Hard water can be softened by atomization

Abstract

Hard water is softened because its use in everyday life can cause various problems in the living environment. However, existing water-softening methods have a number of drawbacks. Here we examined a new water-softening method involving atomization. Three concentrations of calcium-based synthetic hard water and a commercial natural hard water were prepared. Each hard water was sprayed into the atmosphere at 1 MPa from an atomization nozzle and then collected. Part of the collected water was recirculated, and the atomization treatment was continued. After the treatment, a white precipitate and fine bubbles were generated, the Ca²⁺ concentration and electrical conductivity decreased, and the pH increased. Then over time, the white precipitate increased slowly and the Ca²⁺ concentration decreased until finally it almost fell within the recommended range. This water-softening phenomenon was thought to be due to expansion of the gas-liquid interfacial area and generation of fine bubbles with atomization. This method is extremely simple and expected to show high versatility.

Introduction

The hardness of drinking water varies widely according to the water source and whether water-softening methods are effective. Water hardness depends on the weight of calcium and magnesium in the form of calcium carbonate (CaCO₃) and is represented as milligrams of CaCO₃ per liter, as shown in Eq. 1 (WHO, 2017).

  

Because magnesium is usually present in natural groundwater at low concentrations (from approximately 50 mg/L to (rarely) over 100 mg/L), calcium-based hardness usually predominates (National Research Council, 1977). According to the WHO classification, water containing CaCO₃ at concentrations below 60 mg/L is considered to be soft; 60–120 mg/L, moderately hard; 120–180 mg/L, hard; and more than 180 mg/L, very hard (WHO, 2010). Hard water use in everyday life can cause various problems in the living environment. For instance, a precipitate of calcium carbonate scale forms when hard water is heated. Hardness exceeding 200 mg/L CaCO₃ increases the amount of inorganic deposit in plumbing and consumption of soap through interaction of factors such as pH and alkalinity. Therefore water-softening treatment is often performed on hard water. On the other hand, a hardness of less than 100 mg/L CaCO₃ results in corrosion of plumbing because the buffering capacity is low (WHO, 2022). In the current situation, the association between calcium intake and the risk of prostate cancer is not clear, but worrying data have been presented. The formation of calcitriol from vitamin D is inhibited by excessive calcium intake, and the proliferation of prostatic cells may be promoted (Wiseman, 2008). The onset risk of metastatic prostate cancer with a calcium intake of more than 2,000 mg per day rises by nearly 5-fold as compared with a level of less than 500 mg (Giovannucci et al., 1998). However, it is unrealistic to perform water-softening treatment for all tap water in hard water areas because of the high cost. In some high-hardness areas, therefore, water may be treated during the water purification process to reduce hardness, or a water purification and hardness reduction device may be installed as a home water supply system (Lanz and Provins, 2016). Existing water-softening methods vary according to the type of device and the quality of the treated water (Park et al., 2007; Yeon et al., 2004). The crystallization method and the coagulation precipitation method require a pH adjustment as postprocessing with utilization of a strong alkali. For the reverse osmotic membrane (RO membrane) method, it is necessary to drain 2–3 times concentration water than the volume of permeated water at the same time to prevent precipitation of the hardness ingredient. Nanofiltration and electrodialysis employed in the membrane filtration method use a wide range of devices and electronic equipment, so require specialized maintenance of the membrane component by the manufacturers. Because the ion exchange method requires highly concentrated sodium chloride solution for resin regeneration, the wastewater generated is problematic (Park et al., 2007). Because electric regeneration-type desalters (EDRs) are unsuitable for treatment of high-hardness water, it is often combined with RO membrane (Yeon et al., 2004). As mentioned above, there are many problems with existing water-softening methods, and thus any improvement and/or new technical development would be welcome. We have hypothesized that if the gas-liquid interfacial area of hard water could be maximized, the oversaturated CO₂ in hard water volatilized, and the precipitation of CaCO₃ promoted, then water softening could be achieved. Here we report a new extremely simple water-softening method involving atomization.

