Science and Technology at the Ames Laboratory
Spring 1996

The true history of the life and adventures of an Ames Lab invention in the real world.
In 1990 an employee of Sun Up Foods, an orange juice processor, told the Food and Drug Administration (FDA) that the company had set up secret rooms in its facilities to hold tanks of liquid beet sugar, a common adulterant of orange juice. According to the account of this case given in the FDA Consumer, the FDA got a search warrant for one facility but could not find the secret room. During interviews conducted in the following months, FDA investigators learned its location. They obtained a second search warrant, and this time they found the room. Stainless steel pipes hidden in the walls "were set up to look like part of the sewage system, and during a government inspection, the line carrying the sugar could be shut off and the outside pipe closed to conceal the sugar line inside." An FDA compliance officer remarked that it was the most sophisticated adulteration scheme he had ever encountered.

This admittedly comical episode becomes less funny when you realize that Sun Up Foods is estimated to have cheated consumers out of between $10 million and $20 million by substituting inexpensive liquid beet sugar for the more expensive orange juice concentrate.

By the time the FDA was sending in search teams, however, Sun Up was already in trouble. In the spring of 1990, before the FDA searches, Sun Up lost 90 percent of its business when analysis of a customer's product showed it was contaminated with beet sugar.

The analytical instrument that detected the contamination may well have been one invented by Dennis Johnson, a senior chemist at the Ames Laboratory and a professor of chemistry at Iowa State University (ISU). His electrochemical detector, which is sold by Dionex, a company in Sunnyvale, CA, is now the preferred instrument for detecting several classes of chemical compounds important to the food and pharmaceutical industries.

Johnson, an electrochemist, studies the many ways in which the electrical current produced by a chemical reaction is used to measure the amount of the substances participating in the reaction. As Johnson tells the story, when he was a graduate student, "the technique of electrochemistry, which had been successfully applied to the determination of metals, was losing its place in analytical laboratories to atomic spectroscopy, which was more sensitive and more easily applied." But he thought electrochemistry -- "this idea of pushing electrons around," as he calls it -- was so much fun that he was determined to find something else it could do.

"I approached the problem by asking what spectroscopy could not do easily," Johnson says, "and I saw that there was this whole group of compounds -- starches, sugars and complex carbohydrates -- which couldn't be detected by gas chromatography because they aren't volatile and couldn't easily be detected by photometry because they don't absorb light. So here was a huge group for which the analytical technology was very laborious.

"So I said, 'Let's apply electrochemistry to these compounds.' The only work in the literature that was even related to this problem had to do with fuel cells. People were using platinum electrodes to produce electrical energy from methanol and ethanol. They could get some electricity -- a molecule of methanol gives up six electrons when it combines with the oxygen from water to form carbon dioxide -- but the electrode activity quickly fell off, probably because the carbonaceous products of incomplete oxidation stuck to the electrode and prevented fresh compounds from reaching its surface.

"But there's an old electrochemist's trick for cleaning electrodes," Johnson continues. "You alternate the voltage on the electrode: high, low, high, low. This makes oxide on the surface, burning off the carbonaceous compounds in the process, and then dissolves the oxide away, leaving a clean surface. In some ways it was a trivial innovation to build a cleaning cycle into the detection cycle of an analytical instrument, especially since electronics had reached the pitch of sophistication where this was easily done.

"So we began experimenting with electrodes that periodically underwent a cleaning cycle, and one day one of my students came in and said, 'I think I can detect every carbohydrate in the world.' 'Well then,' I said, 'we're in trouble. If you can detect them all, that means you have to separate them before you detect them so you can tell what you're detecting, and there is no really good way to separate carbohydrates.'

"About this time a representative of Dionex visited ISU. When he heard we could detect all the carbohydrates in the world, he said, 'Our company is a separations company; we do chromatography. Why don't we get together on this?''

So the Dionex chemists devised a better separation column for carbohydrates, and their electrical engineers built a commercial version of Johnson's detector. The lineal descendents of this column and detector are sold today by Dionex as the CarboPac series of columns and the ED40 Electrochemical Detector, which is billed as "the most powerful and sensitive electrochemical detector ever developed."

"It's the combination of the separation method and the detection method that's so important," says Roy Rocklin, R&D group leader of electrochemistry at Dionex. "The detection method is 10 to 100 times more sensitive than the method that had been used until then. Moreover, it fits nicely with the separation method. The detection method works for carbohydrates only at high pH, and it just so happens that the separation method for carbohydrates also works only at high pH. So the new and more selective separation method was easily combined with the new and more sensitive detection method. The resulting instrument didn't open up a new area of analytical chemistry, but it was a big step forward in ease-of-use, selectivity and sensitivity, and that meant more people started doing research with these compounds."

