SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 January 2007
Published in Soil Sci Soc Am J 71:86-94 (2007)
DOI: 10.2136/sssaj2005.0232
© 2007 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jeong, C. Y.
Right arrow Articles by Lim, S. K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Jeong, C. Y.
Right arrow Articles by Lim, S. K.
Agricola
Right arrow Articles by Jeong, C. Y.
Right arrow Articles by Lim, S. K.
Related Collections
Right arrow Soil Surface Chemistry
Right arrow Soil Models
Right arrow Soil Chemistry

SOIL CHEMISTRY

Carboxylic Content of Humic Acid Determined by Modeling, Calcium Acetate, and Precipitation Methods

Chang Yoon Jeong

Dep. of Renewable Resources, Univ. of Louisiana, Lafayette, LA 70504

Chan Won Park

National Inst. of Agric. Sci. & Technology, 249 Seodun-dong, Suwon, Korea

Jeong-Gyu Kim* and Soo Kil Lim

Div. of Environ. Sci. and Ecological Eng., College of Life and Environ. Sciences, Korea Univ., 1,5-Ka, Anam-dong, Sungbuk-ku, Seoul, Korea

* Corresponding author (lemonkim{at}korea.ac.kr).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The key to understanding proton equilibria and cation binding on the surface of humic acid is the quantification of the total carboxylic acid contents. Such quantification via titration alone is highly ambiguous and not straightforward due to the complexity and interference of noncarboxylic acids (phenolic acid, thiols, etc.). To gain a greater understanding of such analysis, potentiometric titration data was incorporated into modeling approaches, and they were compared with two other methods: conventional Ca(OAc)2 exchange and the precipitation titration method. Each of these three methods was used to analyze six different humic samples. The humic samples were collected from various sources, such as upland, paddy fields, sediments, and peat. The carboxylic acid contents ranged from 263 to 487 cmolc kg–1. Furthermore, the carboxylic acid content in the humic acid from the Namwon series showed significantly higher values. This was attributed to the relatively higher percentage of organic C in the mother material. Two modeling approaches using titration data were performed. The first was a simple electrostatic model with limited functional group heterogeneity (Model A). The second was our implemented version of Model V with four carboxylic acid group sites and four phenolic hydroxyl group sites. The relative difference in total carboxylic acid contents between the Ca(OAc)2 method and Model A prediction ranged from 0.21 to 7.43%, and that between the Ca(OAc)2 method and Model V prediction was 1.90 to 35.71%. The predicted carboxylic acid content from Model A was more comparable to the results of the Ca(OAc)2 method and the cetyltrimethylammonium method than the Model V prediction. The modeling approaches with the simple ligand model showed a better prediction of the total carboxylic acid content from the potentiometric titration data.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Humic acid is a major fraction of soil organic matter, and its chemistry has been studied extensively. The acidic content of humic acid with respect to the chemical speciation, mobility, and bioavailability of trace metals is important (Christensen et al., 1998). The effect of this acidity (which originates from the different functional groups, mainly carboxylic acids and weak acid groups) must be understood and modeled quantitatively to predict the mobility of cations and heavy metal pollution. Numerous studies have attempted to determine the carboxylic acid content of humic acid to distinguish it from other weak acid groups (Perdue et al., 1980; De Nobili et al., 1989, 1990).

The most commonly used method is the indirect titration method, in which the amount of acetic acid generated from the reaction of Ca(OAc)2 with humic acid is measured. Complexation of Ca2+ with the humic acid increases the release of protons to solution, increasing the apparent acidity (Perdue et al., 1980). The most important step in this method is filtration to remove all solids from the solution before titration. The critical operational problem associated with this method is the elimination of soluble Ca humate salts (Davis, 1982, Perdue, 1985).

De Nobili et al. (1989, 1990) used the chemical precipitation method to determine the carboxyl group of humic substances in solution using cetyltrimethylammonium (CTA+). The number of CTA+ ions bound per unit weight of humic acid increased with pH from 5.5 to about 7.0 and remained fairly constant up to about pH 9. After that threshold, the number of CTA+ ions increased again. They compared the values obtained from the precipitation method with those obtained by the Ca(OAc)2 method. They concluded that the Ca(OAc)2 method overestimated the carboxylic acid content compared with the CTA+ precipitation method.

