Sub 500 nm refractory carbonaceous particles in the polar stratosphere

Eleven particle samples collected in the polar stratosphere during SOLVE (SAGE III Ozone loss and validation 10 experiment ) from January until March 2000 were characterized in detail by high-resolution transmission and scanning electron microscopy (TEM/SEM) combined with energy-dispersive X-ray microanalysis. A total number of 4175 particles (TEM=3845; SEM=330) was analyzed from these samples which were collected mostly inside the polar vortex in the altitude range between 17.3 and 19.9 km. By particle volume, all samples are dominated by volatile particles (ammonium sulfates/hydrogen sulfates). By number, approximately 28 – 82 % of the particles are refractory carbonaceous with sizes 15 between 20 830 nm. Internal mixtures of refractory carbonaceous and volatile particles comprise up to 16 %, individual volatile particles about 9 to 72 %. Most of the refractory carbonaceous particles are completely amorphous, a few of the particles are partly ordered with a graphene sheet separation distance of 0.37 ± 0.06 nm (mean value ± standard deviation). Carbon and oxygen are the only detected major elements with an atomic O/C ratio of 0.11 ± 0.07. Minor elements observed include Si, S, Fe, Cr and Ni with 20 the following atomic ratios relative to C: Si/C: 0.010 ± 0.011; S/C: 0.0007 ± 0.0015; Fe/C: 0.0052 ± 0.0074; Cr/C: 0.0012 ± 0.0017; Ni/C: 0.0006 ± 0.0011 (all mean values ± standard deviation). High resolution element distribution images reveal that the minor elements are distributed within the carbonaceous matrix, i.e., heterogeneous inclusions are not observed. No difference in size, nanostructure and elemental composition was found between particles collected inside and outside the polar vortex. 25 Based on chemistry and nanostructure, aircraft exhaust, volcanic emissions and biomass burning can certainly be excluded as source. The same is true for the less probable, but globally important sources: wood burning, coal burning, diesel engines and ship emissions. Rocket exhaust and carbonaceous material from interplanetary dust particles remain as possible sources of the refractory 30 carbonaceous particles studied. However, additional work is required in order to identify the sources unequivocally. Atmos. Chem. Phys. Discuss., doi:10.5194/acp-2017-278, 2017 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 3 April 2017 c © Author(s) 2017. CC-BY 3.0 License.


Introduction
The chemistry of stratospheric aerosols has been studied for more than half a century (Junge et al., 1961;Junge 1963), and it was discovered that sulfur is the main particle component.Junge and Manson (1961) supposed the particles to consist of ammonium sulfate, and Bigg et al. (1970) suggested sulfuric acid with varying amounts of ammonia.Rosen et al. (1971) strengthened the evidence for the material to be sulfuric acid as most of the material evaporated at the temperature expected for this substance.According to Bigg (1975), the majority of the particles is composed of sulfuric acid with varying amounts of ammonium sulfate.
In addition to the dominating sulfur-rich particles (sulfuric acid, sulfates), refractory particles were reported frequently.Dense, mineral-rich particles presumably originating from the eruption of the Mt.Agung volcano were observed by Mossop (1963Mossop ( , 1965) ) using scanning electron microscopy.However, due to the lack of instrumentation, the chemistry of the particles could not be investigated.Refractory particles with diameters >1 µm were studied in more detail by Zolensky and Mackinnon (1985), and several groups were distinguished: chondrite, silicate, aluminum (Al), aluminum with variable amounts of other elements, iron (Fe) with or without sulfate (S), calcium (Ca)-Al silicates and "other" particles.The silicate particles were dominant and interpreted to be volcanic, probably from the Mt.St. Helens eruption in 1980.In contrast to prior findings, a large refractory stratospheric particle load was recognized by Zolensky et al. (1989).He assumed that this increase was caused by solid rocket exhaust or the re-entry of debris associated to human space flight activity (inoperative satellites, burnt out rocket stages, tools, etc.).According to Sheridan et al. (1994) approximately 97% of all analyzed stratospheric particles were sulfuric acid.Also non-sulfate materials, soot, other C-rich substances and crustal material were detected.Carbonaceous aerosol was found to be widely distributed in the lower and middle stratosphere and interplanetary dust was significantly abundant above 30 km (Renard et al., 2008).Della Corte et al., 2013 found calcium-oxygen (CaO) -rich particles probably originating from a bolide that penetrated the earth' atmosphere.Single particle mass spectrometry (SPMS) brought new insights into the chemistry of stratospheric particles (e.g., Murphy et al., 1998Murphy et al., , 2007Murphy et al., , 2013)).According to these authors, stratospheric particles are dominated by pure sulfuric acid, sulfuric acid internally mixed with material from ablated meteoroids, and mixtures of organic-sulfate particles.A recent SEM study by Ebert et al. (2016) focused on refractory particles in the late winter stratospheric polar vortex.
The main particle groups encountered included Fe-rich, Si-rich, Ca-rich, metal mixtures and Carbon (C)/Si-rich particles.
Refractory particles in the UT/LS (Upper Troposphere / Lower Stratosphere) can act as condensation nuclei for cirrus clouds (Kojima et al., 2004;2005;Cziczo et al., 2013) and as surfaces for heterogeneous chemical reactions in the polar stratosphere which play a significant role in polar ozone depletion (e.g., Peter, 1997;Solomon, 1999;Peter and Grooß, 2012).In addition, the particles can serve as surfaces for the heterogeneous condensation of saturated gases in the polar stratosphere (Saunders et al. 2010(Saunders et al. , 2012;;Bigg, 2012;Frankland et al., 2015, Voigt et al. 2005;Hoyle et al. 2013;Engel et al. 2013).Due to the acidic environment, the particles can (partially) dissolve in the acidic solution droplets (binary HNO3-H2O or ternary HNO3-H2SO4-Atmos.Chem.Phys. Discuss., doi:10.5194/acp-2017-278, 2017 Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2017 c Author(s) 2017.CC-BY 3.0 License.H2O) and, thus, change their freezing properties.Therefore, the dissolved particles in a ternary HNO3-H2SO4-H2O solution could have an important impact on the formation of polar stratospheric clouds (PSCs).As gaseous compounds will condense on refractory particles, they will grow, both in size and mass, which leads to a change in their sedimentation velocity (Fromm et al. 2000;Jost et al., 2004).Therefore, the gaseous compounds can either be redistributed in the stratospheric region or sediment out more quickly.
There are multiple sources which contribute to the stratospheric refractory particle load.Interplanetary dust particles are considered to be the major component of refractory material (Murphy et al., 1998;2007;Plane, 2012).Another important source of stratospheric refractory particles are volcanic eruptions which may either eject material directly into the stratosphere (Vernier et al., 2011) or lead to particle transport through the tropical transition layer (TTL) (Mattis et al., 2010).Further potential sources of stratospheric refractory particles are high-flying aircraft (Fahey et al, 1995;Pueschel et al., 1997, Petzold et al., 1999), rockets (Newman et al., 2001), ablated material from meteorites (Hunten et al. 1980;Turco et al., 1981;Murphy et al., 1998;Cziczo et al., 2001) and all kinds of terrestrial material being lifted and entrained into the stratosphere by the Brewer-Dobson-Circulation (Holton et al., 1995, Austin andLi, 2006).
In summary, a variety of different refractory particle types is observed in the stratosphere.Due to the still limited number of sampling and measurement campaigns, the occurrence as well as the sources of refractory particles is not known precisely.
The present paper first aims at improving the database on the observed particle groups.Second, it is attempted to infer potential sources by a detailed characterization using high-resolution transmission electron microscopy (TEM), scanning electron microscopy (SEM) and energy-dispersive X-ray microanalysis (EDX).

