WO2015138035A1 - Methods for excitation-intensity-dependent phase-selective laser-induced breakdown spectroscopy of nanoparticles and applications thereof - Google Patents

Abstract

A method of characterizing particles by excitation-intensity dependent Laser-Induced Breakdown Spectroscopy (LIBS), including providing a gaseous flow field comprising a gaseous phase and a particle phase, directing to the particle phase a series of laser pulses characterized by an excitation intensity effective to induce breakdown of the nanoparticles without breakdown of the gaseous phase, detecting the emission spectrum from the selective breakdown, and processing the spectrum to obtain measurements on various characteristics of the particles.

Description

METHODS FOR EXCiTATiON-!NTENSITY-DEPENDENT PHASE-SELECTIVE LASER- INDUCED BREAKDOWN SPECTROSCOPY OF NANOP ARTICLES AND APPLICA TIONS

THEREOF

CROSS-REFERENCE TO RELATED APPLICATIONS

6 01| This application claims priority under 35 U.S.C. § 19(e) to LIS, Provisional Patent Application Serial No. 61/917,991, filed on December 19, 2013, which is hereby incorporated by reference in its entirety.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

(0002! This invention was made with gove.nim.eftt support under Grant Nos. W91 1 F-08-I- 0417, W9I iNF-09-1-0! 38, and W91 INF- 10- 1 -0018 by the Army Research Office and through Grant No. N00014-09-1 -0773 by the Office of Naval Research. Accordingly, the Government has certain rights in this invention.

FIELD OF THE DISCLOSURE

(0003! This invention describes methods of characterizin nanoparticies, including aerosols (e.g. elemental composition, size, volume fraction, etc), which can be used, among other systems, during gas-phase synthesis or other pameuiate-momtoring systems,

BACKGROUND OF THE INVENTIO

(0004! Nano-sized metal-oxide particles have great potential in a wide range of applications, including structural additives, purification of air and water, catalysis, chemica! gas sensors, and energy-conversion devices. Gas-phase synthesis methods, including flame processes, are well known to produce metal -oxide nanoparticies less than 10 nm at high production rates with controllable sizes. Gas-phase processes allow for chemical conversion (homogeneous and heterogeneoos) from precursor to metal-oxide monomers; coagulation and coalescence of metal- oxide nanoparticies; and deposition, of particles driven by the.rmophoresis and diffusion. Fast time scales, along with steep temperature gradients, are well-suited for producing metastable phases, with the associated dynamics being far from equilibrium. As such, non-intrusive in-situ

] diagnostics ate essential in understanding growth mechanisms, as well as in process control during nanopo der manufacture via real-time feedback.

{OO05| Although conventional laser-induced breakdown spectroscopy (UBS) has been shown to be a quantitative technique for identifying particles, it cannot quantitatively

characterize them substantially exclusively, as the molecules in the surrounding gas-phase are also broken down and excited. Accordingly; there is a strong need for better methods of

mapping and characterizing nanoparticles during synthesis and other particulate-contaimng systems.

SUMMARY OF THE IN VENTION

{0006} A novel excitation-intensity-dependent phase-selective laser-induced breakdown spectroscopy (LIBS) has been developed to characterize metal-oxide nanoparticles. In contrast to traditional application of LIBS on particles, the power used here is much Sower; and no macroscopic spark is visuall observed. Nevertheless, the low-intensity LIBS shows interesting selecti it — only exciting atoms m particle phase, with no breakdown emission occurring for gas molecules (e.g. metal-organic precursor, air). The emission intensity increases with particle size, plateauiiig as the particles reach a certain size, indicating the absorption efficiency to be size- dependent for small particles. Furthermore, by selecting the excitation wavelength to match the transition lines of the atoms, stronger emission intensity from the selected atoms can be produced to enhance the signal of the selected atoms without increasing the emission from other atoms. This method has been applied to make two-dimensional measurements of volume fraction of nanoparticles in both far and near wall conditions.

10007} The selectivity of the low-intensity LIBS for such application can be advantageous for identifying nanoparticle composition,, tracking nanoparticle formation and presence, and measuring nanoparticle volume fraction during gas-phase synthesis or particulate monitoring. The size-dependent absorption efficiency can be used to measure nanoparticle size in-situ.

}0008] This new method for detecting elements in nanoparticles, wherein only the nanoparticles are dissociated into their constituent aioms/ions (whose electrons are then excited) via laser without macroscopic visible spark, and where no delay time is needed to collect the atomic emissions, presents many applications. It can be used for determining the progress rate of gas-io- article transition processes; for determining the elemental portions i different substances in a gas or liquid dispensed mixture, for example, by changing laser fluence in- between different breakdown thresholds; for determining the particle concentration according to the breakdown threshold; tor determining the particle concentration ^'volunie-fraction according to the signal strength after saturation; for determining the particle size according to the different breakdown threshold; and for determining both the gas phase and particle phase concentrations by combining with conventional LIBS, as well as other laser-based diagnostics such as Raman spectroscopy, laser-induced fluorescence (Li F), laser-induced incandescence (Ϊ..ΪΪ), etc.

