Elemental lead is a known toxic metal that can pose threats to human health and can be found in a variety of sources including lead paints, dust, municipal waste incineration, mining activities, gasoline additives and drinking water [1] -[4] . Lead is a known carcinogen which can lead to serious health problems such as: disruption of kidney function, interfering in synthesis of haemoglobin, interruption in immune system and miscarriages [5] [6] .

In particular, its impact on children’s neurobehavioral and intelligence development, brain cell function and learning abilities has been shown in numerous studies [7] -[11] . Although children’s average blood lead level (BLL) in many countries and areas such as United States, Mexico, Australia, Ecuador and Taiwan has generally been decreasing over the last three decades, lead poisoning in children is still of great concern [12] -[16] .

According to the drinking water quality guideline published by the World Health Organization (WHO), the recommended provisional limit of lead in drinking water is 10 µg/L (ppb). This limit is adopted as the drinking water standard in many countries including Canada, Australia, China and European Union, while in United States the guideline value is 15 µg/L [17] -[21] . Dissolution of lead in drinking water, also known as plumbosolvency, is mainly because of lead containing materials used in water distribution systems and household plumbing, which is influenced by various water quality characteristics. Existing corrosion scales in lead pipes, solders, brass fittings and plumbing fixtures used in drinking water distribution systems are currently considered as a major problem for municipalities [9] [22] [23] . Destabilization of the corrosion scale in the distribution systems is the main source of lead in drinking water. X-ray diffraction (XRD) results of pipes collected from distribution system in City of London, Ontario, Canada by Kim and Herrera [22] have shown the main lead solid phase in the corrosion scale is hydrocerussite (Pb₃(CO₃)₂(OH)₂₎. Additionally, in few cases cerussite (PbCO₃) was also observed. Lead oxides such as plattnerite (PbO₂) and minium (Pb₃O₄) can also be found in the inner layers of the corrosion scale. However, these species are more stable and tend to not dissolve [22] [24] . The corrosion scale can decrease the lead level in drinking water via a passivating layer, which can also trap toxic elements and decrease their concentration in drinking water as well.

Chemical properties of the water passing through the distribution system such as pH, alkalinity, chlorine content, oxidation reduction potential (ORP) and natural organic matters will affect the formation or destabilization of the corrosion scale. Temperature can also affect the stability of corrosion scales [1] [22] [23] . Numerous studies have shown the importance of the water pH in controlling the stability of the corrosion scale. For instance, at pH values less than 7.0, the corrosivity of water is very high and lead concentrations will increase significantly. However, pH effects can be mitigated by presence of alkalinity. In higher value of alkalinity, lead complexes formed with hydroxide and carbonate ions are insoluble and this greatly slows down the kinetics of lead dissolution [22] [23] . Free chlorine, HOCl and OCl⁻, a common disinfectant used in distribution systems, provides a high ORP that has been found to lead to the formation of insoluble solid phases such as PbO₂ and preventing the dissolution of lead species [20] [25] .

So far, researches have primarily focused on the effect of the pH at 20˚C (i.e. [1] [24] [26] ) and there is no research conducted at cold temperature (5˚C). Since there are countries such as Canada experiencing cold temperature about 6 months per year, it is important to investigate the influence of cold temperature on lead dissolution. Therefore, the objective of this paper is to explore the effects of different temperature conditions (20˚C vs. 5˚C) on dissolution of hydrocerussite and cerussite in drinking water at bench scale under various pH and alkalinity (moderate vs. low) conditions.