Sampling
Stratospheric particles were sampled on board of the NASA ER-2 aircraft during the SAGE III Ozone loss and validation experiment (SOLVE), which was conducted in January-March 2000 in Kiruna (Sweden).The Multi-Sample Aerosol Collection System (MACS) (Kojima et al., 2004), a thin-plate low-pressure impactor was used for sampling.The particles were deposited on transmission electron microscopy (TEM) copper (Cu) grids covered with a formvar film.The MACS is designed to collect up to 23 samples, of which the first one always serves as a blank sample for reference.Sampling date, flight characteristics, potential vorticity (PV), maximum altitude, pressure and potential temperature are summarized for the analyzed grids in Table 1.
The meteorological conditions of the early winter (November -January) northern hemispheric polar vortex of 1999/2000 are described in detail by Manney and Sabutis, 2000.The early winter (November -January) northern hemispheric polar vortex of 1999/2000 had much lower averaged temperatures compared to any previously observed Arctic winter.It was weaker than Atmos.Chem. Phys. Discuss., doi:10.5194/acp-2017-278, 2017 Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2017 c Author(s) 2017.CC-BY 3.0 License. the early winter polar vortices of the previous years.The vortex was discontinuous in the middle of December, with a large extent in the upper and small extend in the lower stratosphere.By mid-January, the vortex evolved to be continuous and stable until mid-March (Greenblatt et al., 2002).Jost et al. (2002) describe anomalous single mixing events occurring during that time at the potential temperature (PT) range of 350-500 K.These events are probably the result of mixing between deep vortex and extra vortex air.
The stratospheric particle samples (deposited on TEM grids) taken within the polar vortex, were packed into single plastic boxes and stored in a desiccator prior to analysis.In total, 122 samples from 15 sampling days were collected.

