Steep-Edge Dichroics for Multiplexed Fluorescence: Where Blocking Depth Actually Comes From

Fluorescence Microscopy and Life Science · 2026-10-19 · 3 min read

Multiplexed fluorescence is a signal-to-noise problem disguised as a filter problem. Four or five fluorophores excited at once, emission bands that overlap, and a detector with limited dynamic range: the dichroic has to separate channels that are tens of nanometres apart while passing everything else.

Steep-Edge Dichroics for Multiplexed Fluorescence: Where Blocking Depth Actually Comes From

Two numbers govern the result, and they trade against each other.

Edge steepness

The transition width between reflection and transmission is set by the number of layers, the index contrast available in the materials, and the design approach. A stack with more layers and higher contrast gives a sharper edge, but it also accumulates stress, takes longer to deposit, and becomes more sensitive to thickness errors.

In practice, a steep edge is not a single figure. It is a slope in nanometres from 50% to 90% transmission, and it depends on angle and polarisation. At 45° incidence — the standard geometry for a fluorescence dichroic — s- and p-polarised components see different effective indices and therefore different edge positions. The separation between them is the polarisation splitting, and in a high-NA microscope it shows up as a channel shift between the centre and the edge of the field.

Blocking depth

Reflecting the excitation laser while transmitting emission is only half the job. The dichroic also has to block the excitation from reaching the detector after scattering in the sample, and it has to block emission from adjacent channels from leaking into the wrong one. That is optical density, and it is achieved by adding layers — but the blocking band has to cover the actual excitation wavelength at the actual angle and over the actual cone of light the objective delivers.

This is where many specifications quietly fail. A dichroic quoted at OD 6 for 488 nm at 45° may deliver OD 4 at 40° and OD 3 at the edge of the acceptance cone. For a weak fluorophore next to a strong one, that difference is the whole experiment.

Flatness and wavefront

A thick dichroic stack on a thin substrate bends. In a laser-scanning confocal system the dichroic sits in the illumination path, so transmitted wavefront error directly degrades the point spread function. Buyers who specify 1/10 λ transmitted wavefront on a 1 mm substrate and a 40-layer stack are asking for something that requires a thicker substrate or a compensating design. That conversation is much cheaper before the order than after first article.

What to ask for

  • Edge position and slope, at the working angle, separately for s and p polarisation.
  • Blocking depth over a stated wavelength range, at the working angle and over the acceptance cone.
  • Transmission in the passband including the effect of the AR coating on the second surface.
  • Substrate material, thickness and transmitted wavefront specification.
  • Whether the measured curves come from a spectrophotometer at a single angle or from a cone-angle measurement.

The design trade nobody escapes

Steepness, blocking and wavefront all compete for the same layer budget. A dichroic that is very steep across a wide angular range will be thick; a thick stack will stress the substrate; a stressed substrate will need to be thicker; a thicker substrate will cost more and may not fit the cube. The right answer is a design point, not a maximum, and it can only be found if the application's real tolerances are on the table.

← All 400 articles

Need a Quote or Engineering Review?

Send your drawings, specifications or coating requirements — we reply within 24 hours.

Request a Quote

Related Articles

Request a Quote