Chemosphere 119 (2015) 1141–1147
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Identification and characterization of steady and occluded water in drinking water distribution systems Huiyan Tong a, Peng Zhao a,⇑, Hongwei Zhang a, Yimei Tian a, Xi Chen a, Weigao Zhao a, Mei Li b,⇑ a b
Department of Environmental Engineering, School of Environmental Science and Engineering, Tianjin University, Tianjin, China State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, China
h i g h l i g h t s Characterization of water from the interior of tubercles to the bulk was analyzed. Occluded water is acid and rich in iron, chloride, sulfate, nitrate, and manganese. Steady water acts as a transition layer. Results further understandings of corrosion and secondary pollution in water quality.
a r t i c l e
i n f o
Article history: Received 27 June 2014 Received in revised form 28 August 2014 Accepted 4 October 2014 Available online 25 October 2014 Handling Editor: Shane Snyder Keywords: Drinking water quality Secondary pollution control Corrosion mechanism Steady water Occluded water
a b s t r a c t Deterioration and leakage of drinking water in distribution systems have been a major issue in the water industry for years, which are associated with corrosion. This paper discovers that occluded water in the scales of the pipes has an acidic environment and high concentration of iron, manganese, chloride, sulfate and nitrate, which aggravates many pipeline leakage accidents. Six types of water samples have been analyzed under the flowing and stagnant periods. Both the water in the exterior of the tubercles and stagnant water carry suspended iron particles, which explains the occurrence of ‘‘red water’’ when the system hydraulic conditions change. Nitrate is more concentrated in occluded water under flowing condition in comparison with that in flowing water. However, the concentration of nitrate in occluded water under stagnant condition is found to be less than that in stagnant water. A high concentration of manganese is found to exist in steady water, occluded water and stagnant water. These findings impact secondary pollution and the corrosion of pipes and containers used in drinking water distribution systems. The unique method that taking occluded water from tiny holes which were drilled from the pipes’ exteriors carefully according to the positions of corrosion scales has an important contribution to research on corrosion in distribution systems. And this paper furthers our understanding and contributes to the growing body of knowledge regarding occluded environments in corrosion scales. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Access to safe drinking water is a basic concern for human health. Therefore, the quality of drinking water has been a major issue in the water industry for the last few decades. Meanwhile, with the improvement of people’s living standards, the requirements of drinking water quality are getting higher and higher. The drinking water treatment effluent usually meets the standards ⇑ Corresponding authors at: Department of Environmental Engineering, School of Environmental Science and Engineering, Tianjin University, Tianjin 300072, China. Tel./fax: +86 22 27408298 (P. Zhao). School of the Environment, Nanjing University, 163 Xianlin Avenue, Nanjing 210023, Jiangsu Province, China. Tel./fax: +86 25 89680365 (M. Li). E-mail addresses: [email protected] (P. Zhao), [email protected] (M. Li). http://dx.doi.org/10.1016/j.chemosphere.2014.10.005 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
for chemical and microbiological quality. However, drinking water deteriorates in drinking water distribution systems (DWDS), and results in an increase in turbidity, iron concentration, the total bacterial count, quickly decay of disinfectant residual, and even aesthetic water quality problems (Franzmann et al., 2001; McNeill and Edwards, 2001; Cheng et al., 2005). According to a water quality survey in 36 cities in China, the quality of water in the pipe networks was worse than that of drinking water treatment effluent with turbidity increasing 0.3 NTU, chroma increasing 1.5 degrees, iron concentration increasing 0.02 mg L 1 and the bacteria number increasing 22.6 cfu mL 1 (Wang, 2007). Water quality changes with time and space in the DWDS due to unwanted physical, chemical and biochemical reactions occurring with long residence time (Edwards, 2004; Huck and Gagnon, 2004). Deterioration of
1142
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
water quality in DWDS would lead to consumer complaints and increase health risks. Metallic pipes have been used as the prime choice in the field of urban water supply for decades. The percentage of cast iron and steel pipes used in DWDS are estimated 90% in China, 56.6% in USA, and 67.2% in Italy (Li et al., 2010; S´wietlik et al., 2012). After years of use in DWDS, metallic pipes usually are covered with deposits of corrosion products. Internal corrosion of metallic pipes is common in DWDS, and is more aggravated with the pipe aging. The corrosion scale is one of the main factors causing the deterioration of drinking water quality in DWDS, such as ‘‘discolored water’’ issues (Sarin et al., 2004; Yang et al., 2012). Furthermore, corrosion scales could accumulate many trace inorganic contaminants (e.g., lead, arsenic, nickel, vanadium, and uranium) (Peng and Korshin, 2011; Peng et al., 2012), serve as the breeding ground for microbes, increase the energy required to deliver water, and lead to the pipe damage (Sarin et al., 2004). According to the statistics in China in 2011, the total leakage of water supply pipe network of China reached 60 108 m3 d 1 (Zhuang, 2013). The formation process and physicochemical characteristics of corrosion scales in drinking water pipes are affected by the pipe material, the hydraulic conditions and the contacted water quality that include pH, dissolved oxygen (DO), alkalinity, temperature, concentrations of sulfate, chloride and natural organic matter (NOM), disinfectant type and level, microorganisms, inhibitor applied, hydraulic patterns (Camper, 2004; Gerke et al., 2008; Ray et al., 2010). The process is a complex process. Despite numerous research papers dedicated to understand the corrosion problems, it needs more research to further our understanding of mechanisms and factors affecting corrosion and water quality in the distribution system. Typical iron corrosion scales are porous and have a layered structure: a surface layer, a hard shell-like layer, and a porous core layer (Sarin et al., 2001, 2004). The calculated porosity of the tubercle interiors is 40–54% (Sarin et al., 2001). The porous core contains solid and fluid, which has totally different composition with the drinking water. These facts have not been well analyzed. It was firstly discovered by Baylis (1926) in 1926. He found that the interior of the tubercle was filled with water, and the water had much lower pH and very high concentration of chloride and sulfate compared to flowing water. This was confirmed by Tuovinen et al. (1980) and Nawrocki et al. (2010). According to Nawrocki et al. (2010), the water surrounding the tubercles in the pipe and in the interior of the tubercles is rich in ions and has reductive properties and relatively high concentrations of simple carboxylic acids. In fact, corrosion develops on inner surfaces of the pipes. The water has an acid environment and has high concentration of iron, manganese, chloride and sulfate. It is different with the bulk water. It may open a door for new mechanism of corrosion. This is one possible factor in relation to many pipeline leakage accidents. And contaminants possibly migrate and transform between bulk water and the water inside the tubercles. It may cause the secondary pollution of drinking water quality. However, it is very difficult to sample the water in the interior of the tubercles. And the sample is very little and is easily diluted by stagnant or flowing water. Therefore, the study has been stalled for so many years. The purpose of this paper is to characterize the water in the interior of the tubercles and to compare the water quality amongst the following water, the stagnant water, the water in the exterior of the tubercles, and the water in the interior of the tubercles. With considerations of the two kinds of hydraulic conditions, six types of water samples were collected for the analysis. The pH, inorganic ions (Cl , SO24 , NO3 ), metal ions (Fe, Na, Mg, K, Ca, Mn, Zn, Sr) were measured. The comparison of chemical composition of the six types of water was also analyzed. The findings of this work have an important contribution to the growing body of knowledge of corrosion in the pipe network, better control of secondary pollu-
tion of water pipe network and pipeline leakage problems. And the method of taking occluded water is unique and has an important contribution to research on corrosion in distribution systems.
2. Materials and methods 2.1. Preparation of corrosion pipes Four fragments of corroded pipes obtained from replaced pipes of the working distribution system were used for test, marked as A, B, C and D. Two of the pipes are 100 mm diameter cast iron pipes used about 30 years and the other two pipes are 200 mm diameter steel pipes used about 26 years. Each fragment is about one meter long. The four pipes have different extent of internal corrosion. Four test devices were constructed carefully using the four pipes and the fragments were fixed in the holding racks and placed vertically. Tap water flowed into the system from the top and flowed out from the bottom, and flowed through the specimens as singlepass. The flow rate was 0.2 L s 1.
