|
|
 |
|||||||||||||||||
|
|||||||||||||||||
 |
|||||||||||||||||
| JOURNAL HOME | HELP | CONTACT PUBLISHER | SUBSCRIBE | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Research Paper |
1 Department of Earth Sciences, University of Aarhus, DK-8000 Aarhus C, Denmark, 2 Lehrstuhl für Bodenkunde, Technische Universität München, D-85350 Freising-Weihenstephan, Germany, 3 Institut für Mineralogie und Lagerstättenlehre, RWTH, D-52056 Aachen, Germany, and 4 Department of Physics and Astronomy, University of Aarhus, DK-8000 Aarhus C, Denmark
* E-mail: geopn{at}phys.au.dk
(Received 26 June 2003; revised 4 November 2003)
 |
ABSTRACT |
|---|
 TOP ABSTRACT SITE DESCRIPTIONRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The red soil (Naesset) consists of a 1 m deep patch, ~10 m2 in area. It is one of at least 5–10 isolated spots of limited size which contain hematite and maghemite, and are probably due to local fire.
The four more extended (several hundred m2) red sediments (Salten Skov, Salten, Pot Molle and Laasby) all occur at the feet of slopes and formed from Fe2+-containing seepage water. They contain 25– ~100% Fe oxides, originally consisting of ferrihydrite. The well-drained parts of the deposits are at present all dominated by goethite. Hematite and maghemite were also identified at two sites in the top ~25 cm, one site containing goethite and hematite and probably a little maghemite, and the other goethite and probably hematite. The presence of hematite and particularly maghemite in the surface sediments could be explained by heating of goethite. However, there is no historic or prehistoric evidence of heating activity, and the spatial extension is much wider than that of normal human events. In contrast to the burned soil site, goethite is still present in the upper layers together with a high content of organic matter. Thus, although the present belief is that the deposits formed by heating, site evidence is to the contrary.
KEYWORDS: goethite, hematite, maghemite, heating, XRD, TEM, Mössbauer spectroscopy
Most Danish soils are dominated by colours ranging between 10YR and 7.5YR, with ferrihydrite and goethite being the most common Fe oxides. However, at a number of localities the colours of the materials are as red as 5YR, 2.5YR and 10R (Nørnberg et al. 1991), indicating the presence of hematite (Scheinost & Schwertmann, 1999). Hematite is known from axeric temperate areas in the northern forelands of the Alps (Schwertmann et al., 1982) and known to be related to low water activity of Terre Rosse (Singer et al., 1998). However, Denmark is far north of the line where hematite is normally present in soils (Schwertmann, 1988).
The sites with the strongest red colours are found as spots, often only a few metres across. This type of site is known to be related to prehistoric glass-works, Fe smelters, tree and house burning, charcoal stacks, or ash production. High-intensity fires reach temperatures above 500°C and convert the organic matter completely into a cation-rich white ash, and the yellow Fe oxides into red-brown maghemite and red hematite. The white ash has a strong, short-term effect in raising the pH of the soil to above the zero point of charge of the Fe oxides (pH 7–8), which will enhance the dispersion and thus promote small-particle migration. This effect is known from burned archaeological sites in Denmark, where silt and clay particles have moved downwards after the heating event (J.N. Nielsen, Aalborg Hist. Mus., pers. comm.). A general characteristic for the burned soil sites is that the Fe content is 1–2%, as in nearby unaffected soils.
Another type of red material has so far only been found in areas on Miocene underground deposits in which pyrite is oxidized at the top of the groundwater zone and Fe2+ is brought to the surface by seepage water at the flat feet of slopes. More than ten sites have been found so far, and this type seems to be more widely distributed in Midjutland, Denmark than hitherto known. At the top it consists almost exclusively of Fe oxides with high contents of silt- and clay-size particles, revealing a precipitated sediment rather than a soil. It has a bulk density of <1 g cm–1, i.e. much lower than that of a soil, and appears even in inactive sites as a fluffy deposit. Precipitation of Fe as two-line ferrihydrite close to some of the sites investigated is still active, but this has been almost entirely transformed into a goethite-hematite-maghemite association in the fossil sites investigated here.
