Arsenic does not just kill people in murder mysteries. The metalloid is poisonous, and long-term exposure at low levels leads to health problems such as skin, bladder, lung and prostate cancer. Arsenic exposure has been linked to a host of other ailments and developmental problems.A team from Texas Tech University in Lubbock and the US Geological Survey National Water Quality Laboratory in Denver has developed a detection tool that can help people avoid arsenic. It did so by exploiting a chemiluminescence reaction involving ozone and arsine (AsH3), an arsenic-containing gas, to detect inorganic As(III) and As(V) in water. “We have developed an instrument which is fully automated, sensitive, more affordable and, most importantly, field-deployable,” said team leader Purnendu K. Dasgupta.A schematic shows the setup of an instrument that uses chemiluminescence to detect arsenic in water. The syringe pump (SP) and eight-way distribution valve (DV) send chemicals in sequence into the reactor (R) to generate arsine. The arsenic compound is combined with ozone at the face of a photomultiplier tube (PMT), where chemiluminescence identifies the level of As(III) and As(V) in a 3-ml water sample. SV = solenoid valve; CC = chemiluminescence chamber; OZG = ozone generator; AP = air pump; FC = mass flow controllers (optional); Act C = activated carbon cartridge. Images courtesy of Purnendu K. Dasgupta.Arsenic is hard to avoid because it is in the drinking water at concentrations that can exceed the recommended maximum permissible level of 10 μg/l. Dasgupta, who is now at the University of Texas at Arlington, is well aware of the issue because of its effect on Bangladesh and West Bengal, India, where he grew up and still has relatives. Hundreds of thousands of people in Bangladesh have documented cases of arsenic poisoning resulting from exposure to groundwater.The danger is not confined to Asia. For instance, the US Geological Survey estimates that 32 million people in the US — more than one-tenth of the population — drink water that contains arsenic in concentrations of 2 to 50 μg/l.Unfortunately, detecting arsenic can be difficult. Current methods suffer from such drawbacks as the need for bulky and expensive instruments, the generation of heavy-metal waste and the demand for large sample volumes. Moreover, these techniques often cannot differentiate among the various oxidation states of inorganic arsenic, a crucial bit of information for determining toxicity because As(III) is considered more toxic than As(V). Dasgupta has been working on the problem for nearly 10 years, with a single goal in mind. “A simple and portable instrument for field use is highly desirable,” he said.He has known since the 1980s about the chemiluminescent reaction between ozone and arsine. However, at the time, implementation of the reaction in a detector required 20 ml of water, and it needed extensive sample preparation involving pure gases and liquid nitrogen. The technique also required a state-of-the-art, expensive photodetector. Because the procedure could not be performed in field tests, the idea was abandoned as unworkable a decade or so ago.Dasgupta’s group revisited the reaction, in part because of success that they had had with other chemiluminescence studies. They devised a solution that worked by employing sequential fluid handling, zone fluidics and the different pH dependence of the borohydride reduction of As(III) and As(V) to arsine.The instrument, described in the Oct. 15 issue of Analytical Chemistry, consists of a bidirectional syringe pump and an eight-way distribution valve. The pump injects a water sample, sodium borohydride, sulfuric acid and other solutions from reservoirs into a reactor, following a programmed sequence.The instrument sends the output of the arsenic-borohydride reaction — arsine gas — into a chemiluminescence chamber and then injects ozone generated from the air into the chamber. The cell sits directly on a miniature Hamamatsu photosensor, where electronics amplify the signal that results from chemiluminescence taking place.In a four-minute cycle that uses only 3 ml of the sample, the device determines the total arsenic present with a detection limit of 0.05 μg/l. Because the pH of the reactor can be adjusted, it also can determine the oxidation state of the arsenic that is present. If the pH in the reactor is kept between 4 and 5, only As(III) is converted to arsine in the reaction; if the pH is below 1, both As(III) and As(V) are generated.By running through the detection sequence at different levels of pH, the distribution and the level of each can be determined. The detection limit for As(III) is 0.09 μg/l.Dasgupta estimates that the instrument would cost less than $3000 to build and that it would fit into a suitcase — characteristics that make the approach more suitable than any of the alternatives for fieldwork. And although borohydride is generally unstable and expensive — both problems for a field instrument — he noted that they have devised new borohydride formulations that can be stable for at least a month.For his part, Dasgupta would like to see the instrument move out of the laboratory and into areas where it can do the most good. “We will be happy to see it being used extensively in the field. Mass production of the instrument might be an option through joint partnership with interested companies,” he said.Contact: Purnendu K. Dasgupta, University of Texas at Arlington; e-mail: firstname.lastname@example.org.