1 Introduction

Soil organic matter consists of up to 60–90% of humic substances (HSs). Most stable HSs in soils are incorporated into humic-clay complexes (Stevenson 1994; Laird 2001). They occur in soils, sediments, and aquifers, and it is thought that these HS are irreversibly immobilized onto clay minerals (Kononova 1966). The multiple links between molecular properties of HS and sorptive capabilities have been widely studied. For example, it has been demonstrated that the high molecular weight and low polarity of HS causes high sorption affinity to clay minerals (Baham and Sposito 1994; Specht et al. 2000; Balcke et al. 2002; Khalaf et al. 2003; Ghosh et al. 2009). However, complexity of HS composition and large structural variations impose severe constraints on our molecular understanding of the relationships between HS structure and sorption affinities to clay minerals. As a rule, the sorption of heavy metals, actinides, and hydrophobic organic compounds by humic-clay complexes is much higher as compared to bare clay mineral surfaces (Cruz-Guzman et al. 2003; Nachtegaal and Sparks 2003; Geckeis 2004; Wang et al. 2005; Xu et al. 2006). This effect is frequently related to the substantial change in HS-modified surface properties such as hydrophobicity, surface area, and free energy (Murphy and Zachara 1995). However, quantitative estimates of these changes are very limited.

The wetting contact angle can be used as a quantitative indictor of surface hydrophobicity and free energy. It can be determined using a number of simple and quick techniques (Young 1805; Giljean et al. 2011) including the sessile drop method (both static and dynamic) (Bachmann et al. 2000), the Wilhelmy plate method, the thin-layer wicking method, and the column wicking method (Chibowski and Gonzalez-Caballero 1993). However, the contact angle of colloidal particles is difficult to measure because of small particle sizes (<2 μm in diameter), which in current methods can lead to artifacts (Shang et al. 2008; Wang et al. 2005). Clay mineral type variability can also lead to specific requirements and conditions for contact angle measurements. Contact angles for kaolin and smectite can be measured using both the static sessile drop method and the Wilhelmy plate method (Shang et al. 2008). For this study, the static sessile drop was chosen for determination of contact angles in Ca2+ saturated kaolin and montmorillonite and in their complexes with irreversibly sorbed HS.

The static sessile drop method has been applied primarily for smectites because these clays form a water impermeable film upon swelling. In case of kaolin, such a film contains pores and liquids can infiltrate into the pores, which decreases the contact angle (Wu 2001). Modern approaches, however, can alleviate this problem. To obtain the true contact angle for kaolin, the readings should be accomplished immediately at 0 s, before inhibition begins. The use of a digital goniometer, such as the Kruss Drop Shape Analysis System, allows for determination of the contact angle immediately after the drop is placed on the substrate, so that accurate readings at 0 s can be acquired (Shang et al. 2008).

In this study, we show the applicability of a static sessile drop contact angle method to assess wettability of both 2:1 (smectite) and 1:1 (kaolin) clays, which were modified with four humic materials isolated from soils, peat, and coal.

2 Materials and methods

2.1 Clay minerals

Kaolin clay (Kaolin CF 70) was from Caminauer Kaolinwerk GmbH (Caminau, Germany). Smectite (montmorillonite) clay was used from the bentonite deposit “10 khutor,” located in Khakassia, Russia. Both clays were pretreated in the same way. Specifically, 50 mL of kaolin, at concentrations of 375 g L−1, or montmorillonite, at concentrations of 150 g L−1, were suspended in water and treated by an ultrasonic dispergator for 2 min (22 kHz, 0.1 kWt). The clay size particles (≤2 μm) were separated by centrifugation (at 3000 rpm for 4 min), and 1 M CaCl2 was added into the supernatant until a final concentration of 0.1 M Ca2+ was reached. The obtained suspensions were shaken for 24 h using an overhead shaker, and then the clay particles were precipitated by centrifugation (at 3000 rpm for 5 min). The precipitate was ultrasonically redispersed in 0.001 M CaCl2 at the ratio of 1:20 for kaolin and 1:40 for montmorillonite and mounted to an overhead shaker for 24 h. The procedure was repeated four times. The final suspension concentrations were stored in the dark prior to being measured gravimetrically.

