PREPARATION AND ANALYSISIsotopic Uniformity: After production, the RM tuSUPPLEMENTAL INFORMATIONNational Institute of Standards & Technology Report of Investigation Reference Materials 8562, 8563, 8564 8562: CO2-Heavy, Paleomarine Origin (Carbon Dioxide) 8563: CO2-Light, Petrochemical Origin (Carbon Dioxide) 8564: CO2-Biogenic, Modern Biomass Origin (Carbon Dioxide) These Reference Materials (RMs) are intended to provide carbon dioxide samples of known isotopic composition and uncertainty with 13C/12C and 18O/16O ratios expressed in parts per thousand relative difference (‰) from Vienna Peedee Belemnite (VPDB) or Vienna Standard Mean Ocean Water (VSMOW) [1]. RMs 8562, 8563, and 8564 are not certified, but their use allows comparability of stable carbon and oxygen isotope ratio data obtained by investigators in different laboratories, and provides a means to standardize laboratory gases with measurement traceability to an internationally accepted reference scale. An RM unit consists of two borosilicate glass tubes, each 9 mm in diameter and about 30 cm in length. Each tube contains approximately 400 µmol of gas and is labeled with a unique number. Reference Values: A NIST Reference value is a non-certified value that is the best estimate of the true value; however, the values do not meet NIST criteria for certification and are provided with associated uncertainties that may not include all sources of uncertainty. Six isotopic reference values, along with estimated expanded standard uncertainties (U), are provided in Table 1. Table 1. Reference Values for δ13CVPDB and δ18OVPDB Reference Material δ13CVPDB ± U(a) δ18OnVPDB ± U(a) (‰) (‰) RM 8562 (CO2-Heavy) –3.76 ± 0.07 –8.43 ± 0.44 RM 8563 (CO2-Light) –41.56 ± 0.09 –23.61 ± 0.48 RM 8564 (CO2-Biogenic) –10.45 ± 0.07 +0.06 ± 0.40 (a) U = kuc, where uc is the combined standard uncertainty and k = 2 is the coverage factor [4]. Reference values are calculated on the basis of currently accepted oxygen isotope relationships, where λ = 0.50 and K = 0.008 332 958 2 [2,3]. δ18OnVPDB is the normalized delta value on a scale such that δ18OVPDB of NBS 19 ≡ −2.2 ‰ and δ18OSLAP/VSMOW ≡ −55.5 ‰. [2,5]. Expiration of Reference Values: The reference values for the isotopic composition of RMs 8562, 8563, and 8564 are valid until 31 December 2006, provided they are handled and stored in accordance with the instructions given in this report. Maintenance of Reference Values: The isotopic compositions of these RMs are monitored on a yearly basis. If substantive changes occur to the values before expiration of this report, the National Institute of Standards and Technology (NIST) will notify the purchaser. Return of the attached registration card will facilitate notification. Technical coordination for these RMs was provided by R.M. Verkouteren of the NIST Surface and Microanalysis Science Division. The supporting aspects concerning distribution of these RMs were coordinated through the NIST Standard Reference Materials Program by B.S. MacDonald of the NIST Measurement Services Division. Richard R. Cavanagh, Chief Surface and Microanalysis Science Division Gaithersburg, MD Report Issue Date: 24 February 2005 Robert L. Waters, Jr., Chief See Report Revision History on Last Page Measurement Services Division RMs 8562, 8563, 8564 Page 1 of 5INSTRUCTIONS FOR USE Storage and Use: Until use, it is recommended that these RMs be stored in their original container at ambient temperature (20 ºC to 30 ºC). Before use, the user’s inlet system should be conditioned with carbon dioxide of isotopic composition similar to the RM. To open, the RM requires the proper use of a suitable tube cracker [e.g., 6-9], and after opening should be used immediately for calibration or standardization. If desired, a sample can be stored in a clean glass breakseal or container fabricated with an all-glass stopcock coated with a hydrocarbon-based grease, or other vessel known to preserve the isotopic integrity of carbon dioxide [10]. NOTE: Because very limited quantities of these materials exist, distribution is limited to one unit (two tubes) of each RM per three-year period. Users are strongly advised to prepare their own internal standards for routine quality control and to normalize or compare those standards with these RMs. PREPARATION AND ANALYSIS Preparation: RMs 8562, 8563, and 8564 were prepared by R.M. Verkouteren of the Surface and Microanalysis Science Division. Details of the preparation method are published in reference 11. RM 8562 was prepared in 1995 from carbon dioxide provided by T. B. Coplen, U.S. Geological Survey (USGS) (Reston, VA, USA). This gas originated from natural thermal decomposition of a Jurassic limestone deposit in the Southeastern United States [12], and was sampled through a well in 1982. RM 8563 was prepared in 1996 from SFE-grade carbon dioxide (99.999 % molar purity) obtained through Scott Specialty Gase (Plumsteadville, PA). This gas originated from combustion of byproducts from a petrochemical refinery in the Eastern United States. Lastly, RM 8564 was prepared in 1996 from research grade carbon dioxide (99.995 % molar purity) obtained through Scott Specialty Gases (Plumsteadville, PA). This gas originated from the fermentation of C4 biomass (corn) in a grain distillery in the Central United States. Isotopic Uniformity: After production, the RM tubes were heated to 70 ºC for 24 hours to accelerate potential isotopic exchange of carbon dioxide with trace water vapor or the silicate interior of the tubes, then were allowed to incubate for at least one month before the batch was characterized. Determination of the isotopic heterogeneity of each RM batch was facilitated through a uniform sampling design, where samples were selected for measurement (through their unique identifier) to explore the effects of inherent nuisance factors documented during RM production; fortunately, no factors were significant. Measurements on RM 8562 and RM 8563 were performed at NIST, while measurements on RM 8564 were performed at USGS. No datum was excluded as a statistical outlier. Observed isotopic variability was apportioned between instrumental effects and true sample-to-sample heterogeneity. Table 2. Isotopic Variations in CO2 Reference Materials Reference Material Estimated RM Variation in δ13C