Customized injection techniques for compound-specific isotope analysis of natural gas samples

(1)

HAL Id: hal-01229069

https://hal.archives-ouvertes.fr/hal-01229069

Submitted on 1 Dec 2015

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

Customized injection techniques for compound-specific

isotope analysis of natural gas samples

Michaela Blessing, Eric Proust, Christine Fléhoc

To cite this version:

Michaela Blessing, Eric Proust, Christine Fléhoc. Customized injection techniques for

compound-specific isotope analysis of natural gas samples. Procedia Earth and Planetary Science, Elsevier, 2015,

13, pp.227-231. �10.1016/j.proeps.2015.07.054�. �hal-01229069�

(2)

Procedia Earth and Planetary Science 13 ( 2015 ) 227 – 231

Peer-review under responsibility of the scientific committee of AIG-11 doi: 10.1016/j.proeps.2015.07.054

ScienceDirect

11th Applied Isotope Geochemistry Conference, AIG-11 BRGM

Customized injection techniques for compound-specific isotope

analysis of natural gas samples

Blessing M.*, Proust E., Fléhoc C.

BRGM, LAB-ISO, BP36009, 45060 Orléans cédex 2, France

Abstract

The aim of this study was to develop an easy-to-use, reliable method for measuring the isotopic composition of individual alkanes present in natural gases at concentrations below 1000 ppmv without the need of expensive or complex pre-concentration devices. The proposed method combines pre-concentration of the gaseous compounds (C1 to C5) in a cooled, programmable

temperature injector equipped with a Carbosieve-packed liner followed by compound-specific isotope analysis (CSIA) by gas chromatography – isotope ratio mass spectrometry (GC-IRMS). This method requires only small sample volumes of up to 10 mL and allows reliable isotopic measurements at low concentrations (< 100 ppmv) within the typical precision of GC-IRMS: ≤ 0.5‰ for carbon and ≤ 5‰ and hydrogen isotope analysis, respectively. The proposed injection techniques have been developed and validated using gaseous hydrocarbon samples of different origins.

Peer-review under responsibility of the scientific committee of AIG-11.

Keywords: PTV-injection; natural gas samples; C1-C5 alkanes; isotope analysis; CSIA

1. Introduction

Compound-specific isotope analysis (CSIA) is one of the key monitoring techniques to assess the origin and fate of organic contaminants in soil and groundwater1_{and to study the source and genesis of naturally occurring gases}2_.

Studying the latter has become important in the course of recognizing coalbed methane (CBM) and shale gas reservoirs as important energy resources. Isotopic ratios of hydrocarbons and CO2 can be helpful for determining a)

* Corresponding author. Tel.: +33-23864-3163. E-mail address: m.blessing@brgm.fr

(3)

228 M. Blessing et al. / Procedia Earth and Planetary Science 13 ( 2015 ) 227 – 231

the genetic origin of gases, the type of organic matter from which the gas was derived, b) the maturity level, and c) post-genetic processes such as migration and whether the gas has been subject to alteration3,4_{. Chemical and isotopic}

fingerprinting of groundwater and natural gases may also allow assessing the effect of unconventional gas production on the environment, such as stray gas migration into shallow aquifers5,6_.

When released to the atmosphere, methane (CH4) acts as one of the most prevalent greenhouse gases. Not only

natural gas and petroleum systems but also wetlands and landfills are important sources of atmospheric CH4. Natural

processes in soil such as aerobic bacterial CH4 oxidation can reduce these emissions. In this context, stable isotope

methods can be applied for correlating surface emissions with potential gas sources and for understanding the gas migration pathways and alterations, for example to deliver a conservative estimate of methane oxidation in landfill cover soils7,8_.

Application of compound-specific isotope analysis to natural samples is often limited to relatively high concentrations owing to the fact that precise isotopic measurements by continuous-flow gas chromatography – isotope ratio mass spectrometry (GC-IRMS) require at least 1 nmol of carbon, or 8 nmol of hydrogen, of each individual compound on the GC-column9. Efficient pre-concentration techniques for gaseous hydrocarbons are either costly or quite complex10,11,12,13_{. In this study, we aimed at developing an easy-to-use, reliable method that allows}

measuring the isotopic composition of natural gas alkanes at concentrations below 1000ppmv without the need of expensive or complex pre-concentration devices.