2.2. Water samples collection Six different types of waters were sampled from the four test pipes: (1) flowing water (FW) – that is the water delivered in DWDS, (2) steady water under flowing condition (SWFC) – the water that remains in tubercles’ exteriors when the water in the pipe is flowing, (3) occluded water under flowing condition (OWFC) – the water that remains in tubercles’ interiors when the water in the pipe is flowing, (4) stagnant water (SW) – it is the flowing water that prolonged retention in the pipe, (5) steady water under stagnant condition (SWSC) – the water that remains in tubercles’ exteriors when the water in the pipe is stagnating, (6) occluded water under stagnant condition (OWSC) – the water that remains in tubercles’ interiors when the water in the pipe is stagnating. The six different types of waters are illustrated in Fig. 1. The flowing water was taken by capillary pipettes in the middle of the pipes. The stagnate water was collected by capillary pipettes in the middle of the pipes after one day under stagnant condition. Because corrosion scales were mostly tough and fragile, tiny holes were drilled from the pipes’ exteriors carefully according to the positions of corrosion scales in the pipes, and 1 mL of occluded water from each hole was taken with syringes under flowing and stagnant periods respectively. And 1 mL of steady water was taken by capillary pipettes near the internal pipe wall corresponding to the position of each hole under flowing and stagnant periods respectively. 15 mL of each type of waters were sampled from one pipe fragments and stored at 4 °C prior to analysis.
2.3. Analytical methods The water collected was split into filtered (0.45 lm membrane) and unfiltered samples. Filtered samples were analyzed for dissolved iron, Na, Mg, K, Ca, Mn, Zn, Sr, Cl , SO24 , NO3 . Unfiltered samples were acidified with 1% nitric acid and analyzed for total iron concentration. The filtered and unfiltered water samples were both diluted to analyze the inorganic ions and metal ions. The pH of water samples was measured with the HACH HQ30d flexi portable meter. The inorganic anions (Cl , SO24 , NO3 ) were determined by ion chromatography on DIONEX DX-600 system. The metal ions (Fe, Na, Mg, K, Ca, Mn, Zn, Sr) are determined by Inductively Coupled Plasma Mass Spectrometry (Agilent 7700x) which operated in both the standard and dynamic reaction cell (DRC) modes. All experiments were conducted in triplicate.
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
1143
water under flowing condition and the diffusion process reduces the component differences of water in the interiors and exteriors of corrosion scales. The migration of hydrogen ions to the flowing water is restricted by the corrosion scales. In addition, the pH is affected by the anions that diffuse into the scale. The anions such as chloride and sulfate are enriched in the occluded water under flowing condition, the pH is expected to be low (Sarin et al., 2004). The pH of stagnant water seems not to be substantially affected by occluded water. But the pH of occluded water under stagnant condition is a little lower than that of occluded water under flowing condition except the samples in Pipe C, and is about 2.5 units lower than that of flowing water. Therefore it indicated that the velocity has a great influence on the degree of acidification of occluded water. 3.3. Fe analysis
Fig. 1. The idea of six types of waters in the corroded distribution pipe. (a) Flowing water (FW), steady water under flowing condition (SWFC), occluded water under flowing condition (OWFC), (b) stagnant water (SW), steady water under stagnant condition (SWSC), occluded water under stagnant condition (OWSC).
3. Results and discussion 3.1. General Six different types of waters were sampled from corroded pipes of drinking water distribution systems. All stagnant waters look yellow and contain suspension of iron products of corrosion. Steady waters are all the mixtures of water and particles. All occluded waters look black or brown, and contain suspended products of corrosion. Some of these particles are black and some of them are reddish brown. The results of water quality of the six different kinds of water from four corroded pipes of DWDS are shown in Table 1. Occluded water may be diluted by steady water, despite efforts were made to avoid the unwanted impacts. Steady water may be also diluted by flowing water or stagnant water. But the overall tendency remains the same: occluded water has astonishing different properties from that of flowing water. The property differences are discussed below. 3.2. pH analysis The pH of steady water under flowing condition is close to that of flowing water in all four pipe samples. But the pH of occluded water under flowing condition is much lower than that of flowing water. There are two major processes that affect hydrogen ion concentration in occluded water under flowing condition. The hydrolysis process of metal ions increases the acidification of occluded
The occluded water has very high total iron concentrations, up to 394 mg L 1 in occluded water under flowing condition and 1561 mg L 1 in occluded water under stagnant condition (Fig. 2). This indicates the high corrosion rate in the occluded spaces from one perspective. The high concentration of dissolved iron is also found in occluded water (Fig. 2). The anode reaction of iron corrosion is mainly the formation of ferrous iron from metal. The migration of ferrous irons to the flowing water is restricted due to corrosion scales (Zhu and Guo, 2008). Although high dissolved iron concentration exists in occluded water under flowing condition, flowing water has low concentration of dissolved iron and the total iron concentration meets the national standard of water quality (Fig. 2). These phenomena also indicate that iron release decreased, but the corrosion rate might increase (McNeill and Edwards, 2001). The stagnant water looks yellow and contains suspended particles. High levels of total iron can cause unpleasant smell, taste and ‘‘red water’’ phenomena (Lehtola et al., 2004). The steady water contains low concentrations of dissolved iron but high total iron concentrations (Fig. 2). This is because particles can be attached to the pipe walls of the system by the steady state shear stress. These particles display cohesive-like properties and build up in layers on the pipe wall, conditioned by the usual daily flow patterns within the system. Once changes in the system hydraulics and specific changes in shear stress at the pipe wall happen, particles can be detached from the pipe walls and discoloration material would be mobilized (Husband and Boxall, 2011; Furnass et al., 2013). This also explains why stagnant water carries many suspended iron particles from one perspective. 3.4. Chloride and sulfate analysis The concentrations of chloride and sulfate in occluded water under flowing condition are both much higher than that in flowing water (Fig. 3). The concentrations of chloride and sulfate in occluded water under stagnant condition are even much greater when compared to that in stagnant water (Fig. 3), and both largely increased in comparison with their concentrations in occluded water under flowing condition except the samples in Pipe C (Fig. 3). The mechanism of the enrichment of chloride and sulfate in occluded water is assumed from the following hypotheses. The first hypothesis is maintaining electroneutrality within the occluded water. The second hypothesis takes into account the products of corrosion. Corrosion scale, metal oxides and hydroxides deposition, is a deposited film with anionic selectivity. Anions such as chloride and sulfate are effectively adsorbed on the surface of corrosion scale and diffuse into the scale (Zhu and Guo, 2008). The third hypothesis is the destruction (upon oxygen from air) of green rust. They can be stabilized with chloride and sulfate, but they are labile and sensitive to oxygen. The destruction of green
1144
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
Table 1 The results of waters’ quality from Pipe A, B, C and D.
