Inline Mixers for Chemical Processes
Inline mixers can be defined as a configuration where the materials to be mixed are flowing continuously through the device. Residence time is generally short, on the order of seconds. Oftentimes, plug-flow behavior (minimal back-mixing) can be approached. Inline mixers are used because of the general compactness and simplicity of the equipment relative to tank mixing. With stirred or jet-mixed tanks, inline mixers are often used on recirculation lines to pre-mix an additive before mixing into the tank bulk.
“Mixedness” is used to quantify mixing extent. It is defined as one minus the coefficient of variation (COV). In this context, COV is defined as the standard deviation of a concentration (the minor component) divided by the average concentration after complete mixing. The standard deviation is taken over a cross-sectional plane of the pipe. For turbulent systems, the standard deviation of concentration based on time fluctuations at a point can be used. The average concentration is equivalent to the ratio of minor feed rate to total flow rate. This may not be exactly true if there is a large specific volume change due to mixing, but is generally adequate.
At a mixer exit, a mixedness of at least 95% (COV < 0.05) is considered “well mixed” for non-reacting systems. For reacting systems there are conditions where the rate of mixing, not just the mixedness at the mixer exit, is critical to yield and selectivity. This was outlined in a previous blog Mixing with Reaction; Practical Considerations – Becht.
To the right is a cross-sectional image of the outlet piping from a confined exit swirl nozzle in which dye had been injected at the inlet. The confined exit prevented atomization and introduced a large backflow vortex near the center. The image was acquired using Planar Laser-Induced Fluoresence (PLIF). It is an example where the mixedness was far less thant 95% in the exit piping.
The types of inline mixers available are among the categories discussed below. Note this list is not comprehensive but represents the most common mixers encountered in my 35+ year career.
- No Mixer (pipe turbulence alone)
This option is most desired for very turbulent systems where the components to be mixed are of similar viscosity. As a rule of thumb, if the Reynolds number is at least 100,000, then 100 pipe diameters length is sufficient to mix two streams to at least 95% mixedness. This option is not to be used for mixing-sensitive chemistries, as the mixing is relatively slow.
- Pump Mixing
Oftentimes centrifugal pumps are utilized for mixing, where the minor component is injected near the suction side of the pump. The best mixing, unfortunately, is well away from peak efficiency point of the pump, instead at high head and low flowrate. As in the “No Mixer” case, this option is not to be used for mixing-sensitive chemistries. The residence time is generally low inside of a centrifigal pump, and mixing is not completed at the pump discharge; some additional pipe residence time is required to complete the mixing process. Positive displacement pumps, such as diaphram pumps, will give some mixing but complete mixing at the discharge cannot be expected. (The check valves are a souce of mixing in addition to the limited mixing in the diagragm fluid chamber.)
For high-viscosity systems, gear pumps or special pumps such as progressive cavity pumps are utilized, but very little mixing is expected in these. See Figure 3, showing glycerine being pumped as the bulk flow and a small “rope” of aqueous dye introduced continuously at the centerline. The two flows are completely miscible.
Although Figure 3 shows some slight radial spreading of the dye, it is far from completely mixed.
- Static Mixers
These are a common method for inline mixing. See Table 1 for a partial list. For turbulent systems, mixing can be completed in as little as 4 twisted tape elements, i.e. less than 8 pipe diameters. Laminar systems may require 20 or more elements for twisted tape. These elements are commonly used to enhance heat transfer, in part because the elements conduct heat to/from the wall to provide additional heat transfer area. They are the most versitile static mixers in the sense that they work under laminar, transitional, and tubulent flow conditions.
Another type of element consists of “half moons” that are mounted in opposed positions along the pipe. Like the twisted tape, these elements split the flow into two sections at each element, but additional advantage is claimed because there is impingment on the elements and some backmixing due to flow separation behind the half moons. Backmixing can be important in cases where the rheology is very sensitive to the mixedness. These are most commonly specified for turbulent conditions.
