Physical modelling of ultrasonic degassing in glycerine
Ultrasonic degassing using glycerine, providing a testbed for further study.
degassing of molten glass using glycerine: proof of concept visualised
Glycerine was chosen as it has a similar viscosity to that of molten glass. As the experiments are carried out at room temperature, monitoring the cavitation process is substantially more straightforward. We used two experimental setups where we could visually observe and record the degassing process. By physically modelling ultrasonic degassing of glass using glycerine, we can simplify future process optimisation.
Ultrasonic Cavitation involves applying ultrasonic waves to liquids resulting in the production, growth, pulsation and collapse of microbubbles. Cavitation is initiated once a threshold energy level is reached, which is approximately 0.15 MPa for glycerine. Many thousands of bubbles form. These expand and rapidly collapse, creating high-velocity shock waves and high pressures of several GPa. Localised high temperatures also occur.
Gas bubbles form on the cavitation nuclei and grow by diffusion from the glycerine into the bubble. If the liquid already contains bubbles, then these undergo diffusion growth. Individual bubbles coalesce due to attractive forces between them (Bjerknes and Bernoulli forces). As they grow, their buoyancies increases, and they float to the surface and release the gas to the atmosphere.
Slide the center arrows with your mouse from left to the right and vice versa to see full images before and after ultrasonic degassing.
Videos of the complete process are available on YouTube (please register to watch them): In both setups, bubble growth and bubble transport by acoustic streaming are clearly visible. The acoustic streaming leads to thick bubble cloud formations that eventually reach the surface of the liquid and disperse. In setup 1, cavitation events occurred mainly at the surface of the sonotrode, and the first streamer appeared after approximately 30 seconds. In setup 2, streamers formed throughout the liquid and due to the shape of the container the effects of standing waves, nodal and anti-nodal zones, of high and low energy are observed. In both zones bubbles are seen to move towards each other and coalesce showing significant promise for an improved glass fining process.
Before ultrasonic energy was applied to the samples, they both contained many tiny bubbles making them semi-opaque. Over time, the bubbles were removed, and the glycerine samples clarified. Figure 1 shows the glycerine in experimental setup 1 before and after the experiment. Figure 2 shows the experimental sample and control sample before and after the application of ultrasonics in setup 2. In setup 1, approximately 4.5 litres of glycerine were clarified within ten minutes.
Two experimental setups were used. The first (setup 1) consisted of a 4.5-litre bowl of glycerine containing the ceramic sonotrode. The second (setup 2) was a glycerine filled quartz glass view cell with dimensions H 63 x W 35 x D 10 mm, to which was applied external ultrasonic energy. The second setup also included an identical control sample to which no ultrasonics were applied.
The ceramic sonotrode used in setup 1 has been specially developed to eliminate standing waves at 20 kHz. This eliminates the danger of cracking.
The cavitation patterns were clearly visible in the glycerine in both experimental setups, demonstrating that this physical model is an excellent way to study degassing in viscous media such as molten glass.
Ultrasonic degassing of molten glass has many benefits compared to conventional degassing methods. For example, it is faster, uses considerably less energy, and uses no toxic additives. In addition, physically modelling the process using glycerine provides an ideal way to study further and optimise the process.