The sugar content of food is of considerable interest and importance in modern western society. Sugar is cheap and we are eating more sugar than we did. The evidence certainly suggests that at least some of the rise in obesity and diabetes we now see may well be significantly influenced by this increase in sugar intake. Therefore, if want to reduce both our use and our reliance on sugar as an ingredient in food, we do need to know how much is present in our food.

At this point it is probably best to just confirm a definition of sugar. In food sugar is not just the stuff that you put in your tea. White table sugar is sucrose, which is just one of the six sugars that are usually considered as being present in macro-amounts within food. The six that we tend to group as “sugar” are sucrose, glucose, galactose, fructose, lactose and maltose. The determination of sugar in food is usually performed by chromatography to identify and quantify those sugars present.

The use of chromatography is likely to be restricted to one of two options. The first is HPLC (High Performance Liquid Chromatography) using a refractive index detector, although alternatives for this detection method are available. The second is ion chromatography using HPAEC (High Performance Anion Exchange Chromatography) with an electrochemical detector. Both of these approaches can work well and both can have their difficulties.

It is at this point that I must admit to something of a bias. The very first chromatography system that I ever used was an ion chromatography unit for sugar analysis, and I have been in love with the technique ever since. Both in terms of its stability and its ability to resolve each sugar into a separate measurable peak within a relatively short time it is un-rivalled. It is also wondrously sensitive, although this is a double edged sword from an analytical perspective as many foods contain a lot of sugar which can overload a sensitive system. The price tag associated with these systems can also be prohibitively expensive, but one has to accept that they do not develop and build themselves. Nonetheless, when one does exploit the latest innovations within HPAEC then the results can be most impressive, not to say sexy. The following is taken from a new system that I have recently set up.

It is a mixed sugar standard with each peak representing 0.1g/100ml of each sugar. The seventh sugar in this mix is xylose which has been added as an internal standard. Each component is fully resolved with highly efficient measures of chromatographic separation. It is therefore possible to very accurately determine the concentration of each component within the test solution.

It is unlikely that there is a more significant sentence in this essay than the previous one. It is therefore the test solution that is the most important factor within our analysis. If an analyst does not fully extract the sugar from a test portion then it does not matter which chromatographic technique is employed, or how clever or expensive the equipment is, the results will be inaccurate.

Regardless of the instrumentation used, when determining sugar content a test portion is usually extracted in water, or a water-based extraction solvent. There are many imponderables within this extraction process. There are simple physical parameters to be considered, such as the force of any mechanical action employed to disperse a sample within the extraction solution; or the duration of dispersal and the temperature at which it is carried out. There is also the behaviour of the test material itself – a raw flour contains enzymes that will start breaking down starch into sugar when warm water is added, potentially leading to high results. There is the handling of the test material – experience suggests that sugar will break down within homogenised wet samples stored for a few days in a refrigerator, which will lead to low results. It is important to take into account all such factors.

Such issues are not unique to the analysis of sugars. Any chemical test requiring a similar extraction process will be comparable. However, with sugar content being part of the legal requirement for the nutritional declaration on a food label it is unlikely that there are many other similar analyses performed in such volumes and in such frequencies and on such varied sample matrices. When a laboratory does marry a rugged and effective extraction process with an appropriate chromatographic system then one can be confident that the obtained results will be really sweet.