Vanadium and other elements in Greenland ice cores
M. M. Herron, C. C. Langway Jr, H. V. Weiss, P. Hurley,
R. Kerr and J. H. Cragin
Abstract.
Chemical analysis for Na, Cl, AI, Mn and V of surface snows and deeper ice
core samples from station Milcent, Greenland, indicates a terrestrial or marine origin for these
constituents. Pre-1900 enrichment factors, based on average crustal composition, are high for
Zn and Hg and appear to be related to the volatility of these elements. A comparison of
pre-1900 and 1971-1973 concentrations of V and Hg shows no decided increase due to
industrial production, yet the relative abundance of Zn increased from 12 to 32 over this time
period. The chemical composition of ancient ice is extremely useful in interpreting modern
aerosols.
Vanadium et des autres éléments dans des carottes de glace du Groenland
Résumé.
L'analyse chimique pour Na, Cl, Al, Mn et V des neiges de surface et des
échantillons plus profonds de carottes de glace provenants de la station Milcent, au Groenland,
indique que ces constituants ont une origine et terrestre et marine. Des facteurs d'enrichissement
d'avant 1900, fondés sur une composition de croûte moyenne, sont élevés pour Zn et Hg et
semblent en rapport avec la volatilité de ces éléments. Une comparaison des concentrations de
V et Hg d'avant 1900 et de 1971-1973 ne manifeste aucune augmentation marquée attribuable
à la production industrielle, et pourtant l'abondance relative de Zn a augmenté pendant cette
période de 12 à 32. La composition chimique de la glace ancienne est extrêmement utile à
l'interprétation des aérosols modernes.
The atmospheric concentrations of trace elements, particularly those which may
include an anthropogenic fraction, have been the subject of much recent study
(Zoller et al, 1973; Chester and Stoner, 1974; Zoller et al, 197'4; Duce et al,
1975). The crustal enrichment factor, EF, used to tentatively identify possible
pollutants, is defined as
EFJ5T = WA^SAMPLE
(17A1) CRUST
where W A 1 ) S A M P L E and W A i ) C R U S T are the concentration ratios of the
element X to aluminium in the sample and in average crustal material,
respectively. An element whose aerosol source is primarily crustal weathering
processes should have an EF value near 1. A significantly greater EF indicates
input from local or additional sources, selective volatilization at the source (Zoller
et al, 191 A), selective transport to, or selective removal at the deposition area.
One method of distinguishing naturally high enrichment factors from man's
contribution to the atmospheric burden is the analysis of chemical impurities
within ice cores from the Greenland ice sheet. Pollutant Pb and S have been
identified in modern Greenland snows by comparison with concentrations in
pre-1900 ice strata (Murozumi et al, 1969; Koide and Goldberg, 1971; Weiss
et al, 1975; Cragin et al, 1975). Mercury was identified as a possible pollutant
in north Greenland ice (Weiss et al, 1971a), though work in southern Greenland
disputed this (Weiss et al, 1975). Extreme variations in Hg concentrations in
four north Greenland samples were interpreted as suggesting that the natural
Hg burden was highly variable (Carr and Wilkniss, 1972).
This paper presents the results of a study designed to determine the
elemental abundances of Na, Cl, AI, Mn, V, Zn, and Hg in Greenland ice and
snow over the past several hundred years, and to help shed light on the
significance of modern aerosol analyses.
98
Vanadium and other elements in Greenland ice cores
99
During the 1973 field season of the Greenland Ice Sheet Program a field
camp was established at station Milcent (70°18'N, 44°35'W). A continuous
400-m long 12-cm diameter ice core was recovered and surface pit samples were
collected 2-3 km from the camp (pit 1 was always upwind). The sampler wore a
clean room coat, surgical face mask, hair covering and powder-free polyethylene
gloves throughout the collection, and the pit walls were trimmed back 10 cm
with a precleaned stainless steel shovel. One-year channel samples, as determined
by stratigraphie features, were obtained with a polyethylene scoop and placed in
10-1. polyethylene containers and bagged for shipment frozen to the US. All
bottles and field implements had been precleaned with a leach in 1 % H N 0 3 ,
five rinses with distilled deionized water (DW) and two rinses in double distilled
deionized water (DDW).
In the laboratory, the surface samples were melted in a microwave oven in
their original 10-1. containers and immediately transferred to hot nitric acid
treated (HNAT) one-litre linear polyethylene bottles. These bottles had been
filled with reagent grade HNO s , and heated to 80°C-90°C for periods greater
than 1 h, rinsed five times with DW, twice with DDW and twice with sample
meltwater. This treatment was proven effective in preventing adsorption of 48V
and 203Hg for over 32 d in separate 400-ml aliquots of the pit 4 sample. The
concentrations used were 6.8 ng 48V/kg (5400 cpm) and < 1 ng 203Hg/kg
(2400 cpm).
The ice core samples were cleaned by the 'dry' cleaning procedure (Langway
et al., 191 A), which involves extensive rinsing of the core with DDW in a class
100 clean area. The three deep samples not used for the Hg study were stored
in conventional polyethylene one-litre bottles prepared by leaching with 7 %
Ultrex H N 0 3 for several days at room temperature followed by the usual rinsing
procedure. A concentration of 10 ng 48V/kg was added to a 250-ml aliquot of the
pit 4 sample in one of these bottles and quantitatively recovered after 21-d storage.
