41
Figure 11. Vanadium concentration in barley shoots in relation to aqua regia extractable
vanadium concentration in two soils, Pustnäs (left) and Säby (right). The soils were freshly
spiked (×) or aged (∆) with vanadate(V) salt and amended with two blast furnace slags: M-kalk
(◌) and Merit 5000 (●).
Figure 12. Vanadium concentration in barley shoots (plant V) (left) as a function of the aqua
regia-extractable vanadium concentration in soil and (right) as a function of dissolved vanadium
concentration in soil. Data for freshly spiked (×), aged (∆) and BF slag-amended (●) soils. A
linear regression line fitted the whole dataset (n=81), with R
2
=0.50 (left) and R
2
=0.80 (right).
42
5.3 Vanadium speciation - long-term field study (Paper V)
The vanadium concentrations in the forest soil that had received converter lime
additions in the 1980s were highest in the mor layer (Figure 13). The fraction
of recovered vanadium was estimated by comparing the aqua regia-digestible
vanadium for the whole sampling depth against the added vanadium dose. The
recovery was rather low for all converter lime-amended plots, ranging from 25
to 57%. Uncertainties in the distribution of the lime, the amount of vanadium
recovered with aqua regia and vanadium uptake by vegetation made it difficult
to evaluate the vanadium losses. Considering the strong sorption that has been
established for iron (hydr)oxides (Gäbler et al., 2009; Naeem et al., 2007;
Peacock & Sherman, 2004; Blackmore et al., 1996), higher vanadium
concentrations would have been expected in the mineral soil layers with their
relatively high amount of oxalate-extractable iron and aluminium. The
accumulation in the mor layer suggested either an important role of vanadium-
organic complexes, or the presence of large amounts of unweathered converter
lime. Furthermore, uneven distribution during spreading of the lime may have
caused spatial variations.
Vanadium K-edge XANES spectroscopy was combined with HPLC-ICP-
MS analysis to determine the vanadium speciation in the fresh soil samples.
The distribution of different vanadium oxidation states in environmental
samples has been the subject of several studies (Pyrzynska & Wierzbicki,
2004a), but very few have focused on soils and to the best of my knowledge,
this is the first study to apply these two vanadium speciation methods to soil
samples. Vanadium speciation was evaluated in both the sorbed and the
dissolved phases of the different soil horizons to get a better understanding of
soil properties affecting vanadium speciation in soils.
Figure 13. Vanadium distribution in the Ringamåla soil profile. (Left) aqua regia-extractable
vanadium and (right) 0.01 M CaCl
2
-extractable vanadium.
43
In the mor samples, the vanadium K-edge XANES spectra showed a
predominance of vanadium(IV) (Figure 14), despite the fact that the vanadium
in the converter lime was in the form of vanadium(V). According to the LCF
analysis, the added vanadium was mainly sorbed to the organic matter in the
mor (Table 7). For the dissolved vanadium, determined by HPLC-ICP-MS, the
fraction of vanadium(V) generally increased in the mor layer with increasing
converter lime dose (Figure 15). Vanadium(V) is known to be reduced to
vanadium(IV) by humic substances, but as the pH increases the reduction rate
decreases (Lu et al., 1998). The increasing lime dose increased the soil pH.
This may be the reason for the increasing amount of vanadium(V) in the
dissolved phase of the mor layer in the Ringamåla soil.
Table 7. Results of linear combination fit performed on different layers of the Ringamåla forest
soil, which had been treated with 1.0 kg converter lime m
-2
26 years prior to analysis.
Standard (% of V sorbed)
R-factor
Sample
Inherent V
V+OM
V+Fh
V+HAO
Mor
7
70
23
-
0.00031
Mineral soil, 0-10 cm
21
40
39
-
0.00029
Mineral soil, 10-20 cm
74
-
-
26
0.00122
Figure 14. Vanadium K-edge XANES spectra of Ringamåla reference samples (grey lines) and
samples treated with 1.0 kg converter lime m
-2
(black lines). Blue and orange lines are the
standard samples of VO
2+
(aq)
and H
2
VO
4
-
(aq)
, respectively.
44
Figure 15. Vanadium speciation in the dissolved phase of the Ringamåla soil layers. Dissolved
vanadium was extracted with 0. 01 M CaCl
2
.
In the samples of the 0-10 cm mineral soil amended with converter lime, the
pre-edge peak and E
1/2
of the XANES spectra showed a mixture of
vanadium(IV) and vanadium(V) (Figure 14). For respective samples in the 10-
20 cm layer, the pre-edge peak intensity and the E
1/2
were more similar to the
standard of vanadium(IV). As indicated by the LCF results, the reason for this
difference was the relative concentration of inherent vanadium in the two
layers (Table 7). The inherent vanadium in the mineral soil was represented by
vanadium(IV), which is reported to be located in the octahedral layers of clay
minerals (Mosser et al., 1996; Schosseler & Gehring, 1996; Gehring et al.,
1993). The non-inherent vanadium in the mineral soil contained a larger
fraction of vanadium(V) in comparison to the mor layer. This was due to
sorption to iron and aluminium (hydr)oxides, which involves vanadium(V)
surface complexes (Burke et al., 2013; Peacock & Sherman, 2004).
For the dissolved vanadium in the mineral soil, samples amended with
converter lime consisted mainly of vanadium(V) (Figure 15). This was
probably related to the soil pH and the relatively low concentration of
dissolved organic matter. The reference samples contained only vanadium(IV).
Since the dissolved vanadium concentration in the mineral soil was very low in
the reference samples, it is possible that its speciation was controlled by
vanadyl(IV) complexed to dissolved organic matter. However, the vanadium
speciation in the different soil layers appeared to be controlled by the soil
properties, and not by the oxidation state of vanadium added to the soil.
Dostları ilə paylaş: |