Subscriber access provided by UNIV OF REGINA
Article
Compound-specific sulfur isotope analysis of petroleum gases Ward Said-Ahmad, Kenneth Wong, Monaca Mcnall, Lubna Shawar, Tracey Jacksier, Courtney Turich, Artur Stankiewicz, and Alon Amrani Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b05131 • Publication Date (Web): 27 Jan 2017 Downloaded from http://pubs.acs.org on January 28, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Compound-specific sulfur isotope analysis of petroleum gases Ward Said-Ahmad1, Kenneth Wong2, Monaca Mcnall2, Lubna Shawar1, Tracey Jacksier2, Courtney Turich3, Artur Stankiewicz4, Alon Amrani1* 1
Earth Sciences Institute, Hebrew University, Jerusalem, Israel, *
[email protected] ,phone: +972 54 5502249 (*Corresponding author) 2
Air Liquide, Delaware Research & Technology Center, Newark, DE, USA
3
Schlumberger, 1 Rue Henri Becquerel, Clamart, France 92140
4
Schlumberger, 5599 San Felipe, Houston, TX 77056, USA
We describe a simple, sensitive, and robust method for sulfur isotope ratio (34S/32S) analysis of ppm-level organic sulfur compounds (OSCs) in the presence of percent-level H2S. The method uses a gas chromatograph (GC) coupled with a multicollector inductively coupled plasma mass spectrometer (MC-ICPMS). The GC, equipped with a gas inlet and a valve that transfers the H2S to a thermal conductivity detector (TCD), enables a precise heart cut and prevents the saturation of the MC-ICPMS. The sensitivity and accuracy of the method are better than 0.3‰ for OSCs at a concentration of 25 pmol or 1.4 ppm, and better than 0.5‰ for concentrations ≥ 0.7 ppm OSCs. An order of magnitude increase in sensitivity, with no effect on accuracy, can be achieved if the loop volume (0.5 ml) is changed to 5 ml. High concentrations of methane (95% v/v) and/or H2S ( 20% v/v) had no effect (within 0.5‰) on the precision and accuracy of the gas sample containing 2 ppm OSCs after heart cut. The applicability and robustness of this method is demonstrated on a gas sample (10% v/v H2S) that was produced by pyrolysis of sulfur-rich kerogen. The results show good precision and reveal sulfur isotope variability between individual OSCs that may represent key processes during formation and degradation of OSCs.
INTRODUCTION Sulfur isotope ratio (34S/32S) analysis of trace sulfur compounds in natural gas is potentially useful in recognizing reservoir processes that impact the quality and production of reservoired petroleum, such as thermochemical or microbial sulfate reduction (TSR, MSR) and thermal cracking of source rock or petroleum. Other possible applications include the tracking of atmospheric gases (e.g., carbonyl sulfide) and sulfur sources during refinery processes and the study of trace sulfur gases in the ocean, wetlands, sea ice, and anoxic sediments. The analysis of ppm-level sulfur compounds, such as methanethiol and dimethyl sulfide (DMS), in natural gas is very challenging, owing to high reactivity and possible coelution with nonsulfur hydrocarbons. Measurement of OSCs at the ppm level in a sour gas matrix adds to the analytical complexity because the very high concentrations of H2S (percent level) can cause severe tailing of H2S in the GC column and sulfur saturation of the mass spectrometer. Additionally, analysis of OSCs in a sour gas matrix requires passivated lines throughout the entire analytical system.1 The conventional method for measuring sulfur isotope ratios (34S/32S) uses the combustion of sulfur compounds to SO2 or the conversion to SF6.2,3,4 However, both methods use gas-source isotope ratio mass spectrometry (GC-irMS) and require several micromoles of bulk sample as well as tedious sample preparation. Recently, Hattori et al.5 developed a technique to analyze carbonyl sulfide (COS) using GC-irMS on fragment ions 32S+, 33S+, and 34S+ at nanomole levels. However, this technique is specific for only one compound (COS) and cannot be applied to a complex mixture of organic sulfur compounds such as in natural gas. As a result, S-isotope data for gases with trace organic sulfur is scarce. The development of an S-isotope analysis technique that is based on GC coupled with MC-ICPMS has enabled compound-specific sulfur-isotopic analysis (CSSIA) at the picomole (pmol) level.6 This technique has been applied to several disciplines, from geochemistry to oceanography.7-13 Said-Ahmad and Amrani8 further developed a method for the analysis of the S-isotope ratio of DMS in seawater using the purge-and-trap system coupled to GC/MC-ICPMS, with precision (≤ 0.3‰, 1σ) and accuracy (≤ 0.2‰) for samples with sulfur concentration of 26–179 pmol.8 Therefore, the coupled GC/MCICPMS technique can provide the required selectivity and sensitivity for S-isotope analysis of gases with trace concentrations of organic sulfur.
