Effects of Household Storage and Plumbing Systems on the Levels of Trace Elements in Desalinated Drinking Water in Kuwait ()
1. Introduction
Of all of the countries of the world, particularly the Arab Gulf countries, Kuwait has the most limited fresh-water resources and the lowest levels of renewable fresh water [1]. The desalination of sea water provides the only source of drinking water in Kuwait, and the amount of desalination required increases as the demand for potable water increases. To produce distilled water, multi-stage flash (MSF) distillation is used by desalination plants in Kuwait. The distilled water produced is saturated with calcium carbonate and remineralized by blending it with 5% - 10% brackish water to make it potable. Finally, the remineralized water is chlorinated and pumped to a distribution network. The remineralization process prevents or limits corrosion within the distribution network’s pipes and alters the trace elements (TEs) present and the mineral composition in the desalinated drinking water relative to that produced from surface and ground freshwater sources. In addition, the saturation of the remineralized water with calcium carbonate provides a stable passivation layer that further prevents the corrosion of transfer pipes [2-4]. However, the World Health Organization (WHO) has reported that cardiovascular disease can develop when remineralized desalinated drinking water is consumed [5].
A literature survey indicated that numerous studies have been performed worldwide to evaluate the TEs in drinking water produced from ground or surface water [4, 6-9] or through the thermal desalination of seawater [10- 14]. Desalinated household drinking water (HW) in Kuwait has been extensively studied to evaluate the levels of TEs [12,15], minerals [16], disinfection byproducts (DBPs) known as haloacetic acids [17,18] and halomethanes [19,21-23], and organic contaminants [24]. Moreover, Al-Fraij et al. reviewed the various sources of contamination of drinking water in Kuwait during the transportation to consumers [25] and concluded that the monitoring process is inadequate. Al-Mudahf and AbuShady [15] reported an increase in the levels of some TEs in HW at the consumption point compared with the results of a previous study by Al-Fraij et al. [12] on samples collected at the reservoirs at desalination plants before pumping to the distribution network. There is a worldwide interest in studying the effects of the types and compositions of distribution systems and household utilities on the contamination of drinking water by TEs through the leaching and migration of corrosion products [26-34]. The major sources of indoor TE pollution include brass faucets and fixtures and household plumbing lines and their solders [28,30,32]. Viraraghavan et al. [32] reviewed the impact of the migration of some metals from household plumbing systems on the quality of drinking water. Alam and Sadiq [33] reported an increase in the levels of copper, iron, and zinc in desalinated drinking water in Dhahran, Saudi Arabia, mediated by the transportation to the consumers. Recently, Veschetti et al. [26] studied the migration of trace metals from distribution networks into Italian drinking water and related the high levels of Ni outliers to tap materials, Fe to corrosion processes, and Pb to lead pipes that still existed in old buildings.
In Kuwait, consumers drink water that has passed from the outdoor point of the public distribution system to their taps via their private household plumbing, piping, and storage utilities. The TE content and quality of this water may have changed during its residence and transport through these utilities. The main objectives of this study were 1) to determine the contamination of HW at indoor consumption points by TEs as a result of leaching from the metallic materials of household utilities; 2) to explore the impacts of various household storage and plumbing systems on the levels of TEs and the quality of HW at consumption points; 3) to correlate the variation in TE content to water composition; and 4) to study seasonal variations in TE levels.
2. Materials and Methods
2.1. Reagents and Reference Materials
For all of the required preparations, we used ultra-pure Type I ICP/MS-grade inorganic-free water obtained from a Millipore Ultrapure Water Purification System equipped with a Milli-Q element, a Q-Guard II pack, a Quantum VX cartridge, a 0.22-µm Millipak filter, and a <0.1-µm optimizer filter at the point of use (Millipore, Bedford, MA, USA). Certified high-purity chemical reagents, individual and calibration standard mixes, and quality control and reference materials were obtained from Agilent Technologies (Santa Clara, CA, USA), AccuStandard (New Haven, CT, USA), SPEX CertiPrep, Inc. (Metuchen, NJ, USA), Alfa Aesar (Ward Hill, MA, USA), and Merck (Nottingham, UK).