Several procedures have been used for potentiometric acid–base titrations of humic acids (Cabaniss, 1991; Marshall et al., 1995; Stevenson, 1994; Swift, 1996). They include batch titration (in which case the pH is measured on a series of samples containing increasing amounts), continuous titration using an automatic titrator, and normal acid–base titration with a conventional pH meter. A typical titration curve of humic acid generally shows a featureless smooth line. The titration curve can be divided into three different zones—dissociation of the carboxylic group, phenolic OH and other very weak acid types, and the intermediate area where ionization of carboxylic and weak acid groups overlap (Stevenson, 1994). The proper selection of the end point is difficult in interpreting titration curves of humic acids. Posner (1966) determined the end point to be the maximum rate of change of pH with added alkali. This point can be the initial zone of the overlapping carboxyl and weak acid groups, and it occurs between pH 7.0 and 7.6, depending on the ionic strength of the solution. This ambiguous end point in the titration curve could be explained by assuming that humic acids behave as weak acid polyelectrolytes. Therefore, to obtain a better interpretation of the humic titration curve, it is necessary to use a model to calculate the carboxylic content of humic acid.

Most of the models recently developed explain the affinity behavior of protons on the humic surface. The quantitative interpretation of the potentiometric titrations of humic acid is delicate, however, because humic acid is chemically heterogeneous and contains electrostatic interaction between its functional groups. Most of the developed ion-binding models have made various simple assumptions. The simplest and least satisfactory model treats humic acid as a mixture of simple ligands and does not consider electrostatics. It has been suggested that the fitted constants can be interpreted as average affinity constants that represent classes of functional groups (Pinheiro et al., 1994). Inevitably, these models are limited by their exclusion of electrostatic influences. The simplest form of the electrostatic interaction model included a carboxyl and a phenolic acid group for each class of acid group (Wilson and Kinney, 1977). Much attention has been focused recently on extending this model to include several types of functional groups. Tipping and Hurley (1992) developed the unified cation binding model (Model V) using the work of Marinsky and Ephraim (1986), Ephraim et al. (1986), and Ephraim and Marinsky (1986). A total of eight types of functional groups (four carboxyl groups and four phenolic hydroxyl groups) were invoked in Model V, and single carboxyl and acidic hydroxyl groups were assumed in Model A (Marshall et al., 1995). In addition, Bartschat et al. (1992) developed an oligoelectrolyte model by applying the nonlinear Poisson–Bolzmann equation to an impenetrable, spherical fulvic acid particle, while de Wit et al. (1990, 1993a, 1993b) used a master curve approach.

The objective of this study was to evaluate modeling exercises using potentiometric titration data to calculate the total carboxylic acid content in humic acid and to compare the results to two existing methods, the Ca(OAc)2 method and the precipitation method. Two different models—Model A and Model V—were applied to the experimental data.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Humic Extraction
Humic acids from different types of land and formation of soils, horticultural peat soil, and sediment were prepared (Table 1). Bacsan and Jisan soils were collected at Kimpo and Hwasung province, Korea, respectively. Nonvolcanic (Donghong) and volcanic (Namwon) soils were collected at Jaeju Island, where the background of soil formation is completely different. Horticultural peat soil was selected for humic acid (HA) extraction, and a sediment sample was collected from Chungpyung Lake, upstream of Han River, Korea.


View this table:
[in this window]
[in a new window]

  		Table 1. Physicochemical properties of the soils.

 
Twenty 500-mL bottles were filled with 100 g of sieved soil for humic extraction. Extraction was conducted for 18 h by shaking with enough 0.1 M NaOH (1:10 soil/0.1 M NaOH) to fill a 500-mL glass bottle, and the cap was tightened with O2-free N2 gas. The extract was centrifuged at 9000 x g for 15 min and then separated. The supernatant was immediately acidified to pH 2 with concentrated HCl. The precipitated fraction, which contained humic acid, was separated from the supernatant, which contained the fulvic acid fraction, by centrifuging and packing it into tubes of Visking dialysis membrane (molecular weight cutoff = 12000, Medicell International, London). The dialysis bags were transferred into a mixture of 1% (v/v) HCl and 1% (v/v) HF to remove any metallic contaminants and silica. The acid solution was changed twice at 1-wk intervals. The acid treatment was followed by dialysis against deionized water, and this was repeated with fresh deionized water at 2-d intervals until the conductivity of this solution was <2.0 mS m–1 (comparable to fresh deionized water, which was approximately 1.6 mS m–1). The purified humic acid was freeze-dried and stored in an evacuated desiccator in the dark (Hayes, 1985).

