The sorption of organic carbon onto differing clay minerals in the presence and absence of hydrous iron oxide

Geoderma 209-210 (2013) 15–21

Contents lists available at SciVerse ScienceDirect

Geoderma journal homepage: www.elsevier.com/locate/geoderma

The sorption of organic carbon onto differing clay minerals in the presence and absence of hydrous iron oxide A.R. Saidy a,d,⁎, R.J. Smernik a, J.A. Baldock b, K. Kaiser c, J. Sanderman b a

School of Agriculture, Food & Wine and Waite Research Institute, Waite Campus, The University of Adelaide, Urrbrae, SA 5064, Australia CSIRO Land and Water, Private Bag 2, Glen Osmond, SA 5064, Australia Soil Sciences, Martin Luther University Halle-Wittenberg, von-Seckendorff-Platz 3, 06120 Halle, Germany d Faculty of Agriculture, Lambung Mangkurat University, Banjarbaru 70714, Indonesia b c

a r t i c l e

i n f o

Article history: Received 31 May 2012 Received in revised form 23 May 2013 Accepted 28 May 2013 Available online xxxx Keywords: Soil carbon stabilisation Surface areas Clay mineralogy Binding mechanisms Electrostatic interaction

a b s t r a c t Sorption of organic carbon onto phyllosilicate clays and hydrous iron oxides influences the accumulation and stabilisation of organic carbon in soils. However, the effects of interactions between hydrous iron oxides and phyllosilicate clays on the sorption of dissolved organic carbon (DOC) are poorly understood. We carried out a batch experiment to examine the effects of goethite coatings on kaolinite, illite, and smectite on DOC sorption. The effect of coating illitic clay with different hydrous iron oxides (haematite, goethite, ferrihydrite) on DOC sorption was studied in another experiment. Organic matter extracted from dried medic (Medicago truncatula cv. Praggio) shoot residue was reacted with minerals at DOC concentrations ranging from 0 to 200 mg C L−1 at pH 6.0. The maximum adsorption capacity (Qmax) of phyllosilicate clays, as determined from fits to the Langmuir equation, increased in the order kaolinite b illite b smectite on a mass basis and illite b smectite b kaolinite on a surface area basis. The sorption capacity of kaolinitic clay increased significantly with goethite coating, whereas the sorption capacity of illitic and smectitic clays was not affected by goethite coating. Ferrihydrite coating increased the sorption capacity of the illitic clays, while haematite coating decreased the sorption capacity; goethite-coated illitic clays had a sorption capacity similar to pure illitic clays. Desorption experiments resulted in the removal of 6–14% of the sorbed DOC. The presence of goethite reduced desorption from kaolinitic clays but did not influence desorption from illitic and smectitic clays. The results suggest that interactions of hydrous iron oxides and phyllosilicate clays can modify DOC sorption and desorption, probably by affecting the surface charges. Therefore, sorption and desorption of organic matter from soils may vary with mineral assemblage, with increasing suppression of the contribution of hydrous oxides at circumneutral to slightly alkaline soil reaction. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Sorption of organic matter (OM) onto mineral surfaces is important in a variety of natural environments. The sorption of OM can alter the physicochemical properties of minerals (Angove et al., 2002; Wang and Xing, 2005) and influence the transport and availability of nutrients and contaminants in soils and waters (Xing, 2001). Generally, most OM sorbed to mineral surfaces is hard to remove (Butman et al., 2007; Kahle et al., 2004; Kaiser and Guggenberger, 2007), indicating a high stability of OM-mineral associations. In addition, OM sorbed to clay minerals and oxides decomposes more slowly and to lesser extent than OM either dissolved or not attached to minerals (Kalbitz et al., 2005; Mikutta et al., 2007; Schneider et al., 2010). This suggests that sorption processes influence the accumulation and stabilisation of organic carbon in soils. ⁎ Corresponding author at: School of Agriculture, Food & Wine and Waite Research Institute, Waite Campus, The University of Adelaide, Urrbrae, SA 5064, Australia. Tel.: +61 8 83037436; fax: +61 8 83036511. E-mail address: [email protected] (A.R. Saidy). 0016-7061/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.geoderma.2013.05.026

Clay minerals and oxides play an important role in the sorption of OM in soils. Hydrous iron oxides have been found to be particularly effective in sorbing and stabilising OM in soils (Kaiser and Guggenberger, 2000, 2003; Kaiser et al., 2007). Phyllosilicates or clay minerals have also been shown to be involved in preservation of OM (Balcke et al., 2002; Feng et al., 2005; Kahle et al., 2004). However, the capability of clay minerals to sorb OM is generally less than that of oxides (Chorover and Amistadi, 2001; Kaiser and Guggenberger, 2003). Tombácz et al. (2004), for example, observed that iron oxides (haematite and magnetite) adsorbed more humic acids than clay minerals (kaolinite and montmorillonite). In addition, Meier et al. (1999) found that the maximum amount of sorption of dissolved organic carbon (DOC) to goethite at pH 4.0 was 0.25–0.3 mg C m−2, whereas kaolinite was only able to sorb 0.08–0.10 mg C m−2 under the same conditions. Acidic conditions favour the sorption of DOC to hydrous oxides; their sorption capacity drops with increasing pH (e.g., Gu et al., 1994). Therefore, in agricultural soils, typically being not strongly acidic, the contribution of hydrous oxides to OM binding might be less dominant than in strongly acidic forest soils.

