Chemistry matters. Join us to get the news you need.

If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Eating Arsenic

ACS Meeting News: Efforts to find out how the element gets into food are leading to ways to reduce its presence

by Celia Henry Arnaud
May 6, 2013 | APPEARED IN VOLUME 91, ISSUE 18

Credit: Bill Mindak
Rice of all kinds contains some arsenic.
Credit: Bill Mindak
Rice of all kinds contains some arsenic.

The presence of arsenic in some foods has been reported in the news in the past couple of years, first in juice and then in rice.

In September 2011, the syndicated television program “The Dr. Oz Show” reported levels of arsenic in apple juice above the drinking water standard of 10 ppb. In a follow-up on this alert, Consumer Reports published an analysis of arsenic in commercial apple juice and called for tighter government regulations.

Then the focus switched to rice. Last September, Consumer Reports published its analysis of arsenic in rice. In a coordinated effort, the Food & Drug Administration released preliminary results from its own study of arsenic in rice and rice products on the same day. The studies clearly showed that all rice products contain some arsenic. What they didn’t show so clearly was what can be done about it.

A symposium at the American Chemical Society national meeting, held last month in New Orleans, aimed to cut through some of the confusion by shining its own light on the story of arsenic in water and food, and particularly in rice; presentations covered sources, analytical methods, and possible remediation strategies. The symposium was cosponsored by the Divisions of Agricultural & Food Chemistry, Environmental Chemistry, Agrochemicals, and Small Chemical Businesses.

Arsenic is the 20th most abundant element in Earth’s crust, so its presence in food is not surprising, several speakers noted. “All foods naturally contain arsenic,” said Kenneth J. Reimer, a chemistry professor at the Royal Military College of Canada, in Kingston, Ontario. “As chemists, we should educate the public as such.” But some foods, including rice, seaweed, mushrooms, and seafood, accumulate arsenic at higher levels than others.

“Soil is the primary source of most biologically active trace elements that reach humans through the food chain,” said David B. Smith, chief of the Soil Geochemical Landscapes Project at the U.S. Geological Survey (USGS) in Denver.

Smith described USGS’s recent survey of more than 14,000 soil samples from 4,800 sites in the 48 contiguous states. USGS chemists analyzed samples for 45 major and trace elements, including arsenic. They measured total arsenic by using hydride-generation atomic absorption spectrometry.

Soil arsenic levels varied by about three orders of magnitude, ranging from less than 0.6 mg/kg to more than 1,000 mg/kg. The lowest arsenic levels were in the southeastern corner of the U.S., particularly Florida, which had a median arsenic concentration of 0.6 mg/kg. States with the highest median arsenic concentrations were Ohio (9.7 mg/kg), Montana (8.7 mg/kg), Maine (8.0 mg/kg), South Dakota (7.7 mg/kg), and Pennsylvania (7.7 mg/kg). Individual sites with the highest arsenic levels were in Nevada (1,110 mg/kg) and Montana (588 mg/kg).

The regions with high soil arsenic levels are underlain by bedrock or unconsolidated glacial deposits that also contain high arsenic, Smith said. The study also revealed elevated arsenic levels along the Mississippi River, which is not surprising, he said, because the river drains a large swath of the country. “Input of arsenic from human activities is superimposed on this highly variable natural background, and it is generally very difficult to distinguish the anthropogenic from the natural input,” Smith said.

The largest source of arsenic in the environment from human activities is from the mining and coal-burning industries, said Kevin L. Armbrust, director of the Mississippi State Chemical Laboratory and an associate professor of chemistry at Mississippi State University. Other sources include pesticides, fertilizers, and pressure-treated wood. For example, lead arsenate was once a popular insecticide. Use of it peaked in the 1940s and dropped precipitously after the introduction of DDT, Armbrust said.

Arsenic exists in many forms, the most toxic of which is inorganic arsenic, which is found as either As(III) or As(V)—considered to be equally toxic. Inorganic arsenic can be metabolized to form a variety of organic species, monomethylarsonic acid and dimethylarsinic acid being the most abundant. Other organoarsenical species include arsenobetaine and various arsenosugars. Measurement of the different forms of arsenic is a process known as speciation.

Different forms of arsenic get into cells in different ways, said Lena Q. Ma, a professor in the School of the Environment at Nanjing University, in China, and in the soil and water science department at the University of Florida. For example, As(V) is a phosphorus analog that enters cells via phosphorus transporters and can replace phosphorus in many reactions. In contrast, As(III) gets taken up by aquaporin transporters. It has a high affinity for thiols and can inhibit enzymes through reactions with cysteine.