Another class is called baffled mixers, because mixing is attained by deflecting flow. Compared to the twisted tape type, the surface area is higher per unit length of pipe and hence total length for mixing is shorter and the pressure drop is higher.
Finally, there are vortex style mixers that work under very turbulent conditions. There are wall tabs that generate swirling vortices that enable radial mixing. The main advantage of these types of mixers are the low pressure drop compared to other tubulent options. For very space-constrained cases, an injector is built into a single set of tabs fully mixes in piping downstream.
Table 1. A few types of static mixers.
| Description | Image/Sketch |
| Twisted Tape
Laminar, transitional, and tubulent |
 |
| “Half Moon”
Turbulent |
|
| Baffled
Laminar |
|
| Baffled
Turbulent |
|
| Tab / Vortex Generator
(Kenics® High Efficiency Vortex, HEV) Highly turbulent |
 |
| Compact Tab with Integral Injector
(Sulzer CompaX™, Kenics® UltraTab) Highly turbulent |
 |
- Tee mixers – opposed orientation
For turbulent flows of about the same flow rate, tee mixers with opposed inlet flow are an option that reduces the length of pipe required to complete the mixing. These can also be called impingement mixers. They can be constructed of a simple pipe tee, as shown in Figure 4:
The mixing is completed in the post-exit piping. The two inlet pipes do not need to be of the same diameter e.g. if one flow is smaller within a factor of 10. A static mixer may be used to reduce the mixing length after the tee.
- Tee mixers – side entry
In cases where one flow rate is significantly smaller than the other, a side entry confuguration is used.
For the minor flow, it is often best to use an injector quill (stinger) to introduce it. Commonly, this configuration is used just upstream (1 – 2 pipe diameters) of a static mixer, but can also rely solely on pipe turbulance to complete mixing, with no static mixer.
The reason for use of the injector quill is to prevent “wall-hugging” of the minor flow, which causes the length to mix to be considerably longer than introduction near the centerline of the major flow. Properly sized, sporadic backflow of the major flow into the injector is prevented; sporadic backflow also causes irregular additive flow. Backflow can be particularly harmful if the minor flow is a catalyst, for instance.
- Jets in Crossflow
Jets in crossflow are an effective method for inline mixing, not unlike the side entry tee mixer. The difference is that many jets can be used in parallel to cause the mixing, for cases where both flowrate are large. An example of Planar Laser-Induced Fluoresence images are shown below for a three-jet system. Flournay has done seminal work in design of jet-in-crossflow mixers.
If the injector flow velocity is too low, wall-hugging is experienced similar to the side-entry tee mixer without a injector quill. If the minor flow velocity is too high, the minor flow can can hit the far wall of the pipe, however that is less of a mixing hindrance than wall hugging. Instead, pressure is “wasted” in the injection. Comments on proper design of confined jets is found in.
Below is a sketch of a many-holed jet-in-crossflow industrial-scale mixer, as discussed in its patent. Somewhat counter-intuitively, the major flow is jetted into the minor flow. This mixer has been shown to mix liquids at a rate of thousand of gallons per minute in less than one second.
- Inline Dynamic (stirred) Mixers
These mixers are flow-though devices with a rotating agitator. They can be as small as 1 to 2 impellers or a whole stack of impellers that may differ in design if the fluid (formulation) changes properties when mixed. By exterior pumping, the fluid is forced through the impeller zone(s), where mixing occurs.
 |
 |
Figure 8. Two Types of inline Dynamic Mixers.
Ebara HG (formerly Hayward Gordon) and Dynamix are two manufacturers, among others. Custom fabrication by a local machine shop can also be an option.
For the multi-impeller case, horizontal annular baffles can be used to reinforce compartments and force backflow, i.e., create multiple passes through each impeller.
These mixers are used for challenging problems, such as mixing a viscoelastic minor component into a major flow that is itself non-newtonian.
