By Anat Kartaginer

The accepted use of sodium (Na+), calcium (Ca++) and magnesium (Mg++) phosphate salts in various water treatment processes in scale formation inhibition is considered safe from a health point of view. Since the phosphates used in this application have GRAS status (Generally Recognized As Safe) in the United States, this is an accepted treatment in many Western countries, with the maximum permitted phosphate concentrations varying according to agency as follows:

  • NSF 601: 10 ppm as PO4
  • NSF 422: 10 ppm as PO4
  • FDA/USDA3: 10 ppm as P

These phosphate salts are also prevalent in the food industry for applications such as chelating or sequestering agents, as emulsifiers and as important dietary supplement sources of calcium, magnesium and phosphorus.4

The problem of scale formation is both significant and troublesome in domestic and industrial systems such as potable water supplies, steam generating and water cooling systems.

The mechanism of scale formation includes the following stages:5

  • Attainment of super saturation: Super saturation conditions are achieved when a solution is concentrated beyond the solubility limits of one or more of its constituents. Most of the frequently encountered scaling salts—CaCO3, Mg (OH)2 , CaSO4, exhibit inverse solubility characteristic; i.e., solubility decreases with increasing temperature.
  • Nucleation: Attainment of super saturation alone is not sufficient for precipitation to begin. At relatively moderate super saturation, the solution may enter an induction period, without precipitation. To create a solid phase, an energy barrier has to be overcome. This barrier is particularly important in the early stages of the nucleation process, when only a few molecules or ions in the solution gather to form clusters.
  • Scale layer growth: Subsequent growth of a crystal layer involves several processes as follows: (1) diffusional transport of the crystal forming ions towards the crystallizing layer; (2) incorporation of the ions on growth sites of the crystal lattice; (3) adhesion and removal processes.

The adsorption of anti-scalants on the crystallization surface acts to delay nucleation, to reduce the precipitation rate and distort the crystal structure such that the tenacity of the deposit is weakened. Industrial experience exhibits various models of scale layer growth such as asymptotic, falling rate and linear growth. One study has shown that a dosage of 0.5 ppm of polyphosphate anti-scalant sodium hexametaphosphate (SHMP) results in an increased induction period and the suppression of further layer growth after asymptotic thickness is reached (Figure 1).

Anti-scalant dosage is one of the most effective and widely accepted means for scale suppression. Commonly used anti-scalants are generally derived from three chemical families including condensed polyphosphate, organophosphate and polyelectrolytes.

Polyphosphate can display an effective threshold effect as well as corrosion protection properties.

This paper examines the addition of SHMP (NaPO3)6 for the inhibition of scale formation in domestic filtration and purification systems of both the T4 and T6 types, from which water is produced for either drinking and heating or which supply cooled, room temperature and hot water, respectively.

Materials and methods/T4 unit
The first unit to be assessed was a T4 type water filtration and purification unit incorporating an extruded activated carbon filter and a UV lamp (Figure 2). This unit supplies drinking water at room temperature and it was assessed for the effect of the addition of poly-phosphate on the accumulation of scale in a kettle and for the maximum phosphate concentration released and its comparison to various standards.
Testing was conducted under the following initial conditions in the laboratory:

  • Flow 1.8 L/min (0.5 gpm)
  • Water pressure 6 bars (90 psi)
  • Water quality:
    — Ca hardness as CaCO3 69 mg/L
    — Alkalinity as CaCO3 222 mg/L
    — Total dissolved solids 490 mg/L
    — pH 7.65
    — Temperature 26.5°C (80°F)

Four stainless steel kettles possessing concealed heating elements were boiled repeatedly over a period of three weeks. To assess the efficiency of the phosphate against the control group, the boiling was initiated so that in two kettles (#1 and #2), the water from a T4 unit containing 10 g of polyphosphate incorporated in the carbon filter was boiled; while another two kettles (#3 and #4) acted as the control group, boiling untreated tap water.

