- Email: [email protected]
Comparison of water-soluble and exchangeable forms of Al in acid forest soils
JOURNAL OF
Inorganic Biochemistry Journal of Inorganic Biochemistry 99 (2005) 1788–1795 www.elsevier.com/locate/jinorgbio
Comparison of water-soluble and exchangeable forms of Al in acid forest soils Ondrej Drabek a
a,*
, Lenka Mladkova a, Lubos Boruvka a, Jirina Szakova b, Antonin Nikodem a, Karel Nemecek a
Department of Soil Science and Geology, Czech University of Agriculture in Prague, Kamycka 129, CZ165 21 Prague 6 – Suchdol, Czech Republic b Department of Agrochemistry and Plant Nutrition, Czech University of Agriculture in Prague, Czech Republic Received 31 March 2005; received in revised form 20 June 2005; accepted 27 June 2005 Available online 10 August 2005
Abstract Soil acidification promotes Al release from minerals and parent bedrocks; it also affects Al mobilization and speciation. Speciation of KCl extractable and water-extractable Al in forest soils was done by means of HPLC/IC method. Species Al3+ were the most abundant Al forms in the KCl extracts (around 93%). Prevailing Al forms (more than 70%) in aqueous extracts were þ þ + + Al(X)1+, {i.e., AlðOHÞþ 2 , AlðSO4 Þ , AlF2 , Al(oxalate) , Al(H-citrate) , etc.} species. It is assumed that most of KCl and waterextractable Al is bound in soil sorption complex (i.e., highly dispersed colloidal fraction of the soil solid phase creating negative charge) where majority of Al exists in the form of Al3+ species. The ECEC values, total carbon content and parameters related to soil organic matter composition (N and S content) have apparent effect on Al speciation. The most toxic Al3+ species are more concentrated in the B horizons compared to the A and E horizons. Aqueous extracts simulate Al release to soil solution under normal conditions; it can thus exhibit the actual Al toxicity. On the other hand, KCl extraction describes a potential threat for case of strong disturbance of natural soil conditions. 2005 Elsevier Inc. All rights reserved. Keywords: Aluminium forms; Speciation analysis; High performance liquid chromatography; Forest soils; Acidification
1. Introduction Aluminium (Al) presents an important natural constituent of all soils and its different forms are widely distributed all over the soil profile. Aluminium represents around 8% of the total mineral soil part [1]. Aluminium in most of its forms presents no harm to living organisms. Under certain conditions such as low pH environment, however, Al tends to form toxic species. These species are then potentially toxic to all living organisms including human beings [2]. Toxic effect of * Corresponding author. Tel.: +420 224 382 696/751; fax: +420 234 381 836. E-mail addresses: [email protected] (O. Drabek), boruvka@af. czu.cz (L. Boruvka).
0162-0134/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2005.06.024
certain Al species on different plants is a phenomenon that has been reported by a number of studies (for example [3–7]). The form of soil Al is a key factor of its potential bioavailability and toxicity. Al toxicity to plants qualitatively decreases in this order: polymer Al13 (not in a form of phosphates or silicates), Al3+, þ Al(OH)2+, AlðOHÞþ 2 , AlðOHÞ4 , and AlSO4 (toxicity of þ AlSO4 , however, is not always accepted). Aluminium bound in fluoride or organic complexes and Al(OH)3 is supposed to be non-toxic [8–10]. Even though Al13 seems to be the most toxic Al species, its actual risk in soil environment is probably smaller [11,12]. It can be stated that certain Al species represent a limiting factor for plant growth under acidic conditions [4,6]. Soil acidification presents a serious problem mainly in forest soils located in mountainous regions of the Czech
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
Republic. There are many factors of soil acidification and it is hard to determine their relative share. However, main sources are acidic parent bedrock, soil organic matter (SOM) of acid character, acid deposition and soil leaching due to relatively high precipitations [13,14]. An important role in acidification process is also played by the type of land use, vegetation cover and its management [15,16]. Natural and anthropogenic acidification of soils promotes mineral weathering and thus causes Al release from minerals and parent bedrocks. Progressively released Al (mainly from clay minerals) is then able to occupy more then 70% of exchangeable sorption sites of soil affected by acidification process [17]. As is clear from the above discussed facts and other studies [18,19], a detailed Al speciation is necessary for the risk assessment of its potential toxicity and bioavailability. Combination of selected extracting agents (applied on air dry soil samples) and different analytical techniques for Al speciation can be recommended [20]. To release exchangeable forms of Al from soil sample, KCl solution is used (despite some disagreements, e.g. [21]) most often [20,22–31]. A number of researchers [10,20,32–35] also recommend water with the intention to closely simulate Al speciation in real soil solution. For more detailed Al speciation, chromatographic methods are usually used [36]. Generally, special ionic columns are being used and Al species are then distinguished by their charges [20,32,36–41]. Exploitation of size exclusion chromatography seems to be also promising [36]. The main problem is the identifications of certain Al species and it still lasts, despite enormous scientific progress [20,32,39,42–45]. However, it provides important information on the proportion of Al forms with different toxicity for plants. In soil environment numbers of factors directly influence Al speciation and Al species bioavailability. Important factors are: pH, CEC, SOM composition, ionic strength, environment homogeneity, presence of other cations (such as Ca2+, Mg2+ and K+) and anions (e.g., silicic acids, PO3 4 and F ), soil heterogeneity (vertical and horizontal), climatic factors (precipitation, etc.), vegetation cover (directly influencing SOM composition), land and forest management [46–50]. The aim of this paper is: (1) to compare results of Al speciation in soil extracts with 0.5 M KCl and with water obtained by means of HPLC/IC method; (2) to asses influence of basic soil characteristics on Al speciation in above mentioned extracts; (3) to draw suggestions for further research of Al activity in acidified soils based on obtained information.
2. Materials and methods The Jizera Mountains region was chosen as a model naturally acid area strongly affected by anthropogenic
1789
acidification. Sixty soil samples from Podzols, Cambisols and Leptosols were selected from an original larger set of samples studied in this region [15,51]. Analyzed were 11 samples from organomineral horizon A; 49 samples originated from mineral horizons: 16 samples from E – eluvial; 8 samples from cambic and 25 samples from spodic B horizons. Altitudes of these sites ranged from 400 to 1000 m. Beech (Fagus sylvatica), spruce (Picea abies) forests and grass (Calamagrostis villosa) are the prevailing vegetation cover. Samples were air-dried and passed through 2 mm sieves. Basic soil characteristics were determined by commonly used methods; details are given in [15]. pHH2O and pHKCl were determined potentiometrically [52]. Total contents of C, N and S (C, N, S) were measured by automated analyzer LECO CNS-2000 (MI USA). Effective cation exchange capacity (ECEC) was determined according to Mehlich method with unbuffered 0.1 M BaCl2 extraction solution [53]. At the same time, the concentration of exchangeable base cations (Ca, Mg, K) and exchangeable Al form (Alexch) was determined by means of FAAS (Varian SpectrAA 200). In almost all cases, however, the concentrations of exchangeable Ca, Mg and K were below the detection limit hence this information was not used in statistical analysis. Pseudototal contents of Ca and Mg were measured after soil digestion with aqua regia by means of FAAS (Varian SpectrAA 200) [54]. Air-dried soil samples were extracted by 0.5 M KCl (pH adjusted to 5.8) and by deionized water for Al speciation. In both cases w/v ratio was 1:10 and extracting time was 24 h, under laboratory temperature (approx. 22 C). Extracts were separated from suspension by centrifugation and further purified through chromatography disk filters with pore size 0.45 lm. Speciation of Al forms in 0.5 M KCl and aqueous extracts were determined by means of HPLC/IC method according to [20]. The principle of the method consists in the separation of Al forms according to the value of their positive charge. Final spectroscopy detection in UV range is then enabled by post column derivatization. The HPLC instrument consists of the pump Consta Metric 4100 (LDC Analytical Company) with additional pulse suppressor; automatic sampler Triatlon (Spark); cation column Altech Cation/R SN IC1 818 (Altech); pneumatic sampler RDU 010 (Watrex); T-like conjunction reactor with mixing loop and UVD 200 detector (DeltaChrom). The signal from the detector was processed and stored by chromatographic software CSW (DataApex). Sodium sulphate solution (0.1 M Na2SO4 in 7.5 mM H2SO4, pH about 2.4) was used as the mobile phase. The flow rate was 0.8 mL per minute. The flow rate of the derivatization agent (6 · 104 M Tiron solution in 1 M ammonium acetate solution) was adjusted correspondingly to the flow rate of mobile phase; that means flow rate of approx. 1.6 mL per minute through
1790
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
the detector. Detection of derivatized Al forms was performed online in UV region (310 nm). Standard solutions were prepared from certified Al solution (Analytica) with initial concentration of 1000 mgL1 with respect to extracting agent. The analyses were run under stable laboratory conditions (temperature around 22 C). HPLC/IC method enables to separate Al forms into three different groups according to their charge: þ þ + Al(X)1+ fAlðOHÞ2 , AlðSO4 Þ , AlFþ 2 , Al(oxalate) , Al+ 2+ 2+ 2+ (H-citrate) , etc.}; Al(Y) {Al(OH) , (AlF) , etc.} and Al3+ {Al3+ and transformed hydroxyl Al polymers}. Reference analysis of total Al content extracted by 0.5 M KCl was performed by means of ICP–OES (VARIAN Vista Pro, VARIAN, Australia) under standard conditions. Statgraphics Plus 4.0 for Windows [55] was used to perform correlation and analysis of variance.
