guided methods of growth that produce uniform densities and/or orientations may be used. In one example, nanowires of a template structure may be grouped into various size bins, e.g., short, medium, and long sizes. Long nanowires may be identified as being visible in a top-down SEM image as depicted above. Example densities may be 0.5 to 20 long nanowires per 100 micron squared and 2 to 400 total nanowires per 100 microns squared. In some embodiments, a nanowire density may be determined for a particular mass loading. For example, for a mass loading between 2.5 and 2.9 mg/cm², a top diameter of the Si coating of 4 to 6 microns and a bottom diameter of 0.2 to 0.3 microns, a top nanowire density may be 2×10⁶ to 6×10⁶ long nanowires per centimeter squared.

The substrate is generally a conductive material having a conductivity of at least about 10³ S/m, or more specifically at least about 10⁶ S/m or even at least about 10⁷ S/m, particularly if the nanowire template is rooted to the substrate, and may be used as a current collector in the battery. This may be desirable when the substrate rooted structure is employed as a fully fabricated electrode for a battery or fuel cell. Examples of conductive substrate materials include copper, copper coated with metal oxides, stainless steel, titanium, aluminum, nickel, chromium, tungsten, other metals, metal silicides and other conductive metal compounds, carbon, carbon fiber, graphite, graphene, carbon mesh, conductive polymers, doped silicon or combinations of above including multi-layer structures. The substrate may be formed as a foil, film, mesh, foam, laminate, wires, tubes, particles, multi-layer structure, or any other suitable configuration. In certain embodiments, a substrate is a metallic foil with a thickness of between about 1 micron and 50 microns or more specifically between about 5 microns and 30 microns.

According to various embodiments, a silicon coating may be characterized by one or more of shape (also referred to as morphology), density, and bulk and surface composition. In terms of morphology, the first silicon layer may be generally characterized as non-conformal, having a thickness that is variable in the direction vertical to the substrate.

In some embodiments, a silicon layer has a generally circular symmetry. See, e.g., the top view SEM image in FIG. 4. It should be noted that an array of nanowires having generally circular symmetry includes arrays in which asymmetries may be introduced due to two nanowires being close enough that their coatings abut one another.

FIG. 7 shows a schematic depiction of a representation of a coating as a water drop or a conical frustum. The dimensions d1, d2 and h are labeled, with d1 being the largest diameter of the coating, d2 being the bottom diameter of the coating, and h being the height of the anode after coating. The non-conformal coating (porous non-conformal coating alone, or porous non-conformal coating conformally coated with the dense coating) may be characterized in some embodiments by the following ratios: d1/h of ½ to 1/9, d2/h of 1/400 to 1/70 and a d1/d2 ratio of 50:1 to 1.5:1.

In an example, d1 may be between 4 and 15 microns, or 4 and 12 microns, d2 may be between about 0.2 and 2 microns, and height may be between about 20 and 50 microns, e.g., between about 30 and 40 microns. The coating may extend between 10 and 20 microns above the height of the nanowire in some embodiments. The non-conformal layer substantially coats the nanowire, with the non-conformal coating extending at least most of the length of the nanowires, and in some embodiments, coating the entire template. As described above, there may be discontinuities near or at the root of the nanowire template.

In one example, a nanowire having a diameter of about 10 to 50 nm and a length of between about 10 to 25 microns is coated with silicon, such that after coating the diameter of the nanostructure at the root is 100 to 400 nm, the maximum diameter is 2 to 20 microns, and the total height of the anode is 20 to 40 microns.

In some embodiments, a non-conformal layer may be characterized by a hydrogen (H) content of at least 10%. In some embodiments, a conformal, dense layer may be characterized by a bulk H content of no more than 7%, or no more than 5%.

The non-conformal, porous silicon layers may be deposited by a method such as evaporation or other physical vapor deposition (PVD) method or HWCVD instead of or in addition to PECVD.