The water was boiled in the four kettles simultaneously using the same water source. The amount of water boiled in each session was 1.5 L (0.4 gallons). A total of 173 kettle-boiling sessions were conducted. The kettles were weighed twice. Prior to each weighing, the kettles were dried on an electrical heating platter (Fried model MH-4 No. 8696). The first weighing was performed following the accumulated boiling of 111 L (29 gallons/74 sessions) and the second was performed after the boiling of 295.5 L (78 gallons/173 sessions). The weighing was performed using an electronic scale with a maximum capacity of 8100 ± 0.05 g (17.85+ lbs.).

Additionally, after exposures of six, 12, 48 and 72 hours the release of phosphate was assessed. These measurements were performed by an Israeli Ministry of Health-approved laboratory (Bactochem, Nes Tsiona). The phosphorus tests were performed using a Vista ICP-AES with a limit of detection of 0.02 mg/L as P. To reach the highest potential phosphate release from the polyphosphate incorporated in the hollow annulus of the filter’s circular cylinder, the water was taken from the filter at each stage prior to the sampling; 150 mL of water were introduced in order to flush the water from the UV chamber. Between the various time durations, 300 mL of water were flushed through the system to remove residue(s) between test sessions.

Materials and methods/T6 unit
The second unit assessed was a T6 type filtration and purification unit incorporating an activated carbon filter and a UV lamp, supplying cold, room temperature and hot water (Figure 3).

For the tests, eight T6 units were installed in the laboratory and were evaluated to determine:

  • The effect of the phosphate concentrations released from various initial weights of polyphosphate. All eight units were fitted with filters incorporating polyphosphate in the hollow annulus of the filter’s circular cylinder. The units were left for the first week of the installation without extracting water for a period of 60 hours (simulating no water usage by the consumer over the weekend). Following this, samples were taken. For continual testing of drip phenomena cases at the end of the first stage, the filters were taken out of units 1 and 2 and were replaced with filters not containing polyphosphate.
  • The influence of the addition of polyphosphates at various initial weights on the possible appearance of drips that may appear from the hot water outlet tube during the heating period and without water being taken by the user (Figure 4). 2,000 L (528 gallons) of water were passed through each unit which was visually assessed for the existence or lack of drips. When drips occur, they are collected in the drip receiver.

Drip phenomena
Drip occurrence conditions and prevention were monitored. Hot water tank (7) heating leads to water expanding, eventually reaching the accumulation tank (10). When scale formation is inhibited, dripping is prevented from the outlet tube (3) as opposed to scale growth which would affect the vacuum area and the Venturi effect. Hence, if scale formation is not inhibited, drip phenomena may occur under the following conditions:

  • When the accumulation tank is filled with water.
  • When air bubbles mixed with water from the accumulation tank exit the outlet tube when the water is boiling.

Phosphate release over filter lifespan
Samples were taken at intervals of 500 L (132 gallons) during a total flow of 2,000 L (528 gallons). The phosphate concentrations were determined via a photometer unit (Spectroquant—Nova 60; kit model 1.14848; detection limit 0.03 mg/L PO4).

The maximum phosphate concentrations released after differing durations of exposure and their comparison to different standards were performed. This was conducted at durations of six, 12, 48 and 72 hours. P was measured at the laboratory (using the Vista ICP-AES). Test conditions of the units were a system flow rate of 1.3 L/min. (0.35 gpm) and water pressure of 6 bars (90 psi).

The filter lifespan in the T6 unit (2000 L 528 gallons) is based on a typical consumption of 10 L/d (2.6 gpd). Hence, 300 L/month (79.3 gallons/month) and 1,800 L (476 gallons) for six months. The polyphosphate assessed in both units was Crystal Glassy Phosphate 6R Micromet, solubility 3-5 percent (as percent by weight per month). See Figure 5.

Results/T4 unit
Regarding the effect of addition of polyphosphate on deposition of scale in kettles, as shown in Figure 6, after 173 kettle-boiling sessions or 260 L (69 gallons) of water, equivalent to six weeks’ domestic use (based on an average of four boiling sessions per day), a scale reduction of some 70 percent was observed in the kettles receiving treatment.

As to the assessment of maximum phosphate concentration released and comparison to various standards, (as seen in Table 1) after all time intervals, the phosphate levels in the water did not exceed maximum permitted levels.