|tl0i)9J The broad suitability of the technique for other metal-oxides has been shown for different materials. Moreover, this method is applicable to small particles of a host of

compositions other than metal oxides, such as nitrides, carbides, chlorides, pure metals, etc., albeit with differing requisite excitation power ranges and selectivity dependencies. Note that depending on the breakdown thresholds for the solid and gas phases, the particles may not he limited to nanoparticles and may be larger in size (e.g. microparticles). As such, the technique is particularly useful for small particle identification and monitoring for many systems, including those associated with pollutant emissions and atmospheric sciences. Moreover, the utilization of this diagnostic as an input into real-time feedback in modem control systems can he used to optimize and ensure quality in the large-scale commercial production of user-defined

nanoraaterials, j001 | The present invention thus provides a method of characterizing nanoparticles by

Laser-induced Breakdown Spectroscopy (LIBS). The method comprises; providing a gaseous flow field comprising a gaseous phase and a particle phase, wherein the particle phase comprises the particles to be characterized; exciting the gaseous flow field with a series of laser pulses characterized b an excitation intensity effective to induce breakdown of the particle phase and excitation of the atoms within the particles without the breakdown of said gas phase and excitation of the atoms within the gas phase, so that the excited atoms of the particles emit an atomic emission spectrom characteristic of the elements within the particles; and detecting the emission spectrum produced by the excited atoms within the particles.

{ Ol.tf In some embodiments, the laser pulse comprises a pre-determined excitation wavelength matching an atomic transition line of an elemental species within the particles, 00^"l 2| In some embodiments, the laser pulses are provided by a tunable laser and are characterized by a plurality of pre-determined excitation wa velengths matching the atomic transition line of more than one elemental species within the particles. The plurality of predetermined excitation wavelengths can be generated sequentially or simultaneously.

[0013J In some embodiments, the laser pulses are provided by a plurality of lasers and are characterized by a plurality of pre-determined excitation wavelengths matching the atomic transition Sine of more than one elemental species within the particles. The plurality of predetermined excitation wavelengths can be generated sequentially or simultaneously.

{0014 In some embodiments, the fluence of the laser is adjustable.

{0015J In some embodiments, the fluence of the laser is set at or above saturation of emissio signals.

{0016) In some embodiments, the method of the present invention further comprises processing the emission spectrum to obtain a spatial measurement of the particles based on the emission intensity of the spectrum. In some embodiments, the spatial measurement obtained by processing the emission spectrum is a volume fraction measurement of the particles. In some embodiments, the laser pulses are in the form of a sheet and the emission spectrum is processed to obtain a planar volume fraction measurement of the particles.

(0017| I n some embodiments, the method of the present invention further comprises processing the emissio speetram to obtain a temporal measurement of the nanoparticles.

{00.18) In some embodiments, the method further comprises processing the emission spectrum to obtain a size measurement of the particles based on the emission intensity of the spectrum, provided that the concentratio of the particles is known or constant. |ίΙΘΙ | In some embodiments, the method further comprises processing the emission spectrum to obtain a concentration measurement of the particles based on the emission intensity of the spectrum, provided that the size of the particles is known or constant.

[9020] In some embodiments, the method further comprises processing the emission spectrum to determine the elemental identity of the particles, wherein the fiuenee of the laser is optionally adjusted according to different breakdown thresholds. In some embodiments, the predetermined excitation wavelengths .match one or more atomic lines of known (reference) elements.

[9021] In some embodiments the partic les comprise nanoparticies. In some aspects of this embodiment, the particles consist essentially of nanoparticies. In other embodiments the particles comprise micropariicles. In some aspects of this embodiment, the particles consist essentially of nanoparticies.

[9022] In some embodiments, the gaseous flow field is sampled from a manufacturing process for producing the particles, and said method farther comprises the steps of processing the emissio spectrum to measure a characteristic of the particles and providing a feedback command according to the processed spectrum to the manufacturing process to control the measured characteristic of the particles,

[9023] In some embodiments, the particles are collected from a pollution source or a paniculate-containing atmosphere.

BRIEF DESCRIPTION OF THE DRAWINGS

[9924] Embodiments will be described with reference to the following drawing figures, in which like numerals represent like items throughout the figures, and in which:

[9025] Figure 1 shows a typical schematic of the in-situ laser-based diagnostics as used during flame synthesis of nanoparticies. The setup is similar to that for conventional LIBS; however, the adjustment of laser-excitation intensity is critical in discriminating only the nanoparticle phases, with associated nanoparticle specific characterizations. |0026| Figure 2 provides the typical low-intensity LIBS emission spectra acquired from a laser excitation of 532 nm (28 J/enf). Π atomic spectra from 1ST database is marked by red lines,

[0027} Figure 3A provides the integrated emission intensity (black squares) and calculated particle size (red line) as functions of distance from the tested burner exit, along the axis of symmetry.

(00281 Figure 38 provides the spectra collected at distances of 2 mm 4 mm. 6 mm, 8 ram, and 14mm from die burner exit, respectively.

{0029} Figure 4 provides the emission intensities for different laser excitation powers at two different precursor loading rates. The measuring volume is at a distance of 14 mm from the burner exit.

