WO2017015239A1 - Silicon nanoparticles and related methods of photothermal heating - Google Patents

Abstract

Disclosed herein are silicon nanoparticles (NPs) comprising amorphous silicon and an implanted metal dopant. The NPs are configured to provide improved photothermal (PT) characteristics compared to similarly sized crystalline silicon NPs. Methods of making the silicon NPs and methods of heating and therapy using the silicon NPs are also provided.

Description

SILICON NANOPARTICLES AND RELATED METHODS OF

PHOTOTHERMAL HEATING

CROSS-REFERENCE TO RELATED APPLICATION This application claims the benefit of U.S. Patent Application No. 62/193,892, filed July 17, 2015, the disclosure of which is hereby incorporated by reference in its entirety.

STATEMENT OF GOVERNMENT LICENSE RIGHTS

This invention was made with government support under Grant No. FA9550-12- 1-0400, awarded by the Air Force Office of Scientific Research. The government has certain rights in the invention.

BACKGROUND

Nanoparticle-mediated photothermal (PT) cancer therapy has been a major focus in nanomedicine due to its potential as an effective, noninvasive, and targeted alternative to traditional cancer therapy based on small-molecule pharmaceuticals. Gold nanocrystals have been a primary focus of PT research, which can be attributed to their size tunability, well-understood conjugation chemistry, and efficient absorption of NIR radiation in the tissue transparency window (800 nm to 1 μιη) due to their size-dependent localized surface plasmon resonances.

Recently, semiconducting nanomaterials have gained traction as PT agents owing to their synthetic simplicity and ability to tune their light-absorption properties independent of the nanoparticles' shape and size. Crystalline silicon nanowires (SiNWs) are one such semiconducting nanoparticle that has garnered interest as a PT agent. SiNWs are well suited for PT applications because they are biodegradable and can be made to be highly porous for drug loading and room-temperature photoluminescence applications.

A drawback of using crystalline SiNWs for PT therapy is that silicon is an inefficient absorber of light in the NIR due to its indirect bandgap. However, recent studies have shown that the optical absorption and subsequent PT heating can be drastically increased in crystalline SiNWs via ion implantation. Yet, even with PT heating enhancement via ion implantation, the irradiance needed to heat crystalline SiNWs to temperatures suitable for PT cancer therapy (42°C) is too large and would cause harmful damage to irradiated tissue.

Accordingly, while PT therapies have shown promise, new materials and methods are needed for these therapies to become viable and clinically relevant.

SUMMARY

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This summary is not intended to identify key features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

In one aspect a silicon nanoparticle is provided that includes amorphous silicon and gold.

In another aspect, a silicon nanoparticle is provided that includes amorphous silicon and a metal.

In another aspect, methods of making the nanowires are provided. In one embodiment, the method includes the step of ion implanting a crystalline nanowire with metal ions sufficient to produce a silicon nanoparticle comprising amorphous silicon and a metal.

In another aspect, a method of heating is provided. In one embodiment, the method includes irradiating a silicon nanoparticle as disclosed in any of the embodiments described herein.

In another aspect, a method of heating tissue is provided. In one embodiment, the method includes irradiating a silicon nanoparticle according to any of the embodiments described herein in proximity to the tissue, thereby heating the tissue.

In another aspect, a method of treating cancer in a subject is provided. In one embodiment, the method includes a silicon nanoparticle according to any of the embodiments described herein, in proximity to cancerous tissue in the subject, thereby heating the cancerous tissue and treating cancer in the subject.

DESCRIPTION OF THE DRAWINGS

The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein:

FIGURES 1A-1E. FIGURE 1A: Illustration showing the gold ion-implantation into the tips of the array of SiNWs. FIGURE IB: A helium ion microscope (HIM) image shows a magnified section of a post-implanted SiNW array. A color change of the SiNW array (FIGURE 1C) following silicon (FIGURE ID) and gold (FIGURE IE) ion implantation is evidence of increased light absorption. SRIM simulations of the ion implantation depth profiles are also presented for the silicon (FIGURE ID, inset) and gold (FIGURE IE, inset) ions.

FIGURES 2A-2F. TEM bright field images of (FIGURE 2A) a silicon-implanted and (FIGURE 2B) gold-implanted SiNW. Comparing the SAED of the tip of the silicon- implanted SiNW (FIGURE 2A, top-right inset) with the base of the same SiNW (FIGURE 2A, bottom-right inset) shows partial amorphization throughout the SiNW as predicted by SRIM calculations. EDX spectra of the same regions (FIGURE 2B) also suggest no change in elemental composition. Comparing the SAED of the tip of the gold- implanted SiNW (FIGURE 2C, top-right inset) with the base of the same SiNW (FIGURE 2C, bottom-right inset) suggests that amorphization from gold implantation is contained to one end of the SiNW. EDX spectra of the same regions (FIGURE 2D) further suggest that the gold is constrained to the SiNW tip. Increased amorphization and lattice damage from increasing (FIGURE 2E) silicon and (FIGURE 2F) gold implantation dosage is also evidenced by the presence of an increasing shoulder off the silicon 520 cm^-1 Raman peak.

FIGURES 3A-3C. FIGURE 3A: Sketch of the optical tweezers setup. Stage oscillation moves the surrounding water with respect to the (FIGURE 3B) trapped SiNW undergoing HBM and (FIGURE 3C) produces a peak in the power spectrum at the oscillation frequency. The height of the power spectrum peak is used to calibrate the quadrant photodiode signal, as outlined in ref.

FIGURE 4. Results from the HBM temperature extraction method of optically trapped ion-implanted SiNWs using a 975 nm focused laser source. The error bars represent the standard deviation in the extracted temperature between implanted SiNWs.

FIGURES 5A-5D. (FIGURE 5A) SEM image of a SiNW on a coverslip of a trapping chamber. (FIGURE 5B) A CCD camera image of the same SiNW in the water- filled trapping chamber. Image pixel intensity line profiles with gaussian fits from the (FIGURE 5C) length and (FIGURE 5D) diameter of the same SiNW from FIGURE 5B. Dashed lines across the gaussian peaks indicate the full width at half maximum that is used to calculate the size of the nanowires. Here, the size of the SiNW from SEM analysis is 10.07 μιη long and 290 nm + 36 nm in diameter, whereas the size extraction method gives 10.12 μιη long and 312 nm + 58 nm in diameter.

FIGURES 6A-6D. (FIGURE 6A) Diagram of single SiNW temperature extraction for data presented in (FIGURE 6C). (FIGURE 6B) Diagram of SiNW array temperature measurement for data presented in (FIGURE 6D). (FIGURE 6C) Plot of the average temperature increase for each doping level shown in FIGURE 4 normalized to the maximum temperature increase at each irradiance. (FIGURE 6D) Plot of an array of SiNWs heated in air by an unfocused 975 nm laser with a beam diameter of 3 mm and laser power of 50 mW measured with a thermocouple out of the beam path.

FIGURES 7A-7C. (FIGURE 7A) Video frames of a gold ion-implanted SiNW in Brownian motion before the 975 nm trapping laser is turned on and the SiNW is trapped (FIGURE 7B). (FIGURE 7C) The SiNW is then brought in proximity of other gold ion- implanted SiNWs that are attached to the glass coverslip (above and to the left of bubble) and a bubble is generated.

FIGURE 8. Image of a gold ion-implanted SiNW after an unsuccessful trapping attempt showing a morphology change on one side of the SiNW. Before the trapping attempt, the end of the SiNW was not curved.

FIGURES 9A-9E. Compositional analysis and microstructure of silicon nanowires prepared via metal-assisted chemical etching (MACE). (FIGURE 9A) High- angle annular dark-field image of a single silicon nanowire without silver etching, demonstrating the presence of silver deposits. Scale bar = 50 nm. (FIGURE 9B) High- angle annular dark-field image of a silver-etched silicon nanowire with no detectable silver. Scale bar = 50 nm. (FIGURE 9C) Neutron activation analysis of silicon nanowire array before (i) and after (ii) the silver-etching process. (FIGURE 9D) Atom probe tomography (APT) mass spectrum from a single SiNW, demonstrating no detectable silver signal (Ag⁺: 107 Da, Ag²⁺: 53.5 Da). (FIGURE 9E) Photoluminescence of SiNWs excited by a 975 nm laser source.

