Accurate pH measurement of water samples depends critically on the quality of the standard buffer solutions used to calibrate the pH meter. These solutions serve as the reference against which every water sample is judged, and any error in their preparation propagates directly into the measurement results.
The pH of natural water typically falls between 6 and 8, but industrial discharges can shift it sharply toward either extreme, making reliable calibration essential.
The three buffer systems most commonly used in water quality analysis correspond to the acidic, neutral, and alkaline regions of the pH scale. The potassium hydrogen phthalate buffer, with a pH of approximately 4.01 at 25 °C, is prepared by dissolving 10.12 g of potassium hydrogen phthalate, previously dried at 115 ± 5 °C for 2 to 3 hours, in water and diluting to 1000 mL.
The mixed phosphate buffer at pH 6.86 requires 3.388 g of potassium dihydrogen phosphate and 3.533 g of disodium hydrogen phosphate, both dried at 110 to 130 °C, dissolved together and made up to the same final volume. The borax buffer at pH 9.18 is prepared by dissolving 3.80 g of borax, pre-equilibrated in a desiccator over a saturated sucrose solution, in freshly boiled and cooled water and diluting to 1000 mL.
The water used throughout should be boiled and cooled to remove dissolved carbon dioxide, with a conductivity below 2 × 10⁻⁶ S/cm and a pH between 6.7 and 7.3. Carbon dioxide dissolved from the air can react with borax and alter the pH of alkaline buffers, so the borax solution should be stored in a tightly sealed polyethylene bottle. Potassium hydrogen phthalate solutions, being organic in nature, may develop microbial contamination over time; any visible turbidity indicates the solution should be discarded and freshly prepared.
A further consideration is that buffer pH values are temperature-dependent. When calibrating a pH meter, the temperature of the buffer solution should be recorded, and the instrument's temperature compensation should match the actual measurement conditions. Keeping the temperature difference between the calibration buffer and the water sample within 1 °C minimizes systematic error.

