Following the reading “Steaming milk between false myths and reality: proteins” many have asked me what changes lactose undergoes during milk steaming.

This is because one of the most repeated phrases during coffee courses is “Don’t exceed 70°C when steaming or the sugars will caramelize resulting in perceptible bitterness in the steamed milk”. A phrase now taken for granted by many baristas and just as many industry experts.

But is this really the case? Are there scientific foundations regarding this statement or is it simply an urban legend passed down from generation of baristas to generation?

We have finally reached the third chapter of the series of readings concerning the chemical modifications that occur to milk components during steaming. For those who missed the previous chapters, I recommend starting with “Chemical composition of milk“.

Today with the support of experimental chemist Salvatore Impemba, will we uncover yet another legend or reaffirm a truth?

Let’s start reasoning about some data derived from scientific research regarding lactose, the main sugar in milk….

What Happens to Lactose when it is Heated?

As we have seen previously, lactose dissolved in milk has been found in two forms called α and β which possess the same nutritional properties. At room temperature (about 20°C) there is a predominance of the β form (63%), while at temperatures above 90°C the α form prevails.

When milk is heated, reactions occur between the terminal amino acid portions of proteins and the -COOH groups of lactose. This phenomenon, called lactosylation, begins at 37°C and increases constantly with increasing temperature (6).

During heating, lactose undergoes reactions that can have important consequences on milk: depending on the amount of heat and time to which it is subjected, it changes in odor, color, nutritional values and acidity.

As we have seen, with thermal treatment of milk, lactose can react with the amino groups of proteins (in milk mainly lysine). With the reaching of high temperatures (above 100°C) what is called the Maillard reaction occurs (the same one that leads to the browning of bread, coffee, meat etc…), a complex series of chemical reactions in which the quantity and type of products formed depend on various factors including the pH of the milk, temperature and duration of treatment (1). What is certain is that the chemical composition of milk and its organoleptic properties change due to the formation especially of carbonyl and aromatic compounds.

In the first phase there is the reaction between lactose (the sugar in milk) and the terminal amino groups of proteins (amino groups located at the ends of protein chains) or with free amino acid residues.

Following the first phase there is the formation of Amadori compounds, although this phenomenon does not lead to major changes in coloration and taste, the greatest consequence in this phase is the loss of availability of lysine or free amino acids such as cysteine.

If the reaction continues, the advanced phase of the Maillard reaction is reached with the breakdown of Amadori compounds, here different reaction routes can be obtained with the formation of a wide range of products, some of which are still unknown. In this phase there is a modification in the taste and aroma of milk.

The final stage of the reaction involves the formation of melanoidins (compounds with characteristic brown pigmentation) with tastes and aromas that can be negative such as bitter taste and aromas of burnt, solvent, rancid, sweaty, cabbage or positive such as malted, similar to bread crust, caramel, roasted coffee.

The more the Maillard reaction advances, the less lactose and lysine are available.

New lysine residues are however available as heat degrades proteins, of which lysine is one of the amino acids that compose them.

Since in this case the lysine residues are released from the amino acid chain (protein), to react with lactose it is no longer necessary for them to be at the end of the protein itself.

Studies have shown that heating milk at 120°C for about 6 minutes leads to the release of 3.6% of lysine, while heating at 130°C for 5 minutes leads to the formation of 6.8% of lysine residues. Indirect heating, for example in a water bath, at 115°C for 40 minutes, allows milk to increase the bioavailability of lysine residues reaching up to 13%.

In addition to the Maillard reaction, lactose can undergo reactions that involve its degradation into glucose and galactose (its components).

In light of all this, can lactose really caramelize during the milk steaming process if the 70°C threshold is exceeded?

To answer this question we must first understand what we are talking about when we say “caramelization“.

Both caramelization and the Maillard reaction are non-enzymatic browning reactions.

Caramelization occurs at high temperatures and is strongly influenced by the presence of sugars, water elimination and oxidation. It occurs at high temperatures (about 150°C although some sources report for lactose temperatures slightly above 200°C), in the presence of little water and a high amount of sugar. Caramelization does not involve amino groups.

Imagine putting sugar in a pan and turning on the heat, gradually the sugar will visually tend to “melt” and acquire a brown color transforming into “caramel” through caramelization reactions.

From this definition we can understand that Maillard reactions can be confused at a macroscopic level with “caramelization”, as they too can lead with the prolonging of time to the typical browning of caramel. (5)

Excluding caramelization, we must therefore ask ourselves if Maillard reactions could actually occur during milk steaming once the fateful 70°C is exceeded.

For this we can observe the times and temperatures of lactose degradation studied regarding thermal treatments of milk, as in the presence of these reactions an inevitable decrease (due to degradation) in total content must be observed.

A first study found no evident differences in lactose content in samples of raw milk, milk that was brought to 71.7°C for a time of 15 seconds and milk that was brought to 135°C for a time of 2 seconds (6). A second study detected a minimal variation in lactose content (about 0.5%) in milk treated from 135 to 144°C compared to untreated milk. While a third study found no statistically significant differences in lactose content between raw milk, pasteurized and UHT (7).

These studies, in agreement with each other, confirm that it is unlikely for the times and temperatures involved in milk steaming (significantly lower than those reported above), to think that there could be lactose degradation such as to cause the slightest concern for the barista.

The answer to the fateful question “Can lactose caramelize during milk steaming, once 70°C is exceeded?” based on research and studies conducted leads us to a unanimous answer: “No“.

With the next and final reading of this cycle we will explore the changes that occur to fats when the barista steams milk.

BIBLIOGRAPHY:

1- I. S. Fagerson, Thermal Degradation of Carbohydrates, J. Agr. Food Chem., 1969, 17, 747-750; B. M. Metha, H. C. Deeth, Blocked lysine in dairy products, Comprehensive reviews in food science and food safety, 2016, 15, 206-218; N. Stanciuc, G. Rapeanu, S. Stanciu, Quantitative evaluation of colour development in milk model systems during heat treatment: a kinetic study, Romanian Biotechnological Letters, 2010, 15, 5331-5341.

2- P. A. Finot, R. Deutsch, E. Bujard, The extent of the Maillard reaction during the processing of milk, Prog. Fd. Nutr. Sci., 1981, 5, 345-355.

3- T. Henle, H. Walter, H. Klostermeyer, Evaluation of the extent of the early Maillard-reaction in milk products by direct measurement of the Amadori product lactuloselysine, Unters Forsch, 1991, 193, 122-199.

4- F. J. Morales, C. Romero, S. J. Perez, Fluorescence associated with Maillard reaction in milk and milk-resembling systems, Food Chemistry, 1996, 57, 423-428.

5- M. Van Boekel, Effect of heating on Maillard reactions in milk, Food Chemistry, 1998, 62, 403-414.

6- Pinto G. et al., Lactosylated casein phosphopeptides as a specific heated milks (2011)

7- Pestana J et al., Effects of Pasteurization and Ultra-High Temperature Processes on Proximate Composition and Fatty Acid Profile in Bovine Milk. (2015)