In PECVD processes, according to various implementations, a plasma may be generated in a chamber in which the substrate is disposed or upstream of the chamber and fed into the chamber. Any type of plasma, including capacitively-coupled plasmas, inductively-coupled plasmas, and conductive coupled plasmas may be used. Any appropriate plasma source may be used, including DC, AC, RF and microwave sources may be used.

PECVD process conditions can vary according to the particular process and tool used. A fairly wide range of temperatures, e.g., 180° C. to 600° C., may be used. Pressures are generally low for plasma processes, e.g., ranging from 50 mTorr to 400 Torr, or 200 mTorr to 10 Torr, depending on the process.

In some implementations, the PECVD process is an expanding thermal plasma chemical vapor deposition (ETP-CVD) process. In such a process, a plasma generating gas is passed through a direct current arc plasma generator to form a plasma, with a web or other substrate including the nanowire template in an adjoining vacuum chamber. A silicon source gas is injected into the plasma, with radicals generated. The plasma is expanded via a diverging nozzle and injected into the vacuum chamber and toward the substrate, with a non-conformal layer of amorphous silicon formed on the nanowire template. An example of a plasma generating gas is argon (Ar). In some embodiments, the ionized argon species in the plasma collide with silicon source molecules to form radical species of the silicon source, resulting in deposition on the nanowire template. Example ranges for voltages and currents for the DC plasma source are 60 to 80 volts and 50 to 70 amperes, respectively.

The conformal, dense silicon layers may be deposited by a method as atomic layer deposition (ALD) instead of or in addition to thermal CVD. Any appropriate thermal CVD process may be used, such as low pressure CVD (LPCVD). Temperatures may go as high as the thermal budget allows in some embodiments, as long as no metal silicide is formed around nanowire-substrate interface due to the increase of temperature when using a metal substrate.

Any appropriate silicon source may be used for the non-conformal and conformal silicon layers, including silane (SiH₄), dichlorosilane (H₂SiCl₂), monochlorosilane (H₃SiCl), trichlorosilane (HSiCl₃), and silicon tetrachloride (SiCl₄) to form the silicon layers. Depending on the gas used, the amorphous silicon layer may be formed by decomposition or a reaction with another compound, such as by hydrogen reduction. According to various embodiments, the same or a different silicon source may be used for the each of the layers.

According to various embodiments, the PECVD processes result in non-conformal coatings due to gas phase radical generation of the silicon precursor and gas phase nucleation and condensation on the surfaces of the template. By contrast, the thermal CVD reactions are run at conditions that result in surface reactions. High energy and mobility at the surface results in a conformal layer. In some embodiments, chamber pressure during a thermal CVD process is kept low, e.g., 100 mTorr to 2 Torr to prevent gas phase reactions and non-conformal deposition. Higher pressures, e.g., higher than 2 Torr or 500 Torr can result in non-conformal deposition.

Additional description of depositing active material layers having controlled densities may be found in U.S. patent application Ser. No. 13/277,821, incorporated by reference herein.

As indicated above, the nanostructures may be characterized by a second silicon layer over a first silicon layer, the second silicon layer having a density higher than the density of the first silicon layer. In some embodiments, the first silicon layer may be deposited directly on the nanowire template, with the second silicon layer being the outermost shell of the coated template. However, in some embodiments, other layers may be present. For example, a thin dense silicon layer may be deposited on a nanowire template, followed by a thicker porous layer, and a second thin dense layer.

Still further, a non-Si layer may be the outermost shell of the nanostructure in some embodiments. Examples of layers include metal oxides such as aluminum oxide, titanium oxide, cobalt oxide, and zirconium oxide, metal nitrides, and silicon nitrides. In some embodiments, a thin layer of any of these may be deposited in addition to or instead of the dense Si layer described above.

