Ole K. Borggaard

EVALUATION OF THE FREE ENERGY OF FORMATION OF FE(II)-FE(III) HYDROXIDE-SULPHATE (GREEN RUST) AND ITS REDUCTION OF NITRITE*

The green “ferrosic hydroxide”, “ferroso ferric hydroxide”, or “hydromagnetite”, Fe3(OH)8, claimed to have formed during alkalimetric titrations of acid Fe(II)-Fe(III) sulphate solutions and suggested by many workers to be present in anaerobic soils and sediments, is shown by means of X-ray diffraction to be an Fe(II) -Fe(III) hydroxide-sulphate (sulphate interlayered green rust, GRso4). It has the approximate composition [Fe4IIIFe2III(OH)12][SO4 · 3H2O], and structurally belongs to the pyroaurite-sjogrenite group of layered hydroxides. From solution data, the standard free energy of formation of GRso4 is estimated to −4380 ± 4 kJ · mol−1. Using this value, the reduction of NO2− to N2O, N2, or NH4+ by GRso4 are thermodynamically spontaneous processes, both under standard state conditions and conditions which can exist in natural anoxic waters. In aqueous solutions at pH 7 and containing FeSO4, K2SO4, NaNO2, and freshly precipitated poorly ordered ferrihydrite two types of redox reactions between Fe(II) and NO2− are operating: (1) rapid reduction of NO2− when Fe2+ reacts with ferrihydrite to form GRso4 and (2) a slower reaction in which NO2− is reduced by Fe(II) in the GRso4 lattice. The major oxidation product is goethite (α-FeOOH). Thus, it appears that the reduction of NO2− by GRso4 is not kinetically hindered. If present in soils and sediments, green rusts may participate in abiotic reduction of NO2−. Further work is in progress to study the kinetics of reduction of NO3−by GRso4. This reaction also is thermodynamically spontaneous.

The potential of willow for remediation of heavy metal polluted calcareous urban soils

Growth performance and heavy metal uptake by willow (Salix viminalis) from strongly and moderately polluted calcareous soils were investigated in field and growth chamber trials to assess the suitability of willow for phytoremediation. Field uptakes were 2-10 times higher than growth chamber uptakes. Despite high concentrations of cadmium (>/=80 mg/kg) and zinc (>/=3000 mg/kg) in leaves of willow grown on strongly polluted soil with up to 18 mgCd/kg, 1400 mgCu/kg, 500 mgPb/kg and 3300 mgZn/kg, it is unsuited on strongly polluted soils because of poor growth. However, willow proved promising on moderately polluted soils (2.5 mgCd/kg and 400 mgZn/kg), where it extracted 0.13% of total Cd and 0.29% of the total Zn per year probably representing the most mobile fraction. Cu and Pb are strongly fixed in calcareous soils.

Composition and reactivity of DOC in forest floor soil solutions in relation to tree species and soil type

