Scientists seek answers to food production problems
Many factors can compromise the quality and/or safety of a food product. But one of the most common scenarios that faces a manufacturer is an unexpected problem that comes with a change of supplier.
The irony is that everything seems fine when the ingredient is received. It meets its specification (and this is similar to the previous supplier’s), yet nonetheless, the finished product ends up being far from satisfactory. The problem might be one of flavour, appearance, performance or smell, but whatever it is, the challenge is to find out what has changed as a consequence of changing the supplier, and to take steps to put things right.
Analytical approaches to problem solving
Some problems stand out such as a bad smell, the wrong colour, failure of performance, or presence of foreign bodies. Other problems may not be so obvious e.g. microbial spoilage, adulteration, inauthentic ingredients.
However, whether obvious or not, every problem must be investigated. There are a range of analytical techniques to help find out what has gone wrong and to inform what needs to be done to correct it.
Few manufacturers will have all of these techniques available on-site, so are likely to need to outsource some or all of the investigative work to a specialist laboratory. It makes a lot of sense to find a laboratory to partner with well in advance of a problem occurring, so that they are better able to assist when a problem arises!
Blemishes and discolouration may be the result of chemical contamination or perhaps microbial spoilage; their investigation will likely involve the chemical techniques referred to in the discussion of taints (below). Other cosmetic/aesthetic problems such as sediments or a visible separation of emulsions, will be investigated by the physical techniques applicable to performance problems (also below).
Foreign bodies are a separate case. They can be broadly classified as extrinsic or intrinsic: The former come from external sources either by deliberate or accidental means. Intrinsic foreign bodies include ingredients such as a leaf or stalk, or an ingredient in an unusual/unexpected state.
A broad spectrum of technologies are routinely used in foreign body investigations, but simple light microscopy is often the starting point. It can be used to determine features that are typical or characteristic of the likely source, thereby directing scientists to the sophisticated methods that will provide confirmation.
Different techniques are required for different types of contaminant. For example, a scanning electron microscope fitted with an energy dispersive X-ray (EDX) detector is useful for identifying the elemental composition of items such as metal fragments. Plastic fragments, which may look like glass, require confirmation using a technique known as Fourier transform - infra red spectroscopy (FT-IR).
An inappropriate taste or smell (or discoloration) in a product is usually due to the unexpected/unwanted presence of any number of chemicals. Or there may be an absence of key ingredients (e.g. bitterness blockers), or failure of a taste-masking system (e.g. encapsulation of oils), that might otherwise be expected to keep the unwanted flavours in check.
Taints can come from direct contamination, packaging or the airspace in a bottle/jar. The unwanted chemical may come from microbial spoilage, or perhaps from reaction of ingredients within the product.
Isolating and identifying a tainting chemical is often a challenge because certain taints can be very potent at very low concentrations (e.g. 20 parts per trillion). It depends on the product they are found in but chemicals like halo-anisoles and benzene are especially potent. Whilst certain groups or families of chemicals give rise to specific flavours/odours that an experienced chemist will recognise, there are thousands of potential tainting chemicals in widespread use, so there is never one obvious candidate.
A taint or off-flavour can originate at any point in the lifetime of a product, from production of the raw materials through to eventual consumption. Hence it may be necessary to trace the source of a taint back to any point in the supply chain.
Investigations will usually involve an initial assessment by human senses. If the offending chemical is believed to be organic (i.e. carbon-based) there will follow a chemical extraction procedure, separation via chromatography (either liquid or gas) and identification of the chemical using mass spectrometry or possibly nuclear magnetic resonance (NMR) spectrometry. A metallic contaminant will be detected using inductively coupled plasma (ICP) mass spectrometry or ICP-optimal emission spectroscopy.
This simple description belies the complexity that might be involved. As noted above, taints can be present in ppt concentrations. There is huge skill involved in extracting the taint, at detectable levels, free of other chemicals that might interfere with the analysis, and (in the case of volatile odours) without simply losing the chemical to the atmosphere before it can be identified.