Electron microscopy
A total of 4175 particles (3845 TEM; 330 SEM) from 11 samples were investigated by transmission and scanning electron microscopy.The samples were selected according to meteorological conditions and suitability for electron microscopy (i.e., substrate area covered by particles).
The scope of this study is the detailed characterization of refractory stratospheric particles.According to Ebert et al., 2016, we have classified all particles that are stable (no visible morphological change) under the high vacuum conditions and electron beam excitation in the SEM and TEM as refractory.
The size, morphology, mixing state, nanostructure and chemical composition of 60 refractory particles per sample were studied by TEM using a Philips CM20 instrument (FEI, Eindhoven, The Netherlands) operated at 200 kV electron accelerating voltage.
The chemical composition of the particles (all elements with an at drift X-ray detector (Oxford Instruments, Oxfordshire, United Kingdom) and a measurement time of 20 seconds.Particle size and graphene sheet separation distance were analyzed by the ImageJ (1.48v; Rasband, W. S. National Institutes of Health, USA, 1997-2016) software.Element distribution images were acquired with a JEOL JEM 2100F (JEOL, Tokyo, Japan) operated in Scanning Transmission Electron Microscopy (STEM) mode at an electron acceleration voltage of 200 kV.The instrument is equipped with an energy-dispersive X-ray Silicon-drift detector (Oxford, Wiesbaden, Germany).
In order to detect elements present at low abundance, additional 30 refractory particles per sample were analyzed by SEM using a Quanta 200 FEG instrument (FEI, Eindhoven, The Netherlands) equipped with an energy-dispersive X-ray detector (EDAX, Tilburg, The Netherlands) operated at 15 kV electron accelerating voltage.X-ray spectra were accumulated over fifteen minutes per particle to obtain a low detection limit.These longtime-measurements were not possible in the TEM due to a much higher beam energy leading to particle evaporation, which prevents further particle analysis.Element concentrations were obtained from the X-ray count rates by applying a "standard-less" ZAF correction.Detection limits of element/carbon ratios (at% / at%) for the long-time measurements are as follows: O/C = 0.0034; Si/C = 0.0010; S/C = 0.0008; Cr/C = 0.0008; Fe/C = 0.0009 and Ni/C = 0.0011.
The particles were studied by TEM and SEM without coating.
Potential contamination of the samples was checked by investigating blanks (samples transported in the MACS but not exposed to stratospheric air) for each sampling day.A few titanium (Ti) and zinc (Zn) oxide particles, as well as few pure C particles were encountered on the blank samples.They look similar to some carbonaceous particles being described as contaminants on TEM grids (Harris et al., 2001).However, these particles are different in both size and morphology compared to the particles observed in the impaction spot of the samples.
In order to verify that the small amounts of Fe, chromium (Cr) and nickel (Ni) detected during long-term SEM measurements are not artifacts from the substrate, five points on each substrate far away from particles were analyzed.All three elements were not detected in these substrate measurements.
Another possible artifact that could impair the measurement of elements present at low concentration is scattered radiation within the SEM.To exclude this possibility, one sample was measured at different sample chamber pressures (5x10-3, 200 and 500 Pa).Under the elevated pressures significant scattering of beam electrons occurs in the sample chamber.According to Stokes (2008), the extent of scattering varies between 40 -80 % at 200 Pa and 70 -98 % at 500 Pa.However, even with this large extent of scattering the elements Fe, Cr, and Ni did not show elevated concentrations.In contrast, when increasing the pressure in the vacuum chamber (i.e., increasing scattering of beam electrons) the count rates of these elements decreased.
Thus, it is concluded, that the low Fe, Cr and Ni concentrations detected in individual particles are real, i.e., they are not excited by stray radiation.
The chemical composition of the particles was determined by EDX in TEM (n = 529) as well as SEM (n = 330).Due to the small size of the particles, TEM is the method of first choice when analyzing the particles.As mentioned above, particles were additionally analyzed by SEM.Both measurements led to small but systematic differences in element ratios of O, Si, S, Mg (magnesium), Fe and Al relative to C (Figure S1 of the electronic supplement).For all elements TEM-EDX yielded somewhat higher elemental ratios (relative to carbon) than SEM-EDX.These differences are most likely the result of the different properties of the detectors (i.e., thickness of detector windows) and the highly different acceleration voltage (15 kV in SEM versus 200 kV in TEM).However, as the differences are small, our conclusions are independent of the method used.SEM-EDX data are used throughout the present paper, as the counting time on each particle was much longer in SEM (15 minutes) than in TEM (20 seconds) leading to higher precision as well as lower detection limits.The much lower detection limits of SEM-EDX are important for source identification as minor elements may serve as fingerprint for several anthropogenic and natural sources.Thus, the SEM-EDX data are preferred despite the lower number of particles investigated with this technique.
In addition, the chemical composition of particles could not be analyzed by TEM-EDX on two samples (labeled as G and K) due to the inappropriate position of the impaction spot on the substrate (to close to the Cu grid leading to a very high count rate for Cu).
In order to assess the relative abundance of refractory particles per sample, additional image analysis was performed in TEM.
For this purpose image series before and after TEM analysis were prepared.The destruction respectively morphological change after TEM analysis allows the assignment if individual particles were refractory, volatile or mixtures of both.All particles that are stable after electron beam treatment and not showing any change in morphology are classified as refractory.In contrast, volatile particles evaporate rapidly.In total 3316 particles were analyzed by this method.To ensure unbiased results, the individual images as well as particles for EDX analysis were randomly taken in inner and outer areas of the impaction spot.

Statistical analysis
Censored boxplots show data taking into account the fraction of values below detection limit.Lower and upper quartiles appear as a box, minimum and maximum values as whiskers.
Differences in element ratios between samples collected inside and outside the vortex were tested with the generalized Wilcoxon test (Helsel, 2012) applying a significance-level of 5%.Differences in size, projected area diameter and element ratios between the various samples were tested with the Kruskal-Wallis rank sum test (uncensored data) and the generalized Wilcoxon test (censored data).In all individual tests, s significance level of 5 % was applied.
Detection limits were calculated from counting statistics (background counts plus three times standard deviation of background counts).
All statistical calculations were performed with R (version 3.3.0;R Core Team, 2016) and using the contributed package NADA (version 1.5-6; Lee, 2013).