2.2. Desalination Plants and Sampling
According to official information from the Ministry of Electricity and Water in Kuwait (MEW), five dual-purpose power and desalination plants located along the Kuwaiti coast used MSF distillation to produce a total of ~310 million imperial gallons per day (MIGD) of distillate (yielding 331.5 MIGD of potable water) during the course of this study. To prevent marine fouling inside the distillers, the seawater supply at each plant (a total of ~18.7 mm3·d−1 for the five plants) was chlorinated (~6.0 mg·L-1 total residual chlorine); this seawater was used to cool and feed the MSF distillation units. The pH of the distilled water that was produced was adjusted to 8.3 by carbonation, and the water was blended with 5% - 10% brackish water followed by chlorination to a total residual chlorine value of ~1.2 mg∙L−1. The chlorinated water was pumped first to storage facilities (towers and concrete underground reservoirs) and then to the distribution system network and ultimately to consumer storage facilities. The Az-Zoor plant (with a mean capacity of 115.2 MIGD) completely supplies the Umm Al-Haiman (DUH) location. Together with the Shuaiba plant (36 MIGD), the Az-Zoor plant supplies all of the locations in the Ahmadi Governorate (GAH) and the Mubarak Al-Kabeer Governorate (GMK) with various capacity ratios. The Doha West (110.4 MIGD) and Doha East (50.4 MIGD) plants primarily supply the Shuwaikh Education (CSE) location and all of the residential areas of the Jahra Governorate (GJA). The Shuwaikh (19.5 MIGD) plant primarily supplies the Shuwaikh Sakani (CSW) location. All of the other individual locations in the residential areas of the Capital (GCA), Farwaniya (GFA), and Hawalli (GHA) governorates are supplied by all of the plants at varying ratios depending on temporal demand. The main distribution piping network was constructed from asbestos/ cement or from galvanized iron pipes, and most of this piping was recently replaced with ductile iron pipes. Figure 1 shows both the locations of the desalination plants and the distribution of the sampling points within the locations in the various governorates. Three replicates of each sample were collected and preserved according to US-EPA Methods 300.1 and 200.8 [35]. The samples were delivered to the laboratory in boxes that were cooled with dry ice and were refrigerated at 4˚C until analysis. All of the water samples were analyzed within the recommended holding time [35] or discarded.
Figure 1. Locations of the desalination plants and the distributions of sampling points within the different neighborhoods of the governorates of Kuwait.
2.3. Determination of Analyte Concentrations
The temperature, pH, electrical conductivity (EC), total dissolved solids (TDS), and residual chlorine (Cl) were measured on-site at the time of collection. Inorganic anion concentrations were determined according to USEPA 300.1 method [35] using a Waters gradient HPLC system (Waters, Milford, MA, USA) equipped with a Waters 432 conductivity detector, an Altech 1000 HP Autosuppressor, and a Waters 2690 Separations Module. The levels of 25 TEs, i.e., aluminum (Al), arsenic (As), boron (B), barium (Ba), beryllium (Be), cadmium (Cd), calcium (Ca), chromium (Cr), cobalt (Co), copper(Cu), iron (Fe), lead (Pb), magnesium (Mg), manganese (Mn), mercury(Hg), molybdinum (Mo), nickel (Ni), potassium
(K), antimony (Sb), selenium (Se), sodium (Na), strontium (Sr), titanium (Ti), vanadium (V), and zinc (Zn), were measured according to US-EPA Method 200.8 [35] with an Agilent ICP/MS 7500ce (Agilent, Palo Alto, CA, USA) that was equipped with an octopole reaction system, a micromist nebulizer, a bonnet and shield, and an integrated autosampler. Details regarding the analytical procedures, quality control (QC), quality assurance (QA), and detection limits (DLs) have been reported elsewhere [15,16].
3. Results and Discussion
HW samples were collected from private residences (houses and apartments) and from offices and public buildings after flushing for five minutes. Ninety-nine locations were sampled in 69 neighborhoods in the six governorates of Kuwait. Together, these locations cover more than 95% of the residential areas in Kuwait. At each sampling collection, two samples were collected from two points in each selected building on each occasion: one from the water connection at the point of the distribution network just outside the building (outdoorHW; designated as O-HW) and the second from an inside faucet (tap) after the drinking water had passed through the storage and plumbing facilities as well as the household piping system of the building (indoor-HW; designated as I-HW). At residential or public properties, the storage tanks are mainly roof-top tanks (6 - 12 m3) that are constructed of galvanized iron. These iron tanks are generally being replaced with fiberglass, although ground reservoirs constructed from the same materials (20 - 100 m3) are also used for large buildings. All of the available information on the household storage, piping, and other utilities in the studied premises is given in Table 1, as are the types of outdoor distribution network piping in each area.
Three hundred and twenty-one I-HW samples and the same number of O-HW samples were collected simultaneously between December 2003 and May 2005 with different sampling frequencies (ranging from 1 to 20 sampling events) from 16, 38, 19, 11, 9, and 6 different sampling locations within the residential areas of the GAH, GCA, GFA, GHA, GJA, and GMK governorates, respectively (Figure 1). To achieve the objectives of the study, 12 subareas of higher population density within the 99 sampling locations in the 69 neighborhoods of the governorates were identified and investigated using a higher sampling frequency. Accordingly, 6 - 20 samples were periodically collected from each of the Kifan (CKF), Qibla (CQB), Shuwaikh education (CSE), Shuwaikh Health (CSH), Sulaibekhat (CSL), Sorra (CSO), CSW, DUH, Ardiya (FAR), Riggae (FRI), Salmiya (HSL), and Oyoon (JOY) locations, as indicated in Figure 1. At each of the remaining 87 locations, 1 - 3 samples were collected.