Materials and Methods

Materials　Three concentrations of synthetic hard water were prepared using calcium chloride (CaCl₂) and sodium bicarbonate (NaHCO₃) (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) dissolved in distilled water as follows (Spiro and Jaganyi, 1994): 200 mg/L CaCO₃ (80 mg/L Ca), 375 mg/L CaCO₃ (150 mg/L Ca), and 550 mg/L CaCO₃ (220 mg/L Ca). As natural water, a commercial hard water, “Evian” (Danone S.A., Paris, France), was purchased. The test water samples were stored at 4 °C or 20 °C in glass or polyethylene terephthalate (PET) bottles, which were sealed until use.

Atomization treatment　Schematic diagram of hard water atomization is shown in Fig. 1. The nozzle used for atomization was an “extremely fine mist and ultra-small capacity hollow cone spray nozzle” (1/4M KB 80056 S303-RW, H.IKEUCHI Co., Ltd., Osaka, Japan). According to the product catalogue, the spray angle of the nozzle was 80 degrees within the range 0.7–2 MPa of atomization pressure, and the average diameter of the atomized water droplets was 65–110 μm. Just before the test, 200 mL of each hard water sample was added to a 300-mL beaker. The beaker and the nozzle were then connected by a high purity fluoroplastic tube. A pressurization pump (Flojet PS2011, Xylem Inc. WA, US) was installed in the middle of the tube. Aspirated hard water flowed through the tube under pressure (1 MPa) from the pump, and was sprayed as a mist into the beaker by the nozzle from a height of 30 mm. The atomized water collected in the beaker was recirculated sequentially atomization or stored as a sample for measurements. During atomization, the hard water in the beaker was held at approximately 4°C or 20°C. The time required to atomize each 200 mL sample of hard water (i.e., from the start to the end of atomization) was 47 s. One pass was therefore assumed to be 200 mL/47s and the following atomization time (At) sets were performed: 1 pass (47 s), At 15 min, At 30 min, At 60 min, and At 120 min (only for 550 mg/L CaCO₃). The atomization treatment was not performed for original water. After atomization treatment, each sample of hard water was allowed to stand for 72 h at respective temperatures. All tests were performed in triplicate.

Fig. 1.

Schematic diagram of hard water atomization.

Measurement of calcium ion concentration, electrical conductivity and pH　Analyzers manufactured by HORIBA Advanced Techno Co., Ltd. (Kyoto, Japan) were used to measure the calcium ion (Ca²⁺) concentration (LAQUAtwin-Ca-11), electrical conductivity (LAQUAtwin-EC-33) and pH (LAQUAtwin-pH-33). Each measurement was performed before and after atomization treatment. The measurements after atomization were performed at 0, 24, 48, and 72 h of standing time (St) within the 72-hour standing period.

Measurement of ultrafine bubble particle count and diameter　Ultrafine bubble (UFB) particle count and diameter were measured with a Zetasizer NanoSight NS500 device (Malvern Instruments Ltd., Worcs, UK) which employs a combination of dynamic light scattering and nanoparticle tracking analysis. As typical synthetic hard water, 375 mg/L CaCO₃ (150 mg/L Ca) synthetic hard water was measured before and after the atomization treatment.

Statistical analysis　Two-way ANOVA was used to analyze the main effect and the interaction of the atomization time and the standing time in the standing period (from 0 to 72 h). Data are reported as the mean ± standard error of the mean (SEM). The BellCurve for Excel (version 4.03) statistical software package was used for all statistical analyses.

