We investigated the physical and chemical changes induced in soot aggregates exposed to laser radiation using a scanning mobility particle sizer, a transmission electron microscope, and a scanning transmission x-ray microscope to perform near-edge x-ray absorption fine structure spectroscopy. Laser-induced nanoparticle production was observed at fluences above at and at . Our results indicate that new particle formationproceeds via (1) vaporization of small carbon clusters by thermal or photolytic mechanisms, followed by homogeneous nucleation, (2) heterogeneous nucleation of vaporized carbon clusters onto material ablated from primary particles, or (3) bothprocesses.
© 2007 Optical Society of America
Combustion-generated soot particles are composed of small ( diameter[1, 2, 3, 4, 5, 6]) carbon spheres held together by covalent bonds to form branched-chain aggregates of varied sizes.[3, 7] These small carbonaceous spheres are commonly called “primary particles,” a term that in this context refers to the elementary building blocks of the aggregate without implications regarding origin, and the aggregates are referred to as “mature soot particles” to distinguish them from the smaller ( diameter) nascent particles that are initially formed. The complex structure of soot aggregates, coupled with their typical small sizes , makes soot detection by optical methods challenging.[8, 9, 10, 11, 12, 13, 14] The work presented here focuses on identifying laser-induced changes in aggregate morphology and fine structure, which could have an impact on implementation of laser-based soot detection schemes.
One technique that may be particularly susceptible to laser-induced particle modification is laser-induced incandescence (LII). LII is a widely used optical technique for soot detection. It involves measuring the light emitted by particles heated by laser irradiation to temperatures (2500–4500 K) at which they incandesce (i.e., emit observable radiation). The signal magnitude is correlated with the particle volume fraction, whereas the signal decay rate is assumed to be associated with the primary particle size.[15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28] Accurate measurements of particle volume fraction and size are hindered by limitations in the understanding of the physical mechanisms that determine signal magnitudes and decay rates at operating temperatures and the paucity of information about the properties of soot at such temperatures. The laser fluences typically used and temperatures attained by the particles may be sufficient to lead to significant changes in particle morphology and fine structure, which could have a substantial effect on the LII signal.
To minimize uncertainties in LII particle sizing associated with particle vaporization, many recent studies have relied on low laser fluences for primary particle size measurements. An acceptable upper limit on laser fluence, however, is generally not clearly stated (if known). Often it is assumed that fluences are acceptable.[29, 30, 31, 32] Other recent studies have used laser fluences for primary particle sizing.[33, 34, 35, 36] For LII volume fraction measurements, laser fluences above are routinely used. LII signals increase in magnitude with laser fluence until they reach a threshold at , above which they often become nearly independent of fluence but maintain a dependence on soot volume fraction. Using laser fluences above this threshold value thus allows volume fraction measurements to be made with fewer complications from laser power fluctuations and laser attenuation through scattering or absorbing media.
In this study, we investigated the physical and chemical changes induced in soot aggregates exposed to laser radiation at 532 and over a wide range of laser fluences using results from a scanning mobility particle sizer (SMPS), a transmission electron microscope (TEM), and a scanning transmission x-ray microscope (STXM) to perform near-edge x-ray absorption fine structure (NEXAFS) spectroscopy. Extensive new particle production was observed to occur at fluences above at and above at . Our results indicate that new particle formationproceeds via (1) vaporization of small carbon clusters by thermal or photolytic mechanisms, followed by homogeneous nucleation, (2) heterogeneous nucleation of vaporized carbon clusters onto material ablated from primary particles, or (3) bothprocesses. Soot aggregates appear not to break apart into primary particles during laser heating, aprocess we will henceforth refer to as “disaggregation.” Our results also show that particles retain their original morphology, fine structure, and chemical structure with laser irradiation at fluences below the threshold for new particle formation.
2. Previous Results
The evidence in the literature is inconclusive with respect to whether soot particles disaggregate when laser heated in the visible or IR. Previous work suggested that nanoparticles may be formed by disaggregation of soot aggregates into primary particles when heated with a laser at . Filippov et al. used LII signal decay rates to infer aggregate size distributions. These distributions tended to shift toward smaller sizes with increasing laser fluence. Based on these observations, the authors suggested that aggregates break apart into primary particles during laser heating and concluded that size distributions of primary particles could thus be inferred from LII measurements at intermediate and high laser fluences .
A more recent study by Beyer and Greenhalgh provides evidence that soot particles disaggregate under vacuum when irradiated at . When monitoring time-resolved LII, they observed an increase in the LII signal microseconds after the laser pulse. They attributed this signal increase to previously shielded primary particles embedded in the aggregate, which become exposed during disaggregation. Aggregates or more loosely bound agglomerates also appeared to explode into smaller particles when imaged onto a gated intensified camera.
Other studies indicate that soot particles do not readily disaggregate into primary particles at nm. Vander Wal and co-workers[38, 39, 40, 41] presented TEM images of soot aggregates subjected to laser pulses. Soot samples were either first collected on copper grids and then exposed to laser radiation[38, 39] or irradiated in the flame prior to collection on the grids.[40, 41] Substantial changes in particle fine structure were observed. At intermediate fluences , mass appeared to be lost primarily from the centers of the primary particles, leading to formation of interconnected shell-like spheres. At higher fluences , significant mass loss from the surfaces of the particles occurred, resulting in partial disappearance of primary particle structure. Neither individual primary particles nor their fragments were observed on the TEM grids, however, suggesting that either mass was lost predominantly via sublimation of carbon clusters, or that the primary particles removed from the aggregates defied collection efforts or escaped from the TEM grids. These results are consistent with recent observations from Kock et al. who used an SMPS to measure aggregate size distributions before and after irradiation at 1064 nm. They did not observe a primary particle mode in the size distributions of particles irradiated at fluences between 0.1 and .
Dasch demonstrated that the scattering and absorption by soot decreased during laser heating at and fluences greater than , indicating a reduction in the aggregate particle size. According to model calculations that solved the energy and mass balance equations for laser heating of soot, this reduction in particle size could be explained by carbon sublimation. Dasch concluded that the observed changes in scattering and absorption during laser heating were more likely explained by a decrease in primary particle size by vaporization rather than by disaggregation of the aggregate into primary particles. Witze et al. and Yoder et al. similarly measured changes in (Ref. ) and (Ref. ) laser light elastically scattered from the particles as a result of laser heating. Both groups observed significant decreases in laser scatter at fluences above and speculated that the decrease in scatter resulted from particle-size reduction caused by vaporization. They concluded that this size reduction is insignificant below . Witze et al. confirmed their results with simultaneous extinction measurements. Krüger et al. irradiated a sample of soot at and observed a distortion in the size distribution measured with an SMPS at a laser fluence of . They postulated that this laser-induced reduction in the electric mobility size of particles at the larger end of the distribution could be attributable to carbon vaporization or some other unknown mechanism. They did not observe enhanced volume fractions of smaller diameter particles.
Previous studies demonstrated that carbon atoms are released from soot irradiated at (Ref. ) and that is released by irradiation at 516, 563, and .[47, 48, 49, 50] Using an SMPS to measure size distributions, Stipe et al. observed nanoparticle production from soot aggregates exposed to radiation. Nanoparticle formation was observed at laser fluences as low as . Both mobility diameter and nanoparticle number concentration increased with increasing fluence below , at which point they reached asymptotic values. At higher fluences (between 0.18 and ), the mean size decreased with fluence, and the concentration of new particles was independent of fluence. Because the observed sizes of small particles varied continuously from 20 to , the authors concluded that fragmentation of aggregates into primary particles is not the dominant mechanism for new particle formation. They hypothesized that small carbon clusters form new particles by homogeneous nucleation or by heterogeneous nucleation with larger fragmented primary particles. Incandescence was not observed, indicating that very little of the photon energy is converted to thermal energy. At these wavelengths, production of carbon clusters and other fragments likelyproceeds by direct photodissociation of covalent bonds within primary particles.