- Rotor-Stators
These are high-speed devices and are reserved for challenging process operations for immiscible liquids, including making liquid-liquid emulsions and homogenizing miscible liquids of very different viscosities. In some cases, rotor-stators enable new product formulations, in large part due to the small gaps and high shear rates (up to 100,000 s-1).
In my experience, so-called “tooth & slot” rotor stators are the most common design. The working principle is illustrated in Figure 9 (from the perspective of feed flows into the page), and in this YouTube video: Rotor-Stator Kinematica AG. The two feeds enter concentrically through the center portion of the device. They are forced outward into a rotor that is spinning at rapid speeds, e.g. 3600 rpm, although variable speed is common in order to fine tune the droplet size. The liquids then enter a high-shear zone where the outer boundary is stationary and where the droplets are broken up. The multiphase liquid then passes through the outer holes. The average shear is set by the spacing between the rotor and stator as well as the rotor speed relative to the stator.
Figure 9. Rotor-stator sketch. The feed liquids are forced into the rotor-stator from the front of the device in this sketch.
Industrial applications include:
- Production of paints, coatings, and inks
- Consumer goods formulations
- Skin-care products and cosmetics
- Cleaning products
- Pharmaceutical industries
- Formulation of Ag chemicals
- Food processing
There are many variables to consider in design of a rotor-stator, including rotor & stator hole widths, rotor & stator thickness, spacing between rotor and stator, and number of holes in rotor & stator. The vendor should be consulted to assist in selection. Typically, multiple head designs are available and can be easily changed out in the motor unit. Scaling up is not straightforward, and again the vendor should be consulted. Manufacturers include Ika, Ross, Silverson, Greerco, and MXD Process, among others.
Multiple concentric stages are possible, where the multiphase flow leaving a stator radially encounters another rotor and so on.
They can also be used for incorporation of solids, such as incorporating/dispersing difficult to wet powders into liquids, wet milling of solids in slurries, and grinding and dissolution of solids. A funnel containing solids can be used to induct the solids into the liquid.
Not all rotor-stators are inline; they are also utilized in stirred tanks in conjunction with a traditional pumping impeller to provide a function that pumping impellers can’t.
- Further Considerations in Mixer Selection
There are also certain rules of thumb to avoid problems.
- Baffled static mixers have been known to collapse for very high viscosity systems, such as additive addition to polymer melts, with the associated high pressure drop. Consult with the vendors in these cases.
- For miscible systems, mixing a minor low viscosity component into a bulk of high viscosity will extend the mixing length. When the higher viscosity fluid is the minor flow, the mixing is easier/faster. For example: adding a little honey to hot tea is much easier than adding a little hot tea into honey.
- In some cases, such as dissolving a resin into a solvent, the viscosity is extremely sensitive to the amount of resin dissolved. The lack of back mixing in most static mixers means that fluctuations in pumping rates of the two flows causes unacceptable fluctuations in the mixed product viscosity. A consistent product means that additional mixing is required, such as introducing tote-mixing into a product tote, or relying on mixing in the product container as it is transported to the customer.
- For immiscible liquid systems or gas-liquid systems that are coalescing, a series of static mixers can be used between lengths of open pipe. The static mixers re-split the drops/bubbles to the size needed. A continuous section of static mixers may not be needed. This can be important for reacting systems that need interfacial area.
- When Inline Mixers are Not Used.
There are times when inline mixers should not be used.
- For complex multi-phase formulations, a stirred tank may be a better route. Such formulations have many additives, probably are multiphase and may require a “digestion” or solvent strip step. As mentioned above, this is not to say that some additives can be mixed inline on a recycle line.
- For relatively slow reactions (requiring tens of minutes or longer), the reactants can certainly be premixed by an inline mixer but should be completed in a stirred vessel or wide bed. Otherwise, “miles of pipe” may be required. Part of the decision for this will be whether plug flow is required or if back-mixing is acceptable or required.
Feel free to contact Becht for further discussions on this topic or other topics related to mixing, whether inline, stirred tank, jetted tank, or other options.