Adsorption-based approach to determine the size and mass of humic acids molecules

Abstract

The state of sodium humate in aqueous solution and on a solid surface is studied using various methods. It is shown that in aqueous solution at neutral pH and low ionic strength this substance exists in an associated state. The values of molecular mass of the molecules in the preparation studied (Aldrich) are determined to be 1, 3–6 and 15 kDa. A new adsorption-based approach for initial estimation of the molecular mass and size of humic acid molecules is proposed, which can be used instead of any specific sophisticated equipment. The size of the molecules of the studied preparation estimated in this way (in the range of 2.3 to 9.0 nm), agree well with data obtained by chromatographic and electron microscopic measurements, which confirm the reliability of the proposed method.

Introduction

Humic acids (HAs) are of a paramount importance for various geochemical, biological and biochemical processes, and are used in a number of industrial and scientific applications. This implies the extensive and versatile studies of HA published in the scientific literature. However, many fundamental questions related, in particular, to the physicochemical characteristics of HA molecules are yet to be answered. It is to be noted that the data on the molecular mass and size of HA molecules reported in literature are rather confusing. For example, the values of the molecular mass of HA supplied by Aldrich as determined by gel permeation chromatography determined by various authors were in the range of 1 to more than 100 kDa, see e.g. Ochs et al. (1994), Vermeer and Koopal (1998); also the average value of the molecular mass (21 kDa) obtained in Vermeer and Koopal (1998) was confirmed using a viscosimetry method. In another study (Tanaka, 2012), the same preparation was separated by sequential ultrafiltration into two fractions with a molecular mass of 30–100 and above 100 kDa, respectively. This wide range of values obtained in different studies could be attributed not only to a possible inhomogeneity of the preparations, but also to differences in the experimental conditions. It was shown by Murphy et al. (1994) that the ionic strength, pH value and the cationic charge affect the apparent value of the HA molecular mass calculated from chromatograms.

Similar confusion exists also regarding the size of HA molecules. The values obtained using different methods (Baalousha et al., 2005; Kawahigashi et al., 2005; Manning et al., 2000; Osterberg and Mortensen, 1992) vary within the range from 5 to 850 nm; it should be noted that an uncertainty remains in this regard whether the particles measured in each particular case are separate molecules or supramolecular entities. Quite expectedly, additional problems in the determination of the size and mass of HA molecules are caused by the fact that humic acids exhibit self-organisation properties (Baigorri et al., 2007; Piccolo, 2001). It was shown recently (Tarasevich et al., 2013) that even at concentrations by 3–4 orders of magnitude lower than the critical micelle concentration (CMC) the HAs are capable of binding anionic organic compounds due to the formation of supramolecular structures.

To summarise, studies of the state of HA in solutions and on solid surfaces, their molecular composition and the principal characteristics of their molecules are of relevance. In the present work, we propose a new simple method for the determination of the mass and size of HA molecules.

Experiment

The preparation of the sodium salt of humic acids (purchased from Aldrich) was made without additional purification. The aqueous solutions were prepared with distilled water without addition of electrolyte. This was done to avoid the influence of the electrolytes on the humate structure, following the results presented by Murphy et al. (1994).

The adsorption of sodium humate (HNa) from highly diluted solutions was performed on kaolinite from the Gluchiv deposit (Sumy province, Ukraine) which is characterised by high purity (the mineral content in the sample is 99 %) as studied earlier by Tarasevich (1988). Air-dry samples of kaolinite (mass 0.1 g, fraction < 0.2 mm) were dispersed in 50 cm³ of aqueous HNa solution at pH ≈ 6.0. After shaking the dispersion over 12 hours, the solid phase was separated from the liquid by centrifugation at 8000 r/min. The HNa contents in the aqueous phase was determined spectrophotometrically at λ = 260 nm.