According to various embodiments, each of the first and second silicon layers may have a uniform density. However, in some embodiments, deposition conditions may be tuned during deposition to provide a density gradient in one or both layers. For example, either layer may get denser toward the outer part of the layer. In such embodiments, an average density of the layer may be used to characterize the density of layer as described above.

In some embodiments, silicon may be deposited by HWCVD. In some such implementations, a single Si layer having a density between that typically deposited by PECVD and that typically deposited by TCVD may be deposited. It can also be used to coat both the first and the second Si coatings by tuning the deposition process.

Experimental

Cycling data from nanowires with two layers of Si as described herein, shows that such structures have increased cycle lifetimes over structures that are made using only PECVD or only thermal CVD. FIGS. 8A and 8B are graphs shown capacity vs. cycle number and capacity retention vs. cycle number for electrodes having templated PECVD+thermal CVD (TCVD) Si layers, templated PECVD only Si layers, and template TCVD only layers. Capacity and capacity retention is highest for the PECVD+TCVD Si layers. As described above, the PECVD layer is a more porous, less dense internal layer and the TCVD layer is a denser external layer.

Assembly

FIG. 9A is a plan view of a partially-assembled electrochemical cell that uses electrodes described herein, according to certain embodiments. The cell has a positive electrode active layer 902 that is shown covering a major portion of a positive current collector 903. The cell also has a negative electrode active layer 904 that is shown covering a major portion of a negative current collector 905. Separator 906 is between the positive electrode active layer 902 and the negative electrode active layer 904.

In one embodiment, the negative electrode active layer 904 is slightly larger than the positive electrode active layer 902 to ensure trapping of the lithium ions released from the positive electrode active layer 902 by the active material of the negative electrode active layer 904. In one embodiment, the negative electrode active layer 904 extends at least between about 0.25 millimeters and 7 millimeters beyond the positive electrode active layer 902 in one or more directions. In a more specific embodiment, the negative electrode active layer 904 extends beyond the positive electrode active layer 902 by between about 1 millimeter and 2 millimeters in one or more directions. In certain embodiments, the edges of the separator 906 extend beyond the outer edges of at least the negative electrode active layer 904 to provide the complete electronic insulation of the negative electrode from the other battery components.

FIG. 9B is a cross-sectional view of an electrode stack 900 of the partially-assembled electrochemical cell that uses electrodes described herein, according to certain embodiments. There is a positive current collector 903 that has a positive electrode active layer 902a on one side and a positive electrode active layer 902b on the opposite side. There is a negative current collector 905 that has a negative electrode active layer 904a on one side and a negative electrode active layer 904b on the opposite side. There is a separator 906a between the positive electrode active layer 902a and the negative electrode active layer 904a. The separator sheets 906a and 906b serves to maintain mechanical separation between the positive electrode active layer 902a and the negative electrode active layer 904a and acts as a sponge to soak up the liquid electrolyte (not shown) that will be added later. The ends of the current collectors 903, 905, on which there is no active material, can be used for connecting to the appropriate terminal of a cell (not shown).

Together, the electrode layers 902a, 904a, the current collectors 903, 905, and the separator 906a can be said to form one electrochemical cell unit. The complete stack 900 shown in FIG. 9B, includes the electrode layers 902b, 904b and the additional separator 906b. The current collectors 903, 905 can be shared between adjacent cells. When such stacks are repeated, the result is a cell or battery with larger capacity than that of a single cell unit.

Another way to make a battery or cell with large capacity is to make one very large cell unit and wind it in upon itself to make multiple stacks. The cross-section schematic illustration in FIG. 10A shows how long and narrow electrodes can be wound together with two sheets of separator to form a battery or cell, sometimes referred to as a jellyroll 1000. The jellyroll is shaped and sized to fit the internal dimensions of a curved, often cylindrical, case 1002. The jellyroll 1000 has a positive electrode 1006 and a negative electrode 1004. The white spaces between the electrodes are the separator sheets. The jelly roll can be inserted into the case 1002. In some embodiments, the jellyroll 1000 may have a mandrel 1008 in the center that establishes an initial winding diameter and prevents the inner winds from occupying the center axial region. The mandrel 1008 may be made of conductive material, and, in some embodiments, it may be a part of a cell terminal. FIG. 10B shows a perspective view of the jelly roll 1000 with a positive tab 1012 and a negative tab 1014 extending from the positive current collector (not shown) and the negative current collector (not shown), respectively. The tabs may be welded to the current collectors.

