Four grams of each soil were weighed into 50 mL centrifuge tubes
with four replicates and 40 mL of DI water was added to each tube
prior to shaking overnight. After shaking, each soil suspension
was allowed to settle undisturbed for approximately two hours
before the dispersed colloidal fraction (3mL) was sampled by slowly
pipetting at a depth of 2.5 cm below the solution surface in manner
similar to Miller and Miller (1987)(Miller and Miller, 1987).
The dispersed clay present in each tube was quantified by placing
the pipetted suspension in preweighed, oven-dried aluminum tins
and heating at 110° C to dry the sample prior to reweighing
the pan. Eighty air-dried soil samples reflecting a range of
soil properties were initially screened using the above procedure
and 33 soils were selected for further characterization and identified
as ìproject soilsî. Subsequent WDC measurements
were conducted on field-moist samples of the thirty-three project
soils selected from the initial screening (Table 1).
Analysis of Suspension Properties
The pH and electrical conductivity of the suspensions remaining after sampling for WDC were measured for each suspension prior to filtering through a 0.2µm pore size polycarbonate membrane filter. Dissolved organic carbon (DOC), dissolved cation and anion analyses were performed on the filtrates. The DOC content for each sample was determined using a Shimadzu Organic Carbon Analyzer. The concentrations of the major solution cations (Na+, K+, Mg2+, Ca2+) and anions (Cl-, NO3-, SO42- , PO43-, F-) were determined by ion chromatography using a CD20 conductivity detector (Dionex, Inc.). The sodium adsorption ratio, a property often correlated with the relative dispersiveness of a given solution (Frenkel et al., 1978; Khilar, 1984), was calculated as:
for the average extract cation data with all components expressed
in terms of meq L-1(Bohn et al., 1985; Sposito, 1989). The
reported SAR value reflects the mean sodium adsorption ratio for
the four individual replicates for each soil.
Dissolved and Colloid-Associated Organic Carbon
Samples for the analysis of organic carbon associated with dispersed
colloidal material were collected using a modified version of
the procedure described above. Twenty gram samples of each soil
were weighed into three 250 mL centrifuge tubes. Two-hundred
milliliters of DI water was added to each tube prior to shaking
the sample overnight. After shaking, the samples were allowed
to settle for 2 hours before the top 2.5 cm of the suspension
present in each sample tube was slowly removed using a 50 mL syringe.
The three replicate suspensions for each soil were combined,
quick frozen with liquid N2, and subsequently freeze
dried. The total organic carbon (TOC) for bulk samples of each
soil (<2mm fraction) and the colloid-associated carbon were
determined by dry combustion using the LECO furnace method (Nelson and Sommers, 1982).
Organic Matter Characterization
Considerable research has shown that natural organic matter associated with clay mineral surfaces enhances dispersion of the minerals (Kaplan et al., 1993; Kretzschamar et al., 1993). Fluorescence spectroscopy was used to try to characterize qualitative differences in organic matter associated with water-dispersible clay versus whole soil in a number of the study samples. Because fluorescence spectroscopy is much less time-consuming than other techniques such as NMR spectroscopy, it could serve as a screening tool for selecting samples for more detailed analyses. However, because fluorescence results from electronic transitions of valence electrons, the information on molecular structure and bonding is less specific than in NMR and infrared spectroscopies, especially because organic matter is a heterogeneous mixture of aromatic and aliphatic structures along with non-humified compounds (polysaccharides, proteins, cellulose, etc.).
Last Modified: September 30, 1998
Document Prepared by:
North Carolina Agricultural Research Service
North Carolina State University