The size and electrokinetic potential of HNa particles in aqueous solution were determined by dynamic and electrophoretic light scattering, respectively, using the Zetasizer (Malvern Instruments) with He–Ne laser (λ = 633 nm, maximum power 4 mW) controlled by the software Version 6.20. The measurements were made in the concentration range of 0.32 mg/dm³ to 20 g/dm³ at 25 ℃ and pH 6.0. Each measurement was repeated five times; the results were averaged for subsequent analysis.

The electronic microphotographs of HNa particles were taken by a transmission electron microscope JEM JEOL 1230 with an accelerating potential of 100 kV and a digital camera Gatan. To prepare the samples, a fixed volume of HNa solution (50 mm³) with concentrations of 0.32 mg/dm³ and 20.16 mg/dm³, respectively, was deposited on the surface of the copper blend covered by a carbon pre-shadowed Formvar film. After 10 minutes the main portion of the solution was removed from the blend surface by filter paper, and the residue was dried in a warm air flow.

The molecular mass distribution (MMD) of HNa was measured by a high-performance liquid chromatography method using the liquid chromatograph HP1050/DAD (Hewlett-Packard) in the gel permeation regime with a thermostatically controlled column (Waters) filled by the gel sorbent Ultrahydrogel™ 250 at 25 ℃. To perform the analysis, 5 cm³ of an aqueous HNa solution with a concentration of 80 mg/dm³ was diluted by 5 cm³ of acetonitrile and 5 cm³ of sodium-phosphate buffer (pH 8.6), then an aliquote (1 cm³) was diluted with the mobile phase (sodium-phosphate buffer with 10 % vol. of acetonitrile, pH 6.6) in the volumetric ratio 1:1, immediately followed by the chromatographic measurement.

Results and discussion

To identify correctly the humate particles, their size distribution in aqueous solutions was studied in a wide concentration range using dynamic light scattering. It is seen from Table 1 that the characteristic sizes and volumetric ratio of the humate particles depend on the solution concentration. We believe that the largest particles detected (microparticles, d = 1.6–5.6 µm, see Table 1) are HNa micelles (also sometimes called micelle-like structures). This is indicated by the increase of the ζ potential (Figure 1, curve 1) correlated with the decrease of the surface tension of the solution (Tarasevich et al., 2013) (Figure 1, curve 2); both these trends become apparent when the microparticles appear in the solution and their volumetric ratio increases (see Table 1, C_HNa ≥ 20 mg/dm³). It should be noted that the microparticles (micelles) are detected in HNa solutions even at a concentration which is by 2.5 orders of magnitude lower than the CMC (8 g/dm³).

Table 1.

Dependence of the characteristic diameter and volumetric ratio of humate particles on the HNa concentration C_HNa in the solution.

CHNa, mg/dm3	Nanoparticles		Submicroparticles		Microparticles
	Characteristic diameter d1, nm	Volumetric ratio	Characteristic diameter d2, nm	Volumetric ratio	Characteristic diameter d3, µm	Volumetric ratio
0.32	30–55	0.37	202–467	0.63	–	–
3.20	35–80	0.33	254–492	0.67	–	–
4.80	32–64	0.31	273–409	0.69	–	–
6.40	30–116	0.29	277–582	0.71	–	–
20.16	60–100	0.32	217–312	0.65	4.9–5.6	0.03
80.00	44–149	0.42	256–536	0.54	5.0–5.6	0.04
2000	62–114	0.17	546–695	0.67	2.9–4.6	0.16
20,000	56–95	0.06	269–540	0.13	1.6–4.3	0.81

HNa: sodium humate.

Figure 1.

Dependencies of the ζ potential of humate particles surface (curve 1) and surface tension of HNa solutions (curve 2) on the solution concentration.

The submicroparticles observed in the solution could be identified as supramolecular associates of humate. Even the smallest HNa particles detected in the solution (nanoparticles, d ≥ 30 nm, see Table 1) are obviously supramolecular HA associates rather than individual HA molecules, which is seen from their quite large size.