Results
All particles are located within a characteristic impaction spot.As TEM bright field images show, (Figure 1) volatile particles (initially deposited as droplets) cover a relatively larger area compared to refractory particles.They show high abundances of sulfur and oxygen.Sometimes, a minor nitrogen peak is also observed in the X-ray spectra.These particles are highly unstable under electron bombardment.They most likely consist of ammonium sulfate/hydrogen sulfate and formed from sulfuric/nitric acid.As the presence of sulfates in stratospheric samples is well known (e.g., Sheridan et al., 1989Sheridan et al., , 1994;;Arnold et al., 1998;Murphy et al., 2007;Kremser et al., 2016), these particles are not investigated further.In addition to the sulfates, carbonaceous particles stable under electron bombardment are frequently observed either as individual particles or embedded in the sulfates (Figures 1 and 2).
As the refractory carbonaceous particles in Figure 2a have no distinct shape and surface morphology, only one TEM image is shown.
By number, the refractory particles are dominating in all samples, except for sample G (Figure 3).In most of the samples, less than 10% of the particles are internal mixtures of refractory and volatile particles.Please note that each dropletnow consisting of mainly one large particle surrounded by many smaller onesis counted as one particle.

SEM Analysis of the chemical composition of the refractory carbonaceous particles
The refractory carbonaceous particles always contain besides C also O, Si, and S (Figures 2, 5 and 6).The elements Cr, Fe, and Ni are often found as minor component (Figure 5).As the TEM substrates are made of a formvar foil predominantly consisting of C, O, and traces of Si, these three elements mayat least partlybe an artifact of the substrate.There are three points of evidence which clearly show that the refractory carbonaceous particles observed are not an artifact of the substrate: 1) Much higher carbon X-ray count rates were obtained when measuring particles compared to the pure substrate.
2) Graphene sheets within the particles were observed by high resolution TEM.In contrast, the substrate is always completely amorphous.
3) The refractory carbonaceous particles only occur within the impaction spot.
The refractory carbonaceous particles have a different morphology than the few carbonaceous particles found on blank samples.In addition, the carbonaceous particles encountered on blank samples are often much larger with a size of several µm.Please note that a different foil with much higher O and Si contents was used for sample K. Thus this sample was excluded from figures 5, 7 and 9.The TEM grids consist of Cu leading to strong Cu X-ray peaks in the spectra (Figure 2).Consequently, Cu is excluded from the further analysis and discussion.Mg is only present in a few particles.S is the major component of the volatile material surrounding the carbonaceous particles.Fe is found in the majority (~ 95 %), Cr in about 87 % and Ni in about 49 % of particles.
The spatial distribution of minor elements within the carbonaceous particles was investigated by element distribution images (Figure 6) and by measuring several points on the same particle.C is the most abundant element and is contained in the whole particle.The elements O, Si, Cr and Fe only occur in some regions of the particle.The element Al is only detected in few particles.Due to the low number of X-ray counts, the distribution of Mg and Ni is difficult to assess.S seems to occur in the whole particle and is assumed to be present as surface coating of sulfate.The heterogeneous element distribution was also observed in multiple point measurements (up to 20 points on one particle).
Based on elemental composition, the refractory carbonaceous particles (number n = 330) were classified into four groups ( classification, because this element is found in all spectra.As all particles are either embedded in sulfates or have sulfate surface coatings it is assumed that S is not contained in the carbonaceous material. Element ratios relative to C (at% / at%) are shown in Figure 7 for all samples.The median O/C ratio varies between 0.052 and 0.129.The median Si/C ratios vary between 0.003 and 0.012, but may be influenced by the substrate.For all other elements, the respective median ratios are generally very low (< 0.005).Sample K is not shown due to the different substrate used (with lower C and higher Si content).The differences between the various samples are for all element ratios statistically significant on the 5 % level.Most obvious, sample G has lower element/C ratios than the other samples.

TEM analysis of the particle nanostructure
The nanostructure was investigated by high resolution TEM.All particles are either completely amorphous (Figure 8 a, b) or show only very small regions (less than ten graphene sheets) with ordering (Figure 8 c, d).In the latter case, the graphene sheet separation distance was determined (Table 3).As these measurements are very time-consuming, only 44 particles from 5 samples were investigated.The graphene sheet separation distance varies between 0.19 and 0.60 nm, the mean values of different grains between 0.25 and 0.47 nm.This size range is slightly larger for both lower and upper values than typically found for soot (Vander Wal et al., 2010;Li et al., 2011;Liati et al., 2014;Weinbruch et al., 2016).
The samples, with the exception of one (sample D), were collected inside the polar vortex.Element ratios of the samples inside and outside the vortex are compared in Figure 9.There are no statistically significant differences on the 5% level in element ratios and particle size (Figure 4) between these two cases.

