Development of a Miniature Sensor for Point-of-Care Determination of Mercury

Abstract

Mercury is toxic to human health. In developing countries, artisanal and small scale gold mining (ASGM) entails mixing elemental mercury with crushed sediments to amalgamate gold. The mercury is then burned off, making it inhalable during the process. More than 15 million people -men, women, and children- in Central and South America, Africa and Asia are involved in ASGM. ASGM is responsible for ~40% of elemental mercury emitted into the environment. Mercury can cause mild to severe neurologic toxicity and even death. The US federal Biological Exposure Index (BEI) is currently set at 50 ppb (μg/L) for urine. Blood mercury may include both organic and inorganic forms and must be speciated to quantify the inorganic form. The half-life in blood is 70 days after acute exposure. The half-life of elemental mercury in the urine is ~7 days and requires a spot urine (single, one-time specimen). The current gold standard analytical methods for elemental mercury quantification are AAS and ICP-MS, which are costly and time consuming, and not readily available in countries where testing is needed. Electrochemical point-of-care (POC) sensors can reduce costs and time and could help expedite effective treatment. Our POC system relies on the Square Wave Anodic Stripping Voltammetry (SWASV) electrochemical approach. It is performed on disposable miniature three-electrode sensors (gold WE, gold CE and Ag/AgCl RE). The light weight, small and low-cost miniature USB potentiostat was used for the measurements. The sensor interface with the potentiostat included a vibration motor to provide sample agitation. Each experiment used a single 10 µL droplet of sample. We measured mercury in urine because it is non-invasive measurement, with a simpler organic matrix and no necessary preparation steps. During this work, fresh urine samples were stored in a -20 °C freezer. First, the SWASV conditions were optimized in buffer solution to demonstrate the feasibility of Hg detection on miniature sensors. Then, the validity of the POC system was confirmed by analysis with human urine. Nature and ionic strength of the supporting electrolyte (0.1 M HCl), pH (0.6), deposition potential (0 V), deposition time (900 s), parameters of the square wave scan (amplitude 10 mV, increment 4 mV, scan rate 600 mV/s, period 6 ms), intensity of the motor vibration (3.5 V) and dilution of the urine samples (dilution factor = 4) were perfectioned. A cleaning step in between measurements was introduced to overcome the forming of amalgam of Hg and Au. The cleaning step reduced the standard deviation of measurements, removing all Hg from the surface. Nature, conditions, and duration of it were investigated (chronoamperometry at +1.3 V for 120 s in 0.01 M HCl). After optimization, the limits of quantitation were 15 ppb Hg and 5 ppb Hg respectively in 0.1 M HCl and urine (linear ranges are respectively 15 ppb – 200 ppb Hg and 5 ppb – 20 ppb Hg). Urine broadened the shape of the peak: peak area was the best indicator of Hg concentration in urine. The vibration motor enhanced the sensitivity of the system by allowing all metal ions to enter in contact with the electrode surface and participate to the experiment. The lowest limits of detection were 3.14 ppb Hg in peak height measurements in supporting electrolyte solution (sensitivity 80.2 nA/ppb) and 3.79 ppb Hg in area measurements in urine solutions (15.16 ppb Hg considering a dilution factor 4x for urine; sensitivity 14.795 nW/ppb). The validity of the point-of-care system was confirmed by performing standard addition analysis of unknown Hg concentrations in urine, reaching 85 % accuracy (number of samples n=5). This POC system for mercury detection in urine could offer rapid help for individuals in the field, far from big laboratories. We succeeded in detecting trace concentrations of mercury on miniature sensors, reaching LOQ and LOD below the BEI intervention threshold in urine samples. This is a key innovation: at the moment lab-on-a-chip devices for mercury detection in urine do not exist. Future steps will focus on validation with a large number of samples from Hg-exposed population.