Preparation of Humic Acid Stock Solution
Freeze-dried humic acid was dissolved in a solution of NaOH with the required background electrolyte concentration. Before dissolving the humic acid, the concentration of NaOH in solution was measured by replicated titration of a 25-mL sample against the standard 0.05 M HNO3. The concentration of NaOH was chosen so that the final solution pH was about 10. For example, in a humic acid stock solution of 2.00 g L–1, the OH– concentration was approximately 9 x 10–3 M. The humic acid was dissolved for at least 2 h with continuous stirring under N2 gas to achieve complete dissolution.

Characterization of Humic Samples
Elemental analysis of humic acids was performed by the CHNS elemental analyzer (Carlo Elba, Milan, Italy). The values of E4 and E6, the optical density of the HA solution at 400- and 600-nm wavelengths, respectively, in humic stock solution were measured with a spectrophotometer (Kumada, 1987) and the E4/E6 ratio, {Delta}logK and RF values were obtained ({Delta}logK = logK400 – logK600, where K is the optical density at 400 or 600 nm; and RF = K600(1000/C), where C is the amount (in milliliters) of 0.1 M KnO4 consumed by 30 mL of HA solution used for determining the adsorption spectrum).

Total Acidity
One gram of humic acid was dissolved in 50 mL of 0.20 M NaOH solution. Four replicate aliquots (10.00 mL) of humic acid solution were each mixed with 10 mL of 0.40 M BaCl2 solution. After overnight shaking under N2 gas, these were centrifuged at 2500 x g for 20 min. Four milliliters of supernatant were then mixed with approximately 10.0 mL of deionized water and titrated with 0.10 M HCl solution. Humic acid was dissolved in 10.0 mL of 0.20 M NaOH solution and then mixed with 10.0 mL of 0.40 M BaCl2 solution. This was titrated with 0.10 M HCl to an end point of pH 8.40. A mean value was taken from three replicate titrations. A blank (without HA) was also titrated (Schnitzer, 1972).

Calcium Acetate Method
Five hundred milligrams of humic acid was placed in a 250-mL ground-glass stoppered Erlenmeyer flask, and 100 mL of 0.1 M (CH3COO)2Ca·H2O and 40 mL of CO2-free deionized water was added to the flask. The blank test (without the humic acid) was performed separately. The flask was shaken for 24 h at room temperature, and then 20 mL of supernatant was titrated to pH 9.80 using standardized 0.1 M NaOH after centrifuging at 17400 x g for 20 min.

Cetyltrimethylammonium Method
Cetyltrimethylammonium, a cationic detergent, was used to analyze the carboxylic acid group of humic acid (De Nobili et al., 1990). Standard humic stock solution (0.2 g organic C L–1, pH 7) was added to different amounts of 0.1% CTABr (cetyltrimethylammonium bromide) in test tubes and produced 12 different CTA+/organic C ratios. Suspensions were left standing for 18 h at 25°C in the dark before centrifugation at 17400 x g for 30 min. Absorbance was measured at 400 nm and the number of carboxyl groups was determined to be at the minimum absorbance that coincides with the quantitative precipitation with the same number of CTA+ ions.

Batch Titrations—Downward
Titration aliquots of HNO3 (0.05 M) were dispensed into 50-mL glass bottles with an Orion 960 autoburette (Thermo Electron Corp., Waltham, MA). The volume of titrant was increased by 0.05 mL at each step up to 2.00 mL in duplicate. Ten-milliliter samples of the humic stock solution were added to each bottle. Each bottle was flushed and sealed with the N2 gas and stored at 20°C for 7 d in the dark. The bottles were shaken manually once a day. After 7 d, the pH of the sample solution was measured under the N2 gas.

Batch Titrations—Upward
Upward batch titrations were performed on acidified bulk samples of humic acid stock solution. The initial pH was adjusted to 3.0, and the solution was left standing for 1 h after being well shaken. Ten milliliters of acidified solution was dispensed into 25-mL bottles, and then various amounts of NaOH titrant were added. After 7 d, the titration was completed.