FIGURES 10A-10D. Schematic of laser-trapping instrument and two-photon photoexcitation of both silicon wires and SOSG-EP molecules. (FIGURE 10A) Schematic outlining the components used in the trapping experiments. (FIGURE 10B) Depiction of a nanowire trapped with the focused Gaussian, NIR laser (beam waist = 1.1 μιη) in a solution of SOSG with localized pumping of reacted SOSG. (FIGURE IOC) Silicon-water interface diagram showing the transfer of electrons from surface excitons in silicon to dissolved oxygen molecules. (FIGURE 10D) Structure of SOSG before reaction with singlet oxygen and its associated endoperoxide (SOSG-EP) after reaction with singlet oxygen.

FIGURES 11A-11G. Singlet-oxygen (^₂) generation from silicon nanowires. Digital micrographs of (FIGURE 11 A) a single optically trapped silicon nanowire in SOSG solution, (FIGURE 11B) a single optically trapped silicon nanowire in water alone, (FIGURE 11C) illumination of a chamber of SOSG solution, and (FIGURE 1 ID) a single optically trapped silica bead in a solution of SOSG. (FIGURE HE) Absorption and emission spectra for SOSG-EP; the blue vertical line represents the two-photon wavelength for a 975 nm trapping source. (FIGURE 11F) SOSG-EP emission from a single trapped SiNW as a function of laser irradiance, demonstrating the nonlinearity of SOSGEP excitation. (FIGURE 11G) Comparison between a solution of SiNWs and SOSG and a solution of only SOSG. Both solutions were irradiated by a 975 nm laser at equal powers and, at each data point, were illuminated with a 405 nm diode to obtain the momentary emission spectrum from generated SOSG-EP. The data demonstrate an enhanced generation of singlet oxygen when SiNWs are present.

FIGURES 12A-12D. Singlet-oxygen (^₂) generation from a single optically trapped gold nanowire. (FIGURE 12A) Bright-field micrograph of an electrochemically synthesized gold nanorod (AuNR) optically trapped in a solution of SOSG using the same setup used for SiNWs, diagrammed in FIGURE 10A. (FIGURE 12B) Micrograph of the SOSG-EP emission from the AuNR trapped in (FIGURE 12A). (FIGURE 12C) Micrograph of an optically trapped AuNR in water alone. (FIGURE 12D) Micrograph of the AuNR in (FIGURE 12C) demonstrating no emission in the absence of SOSG. All scale bars = 2 μιη.

DETAILED DESCRIPTION

Disclosed herein are silicon nanoparticles (NPs) comprising amorphous silicon and an implanted metal dopant. The NPs are configured to provide improved photothermal (PT) characteristics compared to similarly sized crystalline silicon NPs. Methods of making the silicon NPs and methods of heating and therapy using the silicon NPs are also provided.

Nanoparticles

In one aspect a silicon nanoparticle is provided that includes amorphous silicon and gold. Fabrication and use for heating of these nanoparticles is disclosed in Example 1. These silicon nanoparticles have the PT heating ability of gold nanoparticles but in a biodegradable material that is primarily silicon with a relatively small amount of gold ions implanted in the nanoparticle.

Unexpectedly, by creating silicon NPs with some, or all, amorphous character, the disclosed NPs provide a greater absorber of clinically relevant radiation (e.g., near-IR and non-damaging intensity), to the extent that clinically relevant heating (e.g., to greater than 42°C) is possible. Silicon-based nanoparticles are typically crystalline, as formed, and thus far there has been no impetus to transform these crystalline NPs into amorphous NPs, as disclosed herein.

By using gold ion implantation on crystalline silicon NPs, the lattice is disrupted and amorphous regions are created that result in significant heating when illuminated with certain wavelengths of light— effectively by an increase in the absorbance cross- section. With sufficient ion implantation energy and dose, the NPs may become entirely amorphous. The implanted ions remain in the lattice and may also contribute to the overall absorbance and PT effect of the NPs. As gold is s significant PT absorber, the use of gold ions is a particularly well-suited ion implantation material. Unlike nanoparticles formed entirely from gold, however, the present silicon nanoparticles are biocompatible and therefore can be introduced into tumors for treating cancer using PT therapy, as will be described in more detail below.

In one embodiment, the silicon NP is substantially (i.e., at least 90%, by weight of the NP) amorphous. In one embodiment, the silicon NP is at least 95% amorphous, by weight of the NP. In one embodiment, the silicon NP is at least 99% amorphous, by weight of the NP.

In other embodiments, a smaller portion of the silicon NP is converted to amorphous silicon, so as to balance the PT effect of amorphous silicon with the benefits of crystalline silicon, including the ability to generate singlet oxygen, as disclosed in Example 2, for potential photodynamic therapy or other reactive oxygen species (ROS) applications. Accordingly, in one embodiment, the silicon NP is from 10% to 90% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 20% to 80% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 30% to 70% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 40% to 60% amorphous, by weight of the NP.

In one embodiment, the gold is at least 1%, by weight, of the nanoparticle. Trace amounts of gold are not sufficient to disrupt the crystalline silicon lattice. However, as discussed in embodiments below, the amount of gold should not exceed a threshold amount where relatively large aggregates are formed. Accordingly, in certain embodiments, the gold is 10% or less, by weight, of the nanoparticle.

In one embodiment, the nanoparticle consists essentially of at least partially amorphous silicon and gold. In a further embodiment the nanoparticle is a nanowire.

The silicon NPs can be formed in any shape that provides PT heating. Representative shapes include spheres, polygons, including cubes, wires, and rods.

In one embodiment, the silicon nanoparticle is a silicon nanowire. Silicon nanowires are particularly useful due to their biodegradable nature and synthetic simplicity. The size (e.g., length) of silicon nanowires can be synthetically tailored and those dimensions translate into absorbance characteristics. Accordingly, known methods can be used to form crystalline silicon nanowires of the desired dimensions. These nanowires can then be ion implanted with gold in order to form partially or completely amorphous silicon nanowires with desired absorbance wavelengths.

In one embodiment, the nanowire is configured to absorb an irradiation wavelength of 650 nm to 1350 nm. In one embodiment, the nanowire is configured to absorb an irradiation wavelength of 785 nm to 900 nm. In one embodiment, the nanowire is configured to absorb an irradiation wavelength of 650 nm to 1000 nm. In one embodiment, the nanowire is configured to achieve heating of greater than 42°C when irradiated.

Furthermore, nanowires maximize surface area compared to spheres, which increases absorbance area. Still further, nanowires facilitate endocytosis, which enables PT within cells.

The length of nanowires affects absorbance properties. In a further embodiment, the silicon nanowire has a length of 100 nm or greater. In a further embodiment, the silicon nanowire has a length of 500 nm or greater. In a further embodiment, the silicon nanowire has a length of 1000 nm or greater. The dimensions of the nanowires can also be defined in terms of aspect ratio. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 2: 1. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 5: 1. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 10: 1. In yet a further embodiment, the nanowire has a length of 100 nm or greater and an aspect ratio according to the embodiments of this paragraph.

In one embodiment, the smallest dimension of the nanoparticle is from 50 nm to 500 nm. As used herein, the smallest dimension of the nanoparticle is the smallest dimension between length, width, height, and diameter.

Turning now to the gold implanted in the silicon nanoparticle, in one embodiment, the gold concentration is from 10¹⁵ to 10²⁵ ions/cm³. This doping range provides the necessary amorphous character to the nanoparticle such that the absorbance cross section is increased and PT heating is increased substantially compared to all-crystalline silicon nanoparticles.

As a result of the ion implantation process, gold ions are deposited ("doped") into the silicon lattice, disrupting the crystal structure and leaving amorphous silicon in the affected volume. At higher ion concentrations, the gold will aggregate. Larger aggregates are typically not desired. In one embodiment, the gold is not present in aggregates larger than 0.5 nm. In one embodiment, the gold is not present in aggregates larger than 1 nm. In one embodiment, the gold is not present in aggregates larger than 2 nm. In one embodiment, the gold is not present in aggregates larger than 3 nm. In one embodiment, the gold is not present in aggregates larger than 4 nm. In one embodiment, the gold is not present in aggregates larger than 1 nm. In one embodiment, the gold is not present in aggregates larger than 5 nm. In one embodiment, the gold is not present in aggregates larger than 10 nm. In one embodiment, the gold is not present in aggregates larger than 15 nm. In one embodiment, the gold is not present in aggregates larger than 25 nm.

