| || || Mataki, Melchior|
| || || Online photochemical oxidation and flow injection conductivity determination of dissolved organic carbon in estuarine and coastal waters|
Institution: University of the South Pacific.
Subject: Organic compounds--Analysis, Chemical oceanography--Fiji--Suva Harbour, Dissolved organic carbon
Call No.: pac QD 272 .F57 M27 1999
Copyright:This thesis may NOT be copied without the authors written permission.
Abstract: The determination of dissolved organic carbon (DOC) using flow injection analysis (FIA) was initially achieved through an FIA method described by (Koshy et al., 1992). Their method involved the photochemical oxidation of DOC using persulfate and the gas diffusion of the carbon dioxide which was measured spectrophotometrically. The above method was reported to suffer from chloride interference when saline samples were analyzed. This thesis presents the results of a research into the interference by chloride on the measurement of DOC using a modified FIA method based on the one described by (Koshy et al., 1992). The work began in the middle of 1997 after considering the importance of DOC as a water quality parameter and the lack of a cheap and reliable method for DOC determination in the South Pacific. Furthermore, it became apparent that there was a need for a baseline study of DOC in the Suva lagoon given the input of organic substances from land based sources via the rivers streams, creeks, storm water outlets and the Kinoya sewage outfall into the lagoon. By preparing corresponding concentrations of inorganic carbon (IC) in distilled deionized water (DDW) and in the 35%o NaCI solution (to simulate seawater) and analyzing them on the modified FIA method, it was established that chloride interfered with the measurement of DOC even with the modified FIA method. The interference by chloride (%CU) was found to be masked by IC to levels as low as 1% in the presence of IC concentrations greater than 8mg/L. The masking effect of IC on chloride interference was facilitated by the rapid decrease in the ratio of chloride to IC as the concentration of IC increased. Because of the generally high levels of IC (near 20mg/L) in estuarine and coastal waters (Manahan, 1994), the masking effect of IC on chloride interference was incorporated to the modified FIA method. With respect to the specific objectives of this project, attempts were made to eliminate or minimize chloride interference using a reduced oxidant (50% oxidant strength solution) boosted with a catalyst (T1O2), a reducing agent (NH2OH.HCI: Hydroxyiaminehydrochloride) and a micro-porous tubing (Accurel PP Type S6/2 0.2um pore size; Enka). When the catalyst was used with the reduced oxidant (50% oxidant strength solution) it raised the DOC recovery back to 100%. Unfortunately, the interference by chloride was also increased in the range of 12% to 5 1 % greater than the levels of chloride interference observed with the 100% oxidant strength solution alone. NH2OH.HCI reduced the chloride signal by 81±3% on its own. In the presence of carbon (IC or DOC), the interference by chloride was reduced in the range of 80% to 98%. On the other hand, chloride interference was reduced by the micro-porous tubing by an average of 95%. It was apparent that the masking effect of IC on chloride interference was comparable to the minimization capacities of the reducing agent and the micro- porous tubing. Furthermore, the choice to utilize the masking effect of IC in the determination of DOC in estuarine and coastal waters did not alter the configuration of the modified FIA method. The FIA method was found suitable for the determination of DOC in estuarine and coastal waters because it has a reasonably high carbon recovery of 98+9%, sample throughput of 60 samples per hour and a method detection limit (MDL) of 0.8mg/L (DOC). The carbon recovery of the modified FIA method described in this work was comparable to the carbon recoveries of a similar FIA method for DOC determination and a Beckman high temperature combustion instrument reported by (Koshy etal., 1992). The MDL (0.8mg/L) of this modified FIA method made it suitable for the measurement of DOC in most estuarine and coastal waters because of the organic input from the land and the level of productivity in these types of waters. Moreover the interference by chloride on the measurement of DOC could be controlled in the FIA method described in this work. The baseline study of DOC indicated that the organic discharge from the greater Suva land mass through the rivers, creeks, storm water outlets and the Kinoya sewage outfall significantly contributed to the DOC present in the Suva lagoon. The assertion was substantiated by the wide variation of DOC in each site on each day of sampling. The detectable DOC levels in the Suva lagoon in the period of study varied from 0.8mg/L to 35mg/L. The trend observed in the short period of study was that the levels of DOC in the sites monitored could be doubled or tripled when the Suva area experienced an average rainfall as low as 0.1mm in the three days prior to sampling. The modified FIA method described in this work could be used for the long term monitoring of DOC in the Suva lagoon and predictions could be made on the lagoon's DOC budget.