a
a
NO3 a
Samples
pH
Total irona
Dissolved irona
Naa
Mga
Ka
Caa
Mna
Zna
Sra
Cl
Pipe A
FW SWFC OWFC SW SWSC OWSC
7.12 7.25 5.12 6.93 7.06 4.72
0.248 13.599 394.409 5.393 10.067 1151.985
0.051 0.044 276.810 0.039 0.044 1014.386
22.420 25.522 19.997 27.163 27.732 20.724
20.451 20.677 15.575 21.271 21.803 15.823
5.705 5.895 4.944 5.710 5.904 4.292
4.262 3.523 2.854 3.815 3.927 2.767
0.011 0.015 2.742 0.184 0.172 8.155
0.046 0.097 0.117 0.034 0.027 0.208
0.317 0.339 0.262 0.375 0.379 0.254
65.041 62.592 325.387 49.630 47.861 537.826
133.572 117.460 517.198 115.359 111.573 1481.130
6.877 6.270 18.292 9.101 8.049 5.035
Pipe B
FW SWFC OWFC SW SWSC OWSC
7.20 7.36 5.46 7.06 6.91 4.75
0.256 13.162 165.400 0.927 7.300 1560.935
0.033 0.031 111.864 0.095 0.009 822.459
33.023 34.325 34.096 32.780 32.168 33.234
18.475 18.357 16.171 19.106 19.732 17.105
3.509 3.515 2.968 3.478 3.190 3.941
3.357 3.469 3.177 2.690 2.798 2.822
0.007 0.008 1.303 0.164 0.127 6.987
0.021 0.035 2.819 0.411 0.247 0.743
0.320 0.347 0.304 0.293 0.293 0.303
31.823 43.014 150.106 66.083 72.047 571.558
83.425 117.410 409.426 110.154 133.439 1554.076
9.097 11.489 10.061 10.786 10.283 9.495
Pipe C
FW SWFC OWFC SW SWSC OWSC
7.03 7.21 5.10 6.91 7.02 5.34
0.270 12.088 126.022 2.374 8.837 165.444
0.050 0.055 94.772 0.039 0.038 97.327
21.165 26.935 20.773 27.646 27.371 22.072
18.867 20.450 16.424 22.278 21.872 17.694
5.780 6.422 4.868 5.958 6.518 4.758
2.997 3.553 3.050 3.882 4.061 3.066
0.009 0.008 1.212 0.145 0.122 3.260
0.041 0.054 0.175 0.028 0.015 0.055
0.315 0.340 0.287 0.379 0.375 0.281
71.157 69.228 171.415 49.597 46.843 144.197
135.936 128.046 269.518 127.626 109.673 234.650
6.893 14.850 16.521 8.497 8.068 7.808
Pipe D
FW SWFC OWFC SW SWSC OWSC
7.18 7.23 5.61 6.97 6.47 4.48
0.205 48.948 270.471 2.245 32.685 1176.266
0.011 0.031 82.434 0.034 0.016 506.343
32.631 29.908 26.619 32.160 34.323 22.779
19.319 15.970 19.476 18.882 19.492 9.339
3.768 3.265 2.700 3.213 3.843 2.522
3.090 3.415 2.990 2.480 2.997 2.166
0.004 0.007 0.747 0.160 0.114 2.495
0.018 0.028 6.812 0.323 0.352 1.197
0.285 0.351 0.279 0.280 0.307 0.170
22.907 44.666 142.684 63.103 69.797 404.146
62.766 117.657 294.515 112.161 142.067 823.490
7.793 12.025 9.333 9.323 8.232 7.793
Concentration unit: mg L
a
SO24
Pipes ID
1
.
Fig. 2. Concentrations of total iron (a) and dissolved iron (b) in flowing water (FW), steady water under flowing condition (SWFC), occluded water under flowing condition (OWFC), (b) stagnant water (SW), steady water under stagnant condition (SWSC), occluded water under stagnant condition (OWSC).
Fig. 3. Concentrations of chloride (a) and sulfate (b) in flowing water (FW), occluded water under flowing condition (OWFC), stagnant water (SW) and occluded water under stagnant condition (OWSC).
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
1145
rust can cause the increase of chloride and sulfate concentrations (Nawrocki et al., 2010). The migration of chloride and sulfate between occluded water and flowing water needs further extensive research. 3.5. Nitrate analysis Nitrate results are a bit complicated. The concentration of nitrate in occluded water under flowing condition is higher than that in flowing water (Fig. 4). But the concentration of nitrate in occluded water under stagnant condition is less than that in stagnant water, and even fewer than that in occluded water under flowing condition (Fig. 4). However, the concentration of nitrate in stagnant water is greater than that in flowing water (Fig. 4). Occluded water located in the interior of corrosion scales can be in contact with Fe0. Nitrates can be reduced by Fe0 and/or ‘‘green rust’’ (Rodríguez-Maroto et al., 2009; Choi et al., 2012). This may be responsible for the decrease of nitrate concentration in occluded water under stagnant condition relative to that in stagnant water (Fig. 4). The mechanisms of the enrichment, reduction and migration of nitrate in drinking water especially in occluded water are not clear and need further study. 3.6. Mn analysis Mn is difficult to remove by the conventional process (Sun, 2006). The presence of Mn can cause aesthetic problems, complaints by consumers, and human health risk (Zeng et al., 2009; Barbeau et al., 2011). Thus, the World Health Organization and People’s Republic of China have issued health-based drinkingwater guidelines of 400 lg L 1 and 100 lg L 1 respectively (WHO, 2004; PRC, 2006). The concentrations of Mn in flowing water and steady water under flowing condition are very low – much lower than the national water quality standard (Fig. 5). However, occluded water under flowing condition contains high concentration of Mn which is even 260 times higher than that in flowing water (Fig. 5). We suppose there are two main possible reasons. One may be that inadequate removal of Mn at the water treatment plant can be introduced into the DWDS as dissolved Mn2+ and change into Mn deposition. Mn deposition can attach to the scale of the pipe under pressurized flow conditions. Deposited MnO2 can be reduced back to Mn2+ under anaerobic conditions, and in acid medium, Mn2+ is the most stable form and other various forms of Mn can be spon-
Fig. 5. Mn concentration in flowing water (FW), steady water under flowing condition (SWFC), occluded water under flowing condition (OWFC), stagnant water (SW), steady water under stagnant condition (SWSC), occluded water under stagnant condition (OWSC).
taneously reverted to Mn2+ (Xiao et al., 2007). In addition, high salinity in occluded water can produce the salt effect, causing the exchange of iron and Mn ion by the same charged ions, such as Na+ and Ca2+. The other reason may be that ductile iron pipes contain no more than 1.5% of Mn except iron, carbon and silicon. When flowing water is prolonged in the pipe, the contents of Mn in stagnant water, steady water under stagnant condition and occluded water under stagnant condition increase obviously (Fig. 5). The highest concentration increase of Mn is about 42 times in stagnant water, and about 16 times in steady water under stagnant condition. Occluded water under stagnant condition contains Mn as many as 8.155 mg L 1. They all largely exceed the national water quality standard. One of the possible reasons may be that the anaerobic environment accelerates the reduction of Mn2+ from deposited MnO2. In addition, the content increase of Mn in stagnant water and steady water under stagnant condition may be derived from occluded water under stagnant condition. Therefore, the high concentration of Mn in occluded water must be noted. The mechanism of the enrichment of Mn needs further researches. 3.7. Sr analysis From our results, we can see Sr exists not only in flowing water but also in occluded water. Ingestion of nonradioactive Sr has been considered a potential threat to human health in recent years (Eikenberg et al., 2001; Langley et al., 2009). And Sr has been listed on the Drinking Water Contaminant Candidate List 3 by the US EPA (CCL3, 2009). Potential concentrations of Sr and mechanisms of release and migration are currently unknown. In a word, Sr in the occluded water should not be neglected. 3.8. New insights on corrosion mechanism and secondary pollution
Fig. 4. Nitrate concentration in flowing water (FW), occluded water under flowing condition (OWFC), stagnant water (SW) and occluded water under stagnant condition (OWSC).
In fact, there are three types of waters in corroded drinking water pipes: bulk water, steady water and occluded water. Ions migrate and transform among the three kinds of water (Fig. 6). Occluded water has an acid environment, contains very high concentration of dissolved iron and also suspended Fe oxides (hydroxides), and is enriched of chloride, sulfate, nitrate, and manganese. Previously, numerous research papers have been dedicated to understand the corrosion processes considering bulk water as the solution medium. However, corrosion develops on inner surfaces of the pipes. The solution medium that the corroded floor contacts
1146
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
Fig. 6. Schematic representation of migration and transformation of main ions among bulk water, steady water and occluded water in corroded drinking water pipes.
directly is occluded water. The property of occluded water is much different to that of bulk water. And green rust is mostly observed in alkaline conditions while it is formed in occluded water which is characterized by a low pH. From the above we can see that the mechanism of corrosion and the formation of corrosion scales should be studied taking the occluded environment into account in the future. And migration and transformation of chloride, sulfate, nitrate, manganese among bulk water, steady water and occluded water should research clearly. Also, strontium in the occluded water should not be neglected during the water quality standard formulation stage.
