THE SOLUBILITY OF IRON IN FOREST SOILS

Piotr Gruba
Department of Forest Soil Science, University of Agriculture in Cracow, Poland

ABSTRACT

The objective of this study was to investigate the mechanisms regulating Fe solubility in mountain forest soils from Southern Poland, with particular emphasis on the relationship between Fe concentration and pH. On evaluating the content of Fe in both the soil solution and soil solid phases, it was found that there is a visible relationship between the solution's pH and iron content, with a lack of correlation between the content of Fe in the solution and soil solid phases (neither Fe_p nor Fe_ox). Furthermore, the saturation degree of soil organic matter with iron, a key element in organic matter solubility/mobility, appears to be pH dependent.

Key words: iron, forest soil, Beskidy Mountains.

INTRODUCTION

The multicharged iron ions play a crucial role in the acidification of, and pedogenesis in, soil derived from sandy materials [10]. Due to its importance to soil formation, Fe (pyrophosphate-extractable form) is commonly used for soil classification in line with what is generally referred to as 'Mokma's Index' [11,15]. The overall concentration of iron in soils is usually very high but, due to its highly limited solubility, the concentration of Fe in groundwater or soil solutions only ranges from 0 to 10 mg L^-1 [13]. The iron content of a soil solution is controlled by the availability of its mineral sources, pH and redox potential, as well as by the presence of chelating agents, especially soil organic matter (SOM).

The most common mineral sources of Fe are amorphous Fe(OH)₃, α-FeOOH (lepidocrocite), α-FeOOH (goethite), and α-Fe₂O₃ (hematite) [7]. Solubilisation of these minerals is described with two parameters: solubility constant (log Ks), and stoichiometry coefficient (n). The log Ks varies from -0.02 for goethite to 3.54 for amorphous Fe(OH)₃ [7].

The general relationship between iron solubility and H⁺ activity (pH) is described by the equation:

Fe (OH)₃ + 3H⁺ = Fe³⁺ + 3 H₂O (1);

thus, the stoichiometry constant (n) for most reactions with Fe minerals equals 3.

In very acid soils (pH<4.0), Fe is released to the soil solution from its complexes with SOM [8]. Metal binding to soil organic matter plays an important role in the solubility of organic colloids in soils and in the podzolization process, especially in the flocculation of organic colloids, moving downwards within the soil profile [4,11]. Podzols usual develop on sands, though they also occur on some materials of a finer texture, such as loams. In general, loams are resistant to podzolization. However, the process may still occur in favourable conditions, such as the absence of carbonates, a high level of precipitation and the presence of an acid organic horizon. Such conditions are common in high elevations of the Polish Beskidy Mountains. The soils are derived from the layers of the Carpathian Flysch – sedimentary rocks, mostly sandstones, as well as shales and mudstones. The most frequent soils are cambisols, covered with a rich plant community, Dentario-glandulosae Fagetum, (mixed forest stands including beech and fir). Large areas of the Beskidy mountains are covered by homogeneous spruce forest (alpine belt), usually present on elevations exceeding 1000m above sea level, as well as on those lower elevations where spruce forests were planted in the 19^th century for their commercial value. The growth of pure spruce forest on relatively fertile soils results in the presence of a very acidic organic horizon on the soil's surface [9] and, in some cases, leads to a somewhat pronounced podzolization in the top soil. This is frequently observed in the Barania Gora range, in the Western Beskidy [9,12]. The large area of spruce stands and very acid soils have given rise to many problems, such as forest decline [16].

The objective of this study is to investigate the mechanism driving the solubility of Fe, with emphasis being placed on relationship between the concentration of soluble Fe in soil solution and the content of this element in soil solid phase.

MATERIALS AND METHODS

Investigation sites
The soils investigated are located in the mountains of the Beskid Slaski range of Barania Gora  (49.41°N; 19.10°E; western Carpathians, southern Poland). The A and B horizons were sampled from six soil profiles, four of which were described as Dystric Cambisols and the remaining two as Pozdol profiles  [15]. The soils have developed in a parent material derived from the layers of the Carpathian Flysch. The parent material is a coarse-grained sandstone with an admixture of feldspars, glauconite and muscovite. Weathering of minerals has given rise to a red colouration brought about by the release of iron (Fe) [3]. For the purposes of this study, the soil profiles were paired, each pair consisting of profiles located in relatively close proximity, but comprising soils developed under two different types of forest vegetation, namely homogenous spruce (Picea abies (L.) Karst.), marked as 'S' and mixed-species ('M'), including spruce (Picea abies (L.) Karst.), beech (Fagus sylvatica L.) and fir (Abies alba L.).