2. Experimental

2.1. Instrumentation

Carbon and hydrogen isotopic compositions of individual compounds were determined by gas chromatography - isotope ratio mass spectrometry (GC-IRMS) consisting of a Trace GC ultra connected to a DeltaplusXP IRMS via a GC Combustion III interface (Thermo Finnigan). For carbon, the interface unit was equipped with a CuO, NiO and Pt containing oxidation tube, maintained at 940°C. For hydrogen the interface used was an empty ceramic tube to allow for the reduction of the analytes to H2 at 1440°C. Chromatographic peak separation was performed on a Pora

PLOT Q GC-column (25m x 0.32mm i.d., 10μm film, Agilent Technologies) run isothermally at 30°C for landfill samples containing only CH4 and CO2. For higher alkane containing samples, such as natural gas samples (CH4,

CO2, C2H6 to C6H14), the chromatographic separation of compounds was achieved by using a temperature program

that started at 33°C (6 min hold), and heated with a rate of 15°C/min to 200°C for an additional 6 min.

Results for CO2 obtained by GC-IRMS were compared to carbon isotope analyses of CO2 independently

performed by Gasbench-IRMS. The system consisted of a gas chromatography-based GasBench II system coupled to a DeltaplusXP IRMS (Thermo Finnigan). Sample injection was performed automatically using a CTC Pal autosampler. Three injections of a pure CO2 reference gas were followed by seven subsequent injections of sample

gas. Gas separation was achieved using a PoraPLOT Q GC-column run at 40°C. Injection times and open split impulses were chosen so as to avoid disturbing the CO2 signal by the N2/O2/Ar-peak that precedes each CO2 peak

(in samples with low CO2 concentration).

Isotopic ratios are expressed in the delta notation (G13_{C, GD), determined relative to a CO}

2 or H2 reference gas,

respectively. These reference gases have been calibrated against defined isotope ratios of the VPDB and VSMOW international standards by a Finnigan MAT 252 dual-inlet IRMS. Applied internal standards were calibrated against NBS-19, VSMOW and SLAP; values determined for our reference gases were δ13_C

PDB -40.9‰ ±0.3‰, and

GDVSMOW -495‰ ±2‰.

2.2. Injection parameters

Gas samples collected in glass ampoules were transferred into He-flushed 12 mL screw-capped Labco vials in a vacuum line (analysis on the same day to avoid storage effects), all other samples could be directly sampled using a gas-tight syringe. The injection system consisted of a programmable temperature vaporizer (PTV) injector (Optic 3, ATAS GL) equipped with a fritted split/splitless-inlet liner for gaseous samples at 2-400μL injection volumes. For low concentrations of gaseous hydrocarbons (< 250 ppmv), the PTV-injector was equipped with a

(4)

Carbosieve-packed fritted inlet liner (obtained from PAS Technology) that allowed injecting larger gas volumes (1-10mL). Injections were performed manually, while injections into the empty liner could be performed rapidly (at-once) and at +40°C, injections using the Carbosieve-packed liner required slow injection at a PTV inlet temperature of -40°C to retain the gas on the liner packing. Cooling of the PTV injector was achieved using a PTV inlet cooling device based on cold gas (N2) produced by a LN2 heat exchanger.

2.3. Gas samples used for testing

x Technical gas mixtures (CH4 to C6H14) at 1000 ppmv and 100 ppmv (Restek/Air Liquide), delivered in SCOTTY

cylinders equipped with a gas pressure regulator with integrated syringe adapter;

x biogas and cover soil gas from a landfill located in Sonzay (France) sampled in aluminium bags from GL Sciences equipped with a septum-sealed port for direct sampling with a syringe;

x coalbed methane from the Monsacro coal mine (Spain) sampled in glass ampoules;

x gas samples extracted from borehole crosscuttings of different Jurassic clay formations, obtained from previously performed degassing studies on core clay samples of the Upper and Lower Toarcian Clay Formation

(Tournemire, France) and the Opalinus Clay Formation (Mont Terri and Schlattingen, Switzerland)14_.