pollution of water quality. Therefore, the occluded environment simulation device should be designed to study the corrosion mechanism in occluded water and the migration of the main cations and anions between bulk water and occluded water in the future. Acknowledgments This research was supported by National Natural Science Foundation of China (No. 51208353). We also thank Rushi Yao for his kind help. References
4. Conclusions An astonishing occluded environment is found inside the corrosion scales of the pipes. Occluded water under flowing condition has an acid environment, contains very high concentration of dissolved iron and also suspended Fe oxides (hydroxides), and is enriched of chloride, sulfate, nitrate and manganese. Occluded water under stagnant condition becomes more acid and the concentrations of iron, chloride, sulfate, manganese are all higher than those in occluded water under flowing condition. But fewer concentration of nitrate in occluded water under stagnant condition is found. Steady water acts as transition layers. They both carry low concentrations of dissolved iron but high total iron concentrations. The property of occluded water is much different to that of bulk water. The results contribute to the growing body of knowledge regarding occluded environments in corrosion. Once the effect of occluded water on the pipeline leakage is known and the control methods are provided, it would reduce the waste of water resources, greatly increase the utility income, and improve the quality of water services. In addition, contaminants migrate among bulk water, steady water and occluded water. This causes the secondary pollution of water quality. And when hydraulic conditions change, not only the property of the water in the bulk but also that of the occluded water change greatly. The role of the hydraulic conditions should be taken into consideration during the study of the mechanism of corrosion and the migration of ions. To sum up, all these results provide a better understanding of the mechanism of corrosion, and confirm necessity of further research on the role of occluded water in corrosion and secondary
Barbeau, B., Carriere, A., Bouchard, M.F., 2011. Spatial and temporal variations of manganese concentrations in drinking water. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 46 (6), 608–616. Baylis, J.R., 1926. Prevention of corrosion and ‘‘red water’’. J. Am. Water Works Assoc. 15 (6), 598–633. Camper, A.K., 2004. Involvement of humic substances in regrowth. Int. J. Food Microbiol. 92 (3), 355–364. Cheng, X., Peterkin, E., Burlingame, G.A., 2005. A study on volatile sulfide causes of odors at Philadelphia’s northeast water pollution control plant. Water Res. 39, 3781–3790. Choi, J., Batchelor, B., Won, C., Chung, J., 2012. Nitrate reduction by green rusts modified with trace metals. Chemosphere 86 (8), 860–865. Edwards, M., 2004. Controlling corrosion in drinking water distribution systems: a grand challenge for the 21st century. Water Sci. Technol. 49 (2), 1–8. Eikenberg, J., Tricca, A., Vezzu, G., Stille, P., Bajo, S., Ruethi, M., 2001. 228Ra/226Ra/ 224Ra and 87Sr/86Sr isotope relationships for determining interactions between ground and river water in the upper Rhine valley. J. Environ. Radioact. 54 (1), 133–162. Franzmann, P.D., Heitz, A., Zappia, L.R., Wajon, J.E., Xanthis, K., 2001. The formation of malodorous oligosulphides in treated groundwater: the role of biofilms and potential precursors. Water Res. 35 (7), 1730–1738. Furnass, W.R., Mounce, S.R., Boxall, J.B., 2013. Linking distribution system water quality issues to possible causes via hydraulic pathways. Environ. Model Softw. 40, 78–87. Gerke, T.L., Maynard, J.B., Schock, M.R., Lytle, D.L., 2008. Physiochemical characterization of five iron tubercles from a single drinking water distribution system: possible new insights on their formation and growth. Corros. Sci. 50 (7), 2030–2039. Huck, P.M., Gagnon, G.A., 2004. Understanding the distribution system as a bioreactor: a framework for managing heterotrophic plate count levels. Int. J. Food Microbiol. 92 (3), 347–353. Husband, P.S., Boxall, J.B., 2011. Asset deterioration and discolouration water distribution systems. Water Res. 45 (1), 113–124. Langley, S., Gault, A.G., Ibrahim, A., Takahashi, Y., Renaud, R., Fortin, D., Clark, I.D., Ferris, F.G., 2009. Sorption of strontium onto bacteriogenic iron oxides. Environ. Sci. Technol. 43 (4), 1008–1014. Lehtola, M.J., Nissinen, T.K., Miettinen, I.T., Martikainen, P.J., Vartiainen, T., 2004. Removal of soft deposits from the distribution system improves the drinking water quality. Water Res. 38 (3), 601–610.
H. Tong et al. / Chemosphere 119 (2015) 1141–1147 Li, Yafeng, Jiang, Baiyi, Zhao, Hongbin, Tao, Cuicui, 2010. Experimental study on internal corrosion of grey cast iron pipeline in water distribution. J. Shenyang Jianzhu Univ. (Natural Science) 26 (2), 321–325. McNeill, L.S., Edwards, M., 2001. Review of iron pipe corrosion in drinking water distribution systems. J. AWWA 93 (7), 88–100. Nawrocki, J., Raczyk-Stanisławiak, U., S´wietlik, J., Olejnik, A., Sroka, M.J., 2010. Corrosion in a distribution system: steady water and its composition. Water Res. 44 (6), 1863–1872. Peng, C.Y., Korshin, G.V., 2011. Speciation of trace inorganic contaminants in corrosion scales and deposits formed in drinking water distribution systems. Water Res. 45 (17), 5553–5563. Peng, C.Y., Hill, A.S., Friedman, M.J., Valentine, R.L., Larson, G.S., Romero, A.M.Y., Reiber, S.H., Korshin, G.V., 2012. Occurrence of trace inorganic contaminants in drinking water distribution systems. J. Am. Water Works Assoc. 104 (3), 53–54. PRC, 2006. State Standard of the People’s Republic of China: Standards for Drinking Water Quality, GB5749-2006. Ray, R.I., Lee, J.S., Little, B.J., Gerke, T.L., 2010. The anatomy of tubercles: a corrosion study in a fresh water estuary. Mater. Corros. 61 (12), 993–999. Rodríguez-Maroto, J.M., García-Herruzo, F., García-Rubio, A., Gómez-Lahoz, C., Vereda-Alonso, C., 2009. Kinetics of the chemical reduction of nitrate by zerovalent iron. Chemosphere 74 (6), 804–809. Sarin, P., Snoeyink, V.L., Bebee, J., Kriven, W.M., Clement, J.A., 2001. Physicochemical characteristics of corrosion scales in old iron pipes. Water Res. 35 (12), 2961–2969. Sarin, P., Snoeyink, V.L., Lytle, D.A., Kriven, W.M., 2004. Iron corrosion scales: model for scale growth, iron release, and colored water formation. J. Environ. Eng. 130 (4), 364–373.
1147
Sun, S., 2006. Removal of stable Mn/Fe from Taihu Lake water by preoxidation with permanganate composite. China Water Wastewater 22 (21), 6–8. S´wietlik, J., Raczyk-Stanisławiak, U., Piszora, P., Nawrocki, J., 2012. Corrosion in drinking water pipes: the importance of green rusts. Water Res. 46 (1), 1–10. Tuovinen, O.H., Button, K.S., Vuorinen, A., Carlson, L., Mair, D.M., Yut, L.A., 1980. Bacterial, chemical, and mineralogical characteristics of tubercles in distribution pipelines. J. Am. Water Works Assoc. 72 (11), 626–635. USEPA, 2009. Drinking Water Contamination Candidate List 3 Final Notice. Vol. 74FR51850, pp. 51850–51862. Wang, L.P., 2007. Study of Iron Stability Problem and Pipes Corrosion in Drinking Water Distribution System. Master Dissertation, Tianjin University. World Health Organization, 2004. Guidelines for Drinking Water Quality Geneva, vol. 1. Xiao, L.P., Liu, W.Y., Lei, L., 2007. Feasibility of fenton oxidation method in removing iron and manganese in acid mine water. J. Liaoning Tech. Univ. 26, 272–274. Yang, Fan, Shi, Baoyou, Junnong, Gu, Wang, Dongsheng, Yang, Min, 2012. Morphological and physicochemical characteristics of iron corrosion scales formed under different water source histories in a drinking water distribution system. Water Res. 46 (16), 5423–5433. Zeng, G., Liang, J., Guo, S., Shi, L., Xiang, L., Li, X., Du, C., 2009. Spatial analysis of human health risk associated with ingesting manganese in Huangxing Town, Middle China. Chemosphere 77 (3), 368–375. Zhu, Y., Guo, X., 2008. Underscales localized corrosion behavior of carbon steel in neutral solution. J. Chin. Soc. Corros. Protect. 5, 003. Zhuang, Y.W., 2013. Study on Leakage Characteristics of Water Distribution Network Using Blind Source Separation Theory. Master Dissertation, Harbin Institute of Technology.
