An accurate, precise, and affordable light emitting diode spectrophotometer for drinking water and other testing with limited resources

Abstract

Spectrophotometers are commonly used to measure the concentrations of a wide variety of analytes in drinking water and other matrixes; however, many laboratories with limited resources cannot afford to buy these very useful instruments. To meet this need, an accurate, precise, and affordable light emitting diode (LED) spectrophotometer was designed and built using best engineering practices and modern circuit design. The cost and performance of this LED spectrophotometer was compared against 4 common commercial spectrophotometers. More specifically, the performance of these spectrophotometers was evaluated from the upper limits of linear range, upper limits of operational range, calibration sensitivities, R2 values, precisions of standards, estimated limits of detection, and percent calibration check standard recoveries for the determinations of iron (Fe), manganese (Mn), and fluoride (F-) in drinking water. This evaluation was done in the United States (U.S.) and India. Our LED spectrophotometer costs $63 United States Dollars (USD) for parts. The 4 commercial spectrophotometers ranged in cost from $2,424 to $7,644 USD. There are no practical differences in the upper limits of linear range, upper limits of operational range, R2 values, precisions of standards, and estimated limits of detection for our LED spectrophotometer and the 4 commercial spectrophotometers. For 2 of the 3 analytes, there is a practical difference in the calibration sensitivities our LED spectrophotometer and the 4 commercial spectrophotometers. More specifically, the calibration sensitivities for Mn and F- using our LED spectrophotometer were 65.2% and 67.0% of those using the 4 commercial spectrophotometers, respectively. In conclusion, this paper describes the design, use, and performance of an accurate, precise, and extremely affordable LED spectrophotometer for drinking water and other testing with limited resources.