ACS Paragon Plus Environment
1
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 12
In the present study, we have extended the capabilities of CSSIA to sour petroleum gases with high percentages of H2S and relatively low concentrations of OSCs. A modified vapor GC with passivated lines equipped with a valve to transfer the gas to a thermal conductivity detector (TCD) was used, enabling a precise H2S peak cut, with no effect on the other sulfur peaks. As a result, analysis of sulfur compounds is possible at sub-ppm levels in the presence of percent-level H2S. MATERIALS AND METHODS Reagents and standards. DMS (>99%), thiophene (99+%), carbon disulfide (CS2, anhydrous ≥99%), benzothiophene (BT, 97%), ethyl methyl sulfide (EMS, 96%), and diethyl sulfide (DES, 98%) were purchased from Sigma-Aldrich (MO, USA). Sulfur hexafluoride (SF6, 500 ppm in helium) was purchased from Praxair (PA, USA). Two gas mixtures were purchased from Air Liquide America (PA, USA); the first gas mixture (Mix 1) had OSCs with concentrations of ～100 ppm and methane as balance gas, and the second gas mixture (Mix 2) was composed of ～21 ppm OSCs and helium as balance gas (Table 1). The sulfur isotope reference materials NBS-127 (BaSO4; δ34S = 21.1‰), IAEA-S-1 (Ag2S; –0.3‰), and IAEA-SO-6 (BaSO4; –34.1‰) were purchased from the National Institute of Standards and Technology (NIST, USA) and were used for calibration of all the in-house standards. Table 1. Gas mixtures composition
Compounds
Formula/ Symbol
Mix 1 ppm
Mix 2 ppm
Hydrogen sulfide
(H2S)
100.0
20.7
Carbonyl sulfide
(COS)
101.0
20.8
Methanethiol
(MeSH)
100.0
20.8
Ethanethiol
(EtSH)
101.0
20.9
Dimethyl sulfide
(DMS)
99.0
20.9
Carbon disulfide
(CS2)
100.0
20.5
2-Propanethiol
(2-PrSH)
99.0
He-BAL
2-Methyl-2-Propanethiol
(2-Me-2-PrSH)
100.0
1-Propanethiol
(1-PrSH)
100.0
Sec-Butyl thiol
(Sec-BuSH)
100.0
Thiophene
(Thiophene)
99.0
1-Butanethiol
(1-ButSH)
99.0
Methane
(Me)
BAL
Propane
(Pr)
50,000
Isobutane
(IsoBu)
20,000
n-Butane
(Bu)
5,000
Carbon dioxide
(CO2)
2,600
Ethane
(Eth)
10,000
Instrumentation. The system employed for the S-isotope analysis of sulfur-containing gases consisted of a gas chromatograph (GC 7890, Agilent Technologies, CA, USA) coupled with a TCD or with a Neptune Plus™ MC-ICPMS (Thermo Scientific, Bremen, Germany). The GC was equipped with a split/splitless injector for direct injection of volatile samples and a heated (70 oC) six-way valve gas inlet system (Valco Instrument Co, TX, USA) for the introduction of gaseous compounds with a computer-controlled actuator (Figure 1). The GC column was then connected (in oven) to a computer-controlled Deans Switch (Agilent Technologies, CA, USA) to perform heart cutting of selected peaks. The effluent from the GC column was diverted either to the TCD or to the MCICPMS via a heated transfer line, hereafter called a GC/MC-ICPMS system (Figure 1). The heated transfer line was a laboratory-built system composed of 1.6-mm OD stainless steel tubing with Ar sample gas flow to the ICP torch