Arsenic removal from water using naturally occurring iron, and the associated benefits on health in affected regions

This thesis presents the results of a detailed study on ‘‘Arsenic Removal from Water using naturally occurring Iron, and the Associated Benefits on Health in Affected Regions’’. The details of the experimental work carried out in the laboratory and field as part of the Ph.D. work are presented and discussed in the 8 accompanied papers.

Arsenic (As) intake with drinking water is a very severe problem in the South East Asia with the Bengal delta being the worst affected area. The occurrence of As in ground water in the Bengal delta is of natural origin and microbial mediated reductive dissolution of Fe hydroxides with adsorbed As and phosphate (P) is the most accepted theory for the release of As to the groundwater. In 1993 WHO lowered the guideline value for As in drinking water from 50 µg/l to 10 µg/l. However only during the past 5 years many industrial countries adopted this lowered guideline value as maximum contaminant level (MCL). On the other hand many developing countries including India and Bangladesh still have 50 µg/l as MCL. As is a documented human carcinogen and the estimated excess life time risk of getting skin cancer even at 10 µg/l level is 6*10^-4. The risk estimates for population of Bengal delta would even be higher because of the
lower average body weight of the person, higher water intakes and lower nutrient intake. However, the reported arsenicosis cases in Bengal delta are lower than expected compared to other parts of the world, despite the relatively higher As concentrations in hair. This Ph.D. study points to shorter duration of exposure to and co-occurrence of iron (Fe) as the two main factors responsible for the observed lower arsenicosis prevalence.

Presence of Fe renders colour to water upon exposure to air and hence people would either leave the water to stand for a couple of hours, to remove Fe, leading to indirect As removals due to co-precipitation with Fe, or, they would not use the water. Both these practices would lead to reduced As exposure. However, the development of colour in the water depends on the concentration and oxidation rate of Fe²⁺. The oxidation rate of Fe²⁺ depends on many factors. At lower dissolved oxygen and/or high P concentrations the colour development is slow. Further the colour of the precipitates in presence of phosphate (P) is pale yellow compared to red-brown in the absence of P. Hence the presence of Fe may not be detected if the distance to tube well is short and people do not store the water. The BGS study and the study by Technical university of Denmark (DTU) showed that the installation of the tube wells increased exponentially in the beginning of 1990’s. This led to usage of drinking water without prior storage due to shorter distance to the tube well. Based on the higher As concentrations in hair and the recent exposure to direct consumption of tube well water, the study indicated that the disaster is yet to be seen and there is a need for immediate measures/action.

Based on the available literature this study identified removal of arsenic using naturally occurring Fe as the most appropriate and accessible mitigation option for the rural setting prevailing in India and Bangladesh. The field results showed that an Fe/As ratio above 80 (M/M) is necessary to achieve As concentrations below 50 µg/l using iron removal units (IRU) and based on the British Geological Survey (BGS) survey this method can be applied at 25% of the tube wells. The field results also showed that simple sedimentation of water with elevated Fe and As concentrations would lead to removal of As and an Fe/As ratio of above 120 (M/M) is required to attain residual As concentrations below 50 µg/l.

Insight into the mechanisms responsible for removal of As with Fe is important to optimise As removals. As occurs mainly in the form of inorganic arsenite (As(III)) in the ground waters of the Bengal delta. As(III) is neutrally charged and is difficult to remove compared to the negatively charged arsenate (As(V)) in the near neutral pH range. The possibility to differentiate between arsenite (As(III)), and arsenate (As(V)), is important to understand the As removal mechanisms involving Fe. Investigations on preservation of samples showed that neither acidification nor acidification and storage at 4ºC were appropriate for preservation of samples. Storage in the dark also did not prove to be sufficient for preservation of samples. Hence, a method based on continuous hydride generation and measuring As using AAS was developed for immediate and
continuous measurement of both As(III) and total As. The results were verified using on-line separation of As species with an anion exchange resin.

Extensive investigations were carried out to understand the removal mechanisms
responsible for As removal in presence of Fe. The results showed that oxidation of As is an important factor furthering the removal of As with Fe and the results indicate that the oxidation is most probably due to heterogeneous oxidation on the surface of Fehydroxides. The properties of the Fe-hydroxides vary greatly depending on the initial oxidation state of Fe and the presence of ions. The results in the present study showed higher As oxidation and removal with Fe initially present as Fe³⁺ compared to Fe²⁺. However, the presence of P had a stronger negative effect on oxidation and removal in presence of Fe³⁺ compared to Fe²⁺. At very high P concentrations both oxidation and removals were completely inhibited with initial Fe³⁺, whereas both oxidation and removal of As(III) occurred though decreased with initial Fe²⁺. Similarly presence of silicate (Si) also had a negative effect on oxidation of As(III). Lower O2 concentrations also had a negative effect on As(III) oxidation in presence of Fe²⁺ and higher pH had a positive effect on As(III) oxidation. The results indicated that maximum oxidation of As(III) occurred within the first 30 minutes of the experiments and similarly maximum As removals also occurred during this time. However, in presence of P and Si prolonged sedimentation of up to 24 hours was required to achieve maximum As removals. Slightly lower As removals were observed in the field compared to the laboratory and could be due to lower O₂ concentration in the field. These results indicate that high Fe/As ratios are necessary if As removal is based on Fe alone. Since oxidation of As(III) is an important factor controlling the As removal capacity of Fe, addition of simple oxidising agents would reduce the required Fe/As ratio. Fe-citrate complex in presence of UV-light as well as KMnO₄ were investigated for their oxidation capability. Citrate could be added as lime and sunlight can be used as UVlight. The results with Fe-citrate complex in presence of UV-light showed that both oxygen and UV-light is necessary for As(III) oxidation. The results further showed that if the citrate/Fe ratio is above 1 then As removals could not be achieved as Fe will not precipitate. Based on these results the method is not suitable for practical applications. KMnO₄ was found to be an appropriate oxidising agent and the results showed that sufficient amount of KMnO₄ necessary for oxidation of both As(III) and Fe²⁺ should be added immediately after pumping the groundwater to achieve satisfactory results. Attaining sufficient surplus can be verified in the field by the development of pale pink colour. The calculations showed that application of sufficient KMnO₄ in presence of high Fe concentrations would alone be enough at 60% of the existing tube wells. The monthly cost of this treatment would be 4 INR (~8 Eurocent) assuming a daily requirement for treated water for a family to be 50 litres. For the rest of the tube wells an extra addition of coagulant is required and could be either in the form of Fe or Al. Even though lower amount of Fe would be required compared to Al, the cost of Fe is higher than Al and hence to reduce the costs an extra amount of Al could be added. To achieve better As removal efficiencies, Al should either be added in very fine powdered or soluble form. The laboratory investigations showed that an aqueous mixture of Al₂(SO₄)₃ and KMnO₄ is stable at pH 11 adjusted with NaOH. The application of chemicals in solution would result in better mixing and the results showed that this mixture would have great potential for treatment of As. However, verification is required in the field.

In all cases the daily generation of As contaminated sludge is very low, however care should be taken in avoiding disposal of the sludge near children’s playing area or edible crops. A suitable way of disposal of this sludge would be to bury it.

Instead of waiting for an expensive solution based on construction of deep tube wells, taking years and reaching few, removal of As based on naturally occurring Fe along with application of KMnO₄ supplemented with Fe/Al could be recommended as a mitigating option for the affected rural areas of the Bengal delta.