THE PATH OF REDUCED NITROGEN TOWARD EARLY EARTH: THE COSMIC TRAIL AND ITS SOLAR SHORTCUTS

Carbonaceous Chondrite (CC) meteorites are asteroidal fragments of near solar elemental abundances that avoided the geological reprocessing of planet formation. CC stones have the appearance of aggregate rocks and contain carbon as soluble and insoluble organic materials (IOMs), intermixed with silicates and other inorganic phases diverse enough in composition and ¹⁸O/¹⁶O fractionation to allow their classifications into various groups. Their studies have provided insights on primordial Solar System constituents as well as temperature and compositional parameters of early solar nebula events (Russell et al. 2006). In regard to the cosmic evolution of carbonaceous materials, CC meteorites' analyses demonstrated that abiotic organic materials preceded biochemistry in the early solar system and could be as diverse as kerogen-like complex macromolecules and simpler soluble compounds varying from polar species such as amino acids or polyols to non-polar hydrocarbons (e.g., Pizzarello & Shock 2010). Furthermore, the finding that CC compounds may display high D-enrichments (Pizzarello & Huang 2005) comparable to those observed in molecular clouds of the interstellar medium (ISM) indicates that HCNO-containing interstellar precursors could survive in asteroidal bodies. Given that most CC parent bodies are known to have been further processed during their early solar residence, with internal heating (∼0°C–125°C), consequent release of liquid water and associated effects on their minerals such as the formation of clays (Brearley 2006), it has been generally understood that the presolar precursors were further processed via aqueous reactions into the organic molecular species found upon meteorites' extraction.

This interstellar-parent body formation hypothesis (Cronin & Chang 1993) has one lacuna, however, and that is that the ¹⁵N anomalies of several CC compounds cannot be easily explained by interstellar processes since large ¹⁵N enrichments in general, and for the ammonia envisioned in some of these reactions (Lerner & Cooper 2005) in particular, have not been detected in the ISM in spite of their prediction. Some of the latest cosmic abundances measured for ¹⁴N/¹⁵N ratios were: 290 ± 40 for the local ISM (Adande & Ziurys 2012), ∼350–850 (Gerin et al. 2009) and 334 ± 50 (Lis et al. 2010) for ammonia specifically, and 424 ± 3 for protosolar environments, taken as the nitrogen composition of the solar wind measured during the Genesis mission (Marty et al. 2010); the ratio of the terrestrial atmosphere is +272. The corresponding values in the geochemical ${\delta }^{15}$N‰ notation used for meteorites (see Table 1) are, respectively: −84.7‰, −222 to −680‰, −186‰, and −374‰, where 0 is the terrestrial atmosphere.

Table 1.  Average δ¹⁵N Values of Largest "Hotspots" in Individual Meteorites' IOM and HT IOM, HT Released Ammonia^a, and Bulk IOM^b

	IOM			HT-IOM
Meteorite Name	${\delta }^{15}$N‰	#	μm2 Area Mounts	${\delta }^{15}$N‰	#	μm2 Area Mounts	${}^{1}{\delta }^{15}$N‰ NH3	${}^{2}{\delta }^{15}$N‰ Bulk IOM
Murchison (CM2)	568 ± 143	5	5400	456 ± 146	1	3050	51.7	−1 ± 7.5
Ivuna (CI1)	1323 ± 146	4	4125	548 ± 130	9	5831	66.4	31.9
Bells (C2-ung)	1412 ± 249	8	1900	1178 ± 120	12	700	455.5	415 ± 1.6
Tagish Lake (C2-ung)	1620 ± 198	7	4675	1436 ± 128	3	4050	140	130 ± 1.5
GRA 95229 (CR2)	1364 ± 118	13	5175	919 ± 241	3	3150	223	153 ± 6.7
Allende (CV3)	nf	⋯	2250	nf	⋯	⋯	nf	−51 ± 0.6
Sutter's Mill (C-ung)	nf	⋯	2220	nf	⋯	⋯	nf	nm

Notes. ${\delta }^{15}$N‰ = [(¹⁵N/¹⁴N_sample/¹⁵N/¹⁴N ${}_{\mathrm{standard}})$−1] × 10³; ¹⁵N/¹⁴N_standard is the ratio of terrestrial atmosphere (3.676·10${}^{-3},\;$ ${\delta }^{15}$N‰ = 0). #, indicates the number of hotspots identified in at least 0.03 mm² of each sample; nf, not found; the acronyms within parentheses after the meteorites' name indicate: their class (C, for carbonaceous, plus the initial letter of the prototypical meteorite, e.g., M for Mighei; I for Ivuna etc.) and petrologic type (1–3, 1 = maximum water alteration). Sutter's Mill and Allende did not display hotspots or released ammonia upon HT treatment and were included as procedural blanks.