Results

Synthetic hard water　The Ca²⁺ concentration, electrical conductivity, and pH of the synthetic hard water are shown in Figs. 2 and 3, and Tables 1 and 2. In all synthetic hard waters, the change in Ca²⁺ concentration was related to the change in electrical conductivity at both water temperatures (Tables 1 and 2). The higher the hardness of synthetic hard water, the more the decrease in Ca²⁺ concentration (Figs. 2A, 2D, 2G, 3A, 3D, and 3G) and electrical conductivity (Figs. 2B, 2E, 2H, 3B, 3E, and 3H) due to atomization treatment. For 200 mg/L CaCO₃ synthetic hard water at both temperatures, the influence of atomization time and standing time was very slight (Figs. 2A and 3A). On the other hand, in the synthetic hard water with 375 mg/L CaCO₃ and 550 mg/L CaCO₃, the decrease in Ca²⁺ concentration was evident immediately after the atomization treatment at both water temperatures. Furthermore, an increase in the atomization time resulted in an increased formation of white precipitate. The longer the standing time and the higher the temperature, the more conspicuous these phenomena became. Therefore, an interaction of atomization time with standing time on the electrical conductivity was more evident at 20 °C compared to 4 °C (Tables 1 and 2). The pH in the original water increased within 24 h at both water temperatures after opening the glass bottle (Figs. 2C, 2F, 2I, 3C, 3F, and 3I). In the synthetic hard water after the atomization treatment, the pH increased within 24 h at both water temperatures. The increased pH tended to return (i.e., decrease) to the original level with prolonged standing time, particularly when the atomization time was shorter. An interaction of atomization time with standing time on pH was particularly evident in the case of synthetic hard water with 550 mg/L CaCO₃, at both water temperatures (Tables 1 and 2). In the case of atomized synthetic hard water, the lowest limit of depressed Ca²⁺ concentration was within a range of approximately 70–100 mg/L Ca (hardness of 175–250 mg/L CaCO₃).

Fig. 2.

Relationship of atomization time and standing time in synthetic hard water (4°C). The values are presented as mean ± SEM (n = 3). A, B, C: 200 mg/L CaCO₃; D, E, F: 375 mg/L CaCO₃; G, H, I: 550 mg/L CaCO₃. Bef.: Before; Aft.; After; At: Atomization time; St: Standing time.

Fig. 3.

Relationship of atomization time and standing time in synthetic hard water (20°C). The values are presented as mean ± SEM (n = 3). A, B, C: 200 mg/L CaCO₃; D, E, F: 375 mg/L CaCO₃; G, H, I: 550 mg/L CaCO₃. Bef.: Before; Aft.; After; At: Atomization time; St: Standing time.

Table 1.Main effects and interaction of atomization time and standing time on calcium ion concentration, electrical conductivity, and pH in synthetic hard water (4°C).

	Statistical significance, p
	Atomization time main effect	Standing time main effect	At × St interaction
200 mg/L CaCO3
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	0.04	< 0.001	ns
pH	0.001	< 0.001	0.001
375 mg/L CaCO3
Ca2+ concentration	< 0.001	< 0.001	0.02
Electrical conductivity	< 0.001	< 0.001	ns
pH	< 0.001	< 0.001	ns
550 mg/L CaCO3
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	< 0.001	ns	0.01
pH	< 0.001	< 0.001	< 0.001

To assess the effects of atomization time and standing time, two-way ANOVA was employed (p < 0.05). ns, not significant (p ≥ 0.05). At: Atomization time; St: Standing time.

Table 2.Main effects and interaction of atomization time and standing time on calcium ion concentration, electrical conductivity, and pH in synthetic hard water (20°C).

	Statistical significance, p
	Atomization time main effect	Standing time main effect	At × St interaction
200 mg/L CaCO3
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	0.003	< 0.001	< 0.001
pH	< 0.001	< 0.001	< 0.001
375 mg/L CaCO3
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	< 0.001	< 0.001	0.002
pH	ns	0.01	ns
550 mg/L CaCO3
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	< 0.001	< 0.001	< 0.001
pH	< 0.001	< 0.001	< 0.001

To assess effects of atomization time and standing time, two-way ANOVA was employed (p < 0.05). ns, not significant (p ≥ 0.05). At: Atomization time; St: Standing time.