2.2 Humic materials

Four humic materials were used in this study: coal (leonardite) humic acids isolated from the commercially available potassium humate (Powhumus, Humintech Ltd., Germany); peat humic acids isolated from a highmoor peat (the Tver region, Russia); soil humic acid isolated from a typical chernozem (WRB 2014) located in the Kursk region, Russia; and soil humic acid isolated from sod podzolic soil (Retisol according to WRB 2014). The samples were designated as Pow, T5, CtK06, and Pw08, respectively. The isolation protocol followed recommendations of the International Humic Substances Society (IHSS) (Swift 1996; Kholodov et al. 2015). Briefly, 1 M HCl was added to homogenized soil sample until the pH declined to 1 to 2, then the solution volume was adjusted to a final ratio of 10 mL liquid phase to 1 g of dry sample using 0.01 M HCl. The suspension obtained was shaken for 4 h and then centrifuged. The solid residue was adjusted to pH 7 using 1 M NaOH, and then 0.1 M NaOH was added until a final liquid to solid ratio of 10:1 (v/v) was reached. The suspension obtained was shaken for 4 h, left overnight, and centrifuged. The supernatant was collected and acidified using 6 M HCl to reach pH 1. The precipitate of humic acids was collected by centrifugation, redissolved in 0.1 M KOH, and then centrifuged at high speed to remove fine suspended particles. The humic acids were precipitated by acidification again, and the precipitate was treated three times with 0.1 M HCl/0.3 M HF. The treated precipitate was desalted by dialysis and brought to dryness under vacuum at 40 °C.

2.3 Structural characterization of humic materials

Elemental analyses (C, H, N) were conducted using a Vario EL analyzer. The ash content was determined by combustion in a quartz tube at 750 °C. Oxygen content was calculated as the difference between total dry weight of organic matter and the portions attributable to C, H, and N. Elemental fractions were calculated on an ash-free basis. The H/C and O/C atomic ratios were calculated as indicators of saturation degree and polarity of HS, respectively.

Size exclusion chromatography (SEC) determinations were performed as described by Perminova et al. (2003). The separations were conducted on a column packed with Toyopearl HW-55S gel (TosoHaas). Five sodium polystyrenesulfonates with known molecular weights were used for column calibration. HS solutions were set at a concentration of 10 mg L−1 of OC by equilibrating with the SEC mobile phase (0.028 M phosphate buffer, pH 6.8) prior to analysis. Ultraviolet (UV) absorbance at λ = 254 nm was used for detection. The number (M n) and weight (M w) average molecular weights and the M w/M n ratio (polydispersity) were calculated from these analyses.

The structural composition of HS was determined by 13C NMR spectroscopy (Kovalevskii et al. 2000; Kholodov et al. 2011). An Avance 400 NMR spectrometer (Bruker) was used for acquiring the spectra after dissolving 50 mg of humic material in 0.6 mL of 0.3 M NaOD/D2O. The spectral assignments were made as described by Hertkorn et al. (2002): 5 to 50 ppm, aliphatic H- and C-substituted C atoms (CAlk); 50 to 108 ppm, aliphatic O-substituted C atoms (CAlk-O); 108 to 145 ppm, aromatic H- and C-substituted C atoms (CAr); 145 to 165 ppm, aromatic O-substituted C atoms (CAr-O); 165 to 187 ppm, C atoms of carboxyl and ester groups (CCOO-H,R); and 187 to 220 ppm, C atoms of quinone and ketone groups (CC=O).