around the capillary column. The transfer line was heated via flexible heating tape to ~200 °C and insulated with glass wool to prevent condensation of analytes. The analytes were carried in a He carrier gas stream from the GC to the Neptune ion source via a fused-silica capillary (DB-Sulfur SCD column, 60 m × 0.32 mm × 4.2 µm, methyl-deactivated or DB1 60 m × 0.32 mm × 1µm, both from Agilent Technologies) within the stainless steel tubing. The capillary extended 4 cm into the Neptune torch such that He and Ar gas streams flowed coaxially through the transfer line and mixed in the torch (Ther-
ACS Paragon Plus Environment
2
Page 3 of 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
mo Fisher Scientific, Germany). A recent study suggested that preheating of the Ar (before it enters the transfer line) was necessary to prevent a drift in the S-isotope values during the heating program of the GC oven.12 However, such an effect was not observed in our laboratory; therefore, heating the Ar as it flowed in the transfer line of our system was sufficient. However, it was noticed that even a small leak in the Ar sample gas connections could facilitate a drift in δ34S values during the heating program of the GC. Therefore, a drift test (see Results section) and the Ar sample gas connections were and must be checked regularly to prevent this effect. A laboratory-built SF6 reference gas injector with six-way valve allowed addition of either a continuous or time-varying stream of SF6 in He to the GC effluent for tuning and calibration (Figure 1c), as described in Amrani et al.6 The SF6 peak sizes and widths were controlled via a mechanical flow controller (Porter Instruments, PA, USA). The S species were atomized and ionized in the plasma source and yielded 32S+ and 34S+ ions that were transferred to the mass spectrometer unit of the GC/MC-ICPMS system for isotope ratio analysis. We did not attempt to measure 33S in this study although it was possible. The Neptune MC-ICPMS system was a double-focusing magnetic-sector instrument equipped with eight moveable Faraday detectors and one fixed detector for simultaneous detection of different masses. The Faraday detectors were positioned to simultaneously collect 32S+ and 34S+. Previous studies have reported that a mass resolving power (m/∆m, 5%–95%) of 3000–6000 was sufficient to resolve isobaric interference.14-16
Figure 1. Schematic layout of instrumentation components: (a) Neptune Plus multicollector inductively coupled plasma mass spectrometer, (b) gas chromatograph, (c) SF6 reference gas injector, and (d) Deans Switch enlargement.
ACS Paragon Plus Environment
3
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 12
The “medium” resolution mode (~ 4000–5000) of the MC-ICPMS was employed, producing two flat-topped but partially overlapping peaks from the exit slit. The S+ ions were measured as a shoulder on the low-mass side of the larger (overlapping) peak.6 The GC, TCD, and MC-ICPMS system parameters are listed in Table 2.