In one embodiment, the amorphous gold is evenly distributed within the silicon nanoparticle. Even gold distribution yields even amorphous character throughout the nanoparticle, which produces consistent PT performance. As used herein, the term "evenly" describes a distribution of gold atoms within the nanoparticle, wherein the number of gold atoms per 5 nm³ volume is within 10% of adjacent 5 nm³ volumes. In one embodiment, at least a portion of the gold is amorphous gold. Amorphous gold is typically extremely small aggregates or non-aggregated atoms. Because the gold is intended to act as a lattice disruptor and not an active species within the nanoparticle, aggregates are not favored in most embodiments. The presence of aggregates typically indicates inefficiency due to over-implantation of ions and thus additional preparation cost and time.

While gold-doped nanoparticles have been primarily disclosed, it will be appreciated that other metals besides gold can also be used to disrupt the crystalline silicon lattice to form at least partially amorphous silicon nanoparticles. Accordingly, in another aspect, a silicon nanoparticle is provided that includes amorphous silicon and a metal. In one embodiment the metal dopant comprises iron. Any metal capable of ion implantation into crystalline silicon sufficient to disrupt the lattice and form at least partially amorphous silicon is compatible with this aspect. Similarly, the previously disclosed embodiments related to gold are compatible with the generic metal aspects.

In one embodiment, the metal is selected from the group consisting of gold, iron, and combinations thereof.

In one embodiment, the silicon NP is substantially (i.e., at least 90%, by weight of the NP) amorphous. In one embodiment, the silicon NP is at least 95% amorphous, by weight of the NP. In one embodiment, the silicon NP is at least 99% amorphous, by weight of the NP.

In other embodiments, a smaller portion of the silicon NP is converted to amorphous silicon, so as to balance the PT effect of amorphous silicon with the benefits of crystalline silicon, including the ability to generate singlet oxygen, as disclosed in Example 2, for potential photodynamic therapy or other reactive oxygen species (ROS) applications. Accordingly, in one embodiment, the silicon NP is from 10% to 90% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 20% to 80% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 30% to 70% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 40% to 60% amorphous, by weight of the NP.

In one embodiment, the metal is at least 1%, by weight, of the nanoparticle. Trace amounts of metal are not sufficient to disrupt the crystalline silicon lattice. However, as discussed in embodiments below, the amount of metal should not exceed a threshold amount where relatively large aggregates are formed. Accordingly, in certain embodiments, the metal is 10% or less, by weight, of the nanoparticle.

The silicon NPs can be formed in any shape that provides PT heating. Representative shapes include spheres, polygons, including cubes, wires, and rods.

In one embodiment, the silicon nanoparticle is a silicon nanowire. Silicon nanowires are particularly useful due to their biodegradable nature and synthetic simplicity. The size (e.g., length) of silicon nanowires can be synthetically tailored and those dimensions translate into absorbance characteristics. Accordingly, known methods can be used to form crystalline silicon nanowires of the desired dimensions. These nanowires can then be ion implanted with metal in order to form partially or completely amorphous silicon nanowires with desired absorbance wavelengths.

The length of nanowires affects absorbance properties. In a further embodiment, the silicon nanowire has a length of 100 nm or greater. In a further embodiment, the silicon nanowire has a length of 500 nm or greater. In a further embodiment, the silicon nanowire has a length of 1000 nm or greater.

The dimensions of the nanowires can also be defined in terms of aspect ratio. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 2: 1. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 5: 1. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 10: 1. In yet a further embodiment, the nanowire has a length of 100 nm or greater and an aspect ratio according to the embodiments of this paragraph.

In one embodiment, the smallest dimension of the nanoparticle is from 50 nm to 500 nm. As used herein, the smallest dimension of the nanoparticle is the smallest dimension between length, width, height, and diameter.

Turning now to the metal is implanted in the silicon nanoparticle, in one embodiment, the metal concentration is from 10¹⁵ to 10²⁵ ions/cm³. This doping range provides the necessary amorphous character to the nanoparticle such that the absorbance cross section is increased and PT heating is increased substantially compared to all- crystalline silicon nanoparticles.

As a result of the ion implantation process, metal ions are deposited ("doped") into the silicon lattice, disrupting the crystal structure and leaving amorphous silicon in the affected volume. At higher ion concentrations, the metal will aggregate. Larger aggregates are typically not desired. In one embodiment, the metal is not present in aggregates larger than 0.5 nm. In one embodiment, the metal is not present in aggregates larger than 1 nm. In one embodiment, the metal is not present in aggregates larger than 2 nm. In one embodiment, the metal is not present in aggregates larger than 3 nm. In one embodiment, the metal is not present in aggregates larger than 4 nm. In one embodiment, the metal is not present in aggregates larger than 1 nm. In one embodiment, the metal is not present in aggregates larger than 5 nm. In one embodiment, the metal is not present in aggregates larger than 10 nm. In one embodiment, the metal is not present in aggregates larger than 15 nm. In one embodiment, the metal is not present in aggregates larger than 25 nm.

In one embodiment, the amorphous metal is evenly distributed within the silicon nanoparticle. Even metal distribution yields even amorphous character throughout the nanoparticle, which produces consistent PT performance. As used herein, the term "evenly" describes a distribution of metal atoms within the nanoparticle, wherein the number of metal atoms per 5 nm³ volume is within 10% of adjacent 5 nm³ volumes.

In one embodiment, at least a portion of the metal is amorphous metal.

Methods of Making the Nanoparticles

In another aspect, methods of making the nanowires are provided. In one embodiment, the method includes the step of ion implanting a crystalline nanowire with metal ions sufficient to produce a silicon nanoparticle comprising amorphous silicon and a metal. In one embodiment, the metal is selected from the group consisting of gold, iron, and combinations thereof. Metal ion-implanting techniques are generally known to those of skill in the art.

In one embodiment, the silicon NP is substantially (i.e., at least 90%, by weight of the NP) amorphous. In one embodiment, the silicon NP is at least 95% amorphous, by weight of the NP. In one embodiment, the silicon NP is at least 99% amorphous, by weight of the NP.

In other embodiments, a smaller portion of the silicon NP is converted to amorphous silicon, so as to balance the PT effect of amorphous silicon with the benefits of crystalline silicon, including the ability to generate singlet oxygen, as disclosed in Example 2, for potential photodynamic therapy or other reactive oxygen species (ROS) applications. Accordingly, in one embodiment, the silicon NP is from 10% to 90% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 20% to 80% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 30% to 70% amorphous, by weight of the NP. In one embodiment, the silicon NP is from 40% to 60% amorphous, by weight of the NP.

In one embodiment, the metal is at least 1%, by weight, of the nanoparticle. Trace amounts of metal are not sufficient to disrupt the crystalline silicon lattice. However, as discussed in embodiments below, the amount of metal should not exceed a threshold amount where relatively large aggregates are formed. Accordingly, in certain embodiments, the metal is 10% or less, by weight, of the nanoparticle.

In one embodiment, the silicon nanoparticle is a silicon nanowire.

The length of nanowires affects absorbance properties. In a further embodiment, the silicon nanowire has a length of 100 nm or greater. In a further embodiment, the silicon nanowire has a length of 500 nm or greater. In a further embodiment, the silicon nanowire has a length of 1000 nm or greater.

The dimensions of the nanowires can also be defined in terms of aspect ratio. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 2: 1. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 5: 1. In a further embodiment, the silicon nanowire has an aspect ratio (length: width) of at least 10: 1. In yet a further embodiment, the nanowire has a length of 100 nm or greater and an aspect ratio according to the embodiments of this paragraph.

In one embodiment, the smallest dimension of the nanoparticle is from 50 nm to 500 nm. As used herein, the smallest dimension of the nanoparticle is the smallest dimension between length, width, height, and diameter.

Turning now to the metal implanted in the silicon nanoparticle, in one embodiment, the metal concentration is from 10¹⁵ to 10²⁵ ions/cm³. This doping range provides the necessary amorphous character to the nanoparticle such that the absorbance cross section is increased and PT heating is increased substantially compared to all- crystalline silicon nanoparticles.

As a result of the ion implantation process, metal ions are deposited ("doped") into the silicon lattice, disrupting the crystal structure and leaving amorphous silicon in the affected volume. At higher ion concentrations, the metal will aggregate. Larger aggregates are typically not desired. In one embodiment, the metal is not present in aggregates larger than 0.5 nm. In one embodiment, the metal is not present in aggregates larger than 1 nm. In one embodiment, the metal is not present in aggregates larger than 2 nm. In one embodiment, the metal is not present in aggregates larger than 3 nm. In one embodiment, the metal is not present in aggregates larger than 4 nm. In one embodiment, the metal is not present in aggregates larger than 1 nm. In one embodiment, the metal is not present in aggregates larger than 5 nm. In one embodiment, the metal is not present in aggregates larger than 10 nm. In one embodiment, the metal is not present in aggregates larger than 15 nm. In one embodiment, the metal is not present in aggregates larger than 25 nm.