3. Results and discussion In the beginning it has to be mentioned that there are not statistically significant differences between content of Al determined according to [22] and the sum of all exchangeable Al species determined by means HPLC/ IC method [20]. However, there are certain differences when ICP–OES as a referential method is used. Results from HPLC/IC and ICP–OES methods show close correlation (r = 0.916 at P < 0.001), but the amount of Al determined by ICP–OES is slightly higher (see Fig. 1). This fact can be explained by the presence of certain strongly complexed Al forms. These species have presumably zero or negative charge and do not react with derivatization agent Tiron (or 8-hydroxyquinoline used by [22]). Obviously, these Al species can be determined by ICP–OES technique but cannot be detected by UV detector. This finding should be taken into account during data processing; mainly if such data are used for Al speciation by means of computer simulation.
Fig. 1. Linear regression model of relationship between Al(KCl) determined by ICP–OES and sum of all forms of Al(KCl) determined by HPLC/IC (R2 = 83.9%).
Water and KCl solution are two most common extracting agents used for Al release from soil sample. Their ionic strength, pH and chemical influence on soil sample are different and thus total Al concentration in aqueous and KCl extracts cannot be compared. However, comparison of Al HPLC/IC speciation in these two extracts reveals some interesting facts useful for Al risk assessment and was never done before. The only similar study [38] reported comparison of Al species determined in soil solution an in CaCl2 soil extract by means of IC. Statistical parameters of speciation results are given in Table 1. Determined amount of Al(X)1+ species ranged from 4.4 to 97.5 mg g1 for aqueous extracts and from 5.8 to 75.6 mg kg1 for the KCl extracts. Amounts of Al(Y)2+ forms ranged from <0.1 mg kg1, which means not detected, to 12.6 mg kg1 for aqueous extracts and from 2.4 to 19.3 mg kg1 for the KCl extracts, respectively. Finally, determined amount of Al3+ species ranged from <0.1 mg kg1, which means not detected, to 11.3 mg kg1 for aqueous extracts and from 159.1 to 1009.3 mg kg1 for the KCl extracts. Despite different chemical nature and extracting power of water and KCl solution, speciation of KCl exchangeable and water-soluble Al forms shows certain similarity, concerning first two peaks (see Figs. 2 and 3). Statistically (paired t test) (Table 2) it was proved that there are small significant differences between forms Al(X)1+ extracted by KCl and Al(X)1+ forms extracted by water. Differences between second peak (forms Al(Y)2+) in both extracts are more apparent. These facts might be explained by the influence of K+ cation on Al(X)1+ complexes. Presumably K+ has an ability to release small share of Al from Al(X)1+ species to solution. This ‘‘free’’ Al is then complexed by ligands present in solution (like organic ligands, F or OH) and determined as a Al(Y)2+. It could be theoretically transformed also to Al3+ species, but this fact was not observed. A similar trend was observed (data not shown) during preliminary tests of KCl extracts ageing. There was a certain shift in speciation. With increasing time of extract storage the content of Al(X)1+ has decreased and the content of Al(Y)2+ increased while the amount of Al3+ remained relatively unchanged. Similar effect of K+ cation on Al release from soil sample treated by KCl describes [56]. Moreover, there is no significant difference between the KCl and aqueous extracts in the sum of Al(X)1+ and Al(Y)2+ forms. This fact supports our hypothesis of influence of K+ cation and thus KCl solution on Al speciation. The third peak, containing the most toxic Al species, was well pronounced in samples extracted by KCl, but very suppressed in aqueous extracts. In some cases this peak was not even detected. Similar findings were reported by [34]. These authors find (by means of capillary electrophoresis – CE) that less than 25% of ICP-detected Al in aqueous
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
1791
Table 1 Basic statistical characteristics of Al forms data set Aqueous extract (mg kg1) 1+
Average Median Std. dev. Minimum Maximum
2+
0.5 M KCl extract (mg kg1) 3+
1+
Al(X)
Al(Y)
Al
Al(X)
32.2 26.4 21.2 4.4 97.5
3.1 2.6 2.0 <0.1 12.6
3.8 3.3 2.4 <0.1 11.3
35.3 29.3 22.4 5.4 103.1
2+
+ Al(Y)
Al(X)1+
Al(Y)2+
Al3+
Al(X)1+ + Al(Y)2+
27.9 24.0 15.7 5.8 75.6
8.5 8.3 3.4 2.4 19.3
509.0 517.1 160.3 159.1 1009.3
36.4 34.2 17.7 11.2 92.8
Fig. 2. Example of typical chromatographic separation of Al forms in the KCl and aqueous extracts. Data originate from two different runs.
extract exists in Al3+ form. In some samples Al3+ was not even detected. Also [57] found, using CE technique, that in aqueous extracts from upper horizons of acid
forest soils less than 14% of total Al exists as ‘‘free’’ Al3+. However, [34] found (by means of CE) that in some samples originating from B horizons up to 60% of ICP-determined Al was present in the form of Al3+ species. Aluminium speciation is schematically described by Fig. 3 and parameters of speciation datasets are given in Table 1; they clearly identify Al3+ forms as the most abundant ones in KCl extracts. Generally, these species have a share around 93% of total Al extracted by KCl and determined by means of HPLC/IC method. In average, 5.2% of total determined Al in the KCl extract exists in the form of Al(X)1+ species; 1.6% exists in the
Table 2 Differences between contents of water and 0.5 M KCl extractable Al forms (values of paired t test) 0.5 M KCl extract /aqueous extract 1+
Al(X) Al(Y)2+ 3+ Al P [Al(X)1+, Al(Y)2+] Fig. 3. Summarized distribution of Al species in the KCl and aqueous extracts (mg kg1).
**,***Significant
respectively.
3.36** 13.34*** 24.55*** 0.75
difference at the probability level 0.01 and 0.001,
1792
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
form of Al(Y)2+ species and 93.2% exists in the form of Al3+ species. Prevailing Al forms in all aqueous extracts are Al(X)1+ species – generally more then 70%. In average, 82.6% of total determined Al in aqueous extract exists in the form of Al(X)1+ species; 8.4% exists in the form of Al(Y)2+ species and 9.0% exists in the form of Al3+ species. These facts are in agreement with findings of [35] who find (using size exclusion chromatography, SE, with UV and ICP–OES detection and fast protein liquid chromatography, FPLC, with ETAAS detection) that 40–55% of total water-soluble Al exist in the form of low molecular mass complexes (Al-citrate, Al-oxalate and AlF2+) and 30–40% of total water-soluble Al is bound to humic substances. Very similar results to our findings were reported by [32], using FPLC–ICP–OES technique, who observed following Al species distribution in aqueous extracts of soil samples: positively charged complexes þ fAlðSO4 Þ and AlFþ 2 g and negatively charged oxalate þ and citrate complexes co-elute with AlðOHÞ2 species 2+ 2+ while AlF co-elutes with Al(OH) species; Al3+ is eluted separately. Researchers [39] found by means of cation exchange FPLC–ETAAS that in aqueous extracts of acid soils AlðOHÞþ 2 is prevailing species while in lysimetric waters from soils is Al bound mainly in humate complexes. There is a strong and significant correlation (r = 0.892, P < 0.001) between Al(X)1+ species extracted by water and the KCl solution. This fact and above discussed observations suggest certain similarity between these species in both extracts. The correlation (r = 0.414, P < 0.01) for Al(Y)2+ species extracted by water and the KCl solution is weaker and less significant. Even less significant and weakest correlation (r = 0.389, P < 0.05) is between Al3+ species in both ex-
tracts. These findings point on increasing differences in the concentrations of Al(Y)2+ and Al3+ species between aqueous and KCl extracts caused by above discussed differences between these two extracting agents; mainly by different pH of final extracts and by exchange reaction between K+ and Al3+ species (bound in soil sorption complex) that occurs during extraction procedure. The soil metal interactions are generally due to the presence of highly dispersed colloidal fraction of the soil solid phase called ‘‘soil sorption complex’’. These soil colloid particles are formed from three kinds of constituents: mineral particles, organic matter and organo– mineral complexes [58]. The crucial feature of the soil colloids is that their surface is negatively charged. Hence, the electric double layer is formed on the surface of the soil colloid particles in contact with soil solution. The soil sorption complex influences chemical and physicochemical properties of the soil including cation exchange capacity, buffering capacity, etc. [59]. The determined values of ECEC and Alexch show that the sorption complex of studied area is saturated by Al from 70% to 90% with negligible base saturation. The same findings were reported in earlier mentioned study [17]. There is a strong correlation (r = 0.922, P < 0.001) between the sum of all KCl extractable Al forms and exchangeable Alexch (from the ECEC determination). All discussed Al forms {Al(X)1+, Al(Y)2+ and Al3+} in both extracts, but mainly in the KCl extracts, closely and positively correlate with determined ECEC values and with Alexch (Table 3). So it can be assumed that most of KCl and water-extractable Al is bound in soil sorption complex. It could be also stated that water compared to KCl solution has not the ability to release major part of Al3+ from soil sorption complex. This suggests that soil water does not release higher share of
Table 3 Correlation coefficients of the relationships between Al forms and other soil characteristics Aqueous extract Al(X)1+ Aqueous extract
0.5 M KCl extract
Al(X)1+ Al(Y)2+ Al3+
0.521*** 0.377**
0.091
Al(X)1+ Al(Y)2+ Al3+
0.892*** 0.437*** 0.599***
0.462*** 0.414** 0.241
Alexch C N S Ca Mg C/N ECEC pHH2O pHKCl *,**,***Significant
0.5 M KCl extract Al(Y)2+ 0.521***
0.713*** 0.477*** 0.406** 0.322* 0.032 0.088 0.053 0.772*** 0.397** 0.357**
0.312* 0.284* 0.293* 0.245 0.103 0.062 0.092 0.389** 0.282* 0.355**
at the probability level 0.05, 0.01, and 0.001, respectively.