16

A.R. Saidy et al. / Geoderma 209-210 (2013) 15–21

The effect of interactions between clay minerals and hydrous oxides on the preservation of organic carbon has received little attention. Hydrous oxides may attach to both clay minerals and organic compounds (Ohtsubo, 1989; Tombácz et al., 2004) to form clay–oxide-organic associations, which may significantly influence the capacity of soils to adsorb DOC. In some studies, sorption of DOC has been found to increase with increasing contents of dithionite-extractable Fe (Fed) in soil (Kahle et al., 2004; Kaiser and Guggenberger, 2000). In a recent study, we reported that at pH 6, the presence of goethite coatings on illitic and smectitic clays did not influence the stabilisation of plant residuederived OC as compared to those clays without goethite addition (Saidy et al., 2012). Hence, the relative contribution of hydrous oxides and phyllosilicate clays on OC sorption remains uncertain. One possible reason could be partial or even complete compensation of the hydrous oxides' positive charge upon their association with negatively charged phyllosilicate clays. Most studies on the sorption of OM on iron oxides have been carried out using soil samples containing different amounts of oxalate or dithionite-extractable Fe (e.g., Benke et al., 1999; Kahle et al., 2004; Kothawala et al., 2009) or in experiments where sorption was measured separately on phyllosilicate clays and oxides (e.g., Feng et al., 2005; Meier et al., 1999; Mikutta et al., 2007). These approaches do not allow for comprehensive testing of different types of phyllosilicate clays and hydrous oxides on the sorption of OM. There are few studies in which sorption has been measured in systems consisting of dissolved OM added to mixtures of clay minerals and hydrous oxides (Fusi et al., 1989; Kaiser and Zech, 1998). This approach enables the interactive effect of hydrous oxides and phyllosilicate clays on OC sorption to be tested directly. The goal of this study was to examine the effect of interactions between clay minerals and hydrous iron oxides on the sorption of plant-derived soluble OM. The objectives were to: (i) determine the sorption–desorption characteristics of OC to different clays, (ii) examine the effect of the addition of iron oxide (goethite) on the sorption– desorption of OC to clays differing in mineralogy, and (iii) examine the effect of adding different hydrous iron oxides on the capability of illitic clay to sorb and desorb OC. 2. Materials and methods 2.1. Clay minerals, hydrous iron oxides and chemical characterisation Kaolinite, illite and smectite clays were obtained from the collection of clay minerals at CSIRO Land and Water, Adelaide, Australia. The clays were prepared by the flocculation of the b2 μm fraction with CaCl2 and removing the excess salt by dialysis until the electrical conductivity (EC) was b10 mS cm−1 and freeze-drying. Goethite (α-FeOOH) was produced as described by Atkinson et al. (1967), by slowly neutralising a FeCl3 solution with NaOH and aging the precipitate at 55 °C for 3 days. Two-line ferrihydrite (5Fe2O3·9H2O) was prepared by neutralising a 0.1 M FeCl3 solution with NaOH (Schwertmann and Cornell, 1991). Haematite (α-Fe2O3) was produced by aging a suspension of fresh ferrihydrite at pH 7 and a temperature of 90 °C (Schwertmann and Cornell, 1991). The nature of the three mineral phases has been confirmed by X-ray diffraction (D5005, Siemens AG/Bruker AXS, Karlsruhe, Germany), and tests for solubility in dithionite–bicarbonate–citrate reagent (Mehra and Jackson, 1958) as well as in acid oxalate solution (Schwertmann, 1964). The specific surface areas (as determined by N2 adsorption–desorption (Nova 4200 analyser, Quantachrome Corp., Boynton Beach, USA)) were 36 m2 g−1 (haematite), 73 m2 g−1 (goethite), and 212 m2 g−1 (ferrihydrite). Hydrous iron oxide coated clays (kaolinite, illite and smectite with goethite; illite with goethite, haematite and ferrihydrite) were prepared through the precipitation of clay (30 g) with hydrous oxide (3 g) in 0.01 M CaCl2. Separate suspensions of clay (kaolinite, illite or smectite) and hydrous oxides (goethite, haematite or ferrihydrite) were prepared

in 0.01 M CaCl2 (1:10 = w:v) and their pH was adjusted to 6.0 with HCl or NaOH. The suspensions were combined, stirred, centrifuged at 5500 g for 10 min, and decanted carefully. The coated clays were then re-suspended in demineralised water and centrifuged at 5500 g for 10 min. This procedure was repeated until the EC of the supernatant was b 100 μS cm−1. The coated clays were freeze-dried and then sieved to b 200 μm. The coated clays had between 8 and 9 g of hydrous oxide per kg clay. The clays without coating received similar treatments to clays with hydrous iron oxide coating. Clay mineralogy was determined by X-ray diffraction (Siemens AG/ Bruker AXS D5000). The specific surface area (SSA) of clay minerals and coated clay minerals was determined by nitrogen adsorption at 77 K and subsequent desorption of nitrogen with a Tristar 5-point BET-instrument on freeze-dried samples. Nitrogen cannot access interlayer spaces of phyllosilicate clays, thus, the SSA is an estimate of external surfaces. Interlayer surfaces do not contribute to the DOC sorption of phyllosilicate clays (Baham and Sposito, 1994), hence external surfaces represent the interfaces where DOC sorption takes place. The cation exchangeable capacity (CEC) of clays and coated clays was determined using the ammonium acetate (pH 7) method (Rhoades, 1982). Exchangeable bases (Na, K, Ca and Mg) were analysed by atomic absorption spectroscopy (AAS) after extraction with ammonium acetate. The contents of dithionite-extractable iron (Fed) were measured using the method of Blakemore et al. (1987). Briefly, 0.5 g aliquots of freeze-dried clays and coated clays were shaken with 1 g sodium dithionite and 50 mL sodium citrate for 16 h. Then 0.05 M MgSO4 was added as a flocculant, the samples were centrifuged and the supernatant decanted. The supernatant was made up to 100 mL with deionized water, and the concentration of Fe was determined by inductively coupled plasma spectroscopy. 2.2. Preparation of dissolved organic carbon (DOC) solutions Dissolved organic carbon (DOC) was extracted from an oven-dried medic (Medicago truncatula cv. Praggio) shoot residue by adding 2 L deionized water to 200 g of ground medic (b2 mm). After 10 min of gentle stirring, the suspension was allowed to settle for 40 h at 22 °C and then filtered through a 0.45-μm membrane filter (Millipore Corporation, USA). The filtrate contained 14.7 g C L−1 and 1.44 g N L−1 (determined by Thermalox total organic C — total N analyser; Analytical Sciences Limited, Cambridge). The pH of the DOC solution was adjusted to 6.0 by addition of HCl or NaOH. Ten DOC solutions ranging in concentration from 0 to 200 mg C L−1 were prepared for sorption experiments by diluting with a solution containing 10 mg NaCl L−1, 20 mg CaCl2·2H2O L−1 and 24 mg K2SO4 L−1. The pH of the diluted DOC solutions was adjusted to 6.0 by addition of HCl or NaOH. 2.3. Sorption experiments Two sets of experiments were carried out: experiment 1 to assess the effect of goethite coatings on DOC sorption onto different phyllosilicate clays, and experiment 2 to examine the effect of the coating of three different hydrous iron oxides (haematite, goethite and ferrihydrite) on DOC sorption onto illitic clay. Sorption experiments were carried out using the batch equilibrium method in triplicate at pH 6.0 by adding 30 mL of DOC solution to 0.03 mg clay or clay-oxide in 50-mL centrifuge tubes. The ratio of 1:1000 (g clay dry weight:mL DOC solution) was chosen to achieve maximum sorption and ensure reliable analysis of the mineralisation of sorbed DOC. The suspensions were shaken at 22 °C for 12 h in the dark. Preliminary tests showed this time to be sufficient to reach equilibrium. Blanks without clay minerals were included to measure the initial DOC concentration. The suspensions were then centrifuged for 30 min at 2000 g, and the supernatants were filtered through 0.45-μm syringe filters (Millex-HV, Millipore Ireland Ltd, Tullagreen, and Carrigtwohill). The concentration of DOC in the filtrate was measured using a Thermalox TOC-TN analyser. The replicate variability in DOC concentration was b3%. The amount of OC sorbed was

