Abstract: | Foamed food products, such as chocolate mousse, ice cream and spread cheese are more and more popular because of their sensorial properties. Their behavior under flow, their stability over time and the perception of their properties by the consumers depend strongly on the gas fraction, the average bubble diameter and the bubble size distribution within the matrix. In this work, the objective is to compare the performances of two different continuous foaming devices. The first one is a rotor-stator unit equipped with pins, whereas the second consists of a narrow annular gap unit simulating a scraped surface heat exchanger. Both devices allow to vary the rotation speed between 200 rpm and 1600 rpm and the gas to liquid flow rate ratio (G/L) between 10/30 and 30/30 (mL.min 1)/(mL.min-1). This allows obtaining overrun values in the final product that vary between 33% and 100%. Moreover, the two devices are connected to an on-line image analysis system, which makes it possible to determine the mean bubble diameter (d32) as well as the size distribution within the foamed product. The model matrix is a Newtonian glucose syrup (1 Pa.s) containing 2% (w/w) whey protein (WPI). The influence of the foaming device, the operating conditions and the formulation viscosity were analyzed using the Weber number. Concerning the gas incorporation, the experimental results showed that both devices allowed a total incorporation of the gas phase with an overrun of 100%. However, when G/L = 30/30, the column made it possible to obtain total gas incorporation only for 800 rpm. In the rotor-stator, total incorporation was difficult only at high rotation speed for G/L = 30/30. For bubble size, results showed that d32 was smaller in the column than in the rotor-stator for G/L = 10/30, but that this difference vanished for higher G/L ratio. In both devices, the minimum diameter obtained was slightly lower than 20 ”m. Lastly, this work highlighted that the dimensionless Weber number remained constant around 0.3 for the both devices, regardless of operating conditions. The results also showed that the energy transmitted to the fluid was significantly higher in the column than in the rotor-stator. Moreover, total energy dissipation, including mechanical friction, was almost identical in the two devices when G was low and for low rotation speeds. Beyond 1200 rpm, energy consumption was higher in the column and this was accentuated for high G/L ratio. In conclusion, the choice of a foaming unit and the operating conditions depends on the desired properties in terms of overrun, stability and bubble size. The energy criterion, often ignored, can also be discriminating. |
