FRONTIER RESEARCH ON EARTH EVOLUTION, VOL. 2

New Noble Gas Isotope Analysis System: Design and Implementation Hajimu Tamura1, Keiko Sato1, Takeshi Hanyu2 and Hidenori Kumagai3 1 Research

Program for Data and Sample Analyses, Institute for Frontier Research on Earth Evolution (IFREE) Program for Geochemical Evolution, Institute for Frontier Research on Earth Evolution (IFREE) 3 Program for Deep Sea Research, Institute for Frontier Research on Earth Evolution (IFREE) 2 Research

1. Specification of Chronos

ure temperature of the crucible but emissivity of the crucible changes during operation due to the accumulating glasses of melted samples. Temperature is usually controlled with electric power and temperature at the outside surface of the core tube. The temperature at the outside surface of the core tube is measured by a HT-THERMIC high temperature sheath thermocouple of Yamari Industries, Ltd. A W5%Re-W26%Re thermocouple wears BeO solid insulator and Mo sheath. The crushing extraction subsystem contains two crushing features (Fig. 1c). One is an electromagnetic crusher consisting of a 38 mm internal diameter stainless tube, a 600 g magnetic stainless hammer, a driving coil as an electromagnet by Nihon Denjisokki Co., Ltd., and a Takasago, Ltd. EX-750L2 regulated power supply as the power source of the electromagnet and another is a valve modified crusher. A Stanford Research Systems, Inc. RGA200 residual gas analyzer is mounted on the crushing extraction subsystem to measure the amount of carbon dioxide and nitrogen. A 10 cm3 volume of the extracted gas is divided for measuring carbon dioxide and nitrogen and residual gas is exposed to a titanium-zirconium getter.

The noble gas isotope analysis system for Chronological Study - Chronos - has been launched for K-Ar dating and isotope analyses for helium, neon and argon. Chronos consists of following subsystems; extraction, standard gas supply, purification, and mass spectrometry. Two extraction subsystems with one initial purification stage, six standard gas tanks, one two-stages purification subsystem, and one NOBLE GAS 5400He static single focus sector type mass spectrometer are included in the system for all-purpose on noble gas isotope analysis for rock samples. Figure 1 is the schematic drawing of Chronos. The ultra high vacuum atmosphere is required for all of the system because of low abundance of noble gas elements and relatively high abundance in air. Stainless steel tubes and copper gasket conflat flanges are employed overall system to realize low leak rate of at most 10-11 ccSTP/min and baking up to 250 ºC. Some components should stand higher temperature and gold ring sealing is employed. 12.7 mm outer diameter stainless tubes and Thermo Vacuum Generators ZCRD20R all metal angle valves are employed to reduce the total volume of the system. The mass spectrometer, NOBLE GAS 5400He of GV Instruments, is the only commercial mass spectrometer for noble gas isotope analysis. It has a 27 cm radius magnetic sector for mass analysis, an electron ion source, and two ion collectors to mount various detectors [Haines et al., 2001]. One Faraday cup detector on the high mass side collector (the High Faraday) and a Daly detector (the Daly) and an ion counting detector on the axial collector were chosen. Mass resolution of 600 for the 10% valley definition is realized in type 5400He to eliminate the effect of HD and H3 from 3He on helium measurement. The ion source works on the condition of trap 400 µA and approximate 4.5 kV accelerating voltage. Other variables are sometimes tuned. The purification subsystem has two purification stages (Fig. 1a, 1b: V2, V3). As described in figure 1b, each stage has one titanium-zirconium getter (Ti-2, Ti-3) and one activated charcoal trap (CH1 and CH2). V3 also has one cold SORB-AC getter of SAES NP-10 and one sintered stainless sieve trap with Advanst Research Systems CSW-204N refrigerator (Cryo Trap). Two extraction subsystems are included in Chronos; a heating extraction subsystem and a crushing extraction subsystem. Each extraction subsystem has one titanium-zirconium getter for initial purification. The heating extraction subsystem employing TH-250T resistance furnace of Horiguchi Iron Works realizes over 2000 ºC in the crucible. This is called as the tantalum furnace because the heater and the core tube are made of tantalum. A glass sample holder is mounted at the top of the tantalum furnace. A molybdenum crucible is sited in the core tube. A pyrometer is employed to meas-