|
SITE DESCRIPTION |
|---|
|
TOPABSTRACTSITE DESCRIPTIONRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
), where the mean annual precipitation is ~700 mm , average evapotranspiration about 375 mm, and mean annual temperature 7.5°C. The geology of the Midjutland area is Weichsel moraine and melt-water deposits overlying Tertiary (Miocene) sediments which contain pyrite. In the upper ground-water zone pyrite is oxidized and supplies Fe to the groundwater. At the feet of slopes this groundwater reaches the surface and Fe precipitates. For site descriptions see Table 1.
|
|
radiation and a diffracted-beam graphite monochromator. The samples were step-scanned using increments of 0.02°2
and a counting time of 20 s per step. The naturally occurring quartz was used as an internal standard for peak positioning and instrumental line broadening except for those samples which were low in quartz. In these, Si was added as an internal standard. Transmission electron microscopy (TEM) was carried out with a 200 kV Philips instrument on samples mounted on carbon-coated Cu grids. Mössbauer spectra were measured using a conventional constant acceleration drive system with the samples at temperatures between 300 and 17 K in transmission geometry. Isomer shifts are given with respect to -Fe at room temperature. Magnetic hysteresis properties were measured using a vibration sample magnetometer. |
RESULTS |
|---|
|
TOPABSTRACTSITE DESCRIPTIONRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
) indicating that alkalinity was added to the soil, probably from ashes. The Fed (<2 mm) maximum is in Bs2. This is below the maximum for slightly podzolized soils, which would be in Bs1. Samples Bs1 and Bs2 contain hematite and maghemite but no goethite (Fig. 3). Except for the Salten, sample II, in which Fed (<2 µm) is comparable to Naesset, the Ald/Fed (<63 µm) ratio in the Fe precipitates is much lower than in the soil due to the availability of Al in soil water, while Al is very low in the natural groundwater when Fe precipitates. Only traces of maghemite could be identified in Bs3, where Fed is also much lower compared to the upper horizons, although the colour is still as red as 2.5YR 4/8. No further analyses were necessary for identification of the Fe oxides. All three Naesset samples are very red, for Danish soils, both in Munsell, redness ratings (Table 2) and in CIE colours (Fig. 2), though none of the three samples nor any others in this investigation reaches the hematite part of the diagram.
|
|
|
) reveals appreciable amounts of both hematite and maghemite present in addition to goethite in the upper layer. This is corroborated by a saturation magnetization of 
S = 3.9(1) Am2/kg, which is considerably higher than the saturation magnetization of bulk hematite (S ~0.4 Am2/kg), indicating the presence of a strongly magnetic component. We find that 5.7 wt.% of maghemite (S ~70 Am2/kg) could account for these findings.
|
) show broad asymmetric lines characteristic of small-particle (nanometer-sized) phenomena, showing superparamagnetic relaxation on the timescale of Mössbauer spectroscopy. The spectra have been analysed in terms of three components (Fig. 5). A sextet labelled Gt is assigned to goethite, a sextet labelled Hm/Mh assigned to hematite/maghemite mixtures and a quadrupole-split component labelled Fe(III) assigned to ferric (super-) paramagnetic Fe oxide(s). The sextet components have been analysed with Split-Lorentzian lineshapes (Morris, 1989). Although there is no physical reason for using this lineshape, it simulates the rather asymmetric spectral lines quite well and enables the determination of area fractions of the different components. We do not try to separate the hematite and maghemite components in view of their rather similar hyperfine parameters and the rather broad asymmetric lineshapes. The hyperfine parameters are given in Table 3. The hyperfine parameters of the Gt component are in good agreement with the parameters of poorly-crystalline goethite (Mørup et al., 1983). The quadrupole shift of this component could not be determined unambiguously at room temperature. The quadrupole shift of the Hm/Mh sextet corresponds neither to that of bulk hematite (
EQ 
–0.2 mm/s at room temperature, +0.4 mm/s below the Morin transition) nor that of maghemite (EQ 0.0 mm/s) (Greenwood and Gibb, 1971), but has an intermediate value of –0.14 mm/s. This is consistent with the interpretation that the complex Hm/Mh sextet results from a mixture of these two minerals and that the hematite does not show a Morin transition due to small-particle phenomena (Schroeer and Nininger, 1967). This value of the quadrupole shift would arise from a mixture of the two minerals in the ratio maghemite/(hematite+maghemite) = 0.30(5). Using the spectral area of the Hm/Mh component (42(2)%) and the total Fe concentration (44.0% Fe2O3), we find that this corresponds to 5.5(5) wt.% maghemite, in good agreement with the amount estimated from the saturation magnetization.