10030! Figure 5 provides the integrated emission intensity as a inaction of precursor

concentration. The measuring volume is at a distance of 14 mm from the burner exit. The insets A and B are TE photos of TiO? nanopaiticles collected from substrate, when the gas-phase precursor loading concentrations are 1 16,4 ppm and 232.8 ppm respectively,

1003! I Figures 6A and 6B provides the emission spectra with 532 nm and 355 tun. excitation, respectively, along with the given settings (e.g. the same laser power, detection settings, etc.). f0032| Figure 7 provides the integrated emission intensity (black squares) from 532nra laser excitation, 355 nm excitation (blue circles), and calculated particle size (red line), as functions of distance from the burner exit, along the axis of symmetry.

[0033J Figure 8 .provides the intensity change with laser power for 497.534 nm emission, along with other emissions at 355 nm laser excitation. The inset shows the intensity change with laser power at 532 nm laser excitation.

[0034 j Figure 9 provides the temporal evolution of emission intensity, as detected by phoiomult.ip.Uer tube (PMT) and recorded by digital oscilloscope.

|0035j Figure 10 provides the increase of signal intensity versus particle volume fraction. The Sine shows the linear fitting of the data with an intercept of 140 ppb. |0036| Figure 1 1 provides the two-dimensional planar measurement of TiC¾ nanoparticie volume fraction produced by Bun sen flame synthesis, using phase-selective LIBS

{0037) Figure 12 provides the two-dimensional planar measurement of TiC¾ nanoparticie volume fraction in stagnation flame synthesis.

[0038] Figure 13 provides the application of the present invention in monitoring and controlling the production of nanopari icles.

(00391 Figure 14 provides the feedback mechanism for determining and refining the parameters for gas-phase synthesis of nanopariicles.

DETAILED DESCRIPTION

{0040) The instant invention describes methods fo diagnosing nanoparticie characteristics (e.g., for metal- oxides, nitrides, carbides, chlorides, pure metals, or mixtures thereof), i»-si.tu and ex-sttu. By adjusting the excitation wavelength and laser fl ence for laser-induced breakdown spectroscopy, only the particle phase undergoes breakdown and there is no macroscopic visible spark in the gas phase. As a result, high resolution spatial measurements of the particles can be accomplished. In addition, because no delay rime is needed for the emission signal detection, high temporal resolution measurements can be achieved.

(00411 ^be methods can be adopted for the detection of nanoparticie composition of metal- oxides as well as other materials such as nitrides, carbides, chlorides, pure metals, etc., albeit with likely differing requisite excitation power ranges and selectivity dependencies,, depending on the nature of the material. This can be determined by one of ordinary skil l in the art without undue experimentation. In addition, depending on the breakdown thresholds for the solid and gas phases, the particles may not he limited to nanopariicles and may be larger in size (e.g.

microparticies).

[0042| The methods of the present invention find broad applications in areas including particle measurement, process control, and environmental monitoring. For example, the methods are particularly useful for small panicle identification and monitoring for many systems, including those associated with pollutant emissions and atmospheric sciences. Moreover, the utilization of this diagnostic as an input into real-time feedback in modem control systems can be used to optimize and ensure quality in the large-scale commercial production of user-defined nanomaterials.

[0043| While the following text may reference or exemplify'^' specific steps of a particle characterization process, it is not intended to limit the scope of the invention to such particular reference or examples. Various modifications may be made by those skilled in the art, in view of practical and economic considerations, such as the size and composition of the particles and the excitation wavelengths of the laser pulses. In order to more clearly and concisely describe the subject matter of the claims, the fol lowing definiti ons are intended to provide gu idance as to the meaning of terms used herein,

|ft044| ^"Hie articles a" and "an" as used herein mean "one or more" or "at least one," unless otherwise indicated. That is, reference to any element of the present in vention by the indefinite article "a" or "an" does not exclude the possibility that more than one of the element is present. f0045| The terra "nanopartide" used here includes solid phase particles as well as liquid phase particles (i.e., naiiodroplets) and liquid phase-solid phase mixtures (i.e., nanosiurries).

[0046} The method of the present invention includes: providing a gaseous flow field comprising a gaseous phase and a particle phase, wherein the particle phase comprises the particles to be characterized; exciting the gaseous flow field with a series of laser pulses characterized by an excitation intensity effective to induce breakdown of the particle phase and excitation of the atoms within the particles without the breakdown of the gas phase and excitation of the atoms within the gas phase, so that the excited atoms of the particles emit an atomic emission spectrum characteristic of the elements within the particles; and detecting the emission spectrum produced by the excited atoms within the particles. The method may further include processing the emission spectrum to determine the characteristics of the particles.

[0047] Because the method includes using in-situ excitation-intensity-dependent laser induced breakdown of the particle phase without breaking down the gas phase in an aerosol, various characteristics can be obtained such as elemental particle information and concentration of particles, which are also useful for determining the progress rate of gas-io-particle transition processes. By selecting the excitation wavelength to correspond to resonance of specific atomic lines, the limit o detection can be significantly lowered down. Hie wave-length can be chose such that only the emission from the atom of interest is enhanced, thus achieving species selectivity. Since relatively low energy is required in this low-intensity LIBS process, species selectivity can be achieved via the use of a tunable laser, such as a dye laser or an array of tunable diode lasers. By selecting die laser fluence carefully according to the breakdown thresholds), this method can be applied to various particle compositions to breakdown the particle phase only for detection, thus with high selectivity.