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Identification and characterization of steady and occluded water in drinking water distribution systems Huiyan Tong a, Peng Zhao a,⇑, Hongwei Zhang a, Yimei Tian a, Xi Chen a, Weigao Zhao a, Mei Li b,⇑ a b
Department of Environmental Engineering, School of Environmental Science and Engineering, Tianjin University, Tianjin, China State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, China
h i g h l i g h t s Characterization of water from the interior of tubercles to the bulk was analyzed. Occluded water is acid and rich in iron, chloride, sulfate, nitrate, and manganese. Steady water acts as a transition layer. Results further understandings of corrosion and secondary pollution in water quality.
a r t i c l e
i n f o
Article history: Received 27 June 2014 Received in revised form 28 August 2014 Accepted 4 October 2014 Available online 25 October 2014 Handling Editor: Shane Snyder Keywords: Drinking water quality Secondary pollution control Corrosion mechanism Steady water Occluded water
a b s t r a c t Deterioration and leakage of drinking water in distribution systems have been a major issue in the water industry for years, which are associated with corrosion. This paper discovers that occluded water in the scales of the pipes has an acidic environment and high concentration of iron, manganese, chloride, sulfate and nitrate, which aggravates many pipeline leakage accidents. Six types of water samples have been analyzed under the flowing and stagnant periods. Both the water in the exterior of the tubercles and stagnant water carry suspended iron particles, which explains the occurrence of ‘‘red water’’ when the system hydraulic conditions change. Nitrate is more concentrated in occluded water under flowing condition in comparison with that in flowing water. However, the concentration of nitrate in occluded water under stagnant condition is found to be less than that in stagnant water. A high concentration of manganese is found to exist in steady water, occluded water and stagnant water. These findings impact secondary pollution and the corrosion of pipes and containers used in drinking water distribution systems. The unique method that taking occluded water from tiny holes which were drilled from the pipes’ exteriors carefully according to the positions of corrosion scales has an important contribution to research on corrosion in distribution systems. And this paper furthers our understanding and contributes to the growing body of knowledge regarding occluded environments in corrosion scales. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Access to safe drinking water is a basic concern for human health. Therefore, the quality of drinking water has been a major issue in the water industry for the last few decades. Meanwhile, with the improvement of people’s living standards, the requirements of drinking water quality are getting higher and higher. The drinking water treatment effluent usually meets the standards ⇑ Corresponding authors at: Department of Environmental Engineering, School of Environmental Science and Engineering, Tianjin University, Tianjin 300072, China. Tel./fax: +86 22 27408298 (P. Zhao). School of the Environment, Nanjing University, 163 Xianlin Avenue, Nanjing 210023, Jiangsu Province, China. Tel./fax: +86 25 89680365 (M. Li). E-mail addresses: [email protected] (P. Zhao), [email protected] (M. Li). http://dx.doi.org/10.1016/j.chemosphere.2014.10.005 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
for chemical and microbiological quality. However, drinking water deteriorates in drinking water distribution systems (DWDS), and results in an increase in turbidity, iron concentration, the total bacterial count, quickly decay of disinfectant residual, and even aesthetic water quality problems (Franzmann et al., 2001; McNeill and Edwards, 2001; Cheng et al., 2005). According to a water quality survey in 36 cities in China, the quality of water in the pipe networks was worse than that of drinking water treatment effluent with turbidity increasing 0.3 NTU, chroma increasing 1.5 degrees, iron concentration increasing 0.02 mg L 1 and the bacteria number increasing 22.6 cfu mL 1 (Wang, 2007). Water quality changes with time and space in the DWDS due to unwanted physical, chemical and biochemical reactions occurring with long residence time (Edwards, 2004; Huck and Gagnon, 2004). Deterioration of
1142
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
water quality in DWDS would lead to consumer complaints and increase health risks. Metallic pipes have been used as the prime choice in the field of urban water supply for decades. The percentage of cast iron and steel pipes used in DWDS are estimated 90% in China, 56.6% in USA, and 67.2% in Italy (Li et al., 2010; S´wietlik et al., 2012). After years of use in DWDS, metallic pipes usually are covered with deposits of corrosion products. Internal corrosion of metallic pipes is common in DWDS, and is more aggravated with the pipe aging. The corrosion scale is one of the main factors causing the deterioration of drinking water quality in DWDS, such as ‘‘discolored water’’ issues (Sarin et al., 2004; Yang et al., 2012). Furthermore, corrosion scales could accumulate many trace inorganic contaminants (e.g., lead, arsenic, nickel, vanadium, and uranium) (Peng and Korshin, 2011; Peng et al., 2012), serve as the breeding ground for microbes, increase the energy required to deliver water, and lead to the pipe damage (Sarin et al., 2004). According to the statistics in China in 2011, the total leakage of water supply pipe network of China reached 60 108 m3 d 1 (Zhuang, 2013). The formation process and physicochemical characteristics of corrosion scales in drinking water pipes are affected by the pipe material, the hydraulic conditions and the contacted water quality that include pH, dissolved oxygen (DO), alkalinity, temperature, concentrations of sulfate, chloride and natural organic matter (NOM), disinfectant type and level, microorganisms, inhibitor applied, hydraulic patterns (Camper, 2004; Gerke et al., 2008; Ray et al., 2010). The process is a complex process. Despite numerous research papers dedicated to understand the corrosion problems, it needs more research to further our understanding of mechanisms and factors affecting corrosion and water quality in the distribution system. Typical iron corrosion scales are porous and have a layered structure: a surface layer, a hard shell-like layer, and a porous core layer (Sarin et al., 2001, 2004). The calculated porosity of the tubercle interiors is 40–54% (Sarin et al., 2001). The porous core contains solid and fluid, which has totally different composition with the drinking water. These facts have not been well analyzed. It was firstly discovered by Baylis (1926) in 1926. He found that the interior of the tubercle was filled with water, and the water had much lower pH and very high concentration of chloride and sulfate compared to flowing water. This was confirmed by Tuovinen et al. (1980) and Nawrocki et al. (2010). According to Nawrocki et al. (2010), the water surrounding the tubercles in the pipe and in the interior of the tubercles is rich in ions and has reductive properties and relatively high concentrations of simple carboxylic acids. In fact, corrosion develops on inner surfaces of the pipes. The water has an acid environment and has high concentration of iron, manganese, chloride and sulfate. It is different with the bulk water. It may open a door for new mechanism of corrosion. This is one possible factor in relation to many pipeline leakage accidents. And contaminants possibly migrate and transform between bulk water and the water inside the tubercles. It may cause the secondary pollution of drinking water quality. However, it is very difficult to sample the water in the interior of the tubercles. And the sample is very little and is easily diluted by stagnant or flowing water. Therefore, the study has been stalled for so many years. The purpose of this paper is to characterize the water in the interior of the tubercles and to compare the water quality amongst the following water, the stagnant water, the water in the exterior of the tubercles, and the water in the interior of the tubercles. With considerations of the two kinds of hydraulic conditions, six types of water samples were collected for the analysis. The pH, inorganic ions (Cl , SO24 , NO3 ), metal ions (Fe, Na, Mg, K, Ca, Mn, Zn, Sr) were measured. The comparison of chemical composition of the six types of water was also analyzed. The findings of this work have an important contribution to the growing body of knowledge of corrosion in the pipe network, better control of secondary pollu-
tion of water pipe network and pipeline leakage problems. And the method of taking occluded water is unique and has an important contribution to research on corrosion in distribution systems.
2. Materials and methods 2.1. Preparation of corrosion pipes Four fragments of corroded pipes obtained from replaced pipes of the working distribution system were used for test, marked as A, B, C and D. Two of the pipes are 100 mm diameter cast iron pipes used about 30 years and the other two pipes are 200 mm diameter steel pipes used about 26 years. Each fragment is about one meter long. The four pipes have different extent of internal corrosion. Four test devices were constructed carefully using the four pipes and the fragments were fixed in the holding racks and placed vertically. Tap water flowed into the system from the top and flowed out from the bottom, and flowed through the specimens as singlepass. The flow rate was 0.2 L s 1.