The results of Stipe et al. are consistent with studies of laser interactions with bulk graphite. When irradiated with a UV laser under vacuum, graphite produces small ionic and neutral clusters. The main products are C, , and and their cationic counterparts.[52, 53, 54, 55, 56, 57, 58, 59, 60, 61] Small clusters initially produced are thought to form larger clusters via thermolecular reactions,[62, 63, 64, 65] and at higher pressures larger clusters and nanoparticles have been observed.[66, 67] The fine structure of the graphite substrate also appears to have an impact on the size of clusters produced. Evidence suggests that single-crystal (i.e., well-ordered) graphite targets produce larger clusters than do amorphous carbon targets.[58, 68] With laser irradiation in the visible and IR, small clusters consisting of a few carbon atoms are produced,[53, 64, 65, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82] but larger clusters (with as many as 100 carbon atoms) are also commonly observed in these wavelength regions under vacuum.[55, 65, 71, 72, 83] At these wavelengths large clusters, fullerenes, and nanoparticles can be formed at higher pressures[62, 63, 74, 84, 85, 86, 87, 88] or when expanded in a buffer gas into a vacuum.[75, 76, 77, 89, 90, 91, 92] Cluster growth rates via termolecular reactions should increase with pressure, thereby leading to small diameter) particle formation at high pressure.[62, 66, 75, 80, 85, 86, 88, 89, 93]
In the present study, we focus on the effects on soot of laser irradiation at 532 and . In addition to making use of an SMPS to measure changes in electric-mobility size distributions following irradiation over a wide range of laser fluences, we use TEM and STXM images to study changes in soot morphology and fine structure and NEXAFS spectroscopy to study changes in chemical structure (carbon hybridization). The results from these studies collectively indicate that new particles produced during laser irradiation of soot are formed by nucleation of small carbon clusters generated by sublimation or photodesorption. Nucleationproceeds homogeneously and∕or heterogeneously onto larger fragments ablated by the laser. New particles appear not to be formed by disaggregation of soot aggregates into primary particles.
3A. Particle Generation
The experimental apparatus, shown schematically in Fig. 1, included a coflow diffusion burner, an annular silica gel diffusion dryer (TSI Model 3062), an atmospheric pressure flow-tube reactor with temperature and humidity probes, a cell with windows for optical access, and a scanning mobility particle sizer (TSI Model 3936L10). The SMPS includes a differential mobility analyzer [(DMA) TSI Model 3085] and a condensation particle counter [(CPC), TSI, Model 3025A]. Soot particles were generated in the flame and cooled to ambient temperatures at atmospheric pressure. The combustion products were dried by passing through the diffusion dryer and introduced into the SMPS after being subjected to a single laser shot at 532 or in the optical cell. The particles were size segregated as a function of their electric-mobility diameter with the DMA and were either counted with the CPC or collected on TEM grids using an electrostatic aerosol sampler (TSI Model 6069). Alternatively the full flow was directed onto a TEM grid or window for analysis using a STXM.
Soot was generated in an ethylene–air diffusion flame produced by a burner with a central fuel nozzle (0.5 cm in diameter) surrounded by a honeycomb structure for the coflow of air. The flow rates were maintained by mass flow controllers (MKS type 1479A for low flow and 1559A for high flow) in the range of 0.22–0.26 standard liters per minute (SLM) ethylene and 27.5 SLM air. The tip of the flame was cut with a cross flow of dry air at 65 SLM to prevent complete combustion of the particles in the flame. With the cross flow turned off, the visible flame height was . The cross flow intersected the flame at above the burner. A small fraction of the total gas flow was extracted into the diffusion dryer and sent to the optical cell, where it was irradiated with a single laser shot at either 532 or . Gas temperature and relative humidity were monitored with a thermohygrometer (Oakton, Model EW-35612-00) inserted through a sidearm in the flow tube.
3B. Optical Setup
The optical section of the flow tube consisted of a six-way stainless-steel cross equipped with optical access windows and an SMPS sampling port. A collimated pulsed laser beam was directed into the cell through a quartz window (tilted at Brewster's angle to minimize scatter) perpendicularly to the aerosol flow. The particles were irradiated with either the fundamental or second harmonic of an injection-seeded Nd:YAG laser (Spectra-Physics Pro-230-10), which produced pulses at a repetition rate of 10 Hz and with a duration of 10 ns at and 7 ns at . The center of the Gaussian beam was passed through a 3 mm diameter ceramic aperture, which was relay imaged to the detection region with a 1:1 positive telescope using two 1 m focal length lenses. This configuration provided a homogeneous tophat spatial profile 3 mm in diameter. The pulse energy was attenuated by use of a half-wave plate followed by two thin-film polarizers and was measured with a surface-absorbing thermal detector (Molectron Model PM10). The linear-flow velocity in the flow tube was adjusted to 6 cm∕s to ensure that particles were not subjected to multiple laser shots. The laser-beam diameter was smaller than the diameter of the aerosol flow tube, and at this flow velocity and with a 10 Hz laser repetition rate, only of the aerosol volume was irradiated.
3C. Transmission Electron Microscope Imaging
For TEM analysis, soot samples were passed through an electrostatic classifier and collected on lacey-carbon film on TEM grids (Electron Microscopy Sciences, LC200-Cu) using an electrostatic aerosol sampler. This sampling configuration allowed soot aggregates to be collected on the grids following segregation by electric-mobility diameter. The optimal collection time was found to be 180 min at a flow rate of 0.3 SLM and a potential of 10 kV on the grid. The particles thus collected were imaged using a JEOL 2000FX electron microscope operated at a 200 kV accelerating voltage. Images were then analyzed to determine the fractal and geometric dimensions of soot aggregates. Particle geometric size was measured on digitized electron micrographs by fitting ellipses to the particle outlines and taking the average of the short and long axes of the best-fit ellipse as the particle diameter. A box-counting method [ImageJ 1.32j (Ref. )] was used for the fractal analysis.
3D. NEXAFS Spectroscopy Using STXM
For NEXAFS analysis, soot samples were collected on TEM grids and windows. Samples containing more and larger particles were required for NEXAFS analysis than for TEM imaging. For this reason, the DMA size filter and electrostatic aerosol sampler were not used for the NEXAFS samples; hence a direct comparison between the TEM and STXM samples is not possible. Laser-irradiated or nonirradiated particles were collected by continuously impinging aerosol flow onto the substrates at a flow rate of 0.4 SLM and collection time of 150 min.
Carbon K-edge NEXAFS spectra of soot particles were acquired using a STXM on beamlines 5.3.2 and 11.0.2 of the Advanced Light Source (Berkeley, California) in a He-filled sample chamber. The majority of the experiments were performed at beamline 5.3.2, which has a bending magnet for an insertion device and produces horizontally polarized light in plane with the sample. For polarization-dependence measurements, a beamline with an elliptically polarized undulator insertion device (11.0.2) was used. For STXM measurements, the x-ray beam was focused with a zone plate onto the sample, and the transmitted radiation was detected. The spot size at the sample is determined by the width of the outermost zone of the zone plate, which was for these experiments. Images were obtained by rastering the sample and collecting transmitted light as a function of sample position. Spectra at each image pixel were obtained from stacks, which are a collection of images recorded at increasing energies across the absorption edge. Carbon K-edge spectra were acquired from 278 to 320 eV with dwell times of 1–2 ms per pixel. Incident beam intensity was measured through the sample-free region of the TEM grid. The logarithm of the intensity ratio between incident and transmitted beams (expressed as a function of energy) provided the NEXAFS spectrum in units of optical density at each image pixel. The NEXAFS spectra were obtained by averaging over regions of the particles.
On coated TEM grids (holey carbon, ultrathin carbon on holey-carbon support, silicon monoxide on Formvar, all from Ted Pella), particles were imaged in regions where the coating was perforated, allowing simultaneous measurement of the incident radiation and measurement of the coating in a region without soot. This holey-carbon spectrum was consistent with those measured from grids without soot and is readily distinguishable from soot in thickness, spectral intensities, and peak positions. For determination of the soot spectra, the holey-carbon region was used to measure the incident radiation. Spectra from each of the support films were examined and used in a similar manner. The soot particle spectra determined this way were similar for different TEM grid coatings.
Measured optical densities [(OD) or absorbance] typically ranged from 0.2 to 0.8; none were greater than 1. The energy was calibrated using the absorption bands of at 292.74 and 294.96 eV. The linear pre-edge background was subtracted from each spectrum. This background arises from scattering due to species that are nonabsorbing at a particular energy and was typically for both irradiated and nonirradiated soot particles. This small pre-edge absorption relative to the absorbance at higher energies ( at 320 eV) confirmed that the primary component of the particles is carbon.
4. Experimental Results
4A. Laser-Induced Size and Structure Changes
Figure 2 shows the measured electric-mobility size distribution of soot particles extracted from the flame (solid curve). This distribution is well described by a lognormal function, where the number concentration as a function of the natural logarithm of the particle diameter n(ln D) is given by97] 97]) of , and geometric standard deviation . Figure 2 also shows distributions of the average geometric diameter (given by an average of the maximum and minimum diameters for each particle) derived from 325 measurements from each of two TEM samples collected with the electrostatic sampler after size selection with the DMA. With the DMA set to collect particles with a mobility diameter of , the median average geometric diameter was determined to be , and for a mobility diameter of , the median average geometric diameter was . Figure 3(a) shows the corresponding TEM image, which demonstrates a typical branched-chain association of primary particles. The average particle fractal dimension is 1.7, which is also typical of mature soot generated in a flame.[2, 4, 13, 14, 98, 99, 100] In Fig. 3(b), the higher-magnification image of primary particles demonstrates fine structure composed of graphite crystallites with better-defined graphene (graphitic) sheets near the particle peripheries than at the centers, which is consistent with thereported fine structure of primary particles.[1, 3, 101, 102, 103, 104] The soot generated and studied in our experiments is similar in morphology and size to other flame-generated soot. Some of the aggregates appeared to have less well-defined primary particles with less turbostratic graphitic order apparent in the fine structure. These particles may have been less mature and may have originated from lower in the flame.