|0048| In some embodiments of the present in ven tion , the seri es of laser pulses include one or more pre-determined excitation wavelengths which may be the same or different from the atomic line of known reference elements. Because each known element may have a set of corresponding atomic lines that can be the same as or overlapping with that of other elements, the number and ran ge of the pre-determined wavelengths can be adj usted to suit the need for the detection of one or more elements. After detection and processing of the emission spectrum of the nanoparticSes in comparison with that of the known reference elements, the elemental identification of the particle composition can be obtained. n some embodiments, the series of laser pulses includes one or more pre-determined excitation wavelengths thai match an atomic transition of an elemental species within the particles to be characterized. The pre-determined wavelengths may also match two or more eieraenial species of the particles. The detection and processing of the resulting emission spectrum will not only serve to confirm the elemental identity of the particles, but also provide other information such as particle size, concentration, and volume fraction, with the advantage of very low detection thresholds. Moreover, the method of the present invention is characterized by lasing, affording extremely low threshold and directional detection at long distances from excitation/detection, using back-scatter

configuration. j 049| The excitation wavelengths ma vary depending on the range of the particle size and other relevant parameters. For example, some wavelengths may be more suitable and sensitive than others for detecting smaller particles. As the size of the particles increases, a different wavelength may become more suitable. Accordingly, the wavelengths can be adjusted according to the range of the particle size, which can be readily determined by one of ordinary skill in the art without undue experimentation.

{OOSOf The laser pulses can be generated from a single laser or multiple lasers, in. some embodiments, the laser pulses have one or more pre-determined wavelengths that match the atomic transition lines of one or more known reference elements. The multiple lasers can be arranged in any suitable manner, including different angles from, the particles to be characterized, different shape (e.g. beam or sheet), and different distance from the particles. In some embodiments, the laser pulses with one or more wavelengths can be generated from a tunable laser, suc as a dye laser, or an. array of tunable diode lasers. As described above, the laser pulses generated from the tunable laser can also have one or more pre-determined excitation

wavelengths that match the atomic transition lines of one or more known reference elements. In some embodiments, the pre-determined wavelengths match the atomic l ines of one or more elemental species of the nanoparticles to be characterized. In exemplary' embodiments, the predetermined wavelengths match .more than one elemental species (e.g. 2, 3, 4, 5, 6, or more elemental species) as references or within the particles, f 0051 The pre-determined excitation wavelengths can be generated sequentially or simultaneously depending on the specific particles and the particular data to be collected. For example, multiple pre-determined excitation wavelengths generated at the same time can be used to detect or confirm the elemental identity of the particles. Meanwhile, multiple pre-determined excitation wavelengths generated sequentially find application on the monitoring of changing particle characteristics, such as particle growth and particle production. Sequential generation of pre-determined wavelengths also offer the benefit of lower intensity and ther efore signal interference from the breakdown of surrounding mediums such as gas phase. However, methods using sequentially or simultaneously generated wavelengths are not mutually exclusive of each other and may both be employed in the same type of particle characterization.

(00521 The fluence of the laser can be adjusted depending on the particle size, the elemental composition of the particle, the excitation wavelength, and other parameters relating to

breakdown thresholds. For example, while certain excitation wavelengths may be suitable for detecting small particles within a certain range, the emission intensity can also be affected by laser fluence which will continue to excite more electrons as the power level increases and may lead to inconsistent results. Therefore, depending on the range of the particle size and tlie predetermined wavelengths, adjustment of laser iluence is desirable. In some embodiments, the fluence remains constant. In some embodiments, the laser fluence is variable. In some embodiments, the laser fluence is set at a. level that corresponds to the saturation of the emission signals. In some embodiments, the laser iluence corresponds to a level at or above the sat ration of the emission signals. The exact extent of laser intensity above the signal saturation can be determined by one of ordinary skill in the art without undue experimentation. In general, keeping the laser fluence above the signal saturation level can help obtain consistent data. Meanwhile, it is still low enough to avoid the breakdown of the surrounding medium such as the gaseous phase,

[0O53| The method of the present invention finds application i various aspects of particle detection. Because the low intensity laser does not affect substances or materials other than the particle phase, the emission spectrum resulting from the selective breakdown of the particles reflects a high spatial resolution measurement of the particles with zero or minimum interference from the surrounding medium (e.g. gaseous phase). For example, under low intensity laser condition, there is no visible macroscopic spark and the affected area (distinct nanoplasmas for each nanopartic!e) is small compared to conventional laser induced breakdown spectroscopy, in addition, minimum delay time is required to detect the emission spectrum or signal because any possible Bremsstrahhing emissions dissipate quickly at nanosecond level so that nanosecond gating delays are sufficient to gate them out. Therefore, the method of the present invention also provides a high temporal resolution measurement of the particles.