Metal coordinating propertiesof DOC (dissolved organic carbon), and henceits influence on heavy metal release andmineral weathering, depend on the compositionand properties of DOC. Tree species producelitter with different chemical composition anddegradability, and these differences mightinfluence the composition and reactivity of DOCin soil solutions. Accordingly, analysis ofcomposition and reactivity of DOC in soilsolution samples collected by centrifugationfrom 16 forest soil O horizons from the fourtree species beech (Fagus sylvatica L.),oak (Quercus robur L.), grand fir (Abies grandis Lindl.), and Norway spruce (Picea abies (L.) Karst.) on two clayey and twosandy soils were carried out. The compositionand properties of DOC were determined bycapillary zone electrophoresis, acid-basetitration, Cu ion titration, total elementalanalysis, IR and UV spectroscopy, and metalrelease assays. Concentrations of DOC rangedfrom 20 to 163 mM with pH ranging from 4.6 to7.3. Norway spruce produced the highest DOCconcentration, and the lowest pH. Carbon inlow-molecular-weight aliphatic carboxylic acids(LACA) accounted for less than 6% of DOC withformic and acetic acids as the most abundantLACAs. The DOC was cation exchanged and protonsaturated to obtain comparable forms of DOC.Titratable carboxylic acid and phenolic groupswere in the range 51 to 82 and 20 to64 mmol·mol−1 C, respectively, with fewerphenolic groups in grand fir DOC as the onlysignificant difference. Infrared spectra offreeze-dried DOC samples suggest low contentsof aromatic C in the DOC especially from grandfir stands. Stability constants, log K of Cu-DOC complexes, determined by Cu ion titrationof DOC samples with fitting of the data to atwo-site binding model, were in the range 5.63to 6.21 for the strong binding sites and 3.58to 4.10 for the weak sites, but with nosignificant effects of tree species or site.Freeze-dried DOC samples were found to consistof 41 to 45% C, 38 to 49% O, 4.4 to 5.4% Hand 1.2 to 2.0% N and C/N ratios in the range26 to 42. Reactivity of DOC in terms of releaserates of Cd, Cu and Fe cations from a soilsample (flow cell experiments) showed nosignificant differences among DOC samples fromdifferent tree species and soil types.Apparently, only minor differences occur inchemical composition and reactivity ofequivalent concentrations of DOC in forestfloor soil solutions irrespective of origin,i.e. four tree species and two soil types. Soilsolution pH and the concentration of DOCproduced by various tree species are thecritical parameters when distinguishing amongtree species in relation to heavy metal releaseand mineral weathering.

Competitive adsorption and desorption of glyphosate and phosphate on clay silicates and oxides

Abstract Competitive adsorption of glyphosate and phosphate on goethite and gibbsite and on illite, montmorillonite and two kaolinites differing in surface area was evaluated. The results show that glyphosate and phosphate are competing for the adsorption sites, but the degree of competition depends on the adsorbent. On goethite the competition is very much in favour of phosphate, on gibbsite the competition is closer, but still phosphate is favoured, while on illite, montmorillonite and kaolinite the competition is almost equal. The amounts of glyphosate and phosphate, which can be adsorbed also depends on the adsorbent: the oxides adsorb more than the clay silicates. The amount adsorbed on kaolinite was dependent on the specific surface area. Changes in the surface area did not affect the competition between glyphosate and phosphate for adsorption sites. The results indicate that differences among soils of different mineralogical composition regarding the adsorption of glyphosate and phosphate can be expected.

Cadmium and copper release kinetics in relation to afforestation of cultivated soil

Afforestation of cultivated soils causes soil acidification and elevated concentrations of dissolved organic matter (DOC) in the soil solution, and hence, aggravate the risk of heavy metal leaching. The kinetics of cadmium and copper release from an unpolluted arable soil applied with forest floor soil solution was investigated in the laboratory, and the release rates correlated to pH and DOC in solution through log-log equations. The soil solution was isolated from Norway spruce (Picea abies (L.) Karst.) by centrifugation, and the solution passed a cation-exchange column to remove metal cations and to protonate the DOC. Soil samples from an arable Ap horizon were placed in completely mixed flow cells, and influent solutions with 0 to 5 mM DOC were applied. The solution pH was adjusted to achieve effluent pH values in the range 3.6 to 6.9 in the flow cells at steady-state conditions. Cadmium release rates were very low at pH . 5 and increased exponentially as pH decreased to ,5. The release rate was correlated to solution pH in a simple model: log(cadmium release rate) 52 0.21 pH 2 15.28 (R 2 5 0.48), and no significant effect of DOC was observed. The kinetics of copper release from the soil was more complicated with effects of both pH and DOC. In experiments without DOC, the release rate of copper was slightly lower at high pH than at low pH. In experiments above pH 5, the presence of 5 mM DOC in the solution increased the release rate of copper. However, the copper release was retarded by DOC in the range pH 3.8 to 5.0, which coincided with a maximum retention of DOC in the flow cells. The release rate of copper was correlated to solution pH and concentration of DOC, including an interaction of pH and DOC: log(copper release rate) 5 0.86 pH 2 1.26 logDOC 1 0.24 pH z logDOC 2 19.26 (R 2 5 0.60). If the changes in soil chemical conditions after afforestation influence the cadmium and copper release rates in a similar way as observed in the flow cell experiments, then the release rate of cadmium will increase exponentially at soil solution pH , 4.5. The inhibition of copper release by DOC observed at pH 3.8 to 5.0 indicates that copper is retained in the soil by interactions with adsorbed organic matter. Copyright