2.4 Preparation of HS-clay complexes

HS stock solutions with concentrations of 1 and 10 g L−1 and containing 1 mM Ca2+ were used for sorption onto Ca-kaolin and Ca-montmorillonite, respectively. The HS solutions were added to clay suspensions at pH 5.5; the pH was adjusted using 0.1 M HCl and 0.1 M NaOH. The final HS concentrations were 0.5 g L−1 for Ca-kaolin and 4 g L−1 for Ca-montmorillonite. These HS concentrations were sufficient for complete saturation of all sorption sites onto kaolin and montmorillonite surfaces, based upon our preliminary experiments. The clay to HS solutions ratio was 1:80. For equilibration, the suspensions were put on an overhead shaker for 24 h and then centrifuged at 6000 rpm. The concentration of adsorbed HS was determined by measuring optical density at 465 nm before and after adsorption. To desorb reversibly bound HS from the clay-humic complexes, fresh 1 mM Ca2+ solution at pH 5.5 was added to the precipitate, resuspended, and shaken for 24 h. This was done six times until the optical density of desorbed HS in the supernatant equaled zero. The humic-clay complexes obtained with irreversibly adsorbed HS were kept in 0.001 M CaCl2 at pH 5.5 in dark. The content of total organic carbon (OC) in the humic-clay complexes was determined and was initially 0.43% in Ca-kaolin and 0.17% in Ca-montmorillonite.

2.5 Static sessile drop method

A sessile drop method was used that has been described elsewhere (Wu 2001; Shang et al. 2008). We used a microscope slide glass (7.6 cm × 2.5 cm) that was cleaned with acetone and distilled water. The HS-clay stock suspensions at a concentration of 1–2% (by weight) were rigorously shaken for several minutes. Then, 50 μL of the suspension was placed onto the slide, left to dry for several days, and then placed in an oven at 105 °C for 2 h immediately prior to analysis. Five to seven HS-clay samples were placed onto one slide.

Contact angles were measured using a goniometer (Drop Shape Analysis System, DSA100, Kruess GmbH, Hamburg, Germany) equipped with a microsyringe steel needle. For static contact angle measurements, the microsyringe needle was positioned at 0.2 mm from the surface of the colloidal film; the volume of water drop used was 1.5 μL. The distance between the needle deposit and standby position was 1.0 mm. After dispensing, the drop shape was monitored with a digital camera for 10 s at 18 frames per second, and the contact angle was recorded. The drop contour was mathematically described by the Young–Laplace equation using DSA100 software, and the contact angle was computed as the slope of the contour line at the three-phase contact point. For each sample, a minimum 15 measurements were made, and the maximum number of replicates was 55. Each sampling was divided into three sub-samplings. The contact angle measurements were carried out on different days for each sub-sampling. Statistical analysis of the data obtained was performed using the software package Statistica for Windows.

3 Results and discussion

3.1 Characteristics of the humic materials used in this study

Four humic materials were used in our study for modification of clay minerals. The sources of humic materials included coal and peat, and two different soils included chernozem and sod-podzolic soil. The structural characteristics of the humic acids isolated from these sources are summarized in Tables 1 and 2.

Table 1 Elemental composition and molecular weight characteristics of the humic materials used in this study
Full size table
Table 2 Structural group composition of the humic materials used in this study as measured by 13C NMR spectroscopy
Full size table

The H/C atomic ratio in the humic materials studied varied from 0.7 to 1.14 (Table 1). An increased contribution of aromatic structures, which are reflected by lower values of H/C ratio, was observed for the coal and chernozem humic materials. The lowest O/C ratio (0.38) was observed for coal HS, whereas the highest one (0.7) was observed for peat HS. These ratios may reflect high contributions of oxygen-rich aliphatic polysaccharide structures into the molecular composition of peat HS as compared to the aromatically rich coal HS. The C/N ratio was highest in coal HS that was depleted in nitrogen. In general, the elemental compositions of the HS under study were in agreement with prior reports for similar sources (Übner et al. 2004; Tikhova et al. 2008; Kholodov et al. 2011; Mylotte et al. 2015).

The molecular weight (M w) characteristics of the HS samples under study are shown in Table 1. The higher M w values were observed for peat and sod-podzolic soil HS as compared to those from coal and chernozem yielding the following ascending trend: coal < chernozem < peat < sod-podzolic soil. This is consistent with the higher content of bulky polysaccharide chains in peat and sod podzolic soil as compared to coal and chernozem HS rich in compact aromatic structures. The values obtained were also well corroborated with those reported in the literature (Perminova et al. 2003; Yuan et al. 2013).