A.R. Saidy et al. / Geoderma 209-210 (2013) 15–21

calculated as the difference between OC in the initial and equilibrium solutions. The pH of the equilibrium solution increased by 0.1–0.3 units, likely due to release of hydroxyl ions during adsorption reactions. Adsorption data were fitted to the Langmuir equation (Eq. (1)). The Langmuir equation expresses a relationship between the amount of substance adsorbed or desorbed (RE) (mg g−1), the final equilibrium solution concentration (Xf) (mg L−1), the binding affinity (k) (L mg−1), and the maximum adsorption capacity (Qmax) (mg g−1).     RE ¼ kQ max X f = 1 þ kX f

ð1Þ

The equation was fitted to the measured adsorption data using the least-square non-linear curve fitting routine in Microsoft Excel® (de Levie, 2001). The values of Qmax and k that gave the smallest residual sums of squares (RSS) were retained. 2.4. Desorption experiments The reversibility of DOC sorption was investigated by exposing mineral-organic associations with the largest DOC loading to a solution free of OC but of similar inorganic composition and pH to that used in the sorption experiment. A 15 mL desorption solution was added to the settled material immediately after sorption, centrifuged for 10 min at 2000 g and decanted carefully to remove OC that did not react with the minerals. The solutions used for desorption were added at a weight-to-volume ratio equal to that used in the sorption experiment. After 12 h shaking at 22 °C in the dark on an end-over-end rotary shaker, the suspension was centrifuged at 2000 g for 30 min, and the supernatant filtered through a 0.45-μm syringe filter. The filtrate was stored (4 °C for 14 h) before the concentration of OC was determined (Thermalox TOC-TN analyser). The desorption data were normalised to sample dry weight. The proportion of desorbable OC (% desorption) for each clay/clay–hydrous iron oxide–OM association was calculated as qdes / qsor × 100%, where qsor is the amount of OC sorbed (mg C g−1) at the largest OC loading (sorption experiment) and qdes is the amount of OC released during desorption (mg C g−1). 2.5. Statistical analysis The ANOVA procedure of GenStat 11th Edition (Payne, 2008) was used to compare the Langmuir parameters of clay–iron oxide associations. In the case of significance in ANOVAs, means were compared by the least significant difference (LSD) multiple comparison procedure at P b 0.05.

17

Coating clays with hydrous oxides resulted in increases in SSA where the SSA of the clay was less than that of the hydrous oxides (kaolinite–goethite, illite–ferrihydrite) and in reduced SSA where the clay's SSA exceeded that of the oxide (illite–goethite, illite–haematite, smectite–goethite). The changes in SSA were in line with added amounts of oxides. The coating also resulted in a decrease in CEC, which suggests that the coating involved electrostatic interactions between the negatively charged clays and the positively charged oxides. Decreases were smallest for kaolinite, moderate for illite, and largest for smectite. It is likely that the drop in CEC of the smectite–goethite association is due to the large accessible surface area of the smectite and a largely complete compensation of goethite's charge (Table 1). The drop in CEC was accompanied by a decrease in exchangeable Ca2+. 3.2. DOC sorption to different clays coated with goethite Sorption isotherms for DOC on the different clays with and without goethite coating are shown in Fig. 1. The sorption isotherms could be modelled reasonably well using the Langmuir equation (Eq. (1) — see Section 2.3), with r2 values ranging between 0.94 and 0.99 (Table 2). The maximum adsorption capacity (Qmax) of DOC on a mass basis for clays without goethite coating increased in the order kaolinite b illite b smectite (Table 2). However, after normalising the amount of OC sorbed to clays to the initial surface area, Qmax increased in the order illite b smectite b kaolinite. Goethite coating of kaolinite significantly increased Qmax but goethite coating of illite and smectite did not significantly change Qmax (Table 2). Despite only representing about 8% of the mass of the kaolinite–goethite association, DOC sorption tripled upon coating with goethite, highlighting mineral's large potential capacity to sorb OM even at the given pH of 6, which fits well with previous findings (e.g., Tipping, 1981). 3.3. DOC sorption to illitic clay with different hydrous iron oxide coatings Sorption isotherms for DOC on illitic clay coated with different hydrous iron oxides are shown in Fig. 2. Again, these could be reasonably fitted by the Langmuir equation, with r2 values ranging between 0.92 and 0.97 (Table 3). Coating with haematite resulted in significant decrease in Qmax on both a mass and surface area basis, whereas ferrihydrite coating increased Qmax on both a mass and surface area basis (Table 3). As stated above, goethite coating did not significantly change Qmax values (Table 3). 3.4. Desorption of OC sorbed to clays and clays coated with hydrous iron oxides

3. Results 3.1. Properties of clays and clay–oxide associations The SSA and CEC of the three uncoated test clays increased in the order kaolinite b illite b smectite, with kaolinite having a much smaller SSA and CEC than the two other clays (Table 1). The dominant exchangeable cation was Ca2 +.