Al3+ 0.377** 0.091
Al(X)1+ 0.892*** 0.462*** 0.512***
Al(Y)2+
Al3+
0.437*** 0.414** 0.287*
0.599*** 0.241 0.389**
0.516***
0.670*** 0.500***
0.512*** 0.287* 0.389**
0.516*** 0.670***
0.500***
0.482*** 0.312* 0.195 0.357** 0.152 0.311* 0.204 0.309* 0.061 0.217
0.760*** 0.553*** 0.439*** 0.377** 0.095 0.356** 0.110 0.748*** 0.139 0.024
0.485*** 0.414** 0.375** 0.357** 0.295* 0.236 0.028 0.562*** 0.221 0.015
0.911*** 0.242 0.087 0.044 0.017 0.494*** 0.284* 0.837*** 0.191 0.095
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
Al3+ from soil sorption complex even under acidic conditions. This trend is also apparent in higher content of Al(Y)2+ species in KCl extracts compared to aqueous extracts. These findings, supported by the results of [17], lead to the conclusion that possible disturbances of soil sorption complex could result in a massive release of highly toxic Al species to soil solution. Moreover, [36,60] point on the fact that even small disturbance of existing equilibrium can cause great enlargement of Al mobilization. From this point of view vulnerability of soil sorption complex should be studied with an emphasis on the influence of liming, introducing Ca cation into the soil system. Total carbon content and parameters related to soil organic matter composition (N and S content) show apparent relationship with Al speciation (Table 3). It is a well known fact that soil organic matter is the major pool of soil sorption sites in these soils low in clay content. Thus, the determination of dissolved organic carbon content and organic matter fractionation could help to reveal the influence of soil organic matter on Al speciation. These parameters will be analysed in further study of this area. Nevertheless, we are able to draw some hypotheses. Low molecular mass organic compounds tend to create stable Al complexes [61,62]. Researchers [61,62] found, using ultrafiltration and size exclusion chromatography, that low molecular mass organic acids, especially citric acid, are present at sufficiently high concentrations to complex a significant fraction of Al in soil solution of podzolic soils. Aluminium complexed by organic matter can be released by water or KCl solution. First peak, in both extracts, containing Al(X)1+ species is significantly influenced by soil organic matter content and composition (correlations with C, N, S). It seems that a large portion of Al(X)1+ species exists in the form of low molecular mass organic complexes. This observation is supported by the findings of [63] and [64] who found, using mathematical simulation models, that the major part (more than 85%) of Al in soil solution is organically bound and on average 4– 30% of this Al in B horizons is bound in the form of low molecular mass organic complexes. Moreover, organic Al complexes were directly detected in KCl soil extract by [28] using NMR technique. However, presence of inorganic Al(X)1+ species cannot be omitted, either. þ Nevertheless, content of relatively toxic AlðOHÞ2 species should be minimal because of low pH in the samples and lack of OH anions [65]. So it can be supposed that toxicity of present Al(X)1+ species is minimal. Our hypothesis concerning composition of Al(X)1+ and Al(Y)2+ forms are in agreement with the findings of [40]. These authors studied aqueous soil extracts by means of FPLC–ICP–OES, microculumn chelating ion-exchange chromatography with ETAAS detection and 8-hydroxyquinoline spectrophotometry. Their results indicate that Al exists in aqueous extracts of acid
1793
forest soils predominately as monomeric species bound to organic molecules and partially as Al-sulphate complexes. Relatively high content of Al(X)1+ species in aqueous extracts (see Table 1) expresses mobility of these forms. There is high annual precipitation, so that Al(X)1+ species mobility might present certain danger in the studied area. Similar trends can be observed for Al(Y)2+ species. However, contribution of Al(Y)2+ forms to Al stress is probably very low, if any, because of their small share on the total extracted Al amount. Moreover, the presence of Y ligands means decreased or zero Al toxicity (content of relatively toxic Al(OH)2+ species is expected to be minimal as it was explained above [65]). These statements are indirectly supported by the findings of [66] who performed Al speciation in forest well drainage waters by means of computer model and found that Al organic and Al fluoride complexes are the predominant species while content of toxic Al3+ or Al hydroxyl forms was negligible. Based on our previous work [20], we had expected that pH and soil vegetation cover would have major influence on Al speciation. This assumption turns out to be a little misleading. Effect of pH is suppressed by the fact that the range of pH in the samples is quite narrow. Values of pHH2O vary from 3.3 to 4.2 and pHKCl varies from 2.8 to 4. The effect of soil vegetation cover and another stand factor – altitude on Al speciation was not observed. The possible explanation is that the differences between various forest covers are not as pronounced as differences between forest soils and agricultural soils [20]. Moreover, we found that chemical characteristics of mineral horizons of forest soils were not affected significantly by the type of vegetation cover [16]. Another factor that significantly influences Al speciation and species distribution is the type of soil horizon. Influence of three types of horizons was evaluated: A – organically enriched, E – depleted by podzolization process, and B – cambic or spodic (Table 4). Al(X)1+ species content in the E horizon was significantly lower Table 4 Significant differences between soil horizons (mg kg1; results of ANOVA multiple range tests, 95% LSD) Horizon
Aqueous extract
0.5 M KCl extract
Mean
H.G.
Mean
Al(X)1+
A E B
45.21 23.65 32.71
b a b
32.37 17.28 32.22
b a b
Al(Y)2+
A E B
4.78 2.75 2.61
b a a
8.99 7.34 8.94
a a a
Al3+
A E B
3.11 2.81 4.62
ab a b
438.00 435.76 580.99
a a b
H.G., homogeneous groups.
H.G.
1794
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
compared to A and B horizons for both types of extracts. There were no differences between A and B horizons. Al(Y)2+ species shows different behaviour. These forms, extracted by water, are more abundant in the A horizons; in contrast in the KCl extract there is no significant difference between A and B horizons. These observations suggest that humic or organically enriched horizons could be an environment favourable to creation of organic forms of Al(X)1+ and Al(Y)2+ species. These mobile Al forms are then distributed in the soil profile according to their relative mobility or are exported to streams [67]. Al3+ species, extracted by KCl, are significantly more concentrated in the B horizons, probably due to the podzolization process and high precipitation [49]. Higher concentration of Al3+ species in the B horizons might be one of the factors causing shallow rooting of trees and consequently their vulnerability to strong winds, drought or frost. Similar conclusions have been reported by [68]. Water extracted Al3+ species follow similar trends; however, Al3+ enrichment in the B horizon is not statistically significant in the aqueous extracts.
4. Conclusions Generally, it can be stated that soil sorption complex of studied area is saturated by Al (from 70% to 90%) and has negligible base saturation. Speciation of KCl-extractable Al shows that under certain conditions great amount of highly toxic Al species can be released to soil solution. Comparison of speciation between water-soluble and KCl-extractable Al forms revealed the fact that water has not the ability to release Al3+ species to a larger extent. Aqueous extracts can therefore simulate Al release to soil solution under normal conditions; it can thus evaluate actual Al toxicity. On the other hand, KCl extraction describes potential threat for a case of strong disturbance of soil conditions. The most abundant forms released by water are Al(X)1+ species and we suppose that X represents mainly organic ligands. This fact suggests that actual Al threat in studied ecosystem is not so dramatic. However, disturbance of existing equilibrium can cause massive release of highly toxic Al3+ species to the soil solution. These species thus can endanger ecosystem by their bioavailability or by entering nearby water streams. Speciation of Al in real soil solution could be recommended for further investigation of the studied area. The comparison of Al speciation in soil extracts and in real soil solution could support utilization of aqueous extract as a simulation of soil solution under defined conditions. Research should be also focused on dissolved organic matter determination and characterization.
