Vienna Standard Mean Ocean Water

 

Abundances of a particular isotope in a substance are usually given relative to some reference material, as a delta in parts per thousand (‰) from the reference. For example, the ratio of deuterium (²H) to hydrogen-1 in a substance x may be given as

${\displaystyle \delta \,^{2}\mathrm {H} _{x/{\text{reference}}}{\text{ (in ‰)}}=\left({\frac {(^{2}\mathrm {H} /^{1}\mathrm {H} )_{x}}{(^{2}\mathrm {H} /^{1}\mathrm {H} )_{\text{reference}}}}-1\right)\cdot 1000}$ ,

where ${\displaystyle (^{2}\mathrm {H} /^{1}\mathrm {H} )_{x}}$  denotes the absolute concentration in x.^[2]

In 1961, pursuing a standard for measuring and reporting deuterium and oxygen-18 concentrations, Harmon Craig of the Scripps Institution of Oceanography in San Diego, California, proposed an abstract water standard. He based the proportions on his measurements of samples taken by Epstein & Mayeda (1953) of ocean waters around the world.^[3] Approximating an average of their measurements, Craig defined his "standard mean ocean water" (SMOW) relative to a water sample held in the United States' National Bureau of Standards called NBS-1 (sampled from the Potomac River^[4]). In particular, SMOW had the following parameters relative to NBS-1:

• δ ²H _SMOW/NBS-1 = 50‰, i.e., an enrichment of 5%;
• δ ¹⁸O _SMOW/NBS-1 = 8‰, i.e., an enrichment of 0.8%.^[3]

Later, researchers at the California Institute of Technology defined another abstract reference, also called "SMOW", for oxygen-18 concentrations, such that a sample of Potsdam Sandstone in their possession satisfied δ¹⁸O _{sandstone/SMOW} = 15.5‰.^[5]

To resolve the confusion, November 1966 meeting of the Vienna-based International Atomic Energy Agency (IAEA) recommended the preparation of two water isotopic standards: Vienna SMOW (VSMOW; initially just "SMOW" but later disambiguated^[5]) and Standard Light Antarctic Precipitation (SLAP).^[6] Craig prepared VSMOW by mixing distilled Pacific Ocean water with small amounts of other waters. VSMOW was intended to match the SMOW standard as closely as possible. Craig's measurements found an identical ¹⁸O concentration and a 0.2‰ lower ²H concentration.^[7] The SLAP standard was created from a melted firn sample from Plateau Station in Antarctica.^[7] A standard with oxygen-18 and deuterium concentrations between that of VSMOW and SLAP, called Greenland Ice Sheet Precipitation (GISP), was also prepared.^[7] The IAEA began distributing samples in 1968, and Gonfiantini (1978) compiled analyses of VSMOW and SLAP from 45 laboratories around the world.^[8] The VSMOW sample was stored in a stainless-steel container under nitrogen and was transferred to glass ampoules in 1977.^[7]

The deuterium and oxygen-18 concentrations in VSMOW are close to the upper end of naturally occurring materials, and the concentrations in SLAP are close to the lower end.^[2] Due to confusion over multiple water standards, the Commission on Isotopic Abundances and Atomic Weights recommended in 1994 that all future isotopic measurements of oxygen-18 (¹⁸O) and deuterium (²H) be reported relative to VSMOW, on a scale such that the δ¹⁸O of SLAP is −55.5‰ and the δ²H of SLAP is −428‰, relative to VSMOW.^[9][10] Therefore, SLAP is defined to contain 94.45% the oxygen-18 concentration and 57.2% the deuterium concentration of VSMOW.^[9] Using a scale with two defined samples improves comparison of results between laboratories.

In December 1996, because of a dwindling supply of VSMOW, the IAEA decided to create a replacement standard, VSMOW2. Published in 1999, it contains a nearly identical isotopic mixture. About 300 liters was prepared from a mixture of distilled waters, from Lake Bracciano in Italy, the Sea of Galilee in Israel, and a well in Egypt, in proportions chosen to reach VSMOW isotopic ratios. The IAEA also published a successor to SLAP, called SLAP2, derived from melted water from four Antarctic drilling sites.^[11] Deviations of ¹⁷O, and ¹⁸O in the new standards from the old standards are zero within the error of measurement.^[12] There is a small but measurable deviation of ²H concentration in SLAP2 from SLAP—δ²H_SLAP2/VSMOW is defined to be −427.5‰ instead of −428‰—but not in VSMOW2 from VSMOW.^[13] The IAEA recommends that measurements still be reported on the VSMOW–SLAP scale.^[14]

The older two standards are now kept at the IAEA and no longer sold.^[15]

All measurements are reported with their standard uncertainty. Measurements of particular combinations of oxygen and hydrogen isotopes are unnecessary because water molecules constantly exchange atoms with each other.

Except for tritium, which was determined by the helium gas emitted by radioactive decay, these measurements were taken using mass spectroscopy.