But opinions are divided over how much the presence of particular arsenic species matters. Inside cells, As(V) is quickly reduced to As(III), said Shannon R. Pinson, a rice biologist at the U.S. Department of Agriculture in Stuttgart, Ark. Although some As(V) remains, most inorganic arsenic in the body will be As(III), regardless of its form in the environment.

Of these arsenic species commonly found in food, the inorganic forms are the most toxic. Arsenobetaine is nontoxic.
Of these arsenic species commonly found in food, the inorganic forms are the most toxic. Arsenobetaine is nontoxic.

It’s widely thought that inorganic arsenic is toxic and organoarsenicals are nontoxic. But this is an oversimplification. For example, the main metabolites of inorganic arsenic are methylated but have been shown to be toxic. The only organoarsenical that has been shown to be safe is arsenobetaine, the main arsenic species found in seafood. For other organoarsenicals, like arsenosugars, there’s not enough evidence to declare a verdict on toxicity. Organoarsenicals probably have a range of toxicities, Reimer said.

Because of such differences in toxicity, a number of researchers in the field believe speciation is a key issue and have therefore been developing new methods and improving old ones for identifying and quantifying arsenic in its various forms.

For example, FDA is focusing on inductively coupled plasma mass spectrometry (ICP-MS) for analyzing and speciating arsenic, said William R. Mindak, a chemist at FDA’s Center for Food Safety & Applied Nutrition, in College Park, Md. FDA’s method for total arsenic analysis in rice and rice products calls for an acid digestion, followed by ICP-MS detection. For speciation, FDA scientists use a dilute nitric acid extraction followed by an isocratic anion-exchange chromatographic separation and ICP-MS detection to measure the two inorganic forms, plus monomethylarsonic acid and dimethylarsinic acid.

The method is not perfect, Mindak acknowledged. For some samples, for example, chromatographic peaks of components overlap with those of As(III) in the total arsenic analysis. To achieve separation, a peroxide oxidation step is used to convert As(III) to As(V), since his team isn’t trying to speciate the samples, Mindak said. Potential ways to improve the separation of arsenic-containing species include the use of different mobile phases, solvent gradients, and smaller particles in the column for the chromatography, he said.

Credit: U.S. Geological Survey
Arsenic is found in surface soil throughout the U.S., as shown in a study by the U.S. Geological Survey. The black dots indicate where soil was sampled; samples were acquired at a depth of 0–5 cm. In addition to state boundaries, rice-producing counties in California, Texas, Louisiana, Mississippi, Arkansas, and Missouri are outlined in black.

The ICP-MS method was originally developed for raw rice, said Sean Conklin, Mindak’s colleague at FDA. But the discovery by Brian Jackson’s group at Dartmouth College that brown rice syrup, a common sweetener, can also contain elevated arsenic levels pushed them to extend the method to many other products such as snack bars, which can be made with rice syrup. These products, which can be gooey and sticky—think bits of fruit, chocolate, and marshmallows—may not be compatible with the original method as written, Conklin said.

One problem with the increased scrutiny of arsenic in food is that many analytical labs are unfamiliar with food as a sample matrix, both Mindak and Conklin said. Measuring total arsenic can be challenging if a lab is not familiar with food analysis, Mindak said, and speciation is even more difficult. Most labs, including some FDA labs, don’t have experience with arsenic speciation or the combination of high-performance liquid chromatography with ICP-MS, he noted.

To address the deficiencies, FDA undertook a multilab method validation exercise to increase its capacity for arsenic speciation analysis. Four FDA labs, two Food Emergency Response Network labs, and two contract labs participated in the exercise. Some problems with incomplete digestions were reported, but the most pervasive problems involved supposedly blank samples that were found to have detectable arsenic levels and improper chromatographic peak integration, Conklin said. Peak integration is how chromatographic data are translated into relative quantities of analytes. With the help of a troubleshooting checklist, Conklin and other participants were able to pinpoint the contamination problem and bring the participating laboratories up to speed, he said.

Labs other than FDA’s are also involved in measuring and speciating arsenic. They have to address the same issues as FDA.

One effective technique for removing interfering species in such samples is the use of ICP-MS with a triple quadrupole mass analyzer, said Amir Liba, an applications biochemist at Agilent Technologies, in Wilmington, Del. In such an analysis, the first quadrupole selectively allows a single mass into a collision cell. Sometimes interfering ions enter the cell along with the desired species. However, carefully controlled reactions with O2 gas in the collision cell followed by mass analysis with the second quadrupole can distinguish arsenic-containing species from interfering species with similar mass-to-charge ratios, such as ArCl+ and doubly charged rare-earth elements.

