Thank you for your reply, Behnam. I will post more data and another image to try and give more details of this scattering phenomenon.

I have spectrometer graphs of a control cell (glass discs without fluid) and spectrometer graphs of cells with different fluid thickness layers. I will post this information soon for you to examine, Behnam.

Thank you for your comments. I considered Mie scattering due to the nano sized particles combined with a magnetic field, but Rayleigh seems more likely.

Note the image seen with a Ferrocell doesn't change with polarized light. Rotating the polarization will increase or decrease brightness but not alter the image shape.

I've attached four spectrometer graphs of cells with increasing layer thickness and one graph of the reference white light. I'll have to re-create the experiment using a cell without fluid (glass reference) and post it later on.

These graph represent an un-activated (no magnetic field) condition.

The first graph represents a 50um glass separation, next is 100um, followed by 150um and 200um. The last graph is the reference white light.

If you have scope to measure the change in frequency what you have observed, then you can use compton equation tos the scattering angle. From this scattering angle, you may get insight as well as new findings 

Thank you both, Behnam and MZ for your responses. First, let me describe the spectrometer as a low cost CCD unit with an accuracy around 2%. The graphs indeed show wavelength, not frequency.

I appreciate the link to TF Heinzs' work and look forward to reading his paper. I've attached another image (screen shot) of the spectrometer graph for a "reference" cell. It's composed of two 2mm thick discs of optical float-glass sealed without ferrofluid in-between. Compare this graph with the white light reference graph (which is a 15W quartz lamp from an optical microscope) and you will see a slight attenuation across sections of the spectrum by insertion of the reference cell between light source and spectrometer.

I will post more charts of inducing a magnetic field into the cell so we can see how the field not only changes the lights' angle of incidence, but also changes the frequency slightly. Then we can analyze the Compton scattering angle from the collected data.

I'd like to point out a major difference between a Ferrocell lens and a contemporary optic lens. The cell is dynamic as opposed to static. The particles move and oscillate when induced with magnetism while irradiated with light. Inside the colloidal suspension, they are free in every axis. Within miliseconds, microscopic particle chains form from the aligned and induced nano-particles. Basically we have created many millions of "slits" that have the freedom to self-align with the lowest potential of the field. See second image of a 2 T cylinder magnet on top of a Ferrocell with a white incandescent lamp 30mm below the center of the cell.

Yes, any quality optic glass has little effect on image output. I've developed a technique where there are no air bubbles or debris within the cell and it maintains a sealed vacuum for at least 10 years (I have active cells from my first experiments).

My limited budget and health issues have prevented me from obtaining new equipment for a while, but we have 10 years of improvements and data to work with now.

For a ferrofluid it is known that refractive index varies with magnetic field. It may be helpful to see changes with varying field. Chain-formation can also give rise to grating effect. In order to eliminate possibility of inelestic scattering it may be useful to use monochromatic source than white light and observe any frequency shift. Blue color can also be accounted by Tydel spectra. In short, further in depth study is required to explain this interesting observation.

Thank you for your clarification and information. I have considered the grating effect to be a major function of the cell, leading to the holographic image we see. And, I'll investigate the Tyndall function further. I've conducted many tests with monochromatic light and have included a few spectrometer graphs for analysis and discussion.

The first graph shows a green LED as reference. The second graph shows a green LED through a medium grade cell. The third graph shows a red LED as reference, while the fourth graph shows a red LED through a medium grade cell.

I've read many research papers on FF after I posed this question. Generally, when the nano-particles are in a state of equilibrium (no light, no magnetic field) they remain separated and individual, giving rise to Rayleigh Scattering. However, when a magnetic field is in induced into them they begin to align and form chains. These chains are more likely to respond with Mie Scattering. So not either, but both!

Mie scattering is actually applicable for large spherical particles and not cylinders.Still sometimes one consider equivalent spherical radius to apply it for cylinders.

Timm Vanderelli

Seems some strange understanding of Rayleigh (an approximation) and Mie (all-encompassing) scattering. Change in wavelength is a Raman effect typically and very weak generally. The Rayleigh approximation (point scatterers) tends to give equivalent to (correct) Mie results at sizes less than approximately λ/10 where λ, lambda, is the wavelength of impinging radiation. So, around 60 nm for a He-Ne laser at 632.8 nm.

Rasbindu V. Mehta Mie scattering is perfectly adapted for cylinders (one of the first solutions after spheres). Peanuts have been modeled too... It is also valid from gas molecules to galaxies.... The issue, of course, is that irregular particles need more than 1 number to perfectly define them. This is why most light scattering theories generate equivalents (scattering in the case of Mie). Take a look at (registration required):

The life of Gustav Mie and the development of the Lorenz-Mie solution to Maxwell’s equations

https://www.malvernpanalytical.com/en/learn/events-and-training/webinars/W140225Gustav-mie-maxwell.html

Mie theory is the basis of the T-Matrix and DDA (Discrete Dipole Approximation) approaches for irregular particles.

"Generally, when the nano-particles are in a state of equilibrium (no light, no magnetic field) they remain separated and individual, giving rise to Rayleigh Scattering". When particles < 100 nm come in contact with one another or surfaces (consider a car windshield and the faint blue haze that you can see with various light illuminations) they become irreversibly aggregated. Take a look at:

Dispersion and nanotechnology

https://www.malvernpanalytical.com/en/learn/events-and-training/webinars/W190528DispNanoMat

and:

Adhesion and cohesion

https://www.malvernpanalytical.com/en/learn/events-and-training/webinars/W180725AdhesionCohesion