Occurrence of refractory carbonaceous particles in the stratosphere
The major finding of the present paper is the high abundance (28-82% by number) of refractory carbonaceous particles (Figure 3), which mostly contain minor amounts of Fe, Cr and Ni heterogeneously distributed within the particles.The high abundance of refractory carbonaceous particles is in good accordance the findings of Weigel et al., 2014.With their measuring device COPAS (COndensation PArticle counting System, e.g.Weigel et al., 2009;Borrmann et al., 2010;Ebert et al., 2016) they detected a high amount of refractory particles (thermostable at 250 °C) within the polar vortex during three stratospheric campaigns (EUPLEX 2003, RECONCILE 2010and ESSenCe 2011;Weigel et al., 2014).
Refractory carbonaceous particles in the polar stratosphere were identified in several earlier studies.Depending on the applied technique different terms were used for such particles.In the present contribution the following nomenclature is used: All particles consisting predominantly of the element carbon are termed carbonaceous, i.e., only the chemical composition is used for classification.The term soot is used for agglomerates of primary particles (20 -50 nm size) predominantly consisting of carbon which show a specific, onion-shell like nanostructure (Buseck et al., 2012).Black carbon (BC) is used for particles strongly absorbing light in a wide spectrum of the visible wavelength (Petzold et al., 2013) with at least 5 m 2 g -1 at a wavelength of 550 nm (Bond et al., 2013).
As indicated by several publications, the presence of soot in the stratosphere was previously identified by scanning and transmission electron microscopy in accordance with the nomenclature outlined above (Pueschel et al., 1992(Pueschel et al., , 1997;;Sheridan et al., 1994;Blake and Kato, 1995;Strawa et al., 1999;Ebert et al., 2016).Chuan and Woods (1984) found carbonaceous particles which might be soot although they were not unequivocally identified as such, due to the lack of high resolution images.Volatile organic and amorphous particles, mainly composed of carbon, were reported by Sheridan et al. (1989).The amorphous particles shown by these authors have diameters of ~2 µm.Thus, the refractory carbonaceous particles of the present paper may be of similar origin but have significantly smaller size.Sheridan et al. (1989) did not specify any potential source of these particles.Testa et al. (1990) found seven poorly graphitized carbon particles in their samples and regarded them as artifacts of carbon films from the TEM grids.As they did not provide images of these particles, they cannot be compared to our findings.However, we can exclude that our particles are substrate contaminants, since such particles did not occur on the blank samples.In addition, the refractory carbonaceous particles were only observed within the impaction spot and not on the clean substrate.
Carbonaceous particles partly containing heterogeneous metallic inclusions were found by Chen et al. (1998).Some of these particles were denominated as soot without providing a more precise description.Thus, it is not clear if the "soot particles" observed by Chen et al. (1998) are similar to soot as defined by us.According to these authors, the "soot particles" most likely stem from aircraft exhaust, as the samples wereat least partlycollected in the exhaust of an aircraft in the lower stratosphere.
Mixed carbon-sulfur particles were observed by Nguyen et al. (2008) at 10 km altitude and were considered to have formed from condensed organic matter.However, particles differ in morphology and chemical composition from the refractory carbonaceous particles encountered in our study.
Stratospheric carbonaceous particles were also often detected by means of mass spectrometry (MS).For example, Murphy et al. (1998Murphy et al. ( , 2013) ) identified carbonaceous particles, with a lower abundance in the stratosphere compared to the upper troposphere.The same group (Murphy et al., 2007) reported the presence of a small fraction of carbonaceous particles (diameter ≤ 300 nm) within the stratosphere.As some of these particles contained potassium, they were assumed to originate from biomass burning.Particles from rocket and space shuttle exhaust (collected in the stratosphere) were investigated by Cziczo et al. (2002).In solid fuel rocket exhaust (Athena rocket and space shuttle boosters) the most frequent observed signals stem from different aluminum oxide species (often with minor amounts of Fe).About 17% of the particles (Athena rocket) were classified as carbonaceous as compared to 1% from space shuttle exhaust.In principle, the carbonaceous particles found by MS could be similar to the carbonaceous particles of the present study.Since MS does not yield images or information on the nanostructure, no definite conclusion can be drawn.Stratospheric carbonaceous particles were also identified by optical measurements.For example, a single particle soot photometer (SP2) was applied by Baumgardner et al. (2004) on board of an aircraft in order to identify ozone loss processes in the polar vortex.In total, 60% of the light absorbing particles incandesce at temperatures above 3500 K and are, thus, interpreted as BC.According to Baumgardner et al. (2004) these particles originate from tropospheric sources rather than aircraft emissions.The same technique was applied by Schwarz et al. (2006) to identify BC in midlatitudes from the boundary layer to the lower stratosphere.Only ≤1% of the particles was classified as BC and no potential source was specified.
Local enhancements of carbonaceous material at altitudes around 25 km were also deduced from simultaneous measurements of a spectrometer on board a satellite as well as from radiance and particle counter data (Renard et al., 2008) obtained on a stratospheric balloon.This material was thought to be injected into the stratosphere by the pyro-convective effect (i.e., fires that have sufficient energy to trigger convection and inject particles to the upper troposphere and lower stratosphere).Some of the particles with submicron size were supposed to originate from vaporized interplanetary material (Renard et al., 2008).Due to the lack of information on morphology, chemistry and microstructure of the particles, a direct comparison of the carbonaceous material deduced from optical measurements with the particles encountered in the present study is not possible.