Figure 2 also shows the mobility-size distribution of soot particles irradiated with a single laser shot at with a fluence of (dotted curve). This distribution demonstrates an additional mode of particle sizes with smaller mobility diameters; this second mode is not apparent in the distribution in Fig. 2 for the nonirradiated soot. Although there is no reason to expect that the size distributions of the new particles and the irradiated aggregate can be strictly described by a lognormal distribution, this type of probability distribution provides a good approximation for the small- and large-mode size distributions. The bimodal distribution shown in Fig. 2 is well described by a sum of lognormal distributions, i.e.,
The size distribution of the aggregates thus demonstrates a small but measurable change upon irradiation. This distortion in the distribution appears as a depletion of larger particles. Figure 4 shows the distributions from Fig. 2 represented as volume concentrations, calculated as if the particles were spherical. The total particle volume (area under the curve) of the irradiated particles is that of the nonirradiated particles, i.e., the large increase in small particles does not violate conservation of mass. Because the particles are actually not spherical, this exercise will not give an accurate representation of total mass loss, but it does suggest that the change in the large-mode distribution with irradiation is probably sufficient to account for the production of new particles.
The number of new particles produced for each aggregate can be estimated according to
Figure 5(a) shows the TEM image of large-mode aggregates irradiated at with a fluence of . This figure demonstrates that the morphology of irradiated particles is similar to that of the nonirradiated soot shown in Fig. 3(a). The primary particles are not as regularly shaped as those that have not been exposed to the laser. These results are generally consistent with previous observations of mass loss accompanied by little change in the aggregate morphology at .[39, 40, 41] The fine structure of these irradiated aggregates appears to demonstrate more long-range order than observed in the nonirradiated particles [Fig. 5(b)]. These aggregates include primary particles with denser rings or folded ribbons of carbon layers that are not perfectly spherical. In previous work, these structures have been identified as annealed carbon material in which small graphene sheets (2–3 nm in length) form more extensive layered structures. At high temperatures, the introduction of five- and seven-membered rings into the hexagonal structure of these graphene sheets leads to curved surfaces. Some of the images show aggregates that appear to have sections with these features and sections that are similar to the original particles, but because only 10% of the particles in the samples have been irradiated, these images may include both irradiated and nonirradiated aggregates.
Figure 6 shows representative TEM images of particles from the small mode produced by the laser at and . Figure 6(a) presents an image of small irregularly shaped particles attached to a section of the lacey-carbon substrate. These particles appear to consist predominantly of carbon material with little or no long-range order. This material may be amorphous carbon, which is a network of carbon atoms with some short-range order (on a length scale) but little or no long-range order exhibiting a mixture of and hybridization.[105, 106, 107, 108] In addition, there appear to be isolated portions of the particles that form hollow layered carbon rings, as shown in the higher-magnification image in Fig. 6(b) (marked with an arrow). Similar results are obtained when the particles are irradiated at at high fluences, as shown in Fig. 6(c).
To elucidate the chemical structure (carbon-bond hybridization and functional groups) of these particles, single-energy STXM images and NEXAFS spectra were collected. NEXAFS spectral features arise from transitions from a core electron to an excited state and provide information about chemical hybridization and bonding. Spectra from the carbon 1s electron lie in the energy region of . Transitions from lie in the range of 285–291 eV. Aromatic compounds exhibit a strong peak at 285–286 eV arising from the transition of unsaturated bonds. Additional transitions from functionalized aromatic groups, such as carbonyl groups, lie in the range of 286–288 eV and may overlap with the transitions of or which begin at . Peaks arising from transitions to are usually broader than the peaks and are superimposed on the photoionization continuum.
Representative STXM images of nonirradiated soot particles and particles irradiated at and are shown in Fig. 7. Figure 7(a) shows an agglomerate of soot aggregates. Such agglomerates likely formed in situ during the samplingprocess. The irradiated particles [Fig. 7(b)] appear to be more compact, thicker, and denser. Such particles were not observed in the nonirradiated or low-fluence samples. Although only of the sample was estimated to be irradiated, particles similar to those shown in Fig. 7(a) were not observed for samples irradiated at high fluences The flow impinging on the substrate thus included both small- and large-mode particles from the irradiated distribution; because an electrostatic classifier was not used to collect the particles, sampling in this flow likely favored collection of the larger mode aggregates. The thickness of each sample was calculated from the atomic scattering factors assuming a density of . Thicknesses were typically for the nonirradiated and low-fluence samples, whereas the laser-irradiated particles were uniform in thickness and were typically thick.
Figure 8 shows representative normalized spectra from a nonirradiated sample and from a sample irradiated at with a fluence of . Measurements were made on ten nonirradiated samples and eight samples irradiated at this wavelength and fluence. Nonirradiated soot aggregates have a spectrum (Fig. 8, dotted line) similar to graphite, with a strong contribution from an component.[107, 108, 110] For comparison, the magic-angle spectrum of highly oriented pyrolytic graphite (HOPG) extracted from Lenardi et al. is shown in Fig. 8 (dashed curve). The graphite spectrum exhibits a strong peak at , which is attributed to the transition of aromatic carbon, and a transition at 292 eV, which corresponds to a carbon transition with enhancement from a sharp exciton transition at 291.7 eV.[110, 112, 113, 114] The exciton resonances have long lifetimes and as a result are very narrow.
The NEXAFS spectrum of nonirradiated soot particles (dotted curve) shows absorption features similar to those of HOPG. As noted above, the strong absorptions in the (285.3 eV) and (292.8 eV) regions point to the dominance of graphitelike sp 2 carbon. The linewidths of both and transitions are broader than those of HOPG, however, indicating additional (and σ*) resonances in the region between 286 and 290 eV. These differences between the highly oriented pyrolytic graphite (HOPG) and soot spectra are attributable to contributions from noncarbonaceous moieties known to be part of combustion-generated particles.[3, 116, 117] Resonances from (287.4 eV), C―H (288.7 eV), and C―OH (290 eV)[115, 118] can account for the additional absorption in the 286–290 eV region.
Particles irradiated with at (Fig. 8, solid curve) have a distinctly different spectrum from either HOPG (magic-angle spectrum) or the nonirradiated sample. The strength of the absorption is reduced, but the material exhibits a strong exciton at 291.7 eV. The appearance of the exciton in the spectrum of laser-irradiated particles could indicate the presence of some short-range order, which could be attributable to formation of annealed structures, such as those shown in Figs. 6(b) and 6(c). Such short-range order (on a length scale of ) is also consistent with amorphous carbon, which may have small regions of clustering of and bonded carbon atoms.[105, 106, 107, 108] Hence, the absorption features suggest a less graphitic structure (relative to nonirradiated aggregates) but, nevertheless, demonstrate short-range order, as indicated by sharp exciton absorption.
The relative amount of graphitic carbon in irradiated and nonirradiated soot samples can be estimated from the ratio of the area under the peak at (A C═C) to the overall area of the spectrum in the region between 280 and according to111] was referenced to the corresponding ratio for the HOPG (magic-angle spectrum) sample, assuming a 100% abundance of carbon in HOPG: 115] i.e.,
From Eq. (6), the average fraction of -hybridized carbon was estimated to be 72% for the nonirradiated soot particles and 21% for the particles irradiated at with a fluence of . The estimated uncertainty is . This latter value is consistent with the composition of amorphous carbon films deposited at room temperature. These results suggest significant changes in the carbon bonding in irradiated samples. The significant differences in particle morphology and spectra observed between these two samples lead us to speculate that the irradiated samples predominantly consist of newly formed nanoparticles.
Alternatively, the NEXAFS spectrum of the irradiated particles shown in Fig. 8 (solid curve) could arise from formation of an oriented carbonaceous material with polarization-dependent spectral intensities. The spectra of the irradiated particles exhibit a decrease in the absorption strength at 285.3 eV as compared to the nonirradiated samples. Similar decreases can be observed in HOPG spectra when measured with the sample rotated out of plane with linearly polarized light at an angle of incidence.[115, 118] In addition, a corresponding increase in the transition intensity of the carbon transition at 292.8 eV should be apparent for the out-of-plane sample. With this orientation, the ratio of the peak intensity at 285.3 eV to that at 292.8 eV can vary by over an order of magnitude with changes in polarization. We did not observe any significant increases in the intensity of the carbon transition for the irradiated particles. Therefore, it appears that the irradiated particles were not fortuitously aligned parallel to the substrate surface, and the decrease in the intensity of the aromatic carbon (285.3 eV) transition is not simply related to the particle alignment. In addition, to test for polarization-dependent spectral intensities, spectra were collected from eight different particles from the sample irradiated at 532 nm with . Variations in the spectral intensities were on the order of a few percent. The ratio of the peak at 285.3 eV to that at 292.8 eV was measured on a single region of this sample with (0.29) vertical, (0.32) horizontal, and (0.27) circularly polarized light. The lack of significant variation in the peak intensities for different particles supports the lack of variability in the spectra with polarization and implies that the carbon is not highly oriented.