Zinc concentrations were measured on the unpreconcentrated samples using
flameless atomic absorption with stopped nitrogen flow during atomization.
Eppendorf pipette tips were found to give erroneously high concentrations until
they were given the hot nitric acid treatment. Mercury concentrations were
measured on one-litre aliquots of the meltwater by neutron activation following
the procedure of Williams et al. (1974). Before radiochemical Hg yields could
be determined, the surface samples were destroyed and a yield of 0.9 was
assumed. Five-millilitre aliquots of the same meltwater were analysed for Na
and CI by instrumental neutron activation analysis (INAA).
The remainder of the samples, 600 ral-21. in volume, were evaporated down
to 20 ml in a precleaned quartz flask in a clean air station. They were transferred
to 50-ml teflon beakers in a glass enclosure under positive filtered N 2 pressure
and evaporated to 5 ml and transferred to HNAT polyethylene irradiation vials
(Weiss et al, 1971b). A spike of 0.2 ng 48V added prior to the preconcentration
was quantitatively recovered. The samples were then analysed for V, Al, and
Mn by INAA. All counting was done with a 75 cm3 Ge(Li) detector coupled to
a 4096 channel analyser and peaks were integrated by computer.
The full data for Milcent snow and deeper ice is presented in Table 1. Each
core sample represents exactly two years' accumulation as determined by
interpretation of the oxygen isotope profile (Dansgaard et al, personal
communication). The elemental concentrations are in /xg/kg.
The Cl:Na ratio of 1.8 throughout the profile matches the ratio of bulk sea
water and confirms the marine origin of these two elements. Based on average
sea-water composition, the contribution of sea spray to the other elements is
minimal.
100
M. M. Herron et al.
Mean crustal enrichment factors using Al as the reference element are
presented in Table 2. The enrichment factors are calculated on the basis of the
mean crustal abundance of Taylor (1964). Also given in Table 2 are enrichment
factors for aerosol collections from Chester and Stoner (1974) for Atlantic
Ocean samples, Duce et al. (1975) for Atlantic Ocean samples, and Zoller et al.
(1974) for South Pole aerosol samples.
TABLE 1.
Elemental concentrations in Milcent snow and ice samples
(all concentrations in fig/kg)
Sample Depth [m]
Age
Na
CI
Al
V
Mn
Zn
Hg
Pit 1
Pit 1
Pit 1
Pit 2
Pit 3
Pit 4
1588
2757
4467
1002
1485
2471
2896
3314
1972-3
1971-2
1971-2
1972-3
1972-3
1972-3
1803-5
1687-9
16003
1864
1815
1717
1675
1631
13
15
9
11
14
12
9
19
14
19
25
26
24
26
14
22
25
23
13
7
7
6
9
18
3
3
7
0.022
0.018
0.008
0.021
0.016
0.010
BD 2
BD 2
0.019
0.215
0.097
0.104
0.125
0.122
0.171
0.075
0.056
0.051
0.215
0.248
0.270
0.211
—
0.174
0.038
0.025
0.070
0.4291
0.3701
0.418 1
0.2901
0.5781
0.881 1
0-1.36
1.36-2.27
1.52-2.75
0-1.51
0-1.37
0-1.38
109.94-111.00
169.43-170.42
250.09-251.00
77.92-78.41
104.49-104.95
155.25-155.69
176.23-176.72
196.50-196.86
0.716
0.318
0.823
0.263
0.445
1
Assumes radiochemical yield of 0.90.
Below detection limit of 6 ng.
3
Estimated age.
2
TABLE 2.
Enrichment factors at Milcent based on average crustal abundances of
Taylor (1964)
Enrichment Factors
Sample
Al
V
Mn
Zn
Hg
Average 1600-1805
Average 1971-1973
South Pole 2
Atlantic Ocean
30°N 3
Atlantic
Westerlies4
1.0
1.0
1.0
1.7
1.1 ± 0.6
1.4
1.5 ± 0.8
1.3 + 0.4
1.4
12 ± 3
32 + 15
69
69 000 1
53 000 ± 11 000
1.0
17
2.6
110
1.0
2.5
1.4
11
1
Assumes average Al concentration of 8.5 /xg/kg in the Milcent core (J. Cragin, personal
communication).
2
From Zoller et al. (1974).
3
From Duce et al. (1975).
4
From Chester and Stoner (1974).
The Hg concentrations are the highest yet reported for Greenland snow and
ice, and there is no indication of significant Hg input in modern times. One
interpretation is that the hot nitric acid treatment of the linear polyethylene
bottles prevented adsorption on the container walls and resulted in more
representative concentrations. Sample contamination is not considered very likely
as samples from Point Barrow, processed and analysed identically to the Milcent
samples, showed low and constant concentrations. The storage containers for
the study of Weiss et al. (1971a) were prepared using a cold nitric acid leaching
Vanadium and other elements in Greenland ice cores
101
(Murozumi et al., 1969) and the samples were stored unacidified for three years
in liquid state prior to analysis for Hg. In the study of Weiss et al. (1975) a cold
acid leach was used and samples were analysed 12 h after melting (Weiss and
Bertine, 1973). Bottle preparation was not discussed in the paper by Carr and
Wilkniss (1972).
