Measuring the ash content is one of the essential tests of any cereals laboratory. But what actually is this ash?
Ash is what remains when a sample of flour has been completely incinerated. It is mineral material. It is present in very small quantities (between 0.3% and 1.8%) in flour.
So why worry about it so much?
Because it turns out that if you cut a grain of wheat down the middle and measured the ash content in all places, you would realize that the floury center is very low in ash (0.3%), while the peripheral part (the bran) is much richer in ash (10 to 20 times more than in the center). So ash content is associated with bran particles, and is a way of measuring the bran contamination of flour. Or, to put it more positively, a way of assessing a flour's purity.
Because, for various reasons, particularly enabling everyone to have access to white flour, history has made it so that regulations oblige millers to sell their flour according to "types" based on the ash content (see Table 1 -French example-). Here we can clearly see the relationship between extraction rate and type (ash content) of flour. What we don't see is the economic impact. But it is easy to understand that the manufacture of Type 45 flours, with an extraction rate of only 68%, generates a production cost that is logically much higher than for a flour extracted at 94%.
The importance for millers is therefore such that it quickly created the need for suitable methods of measuring ash content as precisely and accurately as possible.
Let us start by looking at the benchmark measurement method. It requires one to:
- Accurately weigh a dry balance scoop (we use quartz, or better, platinum).
- Place between 3 and 5g of flour in the balance scoop.
- Weigh the scoop + flour precisely.
- Place them in an oven at above 550°C (sometimes 900°C), either for a set time or until the weight becomes constant (this can take several hours)
- Weigh scoop + residue
- The difference between the mass of the scoop + residue and the scoop mass, compared to the initial flour mass is the ash content.
This incineration-based method is internationally recognized (with small variations), but it remains long and quite restrictive. There is therefore a strong demand from producers and users of flour to have access to methods for determining ash content that are simple, rapid and precise.
When you combine simplicity, speed and precision, the near-infrared spectroscopy (NIRS) method appears to be the alternative of choice. At this stage, however, it is important to note that the mineral materials that make up the ash have no molecular bonds that respond to conventional near-infrared measurement. Thus, they are inert to the wavelengths of light used in NIRS, which makes their rapid measurement physically impossible. Fortunately for millers, where there is more ash, there are also more bran tissue components, such as cellulose. The latter being organic, it absorbs specific wavelengths in NIRS.
But let us be very clear: we do not measure ash in near-infrared spectroscopy. We measure one (or more) organic compound(s) that are correlated with the presence of ash. NIRS measurement is, by nature, an indirect measurement; in the case of ash, it is doubly indirect.
The challenge is all the more difficult because the incineration-based reference method is not the simplest, and it requires great precision of execution in order to obtain repeatable results. This has a direct impact on the quality of prediction models.
Consequently, hoping to obtain more precise results with near-infrared versus the reference method is utopian. Under the conditions described, and when we really understand what “measurement” of ash by NIRS is, obtaining measurement accuracy equivalent to or very slightly higher than that of the reference method already proves to be a tour de force. Asking for performance superior to the reference method supposes a lack of knowledge of the fundamentals of measurement. The worst part is that such results can sometimes be obtained on particular ranges of samples, but this does not stay the course. Let's be reasonable: getting results equivalent to the benchmark method by NIRS is already an extraordinary thing in itself.
It is interesting to see the current trends towards the consumption of products which are increasingly rich in bran. Numerous studies have shown that the consumption of over-refined products (too-white flour) is not necessarily good for health. We need to eat more fiber, and bran is a very good source of fiber. Thus, where previously we tried to avoid the presence of bran, and where we valued the purity of flour, it seems that the tendency is now to prefer flours richer in ash. History is an eternal cycle of renewal.
Table 1: The different types of flour in France