The HS samples were analyzed using 13C NMR spectroscopy in order to obtain direct evidence for structural differences. Data on the structural composition of the HS are given in Table 2. It can be seen that the highest content of aromatic carbon was observed for the coal and chernozem HS. At the same time, the sod-podzolic soil and peat HS had the highest content of heteroatom (O, N) substituted carbons, making up aliphatic moieties of carbohydrates and peptides. This is consistent with the elemental and SEC analyses for the same samples.

In summary, the data summarized above allows us to suggest that the hydrophobicity of the humic materials studied was increasing in relation to an increase in aromatic carbon content in the following order: peat < sod-podzol < chernozem ≈ coal. The humic materials could also be grouped into pairs with similar structural features: “peat–sod-podzolic soil” and “coal–chernozem soil.”

3.2 Modification of clay minerals with the humic materials from different sources

The typical sorption isotherms of the humic materials on clay minerals are shown in Fig. 1. Sorption experiments demonstrated that in case of kaolin, the sorption isotherms reached a plateau at HS equilibrium concentrations ranging from 100 to 300 mg L−1. In case of montmorillonite, the isotherms did not reach a plateau in the whole range of concentrations tested (up to 4000 mg L−1).

Fig. 1
figure 1

Sorption isotherms of coal HS onto clay minerals: a Ca-kaolin, b Ca-montmorillonite

Full size image

To obtain the HS-clay complexes, we used the maximum HS concentration, which was 500 mg L−1 for Ca-kaolin and 4000 mg L−1 for Ca-montmorillonite. The amount of irreversibly sorbed HS on the basis of organic carbon content was as follows (% OC):

HS-Ca-kaolin: K-T5—0.69; K-Pw08—0.63; K-CtK06—1.53; K-Pow—0.92;

HS-Ca-montmorillonite: M-T5—0.67; M-Pw08—1.86; M-CtK06—1.05; and M-Pow—1.36.

There was no clear relationship between the amount of HS irreversibly sorbed onto clay minerals and their structural characteristics once sorbed to kaolin and montmorillonite. This could be due to the phenomenon of selective sorption, where only a portion of the molecular ensemble of HS is adsorbed onto a clay mineral surface.

3.3 Contact angle measurements of the different clay minerals modified with humic substances

Visual images of the water droplets obtained for the bare clays and the HS-clay complexes are shown in Fig. 2. Much steeper contact angles were observed for the montmorillonite complexes as compared to the kaolin ones. In addition, whereas the contact angles of HS complexes with Ca-kaolin were very similar to the bare clay mineral, HS-Ca-montmorillonite complexes had higher contact angles than those of non-modified clay. The corresponding contact angle values are shown in Fig. 3.

Fig. 2
figure 2

Contact angles of Ca-kaolin and its HS-complexes (left panel): K-T5—peat HS, K-Pw08—sod-podzol soil HS, K-CtK06—chernozem HS, K-Pow—leonardite HS; Ca-montmorillonite and its HS complexes (right panel): M-T5—peat HS, M-Pw08—sod-podzol soil HS, M-CtK06—chernozem HS, M-Pow—leonardite HS

Full size image
Fig. 3
figure 3

Box whisker plots of the average contact angle values determined for the surfaces of HS-modified kaolin (a) and montmorillonite (b)

Full size image

The contact angle of non-modified Ca-kaolin (K) was 27°. This value is consistent with prior reports measured by a static sessile drop method for this mineral (Shang et al. 2008). The same is true for Ca-montmorillonite (M): the contact angle of 41° was very close to the previously reported ones (Wu 2001; Shang et al. 2008). HS modification rendered both clays under study—kaolin and montmorillonite—more hydrophobic. In case of kaolin, this increase was very small: from 27° to 31°–32° (t test at α = 0.01). We suggest that low surface area and low cation exchange capacity (Velde and Meunier 2008) might preclude strong binding of Ca-kaolin with acidic functional groups of HS via cation bridges and favor weak hydrophobic bonding between HS and Ca-kaolin surfaces. This assumption is consistent with the larger content of adsorbed HS on Ca-kaolin for more hydrophobic HS from chernozem and coal as compared to more hydrophilic HS of peat and sod-podzolic soil (Badun et al. 2009; Kulikova et al. 2010). As a result of limited modification, we did not observe relationships between the amount of HS irreversibly sorbed to Ca-kaolin and associated properties (Tables 1 and 2).