|
|
). Magnetic susceptibility data are almost identical to Salten Skov, and together with XRD, confirm the mineralogy.
|
). The lowest layer is again very high in Fed (Table 2) and shows an almost pure goethite pattern with somewhat narrower lines than in the upper layer. The Feo/Fed ratio is higher at this site than at any of the other high-Fe oxide sites. Part of this poorly-crystalline material could be a two-line ferrihydrite, which is hidden in the background of the XRD traces. This is also indicated in the Mössbauer spectra.
|
) have been analysed in terms of two magnetic hyperfine field distributions (Hm/Mh and Gt) and a quadrupole-split doublet (Fe(III)), similar to the Salten Skov sample. At room temperature the spectrum is dominated by the Fe(III) component and a weak (roughly 10% of the spectral area) indication of a broad magnetically-ordered component with maximum magnetic hyperfine splitting of 51(1) T. Such a high hyperfine field is inconsistent with goethite. At 200 K the two sextet components are clearly seen and at 16 K the Hm/Mh component can be detected as a shoulder on the high-velocity flanks of lines 1 and 6 of the Gt spectrum. The spectral area of the Hm/Mh component gives a maximum amount of hematite and maghemite of 7 wt.%. The bulk saturation magnetization of the sample is 0.64(5) Am2/kg, which is higher than the saturation magnetization of bulk hematite, suggesting the presence of 0.9(1) wt.% of maghe-mite in the sample. Such a small amount would easily escape detection by XRD.
|
). However, the goethite I301/I111-ratio in sample I and the colour again indicate the presence of hematite. The saturation magnetization of sample I is 0.18(2) Am2/kg and for sample II 0.018(5) Am2/kg, which indicates that sample I is not pure goethite. About 5% of hematite could be responsible or 0.03% maghemite.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSITE DESCRIPTIONRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The Naesset site is one of a number of small isolated spots which are probably burned sites. Here, only hematite and maghemite are present. The Fe content is at the level of the surrounding soils, and the red particles have moved down into the deep B horizon after burning, probably due to pH increase caused by ashes after the burning. This phenomenon is seen in many archaeological sites after heating of surface horizons (J.N. Nielsen, pers. comm.).
The presence of hematite and maghemite in the four Fe oxide deposits in which goethite is the dominant mineral could be explained by heating of goethite. The strong magnetic phase is not present as discrete particles and cannot be separated in a magnetic field. The TEM images reveal that the particles consist of integrated single-mineral particles (Figs 10 and 11).
|
|
S = 22(4) Am2/kg for the Salten Skov sample I and S >7 Am2/kg for the Pot Molle sample I. However, at Salten Skov I we have a definitive XRD identification of maghemite, and nanophase hematite is also inconsistent with our interpretation of a small amount of maghemite at Pot Molle. Although heating is the most obvious explanation of the red-coloured sites, there are indications that the presence of either hematite or both hematite and maghemite might not have their origin in burning. The four sites all have very high Fe contents, and only the high Fe content areas are red and contain hematite or maghemite. If the high Fe content sites were burned like the Naesset soil, it would be likely that the surrounding low-Fe sites would also have burned red, and this is not the case. It is very unlikely that the material could have been re-sedimented in slope processes after burning while maintaining the low bulk density and not becoming mixed up with a high proportion of silicate material. If this material had been heated, one could have expected a high maghemite content due to the organic matter content, which is a prerequisite of maghemite formation from goethite on heating (van der Marel, 1951; Schwertmann & Heinemann, 1959; Stanjek, 1987; Campbell et al., 1993). However, goethite is the main mineral in all the samples.
Comparison of transmission Mössbauer spectra and surface-sensitive Conversion Electron Mössbauer (CEMS) spectra on annealed (225°C) Salten Skov I and the burned Naesset soil show that both samples are more advanced in the transformation to Hm/Mh at the surfaces of the particles (Gunnlaugsson et al., 2002). This surface transformation is not observed on unannealed samples from Salten Skov I on the scale of CEMS, supporting the conclusion that the Naesset is a burned soil, but that Hm/Mh in Salten Skov I did not form by burning.