These conclusions are well supported by the electron microscopic images which exhibit relatively large particles (40 nm to few micrometers, see Figure 2(a)); also at low concentrations of the studied HNa solution smaller particles, less than 10 nm in diameter, are observed (see Figure 2(b)). These smaller particles are obviously formed by the disintegration of larger humate particles (supramolecular associates) during their adsorption on the blend surface. These particles could be presumably identified as individual HA molecules. It should be kept in mind, however, that the drying and evacuation of the samples which result in the loss of bound water could lead to some decrease of the particles; at the same time, the size of particles could increase due to their mutual adhesion. These possible factors, and also insufficient clearness of images at high magnification prevent us from an accurate and reliable determination of the size of primary humate entities (molecules).

Figure 2.

Electron microphotographs of HNa at the concentrations C_HNa = 20.16 mg/dm³ (a) and C_HNa = 0.32 mg/dm³ (b).

More definite information could be obtained from the chromatographic data about the humate molecular mass. It is known that the accuracy of determination of the humate molecular mass depends on the ionic strength of the solution (Murphy et al., 1994). Therefore, to suppress the dissociation of acidic groups and to promote the disintegration of the humate particles into the molecular state, during the preparation of HNa for the MMD analysis, high concentrations of electrolytes containing univalent cations were added to the solution. The chromatogram shown in Figure 3 exhibits a peak at 1 kDa, a sequence of small peaks at 3 to 6 kDa and a peak at 15 kDa. The sizes of these molecules are relatively small; these results also agree with the data obtained by electron microscopy as discussed above.

Figure 3.

HNa molecular mass distribution chromatogram; C_HNa = 80 mg/dm³, detection wavelength λ = 214 nm.

Keeping in mind the influence of the sample preparation process on the MMD results, we performed also dynamic light scattering measurements of the HNa samples under conditions similar to those used in the chromatographic experiments. The smallest particles detected in these measurements were 2.3 nm in size, which obviously corresponds to a molecular mass of 1 kDa; however, it was impossible to determine the size of molecules with larger mass.

In view of the problems outlined above with the determination of size and mass of the primary humic acid particles (molecules), we propose a new simple adsorption-based approach which does not require the use of any sophisticated equipment. This method comprises (i) the adsorption of HA from highly diluted solutions onto the substrate with known specific surface area; (ii) the fitting of the initial portion thus obtained adsorption isotherm to the Langmuir equation; and (iii) the subsequent calculation of the molecular mass or size of HA molecules based on these data. The condition necessary for this method is the formation of a HA monolayer on the sorbent’s surface. To ensure this, the HA molecules in the solution should exist in a non-aggregated state (pH value should be above 5, and additional electrolytes, especially multi-charged cations should be absent). Also a sufficient number of active centres capable for the interaction with HA should be present on the sorbent’s surface. The concentration of the HA solution should be low enough to prevent the adsorption of HA associates at the sorbent’s surface, because at low HA concentrations these associates, if present in the solution, become dispersed into separate molecules when adsorbed on the sorbent’s active centres.

The kaolinite from the Gluchiv deposit used in this study meets the requirements of a sorbent defined above. The specific surface area of its side faces, as determined by electron microscopy is S_S = 6.0 m²/g (Rusko and Yu, 1976). At neutral pH, the behaviour of humic acids is similar to that of negatively charged (anionic) polyelectrolytes (Theng, 2012). Therefore, at low HA concentrations in the solution the adsorption of HA on kaolinite could be treated as an anionic exchange between the OH groups located on the side faces of the mineral crystallites and the acidic groups of HA according to the scheme:

$[\text{Sidesurface}-\text{Al}({\text{H}}_2\text{O})(\text{OH})]+\text{R}-\text{COOH}\to [\text{Sidesurface}-\text{Al}({\text{H}}_2\text{O})(\text{R}-\text{COO})]+{\text{H}}_2\text{O}$

The initial portion of the HNa adsorption isotherm on kaolinite is shown in Figure 4(a). Clearly, in the low concentration range (below the sharp inflection point) the adsorption of individual HNa ions occurs, because these ions possess a maximum affinity to the surface. The inflection point at C_HNa ≈ 5 mg/dm³ indicates the onset of the adsorption of HNa associates.

