False positives for cyanide analysis in wastewaters have been reported. We examined the effects of storage time at high pH and of pH adjustments on the cyanide levels. Cyanide levels changed within the holding time allowed by Standard Methods. We also studied the difference in cyanide levels using two disinfection conditions -- breakpoint chlorination and chloramination. Glycine was used as the precursor to study the cyanide formation pathways. Under breakpoint chlorination conditions, cyanide formation is complete relatively quickly and detectable cyanogen chloride is produced. On the other hand, chloramination yields cyanide through a relatively slow, base-catalyzed reaction. Chloramination followed by dechlorination with sodium arsenite and addition of NaOH results in cyanide levels that increase significantly upon reanalysis in the first 24 hours and then remain relatively constant after that time. Cyanogen chloride (CNCl) was <5 ppb in samples disinfected with chloramination. Mechanisms are proposed that explain the very different cyanide results that are obtained when disinfection is carried out under breakpoint chlorination conditions versus chloramination conditions.
INTRODUCTION AND BACKGROUND
Recent studies in wastewater matrices have suggested the presence of uncharacterized positive interferences affecting the analysis of total cyanide using colorimetric procedures such as EPA 335.4 (U.S. EPA, 1993) and Standard Methods 4500-CN (APHA/AWWA/WEF, 1998). As a result of these studies, attention has begun to focus on the reliability of currently accepted cyanide analytical methods. False positives resulting from cyanide formation during sample storage at high pH have recently been reported by Weinberg, et al. (Weinberg et al., 2005). The city of San Jose concluded that cyanide was being generated after collection, during the preservation of wastewater effluent samples to which NaOH was added to adjust the pH to 12 (City of San Jose, 2004). Studies conducted in the Los Angeles County Sanitation Districts’ laboratories indicated that some of the approved preservation protocols could give rise to cyanide formation in chlorinated wastewater effluent matrices (Khoury et al., 2005).
The Sanitation Districts’ laboratories have carried out extensive studies on cyanide formation by testing samples from several wastewater treatment plants. The results have indicated that cyanide levels are generally below reporting limits when samples are analyzed immediately without pH adjustment, irrespective of the dechlorinating agent used. However, a significant increase (>10 μg/L) of cyanide was found in samples taken after chlorination of the secondary effluent, when dechlorinated with sodium arsenite and then preserved to pH >12 (Khoury et al., 2005).
Standard Methods suggests a holding time of 14 days for samples preserved to pH ≥ 12, using sodium hydroxide to retard the loss of volatile hydrogen cyanide by converting it to its nonvolatile ionic form. The study presented in this paper is focused on the effects of the storage time between sampling and analysis for cyanide within the holding time of 14 days. When samples preserved to pH 12 were analyzed over a period of 48 hours within a Standard Methods recommended holding time, it was observed that the cyanide levels increased with time. This indicates a possibility that an in-situ cyanide generation reaction is in progress at pH12.
Chlorination is a well-developed and widely used process for disinfection. There are numerous pros and cons related to using either chloramination or breakpoint chlorination in disinfecting wastewater streams. Chloramination would decrease several regulated disinfection by-products (DBPs), i.e., total trihalomethanes (THMs), however it would increase the production of the potent carcinogen N-nitrosamines (Mitch and Sedlak, 2002). Breakpoint chlorination, on the other hand, would decrease the production of N-nitrosamines (Mitch and Sedlak, 2002), but may introduce other by-products, i.e., cyanogen chloride (CNCl) (Shang et al., 2000) and THMs. In this study, cyanide generation is examined under two disinfection conditions -- breakpoint chlorination and chloramination.
Possible mechanisms for cyanide formation in water and wastewater treatment processes have been identified in laboratory scale experiments. The mechanism of cyanide and CNCl formation from glycine in water under free chlorine conditions has been reported by Na and Olson (2006). Monochloramine has been shown to react with formaldehyde and eventually yield HCN (Pedersen et al., 1999); organocyanide compounds (cyanocobalamin and coenzyme vitamin B12) release free or metal-complexed cyanide upon chlorination (Yi et al., 2002); solutions of L-serine that were chlorinated and subsequently dechlorinated were shown to produce cyanide (Zheng et al., 2004a); reaction of nitrite with aromatic compounds can produce cyanide (Zheng et al., 2004b); microorganisms have been shown to be capable of producing cyanide (Brandl, 2005); less than stoichiometric chlorination of thiocyanate can liberate free cyanide (Zheng et al., 2004c); and, it was found that phenol reacts with nitrous acid to produce cyanide ions (Adachi et al., 2003). The potential for chloramination to yield cyanide from organic compounds was demonstrated in earlier experiments using synthetic solutions spiked with select precursor organics such as ascorbic acid, humic acid, D-ribose, and 2-furaldehyde (Carr et al., 1997).
Amino acids have been reported as potential precursors of the disinfection byproduct CNCl in chlorinated drinking water (Sawamura Et al., 1982; Hirose et al., 1989). Glycine, among 17 amino acids, has been proved to yield the most CNCl after chlorination (Lee et al., 2006). In this study, glycine was spiked in water samples collected from secondary effluents, before ammonia addition, to study the mechanisms of the production of cyanide and CNCl. Under laboratory
controlled conditions, chlorine and chloramines were dosed according to the wastewater treatment conditions. The major objectives of this study were to confirm the cyanide formation mechanisms and to provide feasible suggestions for wastewater disinfection operations.