^aPizzarello & Williams (2012). ^bAlexander et al. (2007).

Download table as:  ASCIITypeset image

Only limited ¹⁵N isotopic fractionation by ion molecule reactions has been considered possible for interstellar clouds in some studies, based on nitrogen's slight fractional ionization (Terziera & Herbst 2000), while other models predicted even large enrichments, e.g., for nitrile-containing compounds and ammonia in CO depleted ISM regions (Rodgers & Charnley 2008a, 2008b) or based on spin states dependence between N₂ and H₂ in ion-molecules reactions (Wirström et al. 2012). So far, however, ¹⁵N-enrichments in the ISM have been measured only for HCN, and within broad ranges and diverse environments. Observations of H¹³CN and HC¹⁵N rotational transitions estimated ¹⁴N/¹⁵N ratios of 140–360 in protostellar cores (Hily-Blant et al. 2013), added measurement of C¹⁴N and C¹⁵N lead to 120–400 ratios across the galaxy (Adante & Ziurys 2012) and a search for extragalactic HC¹⁵N inferred ratios of 111 ± 17 (Chin et al. 1999).

By contrast, large ¹⁵N-enrichments have been detected consistently in small bodies of the Solar System: spectroscopically for HCN and CN in comets (Bockelée-Morvan et al. 2008; Manfroid et al. 2009), by Secondary Ion Mass Spectroscopy (SIMS) for micrometer sized interplanetary dust particles (IDPs) collected in the high atmosphere (Messenger 2000), and more frequently in meteorites, which, being amenable to direct laboratory analyses, allowed measurements of positive ¹⁵N/¹⁴N enhancements for several N-containing classes of compounds, amines (Pizzarello et al. 1994), amino acids (Engel & Macko 1997; Pizzarello & Holmes 2009; Elsila et al. 2012), ammonia (Pizzarello et al. 2011), and polar hydrocarbons (Pizzarello et al. 1994), with ${\delta }^{15}{\rm{N}}$ values for these molecules reaching ∼250‰. Above all, it was the IOM of CC meteorites that revealed the highest ${\delta }^{15}{\rm{N}}$ isotopic fractionations, which are found in the form of localized submicron-sized areas showing ¹⁵N isotopic anomalies as high as ${\delta }^{15}$N ∼ 5000‰ (Busemann et al. 2006; Briani et al. 2009) and called "hotspots," as for similar anomalies of other elements.

The IOM represents the major portion of the carbon in CCs, up to ∼99% by weight, and is often compared to terrestrial kerogens for its macromolecular complexity, insolubility, and content of H, O, N, S, and P in addition to C. The IOM is highly heterogeneous and several of its chemical and isotopic features are still not fully understood, e.g., its precise chemical composition. It was long believed that polycyclic aromatic hydrocarbons (PAHs) were the primary IOM components (Allamandola et al. 1989), however, recent findings assigned the infrared features attributed to PAHs to combined aromatic and aliphatic components of no fixed arrangements (Kwok & Zhang 2011). Experimental studies have also proposed a new pathway for producing complex presolar aggregates (Schutte et al. 1993), and IOM formation in particular (Cody et al. 2011), via polymerization of formaldehyde, an abundant component of the ISM organic inventory and, presumably, the solar nebula.

