An ethylenediaminetetraacetic acid (EDTA) titration analysis of the both the hardness and calcium concentrations of tap water samples from three San Francisco residences was performed to determine any differences due to geographic location. Four samples for each location (three tap water and a de-ionized (DI) water blank) were tested for each experiment. The water's average hardness was determined to be 1.10 ppm with a standard deviation (SD) of 0.37 ppm, indicating very soft water. Also, Qtest calculations showed that certain hardness data had to be discarded. The average [Ca2+] was 0.28 ppm with a SD of 0.02 ppm. However, the experimental results do not agree with the published standard data at the 95% confidence level.


San Francisco obtains ~90% of its fresh water from the Hetch Hetchy watershed, with watersheds in Alameda, Santa Clara, and San Mateo supplying the remaining amount (1). In the San Francisco Public Utilities Commission 2006 Water Quality Report, it specifies water hardness as [CaCO3], having values of between 6-146 ppm with an average of 66 ppm, and [Ca2+] having values between 3-28 ppm with an average of 15 ppm (1). The U.S. Department of the Interior and the Water Quality Association consider water hardness (in CaCO3 ppm) of < 75 ppm in the soft to lower moderately hard classification ranges, while water hardness > 75 ppm is in the upper moderately hard to very hard ranges (2). In general, Bay Area drinking water is very soft.

The US Environmental Protection Agency (EPA) establishes standards regulating water quality. The primary standards regulate health concerns, while the secondary standards regulate aesthetic concerns. However, the EPA has no regulations for water hardness and calcium level under either of these standards, or elsewhere (3). The federal Safe Drinking Water Act was established in 1974, and last updated in 1996 (1). The effects of water hardness are currently regarded as more of a nuisance than a health or environmental hazard. Hard water can cause mineral scaling in pipes, a concern for industries with water-intensive processes, particularly in areas with hard water. Water hardness can include or involve other chemicals in addition to CaCO3, and this can create problems for aquatic life. The Bay Institutes' publication San Francisco Bay Water Quality Index: Indicator Analysis and Evaluation notes that the toxic effects of certain metals (such as copper and mercury) and trace elements (such as arsenic) are enhanced in soft water (4). Although a number of studies have been conducted in the past 50 years on the relationship between water hardness and cardiac health, the results have been inconclusive (5, 6, 7). Still, the World Health Organization (WHO) is trying to establish guidelines for water quality that include water hardness (8).

This experiment was conducted to determine if there are hardness and calcium concentration differences in the water due geographic location in the city. The three site locations—West, Central, and East—range across the city. The West location is primarily residential, with single family and multi-unit buildings. The Central location is primarily a mix of multi-unit residential and office buildings. The East location is a former industrial corridor that has been changing to residential neighborhoods of renovated industrial buildings and newly constructed lofts. Though plumbing materials are unknown, it is assumed their materials and construction meet standards required by the Safe Drinking Water Act.

EDTA makes an ideal titrant for metal ion titrations because it chelates with the metal ion(s) of interest in a 1:1 relationship, regardless of the metalís specific ionic charge. At equivalence in an EDTA titration for water hardness (the [Ca2+ Mg2+]), the Eriochrome black T indicator changes from red to pure blue. At first, the indicator is bound with some Mg2+ as MgIn-, and this causes the red color. As the EDTA chelates with the Ca2+ and Mg2+, the indicator is freed and thus the solution's color changes to blue.

Ca2+ + Mg2+ + MgIn- (red) + Y4- => CaY2- + MgY2- + In3- (blue)

At equivalence in an EDTA titration for solo [Ca2+], the addition of NaOH solution binds up the Mg2+, enabling the EDTA to bind only to the Ca2+. Once the endpoint is reached, the hydroxyl methyl blue indicator turns blue (2, 9, 10).

Ca2+ + Mg(OH)2 + In- (red) + Y4- => CaY2- + Mg(OH)2- + In3- (blue)


The water was run for 30 seconds prior to collection in a plastic sample bottle. All samples were tested for hardness within 15 hours of collection time and tested for calcium content within 19 hours of collection time. Three solutions and one indicator were made in accordance to R. Neilson's guidelines (2, 11). The standards made were 500.0 mL (± .05) of a standard 0.003 M EDTA solution, 200.0 mL (± .02) of a 50% wt concentrated NaOH solution, and 500 mL (± .5) of a pH 10 ammonia buffer solution. 20.0 mL (± .2) of Erichrome black T indicator (EBT) was made to indicate the hardness experiment's endpoint. Hydroxyl methyl blue (HMB) was used to indicate the calcium concentration titration endpoint, in accordance to E. Wade's guidelines (12).