3. Results and discussion

3.1. Technical gas mixtures

The precision of the G13_{C measurements of C}

1- to C6-alkanes was evaluated by three replicate injections of the

technical gas mixtures. To evaluate the newly proposed PTV-injection technique, various amounts (1 to 10 mL) of the 100 ppmv gas mixture were injected on a Carbosieve-packed liner at -40°C. Standard deviations (SD) are within the typical range for continuous flow-IRMS measurements (≤ 0.5‰). Fig. 1 demonstrates the pre-concentration effect of the here proposed injection method using Carbosieve-packed liners.

Fig. 1. G13_{C measurements of C}

1 to C6-alkanes (100 ppmv each) by GC-IRMS using (a) the conventional technique with a maximum possible

(5)

230 M. Blessing et al. / Procedia Earth and Planetary Science 13 ( 2015 ) 227 – 231

The limit for carbon isotope ratio determination (MDL) was at 30 ppmv for CH4, 15 ppmv for C2H6, and 10

ppmv for C3- to C5-alkanes determined at an injection volume of 10 mL. MDL were evaluated by a linearity test,

following the recommendations given in Jochmann et al.14_{and correspond to a peak height of ≥ 400 mV of the m/z}

44 signal. The newly proposed PTV-injection technique does not allow for isotopic analysis of C6H14-isomers due to

peak splitting. Observed determination limits are valid for direct sampling techniques; samples that need to be transferred on a vacuum line require 2- to 3-times higher concentrations (pressure-dependent). Table 1 shows the accuracy and precision of the proposed PTV-injection technique for GD-measurements compared to the conventional technique by injecting a methane gas standard at different dilution ratios at different volumes.

Table 1. GD-analysis of CH4: Comparison of the conventional and the Carbosieve-injection method (n=4).

CH4 concentration, amount injected, technique applied GD CH4 (‰) SD (‰)

4 vol% (40000 ppmv), 40 μL, conventional injection -177 ±1.8 0.4 vol% (4000ppmv), 400 μL, conventional injection -176 ±1.8 0.02 vol% (200 ppmv), 10 mL, Carbosieve-injection -177 ±1.6

3.2. Natural gas samples

The conventional injection as well as the newly proposed injection technique on a cooled Carbosieve-packed liner have been applied to perform GD and G13

C-analyses of samples containing CH4 and CO2 from a landfill site;

native biogas samples as well as gaseous samples taken at different depths within the 1 m-thick sandy-clay cover have been included. The isotopic values determined for the biogas samples (G13_C

CH4 -53‰ to -60‰; GDCH4 -276‰

to -329‰)are in good agreement with values that have been reported for other European landfills7_{. The}

Carbosieve-method had then been applied to landfill samples with low CH4 concentrations (< 200 ppmv; GDCH4 316‰ to

-324‰). Results for G13_C

CO2 obtained by GC-IRMS had been compared to values of the same samples measured by

GasBench-IRMS and were in perfect agreement; CO2-concentrations included in this test ranged from 350 ppmv to

33 vol%.

The coalbed methane sample of the Monsacro coal mine was used to assess the practical applicability of the new technique over a wide range of different concentrations in a single natural gas sample. The injected volume had to be adjusted for each compound present in this sample to obtain peak heights within the linear range of the IRMS. For compounds with high concentrations within the same sample, the He-backflush function of the interface was used to prevent the compound from entering into the IRMS source (see Table 2 for details).

Table 2. G13_{C-value and height of the m/z-44 peak of C}

1- to C5-alkanes analyzed in a single coalbed methane sample.

Volume injected CH4 [72.4 vol%] C2H6 [2.6 vol%] C3H8 [0.2 vol%] n-C4H10[229ppmv] n-C5H12 [12 ppmv]

3 μL, conventional -48.6‰ ,4860mV <l.d., 225mV n.d. n.d. n.d. 30 μL, conventional BF -27.2‰, 2040mV <l.d., 210mV n.d. n.d. 400 μL, conventional BF BF -23.8‰, 2380mV <l.d., 135mV n.d. 10 mL, Carbosieve BF BF BF -22.1‰, 485mV <l.d., 30mV [concentration given in brackets]; BF = He-backflush; n.d. = not detected; <l.d. = below limit of isotope ratio determination

This example demonstrates that compound-specific isotope analysis of natural gas samples requires the application of different injection techniques which need to be adapted to the concentration of each hydrocarbon present within a single sample. This was also the case for the samples extracted from the borehole crosscuttings of different Jurassic clay formations; the G13

C-values obtained for these samples are presented and discussed in Lerouge et al.15_.