[0054J In some embodiments, processing the emission spectrum of the laser induced breakdown of the partic ies provide a volume fraction measurement of the partic les based on the emission intensity of the spectrum. In exemplary embodiments, the method comprises measuring the planar volume fraction of particies based on the atomic emission with a laser sheet at sufficiently high fluence to breakdown the particies locally. With the laser-generated nanoplasmas (as confirmed by ionic spectra) confined around each nanoparticle and inter-

I I particle distance sufficiently large, particle volume fraction ca be determined with sufficient spatial resolution. 0055J Depending on the range of the particle size, different intensity-particle volume fraction relationship has been observed. Accordingly, suitable equations can be extrapolated from experimental data for different ranges of particles with known characteristics. Processing the emission intensity against the extrapolated equations provides quantitative characterization of the particle volume fraction. In some embodiments, the particles to be characterized are above a cut-off size or within a certain size range, so that the emission intensity and the characteristic (e.g. particle volume fraction} to be measured are in a linear relationship. Exemplary cut-off sizes (particle diameter) include about 5.5 ran, about 6 nm, about 6.5 nm , about 7 nm_> about 7.5 nra, about 8 nm, about 8.5 am , about 9 nm, about 9.5 am, about 10 am, about 10.5 ton, about ί 1 am, about 1 1.5 m, about 12 nm, about 12.5 nm. Similarly in some embodiments, a cut-off volume fraction can be employed to obtain a linear relationship between the emission intensity of the particles and the particle volume traction above the cut-off Exemplary cut-off values of volume fraction include about 80 ppb, about 90 ppb, about 100 ppb, about 1 10 ppb, about 120 ppb, about 130 ppb, about 140 ppb, about 150 ppb, about 160 ppb, about 1 70 ppb, about 180 ppb, about 190 ppb, and about 200 ppb.

|0056J In some embodiments, the laser pulses are in the form of a sheet and processing of the emission spectrum provides a planar volume fraction .measurement of the particles. The two- dimensional measurements of volume fraction of particles are applicable to both far and near wall conditions. The particle volume fraction measurement near walls provides a osetui tool for monitoring various nanoparticle-reiated processes, such as particle deposition, which is important for the synthesis of iiano-films.

[8057) In some embodiments, processing of the emission spectrum from the breakdown of the particles provides a size measurement of the particles based on the emission intensity of the spectrum provided that the concentration of the particles is known or constant. j0058| l a some embodiments, processing of the emission spectrum from the breakdown of the particles provides a concentration measurement of the particles based on the emission intensity of the spectrum provided thai the size of the partic les is known or constant,

[0059| In some embodiments, by increasing the intensity of laser fluence, from no breakdown, to particle-phase breakdown, to all matter breakdown (including gas phase) as a sequence, the concentrations of the particle phase as w ell as gas phase can both be determined. In exemplary embodiments, the particle-phase concentration is determined by low-intensity phase- selective LIBS, and the total concetriration is determined by the normal LIBS, and the difference between them indicates the gas phase concentration.

|0O6O| In some embodiments, processing of the emission spectrum from the breakdown of the nanoparticles provides the elemental identity of the particles. In some embodiments, the emission spectrum reflects saturated signal strength. By comparing the characteristic peaks and intensity thereof with that of known reference elements, one of ordinary skill in the art can readily determine the presence or absence of certain elements in the particles , in particular, when the pre-deteranned excitation wa velengths matches one or more atomic l ines of known { reference} elements, the pattern of the enhanced peaks are expected to be the same between the emission spectrum and the reference spectrum if the reference elements is present. The identity of mul tiple elements can also be determined in this manner.

[0061 In some embodiments, the laser .fluence is optionally adjusted between different breakdown thresholds to suit, the detection of individual or multiple elements. A laser pulse at any particular excitation wavelength may have a constant and changing fluence. Laser pulses with different excitation wavelengths, whether they are generated simultaneously or sequentially, may have the same or different laser fluences, which may change i the same or different patterns. In some embodiments, the fluence is set at or above a level which corresponds to the saturation point of the emission strength.

[00621 In some embodiments, the gaseous flow field is sampled from a manufacturing process for producing the particles. The method further includes the steps of processing the emission spectrum to measure a characteristic of the particles and providing a feedback command according to the processed spectrum to the manufacturing process in order to control the measured characteristic of the particles. The feedback command may require process conditions to be maintained at current levels if the measured characteristic is nominal.

[0063| In some embodiments, the particles are collected from environments associated with pollution emissions or atmospheric sciences. For example, heavy metal-containing particles and other pollutant aanopanicles can be monitored in a particular environment. Regions or iasiaaces where gas-to-particle transitions occur (e.g. micleaticm) can be identified. Accordingly, the methods of the present invention provide a convenient, accurate and efficient means of monitoring various environmental conditions.

|0O64| The methods described herein can be applied to the detection of various types of particles. For example, extensions of the present methods to detect nanodroplets (liquid phase) and nanoslurri.es (liquid solid mixtures) can be made by one of ordinary sfctil in the art without undue experimentation. There are also no particular restrictions on the state of the particles and extensions can be made to solid particles in fluids in general (e.g. surrounded by liquid phase rather than gas-phase). Further, the methods of the present invention can be applied to the detection of particles larger than nanoparticles such as microparticles. Exemplary sizes of nanoparticles and larger particles that can be characterized with the present invention include about 1-10 ora, about 1-25 nm, about 1-50 ran, about 1 -100 om. about 1- 150 tint, about 1-200 nm about 1-300 nm, about 1-400 nm, about 1-500 nm, about 1 -1000 nm, or about 1-2000 nm.

