Het ontstaan van trihalomethanen bij de behandeling van drinkwater met chloor

It has been established that chlorination of naturally coloured waters produces chloroform and other trihalomethanes in concentrations that are considerably higher than most of the organic microcontaminants commonly found in polluted surface waters.The objective of this study was to investigate the origin of the trihalomethanes in drinking water and the mechanism responsible for their formation. Chapter 1 gives a historical survey of the use of chlorine in drinking water treatment. The main application of chlorine originally was disinfection. Soon after, chlorination became common practice for the oxidative removal of earthy tastes and odors. Additionally the bleaching of naturally coloured water was found to be advantageous.As a result of studies of the various reactions between chlorine and inorganic ammonium the breakpoint phenomenon was discovered in 1940. Since then breakpoint chlorination is used for quantitative removal of ammonium from raw water. Until that time alternate methods of removal included biological means such as slow sand filtration. Breakpoint chlorination led to the development of water purification methods that could be accomplished by entirely physico-chemical treatment.However, as the levels of pollution of river Rhine water increased, the experience in Rotterdam was that breakpoint chlorination became less and less effective. Additionally the taste was adversely affected. In 1965 gaschromatography was applied in an attempt to determine the reaction products responsible for the taste problem.Chapter 2 describes the development in our laboratory of a modified headspace vapour analysis method for water. This allowed for the capture of minute amounts of volatile organics before and after chlorination of the river water. The technique basically is a static method in which a gas volume (headspace) is brought into equilibrium with the water sample. The headspace gas is collected on a cold trapping column which contains the same adsorbent as used for the stationary phase in the subsequent gaschromatographic separation. Using this technique for drinking water four large peaks were found which were not present in the untreated water. The four compounds were identified with chloroform, dichloro-bromomethane, chloro-dibromomethane and bromoform.Chapter 3 gives the analytical data obtained from Rotterdam drinking water, first prepared from river Rhine and later from river Meuse. Surprisingly the same four haloforms were found after chlorination of both waters.It was analytically established that the chlorine gas didnot contain trihalomethanes. Neither did it contain bromine in detectable amounts, i.e. less than 100 mg/kg.In an investigation of natural waters chloroform formation was found to occur after chlorination of samples from several unpolluted ponds and. lakes. Those that contained bromide-ions in concentrations higher thans 0.1 mg/l also gave the brominated trihalomethanes. It was suspected that both the organic precursor for chloroform and the inorganic precursor for the brominated compounds were present in all types of natural waters. Chlorination resulted in the oxidation of Br -to HBrO, which in turn is known to react as a halogenating agent.Chapter 4: Since the major soluble organic compounds in natural waters are humic substances - mainly fulvic acids - it was logical to assume that they are the organic precursor. Humic acids contain both carboxylic and phenolic functionalities. Polyphenols are known to make up for the bulk of the degradation products of humic acids. Furthermore several polyhydroxybenzenes are known to give the iodoform reaction. The combination of the latter two facts led to the theory that polyphenolic structures in the humic acid matrix are the reactive sites from which chloroform is formed. To prove this, the principle experiment consisted of chlorination of peat extracts and aqueous solutions of commercial humic acids. Both solutions gave chloroform when treated with chlorine. Addition of inorganic bromide before chlorination resulted in the formation of all four haloforms.Chapter 5 involves an investigation of the specific type of polyhydroxybenzene responsible for chloroform-formation. Compounds in which two hydroxy-groups are in the meta-position gave the highest yields of chloroform.We also found that in the methylethers of resorcinol and phloroglucinol the formation of chloroform is suppressed. This led to the theory that the phenoxide-ion is essential for the formation of the halogenated mono-carbon degradation product.It is a reasonable assumption that in the competetive oxidation and substitution reactions the aromatic nucleus is destroyed to give an alpha-chlorinated ketone. A reaction mechanism is proposed in which the carbon atom in between two meta-positioned OH-groups is chlorinated twice. This is followed by scission into a linear alpha-ketone. This intermediate is further chlorinated to RCCl 3 according to the pathways known to occur in the haloform reaction.In chapter 6 cursory pilot plant studies focused on removal of chloroform by adsorption on activated carbon are described. It was found that the adsorptive capacity of carbon for chloroform was too low to be of interest. Further investigation aimed at the removal of the fulvic acids which are the precursor for the trihalomethanes. Pre-ozonation was tried but had little effect.Some experimental results with macroporous ion exchance resins appear effective in removing a major portion of the precursor for the haloformformation.