Potential Sources
The most likely sources of refractory carbonaceous particles found in the current study include aircraft emissions, extraterrestrial sources, rocket exhaust and explosive volcanic emissions as these sources emit material directly into the stratosphere.In addition, biomass burning can be considered as a possible source, as it was shown by several authors that large fires have sufficient energy to inject particles into the lower stratosphere (Fromm et al., 2000, 2006, Siebert et al., 2000;, Fromm and Servranckx, 2003;Jost et al., 2004;Siddaway et al., 2011).Domestic wood burning, coal combustion, diesel engines and ship exhaust are not expected to significantly contribute to the stratospheric particle load.However, as these sources emit large amounts of carbonaceous material on a global scale (Bond et al., 2004, Gaffney and Marley, 2005, Corbett and Koehler, 2003;Lauer et al., 2007), they need to be discussed shortly.For example Thornberry et al., 2010  tropical tropospheric particles contain high amounts of thermostable carbonaceous material, and it is possible that these particles become mixed within stratospheric air.Although the vertical exchange may be less effective than the direct injection processes, however, it is conceivable that fractions of carbonaceous aerosol material released in the troposphere is vertically transported into the stratosphere by processes such as tropical convection (and lifted further via the Brewer-Dobson circulation), cyclogeneses, warm conveyor belts, tropopause folds and/or isentropic transport.

Aircraft exhaust
High flying aircraft can contribute significantly to the stratospheric aerosol burden.Soot is described as the main particulate exhaust component (Twohy and Gandrud, 1998;Popovitcheva et al., 2000;Smekens et al., 2005;Vander Wal et al., 2010;Liati et al., 2014).The observed soot consisted of primary particles 10 -50 nm in diameter which formed aggregates with diameters of up to approximately 450 nm (Twohy and Gandrud, 1998;Popovitcheva et al, 2000;Smekens et al., 2005;Liati et al., 2014).The nanostructure of the primary particles varied from an onion-shell structure with partial ordering (Popovitcheva et al., 2000;Vander Wal et al., 2010) to completely amorphous (Twohy and Gandrud, 1998).The mean graphene sheet separation distance of the partly ordered primary particles varied between 0.36 and 0.41 nm (Vander Wal et al., 2010;Liati et al., 2014).An atomic O/C ratio of 0.32 was reported by Vander Wal et al. ( 2010).The elements S, Na (sodium), N (nitrogen), Zn (zinc), Ba (barium), Cl (chlorine), K (potassium), Fe and Si were detected at minor concentrations (Vander Wal et al., 2010;Mazaheri et al., 2013).In addition to soot agglomerates, single carbon spheres were found by Mazaheri et al. (2013) in aircraft exhaust.The particles have diameters between 5-100 nm with a median of 35.4 nm.Besides C, minor O, S, Cl, K, Fe and Si were detected by TEM-EDX and Proton Induced X-ray Emission (PIXE) analysis.Aircraft exhaust is excluded as source of the carbonaceous particles encountered in the present study, as we did not observe soot agglomerates.In addition, the chemistry and morphology (basically rounded shapes) of the single carbon particles described by Mazaheri et al. (2013) are different from our particles.

Extraterrestrial particles
With an input of 5-270 tons per year (Plane, 2012), extraterrestrial material is expected to be the major source of refractory stratospheric particles (Murphy et al., 1998;2007).Carbonaceous material is observed in chondrites as well as in interplanetary dust particles (IDPs).
In carbonaceous chondrites, a variety of different carbonaceous constituents is described in previous literature.For example, nanometer-sized carbon-rich flakes, spheres and tubes as well as hollow carbon-rich nanospheres were found (Garvie and Buseck, 2004;Garvie, 2006;Garvie et al., 2008).Most carbonaceous nanospheres are amorphous.Besides C, the only other elements detected are S, N and O. Carbonaceous material in carbonaceous chondrites was also investigated by Aoki and Akai (2008).Different morphologies like "ribbon-film-like carbonaceous material", "spherical carbonaceous globules", "concentric-sphere type carbon material", and "featureless carbon material" were observed.They also describe the occurrence of minor amounts of Cl in many of these particles, an element never observed in our study.Neither morphology nor shape or chemical composition of the particles described above is matching the refractory carbonaceous particles encountered in the present study.Carbon nanoparticles with diameters between 2-10 nm were observed in carbonaceous material isolated from the Allende carbonaceous chondrite (Harris et al., 2000).The particles had either a single outer wall or were surrounded by multiple walls.However, as the particles were mobile under the electron beam, no photographs are shown.A comparison to our particles is thus difficult.Since the authors did not find any other elements except C, most of our particles differ in chemical composition as they contain Fe, Cr, and/or Ni as minor elements.
Carbonaceous material was also found in IDPs (Matrajt et al., 2012).Some of the material observed is similar in size, morphology and nanostructure to our particles.However, in Matrajt et al. (2012) minor elements were not investigated and this parameter cannot be compared.
In summary, the carbonaceous components observed in chondrites differ in chemical composition from most of our particles.
Carbonaceous material contained in IDPs cannot be excluded as source of the refractory particles encountered in the present study, as the minor element concentration of the former is not known.However, if we expect extraterrestrial material to be the major source of our particles, we would also expect to find Mg-rich silicates in our samples, which was not the case.
Furthermore, the occurrence of Fe, Cr and Ni as minor elements in our refractory carbonaceous particles is regarded as hint for an anthropogenic origin.This interpretation is supported by the fact that the observed average atomic ratios of these three elements (Cr/Fe = 0.249, Ni/Fe = 0.167, Cr/Ni = 1.145) are significantly higher than the cosmic element ratios (Cr/Fe = 0.015, Ni/Fe = 0.056, Cr/Ni = 0.278; Palme and Jones, 2005).

