Cheap FDM printed microfluidics (for DNA origami?)

  • When doing DNA origami to form lots of atomically precise nanoscale parts with different shapes/topologies there's the need to mix a lot of different combinations of a lot of different DNA oligomeres.
    AFAIK currently for such large scale mixing projects robotic pipetting systems are used that are not easily accessible.
    Microfluidic systems may have the potential to replace such huge and expensive robotic pipetting systems.

    Here's some new work investigating in how far cheap FDM printing can be used for microfluidics.…1371/journal.pone.0152023

  • I've looked a bit further into microfluidics
    and I now think that this technology will pretty much for certain play a major role in the development towards HT-APM.

    3D printed microfluidics - in contrast to those very expensive and pretty big pipetting systems found at biomedical labs -
    allows at least one order of magnitude cheaper and thus much more accessible
    tinkering with natural self assembly (and of course "unnatural self assembly" too - the desired engineering type that's targeting towards APM)

    Maybe a cost level can be reached that is affordable even for hacker-spaces.
    The main remaining cost point are the "vitamins" around the passive channel structures (like among many others pumps).
    Beside reducing the costs significantly microfluidics also increase operation speed by several orders of magnitude.
    This makes unguided bottom up construction of larger nanosystems much more realistic.

    By now there are lost of videos demonstrating microfluidic principles (bubble handling) on youtube. Very interesting!
    E.g. check out:…view=0&shelf_id=0&sort=dd

    Specifically I am thinking about a matrix like system for programmable high speed automated manufacturing of 3D-DNA-voxel-bricks (there are several papers about those):

    Step 1a) Start with N streams of DNA oglionucleotides (= short DNA snippets) and mix them into the accumulating droplets of M droplet streams. (The droplets grow very pretty big and essentially become stream-segments). For each steam M_i some N_j are retained (How?) the rest passes and mixes.
    Step 1b) The M resulting streams (that deliver consecutive long stream-segments) then are routed through long individual circuits for first internal mixing and then following annealing.
    Step 2) For second order self assembly (this has been experimentally demonstrated - there's a paper on this) repeat the process with the M long stream-segment-bubbles that now contain the desired fully assembled DNA structures (and salt solution streams). Those are now replacing the N DNA oglionucleotide streams that where present in the first processing round. (Obviously if the same channel-circuitry is used for step 2 it needs to be flushed out first.)
    Step 3) Extract the product assemblies and analyze the results.
    Maybe assemble some early productive nanosystem (featuring some hinges and nano-robotics)
    (though with - for the time being - still stealthy and publicly unperceived small scale products)

    Motivated by that exciting prospect I recently began to port the
    discrete element microfluidic designs that where presented in the papers [1] and [2] (linked below) to OpenSCAD.
    This will make it possible to:
    1) quickly adjust the systems components geometries to the specific need at hand
    (I'm not referring to the systems inherent quick adjustability here. Which is a strong point on it's own.)
    2) print the systems components even on cheap FDM printers (which can be tuned to be sufficient for the job as shown in paper [0]).

    0) Title: "Simple and Versatile 3D Printed Microfluidics Using Fused Filament Fabrication" (Cardiff University)
    1) Title: "Discrete Elements for 3D Microfluidics" (2014)
    2) Title: "Predicting the behavior of microfluidic circuits made from discrete elements" (2015)

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!