I am releatively new to COMSOL and self teaching so forgive me if the questions posted here are missing some fundamental understanding. Using COMSOL 5.4.

Proposed Model

I am attempting to simulate part of a single channel for a biosensor that has immobilised antibody (A) on an electrode surface with a microfluidic channel above it which flows the antigen (B) of interest over it - this would allow it to bind and then produce an electrochemical signal. I am wanting to simulate the consumption of the antigen (B) at various flow rates and channel diameters (channel height etc.).

Model so Far

I looked at various model tutorials, including transport and adsorption; protein adsorption and Hydrocarbon Dehalogenation in a Tortuous Microreactor. I tried to mimic what seemed to be the normal method of design by starting with a 0D, time dependant Reaction Engineering model. For the model, I assumed that its an irreversible reaction (there appears to be no data available for the Koff of the antibody I am investigating), the phase is a liquid, the reactive ads species (A) has species concentration locked and I haven't used Arhenious equations (again due to lack of reported data).

For this I tried initially to simulate the reaction:

• A (ads) + B (bulk) => AB(ads) for the antigen being consumed and binding to the surface antibody. However, this led to constant production of AB(ads) without A or B being consumed, to a scale that was much larger than the initial concentrations of either reactant.

• A (ads) + B (bulk) => AB (bulk) If I were to set AB as a bulk species however, I do get a typical 2nd order reaction curve that has the consumption of A mirrored with the production of AB. But this obviously isn't a typical represenation of what the antibody-antigen interaction would look like.

If I were to generate a space dependant model from the second 0D model (given that the 0D time dependant study was a more logical outcome).

I attempted then to produce a channel with 1.6mm sensor surface on the channel base. To this component I added the laminar flow, chemistry and transport of diluted species modules.

Chemistry 1

Chem module followed on very closely to the reaction engineering module. I'm uncertain what settings need changed here. The only thing that confused me was, if I look at each of the individual species and set the reaction rate for the antigen (B), it has the rate expression set by default to = 0 - is this correct or should I be manually changing this value? I had left it to automatic prior.

Transport of Diluted Species

Set using inital concentrations (quantity in scale of nano-100 micro molar); surface reactions 1 designated as the surface for the sensor. Initial concentration set as 0 for cAB, and B = c0 (7.1428e-8 M). Diffusion coeff. set to value of D for B, 0 for AB.

Laminar Flow

fluid properties were set to material (designated as just water from the material catalogue). Initial velocity field set as 0, with the inlet velocity denoting the actual inflow velocity. Pressure set at p0. Outlet and inlets were designated.

Multiphysics

Set flow coupling for source: laminar flow and destination: transport of diluted species.

Mesh

A customised mesh was produced with a free triangular (finer) mesh for the sensor boundary. The remaining domain was covered with a courser swept mesh to cut down on computation time.

Studies

If I were to perform a time dependant study, looking at just the chemistry module, it says I have a concentration of 0 for all species. I'm unsure as to why this is.

If I were to consider just the transport of dilute species, it shows an exponential growth of concentration of AB at t=0min, to immediately 10^3 M on the surface of the sensor, increasing to 10^7 by 60 min.

Help

I would appreciate someone giving guidance on what key parameters I am missing, settings that are incorrectly set at or simply faults I have with my overall model which need reconsidering. Any and all help is appreciated - especially if explained simply or with step by step break downs for those of you who can spare the time.

Many thanks, attached is the messy build of my model if its of use for people to understand the individual settings/parameters used etc. note in model, ads species A is denoted as AS, antigen bulk species B is denoted as TSH after the protein of interest.