Results/T6 unit
Phosphate release from various initial weights and temperatures from a fresh filter was assessed following the completion of the installation of units 1-8 and continued for four days.

As seen in Figure 7, the polyphos-phate solubility increases (consistent with the manufacturer’s specification that at temperatures higher than 100°F (37°C), there is an increase of polyphos-phate solubility)7. Notwithstanding this, the phosphorus solubilized from different initial weights was negligible. In hot water, an average concentration of 1.1 mg/L PO4 was observed (from 10-16 g of polyphosphate), versus 1.2 mg/L PO4 (from 24-30 g of polyphosphate), while in cold water an average concentration of 0.37 mg/L PO4 was observed (from 10-16 g of polyphosphate) in comparison to 0.42 mg/L PO4 (from 24-30 g of polyphos-phate).

The effect of the addition of polyphosphate on the drip phenomena, (i.e., prevention of drips liable to appear owing to the accumulation of scale on the outlet tube) was observed. This experiment was executed over a period of one month during which 100 L (26 gallons) of water were extracted each day during the week with an intervening weekend break. As exhibited in Table 2, the control group (units 1 and 2), which incorporated non-polyphosphate filters, exhibited the appearance of drips after the passing of some 800 L (211 gallons) of water. An addition of 10-16 g or 24-30 g of polyphosphate to the filters (units 3-8) prevented this phenomenon. In fact, no drip phenomena were observed even following the passage of 2,000 L (528 gallons) of water.

In view of the reduction of released phosphate during the life of the filter, over the course of tests performed for assessing the effect of adding polyphosphate for the prevention of drips, which included the passing of 2,000 L (528 gallons) of water through the unit, phosphate concentrations were observed at intervals of 500 L (132 gallons) of hot water added. As exhibited in Figure 8, over the lifespan of the filter, a reduction of some 73 percent in the phosphate concentration (as PO4) was observed.

To determine maximum phosphate concentrations released from T6 units (compared to various standards) after ascertaining that an initial weight of 10-16 g of polyphosphate were sufficient to prevent drips over the lifespan of the filter, a T6 unit was tested by an external laboratory to assess the phosphate concentrations after exposures of six, 12, 48 and 72 hours. As exhibited in Table 3, after all time durations, the maximum concentrations were within the FDA, NSF and USDA standard parameters.

This study was performed to examine the effect of the use of polyphosphate on two factors in two domestic filtration and purification systems: (a) prevention of the deposition of scale in kettles boiled with water taken from the T4 type and (b) prevention of scale formation on the outlet tube in the T6 unit. In addition, the release of phosphate concentrations from various initial weights, at different temperatures, along with the reduction of the phosphorus concentration over the lifespan of the filter, was measured.

Polyphosphate prevented the deposition of some 70 percent of scale in kettles in comparison with the control group. The more water boiled, the greater the difference in the quantity of scale deposited. This difference apparently stems from the crystallization stage. In kettles with untreated water there was a stage of scale growth whereas the treatment with polyphosphate delayed crystallization at an early stage. In the untreated T4 units, the color of the accumulated scale was yellowish, compared to the whitish hue observed in the treated kettles. The difference in the color likely occurred because of the formation of iron and magnesium phosphates in the treated water, where presence of the polyphosphate prevented the formation of iron and magnesium sediments. At all time, in both the T4 and T6 units, the released phosphate concentrations remained within the three American standards. The highest released concentration after 12 hours following installation of the T4 units may stem from the presence of a new granular layer of the material (see Table 1).

As illustrated in Figure 7, the polyphosphate dissolution increases at higher temperatures (consistent with the manufacturer’s specification that at temperatures higher than 100°F (37°C), there is an increase in the polyphosphate solubility). The release of phosphate increases when the initial polyphosphate weight is higher, although the difference is not significant and the lower initial weight (10-16 g) was sufficient to prevent drips from the hot water outlet tube of the units.

Regarding the effect of the addition of polyphosphate on drip phenomena, as shown in Table 2, the control group (units 1 and 2), which incorporated non-polyphosphate filters, exhibited the appearance of drips after the passing of 800 L (211 gallons) of water whereas the addition of 10-16 g or 24-30 g of polyphosphate to the filters (units 3-8) prevented this phenomenon.