It is plausible, in fact, that the IOM is a mixed aggregate of several phases and functional groups as indicated by pyrolytic and Nuclear Magnetic Resonance (NMR) studies (e.g., Sephton et al. 1998; Cody et al. 2002). We showed in previous work that the IOM, while considered resistant to chemical treatment, could release a large suite of solvent-soluble compounds by hydrothermal treatment (HT), i.e., conditions similar to those of terrestrial hydrothermal vents (300°C, 100 MPa) (Yabuta et al. 2007). More recent experiments revealed that several IOMs from meteorites belonging to diverse CC classes also hydrothermally release abundant ammonia (NH${}_{3}),$ which represents in some instances the largest molecular component of the HT extracts, with ¹⁵N enrichments of +45‰ to +455‰. In these studies, the detection of ammonia by Gas Chromatography-Mass Spectrometry (GC-MS; Pizzarello & Williams 2012) and identification of its individual ¹⁵N enrichment by Gas Chromatography-Combustion-Isotope Ratio Mass Spectrometry (GC-C-IRMS; Pizzarello et al. 2011), i.e., by compound specific analytical techniques, excluded interferences by any possible terrestrial contaminants or other co-eluting water-soluble compounds (see below for possible exceptions in regards to CI meteorites in this study).

The findings were surprising in view of the insoluble composition of IOM and the harsh chemical processing involved in isolating the material from the minerals (Cronin et al. 1987), which would easily eliminate any surficial ammonia or ammonia-containing compounds. Even more unexpected, however, was discovering that the ¹⁵N isotopic enrichments of the IOM-released ammonia related closely with the classification of the meteorites (Pizzarello & Williams 2012), i.e., with the mineralogical make-up and alteration history of the stones. Because the mineralogical composition of CC meteorites is considered to be largely the result of pre- and post-accretional processes affecting solar aggregates, their precursors and/or asteroidal parent bodies (e.g., Clayton & Mayeda 1984), these data inferred that such Solar processes were also involved in IOM ammonia's formation/acquisition as well as of its isotopic enrichments.

Aiming to better understand these data, we have analyzed samples of IOM powders from the same meteorite types used in the Pizzarello & Williams (2012) study, both per se and after HT, and measured the extent of ¹⁵N fractionation in ¹⁵N-rich "hotspots" to evaluate their possible isotopic relation to IOM-released ammonia. Although the analytical comparison involved IOM domains of different scales, it was allowed by the blending effect of the demineralization process, which involved protracted acid extractions, repeat rinses, multiple solvents and sonications to yield just few mg g⁻¹ of near to homogeneous IOM powders (see materials and methods). Owing to the heterogeneous distribution of the N-anomalous hotspots in meteoritic IOMs, as seen by several authors (e.g., Briani et al. 2009; Bose et al. 2014), we measured equivalent sizes and several different areas for all samples in this study (Table 1), except for Bells. Analyses of the "hotspots" were performed by nanoscale Secondary Ion Mass Spectrometry (NanoSIMS) on comparatively (Messenger 2000; Busemann et al. 2006) large areas to statistically evaluate the data. Since "hotspots," in spite of their name, have broad fractionation ranges up to terrestrial values, we chose for our study anomalies at or above 3σ from the terrestrial range.

All meteorite powders were extracted sequentially with water, dichloromethane/methanol (9:1, v:v), carbonyl disulfide, and 6NHCl, with drying after each step, and then demineralized as previously described (Cronin et al. 1987). Briefly, powders were treated with 30 ml of 8N HF/3N HCl solution in water, in Teflon containers, at $\leqslant 38$°C for five days with stirring and then repeatedly rinsed with 6N HCl, 3NHCl, and water with intermittent sonications until reaching a stable weight upon drying. The materials were then further extracted in sequence with water and MeOH at 100°C for 2hrs. The meteorites used for the experiments and their weight % recoveries versus starting powders were: Orgueil: 8.0; Ivuna: 4.5; Murchison: 2.7; Bells: 3.0; Tagish Lake: 3.9; GRA 95229: 2.6; Allende: 0.3; Sutters' Mill: 1.8.