Acknowledgements This study was supported by the grants No. 526/05/ 0613 of the Czech Science Foundation and No. 1G57073 of the Ministry of Agriculture of the Czech Republic. The help of Richard Drury with language corrections is gratefully acknowledged.
References [1] K. Wedepohl, Handbook of Geochemistry, 13-A-1 – 13-O-1, Springer-Verlag, Berlin, 1969. [2] D.R.C. McLachlan, Environmetrics 6 (1995) 233–275. [3] M. Schaedle, F.C. Thornton, D.J. Raynal, H.B. Tepper, Tree Physiol. 5 (1989) 337–356. [4] V. Horak, J. Dolejskova, A. Hejtmankova, E. Divisova, R. Bures, Rostl. Vyr. 41 (1995) 77–82 (in Czech). [5] J.M. Fuente-Martinez, L. Herrera-Estrella, Adv. Agron. 66 (1999) 103–120. [6] J.J. Comin, J. Barloy, G. Bourrie, F. Trolard, Eur. J. Agron. 11 (1999) 115–122. [7] Z.M. Xie, Z.H. Ye, M.H. Wong, Environ. Int. 26 (2001) 341–346. [8] J.-P. Boudot, T. Becquer, D. Merlet, J. Rouiller, Ann. Sci. For. 51 (1994) 27–51. [9] V. Horak, J. Dolejskova, A. Hejtmankova, Rostl. Vyr. 41 (1995) 239–245 (in Czech). [10] N. Kozuh, R. Milacic, B. Gorenc, Ann. di Chim. 86 (1996) 99– 113. [11] F. Gerard, J.-P. Boudot, J. Ranger, Appl. Geochem. 16 (2001) 513–529. [12] S. Hiradate, N.U. Yamaguchi, J. Inorg. Biochem. 97 (2003) 26– 31. [13] L. Boruvka, J. Kozak, O. Drabek, Rostl. Vyr. 45 (1999) 229–236. [14] J. Hruska, E. Cienciala (Eds.), Long-Term Acidification and Nutritional Degradation of Forest Soils, Ministry of Environment of the Czech Republic, Prague, 2001 (in Czech). [15] L. Mladkova, L. Boruvka, O. Drabek, Plant Soil Environ. 50 (2004) 346–351. [16] L. Mladkova, L. Boruvka, O. Drabek, Soil Sci. Plant Nutr. 51 (2005) 313–322. [17] L. Blake, K.W.T. Goulding, C.J.B. Mott, A.E. Johnston, Eur. J. Soil. Sci. 50 (1999) 401–412. [18] B. Fairman, A. Sanz-Medel, Tech. Inst. Anal. Chem. 17 (1995) 215–233. [19] K. Pyrzynska, S. Gucer, E. Bulska, Wat. Res. 34 (2000) 359–365. [20] O. Drabek, L. Boruvka, L. Mladkova, M. Kocarek, J. Inorg. Biochem. 97 (2003) 8–15. [21] H.J. Percival, K.M. Giddens, R. Lee, J.S. Whitton, Austr. J. Soil Res. 34 (1996) 769–779. [22] B.R. James, C.J. Clark, S.J. Riha, Soil Sci. Soc. Am. J. 47 (1983) 893–897. [23] D.L. McElreath, R.L. Westerman, G.V. Jonson, Commun. Soil Sci. Plant Anal. 23 (1992) 2493–2510. [24] J. Tichy, Environ. Pollut. 93 (1996) 303–312. [25] G. Porebska, J. Mulder, Eur. J. Soil Sci. 47 (1996) 81–87. [26] Q. Ponette, D. Andre, J.E. Dufey, Eur. J. Soil Sci. 47 (1996) 89– 95. [27] S.R. Gurung, R.B. Stewart, P. Loganathan, P.E.H. Gregg, Soil Technol. 9 (1996) 273–279. [28] S. Hiradate, S. Taniguchi, K. Sakurai, Soil Sci. Soc. Am. J. 62 (1998) 630–636. [29] F. Millan, M.M. Hetier, R. Moreau, J. Petard, M. Burguera, Commun. Soil Sci. Plant Anal. 30 (1999) 183–198.
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795 [30] K.H. Dai, D.D. Richter, Commun. Soil Sci. Plant Anal. 31 (2000) 115–139. [31] O.M. Kachurina, H. Zhang, W.R. Raun, E.G. Krenzer, Commun. Soil Sci. Plant Anal. 31 (2000) 893–903. [32] B. Mitrovic, R. Milacic, B. Pihlar, Analyst 121 (1996) 627–634. [33] M.A. Negrin, M.M. Espino, J.M.M. Hernandez, Eur. J. Soil Sci. 47 (1996) 385–393. [34] R. Vogt, G. Matschonat, Z. Pflanzenernahr. Bodenkd. 160 (1997) 549–554. [35] B. Mitrovic, R. Milacic, Sci. Total Environ. 258 (2000) 183– 194. [36] G. Sposito, The Environmental Chemistry of Aluminum, CRC Press LLC, Boca Raton, 1996. [37] I.R. Willett, Soil Sci. Soc. Am. J. 53 (1989) 1385–1391. [38] M.G. Whitten, G.S.P. Ritchie, I.R. Willett, J. Soil Sci. 43 (1992) 283–293. [39] B. Mitrovic, R. Milacic, B. Pihlar, P. Simonovic, Analysis 26 (1998) 381–388. [40] R. Milacic, N. Kozuh, B. Mitrovic, Microchim. Acta 129 (1998) 139–145. [41] K.I. Tsunoda, T. Umemura, K. Ohshima, S.I. Aizawa, E. Yoshimura, K.I. Satake, Water Air Soil Pollut. 130 (2001) 1589–1594. [42] G. Borman, A. Seubert, Anal. Chim. Acta 332 (1996) 233–239. [43] G. Borman, A. Seubert, Anal. Chim. Acta 386 (1999) 77–88. [44] A. Hils, M. Grote, E. Janßen, J. Eichhorn, Anal. Chem. 364 (1999) 457–461. [45] M. Bush, A. Seubert, Fresenius J. Anal. Chem. 366 (2000) 351– 355. [46] H. Sverdrup, P. Warfvinge, K.A.J. Rosen, Water Air Soil Pollut. 61 (1992) 365–383. [47] C.S. Cronan, D.F. Grigal, J. Environ. Qual. 24 (1995) 209–226. [48] C. Exley, J.D. Birchall, Geochim. Cosmochim. Acta 59 (1995) 1017–1018. [49] J.-P. Boudot, O. Maitat, D. Merlet, J. Rouiller, Sci. Total Environ. 184 (1996) 211–214.
1795
[50] S.A. Quideau, O.A. Chadwick, A. Benesi, R.C. Graham, M.A. Anderson, Geoderma 104 (2001) 41–60. [51] L. Boruvka, L. Mladkova, O. Drabek, V. Vasat, Plant Soil Environ. 51 (2005) 447–455. [52] F. Pospisil, in: Transactions of the 5th International Soil Science Conference, vol. 1. Research Institute of Soil Improvement, Prague, 1981, 135–138. [53] E. Podlesakova, J. Nemecek, V. Sirovy, J. Lhotsky, H. Macurova, O. Ivanek, M. Bumerl, O. Hudcova, K. Voplakal, G. Halova´, F. Blahovec, Soil, Water and Plant Analysis, VUMOP, Prague, 1992 (in Czech). [54] J. Zbiral, Analysis of Soils II-Unified Operating Procedures. UKZUZ, Brno, 1996 (in Czech). [55] Manugistics, Statgraphics Plus for Windows User Manual. Manugistics, Inc., Rockville, MD, 1997. [56] R.A. Dahlgren, W.J. Walker, Geochim. Cosmochim. Acta 57 (1993) 57–66. [57] A. Gottlein, Eur. J. Soil Sci. 49 (1998) 107–112. [58] C. Moor, T. Lymberopoulou, V.J. Dietrich, Microchim. Acta 136 (2001) 123–128. [59] S. Zawadski, Gleboznalstwo, PWRIL, Warszawa, 1999. [60] C. Exley, J. Inorg. Biochem. 97 (2003) 1–7. [61] D.L. Jones, A.M. Prabowo, L.V. Kochian, Plant Soil 182 (1996) 229–237. [62] P.A.W. van Hees, U.S. Lundstrom, R. Giesler, Geoderma 94 (2000) 173–200. [63] P.A.W. van Hees, U.S. Lundstrom, Geoderma 94 (2000) 201–221. [64] P.A.W. van Hees, E. Tipping, U.S. Lundstrom, Sci. Total Environ. 278 (2001) 215–229. [65] G.M. Marion, D.M. Hendriks, G.R. Dutt, W.H. Fuller, Soil Sci. 121 (1976) 76–82. [66] S.P. Bi, S.Q. An, M. Yang, T. Chen, Environ. Int. 26 (2001) 327– 380. [67] C.T. Driscoll, N. van Breemen, J. Mulder, Soil Sci. Soc. Am. J. 49 (1985) 437–444. [68] M. Kazda, L. Zvacek, Plant Soil 114 (1989) 257–267.