2.2. Water samples collection Six different types of waters were sampled from the four test pipes: (1) flowing water (FW) – that is the water delivered in DWDS, (2) steady water under flowing condition (SWFC) – the water that remains in tubercles’ exteriors when the water in the pipe is flowing, (3) occluded water under flowing condition (OWFC) – the water that remains in tubercles’ interiors when the water in the pipe is flowing, (4) stagnant water (SW) – it is the flowing water that prolonged retention in the pipe, (5) steady water under stagnant condition (SWSC) – the water that remains in tubercles’ exteriors when the water in the pipe is stagnating, (6) occluded water under stagnant condition (OWSC) – the water that remains in tubercles’ interiors when the water in the pipe is stagnating. The six different types of waters are illustrated in Fig. 1. The flowing water was taken by capillary pipettes in the middle of the pipes. The stagnate water was collected by capillary pipettes in the middle of the pipes after one day under stagnant condition. Because corrosion scales were mostly tough and fragile, tiny holes were drilled from the pipes’ exteriors carefully according to the positions of corrosion scales in the pipes, and 1 mL of occluded water from each hole was taken with syringes under flowing and stagnant periods respectively. And 1 mL of steady water was taken by capillary pipettes near the internal pipe wall corresponding to the position of each hole under flowing and stagnant periods respectively. 15 mL of each type of waters were sampled from one pipe fragments and stored at 4 °C prior to analysis.
2.3. Analytical methods The water collected was split into filtered (0.45 lm membrane) and unfiltered samples. Filtered samples were analyzed for dissolved iron, Na, Mg, K, Ca, Mn, Zn, Sr, Cl , SO24 , NO3 . Unfiltered samples were acidified with 1% nitric acid and analyzed for total iron concentration. The filtered and unfiltered water samples were both diluted to analyze the inorganic ions and metal ions. The pH of water samples was measured with the HACH HQ30d flexi portable meter. The inorganic anions (Cl , SO24 , NO3 ) were determined by ion chromatography on DIONEX DX-600 system. The metal ions (Fe, Na, Mg, K, Ca, Mn, Zn, Sr) are determined by Inductively Coupled Plasma Mass Spectrometry (Agilent 7700x) which operated in both the standard and dynamic reaction cell (DRC) modes. All experiments were conducted in triplicate.
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
1143
water under flowing condition and the diffusion process reduces the component differences of water in the interiors and exteriors of corrosion scales. The migration of hydrogen ions to the flowing water is restricted by the corrosion scales. In addition, the pH is affected by the anions that diffuse into the scale. The anions such as chloride and sulfate are enriched in the occluded water under flowing condition, the pH is expected to be low (Sarin et al., 2004). The pH of stagnant water seems not to be substantially affected by occluded water. But the pH of occluded water under stagnant condition is a little lower than that of occluded water under flowing condition except the samples in Pipe C, and is about 2.5 units lower than that of flowing water. Therefore it indicated that the velocity has a great influence on the degree of acidification of occluded water. 3.3. Fe analysis
Fig. 1. The idea of six types of waters in the corroded distribution pipe. (a) Flowing water (FW), steady water under flowing condition (SWFC), occluded water under flowing condition (OWFC), (b) stagnant water (SW), steady water under stagnant condition (SWSC), occluded water under stagnant condition (OWSC).
3. Results and discussion 3.1. General Six different types of waters were sampled from corroded pipes of drinking water distribution systems. All stagnant waters look yellow and contain suspension of iron products of corrosion. Steady waters are all the mixtures of water and particles. All occluded waters look black or brown, and contain suspended products of corrosion. Some of these particles are black and some of them are reddish brown. The results of water quality of the six different kinds of water from four corroded pipes of DWDS are shown in Table 1. Occluded water may be diluted by steady water, despite efforts were made to avoid the unwanted impacts. Steady water may be also diluted by flowing water or stagnant water. But the overall tendency remains the same: occluded water has astonishing different properties from that of flowing water. The property differences are discussed below. 3.2. pH analysis The pH of steady water under flowing condition is close to that of flowing water in all four pipe samples. But the pH of occluded water under flowing condition is much lower than that of flowing water. There are two major processes that affect hydrogen ion concentration in occluded water under flowing condition. The hydrolysis process of metal ions increases the acidification of occluded
The occluded water has very high total iron concentrations, up to 394 mg L 1 in occluded water under flowing condition and 1561 mg L 1 in occluded water under stagnant condition (Fig. 2). This indicates the high corrosion rate in the occluded spaces from one perspective. The high concentration of dissolved iron is also found in occluded water (Fig. 2). The anode reaction of iron corrosion is mainly the formation of ferrous iron from metal. The migration of ferrous irons to the flowing water is restricted due to corrosion scales (Zhu and Guo, 2008). Although high dissolved iron concentration exists in occluded water under flowing condition, flowing water has low concentration of dissolved iron and the total iron concentration meets the national standard of water quality (Fig. 2). These phenomena also indicate that iron release decreased, but the corrosion rate might increase (McNeill and Edwards, 2001). The stagnant water looks yellow and contains suspended particles. High levels of total iron can cause unpleasant smell, taste and ‘‘red water’’ phenomena (Lehtola et al., 2004). The steady water contains low concentrations of dissolved iron but high total iron concentrations (Fig. 2). This is because particles can be attached to the pipe walls of the system by the steady state shear stress. These particles display cohesive-like properties and build up in layers on the pipe wall, conditioned by the usual daily flow patterns within the system. Once changes in the system hydraulics and specific changes in shear stress at the pipe wall happen, particles can be detached from the pipe walls and discoloration material would be mobilized (Husband and Boxall, 2011; Furnass et al., 2013). This also explains why stagnant water carries many suspended iron particles from one perspective. 3.4. Chloride and sulfate analysis The concentrations of chloride and sulfate in occluded water under flowing condition are both much higher than that in flowing water (Fig. 3). The concentrations of chloride and sulfate in occluded water under stagnant condition are even much greater when compared to that in stagnant water (Fig. 3), and both largely increased in comparison with their concentrations in occluded water under flowing condition except the samples in Pipe C (Fig. 3). The mechanism of the enrichment of chloride and sulfate in occluded water is assumed from the following hypotheses. The first hypothesis is maintaining electroneutrality within the occluded water. The second hypothesis takes into account the products of corrosion. Corrosion scale, metal oxides and hydroxides deposition, is a deposited film with anionic selectivity. Anions such as chloride and sulfate are effectively adsorbed on the surface of corrosion scale and diffuse into the scale (Zhu and Guo, 2008). The third hypothesis is the destruction (upon oxygen from air) of green rust. They can be stabilized with chloride and sulfate, but they are labile and sensitive to oxygen. The destruction of green
1144
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
Table 1 The results of waters’ quality from Pipe A, B, C and D.