Table 2. GC and MC-ICPMS parameters
Gas Chromatograph
Agilent 7890B
Capillary column
Agilent,DB Sulfur SCD, 60m x 0.32mm ID x 4.2µm
Injector
S/SL, 250 C
Carrier gas
He , 2.5 ml/min
Oven temperature program
40 C (8 min), 10 C/min , 200 C (5 min)
Transfer line temperature
200 C
Reference gas carrier flow
8 ml/min
Detector
TCD
Heater
250 C
Reference flow
20 ml/min
Makeup flow(He)
2 ml/min
Mass Spectrometer
Thermo Neptune plus
Coolant flow (Ar)
16000 ml/min
Auxiliary flow (Ar)
910 ml /min
Sample gas flow (Ar)
1250-1450 ml/min
Extraction voltage
2000 V
Resolution
Medium resolution, ~5000 resolving power (m/Δm, 5%-95%)
RF power
922-1100 W
Sample cone
Nickel 1.1mm aperture
Skimmer cone
Nickel 0.8mm aperture
Ion lenses
Optimized for sensitivity and peak shape
Detection system
Faraday cups
Integration time
0.189 s
o
o
o
o
o
o
Preparation of sulfur standards for GC/MC-ICPMS analysis. Two types of in-house standard mixtures were prepared for this study: in the liquid phase and in the gas phase. The standards were composed of either DMS, CS2, and thiophene or all three in addition to EMS and DES. Liquid standards were injected into a closed flask (100 ml of toluene) with a septum to create a 430 pmol/µl solution (～86 to 6 pmol concentration on column in split 5 ratio). Gas standards were either injected as is (using different split ratios) from gas Mix 1 and Mix 2 cylinders or prepared gravimetrically. Sulfur compounds (CS2, DMS, thiophene, EMS, DES) were measured into weighed syringes and injected into the vacuum-evacuated cylinders. He (99.995%) was then added to the cylinder to a final pressure of 170 psi. In some of the gas standards, H2S (99.99%) was added to variable concentrations using a pressure regulator to control the mix ratio. The 2.25-L cylinders were stainless steel coated with sulfinert® (Restek, PA, USA). These in-house standards were calibrated directly against the international sulfur standards NBS-127, IAEA-S-1, and IAEA-SO-6 using an elemental analyzer (EA) coupled to an isotoperatio MS system (EA-irMS , Delta plus, Thermo Fisher Scientific, Germany) or MC-ICPMS in aqueous solution for DMS as previously reported.8 Data processing. The data processing procedure was described in detail in Amrani et al.6 and was used with small modifications. The GC/MC-ICPMS system produced transient signals that the Neptune software (3.2.1.1) presented as m/z 32
ACS Paragon Plus Environment
4
Page 5 of 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
and 34 ion currents in 189-ms increments. Ricci et al.17 algorithms were employed as Visual Basic code within Microsoft® Excel18 in the current study and other CSSIA studies.6-13 The Visual Basic code used the m/z 32 data current to define chromatographic peaks, with a starting slope of 0.2 mVs-1 and ending slope of 0.4 mVs-1, and this peak definition was applied to the data channels (i.e., m/z 32 and 34). In the current study, the background signals were determined for each data channel independently by averaging 30 points preceding each peak. In cases when peaks coeluted, 5, 10, and 20 points after (rather than before) the peak were used. The background was then subtracted and peak areas were calculated. The isotope ratio 34S/32S was calculated from the relevant peak areas of each compound. Calibrated isotope ratios were obtained by comparison to SF6 reference gas peaks in the same chromatogram. The SF6, in turn, was calibrated against several sulfur standards, as detailed in the Results section. The results are expressed in conventional δ34S notation as a permil (‰) deviation from the V-CDT standard:
δ 34S = (34 Rsample / 34Rstd ) − 1 where 34R is the integrated 34S/32S ion-current ratio of the sample and standard peaks. RESULTS Calibration. The sample-standard bracketing approach was used to calibrate analytes, as is the common practice in most compound-specific analyses of organic samples and was recently used in several CSSIA studies.6-13 Peaks of SF6 reference gas were introduced at the beginning and at the end of the chromatogram, assuming a stable mass bias of the MS during each run.6 The stability of the reference gas (SF6) used as an internal standard was tested to calibrate the δ34S values of the sulfur compounds within the chromatogram. The normal GC program was run from 40 °C, with a temperature ramp of 10 °C/min and a 5-min hold at 200 °C; the total run time was 54 min (Table 2). SF6 (n=19) was injected during the entire GC run. Peak area variation between the SF6 peaks was 2.8% (RSD), and the variability in the S-isotope ratio was