The proportion of OC desorbable from clays without goethite coating increased in the order kaolinite b illite b smectite (Table 4). Goethite coating did not significantly influence the proportion of OC desorbable from clays except for kaolinite, for which there was a small decrease. Coating illitic clays with different hydrous iron oxides only significantly affected the proportion of desorbable OC for ferrihydrite, for which there was a decrease (Table 4).

Table 1 Properties of the clays and hydrous iron oxide coated clays. Properties

Kaolinite

Illite

Smectite

Kaolinite–goethite

Illite–goethite

Illite–haematite

Illite–ferrihydrite

Smectite–goethite

CEC (cmol kg−1) Na+ (cmol kg−1) K+ (cmol kg−1) Mg2+ (cmol kg−1) Ca2+ (cmolkg−1) SSA (m2 g−1) Fed (mg kg−1)

8.9 b0.10 0.2 b0.2 1.3 6.01 129

23.0 0.14 1.5 2.1 12.8 108.0 947

86.9 0.91 1.9 10.5 54.3 169.8 883

7.5 0.16 b0.05 b0.2 0.8 11.48 52,800

21.4 0.30 1.4 4.4 12.2 98.8 55,600

20.8 0.11 1.3 3.8 8.5 95.9 61,900

21.3 0.13 1.3 3.1 5.5 115.6 51,000

73.6 0.17 1.9 9.4 45.2 157.8 55,700

18

A.R. Saidy et al. / Geoderma 209-210 (2013) 15–21

0.60

8 6

0.40 Kaolinte Kaolinite-goethite

4

0.20 2 0.00

0 50

100

150

200

250

C adsorbed (mg C m-2 clay)

C adsorbed (mg C g-1 clay)

0

15 12 9 6

Illite Illite-goethite

3 0 0

50

100

150

200

0

50

100

150

200

250

0

50

100

150

200

250

0

50

100

150

200

250

0.20

0.15

0.10

0.05

0.00

250 0.20

30 25

0.15

20 0.10

15 Smectite Smectite-goethite

10

0.05

5 0

0.00 0

50

100

150

200

250

Equilibrium concentration (mg C L-1) Fig. 1. Isotherms of the sorption of dissolved organic carbon to different clays with and without goethite addition. Symbols denote experimental points while the curves represent Langmuir isotherms fitted to the experimental data.

4. Discussion The greater DOC sorption capacity (on a mass of clay basis) of smectite than those of kaolinite and illite is in agreement with previous studies. Wang and Xing (2005) reported that a 2:1 layer silicate (montmorillonite) adsorbed more than twice as much humic acid as a 1:1 layer silicate (kaolinite). Dontsova and Bigham (2005) also found that the sorption of microbial polysaccharide was smallest with kaolinite and greatest with smectite. Differences in specific surface area (SSA) of clay minerals (Table 1) may account for the differences in DOC sorption by clays since SSA, as assessed by N2 adsorption, gives an estimate

Table 2 Langmuir parameters from fits to sorption of dissolved organic carbon to kaolinite, illite and smectite, with and without goethite coating. Clay

Iron-oxide

Qmax (mg g−1)

Kaolinite Kaolinite Illite Illite Smectite Smectite

None Goethite None Goethite None Goethite

2.82 7.41 13.00 12.34 38.67 37.72

(0.02)⁎ a (0.03) b (0.38) c (0.04) c (1.52) d (1.05) d

Qmax (mg m−2) 0.464 (0.003) 0.622 (0.003) 0.124 (0.001) 0.123 (0.002) 0.233 (0.009) 0.227 (0.006)

c ⁎⁎ d a a b b

k (L mg−1)

r2

0.056 0.035 0.032 0.033 0.013 0.009

0.99 0.99 0.94 0.94 0.96 0.95

(0.006) (0.001) (0.001) (0.001) (0.001) (0.001)

⁎ Values in parenthesis refer to the standard error of mean. ⁎⁎ Similar letters in each column indicate no statistical difference between treatments based on the LSD test at P b 0.05.

of the external mineral surface area potentially available for DOC sorption. Nelson et al. (1992) reported a significant positive relationship between SSA and DOC adsorption capacity for several untreated surface and subsoil horizons. The larger SSA of illitic and smectitic clays compared to the kaolinitic clays have been suggested to favour more and/or stronger OC–clay interactions (von Lützow et al., 2006; Wattel-Koekkoek et al., 2003). Cation contents and other mineralogical properties of the mineral surfaces have also been reported to influence the adsorption of DOC to clay minerals (Feng et al., 2005; Ussiri and Johnson, 2004). Polyvalent exchangeable cations such as Ca2+ and Mg2+, the concentration of which increased in order kaolinite b illite b smectite (Table 1), may function to bridge the negatively charged clay surfaces and negatively charged anionic functional groups of OM. This provides an alternate explanation for the increasing sorption capacity of clays in the order kaolinite b illite b smectite. This possibility is supported by previous studies (e.g., Arnarson and Keil, 2000; Feng et al., 2005; Schlautman and Morgan, 1994) who found that the presence of multivalent cations, such as Ca2+ and Al3+, can enhance OC adsorption through cation bridges. When sorption capacity (Qmax) is expressed on a surface area basis, the order of increasing sorption is completely different (illite b smectite b kaolinite) to the order when expressed on a mass basis (kaolinite b illite b smectite). The former reflects the order of increasing clay surface area charge density, i.e., CEC/SSA, and therefore