Natural hard water　The Ca²⁺ concentration, electrical conductivity, and pH of the natural hard water are shown in Fig. 4 and Table 3. The Ca²⁺ concentration in the natural water was 130 mg/L. Both the changes in the Ca²⁺ concentration and the electrical conductivity were evident immediately after the atomization treatment at both water temperatures. Standing time also had a marked effect at both water temperatures. Therefore, an interaction of atomization time and standing time was found at both water temperatures. The lowest limit of the depressed Ca²⁺ concentration resulting from atomization treatment was within the range of approximately 60–80 mg/L Ca (hardness 150–200 mg/L CaCO₃). In the original water, the pH increased within 24 h after opening the stored PET bottle. Similarly, in the case of atomized natural hard water, the pH increased within 24 h at both water temperatures. The increase in pH was maintained at approximately a plateau until at least 72 h at both water temperatures. There was an interaction of atomization time with standing time at both water temperatures.

Fig. 4.

Relationship of atomization time and standing time in the natural hard water “Evian”. The values are presented as mean ± SEM (n = 3). A, B, C: 4°C; D, E, F: 20°C. Bef.: Before; Aft.; After; At: Atomization time; St: Standing time.

Table 3.Main effects and interaction of atomization time and standing time on calcium ion concentration, electrical conductivity and pH in the natural hard water “Evian”.

	Statistical significance, p
	Atomization time main effect	Standing time main effect	At × St Interaction
4 °C
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	< 0.001	< 0.001	0.03
pH	< 0.001	ns	< 0.001
20 °C
Ca2+ concentration	< 0.001	< 0.001	< 0.001
Electrical conductivity	< 0.001	< 0.001	< 0.001
pH	< 0.001	0.005	< 0.001

To assess the effects of atomization time and standing time, two-way ANOVA was employed (p < 0.05). ns, not significant (p ≥ 0.05). At: Atomization time; St: Standing time.

UFBs　The particle count and diameter of UFBs produced by atomization treatment of the synthetic hard water (375 mg/L CaCO₃) are shown in Fig. 5. The particle count of UFBs increased at both water temperatures temporarily with increased atomization time. However, it decreased after 30 min at 4°C and after 60 min at 20 °C until the start level of atomization. Regardless of the water temperature and the atomization time, both the mean and median particle diameter were approximately 100 μm. Macroscopic observation demonstrated that the water collected in the beaker after atomization was conspicuously cloudy. This remarkable white cloudiness was already evident after only 1 pass (i.e., 47 s of At). After atomization treatment, the white cloudiness rose to the surface of the water within several minutes and then disappeared.

Fig. 5.

Effect of atomization time on UFB generation in synthetic hard water. The values are presented as mean ± SEM (n = 3). At: atomization treatment.

Discussion

The suggested reaction by which CaCO₃ is generated from calcium salts and hydrogencarbonate is shown in Eq. 2 (Hasson et al., 1997; Kim et al., 2002; Macadam and Parsons, 2004).

  

The Ca(HCO₃)₂ is generated in the middle of the reaction, and is decarboxylated by subsequent heating and/or boiling, and as a result, CaCO₃ is precipitated. The Ca(HCO₃)₂ is hydrogencarbonate provided only as an aqueous solution. At lower temperatures and higher carbon dioxide partial pressure, a highly concentrated solution of Ca(HCO₃)₂ is produced. Natural hard water such as the “Evian” used in the present study corresponds to this situation. The CaCO₃ may often precipitate at the bottom of a bottle of commercially available hard water.