Inorganic Biochemistry Journal of Inorganic Biochemistry 99 (2005) 1788–1795 www.elsevier.com/locate/jinorgbio
Comparison of water-soluble and exchangeable forms of Al in acid forest soils Ondrej Drabek a
a,*
, Lenka Mladkova a, Lubos Boruvka a, Jirina Szakova b, Antonin Nikodem a, Karel Nemecek a
Department of Soil Science and Geology, Czech University of Agriculture in Prague, Kamycka 129, CZ165 21 Prague 6 – Suchdol, Czech Republic b Department of Agrochemistry and Plant Nutrition, Czech University of Agriculture in Prague, Czech Republic Received 31 March 2005; received in revised form 20 June 2005; accepted 27 June 2005 Available online 10 August 2005
Abstract Soil acidification promotes Al release from minerals and parent bedrocks; it also affects Al mobilization and speciation. Speciation of KCl extractable and water-extractable Al in forest soils was done by means of HPLC/IC method. Species Al3+ were the most abundant Al forms in the KCl extracts (around 93%). Prevailing Al forms (more than 70%) in aqueous extracts were þ þ + + Al(X)1+, {i.e., AlðOHÞþ 2 , AlðSO4 Þ , AlF2 , Al(oxalate) , Al(H-citrate) , etc.} species. It is assumed that most of KCl and waterextractable Al is bound in soil sorption complex (i.e., highly dispersed colloidal fraction of the soil solid phase creating negative charge) where majority of Al exists in the form of Al3+ species. The ECEC values, total carbon content and parameters related to soil organic matter composition (N and S content) have apparent effect on Al speciation. The most toxic Al3+ species are more concentrated in the B horizons compared to the A and E horizons. Aqueous extracts simulate Al release to soil solution under normal conditions; it can thus exhibit the actual Al toxicity. On the other hand, KCl extraction describes a potential threat for case of strong disturbance of natural soil conditions. 2005 Elsevier Inc. All rights reserved. Keywords: Aluminium forms; Speciation analysis; High performance liquid chromatography; Forest soils; Acidification
1. Introduction Aluminium (Al) presents an important natural constituent of all soils and its different forms are widely distributed all over the soil profile. Aluminium represents around 8% of the total mineral soil part [1]. Aluminium in most of its forms presents no harm to living organisms. Under certain conditions such as low pH environment, however, Al tends to form toxic species. These species are then potentially toxic to all living organisms including human beings [2]. Toxic effect of * Corresponding author. Tel.: +420 224 382 696/751; fax: +420 234 381 836. E-mail addresses: [email protected] (O. Drabek), boruvka@af. czu.cz (L. Boruvka).
0162-0134/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2005.06.024
certain Al species on different plants is a phenomenon that has been reported by a number of studies (for example [3–7]). The form of soil Al is a key factor of its potential bioavailability and toxicity. Al toxicity to plants qualitatively decreases in this order: polymer Al13 (not in a form of phosphates or silicates), Al3+, þ Al(OH)2+, AlðOHÞþ 2 , AlðOHÞ4 , and AlSO4 (toxicity of þ AlSO4 , however, is not always accepted). Aluminium bound in fluoride or organic complexes and Al(OH)3 is supposed to be non-toxic [8–10]. Even though Al13 seems to be the most toxic Al species, its actual risk in soil environment is probably smaller [11,12]. It can be stated that certain Al species represent a limiting factor for plant growth under acidic conditions [4,6]. Soil acidification presents a serious problem mainly in forest soils located in mountainous regions of the Czech
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
Republic. There are many factors of soil acidification and it is hard to determine their relative share. However, main sources are acidic parent bedrock, soil organic matter (SOM) of acid character, acid deposition and soil leaching due to relatively high precipitations [13,14]. An important role in acidification process is also played by the type of land use, vegetation cover and its management [15,16]. Natural and anthropogenic acidification of soils promotes mineral weathering and thus causes Al release from minerals and parent bedrocks. Progressively released Al (mainly from clay minerals) is then able to occupy more then 70% of exchangeable sorption sites of soil affected by acidification process [17]. As is clear from the above discussed facts and other studies [18,19], a detailed Al speciation is necessary for the risk assessment of its potential toxicity and bioavailability. Combination of selected extracting agents (applied on air dry soil samples) and different analytical techniques for Al speciation can be recommended [20]. To release exchangeable forms of Al from soil sample, KCl solution is used (despite some disagreements, e.g. [21]) most often [20,22–31]. A number of researchers [10,20,32–35] also recommend water with the intention to closely simulate Al speciation in real soil solution. For more detailed Al speciation, chromatographic methods are usually used [36]. Generally, special ionic columns are being used and Al species are then distinguished by their charges [20,32,36–41]. Exploitation of size exclusion chromatography seems to be also promising [36]. The main problem is the identifications of certain Al species and it still lasts, despite enormous scientific progress [20,32,39,42–45]. However, it provides important information on the proportion of Al forms with different toxicity for plants. In soil environment numbers of factors directly influence Al speciation and Al species bioavailability. Important factors are: pH, CEC, SOM composition, ionic strength, environment homogeneity, presence of other cations (such as Ca2+, Mg2+ and K+) and anions (e.g., silicic acids, PO3 4 and F ), soil heterogeneity (vertical and horizontal), climatic factors (precipitation, etc.), vegetation cover (directly influencing SOM composition), land and forest management [46–50]. The aim of this paper is: (1) to compare results of Al speciation in soil extracts with 0.5 M KCl and with water obtained by means of HPLC/IC method; (2) to asses influence of basic soil characteristics on Al speciation in above mentioned extracts; (3) to draw suggestions for further research of Al activity in acidified soils based on obtained information.
2. Materials and methods The Jizera Mountains region was chosen as a model naturally acid area strongly affected by anthropogenic
1789
acidification. Sixty soil samples from Podzols, Cambisols and Leptosols were selected from an original larger set of samples studied in this region [15,51]. Analyzed were 11 samples from organomineral horizon A; 49 samples originated from mineral horizons: 16 samples from E – eluvial; 8 samples from cambic and 25 samples from spodic B horizons. Altitudes of these sites ranged from 400 to 1000 m. Beech (Fagus sylvatica), spruce (Picea abies) forests and grass (Calamagrostis villosa) are the prevailing vegetation cover. Samples were air-dried and passed through 2 mm sieves. Basic soil characteristics were determined by commonly used methods; details are given in [15]. pHH2O and pHKCl were determined potentiometrically [52]. Total contents of C, N and S (C, N, S) were measured by automated analyzer LECO CNS-2000 (MI USA). Effective cation exchange capacity (ECEC) was determined according to Mehlich method with unbuffered 0.1 M BaCl2 extraction solution [53]. At the same time, the concentration of exchangeable base cations (Ca, Mg, K) and exchangeable Al form (Alexch) was determined by means of FAAS (Varian SpectrAA 200). In almost all cases, however, the concentrations of exchangeable Ca, Mg and K were below the detection limit hence this information was not used in statistical analysis. Pseudototal contents of Ca and Mg were measured after soil digestion with aqua regia by means of FAAS (Varian SpectrAA 200) [54]. Air-dried soil samples were extracted by 0.5 M KCl (pH adjusted to 5.8) and by deionized water for Al speciation. In both cases w/v ratio was 1:10 and extracting time was 24 h, under laboratory temperature (approx. 22 C). Extracts were separated from suspension by centrifugation and further purified through chromatography disk filters with pore size 0.45 lm. Speciation of Al forms in 0.5 M KCl and aqueous extracts were determined by means of HPLC/IC method according to [20]. The principle of the method consists in the separation of Al forms according to the value of their positive charge. Final spectroscopy detection in UV range is then enabled by post column derivatization. The HPLC instrument consists of the pump Consta Metric 4100 (LDC Analytical Company) with additional pulse suppressor; automatic sampler Triatlon (Spark); cation column Altech Cation/R SN IC1 818 (Altech); pneumatic sampler RDU 010 (Watrex); T-like conjunction reactor with mixing loop and UVD 200 detector (DeltaChrom). The signal from the detector was processed and stored by chromatographic software CSW (DataApex). Sodium sulphate solution (0.1 M Na2SO4 in 7.5 mM H2SO4, pH about 2.4) was used as the mobile phase. The flow rate was 0.8 mL per minute. The flow rate of the derivatization agent (6 · 104 M Tiron solution in 1 M ammonium acetate solution) was adjusted correspondingly to the flow rate of mobile phase; that means flow rate of approx. 1.6 mL per minute through
1790
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
the detector. Detection of derivatized Al forms was performed online in UV region (310 nm). Standard solutions were prepared from certified Al solution (Analytica) with initial concentration of 1000 mgL1 with respect to extracting agent. The analyses were run under stable laboratory conditions (temperature around 22 C). HPLC/IC method enables to separate Al forms into three different groups according to their charge: þ þ + Al(X)1+ fAlðOHÞ2 , AlðSO4 Þ , AlFþ 2 , Al(oxalate) , Al+ 2+ 2+ 2+ (H-citrate) , etc.}; Al(Y) {Al(OH) , (AlF) , etc.} and Al3+ {Al3+ and transformed hydroxyl Al polymers}. Reference analysis of total Al content extracted by 0.5 M KCl was performed by means of ICP–OES (VARIAN Vista Pro, VARIAN, Australia) under standard conditions. Statgraphics Plus 4.0 for Windows [55] was used to perform correlation and analysis of variance.