A.R. Saidy et al. / Geoderma 209-210 (2013) 15–21

Illite

0.15

Illite-goethite 25

C adsorbed (mg C m-2 clay)

C adsorbed (mg C g-1 clay)

30

19

Illite-haematite Illite-ferrihydrite

20 15 10 5

0.10

0.05

0.00

0 0

50

100

150

200

250

0

50

100

150

200

250

Equilibrium concentration (mg C L-1) Fig. 2. Isotherms of the sorption of dissolved organic carbon to illitic clay with different hydrous iron oxide coatings. Symbols denote experimental points while the curves represent Langmuir isotherm fitted to the experimental data.

supports the idea that cation bridging plays an important role. However, the relatively low surface-normalised Qmax and the high maximum surface loading of kaolinite point to other factors being involved. The higher surface loading of kaolinite compared to other clays is in agreement with other studies. Feng et al. (2005) found that the sorption of peat humic acid on a surface area basis was much higher on kaolinite (maxima 0.08–0.43 mg C m−2) than on montmorillonite (maxima 0.006–0.06 mg C m−2). In addition, Kahle et al. (2004) also reported stronger adsorption of DOC to kaolinite than illite, even though illite has a much larger surface area and higher cation exchange capacity (CEC) than kaolinite. Kaiser and Zech (2000) suggested that the greater sorption of OM per unit surface area for kaolinite than for illite is the result of the greater contribution of the AlOH octahedral sheet to the external surface areas of that mineral than to that of illite. The surfaces of octahedral sheets resemble those of Al hydroxides, thus, having hydroxyl groups that can react with organic functional groups the same way as those at the surface of hydrous Al and Fe oxides. The different nature of surfaces may also cause selective sorption of organic molecules. Binding sites on kaolinite surfaces have been reported to be occupied by relatively larger molecules, while 2:1 layer silicate (e.g., illite and smectite) was predominantly covered with compounds with a wider range of molecular weight (Chorover and Amistadi, 2001; Feng et al., 2005). This could contribute to an increased mass of OM sorbed to kaolinitic than to illitic and smectitic clays. The presence of hydrous iron oxides is a major factor in DOC sorption and thereby in the stabilisation of OC in soils. Kahle et al. (2004) found a positive relationship between the content of dithionite-extractable Fe (Fed) and the amount of DOC sorbed to phyllosilicate clays and soil clay fractions. The positive effects of hydrous iron oxides on DOC sorption onto soils were corroborated by Kothawala et al. (2009), who observed a direct proportionality between the amount of extractable Fe and Al and the maximum adsorption capacity (Qmax) of 52 soil samples and 9 soil horizons representing five soil orders. Under the experimental conditions used, we detected an effect of goethite coating on DOC sorption only for kaolinite; there were no statistically significant effects of Table 3 Langmuir parameters from fits to sorption of dissolved organic carbon to illite coated with different hydrous iron oxides. Clay

Iron oxide

Qmax (mg g−1)

Illite Illite Illite Illite

None Goethite Haematite Ferrihydrite

13.00 12.34 11.73 21.59

(0.38)⁎b (0.68) ab (0.01) a (0.43) c

Qmax (mg m−2) 0.124 0.123 0.117 0.130

(0.018) (0.007) (0.003) (0.003)

b⁎⁎ b a c

k (L mg−1)

r2

0.032 0.033 0.032 0.038

0.94 0.94 0.92 0.97

(0.001) (0.001) (0.001) (0.001)

⁎ Values in parenthesis refer to the standard error of mean. ⁎⁎ Similar letters in each column indicate no statistical difference between treatment based on the LSD test at P b 0.05.

goethite coating on DOC sorption for illite and smectite. The strong increase in DOC sorption brought about by the goethite coating of kaolinite clearly underlines the large potential OC sorption capacity of hydrous iron oxides, especially when considering the small amount of goethite involved (~8% of the mass of the kaolinite–goethite assemblage). However, when associated with 2:1 minerals, goethite did not add to their sorption capacity. This demonstrates that the effects of hydrous iron oxides on DOC sorption can be modified by interactions with phyllosilicate clays. One possible reason for the different effects of goethite coating on kaolinite and other phyllosilicate clays (illite and smectite) could be the change in surface area. Coating kaolinitic clay with goethite almost doubled the SSA, whereas the SSA of illitic and smectitic clays decreased slightly with goethite coating (Table 1). The sorption capacity of the goethite-coated 2:1 minerals was about the same as that of the uncoated minerals, despite the reduction in CEC and the loss of divalent cations (Ca2+ in this case) upon coating. That possible loss in potential sorption capacity of the illite and smectite was partly compensated by the oxide coating; however, there was no increase in sorption capacity as found for kaolinite. This could be due to compensation of the goethite's surface charge by the clays' negative charge, which may cause its sorption capacity to decrease. The different effects of oxide coatings on clay minerals on DOC are in agreement with our previous finding (Saidy et al., 2012) of goethite coatings on different clay minerals causing increased OC stabilisation by kaolinitic clays but not by 2:1 minerals.

Table 4 Amounts of sorbed organic carbon, desorbable organic carbon, and % desorbable organic carbon observed at the maximum dissolved organic carbon sorption. Desorption (mg g−1)

% desorption

(0.03)⁎ (0.05) (0.14) (0.08) (0.42) (0.17)

0.22 0.35 1.28 1.27 3.45 3.40

(0.03) (0.01) (0.04) (0.02) (0.11) (0.09)

9.13 5.69 10.86 11.58 13.59 14.37

(0.26) (0.16) (0.26) (0.07) (0.28) (0.46)

b⁎⁎ a c c d d

(0.14) (0.08) (0.14) (0.30)

1.28 1.27 1.28 1.25

(0.04) (0.01) (0.01) (0.01)

11.69 11.58 11.89 6.43

(0.30) (0.07) (0.09) (0.12)

b b b a

Clay

Iron-oxide

Adsorption (mg g−1)

Experiment 1 Kaolinite Kaolinite Illite Illite Smectite Smectite

None Goethite None Goethite None Goethite

2.51 4.44 11.81 10.98 25.33 23.65

Experiment 2 Illite Illite Illite Illite

None Goethite Haematite Ferrihydrite

11.81 10.98 10.73 19.48

⁎ Values in parenthesis refer to the standard error of mean. ⁎⁎ Similar letters in each column indicate no statistical difference between treatment based on the LSD test at P b 0.05.