As shown in Eq. 3, the pH of the water depends on carbon dioxide equilibrium. When Ca(HCO₃)₂ solution is heated, the molecular carbonate (H₂CO₃) and/or bicarbonate (HCO₃⁻) dissolved in it are pyrolized. Then, because the carbonate ion (CO₃²⁻) concentration increases, the reaction shown in Eq. 3 advances to the right, and the pH increases temporarily (Plummer and Busenberg, 1982). Also, precipitation of the CaCO₃ begins at around 85 °C, and the ratio of the CO₂ gas rises. As the CO₂ gas concentration in the fluid phase rises, the Eq. 3 reaction advances to the left, and thus the pH of the solution decreases, being approximately pH 7.4 at 100 °C (Plummer and Busenberg, 1982). In contrast, precipitation of the CaCO₃ is promoted when the CO₂ gas shifts to the gas phase.

  

After opening of the sealed bottle in which hard water had been stored, the pH of both the synthetic water and the natural water increased within 24 h. Then, a slight white precipitate was generated in the 550 mg/L CaCO₃ (220 mg/L Ca) synthetic hard water. This was because CO₃²⁻ dissolved in the water changed to CO₂ and shifted to the gas phase, thus advancing the Eq. 3 reaction to the right and increasing the pH of the aqueous solution (Stumm and Morgan, 1996). Although the elementary composition of the precipitate was not identified, it was estimated to be CaCO₃. The non-atomization treatment, the pH decreased after 24 h and tended to return to the original level. This was thought to be due to the carbon dioxide equilibrium between the aqueous and gas phases defined by Henry's law constant. In contrast, when the atomization treatment was performed, the white precipitate was produced immediately, and the pH of the test water increased. After 24 h, the increased pH tended to be maintained and the white precipitate slowly increased. The higher the hardness and the longer the atomization time, the more conspicuous these phenomena became. This suggests that the shift of CO₂ to the gas phase progressed continuously. Therefore, this water-softening phenomenon is considered to be biphasic. In other words, it is a rapid change occurring immediately after the atomization treatment, followed by a slow change. The former was thought to be due to gas-liquid interfacial area expansion of the hard water with the atomization that occurred momentarily in the air and a large quantity of microbubble (MB) generation, which will be mentioned later. On the other hand, the latter was thought to be due to CaCO₃ crystal formation explained as nucleus formation by accumulation of ions or molecules and subsequent growth, with UFB generation, as mentioned later.

In the present study, the pressure in the tube connecting the pressurization pump and the nozzle for atomization reached 1 MPa. The hyperbaric hard water became microparticulated immediately after passing through the atomization nozzle, and the gas-liquid interfacial area was maximized. At the same time, decompression to atmospheric pressure (approximately 0.1 MPa) occurred instantly. This sudden change was thought to promote deaeration of the oversaturated CO₂ gas. Furthermore, the sudden decompression caused cavitation, the physical process of which resembles boiling. In other words, when the topical pressure in the liquid flow decreased to less than saturated steam pressure momentarily, extremely small “bubble nuclei” smaller than 100 μm are boiled in a liquid and many very small “fine bubbles” (FBs) are generated (Brennen, 1995). The large quantity of UFBs produced by the atomization treatment is evidence of this. The FBs are defined as small bubbles with a diameter of < 100 μm, and can be further classified as microbubbles (1–100 μm) or ultrafine bubbles (< 1 μm) (ISO 20480–1, 2017). The UFBs cannot be confirmed visually because they are colorless and transparent, but the MBs are visible because they cloud the solution (Matsuki et al., 2012; Ma et al., 2015; Takahashi and Chiba, 2007). Also in the present study, remarkable white cloudiness was observed immediately after the start of the atomization treatment. Therefore, it is thought that not only the UFBs but also the MBs were generated in large quantities. The MBs are carried to the surface by the buoyancy and burst rapidly (Matsuki et al., 2012; Ma et al., 2015; Takahashi and Chiba, 2007). It was thought that the MBs were generated from some of the supersaturated CO₂ gas upon cavitation development and were rapidly released from the liquid surface into the gas phase. It is speculated that this caused the sudden decrease of Ca²⁺ concentration in the early stage of this water-softening phenomenon in combination with the gas-liquid interfacial area expansion of the hard water with atomization. On the other hand, UFBs cannot surface because they are strongly affected by Brownian motion (thermal motion) relative to buoyancy. Therefore, they can exist stably in water (Matsuki et al., 2012; Ma et al., 2015; Takahashi and Chiba, 2007). The internal pressure of UFBs soars in inverse proportion to the volume. The UFB interior can be at a very high pressure (10–500 MPa) and very high temperature (1 000–10 000 K) owing to highly localized energy density (Suslick, 1990). Therefore, some UFBs are known to collapse. The phenomenon whereby high energy is generated at the moment of UFB collapse occurs at every location through which a liquid flows and is known to be related to degradation of the instrument's performance and destruction of the material (Dular and Coutier-Delgosha, 2013). The UFBs generated by atomization treatment produce high energy upon collapse and induce dehydration and decarboxylation from Ca(HCO₃)₂ similarly to boiling. This might be associated with the slow precipitation of CaCO₃ after 24 h. However, we cannot draw definitive conclusions because many points regarding the generation and disappearance of cavitation bubbles remain unclear.