Ferrihydrite, which was probably the original precipitate, could have transformed into goethite and hematite in spite of the low mean annual soil temperature and a relatively high humidity as in Denmark. Schwertmann & Murad (1983) describe the effect of pH on the formation of goethite and hematite from ferrihydrite. Their studies show that hematite formation predominantly takes place around neutrality, i.e. at the point of zero charge of the ferrihydrite because the solubility of ferrihydrite is minimal and therefore the dissolution and transformation to goethite is minimal. However, laboratory experiments indicate that hematite could also form in nature at low pH (Cornell & Schwertmann, 2003). This might occur in an environment in which the pzc of ferrihydrite is lowered, e.g. by adsorption of silicate, phosphate and/or simple organic anions (Schwertmann, 1991; Schwertmann & Fechter, 1982; Cornell et al., 1987), although these do not favour the transformation to hematite. It has not been proven that hematite does not form along with goethite, and Schwertmann & Fischer (1966) observed that increasing concentrations of 2-line ferrihydrite promoted hematite formation at the expense of goethite. The molar Fe concentration in water-saturated material of Salten Skov II is ~15 mol/l. This very high level might indicate parallel formation of goethite and hematite, but not that of maghemite.
A possible source of maghemite could be bacterially-produced microcrystallites of magnetite (30 nm). Microorganisms which can successfully couple the oxidation of organic compounds to the reduction of FeIII have been described by Lovley (1990). Their ability to produce extracellular maghemite is much greater (5000-fold) than that of an equivalent biomass of magnetotactic bacteria, which have been identified in soils (Maher & Taylor, 1988; Fassbinder et al., 1990). Ultrafine crystals of magnetite are known to transform into maghemite over years at room temperature (Murad & Schwertmann, 1993). The finer the crystals, the more easily oxidation to maghemite will take place.
So far there is no historic or prehistoric evidence of heating activity, such as pottery, charcoal or burned root systems in the described areas. The spatial extensions of the four high-Fe oxide sites are much wider than those of normal human events (several hundreds to thousands of m2), and in contrast to the burned-soil site, goethite is still present in the upper layers. Annealing experiments on Salten Skov I reveal a surface transformation into Hm/Mh, which is not seen in unannealed samples from nature at Salten Skov. All this is evidence that questions heating, which would be the most obvious explanation.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTSITE DESCRIPTIONRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTSITE DESCRIPTIONRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Barrón V. & Torrent J. (1986) Use of the Kubelka-Munk theory to study the influence of iron oxides on soil colour. Journal of Soil Science, 37, 499–510.
Campbell A.S., Schwertmann U. & Campbell P.A. (1993) Thermal analysis of ferrihydrite. Proceedings of the 10th International Clay Conference, Adelaide, South Australia, Abstracts, p. O-26.
Cornell R.M. & Schwertmann U. (2003) The Iron Oxides. Wiley-VCH, Weinheim, Germany, 664 pp.
Cornell R.M., Giovanoli R. & Schindler P.W. (1987) Effect of silicate species on the transformation of ferrihydrite into goethite and hematite in alkaline media. Clays and Clay Minerals, 35, 21–28.
Fassbinder J.W.E., Stanjek H. & Vali H. (1990) Occurrence of magnetic bacteria in soil. Nature, 343, 161–163.
Greenwood N.N. & Gibb T.C. (1971) Mössbauer Spectroscopy. Chapman & Hall, London, 659 pp.
Gunnlaugsson H.P., Merrison J.P., Mossin L.A., Nørnberg P., Sanden J., Uggerhøj E. & Weyer G. (2002) Can Mössbauer spectroscopy reveal the origin of magnetic soil in Denmark? Hyperfine Interactions, 144/145, 365–370.
Lovley D.R. (1990) Magnetite formation during microbial dissimilatory iron reduction. Pp. 151–166 in: Iron Biominerals (R.B. Frankel and R.P. Blakemore, editors). Plenum Press, New York.
Maher B.A. & Taylor R.M. (1988) Formation of ultrafine-grained magnetite in soils. Nature, 336, 368–370.
Mehra O.P. & Jackson M.L. (1960) Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays and Clay Minerals, 5, 317–327.