3. Results and discussion In the beginning it has to be mentioned that there are not statistically significant differences between content of Al determined according to [22] and the sum of all exchangeable Al species determined by means HPLC/ IC method [20]. However, there are certain differences when ICP–OES as a referential method is used. Results from HPLC/IC and ICP–OES methods show close correlation (r = 0.916 at P < 0.001), but the amount of Al determined by ICP–OES is slightly higher (see Fig. 1). This fact can be explained by the presence of certain strongly complexed Al forms. These species have presumably zero or negative charge and do not react with derivatization agent Tiron (or 8-hydroxyquinoline used by [22]). Obviously, these Al species can be determined by ICP–OES technique but cannot be detected by UV detector. This finding should be taken into account during data processing; mainly if such data are used for Al speciation by means of computer simulation.
Fig. 1. Linear regression model of relationship between Al(KCl) determined by ICP–OES and sum of all forms of Al(KCl) determined by HPLC/IC (R2 = 83.9%).
Water and KCl solution are two most common extracting agents used for Al release from soil sample. Their ionic strength, pH and chemical influence on soil sample are different and thus total Al concentration in aqueous and KCl extracts cannot be compared. However, comparison of Al HPLC/IC speciation in these two extracts reveals some interesting facts useful for Al risk assessment and was never done before. The only similar study [38] reported comparison of Al species determined in soil solution an in CaCl2 soil extract by means of IC. Statistical parameters of speciation results are given in Table 1. Determined amount of Al(X)1+ species ranged from 4.4 to 97.5 mg g1 for aqueous extracts and from 5.8 to 75.6 mg kg1 for the KCl extracts. Amounts of Al(Y)2+ forms ranged from <0.1 mg kg1, which means not detected, to 12.6 mg kg1 for aqueous extracts and from 2.4 to 19.3 mg kg1 for the KCl extracts, respectively. Finally, determined amount of Al3+ species ranged from <0.1 mg kg1, which means not detected, to 11.3 mg kg1 for aqueous extracts and from 159.1 to 1009.3 mg kg1 for the KCl extracts. Despite different chemical nature and extracting power of water and KCl solution, speciation of KCl exchangeable and water-soluble Al forms shows certain similarity, concerning first two peaks (see Figs. 2 and 3). Statistically (paired t test) (Table 2) it was proved that there are small significant differences between forms Al(X)1+ extracted by KCl and Al(X)1+ forms extracted by water. Differences between second peak (forms Al(Y)2+) in both extracts are more apparent. These facts might be explained by the influence of K+ cation on Al(X)1+ complexes. Presumably K+ has an ability to release small share of Al from Al(X)1+ species to solution. This ‘‘free’’ Al is then complexed by ligands present in solution (like organic ligands, F or OH) and determined as a Al(Y)2+. It could be theoretically transformed also to Al3+ species, but this fact was not observed. A similar trend was observed (data not shown) during preliminary tests of KCl extracts ageing. There was a certain shift in speciation. With increasing time of extract storage the content of Al(X)1+ has decreased and the content of Al(Y)2+ increased while the amount of Al3+ remained relatively unchanged. Similar effect of K+ cation on Al release from soil sample treated by KCl describes [56]. Moreover, there is no significant difference between the KCl and aqueous extracts in the sum of Al(X)1+ and Al(Y)2+ forms. This fact supports our hypothesis of influence of K+ cation and thus KCl solution on Al speciation. The third peak, containing the most toxic Al species, was well pronounced in samples extracted by KCl, but very suppressed in aqueous extracts. In some cases this peak was not even detected. Similar findings were reported by [34]. These authors find (by means of capillary electrophoresis – CE) that less than 25% of ICP-detected Al in aqueous
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
1791
Table 1 Basic statistical characteristics of Al forms data set Aqueous extract (mg kg1) 1+
Average Median Std. dev. Minimum Maximum
2+
0.5 M KCl extract (mg kg1) 3+
1+
Al(X)
Al(Y)
Al
Al(X)
32.2 26.4 21.2 4.4 97.5
3.1 2.6 2.0 <0.1 12.6
3.8 3.3 2.4 <0.1 11.3
35.3 29.3 22.4 5.4 103.1
2+
+ Al(Y)
Al(X)1+
Al(Y)2+
Al3+
Al(X)1+ + Al(Y)2+
27.9 24.0 15.7 5.8 75.6
8.5 8.3 3.4 2.4 19.3
509.0 517.1 160.3 159.1 1009.3
36.4 34.2 17.7 11.2 92.8
Fig. 2. Example of typical chromatographic separation of Al forms in the KCl and aqueous extracts. Data originate from two different runs.
extract exists in Al3+ form. In some samples Al3+ was not even detected. Also [57] found, using CE technique, that in aqueous extracts from upper horizons of acid
forest soils less than 14% of total Al exists as ‘‘free’’ Al3+. However, [34] found (by means of CE) that in some samples originating from B horizons up to 60% of ICP-determined Al was present in the form of Al3+ species. Aluminium speciation is schematically described by Fig. 3 and parameters of speciation datasets are given in Table 1; they clearly identify Al3+ forms as the most abundant ones in KCl extracts. Generally, these species have a share around 93% of total Al extracted by KCl and determined by means of HPLC/IC method. In average, 5.2% of total determined Al in the KCl extract exists in the form of Al(X)1+ species; 1.6% exists in the
Table 2 Differences between contents of water and 0.5 M KCl extractable Al forms (values of paired t test) 0.5 M KCl extract /aqueous extract 1+
Al(X) Al(Y)2+ 3+ Al P [Al(X)1+, Al(Y)2+] Fig. 3. Summarized distribution of Al species in the KCl and aqueous extracts (mg kg1).
**,***Significant
respectively.
3.36** 13.34*** 24.55*** 0.75
difference at the probability level 0.01 and 0.001,
1792
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
form of Al(Y)2+ species and 93.2% exists in the form of Al3+ species. Prevailing Al forms in all aqueous extracts are Al(X)1+ species – generally more then 70%. In average, 82.6% of total determined Al in aqueous extract exists in the form of Al(X)1+ species; 8.4% exists in the form of Al(Y)2+ species and 9.0% exists in the form of Al3+ species. These facts are in agreement with findings of [35] who find (using size exclusion chromatography, SE, with UV and ICP–OES detection and fast protein liquid chromatography, FPLC, with ETAAS detection) that 40–55% of total water-soluble Al exist in the form of low molecular mass complexes (Al-citrate, Al-oxalate and AlF2+) and 30–40% of total water-soluble Al is bound to humic substances. Very similar results to our findings were reported by [32], using FPLC–ICP–OES technique, who observed following Al species distribution in aqueous extracts of soil samples: positively charged complexes þ fAlðSO4 Þ and AlFþ 2 g and negatively charged oxalate þ and citrate complexes co-elute with AlðOHÞ2 species 2+ 2+ while AlF co-elutes with Al(OH) species; Al3+ is eluted separately. Researchers [39] found by means of cation exchange FPLC–ETAAS that in aqueous extracts of acid soils AlðOHÞþ 2 is prevailing species while in lysimetric waters from soils is Al bound mainly in humate complexes. There is a strong and significant correlation (r = 0.892, P < 0.001) between Al(X)1+ species extracted by water and the KCl solution. This fact and above discussed observations suggest certain similarity between these species in both extracts. The correlation (r = 0.414, P < 0.01) for Al(Y)2+ species extracted by water and the KCl solution is weaker and less significant. Even less significant and weakest correlation (r = 0.389, P < 0.05) is between Al3+ species in both ex-
tracts. These findings point on increasing differences in the concentrations of Al(Y)2+ and Al3+ species between aqueous and KCl extracts caused by above discussed differences between these two extracting agents; mainly by different pH of final extracts and by exchange reaction between K+ and Al3+ species (bound in soil sorption complex) that occurs during extraction procedure. The soil metal interactions are generally due to the presence of highly dispersed colloidal fraction of the soil solid phase called ‘‘soil sorption complex’’. These soil colloid particles are formed from three kinds of constituents: mineral particles, organic matter and organo– mineral complexes [58]. The crucial feature of the soil colloids is that their surface is negatively charged. Hence, the electric double layer is formed on the surface of the soil colloid particles in contact with soil solution. The soil sorption complex influences chemical and physicochemical properties of the soil including cation exchange capacity, buffering capacity, etc. [59]. The determined values of ECEC and Alexch show that the sorption complex of studied area is saturated by Al from 70% to 90% with negligible base saturation. The same findings were reported in earlier mentioned study [17]. There is a strong correlation (r = 0.922, P < 0.001) between the sum of all KCl extractable Al forms and exchangeable Alexch (from the ECEC determination). All discussed Al forms {Al(X)1+, Al(Y)2+ and Al3+} in both extracts, but mainly in the KCl extracts, closely and positively correlate with determined ECEC values and with Alexch (Table 3). So it can be assumed that most of KCl and water-extractable Al is bound in soil sorption complex. It could be also stated that water compared to KCl solution has not the ability to release major part of Al3+ from soil sorption complex. This suggests that soil water does not release higher share of
Table 3 Correlation coefficients of the relationships between Al forms and other soil characteristics Aqueous extract Al(X)1+ Aqueous extract
0.5 M KCl extract
Al(X)1+ Al(Y)2+ Al3+
0.521*** 0.377**
0.091
Al(X)1+ Al(Y)2+ Al3+
0.892*** 0.437*** 0.599***
0.462*** 0.414** 0.241
Alexch C N S Ca Mg C/N ECEC pHH2O pHKCl *,**,***Significant
0.5 M KCl extract Al(Y)2+ 0.521***
0.713*** 0.477*** 0.406** 0.322* 0.032 0.088 0.053 0.772*** 0.397** 0.357**
0.312* 0.284* 0.293* 0.245 0.103 0.062 0.092 0.389** 0.282* 0.355**
at the probability level 0.05, 0.01, and 0.001, respectively.
