Limited effectiveness of household sand filters for removal of arsenic from well water in North Vietnam

Well water derived from groundwater is important as a domestic water source, including drinking water in Asian countries. However, well water may have a problem of naturally occurring arsenic (As) due to As-enriched aquifers. In fact, arsenic pollution of drinking water from wells has been found in many countries, including Bangladesh, India, Cambodia, Myanmar, and Vietnam (Argos et al. 2012; Kumasaka et al. 2013). Previous studies showed that chronic exposure to As via drinking water can cause various diseases, including cancers, cutaneous pigmented disorders, and diabetes (Nizam et al. 2013; Yajima et al. 2015).

The World Health Organization (WHO) has provided guideline values for the quality of drinking water for barium (Ba), iron (Fe), and manganese (Mn) in addition to As in the present and/or previous editions (WHO 1984, 2008, 2011). Barium naturally coexists with other toxic elements in arsenic-polluted well water in Bangladesh and Vietnam (Frisbie et al. 2009; Kato et al. 2013). Our previous fieldwork studies showed a correlation between As and Ba in well drinking water and in human samples (urine, nails, and hair) in Bangladesh (Kato et al. 2013). Our previous experimental studies in vitro showed that Ba promotes a malignant characteristic of human non-tumorigenic keratinocytes (Thang et al. 2015) and that Ba suppresses As-mediated anti-cancer effects (Yajima et al. 2012). Moreover, Ba was shown to promote hearing loss in mice (Ohgami et al. 2012) and humans (Ohgami et al. 2016) in our experimental study and epidemiological study, respectively. Previous studies showed that Fe promotes all steps of carcinogenesis (Weinberg 1996). In fact, previous studies have shown effects of iron on tumor initiation and tumor growth (Kumasaka et al. 2013; Torti & Torti 2013). A previous epidemiological study also showed that an elevated level of serum iron increases the risk of several carcinomas in humans (Wen et al. 2014). Exposure to Mn in drinking water has been reported to be associated with neurobehavioral disorders in mice (Krishna et al. 2014; Kumasaka et al. 2014). Exposure to Mn from well drinking water is also associated with an increase in infant mortality, cognitive deficit, memory deficit, and lower intelligence scores in humans (O'Neal & Zheng 2015). Thus, high uptake is a problem, which necessitates WHO to provide guidelines for the quality of drinking water, although some of the four elements are needed at trace levels. In addition, high levels of As, Ba, Fe, and Mn are harmful to aquatic life, biota, and the environment as well as human health (Wang 1988; Korte & Fernando 1991; Ventura-Lima et al. 2011; Kalantzi et al. 2013).

Approximately 13 million people in Vietnam (16.5% of the population), who mostly reside in the Red River Delta, are drinking water from wells (Nguyen et al. 2009; Winkel et al. 2011). There have been many reports on pollution of As and other toxic elements in well drinking water in North Vietnam (Agusa et al. 2006, 2014; Berg et al. 2006; Nitzsche et al. 2015). Levels of As in well water are more than 1,000 μg/L (Berg et al. 2001), more than 100-fold higher than the value in WHO health-based guidelines for drinking water. Previous studies also showed increased levels of Fe and Mn in well water in North Vietnam (Agusa et al. 2006; Berg et al. 2006; Nitzsche et al. 2015). However, there is little information on Ba levels in well water in North Vietnam compared to the information on As, Fe, and Mn levels.

Another source of drinking water in Vietnam is harvested rainwater. The quality of harvested rainwater in the Mekong Delta was investigated in a previous study (Wilbers et al. 2013). The quality of well water and that of rainwater were compared in another study in Hanoi (Agusa et al. 2006). However, there has been no report comparing the quality of well water, rainwater, and tap water in North Vietnam.

Household sand filters are commonly used to remediate well water in North Vietnam in order to obtain drinkable water by removing Fe, which causes discoloration and a metallic taste (Berg et al. 2006). Previous studies showed that sand filters are effective for removing As, Fe, and Mn from well water (Berg et al. 2006; Nitzsche et al. 2015). However, information on the effectiveness of sand filters for removal of Ba in well water is limited despite the fact that previous studies showed high levels of Ba in well water in North Vietnam (Agusa et al. 2006; Winkel et al. 2011). To our knowledge, there is also limited information on the effectiveness of sand filters consisting of different components for removal of toxic elements in well water.

In order to assess the health risk of domestic water in North Vietnam, levels of four toxic elements (As, Ba, Fe, and Mn) were compared among well water, rainwater, and tap water in this study. The effectiveness of household sand filters consisting of different components for removal of toxic elements was also investigated.

Water samples were collected at Vinh Tru Town (six hamlets), Nhan Khang commune (three hamlets), and Dong Ly commune (one hamlet) in Ly Nhan District of Ha Nam Province, where pollution of As in well water was previously reported (Nguyen et al. 2009). A total of 20 untreated and 77 sand filter-treated well water samples were collected. In addition, 105 harvested rainwater and seven tap water samples were collected. All samples were collected in polyethylene bottles. Each bottle was filled with sampled water after rinsing out the inside of the bottle with sampled water. Tightly capped bottles were promptly sent from Vietnam to Japan by airplane. Samples were kept at 4 °C and measurements of total levels of As, Ba, Fe, and Mn were completed within 2 weeks after arrival at Nagoya University. Free and informed consent of the participants or their legal representatives was obtained, and the study protocol was approved by the Ethical Committee of Chubu University, Aichi, Japan (approval No. 20008 on July 9, 2008; approval No. 260019 on July 8, 2014) and the Ethical Committee of Nagoya University, Aichi, Japan (approval No. 2013-0070 on July 22, 2013) by following the regulations of the Japanese government.

