HMF (HydroxyMethylFurfuraldehyde) is used as an indicator of heat and storage changes in honey.
While today HMF is used as an indicator of heating or storage at elevated temperatures, it was first used (as early as 1908) as an indicator of the adulteration of honey with invert syrups (syrups of glucose and fructose). Cane sugar (sucrose) is "inverted" by heating with a food acid, and this process creates HMF. However it was quickly realized that heated natural honey also had higher levels of HMF and therefore the interest switched from being an indicator of adulteration, to that of an indicator of heating and storage changes. It should be noted however that high levels of HMF (greater than 100 mg/kg) can still be an indicator of adulteration with inverted sugars.
It should be noted that HMF is not a harmful substance in levels found in food. Many sugar type products (e.g. Jams, Golden Syrup, Molasses etc.) have levels of HMF that are 10-100 times that of honey. Many food items sweetened with high fructose corn syrups, e.g. carbonated soft drinks, can have levels of HMF between 100 and 1,000 mg/kg.
Fresh natural honey can have varying levels of HMF. Normally this is below 1 mg/kg but levels soon start to rise with ambient temperatures above 20°C. It should be noted that temperatures in the beehive can rise to over 40°C during summer months (when the main honey crop is in progress). It is usual for HMF to be below 10 mg/kg in fresh extracted honey. Levels higher than this may indicate excessive heating during the extraction process.
Many countries impose maximum levels for HMF. 40 mg/kg is the maximum level permissible in the EU for table honey. The 1981 Codex alimentarius Standard for Honey mentions HMF as follows:
"1.3 HYDROXYMETHYLFURFURAL CONTENT
The hydroxymethylfurfural content of honey after processing and/or blending shall not be more than 40 mg/kg. However, in the case of honey of declared origin from countries or regions with tropical ambient temperatures, and blends of these honeys, the HMF content shall not be more than 80 mg/kg."
This twofold difference in imposed levels shows that HMF is not a simple subject and that many anomalies arise throughout the varied honey World and key among these is the effect of ambient storage conditions including the effects of storage in the distribution channel after processing and packing in retail containers.
Honey that is traded in a bulk form is usually required to be below 10 or 15mg/kg to enable further processing and then give some shelf life before a level of 40 mg/kg is reached. It is not uncommon for honey sold in hot climates to be well over 100 mg/kg. This is mostly due to the ambient temperatures (over 35°C) that honey is exposed to in the distribution channel.
The time taken to accumulate HMF at varying temperatures has had considerable research. The most important thing to remember is that this is variable. However work by White, Kushnir & Subers in 1964 indicated the following approximate ranges for accumulation of HMF.
Time for 30 mg/kg HMF to accumulate (based on 3 samples)
|30°C||100-300 days||It can be seen that honey held at an ambient temperature over 30°C for 6 months will accumulate many times more HMF than the same honey flash heated to 70°C for 5 minutes and then rapidly cooled.|
|80°C||< 2 hours|
There are 3 main methods in current use for measuring HMF. The oldest of these is "The Winkler" method. This method is a photometric method that was used extensively in Germany. It used to be listed in the AOAC but was dropped some years ago due to concerns regarding the carcinogenic nature of one of the reagents (toluidine). The two methods currently in the AOAC are a direct measurement using HMF's absorbance at 284nm and an HPLC method that uses the same absorbance property. The HPLC method is the most recent method added to the AOAC and is gaining widespread popularity. We use the HPLC method at Airborne.
There are often differences in results between the methods due to unforeseen or unresearched variabilities in some honeys. It is important to use comparable methods when comparing results for specific samples.
Additional References - 107kb text