EXAMPLE

Nanoparticle synthesis

J0065| The h>siru measurements of nanoparticles were performed, on a stagnation swirl flame setup for TiO? synthesis. Briefly, the main parts of the setup included a stainless steel, nozzle (burner) with I S mm inner diameter, an internal swirfer positioned 70 mm upstream of the nozzle outlet, and a water cooled substrate. Premixed gases of O?. CH , and ^'N_¾, were fed into the burner, forming a stagnation swirl flame between the nozzle and substrate, with separation distance fixed at 19 mm. The precursor (titanium tetraisopropoxide, ΤΤΪΡ), was radiantly heated in a bubbler to 353±1 with a temperature-controlled silicon controlled rectifier (SCR) and delivered to the burner by a nitrogen carrier gas flow.

{0066J The second harmonic (532 mil) and third harmonic (so-called 355 mn, which was actually 354.71 mn as measured for our system) of an NdrYAG laser operating at 1 OHz served as excitation source. The detector was an intensified charge coupled device (ICCD) camera, gated to minimize the interference from flame emissions and other sources. The typical gate width is 20 ns to 200 as, and the typical collection rime is 20 s to 400 s (200 shots to 4000 shots) for laser-induced emissions. The typical power used in the low-intensity LIBS measurement is 35 mi/pulse, corresponding to a local f!uence of 28 Van² at the focal point. A typical setup is shown in Figure 1. Components of the setup include; laser source generating laser pulses with suitable excitation wave-lengths (t ); mirrors (2 and 5); beam splitter (3); power meter (4);

aperture (6); photodiode (7); Sengs (8); flame synthesis section (9); beam damp (1.0); 2-D translation (11); notch filter (12); image rotator (13); oscilloscope (14); achroraats (15);

depolarizer (16); computer (17): ICCD camera (18); imaging spectrometer ( 1.9).

Low-intensity LIBS Emission Spectra

{0067 Fo the "low" laser intensity employed here, a selectivity of breakdown is manifested, where breakdown occurs only for nanoparticies hist not for gas molecules. This technique is quite different from that for conventional LIBS, where all the atoms in the measuring volume are excited, eliminating all molecular information and discrimination of the source of atoms (e.g. particle or gas phase). Here, the selectivity makes the technique particularly attractive for any gas-phase synthesis process because the particle phase can be identified at low power with no breakdown in the gas phase where precursor species (that contain metal atoms) are present. Figure 2 provides the typical low-intensity LIBS emission spectra acquired from a laser excitation of 532 nra (28 J/cnr ). Ti atomic spectra from HIST database is marked by red lines. The selectivity is further evinced in Figure 3 A, which shows emission intensity collected along the symmetry axis from distances of 2 mm to 16 mm downstream from the burner exit. No significant peaks were detected at distances less than 4 mm from the burner exit, where there was no particles or with particles smaller than 5 m The peaks started to appear at 4 ram. increased in intensity up to 12 mm, and maintained approximately constant intensity at distances further than 12 mm. Figure 3B displayed emission spectra collected at distances of 2 mm, 4 mm, 6 mm, 8 mm, and 14 mm; and significant increase in peak intensity were discerned.

Emission Intensity and Laser Power

[0068] Saturation is observed when increasing laser power higher than 20 mj/pulse, as shown in figure 4, suggesting that all Ti atoms in the particle phase are completely ablated from the nanoparticles and excited.

Emission Intensity and Precursor Concentration

|¾069| The emission intensities for different precursor concentrations are shown in Figure 5. The measurement volume was located at a distance of 14 mm from the burner exit, using a laser power of 35 mJ/polse, and was confirmed to be above the saturation power. The emission intensity increased fairly linearly with precursor concentration when the precursor concentration was larger than 150 ppm. The particles were smaller for small precursor concentrations, which reduced the absorption efficiency of the laser power. The linearity of the emission intensity curve after 150 ppm suggested that for particles larger than a certain size (-12 nm here), the absorption efficiency remained constant. The absorption efficiency increased with particle size and became constant for particles larger than -12 nm. This effect can be used to determine particle volume .fraction (single point or planer sheet) during synthesis, for particle sizes within a certain range.

Emission Intensity and Wavelength Selectivity

[0070} The emission spectra from the second harmonic (532nm) and from the third harmonic

(354,71 nm) of an injection-seeded Nd:YAG are shown in figures 6A and irrespectively. The markedly strong emission at 497.S34nm was from the resonant excitation where the 354.7tnm laser broke down flame-synthesized titanium-dioxide nanoparticles into their elements and then excited the titanium electrons resonantly (from a ^fG state to ¹ F^i! state, which is the upper state of transition line 497.534nm). I this way, by selecting the excitation wavelength, the limit of detectio can be significantly lowered. Moreover, the wavelength can be chosen such that only emissions from the atoms of interest are enhanced, thus achieving species selectivity. (0071 J The emission intensity, along the axis of symmetry for the setup, from 532 nm and 355 Tim laser excitation, are shown in Fig. 7. The intensity was calculated by integrating the emission from 497 mrs to 502 nra. The emission intensities both showed quick increase at 5- 12 mm distance from burner exit where the flame was located, indicating phase change and particle growth, ^'Nevertheless, it showed some differences tor these two excitation wavelengths. With 532 nm excitation, there was no emission around 500 nm at 2 mm away f om burner exit;

however, with 355 nm excitation, the atomic emission was present even at 2 mm away from the burner exit, albeit with low intensity. Since in this low-swiri flame burner the gases and precursor were mixed at a distance prior to the burner exit to ensure uniform composition, the TTIP precursor encounters <¼; and there was some reaction in the gas phase at - 100 X to form some TK¾ clusters at the burner exit, eventually becoming Ti<½ particles at longer residence times. These clusters can somehow absorb the 355 nm laser light and became excited and broke down, while the 532 nm laser was not well absorbed, due to the absorption efficiency difference for wave-lengths. This different, detection threshold for particle size can be used as a quantitative method for size characterization.

