~ Pergamon PII: S0043-1354(96)00324-7 Wat. Res. Vol. 31, No. 5, pp. 965-972, 1997 © 1997 Elsevier ScienceLtd. All rights reserved Printed in Great Britain 0043-1354/97$17.00+ 0.00 INFLUENCE OF CHARACTERISED NATURAL ORGANIC MATERIAL ON ACTIVATED CARBON ADSORPTION: I. CHARACTERISATION OF CONCENTRATED RESERVOIR WATER GAYLE NEWCOMBE l*, MARY DRIKAS I@, SHOELEH ASSEMF and RONALD BECKETT 2 tAustralia Water Quality Centre, Private Mail Bag, Salisbury, SA 5108, Australia and 2CRC for Freshwater Ecology, Water Studies Centre and Department of Chemistry, Monash University, P.O. Box 197, Caulfield East, Victoria 3145, Australia (Recewed May 1996; accepted & revised form November 1996) Abstract--Natural organic material (NOM) from Myponga Reservoir in South Australia was concentrated and fractionated using ultrafiltration into nominal molecular weight fractions < 500, 500-- 3000, 3000-10,000, 10,000-30,000 and > 30,000. The fractions were characterised using flow field-flow fractionation, ~3CNMR, colour determination and potentiometric titration. The ultrafiltration fractions displayed a gradual transformation from highly coloured, highly branched, high carbohydrate structures to compounds with a prevalence of long chain aliphatic carbon with much lower carbohydrate content and colour. There were no clear trends in the carboxyl content of the fractions, as determined by both NMR and titration. Analysis of the titration data showed evidence of three distinct types of carboxyl groups and the same types of groups were present in each fraction. © 1997 Elsevier Science Ltd Key words--dissolved organic material, natural organic material, humic substances, ~3Cnuclear magnetic resonance, characterisation INTRODUCTION Dissolved natural organic material (NOM) is found in varying concentr~itions in all natural water sources. It is a complex mixture of compounds formed as a result of the breakdown of animal and plant materials in the environment. The composition of the mixture is strongly dependent on the environmental source (Aiken and Cotsaris, 1995); however, some generalis- ations can be made about the structures present in NOM. A range of compounds, from small hydrophilic acids, proteins and amino acids to larger humic and fulvic acids, are constituents in most NOM (Choudry, 1983). The predominant humic and fulvic acids have a relatively high molecular weight, typically in Australian waters from below 500 to > I0,000, and can have both aliphatic and aromatic character (Choudry, 1983; Beckett et al., 1987). Most of the compounds present in NOM carry a charge, generally attributed to carboxylic acid and phenolic groups (Perdue and Lytle, 1983; Perdue et al., 1984), and this causes the larger compounds to behave as polyelectrolytes in solution (Ephraim et al., 1986). The mechanism of transport of NOM through soils into rivers and streams is of importance for *Author to whom all correspondence should be addressed [Fax: (+61) 8259 0228]. catchment management strategies, and the effect of NOM on drinking water treatment is important in terms of public health and water quality. It is therefore of value to gain information on the character of NOM with the aim of determining how this influences soil transport and water treatment processes. The size distribution of the NOM is a significant characteristic which is often measured using size exclusion chromatography or ultrafiltration. Neither technique can fractionate on size alone, and other factors such as structure, charge and polarity influence the results (Wershaw and Aiken, 1985). For example, Kuchler and Miekeley (1994) found that an ultrafiltration membrane of nominal molecular weight 1000 retained a fulvic acid in preference to a humic acid, although the humic acid had a higher molecular weight. Carbon 13 nuclear magnetic resonance (~3C NMR) is a very useful tool in determining the average chemical structure of NOM. Spectra can be obtained for a concentrated liquid, or solid, sample and the chemical shifts of the spectral peaks are assigned to ~3C in different chemical environments. The areas under the peaks are then related to the different percentages of carbon in each of the chemical environments. In this way the per cent of carbon in aliphatic, O-alkyl, aromatic and carbonyl groups can be determined. Wershaw et al. 965