Levels of As, Ba, Fe, and Mn in water samples were quantified by using an inductively coupled plasma-mass spectrometer (ICP-MS; 7500cx, Agilent Technologies Inc., CA, USA) according to the method previously described (Ohgami et al. 2012; Kato et al. 2013). The limits of detection for As, Ba, Fe, and Mn by the ICP-MS are 0.1, 0.1, 3.0, and 0.1 μg/L, respectively.

The Mann–Whitney U test and chi-square test were used for statistical analyses to compare differences between groups. The JMP Pro software package (version 11.0.0; SAS Institute, Cary, NC, USA) was used for all statistical analyses.

Since the levels of the four elements in well water were higher than those in rainwater and tap water, pollution of the toxic elements in well water was then focused on in this study. The median level (14.6 μg/L) of As in well water exceeded the value in the current WHO guidelines (10 μg/L) for drinking water, while the median level of Ba in well water was below the value in the current guidelines (700 μg/L). Although there are no guideline values for Fe and Mn in the present edition, the median levels of Fe and Mn in well water were below the guideline values in the previous editions (300 μg/L for Fe and 400 μg/L for Mn). The percentages of unsafe wells with levels of As, Ba, Fe, and Mn exceeding the values of WHO health-based guidelines for drinking water in the present and/or previous editions were 65%, 10%, 30%, and 20%, respectively (Table 1).

We first compared levels of As, Ba, Fe, and Mn in well water, rainwater, and tap water in North Vietnam. Since the level of As in 65% of the well water samples exceeded the value in WHO health-based guidelines for drinking water, well water is not suitable for drinking water, as was shown in previous studies (Berg et al. 2007; Agusa et al. 2014). Rainwater and tap water seem to be suitable for drinking from the viewpoint of the four toxic elements. Unexpectedly, however, As at a level of 27.8 μg/L was detected in one of seven tap water samples. Our results suggest that continuous monitoring for various kinds of water including tap water is important in the future in North Vietnam.

A previous study showed that the level of As in well water from Ly Nhan District, North Vietnam was 420 μg/L and that the level in sand-filtered water was 23 μg/L, with 80% of the filtered water still containing As at a level higher than 10 μg/L (Agusa et al. 2014). Other studies in Vietnam also showed that sand filters could remove 80% (Berg et al. 2006) to 95% of As (Nitzsche et al. 2015). Sand filters have been reported to be effective for the removal of Mn and Fe with 86.2–99.6% efficiency and nearly 100% efficiency, respectively (Nitzsche et al. 2015). Correspondingly, our results showed that levels of Fe and Mn in well water treated by sand filters were lower than those in untreated well water, suggesting that sand filters are effective for removal of Fe and Mn in well water. In contrast to previous studies (Agusa et al. 2014; Nitzsche et al. 2015), however, levels of As in untreated well water and in well water treated by sand filters were comparable in our study. A previous study showed that Fe/As ratios of ≥50 and >250 are required to reduce As concentrations to levels below 50 μg/L and 10 μg/L, respectively, indicating that the level of Fe in water affects removal of As by sand filters (Berg et al. 2006). The median of Fe/As ratio in well water was about 5 in this study (Figure S1, available with the online version of this paper) and this might be a reason why sand filters could not remove As from well water. Our results suggest that the Fe/As ratio in well water should be examined to determine the effectiveness of sand filters for removal of As. Since a sand filter composed of sand and zero-valent iron filings was shown to be effective for arsenic removal (Leupin & Hug 2005; Leupin et al. 2005; Mehta & Chaudhari 2015), the use of such a sand filter may be one way to overcome the low performance of a sand filter based on the Fe/As ratio. Proper design of a sand filter may also improve the effectiveness for As removal (Leupin et al. 2005). Our results showing a decrease of 24% in Ba in well water treated with sand filters suggest that sand filters can partially remove Ba from well water. To our knowledge, however, there has been no study showing the effectiveness of charcoal and a charcoal-containing sand filter for Ba removal. Further study on their effectiveness for Ba removal is needed.

After confirming pollution of As, Ba, Fe, and Mn in well water in North Vietnam, we found that sand filters could remove Fe and Mn but not As, despite the fact that previous studies showed that sand filters could remove As (Berg et al. 2006; Nitzsche et al. 2015). We then showed that a low ratio of Fe/As in well water in North Vietnam might be one of the reasons for the poor performance of sand filters. Thus, this fieldwork study clarified characteristics of various sand filters being used in North Vietnam.

This study was supported in part by Grants-in-Aid for Scientific Research (A) (15H01743 and 15H02588), (B) (24390157 and 24406002), and (C) (16K11177), Grant-in-Aid for Challenging Exploratory Research (26670525), and Grant-in-Aid for Scientific Research on Innovative Areas (24108001) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), the Mitsui & Co. Ltd Environment Fund, The Mitsubishi Foundation, Ichihara International Scholarship Foundation, Health Sciences Foundation (Kenko-Kagaku Zaidan) and International Fellowship Programs For Foreign Researchers from Takeda Science Foundation. The funding sources had no involvement in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. All authors declare that there is no conflict of interest.