In one embodiment, the amorphous metal is evenly distributed within the silicon nanoparticle. Even metal distribution yields even amorphous character throughout the nanoparticle, which produces consistent PT performance. As used herein, the term "evenly" describes a distribution of metal atoms within the nanoparticle, wherein the number of metal atoms per 5 nm³ volume is within 10% of adjacent 5 nm³ volumes.

In one embodiment, at least a portion of the metal is amorphous metal.

Method of Heating

In another aspect, a method of heating is provided. In one embodiment, the method includes irradiating a silicon nanoparticle as disclosed in any of the embodiments described herein. In one embodiment the silicon nanoparticle includes amorphous silicon and a metal. In one embodiment, the metal is selected from the group consisting of gold, iron, and combinations thereof.

Irradiating the silicon nanoparticle includes directing electromagnetic radiation onto the nanoparticle. The electromagnetic radiation may come from any source, including an LED, laser, or lamp. Any source that can provide the appropriate radiation, including wavelength and intensity, is compatible with the disclosed methods. In one embodiment the source is a narrow-band source, with a particular bandwidth tuned to wavelengths compatible with human tissue. In another embodiment, the source is a broad-band source.

In one embodiment, the method further includes heating a volume in proximity to the silicon nanoparticle by transferring heat from the silicon nanoparticle to the volume. As used herein, the term "in proximity to" is defined as a volume containing the nanoparticle or sufficiently near the nanoparticle to receive heat transferred from the nanoparticle after heating by irradiation. By this step, heating the nanoparticle is used to heat a volume around the nanoparticle so as to provide targeted heat, activated by illumination; such as photothermal therapy. The volume can be liquid, solid, gas, or any combination thereof. In one embodiment, the volume is heated to a temperature of 38°C to 50°C. In one embodiment, the volume is heated to a temperature greater than 42°C.

In one embodiment, the method further includes heating a plurality of the silicon nanoparticles. While a single silicon nanoparticle may be effective in a nano- or micron- scale environment, greater volume can be heated by irradiating a plurality of the silicon nanoparticles.

In one embodiment, irradiating the silicon nanoparticle comprises irradiating at an intensity of 5 MW/cm² to 15 MW/cm².

Selected wavelength ranges for irradiation are desirable as being clinically relevant due to transparency windows in tissues (e.g., mammalian tissue) that allow electromagnetic radiation of select wavelengths to penetrate the tissue and absorb only in the targeted locations where the nanoparticle or plurality of nanoparticles are disposed. Due to the efficient absorption of the nanoparticles, photothermal heating to significant temperatures can be achieved without harming the skin of a treatment subject. In one embodiment, irradiating the silicon nanoparticle comprises an irradiation wavelength of 650 nm to 1350 nm. In one embodiment, irradiating the silicon nanoparticle comprises an irradiation wavelength of 785 nm to 900 nm. In one embodiment, irradiating the silicon nanoparticle comprises an irradiation wavelength of 650 nm to 1000 nm.

In one embodiment, the method further includes the silicon nanoparticle generating singlet oxygen in response to irradiating the silicon nanoparticle. As discussed in Example 2, silicon nanoparticles can be used to generate singlet oxygen.

Methods of Therapy

In another aspect, a method of heating tissue is provided. In one embodiment, the method includes irradiating a silicon nanoparticle according to any of the embodiments described herein in proximity to the tissue, thereby heating the tissue.

As used herein, the term "in proximity to" is defined as a volume containing the nanoparticle or sufficiently near the nanoparticle to receive heat transferred from the nanoparticle after heating by irradiation. By this step, heating the nanoparticle is used to heat a volume around the nanoparticle so as to provide targeted heat, activated by illumination; such as photothermal therapy. In the present aspect, the volume comprises tissue targeted for heating. In one embodiment the silicon nanoparticle includes amorphous silicon and a metal. In one embodiment, the metal is selected from the group consisting of gold, iron, and combinations thereof.

In one embodiment, the tissue is heated to a temperature of 38°C to 50°C. In one embodiment, the tissue is heated to a temperature greater than 42°C.

In one embodiment, the method further includes heating a plurality of the silicon nanoparticles. While a single silicon nanoparticle may be effective in a nano- or micron- scale environment, greater volume of tissue can be heated by irradiating a plurality of the silicon nanoparticles.

In one embodiment, irradiating the silicon nanoparticle comprises irradiating at an intensity of 5 MW/cm² to 15 MW/cm².

Selected wavelength ranges for irradiation are desirable as being clinically relevant due to transparency windows in tissues (e.g., mammalian tissue) that allow electromagnetic radiation of select wavelengths to penetrate the tissue and absorb only in the targeted locations where the nanoparticle or plurality of nanoparticles are disposed. Due to the efficient absorption of the nanoparticles, photothermal heating to significant temperatures can be achieved without harming the skin of a treatment subject. In one embodiment, irradiating the silicon nanoparticle comprises an irradiation wavelength of 650 nm to 1350 nm. In one embodiment, irradiating the silicon nanoparticle comprises an irradiation wavelength of 785 nm to 900 nm. In one embodiment, irradiating the silicon nanoparticle comprises an irradiation wavelength of 650 nm to 1000 nm.

In one embodiment, the method further includes the silicon nanoparticle generating singlet oxygen in response to irradiating the silicon nanoparticle. As discussed in Example 2, silicon nanoparticles can be used to generate singlet oxygen.

The type of tissue heated can be any tissue desired to be heated, including tumors and cancerous tissue. Photothermal treatment of cancer is a developing field and the disclosed methods and materials can be integrated into any known photothermal treatment method. Accordingly, in one embodiment the tissue is a cancerous tissue and the method comprises photothermal therapy to treat the cancer.

Furthermore, in another aspect, a method of treating cancer in a subject is provided. In one embodiment, the method includes a silicon nanoparticle according to any of the embodiments described herein, in proximity to cancerous tissue in the subject, thereby heating the cancerous tissue and treating cancer in the subject. Embodiments of this aspect are disclosed above generally with regard to tissue therapy. Those embodiments are hereby applied to the present aspect of a method of treating cancer in a subject. In one embodiment the subject is a mammal. In another embodiment, the subject is a human.

The following examples are included for the purpose of illustrating, but not limiting, the described embodiments.

EXAMPLES

Example 1: Photothermal Superheating of Water with Ion-Implanted Silicon Nanowires.

Herein, we apply a new method for extracting temperatures of trapped nanoparticles undergoing hot Brownian motion in laser tweezers and show that we can drastically increase PT heating of ion-implanted SiNWs in a single-beam laser trap by using gold as the primary dopant ion to superheat water above 200°C with -10 MW cm^-2 irradiances. Furthermore, we explore the role that the dopant ion plays in the PT heating process by comparing SiNWs implanted with gold ions to those implanted with silicon ions at several implantation doses for each ion. The hot Brownian motion temperature extraction analysis is shown to be a versatile technique for quantifying heating effects that arise during laser-trapping experiments. The ion-implanted SiNW heating results allow for the future possibility of using the gold- implanted SiNWs as a PT cancer agent at clinically relevant laser irradiances.

Nanoparticle-mediated photothermal (PT) cancer therapy has been a major focus in nanomedicine due to its potential as an effective, noninvasive, and targeted alternative to traditional cancer therapy based on small-molecule pharmaceuticals. Gold nanocrystals have been a primary focus of PT research, which can be attributed to their size tunability, well-understood conjugation chemistry, and efficient absorption of NIR radiation in the tissue transparency window (800 nm to 1 μιη) due to their size-dependent localized surface plasmon resonances.