As seen in Figure 8, a reduction of some 73 percent of released phosphate concentration (as PO4) was observed, which emphasizes the importance of timely filter replacement.

The addition of 10 g of polyphos-phate to the T4 units, resulting in the release of 0.44-1.58 mg/L phosphate as PO4, was found to be effective in the prevention of scale crystallization while keeping within the maximum phosphate concentrations permitted by the NSF, FDA and USDA standards.

An addition of 10-16 g of polyphos-phate to the T6 units, releasing phosphate concentrations between 0.25-1.1 mg/L as PO4, was sufficient to prevent the appearance of drips from the hot water outlet pipe. These drips were observed after a flow of only 800 L (211 gallons) through the control group, but were completely eliminated after the passage of 2,000 L (528 gallons) through the units incorporating polyphosphate in the filters.

The differences in the phosphates solubilized from different initial weights of 10-16 g and 24-30 g were negligible in both hot and cold water. Phosphate releases in hot water averaged 1.1 mg/L and 1.28 mg/L for the two sets of initial weights, respectively and for cold water, these concentrations averaged 0.37 mg/L and 0.42 mg/L respectively.

Over the lifespan of the filter, a reduction of some 73 percent in the released phosphate concentration in hot water was observed. For the long-term prevention of drips, timely replacement of the filters is essential.

According to the polyphosphate supplier, up to 5 mg/L PO4 is required to prevent the deposition of scale. Concentrations of 1.58 mg/L for the T4 units and 1.1 mg/L for the T6 units were sufficient to prevent deposition of scale for the specified needs and under the conditions tested.

The author would like to thank Professors Avner Adin of the Department of Soil and Water Sciences at the Hebrew University of Israel and Ronald Gehr of the Department of Civil Engineering at McGill University in Montreal, Quebec, Canada, for their assistance in writing this paper, as well as Danny Targan, Chief Executive Officer and Eldad Maziel, Chief Technology Officer at Tana Water, for their encouragement and support.


1. NSF/ANSI Standard 60 Drinking Water Treatments Chemical- Health effects- Table 2, Chemical for corrosion & scale control-1996a.
2. NSF/ANSI Standard 42 Drinking Water Treatment Units Aesthetic effects- Table 16, Additives intended for scale control, April 4, 2005.
3. Babyak, M. Regulatory Status of Micromet Products, USDA, May 5, 1984.
4. Yannai, H., Expert opinion on the use of sodium (Na+), calcium (Ca++) and magnesium (Mg++) phosphate salts (PO4-3) in drinking water. Haifa, Israel, June 26, 2005. (Toxicologist opinion.)
5. Hasson D., State of the art of scale control in saline and wastewater desalination GWRI Rabin Desalination Research Laboratory, Department of Chemical Engineering. Haifa, Israel
6. Hasson D., Semiat R., Bramson D., Busch M and Limoni-Relis B., Suppression of CaCO3 Scale Deposition by Anti-scalant, Desalination, 118(1998) 285-296.
7. Nu-Calgon, Product Bulletin, Facts about Micromet Treatment. Altom Court, St. Louis, Mo.

About the company
Tana Water has been a major force in the provision of drinking water systems for over 30 years and is one of the world’s most advanced developers and manufacturers of point of use systems. Leading-edge products are matched by a focus on customer service, which comes from long-lasting partnerships with clients.

About the author
Anat Kartaginer, head of the Water Treatment Department at Tana Water of Emek Haela, Israel, manages microbiological, chemical, hydraulic and physical testing for water treatment technology. She is a member of the Israeli Water Association (IWA), International Ultraviolet Association (IUVA) and European Point-Of-Use Drinking Water Association (EPDWA). Kartaginer holds a Bachelor of Science in agronomy and is completing a Master of Science in environmental science from the Hebrew University of Jerusalem. She can be reached at Tana Water, Kibbutz Netiv Ha Lamed Hei, DN Haela 99855, Israel; via telephone +972 29900222; fax +972 29900500; e-mail [email protected] or through the company’s website,



Comments are closed.