In case of Ca-montmorillonite, the contact angles of all HS-montmorillonite complexes were significantly larger (t test at α = 0.01) as compared to Ca-kaolin, which ranged from 53° to 83°. The contact angles varied according to the following ascending trend: Ca-montmorillonite < chernozem HS-montmorillonite < coal HS-montmorillonite < peat HS-montmorillonite < sod-podzol HS-montmorillonite. Moreover, this trend depended on the structural characteristics of the adsorbed HS. Statistically significant correlations were observed with molecular weight (M w) and the content of aromatic carbon in the humic materials used (Fig. 4).

Fig. 4
figure 4

Correlation of the contact angle (CA) of the HS-montmorillonite complexes versus molecular weight of HS (a) and the content of aromatic carbon (b) in the humic materials used

Full size image

The higher molecular weight HS produced more hydrophobic coatings (Fig. 4a). At the same time, reverse correlations were observed for the contact angle (CA) value, based upon the hydrophobicity of HS estimated from the content of aromatic carbon (Fig. 4b). This can be explained by attachment of hydrophilic functional groups of HS (e.g., carboxyls, phenols) to positively charged surfaces of the clay mineral. This could cause a decrease in wettability of the Ca-montmorillonite-HS complexes. Moreover, at pH 5.5 in the presence of Ca2+, an intermolecular aggregation of HS could occur at the surface of the clay mineral. This phenomenon becomes even more pronounced upon drying of organo-mineral complexes during dehydration, which can cause a change in orientation of the organic adsorbate at the clay surface (Jouany 1991), based on the surface free energy properties of the clay-organic complexes. This prior study also observed very little influence from the chemical composition of organic matter onto surface properties of organo-mineral complexes of Ca-montmorillonite.

We conducted factorial analysis of variance (factorial ANOVA) (Table 3) to determine if there are other factors contributing to the value of contact angle.

Table 3 Univariate tests of significance, effect sizes, and powers for contact angles
Full size table

The low estimates of p values obtained in our analysis may indicate significant impacts from other factors (clay mineral and HS type) on the contact angle value based upon the F test. The estimates of effect size and power factors (e.g., partial eta-squared and non-centrality) indicate that apart from the source of HS, the clay mineral type makes a significant contribution on the contact angle value. The statistical analysis confirms that the contact angle values of HS-modified kaolin did not depend on the properties of HS, whereas for montmorillonite complexes, this dependence was much more evident.

4 Conclusions

Application of a sessile drop method demonstrated that irreversible sorption of humic substances onto surfaces of clay minerals is accompanied by a decrease in wettability or an increase in hydrophobicity. This evidence was obtained both for 1:1 (kaolin) and 2:1 (montmorillonite) clays, whereas much more substantial hydrophobization was seen with montmorillonite as compared to kaolin. At least moderate hydrophobization of the hydrophilic surfaces of clay minerals was observed with all types of humic materials used in this study (the contact angle values did not exceed 83°). This may indicate the important role of HS in sustaining soil structure via formation of water stable soil macroaggregates. Recently, Volikov et al. (2016), hypothesized that soil macroaggregates might be considered to be an inverse Pickering emulsion (i.e., an emulsion that is stabilized by solid particles) with water in air stabilized by clay particles that are modified by soil organic matter. This phenomenon might be similar to formation of “dry water” stabilized by hydrophobized particles of silica gels reported by Binks and Murakami (2006). The significance of wettability for water distribution and soil organic matter decomposition was demonstrated by Goebel et al. (2007). Our study provides additional arguments in favor of this hypothesis, including the theoretical rationale for the particularly beneficial effect of soil amendments capable of moderate hydrophobization of clay surfaces, e.g., mineral-adhesive HS (Volikov et al. 2016) or other humic-based hydrophobic amendments (Piccolo and Mbagwu 1999).