The combination of rising population and increasing food demand has placed tremendous pressure on the soil resource in Bangladesh, especially floodplain soils where rice may be grown year round. Although these floodplain soils contain minerals high in potassium (K) such as micas, K deficiency in rice paddy soils is increasing throughout Bangladesh. Harvesting of straw from fields exacerbates the removal of nutrients including K, and intensive cropping with unbalanced fertilizer has resulted in depletion of "readily available" K in large areas. Five representative floodplain soils in rice production were sampled by horizon to determine physical, chemical and mineralogical properties, assess soil K, investigate plant available K, and understand impacts of redox on K fixation and release. Total and nonexchangeable K were determined using concentrated hydrogen fluoride (HF) digestion and 1.0 M HNO3 digestion, respectively, along with NH4OAcextractable K. Cation exchange capacity under both oxidized and reduced conditions was determined. Assessment of adsorption and release of K as a function of redox was done in conjunction with seven different concentrations of potassium chloride (KCl) solution and three different soils. Significant amounts of K were present in all fractions of silts and clays. Potassium concentration was greatest in the coarse clay fraction. High HNO3- extractable K suggested that much of the K in soils was from mica interlayers. Acidic soils contained less K in all fractions than the less weathered, calcareous soils. The clay fractions exhibited mostly mica, vermiculite, smectite and kaolinite. The cation exchange capacity (CEC) of reduced soils was less under oxidized condition, due to collapse of the interlayer in response to increased layer charge upon structural Fe reduction. The adsorption of K was greater for calcareous soils under both oxidized and reduced conditions than for the acidic soils. Less native K was extracted under oxidized conditions than under reduced conditions. Less adsorption under reduced conditions may be due to an increase in solution Fe2+ which can compete with K+ for exchange sites on clays.

Master of Science / Department of Agronomy / Michel D. Ransom / Many soils in southeastern Kansas are characterized by high clay contents and high shrink-swell potentials. Their vertic properties and claypan characteristics cause soil management to be difficult and pose problems for agricultural, environmental, and engineering uses. Thus, collecting more information and improving our understanding of these soils is an important step towards bettering our soil management techniques. The objectives of this study were to examine the morphology, processes of soil genesis, clay mineralogy, micromorphology, and potassium fixation potential of the soils of interest and how these characteristics varied between and within individual pedons. Ten pedons expected to represent varying degrees of vertic expression were selected. Methods included the use of field descriptions, routine soil laboratory characterization, micromorphological investigations, the determination of clay mineralogy by X-ray diffraction, and the measurement of potassium fixation potential. Field morphology reflected the geologic parent materials available in the region. The fine sediments that compose these clayey soils are primarily provided by the Pennsylvanian and Permian shales and limestones underlying this region and the Flint Hills to the west. Dominant pedogenic processes currently at work are clay illuviation and shrink-swell processes. Silty, non-expansive surface soils at all but sites 6 and 7 are thought to buffer the rapid wetting and drying cycles needed for maximum vertic expression. Four of the soils were dominated by smectitic minerals in the clay fraction while the rest exhibited a more mixed mineralogy. Disruption of illuvial clay features by shrink-swell movement was evident in thin section. Striated b-fabrics dominated the micromorphology except in non-expansive surface soils. K fixation of the soil surface was found to be negative in all soils, thus K fixation potential is considered very low. In subsurface horizons, K fixation generally increased with increasing vermiculite content. In addition to limited quantities of K-fixing clay minerals, naturally high K levels limited the amount of K fixation in this study. The information presented can be used to improve our understanding and management of high clay, vertic and claypan soils in southeastern Kansas.

Master of Science / Department of Agronomy / Michel D. Ransom / Factors governing potassium fixation and release are poorly understood. This study was conducted to investigate the effects of clay mineralogy and structural iron oxidation state on potassium fixation. Five reference clays and two soil clays were used to capture a range in mineralogical compositions and potassium behaviors. Reference clays used were illite (IMt-1), kaolinite (KGa-1b), montmorillonite (STx-1b), nontronite (NAu-2), and vermiculite (VTx-1). Soil clays used were from the upper 15 cm of a Belvue loam (BEL) and a Cherokee silt (CHE). Potassium fixation capacities were measured on unaltered as well as sodium dithionite reduced forms of each clay. Ferrous and total iron contents were determined photometrically using 1, 10-phenanthroline. Potassium fixation was measured by potassium saturating the clays and washing off exchangeable and solution potassium with solutions of magnesium chloride; samples were then acid digested and the amount fixed was calculated as the amount of potassium in the acid digestion minus the amount originally in the sample. BEL released potassium rather than fixed it while CHE tended to release potassium in the unaltered form and fix potassium in the reduced form. Structural iron reduction significantly impacted the amounts of potassium fixed by VTx-1 and NAu-2, which had the highest total iron contents of all the clays evaluated. NAu-2 and VTx-1 both on average fixed less than 1 mg K g clay[superscript]-1 in the unaltered form and an average of 6 and 11 mg K g clay[superscript]-1, respectively, in the reduced form. Regardless of being in the unaltered or reduced form, KGa-1b fixed essentially no potassium and IMt-1 and STx-1b fixed intermediate amounts of potassium—2 to 4 mg K g clay[superscript]-1 on average. The effects of clay mineralogy and structural iron oxidation state on potassium fixation can largely be explained through an understanding of layer type, layer charge, and charge distribution. In order for potassium fixation to occur, interlayer sites need to be accessible and available. Generally, the greater the negative layer charge the greater the amounts of fixation, with tetrahedral layer charge favoring fixation more than octahedral layer charge, and layer charge being a function of structural iron oxidation state.