Near-wall mixing by free-stream turbulence

When you put sugar or milk into your coffee, you use a spoon to mix them together by creating a series of swirling motions. This is a basic form of turbulence. Turbulence is a whole bunch of these swirling motions of different size and intensity happening at the same time. We naturally have some idea of how to mix our coffee, but the truth is we don't know what characteristics of the turbulence produce the 'best' mixing in general. Understanding this process has significant ramifications for a variety of real-world applications, including pulling CO2 out of the atmosphere and efficient cooling in power plants. In these applications we are focused on reducing the energy input required so that we can reduce their impact on the environment while making technologies that are cost effective for individuals and businesses. Understanding mixing is also important for our global climate system as the oceans store a significant amount of anthropogenic CO2 and the cycle of O2 between the atmosphere and oceans is required to foster life beneath the surface. WallMix focussed on the fundamental fluid mechanics behind these processes. Using the state-of-the-art facilities at NTNU, we conducted experiments in an open water channel and a closed air channel. Both of these facilities had devices called active grids at their inlet. These devices consisted of rods with diamond shaped wings attached to them. Motors spin these rods and diamonds in predetermined patterns to create turbulence as the flow passes through them. The intensity of the turbulence can be adjusted by changing the pattern the diamonds move in. Laser diagnostics were used to measure the flow in the channels to assess the turbulence and mixing. It was found that turbulence that is more intense improves mixing in the flows. However, the reason it improves the mixing is not what we first suspected. We thought mixing would be improved because the turbulence would break up the layers naturally forming in the flow. While that was true, the real reason mixing was improved was because turbulence enhanced the natural diffusion of passive elements carried in the flow, for example dye. After learning this, we performed experiments in a water channel where the water surface was allowed to interact with the molecules in the air, for example, O2 and CO2. We then changed the levels of turbulence in the water while keeping the surface itself relatively flat. We found that we could change the rate at which O2 transferred from the air into the water by up to 30% by just changing the turbulent motions in the water. This is an important finding because global climate models typically treat the rate that gases dissolve into the oceans as a constant, while a 30% change could have drastic consequences on models and predictions. Moreover, for carbon capture systems where we want as much carbon-laden air as possible to interact with the surface that is scrubbing the carbon, we have demonstrated that tailoring the turbulence can result in substantial gains.

OUTCOMES: The direct outcomes of the project include the publication of 7 open-access journals articles, 1 masters thesis, 1 PhD thesis, significant gains in our understanding of the co-evolution of wall-bounded turbulence and external turbulence, knowledge that should shift the paradigm with respect to interfaces and mixing, and the training and development of three highly-skilled research staff. The published work is all open-access so that it is shared with the community. The major breakthroughs are described in greater detail under “Impacts” below. The project also directly trained a successful PhD graduate, a post-doc who moved on to one of the top engineering universities in the world, and an early career academic who has now been promoted. IMPACTS: The long-term changes that will be brought about by this work are centred around two main topics: (i) identification of interfaces within turbulent flows; and (ii) a broader acceptance of the significant influence turbulence can have on mixing and transport processes. It is common today to equate the a turbulent-turbulent interface (TTI) with the turbulent-non-turbulent interface (TNTI) in how it is detected. Our results clearly demonstrate this is not the case, and this project should be the basis for systemic change in considering what “important” interfaces exist in a turbulent flow. On the role turbulence plays in mixing processes, this is often considered secondary today. It is known that turbulence helps, but compared to other characteristics, e.g., solubility or interface characteristics, turbulence’s role is believed to be second-order. However, we have demonstrated that turbulence can change the gas transfer rate between water and air by 30% by changing nothing other than the turbulence in the water, i.e., no effects on the chemistry. Gains of 30% can be substantial in chemical and thermal processes, and in our climate systems misrepresenting the rate at which gaseous species transfer between the oceans and the atmosphere by 30% can be catastrophic.

Turbulence is omnipresent in engineering flows of practical interest, yet it remains one of the last unsolved problems in classical mechanics. This project seeks to illuminate how turbulence facilitates mixing in a channel. The fundamental research question is straightforward, the possible applications are extensive, and the answer lies in a thorough understanding of the underlying fluid mechanics, making this an ideal basic research project. The research question is: if flow is moving through a channel, how does one mix a substance released in the centre of that channel to the walls in the shortest distance from the release point. The solution lies in a synergy of three areas of contemporary research in fluid mechanics: (i) the generation of bespoke turbulent flows, (ii) instantaneous turbulent structures, and (iii) scalar transport in turbulent flows. Understanding this basic phenomenon has far-reaching repercussions for all in-line mixers. For instance, fluid cooling systems rely on cold fluid being transported toward a surface and warm fluid being transported away. Optimising this process will result in smaller reactors and less energy consumed in the cooling process. Measurements will be conducted in the new water channel at NTNU using simultaneous particle image velocimetry and laser-induced fluorescence. A range of turbulent flows will be generated with an active grid and by the end of the project a conceptual model will be produced that describes the optimal parameters to transport a scalar to the walls of the channel in the shortest distance. In addition to the PI, the project will employ a PhD student and a post-doctoral fellow. It will also involve international collaboration with partners in the UK.