A new CF-IRMS system for the quantification of the stable isotopes of carbon monoxide from ice cores and small air samples

The paper presents a method for the determination of the concentration and isotope ratios of carbon monoxide in ice cores. Although the basic principle of the method is based on published procedures, the application to CO in ice cores is new and the paper contains results from a number of tests providing insight into the potentials and limitations of the methodology. Therefore the paper deserves publication in Atmospheric Measurement Techniques. Nevertheless, there are several issues which need to be addressed before the paper should be published. i) The paper presents results of a variety of blank tests, however it is not always clear which process contributes to the


Introduction
As the major sink for hydroxyl radicals (OH), carbon monoxide plays a key role in global atmospheric chemistry and strongly affects the oxidizing capacity of the atmosphere.The major sources of atmospheric CO in the modern atmosphere are known and incorporation of stable isotopes ( 13 C and 18 O) renders inverse model results for CO budget estimates much more robust compared to inversions of CO mixing ratios only (Bergamaschi et al., 2000a,b).The stable isotopes of CO ( 13 C and 18 O) from air bubbles trapped in ice cores provide valuable information of CO source strength for the past atmosphere, which will enhance the understanding of atmospheric chemistry in the past.However, essentially no observations exist from which the relative source strengths can be estimated.Haan et al. studied the CO mixing ratio in both Antarctic and Greenland ice cores by reduction gas detector (RGD) (Haan et al., 1996(Haan et al., , 1998)) insight to measure ice core CO isotopes.Unfortunately, the volume of air trapped in ice cores is only around 100 mL (STP) per kg ice, and the ice core CO mixing ratio is previously reported to be only ∼50 ppbv (Haan et al., 1996(Haan et al., , 1998)).Measuring CO isotope composition in ice core is extremely challenging since the sample size is very limited, in which case the conventional dual inlet micro-volume analysis method is not applicable.
Continuous-flow isotope ratio mass spectrometry (CF-IRMS) systems have been developed for measuring atmospheric CH 4 (Rice et al., 2001;Merritt et al., 1995;Lowe et al., 1991), atmospheric N 2 O (R öckmann et al., 2003) and volatile organic compounds (VOC) in ambient air (Rudolph et al., 1997).The primary advantage of using CF-IRMS for CO isotope analysis by monitoring the relative abundance of masses 44, 45, and 46 has been shown by Mak and Yang (1998).A recent study by Tsunogai et al. (2002) has shown the advantage of isotope analysis for atmospheric CO using CF-IRMS by simultaneously monitoring the CO + ion currents at masses 28, 29, and 30.However, preparing CO standard gas with known isotope ratios and purifying CO presents challenges.In this study, a new cryogenic vacuum extraction system, which is principally based on the approach originally outlined in previous study (Brenninkmeijer et al., 1993), is designed to analyze isotope ratios of CO for small air samples (∼268-625 pmol CO), specifically for CO in air bubbles trapped in Antarctic ice core by measuring CO-derived CO 2 .

Instrumentation and procedure
A schematic diagram of the online extraction and preconcentration method for carbon and oxygen isotope analyses of ice core CO is shown in Fig. 1.A wet extraction system for ice cores is combined with a cryogenic vacuum system for CO extraction.Ambient air samples are run by filling the sample canisters with air and attaching to the inlet of the cryogenic vacuum system.After passing through the cryogenic cleanup traps with a flow rate of 50 mL/min, atmospheric CO 2 , H 2 O, N 2 O and other condensable species present in air samples are removed due to condensation at liquid nitrogen temperature (Brenninkmeijer, 1993).The purified CO is then selectively and quantitatively oxidized to CO 2 by the Sch ütze reagent, retaining the original 18 O signature of CO (Brenninkmeijer et al., 1993).The Sch ütze reagent is prepared according to the method described by previous studies (Sch ütze, 1949;Smiley, 1949).A bypass is installed for the Sch ütze reactor tube as well, allowing the postrun flushing with helium.The degradation of Sch ütze is indicated by its color change, e.g. from yellow to brown, which informs of the replacement (Mak and Yang, 1998).
The CO-derived CO 2 collected in trap3 is then eluted with cryogenically purified ultrapure helium with flow rate of 40 mL/min for 5 min and cryogenically focused in a microvolume trap of the modified Finnigan Preconcentration unit through a Valco six-port valve (Fig. 1).The cryofocus trap is then heated to room temperature and the CO 2 is loaded into the GC column (25 m×0.25 mm ID Poraplot Q capillary) with helium flow of 1 mL/min from back inlet of GC (Fig. 1).The resolved CO 2 then passes through an open split into the Finnigan MAT Delta-plus isotope ratio mass spectrometer at a flow rate of about 0.4 mL/min.
The detailed calculation of δ 13 C and δ 18 O is based on the two observable ion-current ratios (45/44 and 46/44) as discussed in previous study (Santrock et al., 1985).δ 13 C of the original CO is determined by the δ 13 C of derived CO 2 , whereas δ 18 O is determined based on calibration gas (Brenninkmeijer et al., 1993).The mixing ratio of CO for an air sample is determined by the ratio of peak areas between a sample and a cal-Introduction