Meanwhile, Jack Driscoll and his coworkers at PID Analyzers, an instrument company in Sandwich, Mass., are using a classic method—photoionization detection—to measure arsenic in juice and water. The method is easy to use and doesn’t require much training, Driscoll said. To analyze arsenic species, the researchers convert them to hydrides, separate them with gas chromatography, and detect them with a photoionization detector at levels less than 1 ppb.

Being able to measure arsenic is important, but minimizing its presence in food in the first place is even more significant. Efforts are under way to develop methods of soil and water remediation to do just that.

For instance, arsenic adsorbs on the surface of oxides and hydroxides of iron and manganese, said Philip A. Moore Jr., a soil scientist at USDA in Fayetteville, Ark. Therefore, adding something like iron oxide to soil can decrease the available arsenic, he said.

However, iron and manganese are both redox-active. So under the flooded conditions used for rice cultivation, the soil’s redox potential drops, and the oxides release their adsorbed arsenic.

To sidestep that problem, Moore has turned his attention to aluminum sulfate, otherwise known as alum, which is not redox-active. He has found that adding alum to soil does decrease the availability of arsenic. He noted that the high cost of alum may make it impractical for agriculture but that other less expensive chemicals may work in a similar fashion.

Water management practices also strongly affect the amount of arsenic in rice. The conventional method of growing rice requires flooding the fields. This flooding causes the soil to lose oxygen, Moore said, and results in anaerobic conditions, which favor reduction of arsenic. The resulting As(III) is more soluble than As(V) and thus more available to the plants.

Credit: Merle M. Anders & Anna McClung
Traditional rice cultivation involves flooding the field, but alternative experimental methods such as furrow irrigation could reduce the amount of arsenic in the grain.

“Water has a bigger effect than soil on arsenic levels in rice,” said Rufus L. Chaney, an agronomist at USDA in Beltsville, Md. A study conducted by the agency’s Pinson and her collaborators showed that rice grown under flooded conditions takes up 10 times more arsenic than rice grown without flooding. Drier conditions, however, reduce crop yields and increase cadmium in the rice, Chaney said.

The rice industry has been conducting similar studies in collaboration with USDA to help resolve those issues, said Reece Langley, vice president for government affairs at the USA Rice Federation, an industry trade group.

In one planned study, rice growers aim to determine what water management practices reduce arsenic levels without adversely affecting crop yield and milling characteristics. The study will evaluate four water management techniques—conventional flooding, two approaches to intermittent flooding, and no flooding at all. In one intermittent approach, fields are flooded, drained to mud, and reflooded repeatedly throughout the growing season. In the second intermittent approach, fields are flooded twice over the course of the growing season—at the beginning of the season for a short period; drained, allowing the soil to dry; and then reflooded until the end of the growing season.


In addition to adjusting growing conditions, researchers hope to use breeding or genetic modification to create rice varieties that accumulate less arsenic. So far, Pinson said, 13 genes have been identified that affect arsenic accumulation in rice grains. Nine of those genes are effective in flooded conditions, and a partially overlapping set of six of the genes is effective in unflooded conditions. In addition, another group of nine of the 13 genes affects accumulation of other elements that are correlated with arsenic uptake or detoxification within plants, such as phosphorus, silicon, and sulfur.

But the question of what these arsenic levels mean for consumers still remains to be answered. FDA has analyzed arsenic in more than 1,000 samples of rice and rice products and will be evaluating the safety of those foods. Those results will be available to the public on the agency’s website, said Suzanne Fitzpatrick, the senior adviser for toxicology at FDA in College Park. FDA is also completing a quantitative assessment to determine the risk posed by arsenic in apple juice.

FDA is also working with other government agencies, including USDA and the National Institute of Environmental Health Sciences, to assess the risk from arsenic in food, Fitzpatrick said. FDA doesn’t yet know whether it will recommend levels for particular foods, consumer advice, or both, she said, because risk depends on both hazard and exposure, and exposure depends on individual consumption.

Because arsenic is naturally present in foods, it’s unlikely to be completely eliminated from them. However, food scientists are focused on ensuring that the levels that are there will be safe.



This article has been sent to the following recipient:

Bob Buntrock (May 15, 2013 4:39 PM)
Obviously, testing for As speciation is essential given the wide range of toxicity for As compounds. Also note that State median levels can be deceptive since, from the map, a good share of Maine tests considerably over the 8.0 ppm median value. Of course, speciation is again important.
sandra greve (March 30, 2015 11:18 AM)
after reading , i am not that worried about the wine i drink.......we make alot of our own wine and our fruits come from wi. and florida. i think the red wine for health of my intestines, make more sense to still do a couple glasses a night......we have had so many calls about our homemade wine killing

Leave A Comment

*Required to comment