For the nonirradiated samples, spectra were averaged over a region of the sample; hence long-range order, if any, would be averaged out in the measured spectrum. Although no polarization dependence was anticipated, analogous measurements on the nonirradiated samples yielded values of 0.71 (vertical), 0.74 (horizontal), and 0.67 (circular). This variation indicates the uncertainty of the measurement rather than significant changes in peak intensities.
Although we are unable to definitively identify a mechanism for the differences observed in the NEXAFS spectra between nonirradiated particles and those irradiated at , the significantly altered spectra suggest formation of newly nucleated nanoparticles. This conclusion is consistent with the TEM images of irradiated particles (Fig. 6), which demonstrate isolated regions of carbon with relatively large graphene sheets together with regions where such sheets are not evident. We speculate that the loss of structure observed in the TEM images could be the source of carbon for the newly nucleated particles observed in the STXM samples.
4B. Fluence Dependence of New Particle Formation
The aerosol mobility-size distributions were measured as a function of laser fluence at 532 and . Figure 9 demonstrates the effect of laser fluence on the number concentration and size of small particles produced during laser irradiation at . The solid curves in Fig. 9 represent the best fits to the particle-size distributions of a sum of two log-normal distributions, i.e., Eq. (3). The parameters derived from this analysis are given in Table 1. The full set of experimental data for irradiation is plotted in Fig. 10 in the form of a 3D graph of the measured particle-size distributions as a function of mobility diameter (X axis) and laser fluence (Y axis). The data show that small particles increase in size and number density with increasing fluence and reach a plateau in size and number concentration at higher values. Analogous results were observed for excitation, although the plateau regions in particle number concentration and size were not reached.
The median sizes of the small- and large-mode distributions are shown as a function of fluence in Fig. 11(a). The small-mode distribution increases with fluence at values above for irradiation and for irradiation. Particles with a median size below 8 nm were not observed at either wavelength. At 532 nm, the particle size becomes independent of fluence above . Such a plateau was not observed at at the fluences used in this study. The median particle size of the large-mode distributions starts to decrease with fluence at the value where new particle formation becomes observable. The median size of this mode continues to decrease at fluences below the intermediate values of for irradiation and for irradiation and increases with fluence at higher values. The behavior with fluence of the large-mode particles is mirrored by the total number concentration of these particles [shown in Fig. 11(b)]. At fluences above the threshold for new particle formation, the total volume of large-mode particles decreases with increasing fluence.
Figures 12 and 13 show TEM images of large-mode aggregates irradiated at 1064 and at intermediate fluences. Aggregates irradiated at and (Fig. 12) look similar to the nonirradiated particles (Fig. 3). The aggregate morphology does not change when irradiated at this wavelength and fluence, as demonstrated in Fig. 12(a), and most of the particles appear to have the same polycrystalline graphitic fine structure, as shown in Fig. 12(b). Aggregates irradiated at and (Fig. 13), however, demonstrate features similar to those observed in aggregates irradiated at and (Fig. 5). More long-range order is apparent in the fine structure, which demonstrates layered carbon rings or folded ribbons. The nanoparticles produced under these conditions (Fig. 14) are similar in structure to those produced at higher fluences (e.g., Fig. 6). Figure 14 shows representative nanoparticles produced by laser irradiation at 1064 and at intermediate laser fluences. In many of the particles generated, less long-range order is apparent than in the nonirradiated particles. In the case of irradiation at [Fig. 14(a)], small sections of graphitic structure are visible (circled). With irradiation at some of the new particles have no apparent long-range order [Fig. 14(b)] whereas other particles demonstrate structures with layered carbon planes, rings, or folded ribbons [Fig. 14(c)]. The particles generated at are smaller and more elongated than those formed at similar fluences with irradiation, and the morphology is similar to that of particles formed with high-fluence irradiation (Fig. 6). At higher fluences, however, fewer particles have regions of long-range order, and particles such as those shown in Fig. 14(c) were not seen at higher fluences.
NEXAFS spectra were recorded for soot particles exposed to radiation at laser fluences of 0.01, 0.05, 0.2, and . At fluences , STXM images show no changes in particle morphology, and NEXAFS spectra recorded over 15 different regions from these samples demonstrate no chemical bonding changes with fluence, as shown in Fig. 15. At fluences , distinct changes in the number of particles and particle morphology are observed. As noted above, the large branched agglomerates, such as those shown in Fig. 7(a), are no longer present. At a laser fluence of and greater, the primary particle structure of aggregates is not discernible in the samples collected for the STXM measurements. Rather, a few individual denser particles (submicrometer in size) are apparent [see Fig. 7(b)].
Fluence-dependent NEXAFS spectra indicate that a detectable change in the state of carbon hybridization in particles occurs when irradiated at laser fluences between 0.2 and at . At , this transition is characterized by an apparent decrease in graphitic sp 2-hybridized carbon.[105, 108] These changes are represented by the differences between the spectra shown in Fig. 15(a) and the spectrum shown in Fig. 15(c). Only two spectra were recorded for the sample illuminated at , and both of these spectra looked like the curve presented in Fig. 15(b). These spectra also demonstrate an apparent decrease in graphitic -hybridized carbon by a reduction in the peak at 285.3 eV. In this case, however, the aromatic carbon peak is shifted by 0.4 eV to a lower energy, which may indicate formation of fullerenic carbon. One striking difference between these spectra and those shown in Figs. 15(a) and 15(c) is the appearance of a strong peak at 288.7 eV in conjunction with a decrease in the graphitic carbon, indicating significant contributions from C―H. A single spectrum recorded with the sample irradiated at has similar features and is shown in Fig. 15(b). Although a detailed understanding of the mechanism for these spectral changes is beyond the scope of this paper, spectral changes observed in particles irradiated at fluences support the premise that soot particles are significantly altered by laser irradiation, which may result in the formation of newly nucleated nanoparticles.
4C. Effects of 532 versus 1064 nm Radiation on New Particle Formation
New particle formation was measured using the SMPS at 532 and . More new particle production occurred with the shorter wavelength. The number of new particles produced per aggregate is shown in Fig. 11(c) (left axis). As noted above, the onset of observable nanoparticle production (at which the concentration of small-mode particles is of the original particle concentration) occurs at at and at at . The yield of new particles is higher at than at at all fluences above . At , the yield reaches a plateau at fluences greater than ; such a plateau was not observed at 1064 nm at the fluences used in this study. The yield of new particles is comparable for the two wavelengths at the highest fluences . This dependence on fluence is similar to the observations of Stipe et al. at . In their work, however, the onset of new particle production occurred at , and small-mode particle growth became saturated at . These fluences are substantially lower than the corresponding valuesreported here for longer wavelengths.
The results of calculations of the fractional amount of volatilized carbon mass for laser heating are shown in Fig. 11(c) (right axis). Calculations were performed using a model that solves the energy- and mass-balance equations to account for particle heating by laser absorption, annealing, and oxidation and cooling by conduction to the surrounding atmosphere, radiative emission, sublimation, and multiphoton photodesorption of carbon clusters and . Particle-size reduction during sublimation, photodesorption, and oxidation is also calculated. The model is described in more detail elsewhere.
The amount of volatilized carbon was integrated from 0 to 150 ns, at which point mass loss by sublimation and photodesorption has ceased. Although the model does not currently include laser excitation at wavelengths other than , carbon volatilization is expected to be less efficient at 1064 than at for two reasons: (1) the absorption coefficient is inversely correlated with wavelength and is thus much lower at , and (2) the efficiency of photodesorption is lower at , assuming a two-photon transition at and a three- or four-photonprocess at . The calculated increase in mass loss with fluence at is more gradual than the increase in the number of new particles. The model indicates an earlier onset of substantial mass loss and does not reach a plateau until when the particle has been completely vaporized. New particle production via nucleation will depend on condensation rates. A mechanism for nucleation is not included in the model, and the experimental and modeling results shown here are thus not directly comparable. Nevertheless, they indicate that vaporization of the particle could generate sufficient gas-phase carbon species to account for the observed new particle formation by recondensation at these fluences.