2. Standard gases Chronos is to use both for the K-Ar dating and the noble gas isotope analysis. Hence, different standard gases are required for the isotope analysis of helium, of neon and heavier noble gases, and for the argon analysis for K-Ar dating. The basic standard gas for the noble gas isotope analysis is a diluted air. The most abundant noble gas element in air is argon which is one million times abundant than xenon of the least abundant noble gas element. A diluted air which contains enough xenon for calibration of discrimination involves too much argon for measurement. Three air derived standards, AS, NS, XS were preserved for Chronos with different densities. AS is for K-Ar dating, NS is for calibration for neon isotope analysis and XS is for argon, krypton, and xenon isotope analysis. NS does not contain enough xenon and XS contains too much neon. Making air derived standard gases, a 500 cm3 tank of air (1 atm) was sampled in Akiruno, Tokyo and was mounted to the standard gas supply subsystem. One pipette of the Akiruno air was expanded to Vstd, V2 and V1 then was purified by Ti-1 and Ti-2. One pipette of the purified gas was expanded into AS tank. The residual was expanded into XS tank first and NS tank second. The gas in NS tank was expanded to Vstd, V2, and V1 to adjust density. Helium isotope ratios of usual samples are higher than that of air and non-air derived standard gases are used for calibration. Two standard gases were preserved for Chronos. One is Helium Standard of Japan, HESJ [Matsuda et al., 2002] and another is Kaminoyama Helium Standard [Kumagai, 1999]. HESJ was pro1

FRONTIER RESEARCH ON EARTH EVOLUTION, VOL. 2

duced from the original HESJ tank of ERI, Tokyo University and stored into HS tank. Kaminoyama Helium Standard was produced from a tank from RINS, Okayama University of Science and stored into KS tank. AS is used for both calibration of mass discrimination and sensitivity determination. The argon content of original one pipette and the volume of the pipette were determined. 12 measurements of standard rock samples gave the original content of 4.628±0.511×10-8 cm3STP for 40Ar. Over 50 measurements of AS gas gave the pipette volume of 1.3578cm3. The pipetting trend is shown in Figure 2 and the data points relatively largely scatter around the regression line. The argon background of the system is negligibly small to the content of AS, thus this scattering is not the result of unstable background. There is 5% of standard deviation for the sensitivity of argon, and this balances out to the scattering of AS data points. The stored non-air derived helium standard gases were calibrated by an air derived helium gas. An diluted air was prepared with an activated charcoal trap and the cryo trap to remove heavier noble gases including neon, then was stored into a 3 L stainless tank as the air derived helium standard gas. It is indicated that apparent 3He/4He ratio changes logarithmic to 4He intensity in Chronos and apparent 3He/4He ratios were corrected depending on this relationship [Tamura et al., 2005]. 9 measurements of HS gas gave 3He/4He ratio of 2.832±0.020×10-5 and 23 measurements of KS gas gave that of 7.780±0.198×10-6 (Table 1).

4. Basic Procedure of Noble Gas Analysis A suite of experiments must be planned including analyses of standard gases and known rock samples. As described above, AS is supplied to evaluate the stability of argon sensitivity, XS and NS are supplied to determine the discrimination for heavier noble gases, as HS and KS for helium. It is recommended that the suite includes helium dilution experiments to determine the variation of discrimination factor for 3He/4He ratio as pointed by Tamura et al. [2005] Analytical blanks also must be determined in the same condition as sample analyses. Neon must be removed from helium-neon mixture by a sintered stainless sieve trap at 20K for 10 minutes after removing of heavier gases by a charcoal trap chilled by liquid nitrogen. Neon isotope ratios are varied in helium rich condition as pointed by Hiyagon [Hiyagon, 1989]. Longer trap period causes adsorption of helium to the trap and increase uncertainty of abundance and isotope ratio of helium. If there is too much helium such as over 2×10-7 Torr in V2 + V3, helium in the trap volume should be released until it reduces to 2×10-7 Torr in V2 + V3 or less before release of neon. Trapped neon is released from the trap at 50K. 40Ar and 12C16O must be monitored in neon analysis for later cor2 rection. A charcoal trap cooled to liquid nitrogen temperature is left opened to reduce these peaks interfering neon measurement. Argon and heavier two noble gases also must be separated to prevent rich 40Ar disturbing xenon (and krypton) analysis. Xenon is trapped to a charcoal trap at 250K and almost argon is released. A volume proportional argon remains in the trap and it is recommended to lade out argon until the residual 40Ar is reduced to 2 × 10-8 Torr as partial pressure in the V2 + V3 volume.