Recently, semiconducting nanomaterials have gained traction as PT agents owing to their synthetic simplicity and ability to tune their light-absorption properties independent of the nanoparticles' shape and size. Silicon nanowires (SiNWs) are one such semiconducting nanoparticle that has garnered interest as a PT agent. SiNWs are well suited for PT applications since they are biodegradable, generate singlet oxygen upon NIR photoexcitation, and can be made to be highly porous for drug loading and room- temperature photoluminescence applications. A drawback of using SiNWs for PT therapy is that silicon is an inefficient absorber of light in the NIR due to its indirect bandgap. However, recent studies have shown that the optical absorption and subsequent PT heating can be drastically increased in SiNWs via ion implantation. Yet, even with PT heating enhancement via ion implantation, the irradiance needed to heat SiNWs to temperatures suitable for PT cancer therapy (42°C) is too large and would cause harmful damage to irradiated tissue. Herein, we apply a new method for extracting temperatures of trapped nanoparticles undergoing hot Brownian motion in laser tweezers and show that we can drastically increase PT heating of ion-implanted SiNWs in a single-beam laser trap by using gold as the primary dopant ion to superheat water above 200°C with -10 MW cm^-2 irradiances. Furthermore, we explore the role that the dopant ion plays in the PT heating process by comparing SiNWs implanted with gold ions to those implanted with silicon ions at several implantation doses for each ion. The hot Brownian motion temperature extraction analysis is shown to be a versatile technique for quantifying heating effects that arise during laser-trapping experiments. The ion-implanted SiNW heating results allow for the future possibility of using the gold-implanted SiNWs as a PT cancer agent at clinically relevant laser irradiances.

Before ion implantation, SiNW arrays were made using metal-assisted chemical etching (MACE) of a <111 >, p-type (10 Ω cm), single-crystal silicon wafer. The resulting SiNW array (FIGURES IB and 1C) was subsequently implanted with gold ions at an acceleration energy, incident beam angle, and ion doses that were determined using the simulation package SREVI in order to result in ion penetration depths of -50 nm, which ensures gold ion distributions within the nanowires (FIGURE IE, inset) and not the underlying silicon substrate. For the gold ion implantation, ion doses of 6.25 12

x 10 , 6.25 x 10¹³, 6.25 x 10¹⁴, and 6.25 x 10¹⁵ ions cm^"2 were used, resulting in dopant concentrations of 10 18, 101"9, 1020 , and 102"1¹ ions cm 3. As a control, separate SiNW arrays were implanted with silicon ions at ion doses of 10¹⁴, 10¹⁵, and 10¹⁶ ions cm^-2, resulting in dopant concentrations of 2.5 x 10 , 2.5 x 10 , and 2.5 x 10 ions cm with a penetration depth of « 1.9 μιη (FIGURE ID, inset). As shown in FIGURES ID and IE, the effects of the implantation process can be seen visually as a darker surface, providing qualitative evidence for increased optical absorption. Following implantation, the SiNWs were detached from the substrate into deionized (DI) water via ultrasonic ation. The resulting structure of the ion-implanted SiNWs was characterized using transmission electron microscopy (TEM), select area electron diffraction (SAED), energy-dispersive X-ray spectroscopy (EDX), and Raman spectroscopy. Due to the large penetration depth of the silicon ions during implantation, SiNWs implanted with silicon ions show uniform damage along the entirety of the nanowire, as indicated by the constant SAED patterns showing partial crystalline and partial amorphous character shown in the insets of FIGURE 2A. The EDX spectra in FIGURE 2B reflect that the elemental composition of the nanowire is also constant throughout, showing a strong silicon and an oxygen peak with carbon and copper peaks originating from the lacey carbon TEM grid. Contrary to the silicon-implanted SiNWs, the gold-implanted SiNWs show a localization of gold to a single end of the SiNWs. The SAED pattern (FIGURE 2C, top-right inset) at the tip of the nanowire shown in FIGURE 2C shows diffuse Debye rings, suggesting complete amorphization. The EDX spectrum at the same site (FIGURE 2D, top) shows gold peaks, but those peaks and amorphous SAED pattern diminish as the SiNW is probed further from the tip, as shown in FIGURE 2C. Finally, damage to the single-crystalline silicon lattice from ion implantation can be seen with increasing ion dosage by observing the emergence of an a-Si peak redshifted with respect to the native c-Si Raman peak at 520 cm^{- 1} in the Raman spectra for both silicon and gold ion-implanted SiNW arrays in FIGURES 2E and 2F.

The ion-implanted SiNWs were subsequently trapped in single-beam 975 nm laser tweezers shown in FIGURE 3A. The temperature extraction method developed here uses a piezodriven stage and the quadrant photodiode (QPD) signal from the forward- scattered light from the trapped SiNW to extract a diffusion coefficient of the trapped SiNW. In brief, an individual SiNW was recorded in Brownian motion and subsequently trapped at the center of the chamber as shown in FIGURE 3B. During trapping, the stage was oscillated in the x- and -directions at a frequency of 32 Hz and amplitude of 150 nm. The forward-scattered light was then sent to the QPD in the back focal plane of the trapping system and the signals' power spectra were generated, as shown in FIGURE 3C. The stage oscillation results in a peak in the power spectrum at the oscillation frequency; the amplitude of the peak is used to calibrate the QPD signal from a voltage signal to a displacement signal as described in Tolic-Norrelykke et al. The process is then repeated for increasing trapping irradiances. The resulting power spectra are related to the diffusion coefficient D of the trapped SiNW and are fi t to a known functional form using nonlinear least-squares fitting. The diffusion coefficient can then be related to temperature of the trapped particle by

D = Z- (1)

YNW

where T is the system temperature, ¾ is Boltzmann's constant, and y_NW is the Stokes' drag coefficient for a nanowire and is given by

^= ^ x _i)(n (2)

In Equation (2), η(7) is the temperature-dependent viscosity of the surrounding fluid, and L and D are the SiNW length and diameter which are extracted for each trapped particle using image analysis, as detailed in the Supporting Information.

This treatment, however, assumes that the SiNW-solvent system raises temperature isothermally, giving a thermal average of the hot SiNW and the cooler solvent surroundings. This assumption becomes unsuitable when measuring a trapped SiNW that heats drastically relative to its local solvent environment, giving rise to a large thermal gradient around the SiNW as shown in FIGURE 3B. Theory for the motion of a heated Brownian particle from the fluctuating hydrodynamics of its non-isothermal solvent was previously developed and termed "hot Brownian motion" (HBM). Furthermore, the treatment of HBM to rod-like particles has also been reported and is used for our treatment of SiNWs. Applying the theory to our analysis, we relate the HBM diffusion coefficient to the effective HBM temperature by HBM temperature by analysis, we relate the HBM diffusion coefficient to the effective HBM temperature by

where Z¾BM is the HBM diffusion coefficient, THBM is the HBM temperature, and YHBM(7) is the HBM Stokes' drag. To the leading order of the temperature increment AT= (T_p - T₀), where T_v is the temperature at the surface of the SiNW and To is the initial temperature of the solvent, the temperature dependence of the viscosity on THBM can be neglected, giving the effective temperature

T_HBM = T₀ + ^ AT (4) For a temperature dependence of the solvent viscosity of the form

the HBM Stokes' drag YHBM (T) is related to y_NW by replacing η(7) in Equation (2) with ηΗΒΜ (7), which is related to the viscosity of the solvent at room temperature, ηο, by Equation (6):

Vo

1 + - 121 h_n Z2_] Γ ^T 1 _ ril i_n Hs_ _ ¹²⁵⁶³ _ln2 vo , r AT

VHBM (70 486 L ??_∞J L (6) (T₀ -7V_F)J ^L243 η_∞ 118098 η∞ (T₀-7VF)

Equations (4)-(6) are then used in Equation (3) to obtain Z¾BM, which is subsequently compared to the experimental diffusion coefficient to determine the particle temperature T_v (excluding the temperature discontinuity at the particle's surface from the Kapitza resistance).

An alternative HBM temperature analysis using a semi-phenomenological expression for Z¾BM that approximately accounts for higher order terms in AT yields consistent results, indicating that these higher order corrections are negligible, for our purposes. For example, the VFT viscosity parameters for normal and superheated water were obtained by fitting experimental data and are as follows: η_∞ = 2.664 x lO ⁵ Pa s,

FIGURE 4 shows the HBM temperature extraction results of trapping at least five SiNWs of each implantation type and dose at seven different laser irradiances between 5.3 and 15.6 MW cm at a trapping wavelength of 975 nm. The size distribution of the trapped SiNWs was 455 nm ± 117 nm diameter and 6.65 μιη ± 3.78 μιη. By comparing the silicon-implanted SiNWs with the nonimplanted SiNWs, it is evident that the damage to the crystalline lattice from the implantation process causes an increase in the PT heating efficiency, likely due to the increased density of charge and phonon scattering at the implantation induced defect sites within the SiNW. Moreover, it is important to note that an increase in the silicon ion dosage does not seem to result in a significant change in the SiNW heating, suggesting that the contribution of the lattice damage as a mechanism to the SiNW heating has reached a saturation limit.