Emission intensity and Laser Power

|0072| The emission intensity change with laser intensity was studied and compared to the case with 532 nm laser excitation, as shown in Fig. 8, with the inset showing the emission intensity change with 532 nm laser excitation. For both cases, the measured volume location and precursor concentration were set to 14 mm from the burner e it and 1 16 ppm, respectively, where the particle size was -10 nm. With 355 nra excitation, the laser power changed from 0.5 mJ/puise to 35 mJ/puise, without gas-phase breakdown or visible plasma, which was not- observed until - 40 mJ/pulse. With 532 11m excitation, the LIBS emission intensity saturated at a laser power of 20 mj/pulse, and then remained at the plateau regime until gas broke down at -70 mJ^'/pulse. Nevertheless, the emission intensities from 355 nm laser excitation exhibited no saturation until gas-phase broke down at 40 mJ/pulse. The saturation at 20 mJ/pulse with 532nra excitation corresponded to the complete ablation of the particles, while the low-intensity LIBS with 355 nm excitation continued to excite more electrons resonantly: and the upper state transitions to other energy states, thus leading to the continuous increasing of the 497.534 nra emission, as well as other emissions.

Temporal Evolution of Emission Intensity

{0073 The detailed temporal evolution for different wavelengths as detected by PMT is il lustrated in Fig. 9. The pulse width (F WHM, foil width at half maximum) of the laser beam was fitted to be 10 ns, and the time when the laser pulse reached maximum was set to zero. The emissions around 497.53 nm (resonant emission) increased rapidly when the laser beam was present. However, the regular LIBS emission at 498.17 nm arose slowly but lasted for a longer time. With respect to the laser peak, the emissions around 497.53 nm came about 6 ns later, while the 498.17 nm emission reached its peak about 12 ns later. The emissions around 497.53 nm come 6 ns prior to the emission around 498.17 nm. Since the laser excited the electrons resonantly, it led to the fast increase of the emission, at 497.53 nm; however, this emission was more dependent on the laser and decayed when the laser beam was not sufficiently strong. After the laser pulse ended at -15 ns, the emissions around 497.53 nm lasted for another 15 ns, corresponding approximately to the transition decay time for the Ti (1) 497.534 nm line.

Emission Intensit and Particle Volume Fraction

(00741 The nanoparticle volume fraction measurement is demonstrated in Figure 10, where the particle volume fraction was controlled by different precursor loading rates. The 532nm laser power was 51 ,7 mj /pulse, which was in the saturation regime. The linearity between signal intensity and nanoparticle vol time fraction was quite good. Instead of a strictly proportional relation, there is an intercept due to the size dependent excitation efficiency, it appears that this effect can be well addressed by employing a cut-off volume fraction. Here, by linear

extrapolation, the intercept value is determined to be 140 ppb, corresponding to a particle diameter of 5.6 nm at the measuring point. Below 140 ppb, the intensity significantly deviates from the linear fitting. From 140 ppb to 870 ppb, the linear fitting gives an average accuracy of 8% deviation, which is better for higher volume fraction. For even higher volume fraction in a similar environment, the relation should be even more linear since the size-dependence is weaker.

Two-Dimensional Planar Measurement of Volume Fraction |0075| The two-dimensional planar LIBS for particle volume fraction measurement is demonstrated in figure 1 L The laser beam was shaped to a sheet, and two 500 »m band-pass filters were installed in series in front of the ICCD camera, to assure that inference from elastic scattering is blocked. The whole image was composed of four snapshots at different heights above the Bunsen burner (HAB 7,5-23,5 mm), starting from 3 mm before flame tip to 13 mm downstream from the tip. The rapid formation of TiO? nanoparrieles across the flame front was clearly seen, verif ing the fast chemistry of TOP reaction. The absence of signal before the flame zone demonstrated the phase-selectivity of this technique, showing no excitation when Ti atoms (e.g., as part of the TOP precursor) were in the gas phase. The main increase of signal was accomplished within 1 mm of the flame zone, where particles grew to about 7 nm according to population balance modeling. This portion of the increase came from die size-dependent excitation. As the particles continued coagulating in the post-flame gases, the signal was fairly uniform and stable, indicating the conservation of nanoparticle volume fraction during the coagulation process. The signal decayed at the left and right edges represented die gradient of volume fraction due to diffusion and entrainment. The results further demonstrated the good preservation of spatial resolution. There was a slight increase downstream, which was mainly caused by gas contraction as the burned gas temperature slowl dropped.