The length and width of the electrodes depend on the overall dimensions of the cell and thicknesses of the active layers and the current collectors. For example, a conventional 18650-type cell with 18 mm diameter and 85 mm length may have electrodes that are between about 300 and 1000 mm long. Shorter electrodes corresponding to lower rate/higher capacity applications are thicker and have fewer winds.

A cylindrical design may be used for some lithium ion cells especially when the electrodes can swell during cycling and thus exert pressure on the casing. It is useful to use a cylindrical casing that is as thin as possible while still being able to maintain sufficient pressure on the cell (with a good safety margin). Prismatic (flat) cells may be similarly wound, but their case may be flexible so that they can bend along the longer sides to accommodate the internal pressure. Moreover, the pressure may not be the same within different parts of the cell, and the corners of the prismatic cell may be left empty. Empty pockets may be avoided within lithium ions cells because electrodes tend to be unevenly pushed into these pockets during electrode swelling. Moreover, the electrolyte may aggregate in empty pockets and leave dry areas between the electrodes, negatively affecting lithium ion transport between the electrodes. Nevertheless, for certain applications, such as those dictated by rectangular form factors, prismatic cells are appropriate. In some embodiments, prismatic cells employ stacks of rectangular electrodes and separator sheets to avoid some of the difficulties encountered with wound prismatic cells.

FIG. 10C illustrates a top view of a wound prismatic jellyroll 1020. The jellyroll 1020 includes a positive electrode 1024 and a negative electrode 1026. The white space between the electrodes is the separator sheet. The jelly roll 1020 is enclosed in a rectangular prismatic case 1022. Unlike the cylindrical jellyroll shown in FIG. 12, the winding of the prismatic jellyroll starts with a flat extended section in the middle of the jelly roll. In one embodiment, the jelly roll may include a mandrel (not shown) in the middle of the jellyroll onto which the electrodes and separator are wound.

FIG. 11A illustrates a cross-section of a stacked cell that includes a plurality of cells (1101a, 1101b, 1101c, 1101d, and 1101e), each having a positive electrode (e.g., 1103a, 1103b), a positive current collector (e.g., 1102), a negative electrode (e.g., 1105a, 1105b), a negative current collector (e.g., 1104), and a separator (e.g., 1106a, 1106b) between the electrodes. Each current collector is shared by adjacent cells. A stacked cell can be made in almost any shape, which is particularly suitable for prismatic batteries. The current collector tabs typically extend from the stack and lead to a battery terminal. FIG. 11B shows a perspective view of a stacked cell that includes a plurality of cells.

Once the electrodes are arranged as described above, the cell is filled with electrolyte. The electrolyte in lithium ions cells may be liquid, solid, or gel. The lithium ion cells with the solid electrolyte are referred to as a lithium polymer cells.

A typical liquid electrolyte comprises one or more solvents and one or more salts, at least one of which includes lithium. During the first charge cycle (sometimes referred to as a formation cycle), the organic solvent in the electrolyte can partially decompose on the negative electrode surface to form a SEI layer. The interphase is generally electrically insulating but ionically conductive, thereby allowing lithium ions to pass through. The interphase also prevents decomposition of the electrolyte in the later charging sub-cycles.

