Date of Award


Document Type


Degree Name

Master of Science (MS)

Legacy Department

Environmental Engineering and Science


Karanfil, Tanju

Committee Member

Carraway , Elizabeth

Committee Member

Schlautman , Mark


The main objective of this study was to gain insight to the principles of isolation of natural organic matter (NOM) using reverse osmosis (RO) and subsequent fractionation using resin adsorption chromatography (RAC). Specifically, this study evaluated the RO and RAC methods for NOM characterization from three surface waters with varying physiochemical characteristics. Efficiency of RO was assessed by closing mass balances for dissolve organic carbon (DOC). Mass balances were also closed for dissolved nitrogen (DN), bromide (Br-), nitrate (NO3-), total calcium (Ca), total potassium (K), total manganese (Mn), total iron (Fe), total magnesium (Mg), total aluminum (Al), total copper (Cu), total phosphorous (P), total zinc (Zn), total sulfur (S), and total boron (B). The efficiency of RAC was assessed by closing mass balances for DOC, Br-, and NO3-. Additionally, RO was also evaluated by investigating the effect of pH on NOM isolation, and RAC was evaluated by investigating the effect of column operational parameters (column capacity factor, k', and solute initial concentration, C0) on NOM fractionation. The specific ultraviolet absorbance (SUVA254) of the isolated NOM after RO and the fractionated NOM after RAC was obtained during the study. Based on the high mass recovery of NOM, the RO and subsequent RAC method was an efficient means to isolate and fractionate NOM samples. Efficiency of RO was dependent on both pH and source water chemistry. In general, RO more effectively isolated NOM in high SUVA254 water (~4.9) than low SUVA254 water (~1.9), and showed higher NOM recovery at ambient pH (~7) than at low pH (~4). The pH did not have any significant impact on the mass recovery of DN and various elements. The fractionation of the isolated NOM indicated that the relative amount of the hydrophobic (HPO) fraction decreased with increasing k', thus affecting the overall hydrophobic distribution of NOM. Alternatively, the hydrophobic distribution of NOM fractions was not influenced by varying the C0 between 50 and 150 milligrams per liter (mg/L) at k' of 15. Lastly, the relative amount of the HPO fraction from the small-scale fractionations (at k'15 and C0 of 150 mg/L) agreed well with the HPO fraction from the large-scale fractionations.