Based on studies showing that hydrocerussite (Pb₃(CO₃)₂(OH)₂) and cerussite (PbCO₃) are widely observed in the corrosion scale [22] [26] , batch dissolution experiments were conducted using pure hydrocerussite and cerussite solids to investigate the lead solubility under different pH values (7.0, 8.0 and 9.5) and temperature conditions (20˚C and 5˚C). An environmental chamber was utilized for temperature control of the 5˚C experiments. Trace element grade chemicals and plastic wares were utilized to avoid cross contamination. Batch dissolution experiments were started by adding 0.3 g of either hydrocerussite (Aldrich, 99.44%) or cerussite (Aldrich, 99.99%) to 300 ml sealed polypropylene bottles then filled with solutions. Solutions contained 0.1 M NaNO₃ (Aldrich, 99.995%, trace element basis), 20 mg C/L of dissolved inorganic carbon (CaCO₃ or NaHCO₃ (Sigma-Aldrich, ACS reagent) and 1 mg/L Cl₂ (NaHClO, Aldrich, reagent grade). Solution pH values were adjusted by adding either 2% HNO₃ (Fischer Scientific, trace metal analysis) or 1 M NaOH (Sigma-Aldrich, ACS reagent). The solutions were unbuffered. The bottles filled with solutions with no head space were set up on a shaker (VWR, DS-500E) at 170 rpm and continuously mixed throughout the experiment. A 12 ml Luer Lock syringe (Henkle Sass Wolf) was used for sample collection. Duplicate samples were collected from each bottle and each sample was analyzed in triplicate. Collected samples were 10 ml in volume and filtered through a 0.2 µm membrane filter (VWR, polyethersulfone membrane) with the initial 2 ml discarded and the remainder collected in a 15 ml centrifuge vial (Diamed, polypropylene). The collected filtrate was acidified with 150 µl of 2% HNO₃. Vials were stored in at 4˚C until analyzed with inductively coupled plasma mass spectrometry (ICP-MS) (Agilent Technologies 7700x) or inductively coupled plasma optical emission spectrometry (ICP-OES) (VarianVista-PRO CCD Simultaneous ICP-OES) for lead concentration. For ICP-MS analysis, samples were diluted 5 times with 2% HNO₃ prior to analysis and for ICP-OES samples were nitrified with 1 ml 70% HNO₃ without any dilution. The remainder of the batch solutions was used for measuring free chlorine and pH. Free chlorine and pH were measured respectively with HACH spectrometer (DR 2800) by USEPA DPD method 8021 and HACH pH meter (HQ 40d).

Temperature and pH: Experiments were conducted at 20˚C and 5˚C for cerussite and hydrocerussite over a range of pH values (7.0, 8.0 and 9.5) and in the presence of 140 mg/L NaHCO₃ as dissolved inorganic carbon (DIC) which provides moderate alkalinity in the solutions. Free chlorine, pH and dissolved lead were measured at regular time intervals starting at 15 minutes with the experiments lasting up to 45 days.

Alkalinity: To investigate the possible interference of alkalinity level on lead solubility, a set of experiments was carried out for cerussite and hydrocerussite at pH 8.0 with two different DIC sources (NaHCO₃ versus CaCO₃) at 20˚C versus 5˚C. 140 mg/L of NaHCO₃ was added to the solutions that was equal to 20 mg C/L and provided moderate alkalinity in the solutions. For low alkalinity solutions, 75 mg/L of CaCO₃ was added to the solutions to control alkalinity of solutions in low level because solubility of CaCO₃ is low and the excess amount of CaCO₃ will precipitate.

Free chlorine, pH and dissolved lead were measured at regular time intervals up to 45 days. Table 1 shows a summary of the experimental conditions examined in this research.

Cerussite and hydrocerussite as the major lead species in distribution systems were used in this project to investigate lead dissolution at various temperature and pH. The impact of free chlorine and alkalinity on lead solubility was studied as well. Bench-scale experiments were conducted and the following are the conclusions:

・ Cerussite and hydrocerussite solubility somewhat surprisingly were not impacted significantly by pH value ranging from 7.0 to 9.5. However, there were clear impact associated with temperature and alkalinity level on cerussite and hydrocerussite dissolution and can be considered as good predictors for dissolution.

・ As expected, lower dissolution for hydrocerussite was observed at 5˚C rather than 20˚C.

・ Considering the effect of alkalinity and temperature, at both 20˚C and 5˚C, low alkalinity resulted in higher dissolution of cerussite and hydrocerussite.

・ Hydrocerussite was always found to be more soluble at low alkalinity regardless of temperature.

・ Mix results were found with moderate alkalinity. At 20˚C, hydrocerussite was more soluble than cerussite, whereas at 5˚C cerussite was more soluble than hydrocerussite.

This work was part of a collaborative project between Carleton University and Western University that was funded by Canadian Water Network. We are grateful for the advice of Dr. Nimal De Silva (Ottawa University, Canada) on ICP analysis and for the assistance of Dr. Marie Tudoret, Miaoyi Yan from Carleton University, Canada and Dr. Claire Robinson, Caitlin Kushnir and Daoping Guo from Western University, Canada.