The term "pipeline mixing" covers mixing of materials in a flowing line downstream of a junction. The mixing may involve miscible liquids, immiscible liquids and multi-phase mixtures. Options include just letting materials mingle naturally, using pipe fittings to spur contact, and installing static mixers, spray nozzles or spargers. Static mixers now dominate pipeline mixing — but that doesn't mean they're always the best choice.
Let's consider a recent case that involved choosing a better pipeline mixer for a liquid/liquid service that included mixing both miscible and immiscible liquids.
This application has two mixing steps: 1) mixing two miscible liquid reactants; and 2) adding the reactants to an immiscible liquid catalyst. Some reactions take place at the interface. Others occur inside the catalyst phase after the reactants dissolve into the catalyst. The catalyst-to-reactants ratio is roughly 1:1 by volume; the catalyst has the same volume as the total reactants in the system. Neither the reactant phase nor the catalyst phase is well defined as either a continuous phase or a discontinuous phase.
The idea was to improve yields by more-thorough reactant/reactant and reactants/catalyst mixing. This would increase inter-phase surface area, which would help both types of reaction mechanisms. The current setup relies on a simple pipe junction upstream of the reactors. We evaluated a spray nozzle, a sparger and a static mixer as a possible replacement.
Conventional spray nozzles accelerate a liquid to create a jet. The liquid then breaks up into smaller droplets. The major types of spray nozzles that might be used here are based on (1) rotating flow in a chamber that exits 90° from the liquid inlet, (2) swirl imparted by an internal vane or (3) a narrow stream cut by a spiral blade (pig tail).
These nozzles form droplets primarily through a combination of liquid ligament breakup and slicing of liquid sheets leaving the nozzle. Both mechanisms vary with liquid velocity, surface tension between phases and other physical properties. Jet instability is a key factor in making lots of drops. The little data available show most mixing velocity is shed within 12 in. to 18 in. of the nozzle. No significant droplet formation occurs because the original liquid ligaments or sheets don't form.
A sparger is a pipe with multiple holes that create a pressure drop forcing flow to distribute across the holes. (This pressure drop only is imposed on the liquid being injected, not the entire stream.) With the sparger installed into the main line, the injected flow of one stream would enter the second stream. The sparger could be aligned either across a larger pipe (at 90°) or along the same flow line as the larger pipe.
As with a spray nozzle, enhanced liquid mixing comes from local turbulence created by injecting a high velocity liquid into a second liquid. The mixing is likely at least as good as that of a spray nozzle. Design and installation of a liquid sparger typically is both cheaper and simpler.
Static mixers have become dominant for good reason. They use vanes or blades as elements. This enables mixing to occur at relatively low pressure drop, as little as 10% or 20% that of a sparger. The only potential downsides are that a static mixer often requires a longer straight pipe run for installation and pressure drop is applied to the entire stream.
Overall, the sparger and the static mixer are the best technical choices. Both have proven track records. In contrast, the spray nozzle, which is designed for liquid injection into gas, rarely is used in liquid/liquid services and should be avoided.
Despite this, the plant has opted for spray-nozzle injection for both mixing tasks. It considered spray nozzles proven technology because they have been used in this process by other plants. Here, though, hydraulic constraints limit the pressure drop to a fraction of that at other units; so results may not be as good.
Not agreeing with a decision doesn't free an engineer of the responsibility to help the site derive the most benefit possible from its choice. So, we recommended use of pig-tail-type nozzles. These mechanically "cut" a solid liquid stream into sheets but don't form as uniform droplets as the other types in conventional services. However, their mechanical design is guaranteed to at least do something. The cutting action will improve liquid/liquid mixing somewhat. Also, the cutting edge acts as a minor mixing element in its own right.
ANDREW SLOLEY is a Chemical Processing contributing editor. You can e-mail him at ASloley@putman.net.