Low-molecular-weight aliphatic carboxylic acids in soil solutions under different vegetations determined by capillary zone electrophoresis

Concentrations of low-molecular-weight aliphatic carboxylic acids in soil solution were determined by a newly developed capillary zone electrophoresis method. Soil solution samples were collected by centrifugation of soil from the A horizon of a Danish, homogeneous, nutrient-rich Hapludalf in adjacent forested and arable plots. The forested plots of 0.5 ha were 33-year old stands of beech (Fagus sylvatica L.), oak (Quercus robur L.), grand fir (Abies grandis Lindl.), and Norway spruce (Picea abies (L.) Karst.), while sugar beet (Beta vulgaris L.) and winter wheat (Triticum aestivum L.) were the agricultural crops this year. High variability in soil solution concentrations of metal cations (Al, Ca, K, Mg, Na), monocarboxylic acids (formic, acetic, lactic, and valeric acids), and di- and tricarboxylic acids (oxalic, malic, succinic, and citric acids) were found within each plot. Despite this short-range within-plot variability, higher concentrations of di- and tricarboxylic acids were found in the forested soils than in the arable soils. The vegetation seemed to favour some monocarboxylic acids, but the total monocarboxylic acid concentrations showed little relation to the vegetation. Probably due to much less soil water in the Norway spruce plot, the low-molecular-weight aliphatic carboxylic acid concentrations in the samples from that plot were much higher than those found in samples from the other plots. Carbon in low-molecular-weight aliphatic carboxylic acids only accounts for a few percent of dissolved organic carbon, and no general relation was found between carbon in low-molecular-weight aliphatic carboxylic acids and dissolved organic carbon, although the correlation between carbon in di- and tricarboxylic acids and dissolved organic carbon was significant.

Estimation of soil phosphate adsorption capacity by means of a pedotransfer function.

Abstract The capacity of soil to protect the aquatic environment against pollution is an important soil function. Thus, the phosphate adsorption capacity (PAC) is important in predicting the risk of phosphate pollution of the aquatic environment because of overfertilisation and because of reestablishment of former drained and cultivated wetlands. Moreover, PAC is an essential property in assessing the risk of drain and ground water contamination with the widely used glyphosate herbicide, which may compete with phosphate for adsorption sites on soil solids. As aluminium and iron oxides are the main phosphate adsorbents in many soils, PAC can be predicted by pedotransfer functions based on various aluminium and iron oxide fractions such as oxalate-extractable aluminium and iron (Al o , Fe o ) and dithionite-extractable iron (Fe d ). Accordingly, experimentally determined PAC taken as the sum of oxalate-extractable phosphate and the Langmuir maximum of the phosphate adsorption isotherm was found to be well predicted by the pedotransfer function P calc =0.22×Al o +0.12×Fe o +0.02×(Fe d −Fe o ) for a wide range of noncalcareous soils including Alfisols, Entisols, Inceptisols, Oxisols, Spodosols, and Ultisols from various parts of the world. A set of noncalcareous soil samples from Denmark, Ghana, and Tanzania was used in developing this pedotransfer function, while another set of soils from Canada and Tanzania was used in testing the function. While close relationships between experimentally determined and predicted PACs were obtained with these soils, the function failed, however, to predict PAC of two Tanzanian Andisols.