3. Volumetry The absolute and the relative volumes of each part of the system are important information. A sample gas is lost on a measurement by separating a part of the system to prevent contamination and unexpected sample loss. The residual sample material in the extraction subsystem sometimes absorbs sample gases or discharges contaminant gases. Hence, the valve between the extraction subsystem and the purification subsystem is closed after extraction and a part of helium and neon is lost. Helium is also lost in the helium-neon separation. On the other hand, a sample is sometimes diluted because it has so high density that the detectors cannot measure. The volume information is necessary to account these losses. The basic volumetry method is as follows. A known volume is mounted to the system and gas with known pressure is filled in the volume. Then the gas is expanded to each part of the system and the pressure is measured. The capacitance manometer is used to measure the pressure. This method has serious problems. The air is used as the gas which has known pressure (1 atm), but the air is absorbed by titanium-zirconium getters, activated charcoals, and a SORB-AC getter pump. There are various volumes in the system, such as some cubic centimeters for the minimum and some liters for the maximum. The pressure changes a little when the gas is expanded from a large volume to a small volume, thus poor precision can only get in this situation. Two times volumetries were performed in this method and the result are Table 2. Alternative relative volumes of frequently used pairs are determined using standard gases. The result is shown in Table 3. A systematic sensitivity change depending on intensity is exhibited for Chronos [Tamura et al., 2004] and these alternative relative volumes include the effect of sensitivity change.

5. Basic Correction Methods The intensity of a peak declines in exponential low usually. Hence, the intercept of an exponential curve or a line is employed as the representative value of peak intensity. The average of peak intensities is also employed in extremely low intensity of which actual values show so wide distribution that impossible to determine a trend. The ratio of peaks is usually represented by the average but varies linearly in a few cases. The intercept of the line is employed in the latter cases. At first the variation of helium isotope ratio should be determined. As Tamura et al. [Tamura et al., 2005] says, a log-linear equation is derived from the helium dilution experiment of standard helium gases:

Rv = 1 + alog10V where Rv is the ratio of an actual 3He/4He value at given 4He intensity and the real 3He/4He value. This equation gives correction factors for 3He/4He ratio corresponding given 4He intensities. All helium isotope ratios are arranged to the value at 1 V. Corrections for peak overlapping are also corrected in this stage. As described above, 40 Ar 2+ interference to 20 Ne and 12C16O 2+ to 22Ne are monitored. The ratio of 40Ar2+ / 40Ar+ is 0.3 2 as 12C16O22+ / 12C16O2+ is 0.01. The effect from 12C16O2 is almost negligible. Mass discrimination correction factors are determined for every isotope ratios measured as averages of the results of stan2

FRONTIER RESEARCH ON EARTH EVOLUTION, VOL. 2

dard gas analyses. In noble gas analysis, mass discrimination is constant and corrected with discrimination factors derived from the average in the suite of analyses [Ozima and Podosek, 2002]. Current discrimination factors for helium, neon and argon are shown in Table 4. Analytical blanks are also checked in this stage. As the result of the online, sequential preparation, the blanks easily vary by history of measurements or treatments at sample loading. Typical sets of analytical blanks are shown in table 5 for each of heating and crushing (‘cold’ is essential blank for purification subsystem), but over 1.5×10-9 cm3 STP of 40Ar residual was observed for a fresh crucible, or 8Ra helium was observed with 5×10-9 cm3 STP 4He after crushing of a MORB glass sample. Low blanks for heating are realized by over two hour baking at 1800 - 2000°C after sample loading and one hour baking at 1800 - 2000°C after every analyses. For crushing, several sets of crushing tubes and hammers are applied depending on the sample characteristics. Contents of noble gases are estimated from intensities and sensitivity derived from analyses of standard rock samples and AS standard air after corrections above. Sensitivity of argon in February 2004 is displayed in Sato et al. [Sato et al., 2005] A typical set of relative sensitivities to argon for each noble gas elements are displayed on table 6. Sensitivities should be determined for each suite of experiments. Recommended amounts of sample gas for each noble gas elements are shown in table 7.