A.R. Saidy et al. / Geoderma 209-210 (2013) 15–21

samples and suggested this was due to strong chemisorptive bonding. Similarly, Mikutta et al. (2007) reported that OC primarily sorbing via ligand exchange was less prone to desorption than OC sorbed via cation bridges and van der Waals forces, which dominate the sorption of OC to clay minerals. Differences in the percentage of DOC desorbed from the clay mineral–iron oxide–OM association may also be related to differences in affinity of DOC for the clay mineral–iron oxide associations. Fig. 3 shows that there is a positive relationship between the affinity coefficient of the Langmuir equation (k in Eq. (1) — see Section 2.3) and the proportion of desorbed DOC. A statistical analysis showed this to be a significant correlation for both experiments (r = −0.81; P b 0.001 for experiment 1, and r = −0.73; P b 0.007 for experiment 2). This finding is consistent with the concept that OC that sorbs strongly to minerals is more resistant to desorption (Henrichs, 1995). This result is also in agreement with the observation by Mikutta et al. (2007) that the fraction of desorbable OC was smaller when the affinity of OM for minerals is larger. 5. Conclusions The results of this study indicate that, under the experimental conditions, DOC sorption varied with phyllosilicate clay type; smectite, which has a higher surface area and cation exchange capacity, adsorbed more DOC than kaolinite or illite. Coating clays with goethite increased the sorption capacity of clays, but the difference between clays with and without goethite coating was only significant for kaolinite. Among three hydrous iron oxides tested, only ferrihydrite coating increased the capability of illite to adsorb DOC, clearly indicating that the amount of DOC retained by phyllosilicate clays also depends on the type of

a

Kaolinite Kaolinite-goethite Illite Illite-goethite Smectite Smectite-goethite

20

15

% desorption

There is a second possible explanation for the different effects of the presence of goethite on DOC sorption for different clays that relates to differences in their CEC. At pH 6.0, goethite is slightly below its point of zero charge (PZC) of pH 8.1 (Kaiser and Guggenberger, 2000) and will therefore have a small net positive charge, whereas at pH 6.0, clay minerals have a net negative charge. Thus, the positively charged hydrous iron oxide coatings partly counter the negative layer charge of the clays (Roth et al., 1969; Zhuang and Yu, 2002). Consequently, the CEC of all tested phyllosilicates decreased upon coating with oxides, mainly with a loss of exchangeable Ca2+ (see above). Since they have a much larger CEC than kaolinite, illite and smectite may balance most of the goethite's positive charge, thus suppressing the sorptive interactions with negatively charged organic functional groups. On the other hand, the smaller negative charge of kaolinite may only partly balance the positive charge of the goethite. The resulting increase in positive charge of the kaolinite-oxide surfaces allows for possible sorption of negatively-charge DOC via strong columbic mechanisms (e.g., ligand exchange). The result would be an increase in DOC sorption for kaolinite–goethite associations over that of pure kaolinite. On the other hand, the compensation of goethite's surface charge by illite and smectite reduces the CEC of the clays as well as exchangeable Ca2+, causing reduced sorption capacity of both the clays and the oxide. The net result is a largely unchanged sorption capacity of the clays upon coating with goethite. Among hydrous iron oxides coated onto illite, only ferrihydrite resulted in an increase in DOC sorption, which is in line with the increase in SSA (Table 1). The addition of haematite and goethite to illitic clay led to a decrease in SSA (Table 1), and coincided with a small decrease in DOC sorption for haematite and no significant change for goethite. The decrease in DOC sorption upon haematite coating was coincident with the largest decrease in CEC (Table 1). Again, these different effects on sorption may at least in part be due to the different charge properties of the different hydrous iron oxides. Ferrihydrite has a higher PZC than goethite or hematite (PZC of ferrihydrite at pH 8.6; PZC of goethite at pH 8.1; PZC of haematite at pH 7.6 — Kaiser and Guggenberger, 2003). Therefore, at pH 6.0, ferrihydrite would have a higher positive charge than the other oxides used for coating. Consequently, its surface charge could have been large enough not to be balanced entirely by the illite's surface charge and so the ferrihydrite added to the sorption of the illite–ferrihydrite system. In desorption experiments, most of the DOC sorbed was not removed in a single extraction step. In fact, only 8–14% of sorbed DOC was released from different clays with and without goethite coating and only 6–12% was released from illite with different iron hydrousoxide coatings (Table 4). The effect of hydrous iron oxides on desorption was most evident for the kaolinite–goethite and illite–ferrihydrite systems; for the other oxide coatings, desorption was in the range of the uncoated clays. The percentage of OC released was lower than that reported by Kahle et al. (2004) who observed that 17–50% of initially sorbed OC was released from kaolinite, illite, and montmorillonite by a single extraction step. It should be noted that the desorption was carried out with a solution containing sulfate, an anion that can compete with organic anions for binding sites. Mikutta et al. (2007) reported desorption of 2–35% of OC initially sorbed to clay minerals and goethite with different binding mechanisms. Desorption was smallest for goethite, and higher for the clay minerals. That is well in line with the small reduction in desorption upon goethite coating, especially when considering the small portion of goethite added (~8 mass-%). Since the surface properties of hydrous iron oxides are different from to those of phyllosilicates, the presence of these coating materials was expected to affect the desorption of DOC. In particular, it was anticipated that a smaller proportion of DOC would be desorbed, due to the stronger OM–iron oxide interactions (Benke et al., 1999; Kahle et al., 2004). Kaiser and Guggenberger (2000) reported negligible desorption of DOC initially sorbed at the surface of hydrous oxides and subsoil

10

5

0 0.00

0.02

0.04

0.06

0.08

Affinity coeficient (L mg-1)

b

20 Illite Illite-goethite

15

% desorption

20

Illite-haematite Illite-ferrihydrite 10

5

0 0.030

0.033

0.035

0.038

0.040

Affinity coeficient (L mg-1) Fig. 3. Relation between the proportion of desorbable OC (% desorption) and the affinity coefficient (k) of kaolinite, illite and smectite with and without goethite coating (a); and illite with and without different hydrous iron oxide coatings (b).