a
a
NO3 a
Samples
pH
Total irona
Dissolved irona
Naa
Mga
Ka
Caa
Mna
Zna
Sra
Cl
Pipe A
FW SWFC OWFC SW SWSC OWSC
7.12 7.25 5.12 6.93 7.06 4.72
0.248 13.599 394.409 5.393 10.067 1151.985
0.051 0.044 276.810 0.039 0.044 1014.386
22.420 25.522 19.997 27.163 27.732 20.724
20.451 20.677 15.575 21.271 21.803 15.823
5.705 5.895 4.944 5.710 5.904 4.292
4.262 3.523 2.854 3.815 3.927 2.767
0.011 0.015 2.742 0.184 0.172 8.155
0.046 0.097 0.117 0.034 0.027 0.208
0.317 0.339 0.262 0.375 0.379 0.254
65.041 62.592 325.387 49.630 47.861 537.826
133.572 117.460 517.198 115.359 111.573 1481.130
6.877 6.270 18.292 9.101 8.049 5.035
Pipe B
FW SWFC OWFC SW SWSC OWSC
7.20 7.36 5.46 7.06 6.91 4.75
0.256 13.162 165.400 0.927 7.300 1560.935
0.033 0.031 111.864 0.095 0.009 822.459
33.023 34.325 34.096 32.780 32.168 33.234
18.475 18.357 16.171 19.106 19.732 17.105
3.509 3.515 2.968 3.478 3.190 3.941
3.357 3.469 3.177 2.690 2.798 2.822
0.007 0.008 1.303 0.164 0.127 6.987
0.021 0.035 2.819 0.411 0.247 0.743
0.320 0.347 0.304 0.293 0.293 0.303
31.823 43.014 150.106 66.083 72.047 571.558
83.425 117.410 409.426 110.154 133.439 1554.076
9.097 11.489 10.061 10.786 10.283 9.495
Pipe C
FW SWFC OWFC SW SWSC OWSC
7.03 7.21 5.10 6.91 7.02 5.34
0.270 12.088 126.022 2.374 8.837 165.444
0.050 0.055 94.772 0.039 0.038 97.327
21.165 26.935 20.773 27.646 27.371 22.072
18.867 20.450 16.424 22.278 21.872 17.694
5.780 6.422 4.868 5.958 6.518 4.758
2.997 3.553 3.050 3.882 4.061 3.066
0.009 0.008 1.212 0.145 0.122 3.260
0.041 0.054 0.175 0.028 0.015 0.055
0.315 0.340 0.287 0.379 0.375 0.281
71.157 69.228 171.415 49.597 46.843 144.197
135.936 128.046 269.518 127.626 109.673 234.650
6.893 14.850 16.521 8.497 8.068 7.808
Pipe D
FW SWFC OWFC SW SWSC OWSC
7.18 7.23 5.61 6.97 6.47 4.48
0.205 48.948 270.471 2.245 32.685 1176.266
0.011 0.031 82.434 0.034 0.016 506.343
32.631 29.908 26.619 32.160 34.323 22.779
19.319 15.970 19.476 18.882 19.492 9.339
3.768 3.265 2.700 3.213 3.843 2.522
3.090 3.415 2.990 2.480 2.997 2.166
0.004 0.007 0.747 0.160 0.114 2.495
0.018 0.028 6.812 0.323 0.352 1.197
0.285 0.351 0.279 0.280 0.307 0.170
22.907 44.666 142.684 63.103 69.797 404.146
62.766 117.657 294.515 112.161 142.067 823.490
7.793 12.025 9.333 9.323 8.232 7.793
Concentration unit: mg L
a
SO24
Pipes ID
1
.
Fig. 2. Concentrations of total iron (a) and dissolved iron (b) in flowing water (FW), steady water under flowing condition (SWFC), occluded water under flowing condition (OWFC), (b) stagnant water (SW), steady water under stagnant condition (SWSC), occluded water under stagnant condition (OWSC).
Fig. 3. Concentrations of chloride (a) and sulfate (b) in flowing water (FW), occluded water under flowing condition (OWFC), stagnant water (SW) and occluded water under stagnant condition (OWSC).
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
1145
rust can cause the increase of chloride and sulfate concentrations (Nawrocki et al., 2010). The migration of chloride and sulfate between occluded water and flowing water needs further extensive research. 3.5. Nitrate analysis Nitrate results are a bit complicated. The concentration of nitrate in occluded water under flowing condition is higher than that in flowing water (Fig. 4). But the concentration of nitrate in occluded water under stagnant condition is less than that in stagnant water, and even fewer than that in occluded water under flowing condition (Fig. 4). However, the concentration of nitrate in stagnant water is greater than that in flowing water (Fig. 4). Occluded water located in the interior of corrosion scales can be in contact with Fe0. Nitrates can be reduced by Fe0 and/or ‘‘green rust’’ (Rodríguez-Maroto et al., 2009; Choi et al., 2012). This may be responsible for the decrease of nitrate concentration in occluded water under stagnant condition relative to that in stagnant water (Fig. 4). The mechanisms of the enrichment, reduction and migration of nitrate in drinking water especially in occluded water are not clear and need further study. 3.6. Mn analysis Mn is difficult to remove by the conventional process (Sun, 2006). The presence of Mn can cause aesthetic problems, complaints by consumers, and human health risk (Zeng et al., 2009; Barbeau et al., 2011). Thus, the World Health Organization and People’s Republic of China have issued health-based drinkingwater guidelines of 400 lg L 1 and 100 lg L 1 respectively (WHO, 2004; PRC, 2006). The concentrations of Mn in flowing water and steady water under flowing condition are very low – much lower than the national water quality standard (Fig. 5). However, occluded water under flowing condition contains high concentration of Mn which is even 260 times higher than that in flowing water (Fig. 5). We suppose there are two main possible reasons. One may be that inadequate removal of Mn at the water treatment plant can be introduced into the DWDS as dissolved Mn2+ and change into Mn deposition. Mn deposition can attach to the scale of the pipe under pressurized flow conditions. Deposited MnO2 can be reduced back to Mn2+ under anaerobic conditions, and in acid medium, Mn2+ is the most stable form and other various forms of Mn can be spon-
Fig. 5. Mn concentration in flowing water (FW), steady water under flowing condition (SWFC), occluded water under flowing condition (OWFC), stagnant water (SW), steady water under stagnant condition (SWSC), occluded water under stagnant condition (OWSC).
taneously reverted to Mn2+ (Xiao et al., 2007). In addition, high salinity in occluded water can produce the salt effect, causing the exchange of iron and Mn ion by the same charged ions, such as Na+ and Ca2+. The other reason may be that ductile iron pipes contain no more than 1.5% of Mn except iron, carbon and silicon. When flowing water is prolonged in the pipe, the contents of Mn in stagnant water, steady water under stagnant condition and occluded water under stagnant condition increase obviously (Fig. 5). The highest concentration increase of Mn is about 42 times in stagnant water, and about 16 times in steady water under stagnant condition. Occluded water under stagnant condition contains Mn as many as 8.155 mg L 1. They all largely exceed the national water quality standard. One of the possible reasons may be that the anaerobic environment accelerates the reduction of Mn2+ from deposited MnO2. In addition, the content increase of Mn in stagnant water and steady water under stagnant condition may be derived from occluded water under stagnant condition. Therefore, the high concentration of Mn in occluded water must be noted. The mechanism of the enrichment of Mn needs further researches. 3.7. Sr analysis From our results, we can see Sr exists not only in flowing water but also in occluded water. Ingestion of nonradioactive Sr has been considered a potential threat to human health in recent years (Eikenberg et al., 2001; Langley et al., 2009). And Sr has been listed on the Drinking Water Contaminant Candidate List 3 by the US EPA (CCL3, 2009). Potential concentrations of Sr and mechanisms of release and migration are currently unknown. In a word, Sr in the occluded water should not be neglected. 3.8. New insights on corrosion mechanism and secondary pollution
Fig. 4. Nitrate concentration in flowing water (FW), occluded water under flowing condition (OWFC), stagnant water (SW) and occluded water under stagnant condition (OWSC).
In fact, there are three types of waters in corroded drinking water pipes: bulk water, steady water and occluded water. Ions migrate and transform among the three kinds of water (Fig. 6). Occluded water has an acid environment, contains very high concentration of dissolved iron and also suspended Fe oxides (hydroxides), and is enriched of chloride, sulfate, nitrate, and manganese. Previously, numerous research papers have been dedicated to understand the corrosion processes considering bulk water as the solution medium. However, corrosion develops on inner surfaces of the pipes. The solution medium that the corroded floor contacts
1146
H. Tong et al. / Chemosphere 119 (2015) 1141–1147
Fig. 6. Schematic representation of migration and transformation of main ions among bulk water, steady water and occluded water in corroded drinking water pipes.
directly is occluded water. The property of occluded water is much different to that of bulk water. And green rust is mostly observed in alkaline conditions while it is formed in occluded water which is characterized by a low pH. From the above we can see that the mechanism of corrosion and the formation of corrosion scales should be studied taking the occluded environment into account in the future. And migration and transformation of chloride, sulfate, nitrate, manganese among bulk water, steady water and occluded water should research clearly. Also, strontium in the occluded water should not be neglected during the water quality standard formulation stage.
