Accelerating Food Analysis

Luigi Mondello is an analytical chemist who is intensely involved in the development of comprehensive chromatography instrumentation and methods. The combination of these chromatography approaches with powerful mass spectrometric devices provides the tools that his team and researchers around the world use for the detailed study of samples of high scientific interest. Luigi is as committed to discovering new information on the composition of the world surrounding us as he is about the development of new analytical strategies.

Peter Q. Tranchida is a food chemist, with a great passion for separation science. Peter is a proponent and practitioner of separating samples in two chromatography dimensions - and often adds a third, mass spectrometric, dimension. Such powerful methods provide new insights into old samples. “After each analysis,” Peter says, “I feel like a child opening up a Christmas present.”

Contamination of foodstuffs with petroleum-derived mineral oils is, unfortunately, a common occurrence. Mineral oil saturated hydrocarbon (MOSH; linear and branched alkanes, naphthenes) and aromatic hydrocarbon (MOAH; mainly alkylated) contamination have both been described in numerous investigations. In rats, bioaccumulation of MOSH can lead to the formation of microgranulomas in the liver (1). A limit of 0.6 mg/kg, for MOSH (from n-alkane C10 to C25) contamination in foods, has been recently proposed by the German Federal Ministry of Food, Agriculture and Consumer protection (BMELV). 

Over two decades ago, Grob and co-workers pinpointed lubricating oils, release agents and mineral batching oil used for the production of jute as sources of contamination (2, 3). 

Contamination of foodstuffs with petroleum-derived mineral oils is, unfortunately, a common occurrence. Mineral oil saturated hydrocarbon (MOSH; linear and branched alkanes, naphthenes) and aromatic hydrocarbon (MOAH; mainly alkylated) contamination have both been described in numerous investigations. In rats, bioaccumulation of MOSH can lead to the formation of microgranulomas in the liver (1). A limit of 0.6 mg/kg, for MOSH (from n-alkane C10 to C25) contamination in foods, has been recently proposed by the German Federal Ministry of Food, Agriculture and Consumer protection (BMELV). 

Over two decades ago, Grob and co-workers pinpointed lubricating oils, release agents and mineral batching oil used for the production of jute as sources of contamination (2, 3).  Later, it was discovered that mineral oil can be transferred to dried baby foods from the ink printed on cardboard containers (4) and recent research has further highlighted MOSH and MOAH in foods packaged in cardboard (5). MOSH have even been found in a high proportion of human milk samples [(6) and continue to enter babies bodies throughout weaning, in the form of homogenized baby foods (7). 

Most studies have used rather complicated, time- and solvent-consuming multidimensional liquid-gas chromatography (LC-GC) methods, with flame ionization detection (FID). It is well known that such applications generate humps, defined as unresolved complex mixtures (UCM). No identification information is usually reported on the composition of these UCMs, which is regrettable since they may be important to define the presence of interferences, toxicity and the source of contamination.

With interest in, and awareness of, the toxicological presence of mineral oil in food increasing, we believe that faster, more environmentally-friendly and higher-resolution methods are required, including detection using mass spectrometry. 

Previously, we developed a fast LC-GC-FID method in a study on vegetable oils. This had a run-to-run time of 14 min (8), used a programmed temperature vaporizor as the GC sample introduction system and consumed limited quantities of organic solvents (hexane/CH2Cl2). Disadvantages? We wouldn’t describe the approach as environmentally friendly. In addition, the method suffered from the “blindness” of the FID; however, even an MS system would be challenged by UCMs. 

For baby-food, we initially employed this optimized fast LC-GC-FID method.  Once MOSH contamination was encountered and quantified (levels up to 2 mg/kg were found, considering the BMELV proposed hydrocarbon range), we generated more detailed information using an off-line LC-comprehensive GC (GC×GC)-MS approach. Cryogenic GC×GC, a high-resolution two-dimensional technique, was performed by using an apolar-polar column set. The MOSH components of the UCM found in baby food were nicely separated on the two-dimensional chromatogram: linear and branched alkanes eluted along the same line (separated in the first dimension), while the naphthenes were resolved in the second dimension. At that point, interpretation of the mass spectral data became much easier. 

Of course, an off-line method cannot be defined as rapid but there is no real technological obstacle in the construction of an on-line LC-GC×GC-MS instrument. Ideally, such instrumentation would include an FID unit, with the effluent directed to both detectors. This would provide rapid (fast GC×GC methods can be developed), high-resolution, and information-rich analysis of food contamination by mineral oil. What remains a problem is how to reduce or eliminate the consumption of toxic organic solvents.    

Newsletters

Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.

I have read and understand the Privacy Notice*

Subscribe