A.R. Saidy et al. / Geoderma 209-210 (2013) 15–21

hydrous iron oxide present. These results suggest that the capability of hydrous iron oxides to adsorb DOC can be affected by interactions with phyllosilicate clays; in turn, hydrous oxides seem to modify the sorption of the clay minerals. Only 6–14% of the OC initially sorbed by mineral associations was released in a single extraction step, indicating that most of the DOC sorbs strongly to mineral surfaces. Desorption varied with clay mineralogy and the type of iron oxide in a similar way as for sorption. Since increases in DOC sorption were most evident for clay mineral–hydrous oxide combinations involving either a low-charge clay or high-surface area hydrous oxide; we assume that the extent of the balancing of the hydrous oxides' positive charge by clays' negative charge was crucial for the performance of hydrous oxide coatings in the DOC sorption. The changes in CEC upon coating phyllosilicate clays support that idea. The experimental conditions were set up to reflect those in many agricultural soils, having a pH around 6, where the surface charge of oxides is moderate. However, previous studies on agricultural soils and soil fractions showed OM stabilisation to be governed by hydrous iron oxide (e.g., Bruun et al., 2010). That could mean a substantial portion of the hydrous oxides in soil does form coatings on phyllosilicate clays but resides in associations with OM, as suggested by Fontes et al. (1992). Acknowledgements The authors would like to acknowledge Dr Todd Maddern for assistance in DOC measurement. The financial support in the form of a postgraduate scholarship grant provided by the Directorate of Higher Education, Ministry of National Education, the Republic of Indonesia for first author is also gratefully acknowledged. References Angove, M.J., Fernandes, M.B., Ikhsan, J., 2002. The sorption of anthracene onto goethite and kaolinite in the presence of some benzene carboxylic acids. Journal of Colloid and Interface Science 247, 282–289. Arnarson, T.S., Keil, R.G., 2000. Mechanisms of pore water organic matter adsorption to montmorillonite. Marine Chemistry 71, 309–320. Atkinson, R.J., Posner, A.M., Quirk, J.P., 1967. Adsorption of potential-determining ions at the ferric oxide-aqueous electrolyte interface. Journal of Physical Chemistry 71, 550–558. Baham, J., Sposito, G., 1994. Adsorption of dissolved organic carbon extracted from sewage sludge on montmorillonite and kaolinite in the presence of metal ions. Journal of Environmental Quality 23, 147–153. Balcke, G.U., Kulikova, N.A., Hesse, S., Kopinke, F.D., Perminova, I.V., Frimmel, F.H., 2002. Adsorption of humic substances onto kaolin clay related to their structural features. Soil Science Society of America Journal 66, 1805–1812. Benke, M.B., Mermut, A.R., Shariatmadari, H., 1999. Retention of dissolved organic carbon from vinasse by a tropical soil, kaolinite, and Fe-oxides. Geoderma 91, 47–63. Blakemore, L.C., Searle, P.L., Daly, B.K., 1987. Methods for chemical analysis of soils. New Zealand Soil Bureau, Scientific Report, 80. Department of Scientific and Industrial Research, Lower Hutt, New Zealand. Bruun, T.B., Elberling, B., Christensen, B.T., 2010. Lability of soil organic carbon in tropical soils with different clay minerals. Soil Biology and Biochemistry 42, 888–895. Butman, D., Raymond, P., Oh, N.H., Mull, K., 2007. Quantity, 14C age and lability of desorbed soil organic carbon in fresh water and seawater. Organic Geochemistry 38, 1547–1557. Chorover, J., Amistadi, M.K., 2001. Reaction of forest floor organic matter at goethite, birnessite and smectite surfaces. Geochimica et Cosmochimica Acta 65, 95–109. de Levie, R., 2001. How to Use Excel in Analytical Chemistry and in General Scientific Data Analysis. Cambridge University Press, Cambridge, UK. Dontsova, K.M., Bigham, J.M., 2005. Anionic polysaccharide sorption by clay minerals. Soil Science Society of America Journal 69, 1026–1035. Feng, X.J., Simpson, A.J., Simpson, M.J., 2005. Chemical and mineralogical controls on humic acid sorption to clay mineral surfaces. Organic Geochemistry 36, 1553–1566. Fontes, M.R., Weed, S.B., Bowen, L.H., 1992. Association of microcrystalline goethite and humic acid on some oxisols from Brazil. Soil Science Society of America Journal 56, 982–990. Fusi, P., Ristori, G.G., Calamai, L., Stotzky, G., 1989. Adsorption and binding of protein on clean (homoionic) and dirty (coated with Fe-oxyhydroxides) montmorillonite, illite and kaolinite. Soil Biology and Biochemistry 21, 911–920. Gu, B., Schmitt, J., Chen, Z., Liang, L., McCarthy, J.F., 1994. Adsorption and desorption of natural organic matter on iron oxide: mechanisms and models. Environmental Science and Technology 28, 38–46.