Some examples of non-aqueous solvents suitable for some lithium ion cells include the following: cyclic carbonates (e.g., ethylene carbonate (EC), propylene carbonate (PC), butylene carbonate (BC) and vinylethylene carbonate (VEC)), vinylene carbonate (VC), lactones (e.g., gamma-butyrolactone (GBL), gamma-valerolactone (GVL) and alpha-angelica lactone (AGL)), linear carbonates (e.g., dimethyl carbonate (DMC), methyl ethyl carbonate (MEC), diethyl carbonate (DEC), methyl propyl carbonate (MPC), dipropyl carbonate (DPC), methyl butyl carbonate (NBC) and dibutyl carbonate (DBC)), ethers (e.g., tetrahydrofuran (THF), 2-methyltetrahydrofuran, 1,4-dioxane, 1,2-dimethoxyethane (DME), 1,2-diethoxyethane and 1,2-dibutoxyethane), nitrites (e.g., acetonitrile and adiponitrile) linear esters (e.g., methyl propionate, methyl pivalate, butyl pivalate and octyl pivalate), amides (e.g., dimethyl formamide), organic phosphates (e.g., trimethyl phosphate and trioctyl phosphate), organic compounds containing an S═O group (e.g., dimethyl sulfone and divinyl sulfone), and combinations thereof.

Non-aqueous liquid solvents can be employed in combination. Examples of these combinations include combinations of cyclic carbonate-linear carbonate, cyclic carbonate-lactone, cyclic carbonate-lactone-linear carbonate, cyclic carbonate-linear carbonate-lactone, cyclic carbonate-linear carbonate-ether, and cyclic carbonate-linear carbonate-linear ester. In one embodiment, a cyclic carbonate may be combined with a linear ester. Moreover, a cyclic carbonate may be combined with a lactone and a linear ester. In a specific embodiment, the ratio of a cyclic carbonate to a linear ester is between about 1:9 to 10:0, preferably 2:8 to 7:3, by volume.

A salt for liquid electrolytes may include one or more of the following: LiPF₆, LiBF₄, LiClO₄LiAsF₆, LiN(CF₃SO₂)₂, LiN(C2F5SO2)2, LiCF3SO3, LiC(CF3SO2)3, LiPF4(CF3)2, LiPF3(C2F5)3, LiPF3(CF3)3, LiPF3(iso-C3F7)3, LiPF5(iso-C3F7), lithium salts having cyclic alkyl groups (e.g., (CF2)2(SO2)2xLi and (CF2)3(SO2)2xLi), and combinations thereof. Common combinations include LiPF6 and LiBF4, LiPF6 and LiN(CF3 SO2)2, LiBF4 and LiN(CF3SO2)2.

In one embodiment, the total concentration of salt in a liquid nonaqueous solvent (or combination of solvents) is at least about 0.3 M; in a more specific embodiment, the salt concentration is at least about 0.7M. The upper concentration limit may be driven by a solubility limit or may be no greater than about 2.5 M; in a more specific embodiment, it may be no more than about 1.5 M.

A solid electrolyte is typically used without the separator because it serves as the separator itself. It is electrically insulating, ionically conductive, and electrochemically stable. In the solid electrolyte configuration, a lithium containing salt, which could be the same as for the liquid electrolyte cells described above, is employed but rather than being dissolved in an organic solvent, it is held in a solid polymer composite. Examples of solid polymer electrolytes may be ionically conductive polymers prepared from monomers containing atoms having lone pairs of electrons available for the lithium ions of electrolyte salts to attach to and move between during conduction, such as polyvinylidene fluoride (PVDF) or chloride or copolymer of their derivatives, poly(chlorotrifluoroethylene), poly(ethylene-chlorotrifluoro-ethylene), or poly(fluorinated ethylene-propylene), polyethylene oxide (PEO) and oxymethylene linked PEO, PEO-PPO-PEO crosslinked with trifunctional urethane, poly(bis(methoxy-ethoxy-ethoxide))-phosphazene (MEEP), triol-type PEO crosslinked with difunctional urethane, poly((oligo)oxyethylene)methacrylate-co-alkali metal methacrylate, polyacrylonitrile (PAN), polymethylmethacrylate (PNMA), polymethylacrylonitrile (PMAN), polysiloxanes and their copolymers and derivatives, acrylate-based polymer, other similar solvent-free polymers, combinations of the foregoing polymers either condensed or cross-linked to form a different polymer, and physical mixtures of any of the foregoing polymers. Other less conductive polymers that may be used in combination with the above polymers to improve the strength of thin laminates include: polyester (PET), polypropylene (PP), polyethylene napthalate (PEN), polyvinylidene fluoride (PVDF), polycarbonate (PC), polyphenylene sulfide (PPS), and polytetrafluoroethylene (PTFE).