The European Union member states proposed a guideline about the concentrations of Ca and other minerals in drinking tap water in December, 2018 (Kozisek, 2020). According to this, the lower limit of the Ca concentration necessary for softened water is 30 mg/L Ca (hardness, 75 mg/L CaCO₃) and the recommended Ca concentration from the viewpoint of health risk reduction is 40–80 mg/L Ca (hardness, 100–200 mg/L CaCO₃). The Ca²+ concentration of water obtained using this method presented in this paper is approximately to the recommended range of the Ca²⁺ concentration mentioned above. Because this method can soften hard water to an appropriate hardness easily, it may have high versatility.

The present had several limitations. First, the gas-liquid interfacial area expansion of hard water and MB generation are thought to be involved in the initial stage of the water-softening phenomenon, but the theory remains a matter of speculation. Second, the energy emitted by cavitation bubble collapse might promote the dehydration of Ca(HCO₃)₂ and decarboxylation, but this is also a matter for speculation. Third, it was estimated that the white precipitate was CaCO₃, but its elemental composition, crystal state and crystal formation process were not identified. The mechanism of this water-softening method requires further detailed clarification in the future.

Conflict of interest　　KK and MK have no competing financial or non-financial interests, have not received any consulting fees/honoraria, have no leadership/advisory role in the company, and receive no patent royalties/licensing fees or other benefits (e.g., gifts). However, KK has a right to receive the patent royalties/licensing fee in the future. Because TY is an employee of TKS Co., Ltd., he has a financial competing interest, a non-financial competing interest, has received a salary, has a leadership/advisory role in the company, and has the right to receive the patent royalties/licensing fee in the future.

Author Contributions　　KK conceived and designed the study, collected and analyzed the data, performed project administration, and wrote and edited the manuscript. MK and TY contributed to the study design and data acquisition and analysis.

Funding　　This study was performed as part of a Joint Research Agreement between Gifu University and TKS Co., Ltd. (https://tks-gifu.jp/). The corresponding author (KK) received research funding through Gifu University, which was supported by TKS Co., Ltd.