21

Henrichs, S.M., 1995. Sedimentary organic matter preservation: an assessment and speculative synthesis — a comment. Marine Chemistry 49, 127–136. Kahle, M., Kleber, M., Jahn, R., 2004. Retention of dissolved organic matter by phyllosilicate and soil clay fractions in relation to mineral properties. Organic Geochemistry 35, 269–276. Kaiser, K., Guggenberger, G., 2000. The role of DOM sorption to mineral surfaces in the preservation of organic matter in soils. Organic Geochemistry 31, 711–725. Kaiser, K., Guggenberger, G., 2003. Mineral surfaces and soil organic matter. European Journal of Soil Science 54, 219–236. Kaiser, K., Guggenberger, G., 2007. Sorptive stabilization of organic matter by microporous goethite: sorption into small pores vs. surface complexation. European Journal of Soil Science 58, 45–59. Kaiser, K., Zech, W., 1998. Soil dissolved organic matter sorption as influenced by organic and sesquioxide coatings and sorbed sulfate. Soil Science Society of America Journal 62, 129–136. Kaiser, K., Zech, W., 2000. Sorption of dissolved organic nitrogen by acid subsoil horizons and individual mineral phases. European Journal of Soil Science 51, 403–411. Kaiser, K., Mikutta, R., Guggenberger, G., 2007. Increased stability of organic matter sorbed to ferrihydrite and goethite on aging. Soil Science Society of America Journal 71, 711–719. Kalbitz, K., Schwesig, D., Rethemeyer, J., Matzner, E., 2005. Stabilization of dissolved organic matter by sorption to the mineral soil. Soil Biology and Biochemistry 37, 1319–1331. Kothawala, D.N., Moore, T.R., Hendershot, W.H., 2009. Soil properties controlling the adsorption of dissolved organic carbon to mineral soils. Soil Science Society of America Journal 73, 1831–1842. Mehra, O.P., Jackson, M.L., 1958. Iron oxide removal from soils and clays by a dithionite–citrate system buffered with sodium bicarbonate. Clays and Clay Minerals 7, 317–327. Meier, M., Namjesnik-Dejanovic, K., Maurice, P.A., Chin, Y.P., Aiken, G.R., 1999. Fractionation of aquatic natural organic matter upon sorption to goethite and kaolinite. Chemical Geology 157, 275–284. Mikutta, R., Mikutta, C., Kalbitz, K., Scheel, T., Kaiser, K., Jahn, R., 2007. Biodegradation of forest floor organic matter bound to minerals via different binding mechanisms. Geochimica et Cosmochimica Acta 71, 2569–2590. Nelson, P.N., Baldock, J.A., Oades, J.M., 1992. Concentration and composition of dissolved organic carbon in streams in relation to catchment soil properties. Biogeochemistry 19, 27–50. Ohtsubo, M., 1989. Interaction of iron oxides with clays. Clay Science 7, 227–242. Payne, R., 2008. A Guide to Anova and Design in Genstat. VSN International, Hempstead, UK. Rhoades, J.D., 1982. Cation Exchange Capacity. Methods of Soil Analysis Part 2: Chemical and Microbiological Properties. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), American Society of Agronomy, Inc. and Soil Science Society of America, Inc., Madison, Wisconsin, pp. 149–157. Roth, C.B., Jackson, M.L., Syers, J.K., 1969. Deferration effect on structural ferrous–ferric iron ratio and CEC of vemiculites and soils. Clays and Clay Minerals 17, 253–264. Saidy, A.R., Smernik, R.J., Baldock, J.A., Kaiser, K., Sanderman, J., Macdonald, L.M., 2012. Effects of clay mineralogy and hydrous iron oxides on labile organic carbon stabilisation. Geoderma 173, 104–110. Schlautman, M.A., Morgan, J.J., 1994. Adsorption of aquatic humic substances on colloidalsize aluminium-oxide particles: influence of solution chemistry. Geochimica et Cosmochimica Acta 58, 4293–4303. Schneider, M.P.W., Scheel, T., Mikutta, R., van Hees, P., Kaiser, K., Kalbitz, K., 2010. Sorptive stabilization of organic matter by amorphous Al hydroxide. Geochimica et Cosmochimica Acta 74, 1606–1619. Schwertmann, U., 1964. Differenzierung der Eisenoxide des Bodens durch Extraktion mit Ammoniumoxalat-Lösung. Zeitschrift für Pflanzenernährung, Düngung und Bodenkunde 105, 194–202. Schwertmann, U., Cornell, R.M., 1991. Iron Oxides in the Laboratory. VCH Verlagsgesellschaft, Weinheim, Germany. Tipping, E., 1981. The adsorption of aquatic humic substances by iron oxides. Geochimica et Cosmochimica Acta 45, 191–199. Tombácz, E., Libor, Z., Illes, E., Majzik, A., Klumpp, E., 2004. The role of reactive surface sites and complexation by humic acids in the interaction of clay mineral and iron oxide particles. Organic Geochemistry 35, 257–267. Ussiri, D.A.N., Johnson, C.E., 2004. Sorption of organic carbon fractions by Spodosol mineral horizons. Soil Science Society of America Journal 68, 253–262. von Lützow, M., Kögel-Knabner, I., Ekschmitt, K., Matzner, E., Guggenberger, G., Marschner, B., Flessa, H., 2006. Stabilization of organic matter in temperate soils: mechanisms and their relevance under different soil conditions — a review. European Journal of Soil Science 57, 426–445. Wang, K.J., Xing, B.S., 2005. Structural and sorption characteristics of adsorbed humic acid on clay minerals. Journal of Environmental Quality 34, 342–349. Wattel-Koekkoek, E.J.W., Buurman, P., van der Plicht, J., Wattel, E., van Breemen, N., 2003. Mean residence time of soil organic matter associated with kaolinite and smectite. European Journal of Soil Science 54, 269–278. Xing, B.S., 2001. Sorption of naphthalene and phenanthrene by soil humic acids. Environmental Pollution 111, 303–309. Zhuang, J., Yu, G., 2002. Effects of surface coatings on electrochemical properties and contaminant sorption of clay minerals. Chemosphere 49, 619–628.