The corrosion of construction products that contain metallic materials and contact drinking water (CP-DW) during its production and transportation via the general distribution network and through private household utilities represents a major health concern because of the
Table 1. Types of household storage, piping, and other utilities at various sampling points.
leaching of the contaminants Cd, Cr, Pb, and Ni, which pose serious health threats to the consumers of drinking water [36]. Additionally, as a consequence of corrosion, other elements, including Al, Cu, Fe, and Zn, may leach into the drinking water. To protect the health of consumers of drinking water, the European Directive 98/83 on drinking water quality [37] established rules for monitoring these elements in CP-DW in the European Union. The interactions between the metallic materials of the CP-DW and water may be influenced by several factors that include the surface properties, metallic compositions, and lengths of the pipes as well as their surface/volume ratios, hydraulic conditions, network configurations, water residence times, and the physico-chemical characteristics of the bulk water [38,39]. The most corrosive waters are characterized by low pH levels, high carbon dioxide content, and low mineral content [40]. Each of these factors may affect the leaching of metals differently. In addition, the levels of TEs in the O-HW samples may be influenced, as previously desrcibed [16,17,22,24], by the operational processes that are used at the desalination plants, the nature of the intake seawater, the positions of the plants on the coast of Kuwait, and the nature and type of the distribution networks. In addition to these factors, the type of household storage and piping and plumbing facilities greatly affects the levels of some TEs at the consumer point (I-HW) [26,32,33]. It is also important to note that the sole source of HW in Kuwait is distilled water blended with brackish water. Thus, the differences found in the levels of TEs between O-HW and I-HW samples in various locations and governorates were also likely related to subtle variations in the characteristics of the brackish waters and the blend ratios used during the production process.
The 25 TEs analyzed in the 321 samples of O-HW collected in this study were Al, As, B, Ba, Be, Ca, Cd, Cr, Co, Cu, Fe, Pb, Mg, Mn, Hg, Mo, Ni, K Sb, Se, Na, Sr, Ti, V, and Zn. In addition, we measured the water quality parameters (WQPs) pH, conductivity, TDS, and residual Cl as well as the concentrations of the anions Cl− and
. The levels of these TEs and the WQPs of the IHW as well as their variations among sampling locations and governorates have been presented and discussed previously [15,16]. Table 2 depicts the mean values of the TEs in the O-HW samples from various sampling locations and governorates of Kuwait, and the values of the I-HW samples are included for comparison.
The mean concentrations of Al, Cr, Co, Cu, Fe, Pb, Ni, and Zn (designated as the 8-TE group) were higher in the I-HW samples than in the O-HW samples collected at all sampling locations. However, Be was not detected in any of the samples, and there were no significant differences among the mean values of the remaining 16 TEs in the I-HW samples and those in the O-HW samples at various sampling locations (Table 2). The increases in the mean values of the concentrations of the elements in the 8-TE group in the I-HW samples compared with those in the O-HW samples were attributed mainly to leaching from household utilities and showed significant variations among sampling locations and among elements. Figure 2 illustrates the variations in the increases of the elements in the 8-TE group in all of the I-HW samples (321) collected from the 99 sampling locations in the six governorates in relation to those in the O-HW samples. The highest increases were found for Fe (135%) and Zn (123%), followed by Pb (69%), Co (58%), Cu (42%), Cr (31%), and Al (30%), and Ni exhibited the lowest increase (19%). There were wide variations in the increases for each of these TEs among the sampling locations and among the governorates, as shown in Figure 3.
Figure 3(a) shows that the highest and lowest increases of Al occurred in the I-HW samples of CQB (198%) and CKF (3%) among the locations and in the GCA (46%) and GMK (9%) samples, respectively, among the governorates. Table 2 shows that the highest mean Al values in the I-HW and O-HW samples (34.4 µg∙L−1 and 33.5 µg∙L−1, with ranges of 21.2 - 53.0 and 18.0 - 53.3 µg∙L−1, respectively) occurred at the CKF location, which had the lowest increase of Al (3%). By contrast, the CQB location, which had the highest increase of Al (198%), had significantly lower levels in the I-HW and O-HW samples (mean values of 13.4 and 4.5 µg∙L−1 and ranges of 1.0 - 54.4 and 0.3 - 10.0 µg∙L−1, respectively). Thus, the migration of Al from household utilities and piping was the only source of Al increase in the I-HW samples collected at the CQB location, which was the location with the highest Al increase. In contrast, no Al leaching from household utilities and piping occurred at the CKF location, despite its high mean I-HW value (34.4 µg∙L−1), which was slightly higher than the