Benzene is a carcinogenic compound that is known to be formed when two common ingredients in soft drinks - sodium benzoate (a preservative) and ascorbic acid (vitamin C) - are used together. When placed in acidic conditions, sodium benzoate breaks down into benzoic acid. While the beverage industry says significant efforts have been made to curb the formation of the carcinogen in products, the researchers in the new study say that their findings "indicate that the problem of benzene formation in soft drinks is still a valid topic". What is more, their analysis suggests that there may be other factors at play besides sodium benzoate and ascorbic acid, such as packaging decisions, and (although requiring further study) shelf life and storage conditions. The researchers, from the Belgian Federal Agency for the Safety of the Food Chain, used a headspace gas chromatographic-mass spectrometric method to measure the amount of benzene in 134 samples of soft drinks sold on the Belgian market. They used this method because the samples do not require much preparation. They were just shaken gentle for 30 seconds to mix them up and, in the case of carbonated beverages, degassed with the application of sound for about a minute and a half, until all the bubbles had disappeared. Ten samples were found to have benzene levels above the European limit for drinking water, 1µg L-1. The drinking water level is the most commonly used reference, since no legal limit for beverages in soft drinks exists. However an action level of benzene in soft drinks of 10µg L-1 has been suggested by the Standing Committee on the Food Chain and Animal Health of DG Health and Consumer Protection. One of the soft drinks samples included in this study exceeded this proposed action level. The study does not reveal the identity of the product. The team found that 33 per cent contained no detectable benzene at all, and 47 per cent contained only trace amounts below the quantification level for the method. A four-way ANOVA test was used to test the role of different factors in benzene formation. In addition to the benzoic acid (sodium benzoate) and ascorbic acid, they also looked at acidity regulators (such as citric acid and phosphoric acid) and packaging material. They observed that the interaction between the benzoic acid and the acidity regulator, and ascorbic acid and the acidity regulator, also seemed to play a role. However only one sample analyses used benzoic acid and ascorbic acid but no acidity regulator, so statistical evaluation was not possible. A statistical analysis was made against benzene levels and type of packaging. "This test indicates that significantly higher concentrations of benzene were detected in soft drinks sold in plastic bottles than in canned soft drinks or in those commercialised in glass bottles," wrote the study authors. In addition, 12 of the samples were found to actually be of the same soft drink, all produced by the same manufacturer and all bottled in glass. However, they had different batch numbers and were sampled by different people, on different days and in different locations. The results were wide ranging, but that with the highest benzene content was seen to be the one closest to the expiration date. However, not enough data were available to say for sure that expiration date has a part to play. "Other factors, such as storage conditions, may also play a role and need to be further investigated," said the researchers. Source Journal of Agricultural and Food Chemistry DOI: 10.1021/jf072580q "Monitoring the benzene contents in soft drinks using headspace gas chromatography-mass spectrometry: A survey of the situation on the Belgian market" Authors: Christof Van Poucke, Christ'l Detavernier, Jan Van Bocxlaer, Rudi Vermeylen, Carlos Van Peteghem.