Rocket Exhaust
Rocket exhaust is also a possible source of stratospheric particles.However, literature on particles emitted by rockets is sparse (e.g.Zolensky et al., 1989), and there areto the best of our knowledgeno studies available on carbonaceous particles by electron microscopy.According to Ross and Sheaffer (2014), five different propellant types which use a combination of different oxidizers and fuels are in use: O2/kerosene, O2/H2, NH4ClO4/Al, N2H4/N2O4 and N2O/solid hydrocarbons.Solid rocket motors (SRM) emit characteristic Al2O3 spheres (Strand et al., 1981;Zolensky et al., 1989;Cofer III et al., 1991) and can, thus, be excluded as source of the carbonaceous particles encountered in the present study.Hydrocarbon-fired rockets are powered by kerosene or syntin and can be expected to emit soot.For example, soot particles most likely emitted by a Russian Soyuz booster rocket were found in an aerosol cloud at 20 km (Newman et al., 2001).The soot particles comprised approximately 1/4 to the total particle number, the rest were volatile sulfate particles.The occurrence of carbonaceous material from rocket exhaust was also reported by Cziczo et al. (2002).In the exhaust of an Athena II rocket, the carbonaceous fraction of material was found to be 17% by number (Cziczo et al., 2002).The refractory carbonaceous particles observed by us may in principle originate from rocket exhaust, which could also explain the occurrence of the minor elements Fe, Cr and Ni (rocket debris).However, as carbonaceous rocket exhaust particles were not investigated previously by electron microscopy this question remains open.

Volcanic emissions
Volcanic eruptions are generally not considered to emit carbonaceous material.However, carbonaceous material was found in samples from the Kasatochi (Alaska, 2008), Sarychev (Russia, 2009) as well as Eyafjallajökull (Island, 2010) eruptions (Martinsson et al., 2009;Schmale et al.;2010;Andersson et al., 2013).The carbonaceous material most probably originates from air entrained into the volcanic cloud.The carbonaceous mass fraction in plumes from two volcanoes in Ecuador and Columbia was found to vary between 19-38% (measured by MS; Carn et al., 2011).Sulfur-carbonaceous mixed particles occurred predominantly in the size range below 0.9 µm.In 1999, the year before the campaign, there were four volcanic eruptions with a VEI ≥ 4 (Volcanic Explosivity Index): Soufrière Hills (West Indies) in January, Shiveluch (Russia) in August as well as Guagua Pichincha (Ecuador) and Tungurahua (Ecuador) in October (NOAA, 2017).Still, volcanism seems an unlikely source for our samples, as volcanic eruptions will emit large amounts of silicate particles, which we did not observe.

Biomass burning
A further source of carbonaceous particles in the stratosphere is biomass burning (BB).Particles originating from BB can be lifted to the stratosphere by either the tropical upper tropospheric upwelling or the pyroconvective effect (Fromm et al., 2000;Jost et al., 2004).Three different types of carbonaceous BB particles were described in previous literature (e.g., Pósfai et al., 2003;2004;Kis et al., 2006;Li and Shao, 2009): organic particles with inorganic inclusions, tar ball particles and soot.Organic particles (no specific morphology) contain C and minor O, and are stable under the electron beam.They do not show the typical microstructure of soot (see below).The inorganic inclusions mostly consist of KCl and K2SO4.Tar balls have a typical spherical shape and mainly consist of C and O with minor K, S, Cl and Si contents.Soot consists of chain-like agglomerates of primary particles (10 -100 nm) with a typical onion shell microstructure (graphene sheet separation distance between 0.133 and 0.137 nm).As our particles do not show the characteristics of all types of carbonaceous BB particles described above, this source can be excluded.
The most probable potential sources for carbonaceous particles were already discussed above.For these sources the transport mechanisms into the stratosphere are well known.There are further strong tropospheric sources for carbonaceous particles, which predominately emit particles at ground level.An effective transport to the stratosphere of these particles is unlikely.
Still, they will be discussed briefly, as they areon a global scalemajor sources of carbonaceous material in the lower atmosphere.

Wood burning
Soot is a major component emitted by wood burning.Similar to biomass burning, soot from wood burning consists of agglomerates of spherical primary particles (20 -80 nm diameter) with an onion-shell nanostructure (Kocbach et al., 2005;Torvela et al., 2014, Tumolva et al., 2010).Some primary particles are amorphous (Tumolva et al., 2010).The particles may have a surface coating which is volatile under electron bombardment (Torvela et al., 2014).Carbon is the major element of wood burning soot, O, Na, Si, S, Cl, K, and Ca occur as minor elements (Kocbach et al., 2005;Tumolva et al., 2010).In addition to soot, particulate organic matter (POM) was found in wood burning (Torvela et al., 2014).The POM particles, sometimes described as tar balls are 30 -600 nm in diameter.Judged from the properties described above, both soot and POM are different to the particles observed in our study.Thus, wood burning can be excluded as a source for the refractory carbonaceous stratospheric particles.

