Further Description of the Two Experimental Systems

---

The opposed jet mixer

The opposed jet mixer has some unique attributes that make it very attractive for simulation. There are several reasons for the selection of the opposed jet geometry (see slide 14). Two of these are associated with unaddressed problems associated with DNS calculations in the past on channel, pipe, and boundary layer flows. The first problem is the boundary conditions at the entry to the flow. In the standard approach, the inlet turbulent conditions are unknown and one assumes that the inlet conditions are the same as that obtained from the computation at the exit. This is called cyclic boundary conditions. The second problem is associated with the initial turbulent field in the entire volume. What is sometimes done is to use a representation of the average turbulent velocity profile with an added random component to represent the turbulence. In contrast, the opposed jet system operates with laminar flow inlets, even for conditions where the jets are unstable.Cyclic boundary conditions are not needed. To solve the problem associated with the initial conditions of the field, we have the full-field, time-resolved velocity vectors measured and can give the computation experimentally measured initial conditions. Finally, there are no rotating parts, as in an impeller mixer, that would require the used of a sliding mesh grid.

Our contributions so far have been to obtain detailed flow field measurements for this system over a range of jet Reynolds numbers [Brodkey and Zhao, 1998]. As an example of the type of results, the next visual provides a full 3-D view (21³) of the long-time average (9-minutes) of the two opposing jets at jet Reynolds number of 200 and 4000 (see slide 17). Even with the inlet jet at 200, the flow in the vessel is very unstable. Results have been obtained up to a jet Reynolds number of 5000. Many dynamic examples are also available on our internet pages. For example, there is a sequence showing the vertical or z-vorticity and another that shows the central plans of the experimentally measured time dependent velocity field (see slide 19). Particle paths in the time dependent flow field are also shown (see slide 21),

We have also made progress in the validation of computations for the opposed jet system for much lower Reynolds numbers than shown here [3rd article: Unger et al., 1999]. We compared the full three-dimensional flow field by comparing the probability density functions from both experiments and DNS simulations for laminar, steady-state conditions and for the transitional region. The results are remarkably good under laminar flow conditions that exist for jet Reynolds numbers of 80 and below. For the fully turbulent regime, full DNS and LES computations have been done. Some results are available for this in the next section on mixing computations. Our current effort is nearly finished and is to determine the time (number of computational steps) needed to obtain a statistical steady state so that meaningful measures can be obtained. The next effort will be to try to duplicate actual experiments (still to be completed with the new mu-tech imaging system) by using the experimental measures for the initial conditions.

Some summary results have been published; however, the reporting of time variant, 3-D data fields leaves a lot to be desired for flat page journals that prefer to use black and white. Additional work on the representation of such information by using the Internet is in progress. Our key driving force is simplicity and cost of representation. The same procedures would apply for results from direct numerical simulations (DNS).There has been considerable study of the ability of humans to perceive 3-D objects when observed under oscillatory motions. This is done without using glasses. A very simple stick representation can illustrate the effect.The effect is dramatic! We are also actively working on improved means of representation of such data field; however, we will only provide a link to the work that is in progress and not discuss any details here.

Standard mixing vessels using a rotating frame of reference

Finally, the most difficult mixing vessel simulation and experiments will require extensions of current research. Each of these is underway. Our experiments on mixing will utilize a convective view that will allow measurement of time-resolved, three-dimensional velocity vectors in a mixing vessel, even at high Reynolds numbers. More fundamental and local parameters (e.g., local turbulent kinetic energy) can be obtained that describe inhomogeneities that are of importance in mixing vessels and can be used to describe trailing impeller vortex structures, baffle-fluid interactions, etc. Such measures can be contrasted with overall global parameters as the power per unit volume. Local motions must be used to allow prediction of mixing, especially where selectivity is of importance. These measurements must be made under true dynamic conditions where superimposed larger scale motions can influence finer scale mixing processes. To allow analysis by computational means, the convected view will be implemented. This will require that the wall and baffles be part of a moving grid. The current sliding grid has the impeller rotating.

It is known that incomplete mixing often has strong detrimental effects on the evolution of chemically reactive processes: desired reactions are slowed and sometimes halted before reaching completion, undesired reactions are enhanced, and product selectivity is affected. We know that oscillations can be used to achieve faster mixing for low Reynolds number flows and can enhance turbulent boundary layer heat transfer. For more complex laminar flows and for turbulent mixer flow, the vortex structure that exists along the trailing side of the advancing mixer blades must be modified. Simple changes in impeller speed might be sufficient; however, more likely, dynamic changes in flow or geometry will be needed. Our goal is to identify mixing regimes that yield and reduce waste by-products for chemical reactions. Dynamic flow perturbations will be investigated to elicit global chaos and enhance mixing performance throughout the vessel.In conjunction with others, parallel theoretical developments will be investigated to test the validity of such procedures and to provide a basis for scale-up and design.

Our unique experiments will be to characterize the full-field, time-resolved, velocity vector information in the mixer from a rotating view, locked with the impeller using an existing large rotating table system. By using a rotating view, the rotational convective contribution from the impeller is removed. Thus, the filming rates can be reduced to the point where video technology can provide satisfactory results without blurring images and our PTV technique will be adequate. The three-dimensional stereoscopic measurements will be obtained by using our most recent PTV system that can provide a 60 Hz data rate at a nominal 512²resolution. This system can be run at 120 Hz at half resolution, if needed. The rotating frame of reference system for measurements on a standard mixing vessel is a mechanically complex unit.