Continuous Titrations
The titrations were performed with 10.0 mL of stock humic acid solution from high to low pH at an ionic strength (I) of 0.1 M NaNO3, and temperature was maintained at 25°C using a water jacket that surrounded the titrant vessel. Continuous titration was performed for >2 h, with the subsequent addition of approximately 2.0 mL of titrant.

Surface Charge Calculation
The surface charge on humic acid is estimated using the ion charge balance, which is derived from the titration data (pH and titrant volume). The details of the surface charge equation have been presented elsewhere (Marshall et al., 1995). The distribution of ions between the double layer and the bulk solution are considered in the model calculations (Tipping and Hurley, 1992, Marshall et al., 1995). As a result of this method, a more refined calculation may be applied to humic charge expressions. Tipping and Hurley (1992) expressed the volume (VD) of the diffuse double layer (DDL, in L g–1 of humic substances) by the following equation:

Formula 1[1]
where k is the Debye–Huckel parameter (m–1) and is approximately 3.29 x 109{surd}I, N is Avogadro's number, r is the radius of the humic molecule, and M is the humic molecular weight. The assumed value of the molecular weight was 15000, and the radius of the humic molecule was taken as 1.72 x 10–9 m (Tipping and Hurley, 1992).

The calculation of the concentration of NO3– in the bulk solution, [NO3–]B, is calculated from the total concentration of NaNO3 and HNO3, assuming that the NO3– is confined to the volume outside the DDL:

Formula 2[2]
where VT is the total solution volume and VB is the bulk solution volume.

The amount of Na+ in the DDL (<Na+>D) is calculated by the difference between the total amount of Na+ and the amount of Na+ in the bulk solution:

Formula 3[3]
where the amount of Na+ in bulk solution, <Na+>B, may be calculated from

Formula 4[4]

The concentration of Na+ in the bulk solution is calculated from a charge balance:

Formula 5[5]
The concentration of Na+ in the DDL can be expressed as

Formula 6[6]
The concentration of H+ ions in the DDL ([H+]D) can be derived from a Donnan equation:

Formula 7[7]
where [Mz+] cation concentration with ion charge valence z+. Thus,

Formula 8[8]
and so the absolute value of the humic surface charge (|Z|, cmolc kg–1) can be calculated from

Formula 9[9]
Equation [9] assumes that Na+ and H+ are the only counterions.

Model A
Model A assumed that the humic surface contains a single carboxyl group and a weak acid hydroxyl group (mostly the phenolic hydroxyl group). The functional group heterogeneity of humic acids was described with the variation in the value of the intrinsic acidity constants with third-order polynomial expressions, and the electrostatic interaction factor was expressed as a linear function of log10(I) (Marshall et al., 1995; Jeong, 1998).

A total of 11 adjustable parameters, including total carboxylic acid content, were resolved with the batch titration data of humic surface charge against pH using Marquardt nonlinear optimization (Press et al., 1986). The total acid group concentration (TA) was measured from the humic acid. The total phenolic hydroxyl group concentration was calculated by the difference between TA and the total carboxyl group concentration (TC). Thus, the equation for net humic surface charge (|Z| , cmolc kg–1 HA) is

Formula 10[10]
where KintC and Kint{phi} are the intrinsic constant of the carboxyl group and the phenolic hydroxyl group, respectively, and {omega} is the electrostatic interaction factor.

The two intrinsic acidity constants, carboxylic group and phenolic hydroxyl group (KintC and Kint{phi}) were described as a third-order polynomial function of the square root of ionic strength ({surd}I) and the electrostatic interaction factor, {omega}, was calculated as a linear function of log10I.

Model V
Humic substances were considered to be rigid spheres of uniform size with ion-binding groups positioned on the humic surface in Model V. Model V included the aforementioned site heterogeneity. Site heterogeneity is described as having discrete sites, a range of intrinsic binding constants, and the formation of bidentate sites. The site heterogeneity was expressed with four mostly strong acid sites (referred to as Type A) and four phenolic hydroxyl group sites (referred to as Type B) and used spreading factors (Tipping & Hurley, 1992). The eight intrinsic binding constants (pKi) were expressed in terms of four constants (pKA, pKB, {Delta}pKA, and {Delta}pKB) as follows:

Formula 11[11]

Formula 12[12]
The 12 combinations of proton-binding sites were allowed to form bidentate complexes for rapid computation. The model also accounted for the electrostatic effects within the two classes (carboxyl and phenolic hydroxyl groups). The electrostatic interaction factor ({omega}) is described in terms of the empirical parameters P, Q, and the ionic strength (I) (Tipping and Hurley, 1992).