Problems with the flow of liquids and semi-solids can be investigated using rheological instruments. These instruments apply different shearing forces which can mimic how materials might flow against different surfaces and internally. Such measurements will inform manufacturers of potential processing problems (e.g. when pumping) as well as problems with the behaviour of finished products. The resistance to flow by powders within hoppers, for example, can lead to clogging, or an accumulation of out-of-date material. The propensity for powders to behave in a disadvantageous manner can be predicted via a combination of powder rheology (Shear cell), particle sizing and microscopy.
Performance issues often come down to a problem with particle sizes or shape, as with the instability of an emulsion, or the gritty mouth feel of a product. This can be investigated using laser diffraction instruments, often in combination with microscopy techniques. The size and shape of particles can also be measured using static image analysis, which is a useful alternative to manual microscopy.
It is sometimes possible to observe how ingredients are distributed within a product by using X-ray tomography techniques along with the scanning electron microscope. This can give useful information about why a product performs badly. Even simple light microscopy, used with staining procedures can identify some of the microstructural features that are giving rise to performance problems.
The modern laboratory also has access to a range of instruments that can load products with crushing or stretching forces to investigate structural weaknesses (or strength).
Customers may attribute illness to a particular food item that shows no other signs of being at fault. Clearly, a manufacturer will want to investigate a complaint of this kind, with the possibility of allergen or pathogen contamination being a particular cause for concern.
Microbiological testing can address the latter, however allergen testing can be more complicated. As such, much depends on the nature of the complaint and the likelihood, or otherwise, that specific allergens might have come into contact with the product or its ingredients/packaging.
Tests for allergens usually rely on enzyme-linked immunosorbent assay (ELISA) techniques directed at specific proteins, or DNA techniques that can detect trace amounts of DNA associated with allergenic ingredients. In both cases, but especially with ELISA, it is important to be aware of the potential for interferences inherent to biological assays that can lead to false positive or false negative results. A laboratory must have robust protocols in place to reduce the risk of any false results.
High profile scandals (horsemeat in beef, melamine in milk) remind us that the food supply chain is extremely complex and global, and tracing the origin of ingredients is near impossible. Moreover, not every player in the supply chain is necessarily honest or legitimate.
Whether there has been a specific incident, notification of a wider industry-concern or merely a desire on behalf of the manufacturer to protect their interests (perhaps with a new supplier), there are analytical approaches that will assist in determining the authenticity of a particular supply. However, this is very much dependent on the ingredient in question and authenticity is not always easy to prove.
Testing for authenticity in the case of meat and fish is relatively routine, using DNA methods that can target gene sequences found in one species but not in another. However the quantification of cross contamination in meat against the level set by the FSA still requires specialist knowledge and testing.
Other authenticity issues are more complex. For example, olive oil has well defined acceptable ranges for a variety of naturally occurring compounds giving the analyst a set of parameters that can be measured to assess authenticity.
Other techniques such as isotope ratio analysis have become established but these depend on building a large database of samples, at considerable cost. So new approaches such as non-targeted screening (NTS) are now being applied. Rather than looking for specific chemical compounds, analysts are applying chemo-metric or food-omic approaches to identify differences between samples and a dataset of known authentic ingredients. This 'probabilistic' approach to testing is useful for identifying 'outliers', allowing analysts to focus on the suspect samples more easily, and to devote time to more careful scrutiny of these.
Where adulteration is suspected, new methods may need to be developed as a matter of urgency to address the specific problem. In the melamine in milk example, a nitrogen-rich chemical had been added to milk to fool a test that judged quality by nitrogen content. That particular test had no means of detecting the source of nitrogen, and an entirely different way of testing milk was needed for this fraud to come to light.
It is impossible to give a definitive list of problems and solutions, even in a very broad sense. Each problem needs to be assessed and investigated on its own merits.
It is also impossible for every manufacturer to equip itself with the resources and expertise that would be needed to investigate every potential problem. Rather every manufacturer should seek to partner with an expert laboratory that understands its processes, and can work with them to highlight potential weaknesses, and understand vulnerabilities.
At least this approach will ensure that when a problem does occur, whatever it happens to be, there is help at hand to identify what has gone wrong, and recommendations on what needs to be done to put it right.