For the gold ion-implanted SiNWs, the lowest dosage shows temperatures comparable to those of the silicon ion-implanted SiNWs. However, as the gold ion concentration increases, the temperature of the trapped SiNWs also increases dramatically to a point of superheating water above 200°C. Superheating water without bubble formation by trapping gold nanoparticles with 10 nm radii of curvature has recently been reported and was attributed to the large thermal energy needed to overcome the large Maxwell- Young interfacial surface pressure at the surface of the trapped nanoparticle. The increased PT heating efficiency as a function of the increased gold-ion dosage is likely due to the increased light absorption at 975 nm as well as an increased free carrier density to respond to the generated internal electric fields within the nanowire. The mechanism for increased absorption from the gold ion-implanted SiNWs is unclear, but could be due to the formation of gold-related deep level traps in silicon. The increase in the error bars for the highly dosed gold-implanted SiNWs suggests that the amount of gold accumulation between SiNWs varies greatly and causes a large spread in PT heating enhancement from nanowire to nanowire. This is likely caused by the irregularity of the fabricated SiNW array as shown in FIGURE IB and consequently results in an inhomogeneous ion implantation into the SiNWs. Furthermore, during trapping of the highest dosage gold-implanted SiNWs, we also noticed both changes in the nanowire morphology after trapping as well as bubble formation when the trapped SiNW was brought in proximity with other gold-implanted SiNWs attached to the surface (FIGURES 7A-8).

Future research into further enhancement of PT heating efficiencies of ion- implanted SiNWs could be realized by altering the gold implantation profile to be uniform throughout the SiNW length. Another approach to increasing PT heating efficiencies would be to follow gold implantation with a thermal annealing step, which may allow for formation of gold nanoclusters within the SiNWs of a size that could allow them to be more efficient absorbers of the incident NIR radiation and inhibit ion migration to the SiNW surface. Likewise, alternative methods to obtain metallic nanoclusters, such as silver nanoclusters with tunable and uniform sizes, could also drastically increase PT heating efficiencies while removing the need for an implantation processing step.

In conclusion, we have detailed a local temperature profiling method for particles trapped in a single-beam laser trap and demonstrated its use as a versatile method to study the superheating of water from optically trapped nanoparticles and their efficacy as PT agents. We have also shown that the PT heating efficiency of SiNWs can be increased by causing lattice damage through ion implantation that acts as scattering sites that impede thermal and charge transport. Additionally, the efficiency of SiNW PT heating was shown to increase significantly by implanting with gold ions that drastically increase absorption at the 975 nm trapping wavelength and cause superheating of water of over 200°C at the trap site. We expect that this HBM temperature extraction analysis method for laser tweezers will be used to further explore nanoparticles for PT therapy applications as well as to inform how to quantify laser-induced heating effects during laser-trapping experiments in various fields of study including cavity optomechanics and single-molecule biophysics.

Experimental

Silicon Nanowire Synthesis: Silicon nanowires were synthesized with a range of diameters using metal-assisted chemical etching (MACE) of single-crystalline, <111> p- type (boron) 10 Ω cm silicon wafers. The silicon wafer chips were submerged in a 1: 1 v/v solution of HF (10 M):AgN0₃ (0.04 M) at room temperature and pressure for 24 h. The etched Si wafer chips were then submerged in a silver etch containing a 1: 1 v/v solution of NH 4 OH (30%):H₂O2 (30%) at room temperature and pressure for 30 min. Following the silver etch, the silicon nanowires were rinsed multiple times in DI water and dried with nitrogen gas.

Ion Implantation: Silicon nanowire arrays were implanted with gold and silicon ions using a 3 MeV tandem accelerator from National Electrostatic Corporation (9SDH-2). The acceleration energy, incident beam angle, and ion doses were chosen based on calculations using the simulation package SRTM. For the gold ion implantation, an acceleration voltage of 500 keV at a 68° incidence angle was used. Ion doses of 6.25 x 10¹², 6.25 x 10¹³, 6.25 x 10¹⁴, and 6.25 x 10¹⁵ ions cm^"2 resulted in dopant concentrations of 10 18 , 1019 , 1020 , and 1021 ions cm—3 with a mean penetration depth of -50 nm. For the silicon implantation, an acceleration energy of 2 MeV at an incident angle normal to the sample was used. Ion doses of 10¹⁴, 10¹⁵, and 10¹⁶ ions cm^-2 resulted in dopant concentrations of 2.5 x 10 18 , 2.5 x 1019 , and 2.5 x 1020 ions cm—3 with a mean penetration depth of -1.9 μιη.

Transmission Electron Microscopy: Bright field images were taken on a FEI

Tecnai G2 F20 at an acceleration voltage of 200 keV. Select area electron diffraction (SAED) images were taken with camera lengths of 490 and 600 mm. EDX spectra were obtained with a 60 s acquisition time at a rate of approximately 2000 counts s^_1. The spectra were then processed by subtracting the background and using a multipeak fitting software to identify and smooth the peaks.

Helium Ion Microscopy: Helium ion microscopy was performed using Zeiss Orion Plus Helium Ion Microscope at Environmental Molecular Sciences Laboratory (EMSL) in Pacific Northwest National Laboratory, Richland, WA. The He⁺ ion beam was accelerated using a voltage of 30 keV with a dwell time of 30 μ8 and blanker current of 1.3 pA.

Raman Spectroscopy: Raman spectra were acquired on a home-built setup using a Coherent Compass 532 nm laser with 19 mW of output focused onto the sample with a Mitutoyo 50x objective (NA = 0.55). Back- scattered photons were collected through the same objective and focused onto an Acton SpectraPro 500i spectrometer with a Princeton Instruments liquid nitrogen-cooled silicon charge coupled device. Wavenumber values were corrected using modes from a cyclohexane sample measured in the same setup.

Optical Trap Setup: The laser tweezers setup is a highly modified modular laser tweezers kit (Thorlabs, OTKB), where the original condenser lens was replaced with a lOx Mitutoyo condenser (Plan Apo infinity-corrected long WD objective, Stock No. 46-144). The quadrant photodiode and piezostage were interfaced to the computer through a DAQ card (PCIe-6361 X Series, National Instruments) and controlled through custom-built Matlab software graphical user interface. Experimental chambers were prepared by transferring several microliters of the silicon nanowire/water dispersion via pipette into a chamber consisting of a glass slide and glass covers lip. The edges of the glass slide and the glass coverslip were then sealed with a 150 μιη thick adhesive spacer (SecureSeal Imaging Spacer, Grace Bio-labs). The ion-implanted silicon nanowires were trapped at the center of the chamber (-75 μιη from the surface), while voltage traces were recorded at the quadrant photodiode for 3 s at a sample rate of 100 kHz. The quadrant photodiode voltage signal was calibrated by oscillating the piezostage at 32 Hz and an amplitude of 150 nm peak-to-peak during signal acquisition, as outlined in ref. Trapping data were acquired using a 975 nm pigtailed Fiber Bragg Grating (FBG) stabilized single mode laser diode source.

SiNW Size Analysis. The hot Brownian motion (HBM) temperature extraction method detailed in the main manuscript necessitates knowledge of the particle size for the Stokes' drag coefficient of the SiNW (yi y, Equation 2) before a temperature can be calculated. However, often the SiNW diameter lies below the diffraction limit of the illumination source. Therefore, care needs to be taken when using image analysis techniques to determine the diameter of the SiNWs. In order to account for this, a method for determining the size of a SiNW in the laser trap chamber was developed and compared with SEM analysis as shown in FIGURES 5A-5D.

To extract the dimensions of the SiNW, a calibrated CCD camera image is taken of the SiNW in focus lying flat against the chamber surface. The image is then converted to grayscale and pixel value intensity line profiles are drawn across the length and diameter of the SiNW as shown in FIGURE 5C and 5D. The edges of the SiNW show up as a peak in pixel intensity in the line profiles. These peaks are fit to a gaussian curve and the full width at half max (FWHM) is used as the start and stop points for the length and diameter measurement. Lastly, the procedure was then repeated ten times in order to decrease uncertainty.

In order to test the effectiveness of this method, five SiNWs where imaged in a trapping chamber and their sizes were calculated using the method outlined above. Then, the sample chamber was dehydrated and taken to an SEM for high resolution imaging. Lastly, SiNW sizes were determined from SEM data and compared to the size extraction method. Results of the study showed that the analysis worked to within 25% uncertainty for nanowires greater than 250 nm in diameter, with the uncertainty decreasing as the diameter increases. Therefore, no data in the manuscript was presented for SiNWs with diameters less than 280 nm. As an example, the SiNW shown in FIGURE 5A has a length of 10.07 μιη and a diameter of 290 nm + 36 nm as determined by SEM. Using the image analysis method, the results show 10.12 μιη long and 312 nm + 58 nm in diameter which is within the SEM error.