Small carbon clusters (predominantly C, , and ) may thus sublime or photodesorb from soot aggregates and homogeneously nucleate to form new particles. This mechanism is consistent with observations of laser-induced particle production from graphite, particularly at wavelengths in the visible[75, 84, 87, 88, 89] and infrared[62, 74, 85, 86] regions at ambient pressures above 1 Torr or in free jet expansions into vacuum. Previous work on graphite has shown that small carbon clusters can react to form larger clusters,[62, 63, 64, 65, 91] which can then act as nucleation sites for particle growth. As indicated by the combined results of TEM images, STXM images, and NEXAFS spectra, thisprocess leads to apparent changes in the hybridization. These particles may partially anneal to form small regions with more ordered structure. Alternatively, under some conditions, surface ablation of the primary particle may produce graphitic fragments, which then serve as nucleation sites for new particle formation. Vaporization of such crystallites from graphite with semi-isolated graphite platelets has been hypothesized as a mechanism for generation of large carbon clusters.
The larger aggregates do not appear to be effective sites for heterogeneous cluster nucleation, e.g., the large regions of disordered carbon in Fig. 6 are not apparent on the aggregate in Fig. 5. At all fluences above the threshold for new particle formation, the median size of the large-mode aggregates is smaller than that of the nonirradiated particles, as shown in Fig. 11(a). A possible explanation is that at such high fluences the particles become superheated, pressure and temperature gradients are established at the surface, and convective transport of carbon clusters away from the surface leads to a supersonic expansion.[119, 121] These pressure waves and large flux of carbon away from the surface hinder transport of clusters back to the surface and reduce the rate of cluster recondensation on the original aggregate. Figure 11(a) also shows a growth in the median diameter of the large-mode aggregates with fluence following a minimum at intermediate fluences, suggesting that some amount of recondensation may occur when the cluster concentration is very high. Caution should be used in drawing such conclusions from these data, however, because this behavior of the large-mode median size and the anticorrelated behavior of the number of large-mode particles may be an artifact of the fittingprocedure. Uncertainties in these parameters are more significant for the large mode, for which the distribution was truncated. In addition, if significant recondensation were occurring on large-mode aggregates, TEM images showing material with no long-range order coating the aggregates might be expected but was not observed.
Our conclusions are consistent with those of Dasch, Witze et al., and Yoder et al. who measured changes in scattering and absorption cross sections during laser irradiation of soot and hypothesized that particle size reduction at fluences above occurred by vaporization of the particle. Our conclusions are also consistent with those of Vander Wal et al.[38, 39, 40, 41] who deduced from TEM images of laser-irradiated soot that aggregates do not disaggregate into primary particles upon laser irradiation but at high fluences lose mass via vaporization of primary particles.
Our results suggest that complete disaggregation of the aggregate into primary particles followed by reaggregation does not occur. Such disaggregation and reaggregation would likely lead to significant changes in the electric-mobility aggregate size distribution, which we do not observe. In addition, the small-mode component does not have the same physical and chemical structures as the original primary particles and increases in size with laser fluence, suggesting that, if disaggregation and reaggregation took place, it would have to be accompanied by a mechanism such as vaporization and recondensation. Some fragmentation of large aggregates into sections larger than a primary particle may occur, however, which might explain the decrease in median size [Fig. 11(a)] accompanied by an increase in total number concentration [Fig. 11(b)] of the large-mode distribution at intermediate fluences. Thisprocess may be overwhelmed by the effects of vaporization at higher fluences. Our conclusions are thus inconsistent with those of Filippov et al. who proposed that particles completely disaggregate during laser irradiation. The work of Beyer and Greenhalgh also suggested disaggregation of soot aggregates or agglomerates after laser heating; this work was performed under vacuum, however, and the particle dynamics could be strongly influenced by ambient pressure.
Our measurements of new particle formation are inconsistent with those of Kock et al. who recorded particle electric-mobility size distributions of laser-irradiated soot and did not observe new particle formation. They used fluences in the range of 0.1–0.6 at 1064 nm, for which we measure significant nanoparticle generation. Krüger et al. also did not observe new particle formation upon laser irradiation of soot in similar experiments, but their laser fluence at ) was approximately at our measured threshold for new particle generation, where our observations suggest that new particles are just starting to be observable with the SMPS. Vander Wal et al.[38, 39, 40, 41] did not observe new nanoparticles in TEM images of particles that did show apparent mass loss via vaporization of primary particles at high fluences, but nanoparticles may not have been collected efficiently on their TEM grids.
We have demonstrated the formation of carbonaceous nanoparticles during pulsed laser irradiation of soot at 532 and . New particles are produced at fluences above at and at (Fig. 11). The mean size and number density of these new particles, as measured by SMPS, increase with increasing fluence (Fig. 11). At high fluences for irradiation, for irradiation) new particle formation is accompanied by a decrease in number concentration of large-mode aggregates [Fig. 11(b)]. At intermediate fluences an increase in large-mode aggregate number concentration is accompanied by a decrease in large-mode median size, which may indicate that some fraction of the large aggregates break apart into fragments larger than a primary particle. At higher fluences an increase in the large-mode median diameter and decrease in large-mode number concentration with fluence may indicate recondensation of volatile carbon on the aggregates at high carbon cluster concentrations.
NEXAFS spectra show that the nonirradiated particles are predominantly graphitic -hybridized carbon and that particles irradiated at fluences above the threshold exhibit significant changes in this peak intensity (Fig. 8). In TEM images, small diameter) particles produced by irradiation of soot consist of extended regions without any obvious long-range order and smaller isolated regions of carbon with significant long-range order (Figs. 6 and 14). These ordered regions contain small graphite crystallites or partially annealed carbon that forms small, layered carbon rings or ribbons. Large particles diameter) irradiated at fluences above the threshold fluence consist of primary particles that appear to be similar in size and structure to the nonirradiated primary particles [Fig. 5(b)]. These large particles can also include primary particles with denser layered carbon rings [Fig. 5(c)], which may result from partial annealing. STXM images suggest that these irradiated aggregates are smaller and denser than the nonirradiated particles (Fig. 7).
These results suggest that particle growthproceeds through recondensation of small carbon clusters (e.g., C, , and ) by homogeneous nucleation, heterogeneous nucleation onto graphitic fragments of primary particles or a combination of both homogeneous and heterogeneous nucleation. Nanoparticle growth via homogeneous or heterogeneous nucleation of small carbon clusters is consistent with model predictions of the onset and extent of carbon volatilization by sublimation and photodesorption mechanisms. Reduced efficiency of particle formation at over is likely due to a smaller absorption coefficient and lower photodesorption efficiency at the longer wavelength. These results are consistent with previous studies of laser irradiation of soot particles and bulk graphite.
Laser-induced vaporization of soot particles and the resulting new particle formation could have significant implications for implementation of LII incandescence for soot detection. For volume fraction measurements, LII is frequently performed at fluences above the threshold for new particle formation. These results demonstrate that fluences as low as at could lead to nonnegligible mass loss of the irradiated particle and suggest that particle size measurements using LII and laser-based scattering and extinction measurements of soot particles should be confined to fluences well below this fluence. Morphology and fine structure do not appear to be strongly influenced by irradiation at fluences below the threshold value for new particle formation.
The work performed at Sandia was supported by the Division of Chemical Sciences, Geosciences, and Biosciences, the Office of Basic Energy Sciences (BES), the U.S. Department of Energy (DOE). Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the National Nuclear Security Administration under contract DE-AC04-94-AL85000. We thank Michael Gershenzon for his considerable contributions to the paper, Matt Boisselle for technical support, Pete Witze for stimulating discussions, Daniel Strong for the rendition of the experimental setup in Fig. 1, and Laurence Garvie for helpful suggestions for improving this manuscript. The work at the Advanced Light Source (ALS) of Lawrence Berkeley National Laboratory (LBNL) was supported by the DOE, BES, the Division of Chemical Sciences, Geosciences, and Biosciences, and the Division of Materials Sciences under contract DE-AC03-76S00098. M. K. Gilles and A. V. Tivanski thank A. D. Kilcoyne and T. Tyliszczak for support at beamlines 5.3.2 and 11.0.2, respectively. M. K. Gilles and A. V. Tivanski acknowledge funding from the DOE, Office of Science, Atmospheric Science Program. We gratefully acknowledge the use of facilities within the Center for Solid State Science at Arizona State University (ASU). The work at ASU was funded by the National Science Foundation under grant ATM-0531926. The authors gratefully acknowledge the perceptive and insightful comments of the reviewer.
1. J. Lahaye and G. Prado, “Morphology and internal structure of soot and carbon blacks,” in Particulate Carbon: Formation during Combustion, D. C. Siegla and G. W. Smith, eds. (Plenum, 1981), pp. 33–35.