Formula 13[13]

There were seven total fitted parameters, including the total carboxyl group contents (TC), in the Model V application. Model V was incorporated into WHAM (Windermere humic aqueous model) to simulate catchment chemical behavior (Tipping, 1994). Tipping (1998) discussed the improved model of Model V (Model VI). Model VI demonstrated a more distributional approach in describing binding sites for metals. However, Model VI used the same base of discrete-site and electrostatic formulation and proton bindings. In this research, an alternative Model V was applied to analyze the parameters for proton and total carboxylic contents. The amount of the phenolic hydroxyl group (as weaker acids) was calculated from the difference between the amount of the total acid groups and the amount of the total carboxylic acid group (Jeong, 1998). Tipping and Hurley (1992) assumed that the total amount of phenolic hydroxyl groups was 50% of the content of the total carboxylic acid group.

Statistics
The four different methods of quantifying the content of total carboxylic acid in each sample were compared using the one-way analysis of variance and the Tukey–Kramer honestly significant difference (HSD) test with a significance level of p < 0.05. The JMP 5.0.1 statistical program was used for all analyses (SAS Institute, 2002).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Elemental Composition and Spectroscopic Properties in Humic Acid
The ash content of the extracted HAs was <0.5% in most of the samples. The C and N elemental composition of the humic acid varied from 50.2 to 61.1% and 2.97 to 5.05%, respectively. Proton and O ranged from 4.01 to 4.69% and from 32.4 to 41.4%, respectively (Table 2). The HA from Namwon, the volcanic ash soil, had a relatively higher percentage of C but was lower in N content than the other HAs. The HA from Jisan, a paddy soil, showed relatively higher N content than the other HA samples. Regarding HA in soil, Schnitzer and Khan (1972) also reported that C generally varied from 53.8 to 58.7%, H varied from 3.20 to 6.20%, N varied from 0.80 to 4.30%, O varied from 32.8 to 38.3%, and S varied from 0.10 to 1.50%. The Namwon HA presented a relatively high degree of humification (Table 2). The spectral analysis showed that HA samples from Bacsan, Donghong, and the sediment presented similar E4/E6 ratio and RF values.


View this table:
[in this window]
[in a new window]

  		Table 2. Elemental composition and spectroscopic properties in humic acid (HA) samples.{dagger}

 
Batch and Continuous Titrations
Figure 1 presents the results of charge development curves in the batch downward, upward, and continuous downward titrations with Namwon HA at 0.1 M NaNO3 ionic background. Discrepancy was observed in the results of the batch and continuous titrations. Two different batch titrations, the downward and the upward experiments, were coincidental. The results from these batch titrations showed a larger surface charge than from the continuous titration. Similar results were observed from the other five HA samples. Marshall et al. (1995) discussed that the discrepancy of results between the batch and the continuous titration methods may be the result of the different protonation and deprotonation reactions on humic surfaces during each method. Evidence exists that the continuous titration did not reach an equilibrium state within the time period of the experiment (about 2 h). The batch downward titration was adopted for this research since the batch downward and the batch upward titrations did not show hysteresis.


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 1. Comparison of charge development curves in batch downward, batch upward, and continuous titration for Namwon humic acid with a 0.1 M NaNO3 ionic background.

 
Surface Charge Analysis and Parameter Estimation from Models
Figure 2 shows the corresponding model charge development curves produced using Model A. Solid lines present the fit of Model A, which uses all three or four titrations analyzed at different ionic strengths (0.01, 0.05, 0.1, and 3.0 M NaNO3) as a single combined data set. This model described the dissociation of all six HAs well through the range of ionic strengths. The effect of changing the ionic strength from 0.01 to 1.0 M is shown as the surface charge on HA became more negative as the ionic strength increased. The difference of the surface charge on HA between 0.01 and 1.0 M NaNO3 background (based on pH 5.00) presented an average of 30.0 cmolc kg–1. The effect of ionic strength was greater in the Namwon HA and smaller in the Donghong HA. The difference was 100 cmolc kg–1 in Namwon HA and 15 cmolc kg–1 in Donghong HA at pH 5.00. By changing the ionic strength in these experiments, it was found that the charge on the surface of humic acid was probably dependant on the origin of the humic materials (Table 2).