Al3+ 0.377** 0.091
Al(X)1+ 0.892*** 0.462*** 0.512***
Al(Y)2+
Al3+
0.437*** 0.414** 0.287*
0.599*** 0.241 0.389**
0.516***
0.670*** 0.500***
0.512*** 0.287* 0.389**
0.516*** 0.670***
0.500***
0.482*** 0.312* 0.195 0.357** 0.152 0.311* 0.204 0.309* 0.061 0.217
0.760*** 0.553*** 0.439*** 0.377** 0.095 0.356** 0.110 0.748*** 0.139 0.024
0.485*** 0.414** 0.375** 0.357** 0.295* 0.236 0.028 0.562*** 0.221 0.015
0.911*** 0.242 0.087 0.044 0.017 0.494*** 0.284* 0.837*** 0.191 0.095
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
Al3+ from soil sorption complex even under acidic conditions. This trend is also apparent in higher content of Al(Y)2+ species in KCl extracts compared to aqueous extracts. These findings, supported by the results of [17], lead to the conclusion that possible disturbances of soil sorption complex could result in a massive release of highly toxic Al species to soil solution. Moreover, [36,60] point on the fact that even small disturbance of existing equilibrium can cause great enlargement of Al mobilization. From this point of view vulnerability of soil sorption complex should be studied with an emphasis on the influence of liming, introducing Ca cation into the soil system. Total carbon content and parameters related to soil organic matter composition (N and S content) show apparent relationship with Al speciation (Table 3). It is a well known fact that soil organic matter is the major pool of soil sorption sites in these soils low in clay content. Thus, the determination of dissolved organic carbon content and organic matter fractionation could help to reveal the influence of soil organic matter on Al speciation. These parameters will be analysed in further study of this area. Nevertheless, we are able to draw some hypotheses. Low molecular mass organic compounds tend to create stable Al complexes [61,62]. Researchers [61,62] found, using ultrafiltration and size exclusion chromatography, that low molecular mass organic acids, especially citric acid, are present at sufficiently high concentrations to complex a significant fraction of Al in soil solution of podzolic soils. Aluminium complexed by organic matter can be released by water or KCl solution. First peak, in both extracts, containing Al(X)1+ species is significantly influenced by soil organic matter content and composition (correlations with C, N, S). It seems that a large portion of Al(X)1+ species exists in the form of low molecular mass organic complexes. This observation is supported by the findings of [63] and [64] who found, using mathematical simulation models, that the major part (more than 85%) of Al in soil solution is organically bound and on average 4– 30% of this Al in B horizons is bound in the form of low molecular mass organic complexes. Moreover, organic Al complexes were directly detected in KCl soil extract by [28] using NMR technique. However, presence of inorganic Al(X)1+ species cannot be omitted, either. þ Nevertheless, content of relatively toxic AlðOHÞ2 species should be minimal because of low pH in the samples and lack of OH anions [65]. So it can be supposed that toxicity of present Al(X)1+ species is minimal. Our hypothesis concerning composition of Al(X)1+ and Al(Y)2+ forms are in agreement with the findings of [40]. These authors studied aqueous soil extracts by means of FPLC–ICP–OES, microculumn chelating ion-exchange chromatography with ETAAS detection and 8-hydroxyquinoline spectrophotometry. Their results indicate that Al exists in aqueous extracts of acid
1793
forest soils predominately as monomeric species bound to organic molecules and partially as Al-sulphate complexes. Relatively high content of Al(X)1+ species in aqueous extracts (see Table 1) expresses mobility of these forms. There is high annual precipitation, so that Al(X)1+ species mobility might present certain danger in the studied area. Similar trends can be observed for Al(Y)2+ species. However, contribution of Al(Y)2+ forms to Al stress is probably very low, if any, because of their small share on the total extracted Al amount. Moreover, the presence of Y ligands means decreased or zero Al toxicity (content of relatively toxic Al(OH)2+ species is expected to be minimal as it was explained above [65]). These statements are indirectly supported by the findings of [66] who performed Al speciation in forest well drainage waters by means of computer model and found that Al organic and Al fluoride complexes are the predominant species while content of toxic Al3+ or Al hydroxyl forms was negligible. Based on our previous work [20], we had expected that pH and soil vegetation cover would have major influence on Al speciation. This assumption turns out to be a little misleading. Effect of pH is suppressed by the fact that the range of pH in the samples is quite narrow. Values of pHH2O vary from 3.3 to 4.2 and pHKCl varies from 2.8 to 4. The effect of soil vegetation cover and another stand factor – altitude on Al speciation was not observed. The possible explanation is that the differences between various forest covers are not as pronounced as differences between forest soils and agricultural soils [20]. Moreover, we found that chemical characteristics of mineral horizons of forest soils were not affected significantly by the type of vegetation cover [16]. Another factor that significantly influences Al speciation and species distribution is the type of soil horizon. Influence of three types of horizons was evaluated: A – organically enriched, E – depleted by podzolization process, and B – cambic or spodic (Table 4). Al(X)1+ species content in the E horizon was significantly lower Table 4 Significant differences between soil horizons (mg kg1; results of ANOVA multiple range tests, 95% LSD) Horizon
Aqueous extract
0.5 M KCl extract
Mean
H.G.
Mean
Al(X)1+
A E B
45.21 23.65 32.71
b a b
32.37 17.28 32.22
b a b
Al(Y)2+
A E B
4.78 2.75 2.61
b a a
8.99 7.34 8.94
a a a
Al3+
A E B
3.11 2.81 4.62
ab a b
438.00 435.76 580.99
a a b
H.G., homogeneous groups.
H.G.
1794
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795
compared to A and B horizons for both types of extracts. There were no differences between A and B horizons. Al(Y)2+ species shows different behaviour. These forms, extracted by water, are more abundant in the A horizons; in contrast in the KCl extract there is no significant difference between A and B horizons. These observations suggest that humic or organically enriched horizons could be an environment favourable to creation of organic forms of Al(X)1+ and Al(Y)2+ species. These mobile Al forms are then distributed in the soil profile according to their relative mobility or are exported to streams [67]. Al3+ species, extracted by KCl, are significantly more concentrated in the B horizons, probably due to the podzolization process and high precipitation [49]. Higher concentration of Al3+ species in the B horizons might be one of the factors causing shallow rooting of trees and consequently their vulnerability to strong winds, drought or frost. Similar conclusions have been reported by [68]. Water extracted Al3+ species follow similar trends; however, Al3+ enrichment in the B horizon is not statistically significant in the aqueous extracts.
4. Conclusions Generally, it can be stated that soil sorption complex of studied area is saturated by Al (from 70% to 90%) and has negligible base saturation. Speciation of KCl-extractable Al shows that under certain conditions great amount of highly toxic Al species can be released to soil solution. Comparison of speciation between water-soluble and KCl-extractable Al forms revealed the fact that water has not the ability to release Al3+ species to a larger extent. Aqueous extracts can therefore simulate Al release to soil solution under normal conditions; it can thus evaluate actual Al toxicity. On the other hand, KCl extraction describes potential threat for a case of strong disturbance of soil conditions. The most abundant forms released by water are Al(X)1+ species and we suppose that X represents mainly organic ligands. This fact suggests that actual Al threat in studied ecosystem is not so dramatic. However, disturbance of existing equilibrium can cause massive release of highly toxic Al3+ species to the soil solution. These species thus can endanger ecosystem by their bioavailability or by entering nearby water streams. Speciation of Al in real soil solution could be recommended for further investigation of the studied area. The comparison of Al speciation in soil extracts and in real soil solution could support utilization of aqueous extract as a simulation of soil solution under defined conditions. Research should be also focused on dissolved organic matter determination and characterization.
Acknowledgements This study was supported by the grants No. 526/05/ 0613 of the Czech Science Foundation and No. 1G57073 of the Ministry of Agriculture of the Czech Republic. The help of Richard Drury with language corrections is gratefully acknowledged.