Conclusions References
Tables Figures

Back Close
Full Screen / Esc

Printer-friendly Version
Interactive Discussion ibration gas.The precision of CO mixing ratios for air samples results reported in this study is better than 5% (±1σ) based on reproducibility of 4 to 10 runs for field samples.The Finnigan Precon system (Mak and Yang, 1998) is replaced by a new cryogenic vacuum system for measuring CO in air bubbles of ice core samples.A further detailed procedure for ice extraction will be discussed later in this paper.The ice core sample is placed in a 1.2 L glass container with a 4.5 inch glass flange, viton o-ring and horseshoe clamp.A four-loop coil trap between the container and the inlet of the cryogenic vacuum system is used to remove water vapor.Upon melting, the air sample released from around 1 kg ice core is around 100 mL (STP), giving an initial pressure in the headspace of about 300 mbar.This pressure drops to 60 mbar after air sample is processed for 5 min.A 150 mL glass flask filled with zero air (CO free air) is expanded to the ice container for building up pressure to around 500 mbar after processing the air sample for 5 min.The disadvantage for this method is the large blank from Sch ütze reagent due to long process time, which largely limits the precision of isotope ratios measured using this technique.We tested the use of molecular sieve 5A and 13X held at liquid nitrogen temperature is used to condense air released from ice cores in order to shorten the process time and decrease the Sch ütze blank, however isotopic fractionation of both carbon (2-3‰) and oxygen (1-4‰) was evident during the elution process.This isotopic fractionation was not reported in Tsunogai et al. (2002).Use of a Teflon beaker liner (Welch Fluorocarbon Inc.) was also investigated as its volume is adjustable and thus builds up pressure in the headspace by shrinking and pushing water upward.This can counteract the pressure reduction during processing.However, permeability of CO across Teflon film limits the use of this material.Similarly, Tedlar (polyvinyl fluoride, PVF), another polyfluorocarbon but with low CO permeability, was also tested.However, it is quite difficult to engineer a Tedlar liner because of its rigidity.
In the end, while the idea is intriguing and may warrant further study, we opted to keep our sampling protocol simple with the dilution process and characterize the Sch ütze blank to a high degree of precision.Clearly, an old fashioned mercury toepler pump could be ideal to solve this problem by reducing extraction time and Sch ütze blank.1).
As for ice core analysis, the system blank includes the Sch ütze blank and ice blank.The former is 60 pmol for 10 min collection time.The latter is determined by tests base on artificial bubble free ice from LGGE, France (Table 4).High ice blank of 140 pmol CO was found resulting from CO adsorption on the ice surface, which decreased significantly after the ice was flushed with zero air for a few times to a level of 20 pmol CO.
The ice blank was subjected to light tests as well.Sample environment with or without light exposure were compared.The isotope composition (δ 13 C=−15‰) of the ice blank and the slight difference of blank signal between light and dark environment (Table 4) indicates the ice blank is from air diffusion/adsorption not in situ photochemical production.

Ice core introduction
South Pole ice core samples provided by National Ice Core Lab (NICL) were dry drilled in 2004 by Jihong Cole-Dai.They contain very low contamination since no fluid was used for drilling.The ice core samples are semi cylindrical with average length of 17 cm and radius of 5 cm.The ice core samples were first trimmed with a band saw to remove several mm of surface, and then scraped with a stainless steel disposable scalpel to remove additional 1-2 mm of surface.Prior to use, the container was cleaned with Introduction

Conclusions References
Tables Figures

Back Close
Full

Calibration gas test
Experiments were performed to check the reliability and precision of stable C and O isotope data using a calibration gas of known concentration and isotopic composition (130 ppbv; δ 13 C=−45.56‰VPDB; δ 18 O=−1.94‰VSMOW).Concentration and δ 13 C and δ 18 O values were determined by using whole air extraction of several hundred liters of air for 14 CO analysis (Mak et al., 1994) followed by dual-inlet microvolume analysis with the Delta Plus IRMS (Mak et al., 1994).The δ 18 O of Sch ütze reagent is derived with the δ 18 O of CO for the calibration gas and the δ 18 O of the produced CO 2 in cryogenic vacuum extraction system.As shown in Fig. 2, the mean value of δ 13 C(CO) is −45.40‰(VPDB) with a standard deviation of 0.33‰ (±1σ), for 47 runs of the measurements with CO amount ranging from 232 pmol to 4 nmol, and the mean value of δ 18 O(CO 2 ) is 0.42‰ (VSMOW) with a standard deviation of 0.32‰ (±1σ).calibration runs (0.42‰), and δ 18 O Cal CO is the isotopic signature of the CO calibration runs (−1.94‰,VSMOW), which was determined based on conventional dual inlet micro-volume analysis method (Brenninkmeijer et al., 1993;Mak and Brenninkmeijer, 1998).

