Abstract
The objective of this study was to evaluate the suitability of using woodchips as a carbon source for denitrification in intermittently operated household slow sand filters. This was accomplished through a series of batch and column experiments where fishpond bacteria were used to inoculate the reactors, and the resulting nitrate and nitrite concentrations observed over 24 h. Results from batch experiments in 100 mL serum bottles loaded with 0.5, 3 and 5 g of woodchip material, and 200 mg/L nitrate feed achieved 78%, 100% and 90% nitrate removal respectively. Nitrite concentrations in the batch reactors were 55, 0 and 98 mg/L after 24 h. A second set of batch experiments where 3 g of column sand was used as an inoculum showed a drop in nitrate from 200 mg/L to 182 mg/L. Three columns were constructed and fed intermittently with 3 L of nitrate contaminated water daily. The filter columns consisted of a woodchip only column, a sand only column and a mixed sand and woodchip column. After allowing them to stand for 24 h after being inoculated with denitrifying bacteria, the three columns were able to achieve complete denitrification within the 24 h filter run for a 50 mg/L nitrate feed. Nitrite production peaked at 2.4mg/L during this filter run. For a 200 mg/L nitrate feed, complete denitrification was achieved in the woodchip only column, effluent from the mixed and sand columns measured at 45 and 156 mg/L of nitrate, and 8 and 4 mg/L of nitrite respectively. A heterotrophic plate count gave counts of 9x103, 4.8x105, 9.9x104 and 3.3x105 cfu/mL for the feed, woodchip, sand, and mixed effluent filter streams at 200 mg/L nitrate feed.