(6)

Acknowledgements

This study received financial support of the research division (DR) of the BRGM. G. Bellenfant, A. Leynet and C. Lerouge are thanked for providing natural gas samples.

References

1. Negrel P, Blessing M, Millot R, Petelet-Giraud E, Innocent C. Isotopic methods give clues about the origins of trace metals and organic pollutants in the environment. Trends Anal Chem 2012; 38 143-153.

2. Giustini F, Blessing M, Brilli M, Lombardi S, Voltattorni N, Widory D. Determining the origin of carbon dioxide and methane in the gaseous emissions of the San Vittorino plain (Central Italy) by means of stable isotopes and noble gas analysis. Appl Geochem 2013; 34 90-101. 3. Baylis SA, Hall K, Jumeau EJ. The analysis of the C1-C5 components of natural-gas samples using gas-chromatography combustion isotope

ratio mass-spectrometry. Org Geochem 1994; 21 777-785.

4. Tilley B, Muehlenbachs K. Gas maturity and alteration systematics across the Western Canada Sedimentary Basin from four mud gas isotope depth profiles. Org Geochem 2006; 37 1857-1868.

5. Rowe D, Muehlenbachs K. Isotopic fingerprints of shallow gases in the Western Canadian sedimentary basin: tools for remediation of leaking heavy oil wells. Org Geochem 1999; 30 861-871.

6. Jackson RB, Vengosh A, Darrah TH, Warner NR, Down A, Poreda RJ, Osborn SG, Zhao K, Karr JD. Increased stray gas abundance in a subset of drinking water wells near Marcellus shale gas extraction. Proc Natl Acad Sci USA 2013; 110 11250-11255.

7. Bergamaschi P, Lubina C, Konigstedt R, Fischer H, Veltkamp AC, Zwaagstra O. Stable isotopic signatures (delta C-13, delta D) of methane from European landfill sites. J Geophys Res 1998; 103 8251-8265.

8. Liptay K, Chanton J, Czepiel P, Mosher B. Use of stable isotopes to determine methane oxidation in landfill cover soils. J Geophys Res 1998;

103 8243-8250.

9. Schmidt TC, Zwank L, Elsner M, Berg M, Meckenstock RU, Haderlein SB. Compound-specific stable isotope analysis of organic contaminants in natural environments: a critical review of the state of the art, prospects, and future challenges. Anal Bioanal Chem 2004; 378 283-300.

10. Rice AL, Gotoh AA, Ajie HO, Tyler SC. High-precision continuous-flow measurement of delta C-13 and delta D of atmospheric CH4. Anal Chem 2001; 73 4104-4110.

11. Schaefer H, Whiticar MJ. Measurement of stable carbon isotope ratios of methane in ice samples. Org Geochem 2007; 38 216-226. 12. Huiban Y, Noirez S, Prinzhofer A, Girard JP, Chappellaz J. Chemical and isotopic analysis of hydrocarbon gas at trace levels Methodology

and results. Chem Geol 2009; 265 363-368.

13. Yamada K, Ozaki Y, Nakagawa F, Tanaka M, Yoshida N. An improved method for measurement of the hydrogen isotope ratio of atmospheric methane and its application to a Japanese urban atmosphere. Atmos Environ 2003; 37 1975-1982.

14. Jochmann MA, Blessing M, Haderlein SB, Schmidt TC. A new approach to determine method detection limits for compound-specific isotope analysis of volatile organic compounds. Rapid Commun Mass Spectrom 2006; 20 3639-3648.

15. Lerouge C, Blessing M, Fléhoc C, Gaucher EC, Henry B, Lassin A, Marty N, Matray JM, Proust E, Rufer D, Tremosa J, Vinsot A. Dissolved CO2 and alkane gas in clay formations. Procedia Earth Planet Sci 2015; this volume.