Field sample measurement
The accuracy and reproducibility of the above analysis method was also tested in processing field samples.Air samples collected from two sites: the Mauna Loa Observatory and Westmann Islands, Iceland, with the whole volume on the order of 600 L (STP) were collected into high pressure aluminum air cylinders using a previously published protocol (Mak and Brenninkmeijer, 1998).These field samples were first processed with off-line extraction and analyzed with conventional dual inlet-microvolume method.Aliquot of remaining samples were then processed with the new on-line analysis.Measurements based on the on-line extraction CF-IRMS are compared with the off-line results (Table 1).The 0.2‰ standard deviation for δ 13 C and 0.6‰ standard deviation for δ 18 O obtained from the new on-line analysis method for these field samples indicate reliable reproducibility.Moreover, no significant difference for δ 13 C or δ 18 O is found between the off-line dual inlet microvolume analysis and the on-line extraction CF-IRMS analysis, implying the reliable accuracy of the new on-line method.

South Pole ice core measurement
As described in the methodology section, six South Pole ice core samples (∼1 kg for each sample, with depth ranging from 133 to 177 m) were processed with this on-line technique.Results including both CO mixing ratios and isotope ratios were shown in the following table (Table 2).All these CO mixing ratio and isotope data are raw data and have not been calibrated for gravitational fractionation.
with deionized water, then heated overnight at 90 • C. The ice core sample was weighted and sealed in the container and immersed in cooling bath held at −20• C. The ice and the container were repeatedly evacuated and flushed with zero air for 30 min.After final pumping, the ice was melted in hot water bath held at temperature of 60-70 • C. The occluded air extracted from the ice was then processed through the cryogenic extraction line and followed CF-IRMS.
This δ18 O(CO 2 ) value was used for determining the δ 18 O of the Sch ütze reagent and δ 18 O(CO) for samples based on the following equation:δ 18 O(‰, VSMOW) sample = 2δ 18 O CO 2 − 2δ 18 O Cal CO 2 −δ 18 O Cal CO (1)where δ 18 O CO 2 is the isotopic signature of the CO-derived CO 2 with one of the oxygen atoms from the Sch ütze reagent, δ 18 O Cal CO 2 is the average value obtained from the Figure 2. 271
Diagnostic experiments have been conducted on the cryogenic extraction system, and results indicate this new technique produces reliable isotope ratios for 100 mL air samples with CO mixing ratio as low as 60 ppbv (∼268 pmol CO).The diagnostic tests include analytical blank tests, precision tests and accuracy tests.Analytical blank is lower than 40 pmol CO for 100 mL air sample (see Table3).Analytical precision of 2 ppb (±1σ) for CO mixing ratio, 0.2‰ (±1σ) for δ 13 C and 0.6‰ (±1σ) for δ 18 O can be obtained for sample size of 268 pmol CO (100 mL with CO mixing ratio of 60 ppbv).Meanwhile, both δ 13 C and δ 18 O from CF-IRMS were compared with those from dual inlet IRMS to examine the accuracy of this new technique (Table

Table 1 .
sample size as small as 220 pmol, making it especially useful in analyzing air samples with very limited amount, such as measuring stable isotopes for atmospheric CO trapped in ice core.Six South Pole ice core samples were measured for CO mixing ratio, δ 13 C(CO) and δ 18 O(CO) based on this new on-line technique.This is the first report on measuring ice core samples for CO isotopes.Introduction Lowe, D. C., Martin, R. J., and Clarkson, T. S.: A novel method for compound specific determination of delta C-13 in volatile organic compounds at ppt levels in ambient air, Geophys.Res.Lett., 24, 659-662, 1997.Santrock, J., Studley, S. A., and Hayes, J. M.: Isotopic analyses based on the mass spectrum of carbon dioxide,Anal.Chem., 57, 1444-1448, 1985.Comparison of δ 13 C and δ 18 O results of field CO samples obtained by dual inlet microvolume analysis and on-line extraction CF-IRMS.
The consistency of CO mixing ratio and isotope ratios between the duplicate South Pole ice core samples (165 m) indicate the above analytical technique produces consistent results for ice core

Table 3 .
Sch ütze blank of the cryogenic system tested with zero air.