EFFECT OF ENDOPHYTIC FUNGI ON CADMIUM TOLERANCE AND BIOACCUMULATION BY FESTUCA ARUNDINACEA AND FESTUCA PRATENSIS

Endophytic fungi are a group of fungi that live asymptomatically inside plant tissue. These fungi may increase host plant tolerance to biotic and abiotic stresses. The effect of Neotyphodium endophytes in two grass species (Festuca arundinacea and Festuca pratensis) on cadmium (Cd) tolerance, accumulation and translocation has been our main objective. The plants were grown in a hydroponic system under different Cd concentrations (0, 5, 10, and 20 mg L−1) for 6 weeks. They were also grown in soil spiked with different concentrations of Cd (0, 10, 20, and 40 mg kg−1) for 2 months. The results from all Cd treatments showed higher biomass production (12–24%) and higher potential to accumulate Cd in roots (6–16%) and shoots (6–20%) of endophyte-infected plants than endophyte-free plants. Cadmium accumulation by plants indicated that the grasses were capable of Cd hyperaccumulation, a property that was augmented after endophyte infection. Maximum photochemical efficiency of photosystem II (Fv/Fm) revealed that Cd stress was significantly reduced in endophyte-infected plants compared to non-infected ones.

EFFECT OF KCl AND CaCl2 AS BACKGROUND ELECTROLYTES ON THE COMPETITIVE ADSORPTION OF GLYPHOSATE AND PHOSPHATE ON GOETHITE

Competitive adsorption between glyphosate and phosphate on goethite was evaluated. The influence of background electrolyte on the adsorption of glyphosate and phosphate was also investigated by using 0.01 M KCl, 0.1 M KCl and 0.01 M CaCl2 as background electrolytes. Experiments showed that phosphate displaced adsorbed glyphosate from goethite, whereas glyphosate did not displace phosphate. Results also showed that the background electrolyte had a strong effect on phosphate adsorption, but little effect on glyphosate adsorption. Thus, there are differences between the adsorption of glyphosate and phosphate. The study also showed that 0.01 M KCl caused dispersion of goethite, resulting in inefficient filtering, and that phosphate precipitated as calcium phosphates in 0.01 M CaCl2 background electrolyte solutions. The results suggest that 0.1 M KCl is a more suitable background electrolyte to determine competitive adsorption processes involving glyphosate and phosphate.

PHOSPHATE AND GLYPHOSATE ADSORPTION BY HEMATITE AND FERRIHYDRITE AND COMPARISON WITH OTHER VARIABLE-CHARGE MINERALS

Adsorption by synthetic 2-line ferrihydrite and hematite of glyphosate and phosphate, separately and together, was compared with adsorption results for goethite, gibbsite and two kaolinites in order to determine adsorption differences and similarities, in particular competition and phosphate preference, of these variable-charge minerals. Hematite rapidly adsorbed both compounds, while adsorption by ferrihydrite was slow, in particular of glyphosate, probably because of very slow diffusion of the bulky glyphosate molecules into interior sites in ferrihydrite particles. Accordingly, the Langmuir adsorption capacity of glyphosate (GAC) was considerably smaller (1.85 µmol m−2)than GAC for hematite (2.61 µmol m−2). The phosphate adsorption capacities (PAC) for ferrihydrite and hematite were more alike, 2.91 µmol m−2 and 2.85 µmol m−2, respectively. Differences between surface coordination (mono- or bidentate) may also contribute to the observed differences but conflicting information about the nature of the surface complexes makes this a difficult contributary factor to assess. The minerals were found to exhibit great variation in extent of competition and phosphate preference. Little competition and phosphate preference characterized hematite adsorption, while phosphate almost completely outcompeted glyphosate on goethite; ferrihydrite adsorption fell between these extremes. These differences may be attributed to different numbers of common (competitive) and specific (selective) adsorption sites on the three Fe oxides with a decreasing number of common sites in the order: goethite>>ferrihydrite>hematite, i.e. almost all goethite sites are common but with strong phosphate preference, while most hematite sites are specific for either glyphosate or phosphate. Alternatively, the result may be explained by adsorption in more planes, e.g. glyphosate adsorption onto the inner-Helmholtz-plane-adsorbed phosphate. For all six minerals compared, desorption of glyphosate following phosphate addition was found to be significantly correlated with the difference between the amounts of phosphate and glyphosate adsorbed indicating that this difference may be used as a competition index for predicting the influence of phosphate on glyphosate adsorption.