2. Ü. Ö. Köylü and G. M. Faeth, “Structure and overfire soot in buoyant turbulent diffusion flames at long residence times,” Combust. Flame 89, 140–156 (1992). [CrossRef]
3. H. X. Chen and R. A. Dobbins, “Crystallogenesis of particles formed in hydrocarbon combustion,” Combust. Sci. Technol. 159, 109–128 (2000). [CrossRef]
4. B. Hu, B. Yang, and Ü. Ö. Köylü, “Soot measurements at the axis of an ethylene∕air nonpremixed turbulent jet flame,” Combust. Flame 134, 93–106 (2003). [CrossRef]
5. R. L. Vander Wal, T. M. Ticich, and A. B. Stephens, “Can soot primary particle size be determined using laser-induced incandescence?,” Combust. Flame 116, 291–296 (1999). [CrossRef]
6. L. H. van Poppel, H. Friedrich, J. Spinsby, S. H. Chung, J. H. Seinfeld, and P. R. Buseck, “Electron tomography of nanoparticle clusters and implications for atmospheric lifetimes and radiative forcing of soot,” Geophys. Res. Lett. 32, L24811 (2005). [CrossRef]
7. J. Lahaye and F. Ehrburger-Dolle, “Mechanisms of carbon black formation: correlation with the morphology of aggregates,” Carbon 32, 1319–1324 (1994). [CrossRef]
8. T. L. Farias, M. G. Carvalho, Ü. Ö. Köylü, and G. M. Faeth, “Computational evaluation of approximate Rayleigh–Debye–Gans∕fractal-aggregate theory for the absorption and scattering properties of soot,” J. Heat Transfer 117, 152–159 (1995). [CrossRef]
10. Ü. Ö. Köylü, “Quantitative analysis of in situ optical diagnostics for inferring particle∕aggregate parameters in flames: Implications for soot surface growth and total emissivity,” Combust. Flame 109, 488–500 (1996). [CrossRef]
11. J. C. Ku and K.-H. Shim, “A comparison of solutions for light scattering and absorption by aggregated and arbitrarily-shaped particles,” J. Quant. Spectrosc. Radiat. Transfer 47, 201–220 (1992). [CrossRef]
12. J. C. Ku and K.-H. Shim, “Optical diagnostics and radiative properties of simulated soot aggregates,” J. Heat Transfer 113, 953–958 (1991). [CrossRef]
13. Ü. Ö. Köylü and G. M. Faeth, “Fractal and projected structure properties of soot aggregates,” Combust. Flame 100, 621–633 (1995). [CrossRef]
14. C. M. Sorensen, “Light scattering by fractal aggregates: a review,” Aerosol Sci. Technol. 35, 648–687 (2001).
15. R. J. Santoro and C. R. Shaddix, “Laser-induced incandescence,” in Applied Combustion Diagnostics, K. Kohse-Höinghaus and J. B. Jeffries, eds. (Taylor & Francis, 2002), pp. 252–286.
16. J. A. Pinson, D. L. Mitchell, and R. J. Santoro, “Quantitative, planar soot measurements in a D. I. diesel engine using laser-induced incandescence and light scattering,” in Proceedings of the SAE (SAE, 1993), paper 932650.
17. J. A. Pinson, T. Ni, and T. A. Litzinger, “Quantitative imaging study of the effects of intake air temperature on soot evaluation in an optically-accessible D. I. diesel engine,” in Proceedings of the SAE (SAE, 1994), paper 942044.
18. R. L. Vander Wal and D. L. Dietrich, “Laser-induced incandescence applied to droplet combustion,” Appl. Opt. 34, 1103–1107 (1995). [CrossRef]
21. P. Roth and A. V. Filippov, “In situ ultrafine particle sizing by a combination of pulsed laser heatup and particle thermal emission,” J. Aerosol Sci. 27, 95–104 (1996). [CrossRef]
22. B. Quay, T.-W. Lee, T. Ni, and R. J. Santoro, “Spatially resolved measurements of soot volume fraction using laser-induced incandescence,” Combust. Flame 97, 384–392 (1994). [CrossRef]
23. C. R. Shaddix and K. C. Smyth, “Laser-induced incandescence measurements of soot production in steady and flickering methane, propane, and ethylene diffusion flames,” Combust. Flame 107, 418–452 (1996). [CrossRef]
24. A. V. Filippov, M. W. Markus, and P. Roth, “In situ characterization of ultrafine particles by laser-induced incandescence: sizing and particle structure determination,” J. Aerosol Sci. 30, 71–87 (1999). [CrossRef]
25. K. Inagaki, S. Takasu, and K. Nakakita, “In-cylinder quantitative soot concentration measurement by laser-induced incandescence,” in Proceedings of the SAE (SAE, 1999), paper 1999-1901-0508.
26. S. Schraml, S. Will, and A. Leipertz, “Simultaneous measurement of soot mass concentration and primary particle size in the exhaust of a DI diesel engine by time-resolved laser-induced incandescence,” in Proceedings of the SAE (SAE, 1994), paper 1999-1901-0146.
27. D. J. Bryce, N. Ladommatos, and H. Zhao, “Quantitative investigation of soot distribution by laser-induced incandescence,” Appl. Opt. 39, 5012–5022 (2000). [CrossRef]
28. C. Allouis, A. D'Alessio, C. Noviello, and F. Beretta, “Time resolved laser induced incandescence for soot and cenospheres measurements in oil flames,” Combust. Sci. Technol. 153, 51–63 (2000). [CrossRef]
29. T. Schittkowski, B. Mewes, and D. Brüggemann, “Laser-induced incandescence and Raman measurements in sooting methane and ethylene flames,” Phys. Chem. Chem. Phys. 4, 2063–2071 (2002). [CrossRef]
30. R. Starke, B. Kock, and P. Roth, “Nano-particle sizing by laser-induced incandescence (LII) in a shock wave reactor,” Shock Waves 12, 351–360 (2003). [CrossRef]
31. A. Boiarciuc, F. Foucher, and C. Mounaïm-Rousselle, “Soot volume fraction and primary particle size estimate by means of the simultaneous two-color-time-resolved and 2D laser-induced incandescence,” Appl. Phys. B 83, 413–421 (2006). [CrossRef]
32. F. Liu, M. Yang, F. A. Hill, D. R. Snelling, and G. J. Smallwood, “Influence of polydisperse distributions of both primary particle and aggregate size on soot temperature in low-fluence LII,” Appl. Phys. B 83, 383–395 (2006). [CrossRef]
33. S. Schraml, S. Dankers, K. Bader, S. Will, and A. Leipertz, “Soot temperature measurements and implications for time-resolved laser-induced incandescence (TIRE-LII),” Combust. Flame 120, 439–450 (2000). [CrossRef]
34. B. Axelsson, R. Collin, and P.-E. Bengtsson, “Laser-induced incandescence for soot particle size and volume fraction measurements using on-line extinction calibration,” Appl. Phys. B 72, 367–372 (2001).
35. C. Allouis, F. Beretta, and A. D'Alessio, “Sizing soot and micronic carbonaceous particle in spray flames base on time resolved LII,” Exp. Therm. Fluid Sci. 27, 455–463 (2003). [CrossRef]
36. T. Lehre, B. Jungfleisch, R. Suntz, and H. Bockhorn, “Size distributions of nanoscaled particles and gas temperatures from time-resolved laser-induced incandescence measurements,” Appl. Opt. 42, 2021–2030 (2003). [CrossRef] [PubMed]
37. V. Beyer and D. A. Greenhalgh, “Laser induced incandescence under high vacuum conditions,” Appl. Phys. B 83, 455–467 (2006). [CrossRef]
38. R. L. Vander Wal and M. Y. Choi, “Pulsed laser heating of soot: morphological changes,” Carbon 37, 231–239 (1999). [CrossRef]
39. R. L. Vander Wal, C. Y. Choi, and K. O. Lee, “The effects of rapid heating of soot: implications when using laser-induced incandescence for soot diagnostics,” Combust. Flame 102, 200–204 (1995). [CrossRef]
40. R. L. Vander Wal, T. M. Ticich, and A. B. Stephens, “Optical and microscopy investigations of soot structure alterations by laser-induced incandescence,” Appl. Phys. B 67, 115–123 (1998). [CrossRef]
41. R. L. Vander Wal and K. A. Jensen, “Laser-induced incandescence: excitation intensity,” Appl. Opt. 37, 1607–1616 (1998). [CrossRef]