Temperature Extraction Comparison. In order to validate the results from our temperature extraction method, we also compared the heating trend of optically trapped SiNWs (FIGURES 4 and 6C) with the trend for an array of the SiNWs (FIGURE 6D) heated in air by an unfocused 975 nm laser with a beam diameter of 3 mm and laser power of 50 mW measured with a thermocouple out of the beam path (FIGURE 6B). The similarity in the heating trends suggests that the temperature extraction method detailed in the main manuscript is a promising method for extracting HBM temperature information from individually trapped nanoparticles in a single-beam laser tweezer. Example 2: Singlet-Oxygen Generation from Semiconducting and Metallic

Nanostructures

Photodynamic therapy has been used for several decades in the treatment of solid tumors through the optical generation of chemically reactive singlet-oxygen molecules (^C^). Recently, nanoscale metallic and semiconducting materials have been reported to act as photosensitizing agents with additional diagnostic and therapeutic functionality. To date there have been no reports of observing the generation of singlet-oxygen at the level of single nanostructures, particularly at near- infrared (NIR) wavelengths. Here we demonstrate that NIR laser tweezers can be used to observe the formation of singlet oxygen produced from individual silicon and gold nanowires via use of a commercially available reporting dye. The laser trap also induces two-photon photoexcitation of the dye following a chemical reaction with singlet oxygen. Corresponding two-photon emission spectra confirms the generation of singlet oxygen from individual silicon nanowires at room temperature (30°C), suggesting a range of applications for investigating semiconducting and metallic nanoscale materials for solid tumor photoablation.

Photodynamic therapy has been used over the past two decades to clinically treat metastatic tumors through the conversion of naturally available triplet-oxygen molecules

( 30₂) into highly reactive singlet oxygen ( 10₂). Singlet oxygen then reacts with a range of proteins, lipids, and nucleic acids in vivo to prevent angiogenesis within and metastasis from solid tumors. In this process, visible light is absorbed by an intermediate photosensitizing molecule (e.g., Photofrin) to create electronic excited states, which then transfer their energy to 30₂ molecules, converting them to 10₂ through a triplet-triplet annihilation process.

Even though photodynamic therapy (PDT) has been studied for well over a century, it has been in clinical use only since 1993, and currently there are a dozen FDA- approved photosensitizers based on small molecules. Two of the primary challenges of using molecular PDT agents is their long in vivo half-lives and also their nonspecific distribution throughout the body of a patient. This often leads to extended (week-long) periods of time in which a patient must avoid direct exposure to sunlight. For this reason, researchers have been studying novel ways of targeting tumor cells including micellar and liposomal systems, carbon nanotubes, gold nanoparticles, and porous silica or silicon materials as well as exploring new solid-state nano materials, including metals and semiconductors, as direct photosensitizers with predominant excitation wavelengths in the visible spectral window. In particular, silicon nanocrystals have been shown to sensitize singlet oxygen and have the benefit of biodegradability as opposed to inert materials such as gold nanocrystals.

The shape of nanoscale materials has also been shown to affect uptake within targeted tumor cells due in part to the high surface-area-to-volume ratio of one- dimensional materials. Furthermore, recent theoretical calculations have shown that silicon nanowires can be designed to exhibit significant morphology-dependent resonances during photothermal laser heating. In this way, the shape of nanostructures can be optimized to enhance the uptake and optical absorption relative to small-molecule pharmaceuticals with fixed sizes and absorption coefficients.

The phototoxicity of silicon nanomaterials under illumination with wavelengths in the blue spectral region (450 < λ < 500 nm) has also been demonstrated. Blue light, however, lies outside the near-IR (NIR) biological transparency window, resulting in a low tissue penetration depth, which has motivated recent experiments with photoexcitation of both gold nanocrystals that can absorb light via plasmon resonance and upconverting nanocrystals that excite surface-grafted PDT molecules following the upconversion of rare-earth ions in nanocrystalline fluoride host materials. Photoexcitation of silicon nanostructures in the NIR would increase the penetration depth and expand silicon's potential as a PDT agent. We demonstrate here that NIR laser radiation (λ = 975 nm) can be used to optically excite individual silicon and gold nanowires for the local generation of ^l0₂ within a single-beam optical trap.

RESULTS AND DISCUSSION

Silicon nanowires (SiNWs) are synthesized from p-type silicon wafers using metal-assisted chemical etching (MACE) with silver as the active etching metal. High- angle annular darkfield transmission electron microscopy (HAADF TEM) was used to confirm the presence of silver domains on SiNWs following their synthesis (FIGURE 9A). TEM imaging also confirms that the silver nanocrystals can be removed following an aqueous etching step that dissolves silver (FIGURE 9B). Given that metallic nanocrystals have recently been demonstrated to generate singlet oxygen, it must be implicitly ruled out that there is no residual silver to interfere with the experimental measurements.

Neutron activation analysis (NAA) is a sensitive measure of metallic elemental composition and was used to analyze the residual amount of silver both with and without the silver etch (FIGURE 9C). The distinct gamma-ray emission from the Ag isotope decay was used to identify and quantify the silver concentration in the SiNWs. Before silver etching, the Ag concentration is measured to be 291 ppm, whereas after silver etching the amount of Ag is below the detection limits of NAA (<100 parts per trillion).

The surface of silicon nanowires following MACE in hydrofluoric acid is expected to be hydrogen terminated; however the metal-etching step is highly oxidative and may also oxidize the surface of silicon nanowires. Fourier- transform infrared absorption spectroscopy on silicon nanowires following the silver-etching step reveals the presence of a silicon dioxide passivation layer that is also confirmed by oxygen kedge scanning transmission X-ray absorption microscopy (see Supporting Information). Atom- probe tomography (APT) measurements on silicon nanowires further confirm the absence of silver in the bulk of the nanowires following the silver etch, as well as the presence of a surface oxide layer (FIGURE 9D).

Laser trapping of individual nanowires was performed using a custom instrument shown in FIGURE 10A. A 975 nm laser is expanded to overfill the back aperture of a lOOx oil immersion objective (NA = 1.25) and then focused into a chamber with an aqueous nanowire suspension. The silicon nanowires are trapped and photoexcited by the NTR laser (FIGURE 10B), which is above silicon's band gap, to produce triplet excitons, as evidenced by the observation of excitonic emission from SiNWs after Ag etching (FIGURE 9E). The excitons then diffuse to the surface and, through a Dexter electron exchange mechanism, excite molecular oxygen in solution to the singlet state (FIGURE IOC) via a triplet-triplet annihilation process. Singlet oxygen molecules have been reported to have a decay time of 3.7 μ8 in water, leading to diffusion distances on the order of 250 nm from the silicon-water interface at room temperature. Singlet-oxygen excited states can also relax to their ground state through emission of a photon at 1270 nm. However, their long lifetimes at room temperature also allow for reactions with singlet oxygen sensor green (SOSG) molecules, which selectively exclude reaction with superoxide anions and peroxide, to make a 1,4-endoperoxide, SOSG-EP (FIGURE 10D). In addition to trapping single silicon nanowires and photoexciting triplet excitons in silicon, the laser trap also has a sufficient irradiance (~MW/cm ) for two-photon photoexcitation (2PPE) of SOSG-EP to the electronic excited state, SOSG-EP* (FIGURE HE). The green emission from SOSG-EP* can be visualized through use of a CCD camera (FIGURE 11 A). The potential for visible photoluminescence from the porous SiNWs was eliminated as a possible source of emission by performing an identical control experiment with optically trapped SiNWs in the absence of SOSG (FIGURE 11B). It also has been demonstrated recently that there is negligible (<5°C) photothermal heating of optically trapped SiNWs via analysis of their Brownian dynamics.

Generation of singlet oxygen directly from SOSG has been reported previously using UV or visible irradiation. An additional control experiment was performed to test for potential photogeneration of 0₂ from SOSG molecules themselves at an identical NIR irradiance (FIGURE 11C). Clearly, the NIR source is unable to excite 0₂ directly or through photosensitization of the SOSG molecule. Similarly, a final control experiment combining SOSG and Si0₂ microspheres (FIGURE 11D) exhibits no detectable 0₂ generation at the bead's fully oxidized surface, which is expected since Si0₂ is an insulator and would not generate the excitons necessary to sensitize ^x0₂. A quadratic dependence on the integrated emission from SOSG as a function of laser power is observed (FIGURE 1 IF), indicating that SOSG is excited by a two-photon mechanism.