2.1. Hydrothermal Experiments

2.2. NanoSIMS Analyses

2.3. Materials

Data are listed in Table 1 and summarized in Figure 1; they present a quantitative overview of the ¹⁵N isotopic enrichments detected before and after HT treatment from equivalent areas of individual IOMs, with their standard deviations and in comparison to those of their IOM-released ammonia and bulk IOM¹⁵N (Alexander et al. 2007). Figure 2 gives a sample NanoSIMS image collected in the study.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

As shown, ¹⁵N anomalies are found to be ubiquitous in all IOMs analyzed, except for the Allende and Sutter's Mill meteorites, whose IOMs did not release ammonia upon HT (e.g., Pizzarello & Williams 2012; Pizzarello et al. 2013), for which no anomalies were found and whose data were taken as baseline reference. The ¹⁵N anomalies are reduced in number after HT to an extent that is compatible with their release of ammonia.

The correlation between IOMs ${\delta }^{15}$N‰ reduction upon HT and IOM-released ammonia is given in Figure 3(a), it shows a trend of the data that is consistent with our proposal with the exception of Ivuna's. The remaining variation from linearity, e.g., for GRA, can be reasonably attributed to the small IOM amounts involved in the analytical technique, the inherent difference in samples' areas between repeat analyses and, as referenced above, the known heterogeneity of the IOM.

Zoom In Zoom Out Reset image size

Ivuna IOM data have instead clearly different values and, even if showing hotspots losses upon HT, gave larger ${\delta }^{15}$N-reductions compared to released ammonia than other meteorites. We measured IOM and HT IOM area (2500 μm²) of Orgueil, a meteorite of the same type as Ivuna, and detected an even larger ¹⁵N-reduction, 1075 (not listed because the smaller quantities available for these IOMs prevented repeat analyses). Both meteorites are classified as CI, a rare type of stones, of which only four falls are known and one specimen (Alais, fallen in 1806) is no longer available (Sears & Dodd 1988). CIs show in their predominant clay composition and absence of chondrules, a pervasive asteroidal alteration by water, and have been at times likened to comets (e.g., Lodders & Osborne 1999). The deviation from the trend displayed in 3a by both CIs suggests that their protracted parent body alteration might have affected bondings within their IOMs and led to a larger release of organic compounds besides ammonia upon HT.

In fact, the search of solvent extracts of hydrothermally treated powders analyzed previously (Pizzarello & Williams 2012) revealed a suite of numerous branched (methyl-, dimethyl-, ethyl-) pyridines in both CIs and Murchison, but not in other types of meteorites. Some of these N-containing one-ring hydrocarbons, already miscible in water (Merck Index, 10th ed.), could have been more easily released from the polymer upon HT and collected in the ammonia containing water phase. The isotopic composition of the pyridines was not assessed at the time but, were they also enriched in ¹⁵N, their HT release would provide a reasonable explanation for the higher overall ¹⁵N value found for CI hotspots in this study.

Figure 3(b) shows the plot of hotspots in ${\delta }^{15}{\rm{N}}$ depleted H-IOM versus the ¹⁸O compositions of the stones and is linear. CC display a wide range of oxygen isotopic composition between types as a consequence of the interaction of primordial isotopic reservoirs in the solar nebula, subsequent asteroidal processes (Clayton & Mayeda 1999), characterize their minerals and also define their classification or absence of it, such as for TL and Bells. That the IOM ¹⁵N composition may correlate so clearly with the ${\delta }^{18}{\rm{O}}$ ratios of the meteorites' matrices after the loss of ammonia appears to point to a later acquisition stage for the molecule within the pre-accretionary history of their parent bodies.

The data in Table 1 fall within the broad ranges of ¹⁵N anomalies reported for solar materials in previous studies, e.g., for a regolith in the metal-rich chondrite Isheyevo that had astounding variations (−310 to +4900‰; Briani et al. 2009), for IDPs from the comet 26P/Grigg–Skjellerup dust stream (+70 to +485‰, for the bulk and +410 to +1400‰ for hotspots; Busemann et al. 2006) and the +480 for the Dragonfly IDP (Messenger 2000).

For meteoritic IOM in particular, its bulk ¹⁵N composition was determined for several meteorites by Alexander et al. (2007) and the pertinent values from that study are listed in Table 1. IOM ¹⁵N anomalies were also determined by Busemann et al. (2006) for a group of meteorites similar to the ones we studied. The study provided only the single largest average values of the anomalies for each IOM and they were found to wary from +1740 to +2000‰ for three CR2 meteorites, to +410‰ for Tagish Lake and +3200‰ for Bells. Also, a more recent attempt to characterize isolated ¹⁵N-rich hotspots for the CO QUE 97416 (an Ornans-type, a thermally altered class of meteorites) found nitriles to be the carrier phase of the anomaly (Bose et al. 2014).