References

•  Brennen,  C.E. (1995). CAVITATION AND BOILING. In “Cavitation and Bubble Dynamics.” ed. by Oxford University Press, NY., p.21.
•  Dular,  M. and  Coutier-Delgosha,  O. (2013). Thermodynamic effects during growth and collapse of a single cavitation bubble. J. Fluid Mech., 736, 44–66.
•  Giovannucci,  E.,  Rimm,  E.B.,  Wolk,  A.,  Ascherio,  A.,  Stampfer,  M.J.,  Colditz,  G.A., and  Willett,  W.C. (1998). Calcium and fructose intake in relation to risk of prostate cancer. Cancer Res. 58, 442–427.
•  Hasson,  D.,  Bramson,  D.,  Limoni-Relis,  B., and  Semiat,  R. (1997). Influence of the flow system on the inhibitory action of CaCO₃ scale prevention additives. Desalination 108, 67–79.
• ISO 20480–1:2017. Fine bubble technology — General principles for usage and measurement of fine bubbles — Part 1: Terminology. International Organization for Standardization, Geneva (2017).
•  Kim,  W.T.,  Bai,  C., and  Cho,  Y.I. (2002). A study of CaCO₃ fouling with a microscopic imaging technique. Int. J. Heat. Mass. Transfer. 45, 597–607.
•  Kozisek,  F. (2020). Regulations for calcium, magnesium or hardness in drinking water in the European Union member states. Regul. Toxicol. Pharmacol. 112, 104589.
•  Lanz,  B. and  Provins,  A. (2016). The demand for tap water quality: Survey evidence on water hardness and aesthetic quality. Water Resour. Econ. 16, 52–63.
•  Ma,  J.,  Xu,  C.S.,  Gao,  F.,  Chen,  M.,  Li,  F., and  Du,  L.F. (2015). Diagnostic and therapeutic research on ultrasound microbubble/nanobubble contrast agents. Mol. Med. Rep. 12, 4022–4028.
•  Macadam,  J. and  Parsons,  S.A. (2004). Calcium carbonate scale control, effect of material and inhibitors. Water Sci. Technol. 49(2), 153–159.
•  Matsuki,  N.,  Ichiba,  S.,  Ishikawa,  T.,  Nagano,  O.,  Takeda,  M.,  Ujike,  Y., and  Yamaguchi,  T. (2012). Blood oxygenation using microbubble suspensions. Eur. Biophys. J. 41, 571–578.
• National Research Council (1977). Drinking water and health. Washington, DC, National Academy of Sciences.
•  Park,  J.S.,  Song,  J.H.,  Yeon,  K.H., and  Moon,  S.H. (2007). Removal of hardness ions from tap water using electromembrane processes. Desalination 202, 1–8.
•  Plummer,  L.N. and  Busenberg,  E. (1982). The solubilities of calcite, aragonite and vaterite in CO₂-H₂O solutions between 0 and 90°C, and an evaluation of the aqueous model for the system CaCO₃-CO₂-H₂O. Geochim. Cosmochim. Acta. 46, 1011–1040.
•  Spiro,  M. and  Jaganyi,  D. (1994). Kinetics and equilibria of tea infusion. Part 10—The composition and structure of tea scum. Food Chem. 49, 351–357.
•  Stumm,  W. and  Morgan,  J.J. (1996). Chemical Equilibria and Rates in Natural Waters. In “Aquatic Chemistry.” 3rd ed. by John Wiley & Sons, NY, p. 150.
•  Suslick,  K.S. (1990). Sonochemistry. Science 247, 1439–1445.
•  Takahashi,  M.,  Chiba,  K., and  Li,  P. (2007). Free-radical generation from collapsing microbubbles in the absence of a dynamic stimulus. J. Phys. Chem. B. 111, 1343–1347.
• WHO (2010). Hardness in Drinking-water: Background document for development of WHO Guidelines for Drinking-water Quality. World Health Organization. https://apps.who.int/iris/handle/10665/70168
• WHO (2017). Guidelines for Drinking-water Quality: Fourth edition incorporating the first addendum. World Health Organization, Geneva; 2017.
• WHO (2022). Guidelines for Drinking-water Quality: Fourth edition incorporating the first and second addenda. World Health Organization, Geneva; 2022.
•  Wiseman,  M. (2008). The second World Cancer Research Fund/American Institute for Cancer Research expert report. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Proc. Nutr. Soc. 67, 253–256.
•  Yeon,  K.H.,  Song,  J.H.,  Kim,  J.B., and  Moon,  S.H. (2004). Preparation and characterization of UV-grafted ionexchange textiles in continuous electrodeionization. J. Chem. Technol. Biotechnol. 79, 1395–1404.