Enhanced singlet-oxygen generation within a suspension of SiNWs is also observed at much lower laser irradiance. A cuvette with a concentrated suspension of SiNWs and SOSG was placed in the path of the 975 nm laser beam at a constant irradiance of 8 W/cm (beam power = 260 mW, spot size = 2 mm) and pumped briefly at regular intervals with a low power 405 nm lamp to monitor changes SOSG-EP concentration. Comparison with a cuvette containing SOSG alone shows a significant increase in 0₂ production (FIGURE 11G).

The approach demonstrated above for analyzing single silicon nanowires may also be extended to other nanoscale materials including noble metals. Gold nanocrystals have been shown to induce significant photothermal superheating of water when optically trapped, and several recent reports have indicated that noble metal nanoparticles are capable of photogeneration of singlet oxygen. In FIGURES 12A-12D we show that it is possible to observe the generation of singlet oxygen from single gold nanorods that have been synthesized through electrochemical deposition within track-etched polycarbonate membranes. To confirm that the observed emission is from SOSG-EP and not 2PPL from the AuNRs, a single AuNR was trapped without SOSG (FIGURE 12C), and long exposure micrographs showed no observable emission (FIGURE 12D). CONCLUSION

We have demonstrated that a near-infrared laser trap can be used to observe the photogeneration of 0₂ molecules from individual silicon and gold nanowires that are suspended within an aqueous trapping medium. To our knowledge, there have been no prior reports of observing the photosensitization of 0₂ at the level of single nanostructures. Although the optical absorption coefficient of silicon at NIR wavelengths is not as high as for visible wavelengths, the use of NIR radiation allows for deeper tissue penetration. Furthermore, this work also suggests that singlet oxygen may also be generated during photothermal heating of gold and semiconductor nanostructures, including recently reported silicon/gold composite nanostructures that have been investigated for in vivo solid tumor photothermal ablation. The NIR generation of 0₂ from silicon nanowires may also affect the long-term stability of catalysts used for solar energy conversion. Future studies will investigate how the efficiency of 0₂ generation is affected by size-dependent morphology-dependent resonances.

METHODS

Nanowire Synthesis. Silicon nanowires were synthesized using metal-assisted chemical etching methods. Briefly, a (111), B-doped silicon wafer with a resistivity of 11 Ω cm (Silicon Sense) is immersed in a 1: 1 solution of 10 M HF/0.04 M AgN0₃ for 3 h, resulting in an array of vertically aligned nanowires with lengths around 10 μιη and a range of diameters on the order of 100 nm. After rinsing with DI water, the etched wafers are placed into a 1: 1 30% NH₄OH/28% H₂0₂ solution for 5 min to dissolve the silver film and deposits from the nanowires. Gold nanowires were synthesized electrochemically using a 10 μιη thick polycarbonate track etch membrane with pores of 1 μιη diameter (Sterlitech) to control the dimensions of the grown nanowires. The polycarbonate membrane is subsequently dissolved in chloroform to obtain a suspension of the gold nanowires.

Single-Beam Laser Trapping. A 330 mW 975 nm laser diode (Thorlabs PL980P330J) was expanded to overfill the back aperture of a lOOx Nikon oil immersion objective (NA: 1.25), which was focused into a chamber consisting of a #1 glass coverslip and a 1 mm thick glass slide with a 120 μιη spacer (Grace Bio-Laboratories, #654002). Forward scattered light was focused onto a quadrant photodiode (Thorlabs, PDQ80A). Brownian analysis of a trapped particle was analyzed using custom code written using Matlab. Neutron Activation Analysis. Compositional analysis was performed pre- and postcleaning via a TRIGA Mark II nuclear reactor. The samples were irradiated for 30 min operating at a 100 kW thermal power, 4 x 10 12 neutrons/cm 2 s thermal flux, and

4.8 x 10 12 neutrons/cm 2 s fast and epithermal flux. Each sample was counted while positioned near the surface of a 25 cm , trapezohedral, germanium, lithium-drifted semiconductor detector (Nuclear Diodes), which is constantly cooled by liquid nitrogen (77 K). Dead time between end of irradiation and start of collection was 19 h 48 min.

Atom-Probe Tomography . The needle specimens for APT were prepared by lifting out single nanowires and attaching them onto Si microtip arrays followed by Pt capping and annular milling in an FEI Helios Nanolab 600 dual-beam FIB system. Compositional analysis by APT was performed using a CAMECA LEAP 4000 XHR instrument for a total of 3.8 million ion counts. The sample temperature was maintained at 40 K, and the evaporation rate was maintained at 0.005 atom per pulse. APT analysis was conducted using a picosecondpulsed, 355 nm UV laser (20 pJ, 100 kHz) focused onto the apex of the SiNW needle specimen.

Scanning Transmission X-Ray Microscopy (STXM). Oxygen K-edge STXM analysis was performed at the Lawrence Berkeley National Laboratory Advanced Light Source beamline 5.3.2.2. Samples for STXM analysis were prepared by drop casting a solution of SiNWs onto a lacy-carbon TEM grid and placing the grid in the STXM beamline under a one-third atmosphere of helium. Soft X-rays (250-800 eV) were focused onto the sample and rastered with a spot size of 31 nm using 25 nm Fresnel zone plates.

Singlet-Oxygen Detection. Singlet oxygen sensor green was used to detect singlet oxygen from optically trapped nanoparticles. SOSG is prepared by dissolving 100 mg in 66 μΐ_^ of methanol and then diluting in Millipore (18.6 ΜΩ) water to obtain the appropriate concentration. SOSG-EP* emission is visualized with 20 s exposures using a Thorlabs CCD camera (DCU224C), and emission spectra were obtained by collecting emission into an Acton SpectraPro 500i spectrograph and dispersing onto a Princeton Instruments liquid-nitrogen-cooled silicon detector array.

While illustrative embodiments have been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention.

Claims

CLAIMS The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A silicon nanoparticle comprising amorphous silicon and gold.

2. The silicon nanoparticle according to Claim 1, wherein the silicon nanoparticle is a silicon nanowire.

3. The silicon nanoparticle according to Claim 2, wherein the silicon nanowire has a length of 100 nm or greater.

4. The silicon nanoparticle according to Claim 2, wherein the silicon nanowire has an aspect ratio (length: width) of at least 2: 1.

5. The silicon nanoparticle according to Claim 1, wherein the smallest dimension is from 50 nm to 500 nm.

6. The silicon nanoparticle according to Claim 1, wherein the gold concentration is from 10¹⁵ to 10²⁵ ions/cm³.

7. The silicon nanoparticle according to Claim 1, wherein the gold is not present in aggregates larger than 0.5 nm.

8. The silicon nanoparticle of Claim 1, wherein the gold is evenly distributed within the silicon nanoparticle.

9. The silicon nanoparticle of Claim 1, wherein the silicon is substantially amorphous.

10. The silicon nanoparticle of Claim 1, wherein at least a portion of the gold is amorphous gold.

11. The silicon nanoparticle of Claim 1, wherein the gold is at least 1%, by weight, of the nanoparticle.

12. A method of heating, comprising irradiating a silicon nanoparticle a silicon nanoparticle according to any of Claims 1-11.

13. The method of Claim 12, further comprising heating a volume in proximity to the silicon nanoparticle by transferring heat from the silicon nanoparticle to the volume.

14. The method of Claim 12, wherein the volume is heated to a temperature of 38°C to 50°C.

15. The method of Claim 12, further comprising heating a plurality of the silicon nanoparticles.

16. The method of Claim 12, irradiating the silicon nanoparticle comprises irradiating at an intensity of 5 MW/cm² to 15 MW/cm².

17. The methods of Claim 12, wherein irradiating the silicon nanoparticle comprises an irradiation wavelength of 650 nm to 1350 nm.

18. The method of Claim 12, further comprising the silicon nanoparticle generating singlet oxygen in response to irradiating the silicon nanoparticle.

19. A method of heating tissue comprising irradiating a silicon nanoparticle according to any of Claims 1-11 in proximity to the tissue, thereby heating the tissue.

20. The method of Claim 19, wherein the tissue is heated to a temperature of 38°C to 50°C.

21. A method of treating cancer in a subject comprising irradiating a silicon nanoparticle according to according to any of Claims 1-11 in proximity to cancerous tissue in the subject, thereby heating the cancerous tissue and treating cancer in the subject.

22. The method of Claim 21, wherein the cancerous tissue is heated to a temperature of 38°C to 50°C.