• Deuterium (²H / ¹H) – 155.76±0.05 ppm,^[16][17] about 1 in 6420 hydrogen atoms
• Tritium (³H / ¹H) – 18.5±3.6 TU^[a] = (1.85±0.36)×10⁻¹¹ ppm, measured on 16 September 1976,^[8][16] about 1 in 5.40×10¹⁶ hydrogen atoms
• Oxygen-18 (¹⁸O / ¹⁶O) – 2005.20±0.45 ppm,^[16][18] about 1 in 499 oxygen atoms
• Oxygen-17 (¹⁷O / ¹⁶O) – 379.9±0.8 ppm,^[16][19] about 1 in 2640 oxygen atoms

Based on the results of Gonfiantini (1978), the IAEA defined the delta scale with SLAP at −55.5‰ for ¹⁸O and −428‰ for ²H. That is, SLAP was measured to contain approximately 5.55% less oxygen-18 and 42.8% less deuterium than does VSMOW, and these figures were used to anchor the scale at two points.^[8] Experimental figures are given below.

• ²H / ¹H – 89.02±0.05 ppm, δ := −428.5 ± 0.4‰,^[16][17] about 1 in 11230 atoms
• ³H / ¹H – 374±9 TU^[a] = (3.74±0.09)×10⁻¹⁰ ppm, measured on 16 September 1976,^[8] about 1 in 2.67 × 10¹⁵ atoms
• ¹⁸O / ¹⁶O – 1893.91±0.45 ppm, δ := −55.5‰,^[16] about 1 in 528 atoms
• ¹⁷O / ¹⁶O – δ = −28.86 ± 0.1‰,^[16] about 1 in 3700 atoms

The concentrations of ¹⁷O, and ¹⁸O are indistinguishable between VSMOW and VSMOW2, and between SLAP and SLAP2. The specification sheet gives the standard errors in these measurements.^[20] The concentration of ²H is unchanged in VSMOW2 as well, but is slightly increased in SLAP2. The IAEA reports:

• δ²H_SLAP2/VSMOW = −427.5 ± 0.3‰,^[21] (Compare −428‰ for SLAP.)

On 6 July 2007, the tritium concentration was 3.5±1.0 TU in VSMOW2, and 27.6±1.6 TU in SLAP2.^[22]

• δ ²H _GISP = −189.5 ± 1.2‰
• δ ¹⁸O _GISP = −24.66 ± 0.09‰
• δ ¹⁷O _GISP = −12.71 ± 0.1‰^[23]

The VSMOW–SLAP scale is recommended by the USGS, IUPAC, and IAEA for measurement of deuterium and ¹⁸O concentrations in any substance.^[24][25][9] For ¹⁸O, a scale based on Vienna Pee Dee Belemnite can also be used.^[9] The physical samples, which are distributed by the IAEA and U.S. National Institute of Standards and Technology, are used to calibrate isotope-measuring equipment.^[26]

Variations in isotopic content are useful in hydrology, meteorology, and oceanography.^[27] Different parts of the ocean do have slightly different isotopic concentrations: δ ¹⁸O values range from –11.35‰ in water off the coast of Greenland to +1.32‰ in the north Atlantic, and δ ²H concentrations in deep ocean water range from roughly –1.7‰ near Antarctica to +2.2‰ in the Arctic. Variations are much larger in surface water than in deep water.^[28]

In 1954, the International Committee for Weights and Measures (CIPM) established the definition of the Kelvin as 1/273.16 of the absolute temperature of the triple point of water. Waters with different isotopic compositions had slightly different triple points. Thus, the International Committee for Weights and Measures specified in 2005^[29] that the definition of the kelvin temperature scale would refer to water with a composition of the nominal specification of VSMOW.^[30] The decision was welcomed in 2007 by Resolution 10 of the 23rd CGPM.^[31] The triple point is measured in triple-point cells, where the water is held at its triple point and allowed to reach equilibrium with its surroundings. Using ordinary waters, the range of inter-laboratory measurements of the triple point can be about 250 μK.^[32] With VSMOW, the inter-laboratory range of measurements of the triple point is about 50 μK.^[33]

After the 2019 revision of the SI, the kelvin is defined in terms of the Boltzmann constant, which makes its definition completely independent of the properties of water. The defined value for the Boltzmann constant was selected so that the measured value of the VSMOW triple point is identical to the prior defined value, within measurable accuracy.^[34] Triple-point cells remain a practical method of calibrating thermometers.^[33]

• International Temperature Scale of 1990
• Properties of water

• International Atomic Energy Agency – IAEA
• ITS-90 – Swedish National Testing and Research Institute
• ITS-90 – Omega Engineering
• Scripps Institution of Oceanography Archived 2013-06-07 at the Wayback Machine
• Temperature Sensors – information repository
• Scientific data of water – London South Bank University