Soil solution analysis
Samples of soil solution were collected using a zero-tension lysimeter. The sampling was performed 7–9 times during the period from May to September 2006. No special sample preservation treatment was applied during the collection period. Directly after collection, the samples were filtrated by means of a 0.45 µm Millipore membrane and kept in a refrigerator (2-3°C) for the purpose of analysis. The samples were analyzed in terms of pH (by means of a combination electrode), dissolved organic carbon (DOC) (using a SHIMADZU 3000 DOC analyzer) and total dissolved Fe (by means of the AAS technique).

Soil analysis For each profile, soil samples were taken from the two horizons directly overlying the lysimeter's collectors. The samples were then air dried and sieved (2 mm). The pH of the soil was measured in suspensions of distilled water and 1 M KCl, respectively (1:2.5 mass-to-volume ratio), after a 24-hour equilibration period (pH_H2O, pH_KCl), by means of a combination electrode. The total soil organic C (C_t) was measured using an LECO 2000 CNS analyzer. The contents of organically bound Fe and soil humic substances were determined by the extraction of soil with 0.1 M Na₄P₂O₇ (Fe_p and C_p). The total amorphous forms of Fe were determined using acid oxalate as an extractant (Fe_ox) [2].

Calculations
The software used for the chemical calculations was Visual MINTEQ version 2.53. The total amount of dissolved Fe was employed as an input parameter, applying the assumption that it was Fe³⁺ only. The solution’s ionic strength was calculated on the basis of the elements determined and the dissolved organic carbon (DOC). The parameters of the linear relations shown in the figures below were calculated in accordance with Webster [14].

RESULTS AND DISCUSSION

General description of the soils
The six soil profiles (three pairs) investigated represent two types of soil (Table 1). The profiles in the first pair (located under spruce and mixed forest, respectively) are dystric cambisols [15], the texture being heavy loam, with 20% to 80% of skeleton (sandstone). The second pair of profiles (soil 3 and 4, see Table 1) are podzolic soils; with sandy loam texture. The AE horizons were covered by a thick organic horizon (approximately 10cm in depth). In both the AE and the B horizons, large pieces of bedrock were present. The third pair of profiles are dystric cambisols, again having the texture of loams.

In the cambisol pairs investigated (1-2 and 5-6), the A horizons under homogeneous spruce forest were more acidified than those in mixed forest (Table 2). The reverse is true for the pair of stands with podzol. From the statistical point of view, the set of six profiles is not sufficient to allow comparison of the acidifying influence of spruce and mixed forest. Nevertheless, the data from the investigations carried out by Maciaszek et al. [9] on a large number of profiles in this region confirm that the top soils under spruce forest have a slightly, but significantly, lower pH than the topsoils under mixed forest with beech. The same author also reports that the soil under spruce gathers more organic matter in upper horizons, despite the fact that, overall, the stock of organic carbon in the profiles under spruce and mixed forest is similar.

Iron in the soil solid phase
Iron in the soil solid phase, either Fe_p or Fe_ox, is predominantly associated with soil organic matter and soil pH (Table 2). The determined contents of Fe_p and Fe_ox were highly correlated (R² = 0.95) and close to a 1:1 relationship (not shown in the figure). However, neither Fe_p nor Fe_ox were correlated with the content of soil humic substances (C_p, Table 2) . As this is in parallel with aluminium chemistry, the explanation given for aluminium [5,6] should also hold for Fe. Consequently, and in line with aluminium chemistry, the amount of Fe bound to soil organic matter should depend on the degree of dissociation of its functional groups (mostly carboxylic). Fe_ox is usually assumed to be an amorphous form of Fe; however, as stated above, acid oxalate predominantly extracts Fe from soil organic matter and from amorphous minerals, wherever such minerals exist.

Iron in soil solutions
The pH of the soil solution is consistent with the pH of the soil, the correlation coefficient being 0.91 (Table 2, Table 3, correlation not show in the figure). Thus it can be assumed that the soil solutions collected were close to equilibrium with the soil solid phase. Due to the limited solubility of minerals, the concentrations of the majority of the elements determined in the solutions (Table 3) being investigated were very small, when compared with their amounts in the soil solid phase (Table 2).

All the solutions investigated were undersaturated with goethite. The soils investigated are young and more likely to contain amorphous Fe minerals (not ordered). The data given in the figure (Figure 1) also indicate that samples from 3B, 6A, 5B, 4B 2B, 1A, 6B and 1B are close to being in equilibrium with amorphous ferrihidrite (pK = 3.2, n = 3). As was mentioned in the site description section, the weathered sandstones of the Barania Gora range are reddish, due to the release of Fe. In fact, iron oxides were not detected in the parent material of several profiles presented by Brożek and Zwydak [1], where the main mineral, apart from quartz, was illite.

On the other hand, the samples with pH <4.0 (Figure 1) are undersaturated with ferrhidrite; the solubility of Fe is controlled by exchange reactions with soil organic matter [8].