Morris R.V, Agresti D.G., Lauer H.V., Newcomb J.A., Shelfer T.D. & Murali A.V. (1989) Evidence for pigmentary hematite on Mars based on optical, magnetic, and Mössbauer studies of superparamagnetic (nanocrystalline) hematite. Journal of Geophysical Research B, 94, 2760–2778.
Murad E. & Schwertmann U. (1993) Temporal stability of a fine-grained magnetite. Clays and Clay Minerals, 41, 111–113.
Mørup S., Madsen M.B., Franck J., Villadsen J. & Koch C.J.W. (1983) A new interpretation of Mössbauer spectra of microcrystalline goethite: ‘Super-ferromagnetism’ or ‘super-spin-glass’ behaviour? Journal of Magnetism and Magnetic Materials, 40, 163–174.
Nørnberg P., Wulf Petersen J. & Koch C.B. (1991) Present day formation of Hematite in Danish soils. Proceedings of the 7th Euroclay Conference, Dresden, pp. 801–804.
Scheinost A.C. & Schwertmann U. (1999) Color identification of iron oxides and hydroxysulfates: use and limitations. Soil Science Society of America Journal, 63, 1463–1471.
Schroeer D. & Nininger R.C. (1967) Morin transition in -Fe2O3 microcrystals. Physical Review Letters, 19, 632–634.
Schwertmann U. (1964) Differenzierung der Eisenoxide des Bodens durch Extraktion mit Ammoniumoxalat-Lösung, Zeitschrift für Pflanzenernährung, Düngung und Bodenkunde, 105, 194–202.
Schwertmann U. (1988) Occurrence and formation of iron in various pedoenvironments. Pp. 267–302 in: Iron in Soils and Clay Minerals (J.W. Stucki, B.A. Goodman and U. Schwertmann, editors). Reidel Publishing Company, Dordrecht, The Netherlands/ Boston, USA, NATO ASI Series 217.
Schwertmann U. (1991) Solubility and dissolution of iron oxides. Plant and Soil, 130, 1–25.
Schwertmann U. & Fechter H. (1982) The point of zero charge of natural and synthetic ferrihydrite and its relation to adsorbed silicate. Clay Minerals, 17, 471–476.
Schwertmann U. & Fischer W.R. (1966) Zur Bildung von -FeOOH und -Fe2O3 aus amorphem Eisen(III)-hydroxid. III. Zeitschrift für anorganische und allgemeine Chemie, 346, 137–142.
Schwertmann U. & Heinemann B. (1959) Über das Vorkommen und die Entstehung von Maghemit in nordwestdeutschen Böden. Neues Jahrbuch für Mineralogie Monatshefte, 174–181.
Schwertmann U. & Murad E. (1983) Effect of pH on the formation of goethite and hematite from ferrihydrite. Clays and Clay Minerals, 31, 277–284.
Schwertmann U., Murad E. & Schulze D.G. (1982) Is there Holocene reddening (hematite formation) in soils of axeric temperate areas? Geoderma, 27, 209–223.
Singer A., Schwertmann U. & Friedl J. (1998) Iron oxide mineralogy of Terre Rosse and Rendzinas in relation to their moisture and temperature regimes. European Journal of Soil Science, 49, 385–395.
Stanjek H. (1987) The formation of maghemite and hematite from lepidocrocite and goethite in a Cambisol from Corsica, France. Zeitschrift für Pflanzenernährung und Bodenkunde, 150, 314–318.
Van der Marel H.W. (1951) Gamma ferric oxide in sediments. Journal of Sedimentary Petrology, 21, 12–21.
This article has been cited by other articles:
|
|
 |
|
  R. Zanelli, M. Egli, A. Mirabella, M. Abdelmoula, M. Plotze, and M. Notzli 'BLACK' SOILS IN THE SOUTHERN ALPS: CLAY MINERAL FORMATION AND TRANSFORMATION, X-RAY AMORPHOUS Al PHASES AND Fe FORMS Clays and Clay Minerals, December 1, 2006; 54(6): 703 - 720. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| JOURNAL HOME | HELP | CONTACT PUBLISHER | SUBSCRIBE | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), quadrupole shift/ splitting (
for sextets)), line width (
) and spectral area.
