Skip to content

Microfluidics case study

Available (limited)

Microfluidic cell-sorting chip design

Separating cells by size and deformability without labels is a core microfluidic capability. A demanding case is isolating circulating tumor cells (CTCs, roughly 17-20 um) from the whole-blood background (red cells ~7.5 um, white cells ~12 um). Octara sizes these separations against a cited cell library and simulates the streams.

Open the free StudioRequest access
Simulated cell-sorting streams separating large from small cells
cell sorting - simulated streams

The problem

A size-based sorter must push large and small cells onto different streamlines and keep them there to distinct outlets, in a background that is crowded and deformable.

Architecture

  • Passive size-based methods: inertial spiral (Dean-flow focusing), deterministic lateral displacement, or pinched-flow fractionation.
  • A cited cell library sets diameters, density and deformability.
  • Outlets are placed at the focused-stream positions for large vs small cells.

Assumptions

  • Dilute suspension, size-based cutoff
  • Laminar carrier flow
  • Particles follow streamlines except for inertial / Dean migration

Model

Inertial focusing scales with the particle Reynolds number; a curved channel adds a Dean secondary flow that focuses particles by size; the geometry is tuned so the target size lands on a different outlet from the background.

Particle Reynolds: Re_p = rho U a^2 / (mu D_h)Dean number: De = Re * (D_h / 2R)^(1/2)Separation: size cutoff a* placing large cells (CTC) and small cells (blood) on different streams

Simulation results

  • Distinct outlet streams for large versus small cells in the simulated separation.
  • The design is sized against real cell diameters, not arbitrary values.
  • Flow and residence time are checked with the lumped-hydraulic solver.

Limitations

  • Size overlap (large white cells vs small CTCs) limits achievable purity.
  • Deformability, clogging and hematocrit matter and are not fully captured.
  • Predictions are computational - validate with spiked samples and imaging.

Try it in the free Studio

Build a network like this and solve it for pressure, flow and Reynolds number in your browser - the same lumped-hydraulic model, no account needed.

Open the free StudioBack to Microfluidics

More microfluidics case studies

Microfluidic simulation softwareA reduced-order microfluidic simulation model: solve pressure, flow split and Reynolds number across a chip with a validated lumped-hydraulic network, before you fabricate.AI-assisted microfluidic designFrom a plain-language objective to an editable microfluidic chip architecture. Octara's design assistant orchestrates a real component library, placement and routing engine - the model routes intent, the engines do the work.Microfluidic concentration-gradient generatorDesign a Christmas-tree (Pascal-triangle) microfluidic gradient generator: matched-resistance splitting, serial mixing and defined outlet concentrations - with the hydraulic model that sets the geometry.LAMP chip design for isothermal amplificationDesign a lab-on-chip for LAMP isothermal nucleic-acid amplification: metering, reagent mixing, evenly-filled reaction chambers held at temperature, and a readout - with the hydraulic balance that fills chambers evenly.