The relationship between the dissolved Fe and pH_r solution suggest a mechanism similar to the one described for aluminium solubility. The degree to which soil humic substances are saturated with iron ions depends on the outcome of the competition of Fe and H in the protonation/deprotonation of functional groups, predominantly carboxylic. The value of Fe_p/C_p ratio should thus be as pH-dependent as it is for Al [5]. To test this hypothesis, the data were plotted to draw up Figure 2. Indeed, the pattern visible in Figure 2 is in accordance with the results obtained by Gruba and Mulder [5]. By analogy, the Fe solubility seems to be a factor independent of the soil type (in this study, podzolic or cambic) or the type of forest vegetation (in this study, spruce monoculture or mixed forest).

The amount of dissolved Fe was also a factor independent of the Fe fixed in the soil solid phase (with very little correlation between Fe_r and Fe_p or Fe_ox, see Table 4); there is, however, a strong correlation with the pH of the soil or soil solution and the degree to which the soil organic mater is saturated with metal ions.

CONCLUSIONS

The investigation demonstrates that the solubility of iron is characterized by a number of similarities to the chemistry of aluminium, i.e. the amount of soluble Fe³⁺ is correlated with the pH; it occurs irrespective of the amount of Fe in the soil solid phase (Fe_p, Fe_ox); it is released from complexes with soil organic matter during competition with H⁺; it occurs irrespective of soil type or vegetation type.

In soils with pH >4.0, the soluble Fe is close to an equilibrium with amorphous Fe(OH)₃. Wherever the pH is below 4.0 (the A, AE horizons), the Fe solubility is controlled more by the exchange reaction with soil organic matter than by solubility of the minerals.

The soluble Fe reaches quite low concentrations in soil solutions, being, on average, approximately 10 times lower than the concentration of Al. Its role in the precipitation of organic colloids (podzolization process) thus seems to be limited.

ACKNOWLEDGEMENTS

This work was co-financed by the State Committee for Scientific Research (Grant 2P06S 051 27). Laboratory analysis was supported by Polish-Norwegian Research Fund (PNRF-68-AI-1/07).

REFERENCES

1. Brozek S., Zwydak M., 2003. Atlas gleb leśnych Polski [Atlas of forest soils of Poland]. Centrum Informacyjne Lasów Państwowych, Warszawa [in Polish].

2. Burman P., Van Lagen B., Velthorst E.J., 1996. Manual for soil and water analysis. Backhuys Publishers Leiden, 314.

3. Burtan J., 1973. Objaśnienia do szczegółowej mapy geologicznej Polski. Arkusz Wisła. [Explanation to detailed geological map of Poland. Sheet Wisla] 1:50000. Polish Geol. Inst., Warszawa [in Polish].

4. Buurman P., 1984.Carbon/sesquioxide ratios in organic complexes and the transition albic-spodic horizon Journal of Soil Science 35, 254–260.

5. Gruba P., Mulder J., 2008. Relationship between aluminum in soils and soil water in mineral horizons of a range of acid forest soils. Soil Sci. Soc. Am. J., 72, 1150–1157.

6. Johnson E.C., 2002. Exchange properties of acid forest soils of the northern USA. European Journal of Soil Science, 53, 271–282.

7. Lindsay W.L., 1979. Chemical equilibria in soils. Wiley-Interscience, New York, 449.

8. Lindsay W.L., 1991. Iron oxide solubilization by organic matter and its effect on iron solubility. Plant and Soil, 130, 27–34.

9. Maciaszek W., Gruba P., Januszek K., Lasota J., Wanic T., Zwydak M., 2000. Degradacja i regradacja gleb pod wpływem gospodarki leśnej na terenie Żywiecczyzny [Degradation i recovery of soils under the forest management in the Zywiec region]. Wydawnictwa AR w Krakowie, 88 [in Polish].

10. McBride M., 1994. Environmental chemistry of soils. Oxford University Press, NY.

11. Mokma D.L., 1983. New chemical criteria for defining the spodic horizon. Soil Sci. Soc. Am. J. 47, 972–976.

12. Samec P., Vavříček D.,  Bojko J., Zid T. 2010. Soil component vectors in the southwestern Silesian Beskids. EJPAU, 13(2), #14.

13. Sparks D.L., 2003. Environmental soil chemistry. Second edition. Academic Press. San Diego, CA.

14. Webster R. 1997. Regression and functional relations European Journal of Soil Science, 48, 557–566.

15. World Reference Base For Soil Resources. FAO, ISRIC and ISSS, 2006.

16. Wróbel S., 1995. Zakwaszenie Czarnej Wisełki i eutrofizacja zbiornika zaporowego Wisła-Czarne [Acidification of Czarna Wisełka and eutrophication of Wisła-Czarne reservoir.] Centrum Informacji Naukowej, Kraków, 131–142 [in Polish].

Accepted for print: 30.11.2010

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.