References [1] K. Wedepohl, Handbook of Geochemistry, 13-A-1 – 13-O-1, Springer-Verlag, Berlin, 1969. [2] D.R.C. McLachlan, Environmetrics 6 (1995) 233–275. [3] M. Schaedle, F.C. Thornton, D.J. Raynal, H.B. Tepper, Tree Physiol. 5 (1989) 337–356. [4] V. Horak, J. Dolejskova, A. Hejtmankova, E. Divisova, R. Bures, Rostl. Vyr. 41 (1995) 77–82 (in Czech). [5] J.M. Fuente-Martinez, L. Herrera-Estrella, Adv. Agron. 66 (1999) 103–120. [6] J.J. Comin, J. Barloy, G. Bourrie, F. Trolard, Eur. J. Agron. 11 (1999) 115–122. [7] Z.M. Xie, Z.H. Ye, M.H. Wong, Environ. Int. 26 (2001) 341–346. [8] J.-P. Boudot, T. Becquer, D. Merlet, J. Rouiller, Ann. Sci. For. 51 (1994) 27–51. [9] V. Horak, J. Dolejskova, A. Hejtmankova, Rostl. Vyr. 41 (1995) 239–245 (in Czech). [10] N. Kozuh, R. Milacic, B. Gorenc, Ann. di Chim. 86 (1996) 99– 113. [11] F. Gerard, J.-P. Boudot, J. Ranger, Appl. Geochem. 16 (2001) 513–529. [12] S. Hiradate, N.U. Yamaguchi, J. Inorg. Biochem. 97 (2003) 26– 31. [13] L. Boruvka, J. Kozak, O. Drabek, Rostl. Vyr. 45 (1999) 229–236. [14] J. Hruska, E. Cienciala (Eds.), Long-Term Acidification and Nutritional Degradation of Forest Soils, Ministry of Environment of the Czech Republic, Prague, 2001 (in Czech). [15] L. Mladkova, L. Boruvka, O. Drabek, Plant Soil Environ. 50 (2004) 346–351. [16] L. Mladkova, L. Boruvka, O. Drabek, Soil Sci. Plant Nutr. 51 (2005) 313–322. [17] L. Blake, K.W.T. Goulding, C.J.B. Mott, A.E. Johnston, Eur. J. Soil. Sci. 50 (1999) 401–412. [18] B. Fairman, A. Sanz-Medel, Tech. Inst. Anal. Chem. 17 (1995) 215–233. [19] K. Pyrzynska, S. Gucer, E. Bulska, Wat. Res. 34 (2000) 359–365. [20] O. Drabek, L. Boruvka, L. Mladkova, M. Kocarek, J. Inorg. Biochem. 97 (2003) 8–15. [21] H.J. Percival, K.M. Giddens, R. Lee, J.S. Whitton, Austr. J. Soil Res. 34 (1996) 769–779. [22] B.R. James, C.J. Clark, S.J. Riha, Soil Sci. Soc. Am. J. 47 (1983) 893–897. [23] D.L. McElreath, R.L. Westerman, G.V. Jonson, Commun. Soil Sci. Plant Anal. 23 (1992) 2493–2510. [24] J. Tichy, Environ. Pollut. 93 (1996) 303–312. [25] G. Porebska, J. Mulder, Eur. J. Soil Sci. 47 (1996) 81–87. [26] Q. Ponette, D. Andre, J.E. Dufey, Eur. J. Soil Sci. 47 (1996) 89– 95. [27] S.R. Gurung, R.B. Stewart, P. Loganathan, P.E.H. Gregg, Soil Technol. 9 (1996) 273–279. [28] S. Hiradate, S. Taniguchi, K. Sakurai, Soil Sci. Soc. Am. J. 62 (1998) 630–636. [29] F. Millan, M.M. Hetier, R. Moreau, J. Petard, M. Burguera, Commun. Soil Sci. Plant Anal. 30 (1999) 183–198.
O. Drabek et al. / Journal of Inorganic Biochemistry 99 (2005) 1788–1795 [30] K.H. Dai, D.D. Richter, Commun. Soil Sci. Plant Anal. 31 (2000) 115–139. [31] O.M. Kachurina, H. Zhang, W.R. Raun, E.G. Krenzer, Commun. Soil Sci. Plant Anal. 31 (2000) 893–903. [32] B. Mitrovic, R. Milacic, B. Pihlar, Analyst 121 (1996) 627–634. [33] M.A. Negrin, M.M. Espino, J.M.M. Hernandez, Eur. J. Soil Sci. 47 (1996) 385–393. [34] R. Vogt, G. Matschonat, Z. Pflanzenernahr. Bodenkd. 160 (1997) 549–554. [35] B. Mitrovic, R. Milacic, Sci. Total Environ. 258 (2000) 183– 194. [36] G. Sposito, The Environmental Chemistry of Aluminum, CRC Press LLC, Boca Raton, 1996. [37] I.R. Willett, Soil Sci. Soc. Am. J. 53 (1989) 1385–1391. [38] M.G. Whitten, G.S.P. Ritchie, I.R. Willett, J. Soil Sci. 43 (1992) 283–293. [39] B. Mitrovic, R. Milacic, B. Pihlar, P. Simonovic, Analysis 26 (1998) 381–388. [40] R. Milacic, N. Kozuh, B. Mitrovic, Microchim. Acta 129 (1998) 139–145. [41] K.I. Tsunoda, T. Umemura, K. Ohshima, S.I. Aizawa, E. Yoshimura, K.I. Satake, Water Air Soil Pollut. 130 (2001) 1589–1594. [42] G. Borman, A. Seubert, Anal. Chim. Acta 332 (1996) 233–239. [43] G. Borman, A. Seubert, Anal. Chim. Acta 386 (1999) 77–88. [44] A. Hils, M. Grote, E. Janßen, J. Eichhorn, Anal. Chem. 364 (1999) 457–461. [45] M. Bush, A. Seubert, Fresenius J. Anal. Chem. 366 (2000) 351– 355. [46] H. Sverdrup, P. Warfvinge, K.A.J. Rosen, Water Air Soil Pollut. 61 (1992) 365–383. [47] C.S. Cronan, D.F. Grigal, J. Environ. Qual. 24 (1995) 209–226. [48] C. Exley, J.D. Birchall, Geochim. Cosmochim. Acta 59 (1995) 1017–1018. [49] J.-P. Boudot, O. Maitat, D. Merlet, J. Rouiller, Sci. Total Environ. 184 (1996) 211–214.
1795
[50] S.A. Quideau, O.A. Chadwick, A. Benesi, R.C. Graham, M.A. Anderson, Geoderma 104 (2001) 41–60. [51] L. Boruvka, L. Mladkova, O. Drabek, V. Vasat, Plant Soil Environ. 51 (2005) 447–455. [52] F. Pospisil, in: Transactions of the 5th International Soil Science Conference, vol. 1. Research Institute of Soil Improvement, Prague, 1981, 135–138. [53] E. Podlesakova, J. Nemecek, V. Sirovy, J. Lhotsky, H. Macurova, O. Ivanek, M. Bumerl, O. Hudcova, K. Voplakal, G. Halova´, F. Blahovec, Soil, Water and Plant Analysis, VUMOP, Prague, 1992 (in Czech). [54] J. Zbiral, Analysis of Soils II-Unified Operating Procedures. UKZUZ, Brno, 1996 (in Czech). [55] Manugistics, Statgraphics Plus for Windows User Manual. Manugistics, Inc., Rockville, MD, 1997. [56] R.A. Dahlgren, W.J. Walker, Geochim. Cosmochim. Acta 57 (1993) 57–66. [57] A. Gottlein, Eur. J. Soil Sci. 49 (1998) 107–112. [58] C. Moor, T. Lymberopoulou, V.J. Dietrich, Microchim. Acta 136 (2001) 123–128. [59] S. Zawadski, Gleboznalstwo, PWRIL, Warszawa, 1999. [60] C. Exley, J. Inorg. Biochem. 97 (2003) 1–7. [61] D.L. Jones, A.M. Prabowo, L.V. Kochian, Plant Soil 182 (1996) 229–237. [62] P.A.W. van Hees, U.S. Lundstrom, R. Giesler, Geoderma 94 (2000) 173–200. [63] P.A.W. van Hees, U.S. Lundstrom, Geoderma 94 (2000) 201–221. [64] P.A.W. van Hees, E. Tipping, U.S. Lundstrom, Sci. Total Environ. 278 (2001) 215–229. [65] G.M. Marion, D.M. Hendriks, G.R. Dutt, W.H. Fuller, Soil Sci. 121 (1976) 76–82. [66] S.P. Bi, S.Q. An, M. Yang, T. Chen, Environ. Int. 26 (2001) 327– 380. [67] C.T. Driscoll, N. van Breemen, J. Mulder, Soil Sci. Soc. Am. J. 49 (1985) 437–444. [68] M. Kazda, L. Zvacek, Plant Soil 114 (1989) 257–267.