[MTT51-08] Examination of analytical accuracy of GC/IRMS: Nitrogen isotopic composition of 15N-enriched amino acids
キーワード：窒素安定同位体比、アミノ酸、地球外物質
Nitrogen consists of two stable isotopes (14N: 99.63% and 15N: 0.37%). The ratio of 15N/14N is expressed as δ15N notation in per mill (‰) relative to the standard ratio in terrestrial atmospheric air (Air). The variation of δ15N in the terrestrial materials generally falls within ±20‰, but even such little variations tell us a lot of information for studying environmental and physicochemical processes that the molecules has experienced. Unlike the terrestrial materials, extraterrestrial materials, especially pristine solar system materials such as comets and chondrites, exhibit extremely high δ15N values more than +1000‰ [e.g., 1], and even +5000‰ as a nanometer scale materials (which is called a hot spot [e.g., 2]).
However, there are problematic issues on the analytical accuracy in calculating the δ15N values of such 15N-enriched materials, because the conventional δ15N analysis has been developed for measuring the ratio in terrestrial materials and focusing very little amount of 15N (~0.4%) in them. For the δ15N analysis by gas chromatography / isotope-ratio mass spectrometry (GC/IRMS), the nitrogen in the organic materials are converted to N2. There are three combinations of the nitrogen isotopes: 14N≡14N (m/z 28), 15N≡14N (m/z 29), 15N≡15N (m/z 30). Among these three combinations, it is assumed that the existence of 15N≡15N (m/z 30) are negligible, because the amount of 15N on the Earth is very small against that of 14N. On the other hand, the applicability of this assumption has not been examined to the isotope analysis of 15N-enriched materials. If 15N≡15N are produced significantly in the analytical procedure, the δ15N calculation will include a large uncertainty.
To examine the accuracy in the δ15N analysis for 15N-enriched materials, we evaluate the GC/IRMS results for several 15N-labelled amino acid samples. The samples were prepared by mixing 15N-labelled and non-labeled amino acids, resulting in a variation in the δ15N value from 0‰ to +1000‰ based on the mathematical calculation from the mixing ratios. The amino acid samples were then derivatized and analyzed by GC/IRMS, according to the standard procedure [3]. The results reveals that the measured δ15N values certainly deviate from the calculated δ15N values, with a linear relationship between the δ15N values and the deviation: the deviation is close to zero for the 0‰ samples but reaches ~ +90‰ for the +1000‰ sample accompanied by m/z 30 signals on the GC extracted ion chromatograms. This deviation would be critical to understand the physicochemical processes occurring in the extraterrestrial environments. These results demonstrate the importance of the inclusion of the m/z 30 signal area for the calculation of the δ15N values of 15N-enriched materials. Further detailed studies including memory effects [4] are necessary to investigate the factors on potential inaccuracy to form 15N≡15N during GC/IRMS analysis.
References: [1] Furi & Marty (2015) Nature Geosci. 8, 515-522. [2] Briani et al. (2009) PNAS 106, 105222-10527. [3] Chikaraishi et al. (2010) in Earth, Life, Isotopes, Kyoto Univ. Press, pp.367-386. [4] Petzke & Metges (2012) Rapid Commun. Mass Spectrom., 26, 195-204.
However, there are problematic issues on the analytical accuracy in calculating the δ15N values of such 15N-enriched materials, because the conventional δ15N analysis has been developed for measuring the ratio in terrestrial materials and focusing very little amount of 15N (~0.4%) in them. For the δ15N analysis by gas chromatography / isotope-ratio mass spectrometry (GC/IRMS), the nitrogen in the organic materials are converted to N2. There are three combinations of the nitrogen isotopes: 14N≡14N (m/z 28), 15N≡14N (m/z 29), 15N≡15N (m/z 30). Among these three combinations, it is assumed that the existence of 15N≡15N (m/z 30) are negligible, because the amount of 15N on the Earth is very small against that of 14N. On the other hand, the applicability of this assumption has not been examined to the isotope analysis of 15N-enriched materials. If 15N≡15N are produced significantly in the analytical procedure, the δ15N calculation will include a large uncertainty.
To examine the accuracy in the δ15N analysis for 15N-enriched materials, we evaluate the GC/IRMS results for several 15N-labelled amino acid samples. The samples were prepared by mixing 15N-labelled and non-labeled amino acids, resulting in a variation in the δ15N value from 0‰ to +1000‰ based on the mathematical calculation from the mixing ratios. The amino acid samples were then derivatized and analyzed by GC/IRMS, according to the standard procedure [3]. The results reveals that the measured δ15N values certainly deviate from the calculated δ15N values, with a linear relationship between the δ15N values and the deviation: the deviation is close to zero for the 0‰ samples but reaches ~ +90‰ for the +1000‰ sample accompanied by m/z 30 signals on the GC extracted ion chromatograms. This deviation would be critical to understand the physicochemical processes occurring in the extraterrestrial environments. These results demonstrate the importance of the inclusion of the m/z 30 signal area for the calculation of the δ15N values of 15N-enriched materials. Further detailed studies including memory effects [4] are necessary to investigate the factors on potential inaccuracy to form 15N≡15N during GC/IRMS analysis.
References: [1] Furi & Marty (2015) Nature Geosci. 8, 515-522. [2] Briani et al. (2009) PNAS 106, 105222-10527. [3] Chikaraishi et al. (2010) in Earth, Life, Isotopes, Kyoto Univ. Press, pp.367-386. [4] Petzke & Metges (2012) Rapid Commun. Mass Spectrom., 26, 195-204.