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 2. Proton affinity curves for humic acids (HA): plots of the observed and fitted charge vs. pH using Model A.

 
Table 3 shows that the eight resolved values were applied to the polynomial equations, and the intrinsic acidity constant was calculated at a given I. The electrostatic interaction factor ({omega}) also derived from the resolved values of P and Q. (The resolved values of carboxylic acid contents are described below.) Model A presented the dissociation of all six humic samples very well (average residual standard deviation [RSD] = 0.895 cmolc kg–1) and showed the systematic relationship between the intrinsic acidity constants and the ionic backgrounds.


View this table:
[in this window]
[in a new window]

  		Table 3. Values resolved for the parameters and the (–log 10) intrinsic dissociation constant for carboxyl groups (pKintC) and weak acidic groups (pKint{phi}), and the electrostatic interaction factor ({omega}), using Model A, at ionic strengths (I) of 0.01, 0.1, 0.5, and 1.0. The polynomial expressions for the parameters and the linear function of the electrostatic interaction factor are pKintC = a –bI1/2 + cI –dI3/2, pkint{phi} = e –fI1/2 + gI – hI3/2, and {omega} = P–Q log10I. The residual standard deviation (RSD) of the model fit is shown.

 
Figure 3 shows that the surface charge developments using our implementation of Model V fitted the same batch titration data as was applied to Model A. This model provided a good fit to the data across the range of ionic strengths used (average RSD = 1.262 cmolc kg–1). The resolved parameters are shown in Table 4. Model V has fewer fitted parameters, seven in all, including total carboxylic acid contents, compared with Model A, which requires 11. The assumption of the distribution of pKint values using a hypothetical median has reduced the number of optimizing parameters.


Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 3. Proton affinity curves for humic acids (HA): plots of the observed and fitted charge vs. pH using Model V.

 

View this table:
[in this window]
[in a new window]

  		Table 4. Values resolved for the (–log10) intrinsic dissociation constant for carboxyl groups (pKintC), the spreading factor for carboxyl groups ({Delta}pKintC), the (–log10) intrinsic dissociation constant for weak acidic groups (pKint{phi}), and the spreading factor for weak acidic groups ({Delta}pKint{phi}), using Model V; P and Q are the resolved empirical values in the equation for the electrostatic interaction factor [{omega} = P log10(I) exp(Q|Z|) where I is the ionic strength].The residual standard deviation (RSD) of the model fit is shown.

 
Carboxylic Acid Determination
Table 5 shows the total carboxylic acid group contents as determined by the different methods. The values of total carboxylic acid contents are the mean of three analyses, and mean values obtained from different analysis methods were statistically compared with each humic sample (Table 5). The most common method of analysis uses Ca(OAc)2, and the amount of acetic acid produced from the reaction with humic acid is determined by titration with a standard alkaline solution. With this method, total acidity ranged from 409 to 733 cmolc kg–1, and total carboxyl acid content ranged from 268 to 484 cmolc kg–1 (Table 5). The total carboxylic acid contents of the humic samples were from 54 to 66% of the total acidity of the humic materials. This result was comparable with other published data (Milne et al., 2001), in which carboxylic groups were estimated to be from 64 to 66% of the total acidity of soil's humic acid. The Namwon HA showed relatively higher values and Jisan HA presented relatively lower values in total carboxylic acid content. This result may indicate that the total carboxylic acid contents in different humic acids were related to the physicochemical properties of the parent soils. Namwon, a volcanic ash soil, contained a relatively higher percentage of organic C than Jisan, a paddy soil (Table 1). In addition, relatively higher RF and lower {Delta}logK values for the Namwon HA may be due to a higher degree of humificaton than the Jisan HA (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 5. The comparison of the carboxyl group analysis in humic acids (HA). The values are the mean of three analyses.