42. B. F. Kock, Ph.D. dissertation (Universität Duisburg-Essen, 2006).
44. P. O. Witze, S. Hochgreb, D. Kayes, H. A. Michelsen, and C. R. Shaddix, “Time-resolved laser-induced incandescence and laser elastic scattering measurements in a propane diffusion flame,” Appl. Opt. 40, 2443–2452 (2001). [CrossRef]
45. G. D. Yoder, P. K. Diwaker, and D. W. Hahn, “Assessment of soot particle vaporization effects during laser-induced incandescence with time-resolved light scattering,” Appl. Opt. 44, 4211–4219 (2005). [CrossRef] [PubMed]
46. V. Krüger, C. Wahl, R. Hadef, K. P. Geigle, W. Stricker, and M. Aigner, “Comparison of laser-induced incandescence method with scanning mobility particle sizer technique: the influence of probe sampling and laser heating on soot particle size distribution,” Meas. Sci. Technol. 16, 1477–1486 (2005). [CrossRef]
47. C. J. Damm, D. Lucas, R. F. Sawyer, and C. P. Koshland, “Characterization of diesel particulate matter with excimer laser fragmentation fluorescence spectroscopy,” Proc. Combust. Inst. 29, 2767–2774 (2002). [CrossRef]
48. P.-E. Bengtsson and M. Aldén, “Soot-visualization strategies using laser techniques,” Appl. Phys. B 60, 51–59 (1995). [CrossRef]
49. C. B. Stipe, B. S. Higgins, D. Lucas, C. P. Koshland, and R. F. Sawyer, “Soot detection using excimer laser fragmentation fluorescence spectroscopy,” Proc. Combust. Inst. 29, 2759–2766 (2002). [CrossRef]
50. J. Walewski, M. Rupinski, H. Bladh, Z. S. Li, P.-E. Bengtsson, and M. Aldén, “Soot visualisation by use of laser-induced soot vapourisation in combination with polarisation spectroscopy,” Appl. Phys. B 77, 447–454 (2003). [CrossRef]
51. C. B. Stipe, J. H. Choi, D. Lucas, C. P. Koshland, and R. F. Sawyer, “Nanoparticle production by UV irradiation of combustion generated soot particles,” J. Nanopart. Res. 6, 467–477 (2004). [CrossRef]
52. F. Kokai and Y. Koga, “Time-of-flight mass spectrometric studies on the plume dynamics of laser ablation of graphite,” Nucl. Instrum. Methods Phys. Res. B 121, 387–391 (1997). [CrossRef]
53. F. Kokai, “Optical emission spectra from laser ablation of graphite at 266 nm and 1064 nm under a magnetic field,” Jpn. J. Appl. Phys. 36, 3504–3509 (1997). [CrossRef]
54. R. W. Dreyfus, R. Kelly, and R. E. Walkup, “Laser-induced fluorescence study of laser sputtering of graphite,” Nucl. Instrum. Methods Phys. Res. B 23, 557–561 (1987). [CrossRef]
55. J. J. Gaumet, A. Wakisaka, Y. Shimizu, and Y. Tamori, “Energetics for carbon clusters produced directly by laser vaporization of graphite: dependence on laser power and wavelength,” J. Chem. Soc. Faraday Trans. 89, 1667–1670 (1993). [CrossRef]
56. D. J. Krajnovich, “Laser sputtering of highly oriented pyrolytic graphite at 248 nm,” J. Chem. Phys. 102, 726–743 (1995). [CrossRef]
57. H. C. Ong and R. P. H. Chang, “Effect of laser intensity on the properties of carbon plumes and deposited films,” Phys. Rev. B 55, 13213–13220 (1997). [CrossRef]
58. Y. Yamagata, A. Sharma, and J. Narayan, “Comparative study of pulsed laser ablated plasma plumes from single crystal graphite and amorphous carbon targets. Part 1. Optical emission spectroscopy,” J. Appl. Phys. 88, 6861–6867 (2000). [CrossRef]
59. P. T. Murray and D. T. Peeler, “Dynamics of graphite photoablation: kinetic energy of the precursors to diamond-like carbon,” Appl. Surf. Sci. 69, 225–230 (1993). [CrossRef]
60. M.-A. Bratescu, Y. Sakai, D. Yamaoka, Y. Suda, and H. Sugawara, “Electron and excited particle densities in a carbon ablation plume,” Appl. Surf. Sci. 197–198,257–262 (2002). [CrossRef]
61. T. Shinozaki, T. Ooie, T. Yano, J. P. Zhao, Z. Y. Chen, and M. Yoneda, “Laser-induced optical emission of carbon plume by excimer and Nd:YAG laser irradiation,” Appl. Surf. Sci. 197–198,263–267 (2002). [CrossRef]
62. F. Kokai, K. Takahashi, M. Yudasaka, and S. Iijima, “Emission imaging spectroscopic and shadowgraphic studies on the growth dynamics of graphitic carbon particles synthesized by CO2 laser vaporization,” J. Phys. Chem. B 103, 8686–8693 (1999). [CrossRef]
63. T. Moriwaki, M. Kobayashi, M. Osaka, M. Ohara, H. Shiromaru, and Y. Achiba, “Dual pathway of carbon cluster formation in the laser vaporization,” J. Chem. Phys. 107, 8927–8932 (1997). [CrossRef]
64. K. Sasaki, T. Wakabayashi, S. Matsui, and K. Kadota, “Distributions of C2 and C3 radical densities in laser-ablation carbon plumes measured by laser-induced fluorescence imaging spectroscopy,” J. Appl. Phys. 91, 4033–4039 (2002). [CrossRef]
65. K. Shibagaki, T. Kawashima, K. Sasaki, and K. Kadota, “Formation of positive and negative carbon cluster ions in the initial phase of laser ablation in vacuum,” Jpn. J. Appl. Phys. 39, 4959–4963 (2000). [CrossRef]
66. M. Ullmann, S. K. Friedlander, and A. Schmidt-Ott, “Nanoparticle formation by laser ablation,” J. Nanopart. Res. 4, 499–509 (2002). [CrossRef]
67. Z. Márton, L. Landstrom, and P. Heszler, “Early stage of the material removal during ArF laser ablation of graphite,” Appl. Phys. A 79, 579–585 (2004). [CrossRef]
68. R. M. Mayo, J. W. Newman, Y. Yamagata, A. Sharma, and J. Narayan, “Comparative study of pulsed laser ablated plasma plumes from single crystal graphite and amorphous carbon targets: Part II. Electrostatic probe measurements,” J. Appl. Phys. 88, 6868–6874 (2000). [CrossRef]
69. H. Kamezaki, K. Tokunaga, S. Fukuda, N. Yoshida, and T. Muroga, “Pulse high heat flux experiment with laser beams on graphite,” J. Nucl. Mater. 179, 193–196 (1991). [CrossRef]
70. K. A. Lincoln and M. A. Covington, “Dynamic sampling of laser-induced vapor plumes by mass spectrometry,” Int. J. Mass Spectrom. Ion Phys. 16, 191–208 (1975). [CrossRef]
71. T. Wakabayashi, T. Momose, and T. Shida, “Mass spectroscopic studies of laser ablated carbon clusters as studied by photoionization with 10.5 eV photons under high vacuum,” J. Chem. Phys. 111, 6260–6263 (1999). [CrossRef]
72. J. Berkowitz and W. A. Chupka, “Mass spectrometric study of vapor ejected from graphite and other solids by focused laser beams,” J. Chem. Phys. 40, 2735–2736 (1964). [CrossRef]
74. K. Sasaki, T. Wakasaki, and K. Kadota, “Observation of continuum optical emission from laser ablation carbon plumes,” Appl. Surf. Sci. 197–198,197–201 (2002). [CrossRef]
75. E. A. Rohlfing, “Optical emission studies of atomic, molecular, and particulate carbon produced from a laser vaporization cluster source,” J. Chem. Phys. 89, 6103–6112 (1988). [CrossRef]
76. M. Anselment, R. S. Smith, E. Daykin, and L. F. Dimauro, “Optical emission studies on graphite in a laser∕vaporization supersonic jet cluster source,” Chem. Phys. Lett. 134, 444–449 (1987). [CrossRef]
77. E. A. Rohlfing, D. M. Cox, and A. Kaldor, “Production and characterization of supersonic carbon cluster beams,” J. Chem. Phys. 81, 3322–3330 (1984). [CrossRef]
78. M. Jeunehomme and R. P. Schwenker, “Focused laser-beam experiment and the oscillator strength of the Swan system,” J. Chem. Phys. 42, 2406–2408 (1965). [CrossRef]
79. A. M. Keszler and L. Nemes, “Time averaged emission spectra of Nd:YAG laser induced carbon plasmas,” J. Mol. Struct. 695–696,211–218 (2004). [CrossRef]
80. J. A. Howe, “Observations on the maser-induced graphite jet,” J. Chem. Phys. 39, 1362–1363 (1963). [CrossRef]
81. L. Nemes, A. M. Keszler, J. O. Hornkolh, and C. G. Parigger, “Laser-induced carbon plasma emission spectroscopic measurements on solid targets and in gas-phase optical breakdown,” Appl. Opt. 44, 3661–3667 (2005). [CrossRef] [PubMed]