Based on the knowledge and hypotheses available to date on the chemical evolution of biogenic elements, their molecules identified outside the Earth and the mineralogy present in early solar system bodies such as asteroids, three main features of our results cannot be accounted for directly and are in need of further explanation: (1) the presence of freeable ammonia in insoluble materials; (2) the ¹⁵N enrichment of the IOM ammonia; and (3) the dependence of IOM ammonia ¹⁵N enrichments with the mineralogy of the stones. We argue that the following sequence and/or progression of events relying on plausible solar processes would account for our results.

As presently understood (e.g., Levy 1985), the protosolar nebula formed from a collapsing molecular cloud in the local ISM and formed a nebula with a star at its center, the Sun, while left-over matter containing dust and molecules of the cloud reached different and farther temperature zones in a spinning disk, all the while clumping, condensing and being subjected to sometimes intense radiation, e.g., during the T-Tauri phase (Elmegreen 1979). ¹⁵N-enriched ammonia formed during these stages by vacuum UV irradiation of N₂ and H₂ condensed mixtures. Macromolecular assemblages involving formaldehyde polymerization, both inherited from the cloud and newly formed in cooler regions of the protosolar or solar nebula, readily incorporated the ¹⁵N-enriched ammonia in concentrated "hotspots," adsorbed upon and/or intermixed with the mineral precursors present in those regions and was eventually incorporated by assembling small bodies such as comets and asteroids.

Later the asteroids, whose material was left unable to coalesce into a planet under the gravitational tag of earlier-formed giant planets, incorporated ¹⁵N-enriched ammonia, both bound and free, and used it after developing a hydrous environment for the formation of ¹⁵N-enriched compounds such as amino acids via known reactions (e.g., Pizzarello & Shock 2010). Comets have long been known to contain ammonia (Wyckoff et al. 1991) and recent measurements revealed ${\delta }^{15}$N‰ up to ∼+1000, with the authors implying solar nebular fractionation effects to justify the high value (Rousselot et al. 2014; Shinnaka et al. 2014). Gaseous planets such as Titan, where ammonia has an estimated ${\delta }^{15}$N‰ $\geqslant$ +397, may have incorporated this solar ammonia as well (Mandt et al. 2014).

Recent findings support this scenario. The isotope selective photodissociation of N₂ by UV irradiation was assessed theoretically (Heays et al. 2014) and ¹⁵N enrichments of ∼12,000‰ were produced experimentally by 90 nm UV radiation of an N₂ H₂ mix (Chakraborty et al. 2014). As the authors note in the latter reference, the photoionization of N₂, H₂, and H are 14.534, 15.42, and 13.6 eV, respectively, i.e., greater than the energy used in their experiments, but they point out that the ionization energy of NH₃ is 10.2 eV and may lead to photoionization at longer wavelengths, such as those used in these experiments, as well as introduce additional isotope effects. We would assume that the environment of the early nebula, far more complex than experimental conditions and thought to be highly heterogeneous, could have intervened significantly in these reactions. The protosolar disk could contain mixtures of simple molecules such as H₂O ice, CH₄, CO₂, OH, or HCN (e.g., Carr & Najita 2008), which, if exposed to stellar UV photons or cosmic rays breaking atomic bonds, would produce reactive ions and radicals within the ice-rocky matrix. Subsequent to N₂ photodissociation, fractionation caused by exchange reactions between atomic ¹⁵N and other N-containing ions would form several intermediate ¹⁵N-enriched ionization products in the solar nebula: ${{\rm{N}}}^{+},$ NH${}^{+},$ NH${}_{2}^{+},$ NH${}_{3}^{+},$ NH${}_{4}^{+}.$ Photochemical self-shielding effects, similar to those observed for O isotopes in CCs' mineral component (Clayton 2002), could also have been responsible for producing N-anomalies in the protosolar nebula and produce ¹⁵N-rich ammonia in the presence of appropriate ice-rock mixtures. The temperature in the disk can be as high as 1000 K (Chiang & Goldreich 1997), depending on the proximity to the star and the stage of the star (e.g., the T-Tauri stage), which may be sufficient to help initiate reactions such as the formation of NH₂ from NH (NIST data base), however, the majority of other reactions to produce ammonia require low temperatures of ∼10 K temperatures, typical in the outer parts of the solar system or midplane of the protosolar disk.