The crustal enrichment factors for V and Mn in Milcent ice are near 1 and
do not show any increase in modern deposits. This suggests a common origin
due to crustal weathering. Duce et al. (1975) reported a vanadium EF of 17 for
aerosol samples collected at 30°N over the Atlantic Ocean attributable to the
burning of V-enriched fossil fuels. Zoller et al. (1974), on the other hand, found
a vanadium EF of 1.4 for aerosol samples at the South Pole. The fact that fossil
fuel-derived V does not appear to reach the polar regions may be related to the
relatively low volatility of V compounds.
The pre-1900 zinc EF in Milcent ice is 12, increasing to 32 for 1971-1973
snows. This indicates not only that zinc is selectively volatilized, transported to
and/or precipitated in Greenland, but also that the atmospheric load of Zn has
increased in modern times. The fossil fuel mobilization of V and Zn has been
estimated at 12 and 7 x 109 g/year respectively (Bertine and Goldberg, 1971).
Yet the Milcent evidence suggests that industrial Zn and not V reaches the
Greenland ice sheet.
At present it appears that elemental enrichment factors in pre-1900 Greenland
ice may be explained adequately in terms of crustal abundances and relative
volatilities. Thus Hg and Zn have high EFs while those of V and Mn are near
unity. Apparently the presence or absence of a pollution fraction in modern
snows is also affected by relative volatilities. Efforts to expand the list of
enrichment factors of ancient and modern precipitation are presently underway.
Acknowledgements.
The authors gratefully thank E. D . Goldberg and M. Koide for
laboratory use. This work was supported by National Science Foundation's Office of Polar
Programs.
REFERENCES
Bertine, K. K. and Goldberg, E. D. (1971) Fossil fuel combustion and the major sedimentary
cycle. Science 173, 233-235.
Carr, R. A. and Wilkniss, P. E. (1972) Mercury in the Greenland ice sheet: Further data.
Science 181, 843-844.
Chester, R. and Stoner, J. H. (1974) The distribution of Mn, Fe, Cu, Ni, Co, Ga, Cr, V, Ba,
Sr, Sn, Zn and Pb in some soil-sized particulates from the lower troposphere over the
world ocean. Mar. Chem. 2, 1273-1285.
Cragin, J. H., Herron, M. M. and Langway, C. C. Jr (1975) The chemistry of 700 years of
precipitation at Dye 3, Greenland. US Army Cold Regions Research and Engineering
Laboratory Research Report 74-16.
Duce, R. A., Hoffman, G. L. and Zoller, W. H. (1975) Atmospheric trace metals at remote
northern and southern sites : Pollution or natural ? Science 187, 59-61.
Koide, M. and Goldberg, E. D. (1971) Atmospheric sulphur and fossil fuel combustion. / .
Geophys. Res. 76, 7689-7696.
Langway, C. C. Jr, Herron, M. and Cragin, J. H. (1974) Chemical profile of the Ross Ice
Shelf at Little America V, Antarctica. / . Glaciol. 13, 431-435.
Murozumi, M., Chow, T. J. and Patterson, C. (1969) Chemical concentrations of pollutant
lead aerosols, terrestrial dusts and sea salts in Greenland and Antarctica snow strata.
Geochim. et Cosmoch. Acta 33, 1247-1294.
Taylor, S. R. (1964) Abundance of chemical elements in the continental crust: A new table.
Geochim. et Cosmoch. Acta 28, 1273-1285.
Weiss, H. V., Koide, H. and Goldberg, E. D . (1971a) Mercury in a Greenland Ice Sheet:
Evidence of recent input by man. Science 174, 692-694.
Weiss, H. V., Koide, M. and Goldberg, E. D . (1971b) Selenium and sulphur in a Greenland
Ice Sheet. Science 172, 261-263.
102
M. M. Herron étal.
Weiss, H. V. and Bertine, K. K. (1973) Simultaneous determination of manganese, copper,
arsenic, cadmium, antimony, and mercury in glacial ice by radioactivation. Analyt.
Chimica Acta 65, 253-259.
Weiss, H. V., Bertine, K. K., Koide, M. and Goldberg, E. D. (1975) The chemical composition
of a Greenland glacier. Geochim. et Cosmoch. Acta 39, 1-10.
Williams, P. M., Robertson, K. J., Chew, K. and Weiss, H. V. (1974) Mercury in the south
polar seas and in the northeast Pacific Ocean. Mar. Chem. 2, 287-299.
Zoller, W. H., Gordon, G. E., Gladney, E. S. and Jones, A. G. (1973) The sources and
distribution of vanadium in the atmosphere. Adv. in Chem., series 123, 31-47.
Zoller, W. H., Gladney, E. S. and Duce, R. A. (1974) Atmospheric concentrations and sources
of trace metals at the South Pole. Science 183, 198-200.