Ozima, M. and F. A. Podosek, Noble Gas Geochemistry, 286 pp., Cambridge University Press, Cambridge, United Kingdom, 2002. Sato, K., H. Tamura, H. Kumagai and T. Hanyu, Application of K-Ar dating system to be performed by new noble gas mass spectrometry and its calibration from standard air analysis, In Frontier Research on Earth Evolution, IFREE Report 2003-2004, 2, 2005. Tamura, H., H. Kumagai, K. Sato and T. Hanyu, Systematic Discrimination for 3He/4He depending on 4He Intensity, J. Mass Spectrom. Soc. Jpn., submitted on January 20th 2005. Tamura, H., T. Hanyu, K. Sato and H. Kumagai, Evaluation for Sensitivity Linearity of 5400He Noblegas Mass Spectrometer in Yokosuka HQ, JAMSTEC (No.SV042), 2004 Annual Meeting of Isotope Ratio Division of Mass Spectrometry Society Japan Annual Meeting of Isotope Ratio Division of Mass Spectrometry Society Japan, Mass Spectrometry Society Japan, Yamagata, Japan, 24-26 November, 2004.

6. Conclusion Chronos, the noble gas isotope analysis system of IFREE, has been designed and implemented for the noble gas isotope analysis and K-Ar dating. The basic procedure of the analysis for helium, neon and argon isotopes and the argon analysis for K-Ar dating are minimally established. Acknowledgments. We thank Prof. Ichiro Kaneoka of ERI, Tokyo University and Prof. Hironobu Hyodo of RINS, Okayama University of Science for their supply of helium standards. Prof. Ichiro Kaneoka also supplied his mineral standard EB-1 for us. We also thank Professors Kazuo Saito and Naoyoshi Iwata of Yamagata University for their supply of various rock and mineral standards for K-Ar dating, Professors Kyoichi Ishizaka and Takahiro Tagami of Kyoto University for their supply of Bern4B biotite standard for K-Ar dating, Dr. Kozo Uto, Dr. Tomoaki Sumii, and Dr. Masafumi Sudo of GSJ for their supply of SORI93 biotite standard for K-Ar and Ar-Ar dating.

References Kumagai, H., Variation of noble gas signatures controlled by tectonic conditions and magmatic processes: a case study for an area around the Rodriguez Triple Junction in the Indian Ocean, Ph.D thesis, 129 pp, Tokyo University, December 1999. Matsuda, J., T. Matsumoto, H. Sumino, K. Nagao, J. Yamamoto, Y. Miura, I. Kaneoka, N. Takahata, and Y. Sano, The 3He/4He ratio of the new internal He Standard of Japan (HESJ), Geochemical Journal, 36, 191-195, 2002. Haines, R. C., A. N. Eaton and S. N. Dudd, SPECIFICATION NOTE 701 Micromass 5400 Static Vacuum Mass Spectrometers, Micromass UK Ltd., Manchester, UK, Sept. 2001. Hiyagon, H., Neon Isotope Measurement in the Presense of Helium, Mass Spectroscopy, 37, 325-330, 1989.

3

FRONTIER RESEARCH ON EARTH EVOLUTION, VOL. 2

Figure 1. Schematic Drawing of Chronos

4

FRONTIER RESEARCH ON EARTH EVOLUTION, VOL. 2

Table 2. Volume

Table 3. Alternative relative volumes Figure 2. Trend of AS

Table 1. Helium standard gases

Table 4. Degree of discrimination for helium, neon and argon

Table 5. Analytical blanks for most abundant isotopes for noble gas elements (cm3 STP)

Table 6. Relative sensitivities to argon

Table 7. Recommended amount for analysis (cm3 STP)

5