Coal burning
A variety of different carbonaceous particles was observed during coal burning.Most carbonaceous particles are soot particles, i.e., fractal-like agglomerates (0.1 -1 µm size) consisting of 10 -50 nm diameter primary particles.The primary particles show a characteristic onion-shell structure (Chen et al., 2004(Chen et al., , 2005)).Some of the soot agglomerates have inorganic inclusions or inorganic particles on their surface, containing Mg, Ca, Sr (strontium), Ba (barium) and Na (Chen et al., 2005).Char particles associated with ultrafine titanium oxide particles were also found (Chen et al., 2005).Sometimes ultrafine Ti, Al, Fe and Ca particles were embedded in large char particles.In addition, graphitic fiber structures that are either straight or curved were encountered.Two of the three types of carbonaceous particles described for coal combustion (soot agglomerates and graphitic fibers) are certainly different from our particles.As images of char particles were not provided by Chen et al. (2005) we cannot compare their results to the refractory carbonaceous particles of the present study.We regard coal combustion as an implausible source.

Ship emissions
Ship emissions also contain soot agglomerates (Popovicheva et al., 2012;Lieke et al., 2013).Their chemical composition is dominated by C and O, with small amounts (< 1 wt %) of additional elements; e.g.V, S, Cl, Ca and Si (Popovicheva et al., 2012).Furthermore, spherical char particles with diameters of 0.2 -1 µm are found to be characteristic for ship emissions (Popovicheva et al., 2012).The morphology of these particles shown in SEM photomicrographs is different to the particles found in our study.Lieke et al., 2013 found amorphous carbonaceous material filling cavities of larger soot aggregates.The characteristics described for ship exhaust particles are significantly different to the particles found in our study, leading to the conclusion that that ship exhaust can be excluded as the source.

Summary
The major finding of the present study is the exceptional high abundance of refractory carbonaceous particles collected in and outside of the stratospheric polar vortex.By number, up to 82 % of the particles were found to be externally mixed refractory carbonaceous particles.In addition, approximately 10% of the particles are internal mixtures of refractory carbonaceous particles and sulfates.Most of the carbonaceous particles show small and variable amounts of the elements Fe, Cr and Ni.
The exact source of the refractory carbonaceous particles remains unclear and can only be confined by exclusion.The most likely sources are rocket exhaust and extraterrestrial material.
Atmos.Chem.Phys.Discuss., doi:10.5194/acp-2017-278,2017   Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2017 c Author(s) 2017.CC-BY 3.0 License.The size distribution of the refractory carbonaceous particles is shown in Figure4.Approximately 98 % of the particles have an equivalent projected area diameter (Dpa) below 500 nm (range 20 -830 nm).The size of the particles slightly increases with time during the campaign, sample J shows the largest median particle sizes.

Figure 1 :
Figure 1: TEM bright field image of a typical sample (sample I; 17.3 km altitude), before (a) and after evaporation (b).Particles evaporating under electron bombardment are marked with blue circles.They consist of sulfates/hydrogen sulfates.Red circles indicate stable carbonaceous particles.Green circles show refractory carbonaceous particle internally mixed with volatile sulfates/hydrogen sulfates.5

Figure 2 :
Figure 2: TEM bright field image (a) of a typical refractory carbonaceous particle from sample H (19.1 km altitude).The image is representative for all particles.The morphology is not depending on chemical composition, size, morphology or nanostructure.Energy-dispersive X-ray spectra of (b) a typical refractory carbonaceous particle with Fe, Cr and Ni, (c) Fe and 5 Cr, (d) Fe and (e) without any other minor constitute.The particle predominantly consists of C and O. Minor amounts of Si are always present and may partly be an artifact of the substrate.Cu is an artifact from the TEM grid.

Figure 3 :
Figure 3: Relative number abundance of refractory carbonaceous and volatile particles as well as internal mixtures of both.

Figure 4 :
Figure 4: Boxplot of particle size (equivalent projected area diameter Dpa).Lower and upper quartiles appear as a box, minimum and maximum values as whiskers.5

Figure 5 :Figure 6 :
Figure 5: Median chemical subcomposition (atom %, without C and O) of refractory carbonaceous particles determined by SEM-EDX (30 particles per sample).Sample K was excluded from the figure due to a different substrate with higher Si content.

Figure 7 :
Figure 7: Censored boxplots of element ratios relative to C (atom %) determined by SEM-EDX (30 particles per sample).Sample K is not shown due to the different substrate used.Lower and upper quartiles appear as a box, minimum and maximum values as whiskers.Values below detection limit (horizontal line) are not shown.

Figure 8 :
Figure 8: High resolution TEM image of individual refractory carbonaceous particles from sample G (17.4 km altitude): (a, b) completely amorphous particles, (c, d) particles showing small regions with ordering.

Figure 9 :
Figure 9: Censored boxplots of element ratios (atom %) relative to C, separately for outside and inside the polar vortex.Sample K was excluded from the analysis due to the different substrate used.Lower and upper quartiles appear as a box, minimum and maximum values as whiskers.Values below detection limit (horizontal line) are not shown.

Table 2
).Only few particles consist of C, O, and Si only.Many particles contain additional Cr, Fe and Ni (n = 131) or Cr and Fe (n = 125).Particles only containing additional Fe are rare (n = 22).Please note that the element S was not used for particle Atmos.Chem.Phys.Discuss., doi:10.5194/acp-2017-278,2017 Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2017 c Author(s) 2017.CC-BY 3.0 License.