The orange juice problem illustrates the merits of the electrochemical detector. Orange juice is typically adulterated with sweeteners technically known as invert sugars, which are made by a chemical reaction that partially converts the simple sugar sucrose into glucose and fructose. The reaction can be controlled to produce a glucose/fructose/sucrose ratio matched to the ratio in orange juice. And if this is done, the simple sugars in themselves cannot be used to distinguish pure orange juice from an adulterated product.

Sugar cane and oranges produce sucrose by different metabolic pathways, however, so if the sucrose in the invert sugar comes from cane sugar, there is a clever and well-established way of detecting it. But beets and oranges produce sucrose by similar metabolic pathways, and so adulteration with beet sugar is not detectable by this technique. Johnson's electrochemical detector neatly solves this problem because it can detect raffinose, a sugar molecule present in beet invert sugar but absent in pure orange juice. A raffinose peak is a reliable marker of adulteration with beet sugar.

Sugars can also be used to detect the adulteration of coffee, molasses, honey and other foods. For example, if coffee beans are mixed with husks, hulls, or parchments (non-bean parts of the coffee cherry), the processed soluble coffee will contain high levels of the sugar xylose. A recent study of samples of commercial soluble coffees concluded that pulsed amperometric detection, as Johnson's technique is technically known, is "the simplest and most powerful technique" for the routine analysis of soluble coffee. It also showed that 12 percent of 700 samples of commercial soluble coffee, or 84 samples, were adulterated.

Some of the most interesting applications for the detector involve the study of glycoproteins. Proteins are made up of two types of molecules: amino acids and carbohydrates. The amino acids largely determine the function of the protein and, until recently, the role of the attached carbohydrates was poorly understood. The increased selectivity and sensitivity of Johnson's analytical method has allowed researchers to learn more about these carbohydrates.

One interesting glycoprotein is tPA, or tissue plasminogen activator, a clot-dissolving substance that, when properly administered, can greatly reduce the damage done by a heart attack. This compound dissolves blood clots by activating the enzyme plasmin, which degrades the fibrin strands that constitute the bulk of a clot. Tissue plasminogen activator exists in two common variants, which differ in their sugar content. The activity of one type has been shown to be as much as 50 percent greater than the activity of the other, and this variation is of concern because excessive tPA can cause dangerous hemorrhaging. The combination of liquid chromatography and pulsed amperometric detection is used as a quality-control tool during the production of tPA to ensure that the drug's therapeutic activity is predictable.

Pulsed amperometric detection excels at the determination of carbohydrates and glycoproteins, but Johnson soon realized it could be used to detect other substances as well. In the late 1980s and 1990s he and his students published articles about the detection of amino acids, sulfur compounds, alkanol- amines (compounds used as emulsifiers in the pharmaceutical industry), and aliphatic amines (breakdown products of proteins).

Johnson's latest work with aliphatic amines has interesting although slightly stinky applications. When fish spoil, the degradation of their proteins by bacteria produces organic amines, including diamines with the evocative common names of cadaverine and putrescine. The presence of these compounds renders the fish unpalatable, and they are also thought to be responsible for food poisoning.

Conventional schemes for the detection of amines are cumbersome and slow. By contrast, the electrochemical method of detecting diamines is quick and simple, as demonstrated by a recent study of the analysis of Scombridae and Clupeidae, otherwise known as tuna and sardines.

The ability to detect amines might also have medical applications. When the body rejects a transplanted organ, it begins to break apart the protein of which the organ is composed, and trace levels of cadaverine and putrescine appear in the blood. These compounds, therefore, might serve as early warning signs of organ rejection or even, perhaps, of early stage cancer.

Johnson clearly relishes the opportunities his detector has afforded him to talk to people about the chemistry of food and the chemistry of life. One of his favorite stories has to do with brandy. "I was talking to a lady from home economics," says Johnson, "who's an expert in wine. Her specialty is the back of the lip and the tip of the tongue. We were looking at brandies that were high in propylene glycol, which worried me at first because ethylene glycol, best known for its use as an anti-freeze, is toxic. But she explained that propylene glycol is an essential component of brandy. Connoisseurs swirl the brandy in a snifter to look at its streaking, and glycols control viscosity and therefore control streaking.

"So I told her facetiously that we could rig up a kind of artificial brandy connoisseur that would be able to tell whether the brandy was viscous, based on the glycol levels, and whether it was sweet or dry, based on the sugar levels."

But the main satisfaction Johnson derives from the electrochemical detector is the simple one of having invented something that has proved to be of genuine use in the workaday world.

For more information: Dennis Johnson, 515-294-7530, dcj@iastate.edu
Current research funded by: National Science Foundation and Dionex Corp.

Last Revision: 4/17/98 mab

Contents | Ames Lab Home Page | Comments | Search | Disclaimer