[6076} In order to demonstrate the measurement near walls, where particle dynamics as affected by diffusion and theraiophoresis is always of great interest, a substrate was placed perpendicularly to the issuing Bunsen flame; and the laser sheet was positioned to investigate the boundary layer. The image of nanoparticle volume fraction in the stagnation flame was shown in Fig. 12. The decay of signal downstream of the diverging flow showed the decrease of nanoparticle volume fraction due to the entrainment of ambient particle-free gas, as well as deposition loss to the substrate. The gradient in the boundary layer can be resol ved. Interference from bright reflections from the substrate surface was blocked by two tandem band-pass filters. The measurement of nanoparticle volume fraction near walls is very useful for the study of nanoparticle deposition, which is important for the synthesis of iiatio- films.

Monitoring and Controlling the Production of NanoparticJes |O077| Using excitaiion-imeiisity-dependent laser-induced breakdown spectroscopy, heavy metal-containing particles and other poi infant nanoparticles can be monitored in a particular environment. Regions or instances where gas-to-panicle transitions occur (e.g. nueieation can be identified.

[0078| Referring now to Figure 13, a method 1300 is provided. First, nanoparticles are produced using known gas-phase synthesis methods. For example, gas flow rates, ambient pressure, precursor loading, substrate temperature, and the like, are parameters that are involved to synthesize nanoparticles 1302. In-situ low-intensity phase-selective LIBS measurement is conducted 1304. The characteristics of the nanoparticles are determined 1306. If the nanopartic!e characteristics are in the proper range (1308: Yes) then the nanopartiele production, process is concluded 1310. If the nanoparticles are not hi the proper range (1308: No), then the process returns to step 1302 for refinement of the parameters for gas-phase synthesis of nanoparticles. Details on the feedback mechanism for determining and refining the parameters for gas-phase synthesis of nanoparticles are provided in Figure 14.

Claims

We claim^*.

.1. A method of characterizing particles by Laser-Induced Breakdown Spectroscopy (UBS), said method comprising: providing a gaseous flow field comprising a gaseous phase and a. particle phase, wherein said particle phase comprises the particles to be characterized; exciting said gaseous flow field with a series of laser pulses characterized by an excitation intensity effective to induce breakdown of said particle phase and excitation of the atoms within said particles without the breakdown of said gas phase and. exci tation of the atoms within said, gas phase, so that said excited atoms of said particles emit an atomic emission spectrum

characteristic of the elements within said particles; and detecting the emission spectrum produced by the excited atoms within said particles,

2. The LIBS method of claim I , wherein said laser pulse comprises a pre-detennmed excitation wavelength matching an atomic transition line of an elemental species within said particles.

3. The LIBS method of claim 1 , wherein said, laser pulses are provided by a tunable laser and are characterized by a plurality of pre-deterrnined excitation wavelengths matching the atomic transition Sine of more than one elemental species within said particles.

4. The L IBS method of claim 1 , wherein said laser pulses are provided by a plurality of lasers and are characterized by a plurality of pre-detemiined excitation wavelengths matching the atomic transition line of more than one elemental species within said particles.

5. The LIBS method of claims 3 or 4, wherein said plurality of predetermined excitation wavelengths are generated sequentially.

6. The LIBS method of claims 3 or 4. wherein said plurality of predetermined excitation wavelengths are generated simultaneously.

7. The LI BS method of claim 1 , wherein the fluence of the laser is adjustable,

8. The LI BS method of claim 1, wherein the ^"fluence of the laser is set at or above saturation of emission signals.

9. The LIBS method of claim 1 , farther comprising processing the emission spectrum to obtain a spatial measurement of the partic les based on the emission intensity of said spectrum,

10. The LIBS method of claim 3 , further comprising processing the emission spectrum to obtain a temporal measurement of the particles.

11. The LIBS method of claim. 9, wherei n the spatial measurement obtained by processing said emission spectrum is a volume fraction measurement of said particles,

12. The LI BS method of claim 1 1, wherein the laser pulses are in the form of a sheet and the emission spectrum is processed to obtain a planar volume fraction measurement of the particles,

13. The LI BS method of claim 1 , further comprising processing the emission spectrum to obtain a size measurement of the particles based on the emission intensity of said spectrum, provided that the concentration of the particles is known or constant.

14. The LIBS method of claim 1 , further comprising processing the emission spectrum to obtain a concentration measurement of the particles based on the emission intensity of said spectrum, provided that the size of the panicles is known or constant.

15. The LI BS method of c laim 2, further comprising processing the emission spectrum to determine the elemental identity of the particles, wherein the fluence of the laser is optionally adjusted according to different breakdown thresholds.

16. The LIBS method of claim 15, wherein said predetermined excitation, wavelengths matches one or more atomic lines of known (reference) elements.

17. The LIBS method of claim 1 , wherein said gaseous flow field is sampled from a

manufacturing process for producing said particles, atid said method further comprises the steps of processing the emission spectrirm to measure a characteristic of said particles and providing a feedback command according to the processed spectrum to said manufacturing process to control the measured characteristic of the particles,

18. The LIBS method of claim 1 , wherein the particles are collected from a pollution source or a particle-containing atmosphere.

19. The LIBS method of claim 1, wherein the particles are nanoparticles.

20. The LIBS method of claim 1, wherein the particles are n ieraparticies.