FIG. 12 illustrates a cross-section view of a wound cylindrical cell, in accordance with one embodiment. A jelly roll comprises a spirally wound positive electrode 1202, a negative electrode 1204, and two sheets of the separator 1206. The jelly roll is inserted into a cell case 1216, and a cap 1218 and gasket 1220 are used to seal the cell. It should be noted that in certain embodiments a cell is not sealed until after subsequent operations. In some cases, cap 1218 or cell case 1216 includes a safety device. For example, a safety vent or burst valve may be employed to open if excessive pressure builds up in the battery. In certain embodiments, a one-way gas release valve is included to release oxygen that has been released during activation of the positive material. Also, a positive thermal coefficient (PTC) device may be incorporated into the conductive pathway of cap 1218 to reduce the damage that might result if the cell suffered a short circuit. The external surface of the cap 1218 may be used as the positive terminal, while the external surface of the cell case 1216 may serve as the negative terminal. In an alternative embodiment, the polarity of the battery is reversed and the external surface of the cap 1218 is used as the negative terminal, while the external surface of the cell case 1216 serves as the positive terminal. Tabs 1208 and 1210 may be used to establish a connection between the positive and negative electrodes and the corresponding terminals. Appropriate insulating gaskets 1214 and 1212 may be inserted to prevent the possibility of internal shorting. For example, a Kapton™ film may be used for internal insulation. During fabrication, the cap 1218 may be crimped to the cell case 1216 in order to seal the cell. However, prior to this operation, electrolyte (not shown) is added to fill the porous spaces of the jelly roll.

A rigid case is typically used for lithium ion cells, while lithium polymer cells may be packed into flexible, foil-type (polymer laminate) cases. A variety of materials can be chosen for the cases. For lithium-ion batteries, Ti-6-4, other Ti alloys, Al, Al alloys, and 300 series stainless steels may be suitable for the positive conductive case portions and end caps, and commercially pure Ti, Ti alloys, Cu, Al, Al alloys, Ni, Pb, and stainless steels may be suitable for the negative conductive case portions and end caps.

In addition to the battery applications described above, the nanostructures may be used in fuel cells (e.g., for anodes, cathodes, and electrolytes), hetero-junction solar cell active materials, various forms of current collectors, and/or absorption coatings.

This invention has been described herein in considerable detail to provide those skilled in the art with information relevant to apply the novel principles and to construct and use such specialized components as are required. However, it is to be understood that the invention can be carried out by different equipment, materials and devices, and that various modifications, both as to the equipment and operating procedures, can be accomplished without departing from the scope of the invention itself

Claims

1. A method of making an anode for a lithium battery, comprising the steps of: providing a substrate;growing nanowires from the substrate;depositing a first silicon layer over the nanowires using a PECVD method;depositing a second silicon layer over the first silicon layer, the nanowires, and the substrate using a thermal CVD method.

2. The method of claim 1, wherein the PECVD method is an expanding thermal plasma (ETP) method.

3. The method of claim 1, wherein the nanowires are silicide nanowires.

4. The method of claim 1, wherein chamber pressure during the thermal CVD method is less than about 2 Torr.