For each test, each site had three 50.0 mL (± .05) samples and one 50.0 mL (± .05) blank of DI water. To determine water hardness, each 50.0 mL sample was treated with 3.0 mL (± .003) buffer solution and 10 drops EBT. To determine [Ca2+], each sample received 0.2 g (± 1) HMB indicator and 5 drops NaOH solution, the flask swirled for two minutes and sealed with airtight wrap to prevent oxidation.

Results and Discussion

To determine [Ca2+ Mg2+] and [Ca2+], first the concentration of the standard EDTA solution was determined in both molarity and millimole units, as follows:

0.6016 g (± 1) EDTA × 372.24 g/mole = 0.002 EDTA moles (1)

0.002 EDTA moles ÷ .500 L = 0.003232 M, or 3.232 mMoles/L (2)

The average titrated volume for each test was multiplied with the [EDTA] to obtain the EDTA mMoles, as this value equates to [Ca2+ Mg2+] and [Ca2+] for the respective test. The mMoles value was multiplied to the formula weight (100.09 g/mol for hardness, 40.08 g/mol for calcium) to obtain the respective ppm value.

The hardness test data is reported in Table 1. The average hardness concentrations for each of the locations were as follows: West, 1.10 ppm; Central, 1.10 ppm; East, 3.01 ppm. Given a wide range of values titrated for the East site (as indicated by the three-fold increase of ppm concentration), a Qtest was performed to check the validity of the data. The Qcalc value was greater than the Qtable value for three data points at the 90% confidence interval level (0.98 > 0.94), thus the East site data was disregarded. The average hardness concentration of the remaining two locations was 1.10 ppm, with a SD of 0.37 ppm. A Case #1 Ttest was performed at the 95% confidence interval level using the minimum hardness value (6.0 ppm) as reported by the SFPUC, as follows:

Tcalc = |1.10 − 6.0| ÷ 0.37 × √2 = 18.73 (3)

However, the Tcalc value is greater than the Ttable value (18.73 > 4.303), indicating that the values do not agree within a 95% confidence interval.

For calcium concentration, the overall average values for the three locations are in better agreement. The calcium test data is reported in Table 2. The average calcium concentrations for each site were as follows: West, 0.34 ppm; Central, 0.29 ppm; East, 0.22 ppm. The overall average value for calcium concentration was 0.28 ppm with a SD of 0.02 ppm. A Case #1 Ttest was performed at the 95% confidence interval level, using the minimum 3.0 ppm value reported by the SFPUC as follows:

Tcalc = |0.28 − 3.0| ÷ 0.02 × √3 = 235.6 (4)

Again, the Tcalc value is greater than the Ttable value (235.6 > 3.182), indicating that the values do not agree within a 95% confidence interval.

During testing, the East site's water gave unexpected results. If only the West and Central sites are considered, the average and SD values for hardness decrease to 1.10 ppm and 0.37 ppm, respectively, and the average and SD values for calcium concentration slightly increase to 0.32 ppm and 0.0 ppm, respectively. In particular, the East site calcium concentration blank developed a violet color upon standing, and adding 0.2 mL (± .03) of EDTA enhanced this violet color, instead of changing it to blue. It is unknown what chemical reaction caused the East site blank to turn violet. Further tests would need to be performed to determine if the East site's idiosyncratic results are indeed due to chemical differences in the water perhaps from the location's site in a former industrial corridor, or due to experimental error.


An EDTA titration was performed on tap water samples from three residences in San Francisco to determine total hardness and total calcium concentrations. EDTA was chosen as the titrant due to the stable 1:1 chelation relationship between EDTA and metal ions. The mean and SD values were calculated for each location in the separate experiments, and also mean and SD values calculated for each experiment that includes 1) all three locations' data and 2) only the first two locations' data, as the third location's data showed idiosyncrasies. The results show an average hardness value of 3-5 ppm less than the lowest hardness value reported by the SFPUC (6 ppm). The average calcium differed by 0.04 ppm, dependent on whether the East site data is included or not. Results of the Ttest and Qtest calculations show the data is largely invalid, and the experiment should be redone.