Transflooder
The idea here is sort of a microfluidic analog to a general purpose electronic computer. Easily re-programmable to do vastly different tasks.
There are several levels here.
Ordered in terms of system throughput
- Red: Chip integrated mirofluidics … this needs a dedicated chip fab facility
- Orange: (DLP resin) 3D printed integrated mirofluidics (hardcoded dense microfluidic functionality in 3D but produceable locally)
- Yellow: (DLP resin) 3D printed composable "LEGO" (small cubes with ports and certain functions)
- Green: Pipetting (and on chip electrostatic droplet manipulation)
- Blue: (FFF/SLS) 3D printed (or conventionally machined) high througput fluidics
Ordered in terms of system complexity
Same order except blue ends up in the center between yellow and orange.
Crudely FFF or SLS printed microfluidics can pack complexity (meant per volume) somewhere between these two spots.
Thus the resulting systems would not be ginormous in size (also hurting throughput, beside the natural throughput ranking for simple processings).
(wiki-TODO: Was there more thinking to this?)
Use cases
Beside one-pot-chemisty (particulatly one-pot self-assembly)
more complex process with iterative introduction and wash out steps may be enabled.
See: Iterative self-assembly
Temporary attachment to surfaces is one critical tool for these kinds of synthesis processes (aka assays).
Big columns are often used to get enough surface area and yield.
Here the focus is mostly analytics an nano-quantities so tings can be smaller.
Still getting large surface area might remain a challenge.
E.g. termination control for peptides/proteins
Maybe interesting things to try and explore include
various approached to termination control for de-novo proteins:
- circumsembly evading exponential drop-off in yield but still extending on a chain with a small set of next-up interfaces,
evading the specificity vs activity small orthogonal binding interface matrix dilemma - squigglesembly terminaltion contrilled 2D (and perhaps 3D) structures by assemblying squiggly 1D shapes
- algorithmic self assembly extended to the peptide/protein space
Analytics (why factored out)
Quite some analytics are factored out as there are specialized existing monolithic devices for these.
Transfer to these can be a bottleneck.
Increasingly hard to integrate (no claim on completeness):
- SPM microscopy
- Confocal microscopy
- Cryo EM
Cryo EM side-notes
Some are big expensive and nontrivially to operate. E.g. for cryso TEM ome beeds access too a modern high res TEM worh well >1M$ (2026). There devices are researc institutions sometimes time on these can be rented. There is also very much manual non automated prep needed in shock freezing and inserting samples. As has been done extensivels for structural DNA nanotechnology. Stained for crude pictures of individual results. Unstained more for averages over ensembles by comuting many imaging results to real space.
Good learning resources here:
Leave to youtube playlist: Cryo-EM_ Grant Jensen
Delineation
There is the concept of a "chemputer" advertised by Leroy Cronin.
- attacking the reproducability crisis in chemistry
- partly motivated by researching the origin of life
- focus on more broad and general chemistry
- focus on a bit bigger quantities
- (wiki-TODO: Check these assumptions more thoroughly.)
This here is likely a bit different with a specific focus on:
- foldamers for nanosystems as quite narrow subspace of chemistry (rather than more general chemistry)
- very low quantities mostly for research and analysis,
in the context of a mixed path extremely low product quantities might be sufficient),
the grey part in the graphic leads from low throughput research results to high throughput high quantity product though
if that is a desiderata for sellable side products. - mostly motivated by helping along
incremental path - bootstrapping - pathways
If anything a "chemputer" would likely be closest to (or could interface with) the (blue marked) high throughput fluidics.
This matches the in dashed box indicated "foldamer pre-synthesis in high quantity"
Related
External links
- Assay & Analytical chemistry
- Microfluidics
- Automated synthesis
- Leroy Cronin, the guy proposing the here tangentially related chemputer.
When taken the wrong way this "transflooder" concept seems uncomfortably close to Theranos.