pollution of water quality. Therefore, the occluded environment simulation device should be designed to study the corrosion mechanism in occluded water and the migration of the main cations and anions between bulk water and occluded water in the future. Acknowledgments This research was supported by National Natural Science Foundation of China (No. 51208353). We also thank Rushi Yao for his kind help. References
4. Conclusions An astonishing occluded environment is found inside the corrosion scales of the pipes. Occluded water under flowing condition has an acid environment, contains very high concentration of dissolved iron and also suspended Fe oxides (hydroxides), and is enriched of chloride, sulfate, nitrate and manganese. Occluded water under stagnant condition becomes more acid and the concentrations of iron, chloride, sulfate, manganese are all higher than those in occluded water under flowing condition. But fewer concentration of nitrate in occluded water under stagnant condition is found. Steady water acts as transition layers. They both carry low concentrations of dissolved iron but high total iron concentrations. The property of occluded water is much different to that of bulk water. The results contribute to the growing body of knowledge regarding occluded environments in corrosion. Once the effect of occluded water on the pipeline leakage is known and the control methods are provided, it would reduce the waste of water resources, greatly increase the utility income, and improve the quality of water services. In addition, contaminants migrate among bulk water, steady water and occluded water. This causes the secondary pollution of water quality. And when hydraulic conditions change, not only the property of the water in the bulk but also that of the occluded water change greatly. The role of the hydraulic conditions should be taken into consideration during the study of the mechanism of corrosion and the migration of ions. To sum up, all these results provide a better understanding of the mechanism of corrosion, and confirm necessity of further research on the role of occluded water in corrosion and secondary
Barbeau, B., Carriere, A., Bouchard, M.F., 2011. Spatial and temporal variations of manganese concentrations in drinking water. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 46 (6), 608–616. Baylis, J.R., 1926. Prevention of corrosion and ‘‘red water’’. J. Am. Water Works Assoc. 15 (6), 598–633. Camper, A.K., 2004. Involvement of humic substances in regrowth. Int. J. Food Microbiol. 92 (3), 355–364. Cheng, X., Peterkin, E., Burlingame, G.A., 2005. A study on volatile sulfide causes of odors at Philadelphia’s northeast water pollution control plant. Water Res. 39, 3781–3790. Choi, J., Batchelor, B., Won, C., Chung, J., 2012. Nitrate reduction by green rusts modified with trace metals. Chemosphere 86 (8), 860–865. Edwards, M., 2004. Controlling corrosion in drinking water distribution systems: a grand challenge for the 21st century. Water Sci. Technol. 49 (2), 1–8. Eikenberg, J., Tricca, A., Vezzu, G., Stille, P., Bajo, S., Ruethi, M., 2001. 228Ra/226Ra/ 224Ra and 87Sr/86Sr isotope relationships for determining interactions between ground and river water in the upper Rhine valley. J. Environ. Radioact. 54 (1), 133–162. Franzmann, P.D., Heitz, A., Zappia, L.R., Wajon, J.E., Xanthis, K., 2001. The formation of malodorous oligosulphides in treated groundwater: the role of biofilms and potential precursors. Water Res. 35 (7), 1730–1738. Furnass, W.R., Mounce, S.R., Boxall, J.B., 2013. Linking distribution system water quality issues to possible causes via hydraulic pathways. Environ. Model Softw. 40, 78–87. Gerke, T.L., Maynard, J.B., Schock, M.R., Lytle, D.L., 2008. Physiochemical characterization of five iron tubercles from a single drinking water distribution system: possible new insights on their formation and growth. Corros. Sci. 50 (7), 2030–2039. Huck, P.M., Gagnon, G.A., 2004. Understanding the distribution system as a bioreactor: a framework for managing heterotrophic plate count levels. Int. J. Food Microbiol. 92 (3), 347–353. Husband, P.S., Boxall, J.B., 2011. Asset deterioration and discolouration water distribution systems. Water Res. 45 (1), 113–124. Langley, S., Gault, A.G., Ibrahim, A., Takahashi, Y., Renaud, R., Fortin, D., Clark, I.D., Ferris, F.G., 2009. Sorption of strontium onto bacteriogenic iron oxides. Environ. Sci. Technol. 43 (4), 1008–1014. Lehtola, M.J., Nissinen, T.K., Miettinen, I.T., Martikainen, P.J., Vartiainen, T., 2004. Removal of soft deposits from the distribution system improves the drinking water quality. Water Res. 38 (3), 601–610.
H. Tong et al. / Chemosphere 119 (2015) 1141–1147 Li, Yafeng, Jiang, Baiyi, Zhao, Hongbin, Tao, Cuicui, 2010. Experimental study on internal corrosion of grey cast iron pipeline in water distribution. J. Shenyang Jianzhu Univ. (Natural Science) 26 (2), 321–325. McNeill, L.S., Edwards, M., 2001. Review of iron pipe corrosion in drinking water distribution systems. J. AWWA 93 (7), 88–100. Nawrocki, J., Raczyk-Stanisławiak, U., S´wietlik, J., Olejnik, A., Sroka, M.J., 2010. Corrosion in a distribution system: steady water and its composition. Water Res. 44 (6), 1863–1872. Peng, C.Y., Korshin, G.V., 2011. Speciation of trace inorganic contaminants in corrosion scales and deposits formed in drinking water distribution systems. Water Res. 45 (17), 5553–5563. Peng, C.Y., Hill, A.S., Friedman, M.J., Valentine, R.L., Larson, G.S., Romero, A.M.Y., Reiber, S.H., Korshin, G.V., 2012. Occurrence of trace inorganic contaminants in drinking water distribution systems. J. Am. Water Works Assoc. 104 (3), 53–54. PRC, 2006. State Standard of the People’s Republic of China: Standards for Drinking Water Quality, GB5749-2006. Ray, R.I., Lee, J.S., Little, B.J., Gerke, T.L., 2010. The anatomy of tubercles: a corrosion study in a fresh water estuary. Mater. Corros. 61 (12), 993–999. Rodríguez-Maroto, J.M., García-Herruzo, F., García-Rubio, A., Gómez-Lahoz, C., Vereda-Alonso, C., 2009. Kinetics of the chemical reduction of nitrate by zerovalent iron. Chemosphere 74 (6), 804–809. Sarin, P., Snoeyink, V.L., Bebee, J., Kriven, W.M., Clement, J.A., 2001. Physicochemical characteristics of corrosion scales in old iron pipes. Water Res. 35 (12), 2961–2969. Sarin, P., Snoeyink, V.L., Lytle, D.A., Kriven, W.M., 2004. Iron corrosion scales: model for scale growth, iron release, and colored water formation. J. Environ. Eng. 130 (4), 364–373.
1147
Sun, S., 2006. Removal of stable Mn/Fe from Taihu Lake water by preoxidation with permanganate composite. China Water Wastewater 22 (21), 6–8. S´wietlik, J., Raczyk-Stanisławiak, U., Piszora, P., Nawrocki, J., 2012. Corrosion in drinking water pipes: the importance of green rusts. Water Res. 46 (1), 1–10. Tuovinen, O.H., Button, K.S., Vuorinen, A., Carlson, L., Mair, D.M., Yut, L.A., 1980. Bacterial, chemical, and mineralogical characteristics of tubercles in distribution pipelines. J. Am. Water Works Assoc. 72 (11), 626–635. USEPA, 2009. Drinking Water Contamination Candidate List 3 Final Notice. Vol. 74FR51850, pp. 51850–51862. Wang, L.P., 2007. Study of Iron Stability Problem and Pipes Corrosion in Drinking Water Distribution System. Master Dissertation, Tianjin University. World Health Organization, 2004. Guidelines for Drinking Water Quality Geneva, vol. 1. Xiao, L.P., Liu, W.Y., Lei, L., 2007. Feasibility of fenton oxidation method in removing iron and manganese in acid mine water. J. Liaoning Tech. Univ. 26, 272–274. Yang, Fan, Shi, Baoyou, Junnong, Gu, Wang, Dongsheng, Yang, Min, 2012. Morphological and physicochemical characteristics of iron corrosion scales formed under different water source histories in a drinking water distribution system. Water Res. 46 (16), 5423–5433. Zeng, G., Liang, J., Guo, S., Shi, L., Xiang, L., Li, X., Du, C., 2009. Spatial analysis of human health risk associated with ingesting manganese in Huangxing Town, Middle China. Chemosphere 77 (3), 368–375. Zhu, Y., Guo, X., 2008. Underscales localized corrosion behavior of carbon steel in neutral solution. J. Chin. Soc. Corros. Protect. 5, 003. Zhuang, Y.W., 2013. Study on Leakage Characteristics of Water Distribution Network Using Blind Source Separation Theory. Master Dissertation, Harbin Institute of Technology.