Figure 4.

Initial section of the HNa adsorption isotherm on kaolinite (a) and its processing in the Langmuir equation coordinates; (b) α is the slope angle.

From the initial portion of the isotherm re-plotted in the Langmuir equation coordinates (see Figure 4(b)) the arbitrary monolayer capacity is calculated as a_m = 1/tgα = 1/0.43 = 2.33 mg/g. The HNa concentration on the kaolinite side faces amounts to [C] = a_m/S_S = 2.33/6.0 = 0.388 mg/m²= 3.88∙10⁻²² g/nm². Therefore, the amount of humate adsorbed on the area S = 4.15 nm² which corresponds to the area of the smallest humate particle detected by dynamic light scattering (2.3 nm in size) is P = [C]∙S = 3.88∙10⁻²²∙4.15 = 16.1∙10⁻²² g. The molecular mass of this HA entity is M = P∙N_A = 16.1∙10⁻²² 6.023∙10²³= 969.8 g/mol, which is very close to the value of 1 kDa estimated from MMD measurements.

A reciprocal calculation could also be performed. The maximum molecular mass obtained by MMD measurements (M = 15 kDa) is P = M/N_A = 15000/6.023·10²³= 2490.45·10⁻²³ g. The area covered by this molecule on the kaolinite surface is S = P/[C] = 2490.45·10⁻²³/3.88·10⁻²²= 64.19 nm², which (assuming spherical molecules) yields a diameter of 9.04 nm. This value agrees well with the size of primary entities (molecules) of HNa (< 10 nm) estimated using transmission electron microscopy.

Conclusions

Various methods are used to study the state of HNa (Aldrich) both in aqueous solution and on solid surfaces. It was shown by dynamic light scattering that in aqueous solution at neutral pH values and low ionic strength the HNa exists in the associated state. In the concentration range of 0.32 to 20000 mg/dm³ the size of HNa particles varies from 30 nm to 5.6 µm, and the distribution of particles with respect to size depends on the concentration.

Using electrophoretic light scattering it was found that the electrokinetic potential of HNa particles increases with the increase of the solution concentration. It is essential that a sharp increase of the ζ potential which correlates with the decrease of surface tension of the solution was observed when the particles of micron size (micelles) appear in the solution at a concentration by 2.5 orders of magnitude lower than the CMC.

The electron microscopy data show that when the HNa is adsorbed from dilute solutions (0.32 mg/dm³), particles less than 10 nm in diameter are present on the solid surface of the copper blend covered by a carbon pre-shadowed Formvar film. For the adsorption from more concentrated solutions particles from 40 nm to several micrometers in size were detected.

Using a high-performance liquid chromatography method it was found that the samples studied contain molecules with masses of 1, 3–6 and 15 kDa.

The adsorption-based approach proposed in this study enables one to determine the mass and the size of humate molecule. The method involves the adsorption of HA from very diluted solutions onto a substrate with known surface area, followed by the fitting of the initial section of the isotherm by the Langmuir equation to determine the arbitrary monolayer capacity and the surface concentration of HA, and by the subsequent calculation of the HA molecular mass from its known size or, vice versa, by the calculation of the size of HA molecule from its known molecular mass.

The proposed method is promising for the initial estimation of the molecular mass and size of HA molecules. The method does not require any special equipment (electron microscope, laser analyser, liquid chromatograph). The mathematical processing of the experimental data involves only the well-known and easily treatable Langmuir equation and enables one to obtain the required values in a straightforward way.

The sizes and molecular masses of HNa obtained in this way agree well with the data obtained by chromatographic and electron microscopic measurements, which confirm the reliability of the proposed new simple method. In particular, the calculations show that the diameter of the smallest molecules of the samples studied (with molecular mass 1 kDa) is 2.3 nm, while for the largest molecules (15 kDa) the size is 9.0 nm.