82. S. S. Harilal, R. C. Isaac, C. V. Bindhu, V. P. N. Nampoori, and C. P. G. Vallabhan, “Optical emission studies of C2 species in laser-produced plasma from carbon,” J. Phys. D 30, 1703–1709 (1997). [CrossRef]
83. A. O'Keefe, M. M. Ross, and A. P. Baronavski, “Production of large carbon cluster ions by laser vaporization,” Chem. Phys. Lett. 130, 17–19 (1986). [CrossRef]
84. G. F. Lorusso, V. Capozzi, P. Milani, A. Minafra, and D. Lojacono, “UV spectra of graphite microparticles produced by laser vaporization,” Solid State Commun. 85, 729–734 (1993). [CrossRef]
85. F. Kokai, K. Takahashi, D. Kasuya, A. Nakayama, Y. Koga, M. Yudasaka, and S. Iijima, “Laser vaporization synthesisof polyhedral graphite,” Appl. Phys. A 77, 69–71 (2003). [CrossRef]
86. S. M. Kimbrell and E. S. Yeung, “Real-time particle size measurements in laser-generated plumes by Mie scattering,” Appl. Spectrosc. 43, 1248–1251 (1989). [CrossRef]
87. S. Iijima, T. Wakabayashi, and Y. Achiba, “Structures of carbon soot prepared by laser ablation,” J. Phys. Chem. 100, 5839–5843 (1996). [CrossRef]
88. T. Ishigaki, S. Suzuki, H. Kataura, W. Krätschmer, and Y. Achiba, “Characterization of fullerenes and carbon nanoparticles generated with a laser-furnace technique,” Appl. Phys. A 70, 121–124 (2000). [CrossRef]
89. E. A. Rohlfing and D. W. Chandler, “Two-color pyrometric imaging of laser-heated carbon particles in a supersonic flow,” Chem. Phys. Lett. 170, 44–50 (1990). [CrossRef]
90. H. W. Kroto, J. R. Heath, S. C. O'Brien, R. F. Curl, and R. E. Smalley, “: buckminsterfullerene,” Nature 318, 162–163 (1985). [CrossRef]
91. D. M. Cox, K. C. Reichmann, and A. Kaldor, “Carbon clusters revisited: the ‘special’ behavior of C60 and large carbon clusters,” J. Chem. Phys. 88, 1588–1597 (1988). [CrossRef]
92. M. Pellarin, E. Cottancin, J. Lermé, J. L. Vialle, and M. Broyer, “Coating and polymerization of C60 with carbon: a gas phase photodissociation study,” J. Chem. Phys. 117, 3088–3097 (2002). [CrossRef]
93. S. Suzuki, H. Yamagachi, R. Sen, H. Kataura, W. Krätschmer, and Y. Achiba, “Time and space evolution of carbon species generated with a laser furnace technique,” AIP Conf.Proc. 590, 51–54 (2001). [CrossRef]
94. “Imageprocessing and analysis in Java,” http://rsb.info.nih.gov/ij/.
95. A. L. D. Kilcoyne, T. Tyliszczak, W. F. Steele, S. Fakra, P. Hitchcock, K. Franck, E. Anderson, B. Harteneck, E. G. Rightor, G. E. Mitchell, A. P. Hitchcock, L. Yang, T. Warwick, and H. Ade, “Interferometer-controlled scanning transmission x-ray microscopes at the advanced light source,” J. Synchrotron Radiat. 10, 125–136 (2003). [CrossRef] [PubMed]
97. J. H. Seinfeld and S. N. Pandis, Atmospheric Chemistry and Physics from Air Pollution to Climate Change (Wiley, 1998).
98. T. T. Charalampopoulos and H. Chang, “Agglomerate parameters and fractal dimension of soot using light scattering-effects of surface growth,” Combust. Flame 87, 89–99 (1991). [CrossRef]
99. Ü. Ö. Köylü, Y. C. Xing, and D. E. Rosner, “Fractal morphology analysis of combustion-generated aggregates using angular light scattering and electron microscope images,” Langmuir 11, 4848–4854 (1995). [CrossRef]
100. J.-S. Wu, S. S. Krishnan, and G. M. Faeth, “Refractive indices at visible wavelengths of soot emitted from buoyant turbulent diffusion flames,” J. Heat Transfer 119, 230–237 (1997). [CrossRef]
101. W. S. Bacsa, W. A. de Heer, D. Ugarte, and A. Châtelain, “Raman spectroscopy of closed-shell carbon particles,” Chem. Phys. Lett. 211, 346–352 (1993). [CrossRef]
102. R. H. Hurt, G. P. Crawford, and H.-S. Shim, “Equilibrium nanostructure of primary soot particles,” Proc. Combust. Inst. 28, 2539–2546 (2000). [CrossRef]
103. R. L. Vander Wal, “A TEM methodology for the study of soot particle structure,” Combust. Sci. Technol. 126, 333–357 (1997). [CrossRef]
104. T. Ishiguro, Y. Takatori, and K. Akihama, “Microstructure of diesel soot particles probed by electron microscopy: first observation of inner core and outer shell,” Combust. Flame 108, 231–234 (1997). [CrossRef]
105. M. S. Dresselhaus, G. Dresselhaus, and P. C. Eklund, Science of Fullerenes and Carbon Nanotubes (Academic, 1996).
106. J. Robertson, “Amorphous carbon,” Adv. Phys. 35, 317–374 (1986). [CrossRef]
107. R. Gago, I. Jiménez, and J. M. Albella, “Detecting with x-ray absorption spectroscopy the modifications of the bonding structure of graphitic carbon by amorphisation, hydrogenation and nitrogenation,” Surf. Sci. 482–485,530–536 (2001). [CrossRef]
108. R. Gago, M. Vinnichenko, H. U. Jäger, A. Y. Belov, I. Jiménez, N. Huang, H. Sun, and M. F. Maitz, “Evolution of sp2 networks with substrate temperature in amorphous carbon films: experiment and theory,” Phys. Rev. B 72, 014120 (2005). [CrossRef]
109. B. L. Henke, P. Lee, T. J. Tanaka, R. L. Shimabukuro, and B. K. Fuikawa, “Low-energy x-ray interaction coefficients: photoabsorption, scattering, and reflection −E = 100–2000 eV, Z = 1–94,” At. Data Nucl. Data Tables 27, 1–144 (1982). [CrossRef]
110. R. A. Rosenberg, P. J. Love, and V. Rehn, “Polarization-dependent C(K) near-edge x-ray-absorption fine structure of graphite,” Phys. Rev. B 33, 4034–4037 (1986). [CrossRef]
111. C. Lenardi, M. Marino, E. Barborini, P. Piseri, and P. Milani, “Evaluation of hydrogen chemisorption in nanostructured carbon films by near edge x-ray absorption spectroscopy,” Eur. Phys. J. B 46, 441–447 (2005). [CrossRef]
112. R. Ahuja, P. A. Brühwiler, J. M. Wills, B. Johansson, N. Mårtensson, and O. Eriksson, “Theoretical and experimental study of the graphite 1s x-ray absorption edges,” Phys. Rev. B 54, 14396–14404 (1996). [CrossRef]
113. R. F. Willis, B. Fitton, and G. S. Painter, “Secondary-electron emission spectroscopy and the observation of high-energy excited states in graphite: theory and experiment,” Phys. Rev. B 9, 1926–1937 (1974). [CrossRef]
114. F. L. Coffman, R. Cao, P. A. Pianetta, S. Kapoor, M. Kelly, and L. J. Terminello, “Near-edge x-ray absorption of carbon materials for determining bond hybridization in mixed sp2∕sp3 bonded materials,” Appl. Phys. Lett. 69, 568–570 (1996). [CrossRef]
115. J. Stöhr, NEXAFS Spectroscopy (Springer, 1996).
116. M. B. Fernandes, J. O. Skjemstad, B. B. Johnson, J. D. Wells, and P. Brooks, “Characterization of carbonaceous combustion residues: I. Morphological, elemental and spectroscopic features,” Chemosphere 51, 785–795 (2003). [CrossRef] [PubMed]
118. G. D. Cody, H. Ade, S. Wirick, G. D. Mitchell, and A. Davis, “Determination of chemical-structural changes in vitrinite accompanying luminescence alteration using C-NEXAFS analysis,” Org. Geochem. 28, 441–455 (1998). [CrossRef]
119. H. A. Michelsen, “Understanding and predicting the temporal response of laser-induced incandescence from carbonaceous particles,” J. Chem. Phys. 118, 7012–7045 (2003). [CrossRef]
120. Y. P. Yang, P. Xia, A. L. Junkin, and L. A. Bloomfield, “Direct ejection of clusters from nonmetallic solids during laser vaporization,” Phys. Rev. Lett. 66, 1205–1208 (1991). [CrossRef] [PubMed]
121. H. A. Michelsen, P. O. Witze, D. Kayes, and S. Hochgreb, “Time-resolved laser-induced incandescence of soot: the influence of experimental factors and microphysical mechanisms,” Appl. Opt. 42, 5577–5590 (2003). [CrossRef] [PubMed]