As to the incorporation of ammonia into a developing and/or developed formaldehyde-type polymer, experimental polymerization reactions of formaldehyde found ammonia addition to be indispensable to (Schutte et al. 1993) or at least increase its rate of polymerization (Cody et al. 2011). On the basis of more recent data by Kebukawa et al. (2013), who followed the formose polymerization in the laboratory under diverse conditions and detailed its composition by several spectroscopic techniques, the many functional groups predicted throughout such polymer, e.g., carbonyls (R-CO-R'), would have made ammonia addition through its lone pair of electrons ($:{\mathrm{NH}}_{3})$ not only possible but abundant (Breslow 1959; Appayee & Breslow 2014).

The stability of IOM-bound ammonia during the extensive washes and rinses of the demineralization procedures can also be accounted for by the above experimental findings. For example, the Cody group conducted N-XANES spectroscopy of a formaldehyde polymer and detected nitriles when the polymer was synthesized in the presence of ammonia (Kebukawa et al. 2013). Nitriles will hydrolyze during acid washes but the product carboxamides would be stabilized by CO⁻=NH⁺${}^{-}$ ${}^{}$resonance, survive demineralization procedures, and release ammonia upon HT; formamide in particular comes to mind (Harada 1967).

Albeit more tentatively, we could also assume that ammonia retention could have benefited from the very complexity of the IOM assemblages. Many functional groups are observed in the IOM by NMR, such as carboxyls (COOH; Cody et al. 2002), and model experiments of formaldehyde polymerization have shown disordered sequences of phenols and units with phenolic rings (Pizzarello et al. 2013); all of these O-containing, highly branched and interlocked assemblages might have helped caging amide groups though hydrogen bonding at the very low pH conditions for isolating the IOM but allowed the release of ammonia upon HT.

The experimental data presented here offer new insights on the possible distributions and molecular evolution of volatile N-containing compounds in the Solar System. The hypothesis we propose also offers a reasonable explanation for the presence of highly ¹⁵N-enriched molecules in the IOM, a materials inherited from the ISM, as well as for the chance of these molecules and materials to associate with mineral phases in different zones of the nebula. The interesting isotopic coherence discovered for two CI IOMs also points to the desirability of similar analyses for several meteorites of the same class, as we are planning.

Nitrogen is, after H and the noble gases, one of the most abundant elements in cosmic environment, the Solar System and, on the Earth, a major constituent of the biosphere. The possibility that CC meteorites seeded the Earth with prebiotic precursors has often been offered, however, the caveat raised for these proposals has been the lack of explanation for how an heterogeneous and random mix of abiotic compounds, in the thousands (Schmitt-Kopplin et al. 2010) and minute amounts (μg g⁻¹ at best; Pizzarello & Shock 2010), could have worked their way into the exacting selectivity of even the simpler biochemistry.

That ammonia could be released from the IOMs of meteorites in μg mg⁻¹ (Pizzarello et al. 2011) and the findings that such simple but essential prebiotic molecule can form abundantly in solar environments, be incorporated in asteroids and delivered to early Earth environments, open a new understanding of the prebiotic molecular inventory in our early planet as well as inform our estimates for the possible habitability of other planetary systems.

Funding for this work was provided by the NASA Exobiology and Astrobiology Program, with grant # NNX10AT81G to S.P.. M.B. thanks Dr. Peter Williams (NSF ARRA-960334) and Dr. Rick Hervig (NSF EAR-0948878 & NSF EAR-1352996) for supporting her during this work. S.P. is indebted to Ana L. Moore for fruitful discussions and the ASU Center for Meteorite Studies for providing the indispensable distilled HF solutions. Both authors are sincerely thankful to an anonymous reviewer for a prompt and careful revision.