 
The precipitation–titration curves of HA with CTA+ at pH 7.0 and absorbance at 400 nm presented a distinctive precipitated end point (Fig. 4 ). The results of total carboxylic acid contents from different HAs ranged from 294 to 487 cmolc kg–1 with the CTA+ method (Table 5). Quantitative precipitation was conducted on HA with CTA+ reaction in solutions. De Nobili et al. (1990) suggested that the CTA+ method may overcome the apparent interference from the phenolic groups in the carboxylic acid group determination. The differences in the values obtained by the Ca(OAc)2 method and the CTA+ precipitation method varied from sample to sample and ranged between 0.3 and 11.6%, a much lower range than that calculated by De Nobili et al. (1990), which was 1.0 to 45%. Swift (1996) discussed that the CTA+ precipitation method needed to be tested for general use; however, the Bacsan HA data, Donghong HA data, and peat HA data indicated no significant differences between the Ca(OAc)2 and CTA+ methods when compared (Table 5).


Figure 4
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 4. Solubility curves of humic acids (HA) in the presence of different amount of cetyltrimethylammonium (CTA+) at pH 7.0.

 
The value of the carboxylic acid content was resolved with 10 other fitted parameters from the humic surface charge calculations in Model A (Table 5). The results from the Ca(OAc)2 method and the results from Model A regarding the carboxylic acid content in Jisan HA and Namwon HA showed no significant differences. The relative difference between the values from the Ca(OAc)2 method and the resolved values from Model A ranged from 0.21 to 7.43%. Model V predicted relatively lower values of the carboxylic acid groups when compared with the other methods except for the humic sample from peat. The resolved values obtained from Model V were significantly different from the values obtained from the other three methods when results were compared statistically (Table 5). For example, the relative differences between the Ca(OAc)2 method and Model V ranged from 1.90 to 35.71, a range that is comparable with the results of De Nobili et al. (1990).

The models used in this research, Model A and Model V, both adhere to the same basic principle of iteratively adjusting parameters until the lowest RSD values are found; however, Model A fitted values for a single carboxyl group acid dissociation constant (KintC) and a weak acid dissociation constant (Kint{phi}), and Model V assumed an equal relative abundance of group types within the carboxyl and weak acidic group classes. Marshall et al. (1995) investigated the changing effect of the number of group types on Model V. They fitted the data with a different numbers of group types in each class. They concluded that four different binding group types was slightly better than three different binding group types; however, choosing fewer than three different binding group types resulted in a sharp reduction in the goodness of fit to the data. Thus, Model V is more mechanistically realistic than Model A. The number of binding group types chosen for this model, however, does not indicate the actual heterogeneity of the functional groups of humic acid. In short, although Model A requires more parameters, it is simpler in its representation of humic acid functional group behavior. It seems that the total carboxylic acid contents obtained from Model A were more equivalent than the values from Model V when the resolved values from the models were compared with the values from the chemical analyses.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
This study demonstrated that HA extracted from a relatively elevated degree of humified soil showed higher total acidity and total carboxylic acid content. The effect of ionic strength was greater in HA extracted from soil that contained a high amount of organic matter, and humic surface charge was dependant on the origin of the humic materials. The contents of the carboxylic acid group derived from the two chemical analysis methods were equivalent. The total carboxylic acid contents from Model A, a model that is simpler but uses more fitted parameters, were more alike than those contents from two different chemical analysis methods. The Model V prediction of the carboxylic acid contents showed relatively higher discrepancy between the model and the two chemical analysis methods; however, the modeling approaches with potentiometric titration data shows some potential for further development in the analysis of carboxylic acid contents in soil humic samples.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Abbreviations: CTA+, cetyltrimethylammonium; DDL, diffuse double layer; HA, humic acid; RSD, residual standard deviation.

Received for publication July 14, 2005.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
Z. He, T. Ohno, F. Wu, D. C. Olk, C. W. Honeycutt, and M. Olanya
Capillary Electrophoresis and Fluorescence Excitation-Emission Matrix Spectroscopy for Characterization of Humic Substances
Soil Sci. Soc. Am. J., September 1, 2008; 72(5): 1248 - 1255.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jeong, C. Y.
Right arrow Articles by Lim, S. K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Jeong, C. Y.
Right arrow Articles by Lim, S. K.
Agricola
Right arrow Articles by Jeong, C. Y.
Right arrow Articles by Lim, S. K.
Related Collections
Right arrow Soil Surface Chemistry
Right arrow Soil Models
Right arrow Soil Chemistry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome