Modeling of a microfluidic membrane-based preconcentrator (Advanced Technology / Chemical Engineering)

Student assignment: Modeling of a microfluidic membrane-based preconcentrator

In previous research in collaboration with the Radboud University in Nijmegen, a microfluidic chip for NMR spectrometry has been developed, which operates with less than 1 microliter liquid sample [1]. An intrinsic problem with NMR is the low sensitivity. Therefore, measurement times for small samples with low concentrations can become unacceptably long. For example, in the experiments on human cerebrospinal fluid in ref. [1], in which 600 nL of sample was measured in a microfluidic NMR system, continuous collection of NMR scans for almost 18 hours was necessary to detect metabolites at 1 mM concentration. An important method to decrease measurement time is by increasing the analyte concentration: a 10 times higher concentration reduces measurement time by a factor of 100. For this purpose, we have investigated an in-line microfluidic concentrator [2] which is based on evaporation of solvent through a microporous membrane, see figure (left: membrane with pores, right: one pore).


Previously an analytical model was developed for the preconcentrator, which predicts an asymptotic dependence of concentration factor on flow rate. Due to this, very accurate control of the liquid feed is crucial, especially when the flow rate approaches the asymptote, and the smallest variations in liquid flow rate locally lead to too fast concentration of the solution followed by crystallization in the pores. The probable reason for this problem is that dissolved substances do not have enough time to diffuse away from the pore opening where the solvent is evaporating quickly, and the solution becomes supersaturated, so that crystals will form in the opening, which blocks the pore. This changes membrane porosity in an unpredictable way and limits the maximum attainable concentration factor to ca. 6 to 8.

The main goal of this assignment is to come to an optimized concentrator design, by modeling flow, diffusion and evaporation. This assignment consists of the following tasks:

i) develop a numerical model which describes the evaporation of solvent at a circular pore in combination with the diffusion of the dissolved molecules away from the pore;

ii) include the flow of the liquid that is present on one side of the membrane, as well as the flow of sweeping gas on the other side of the membrane, into the model;

iii) formulate design rules for an improved membrane-based preconcentrator.

The assignment can be extended with experimental (microscopic or microfluidic) work on the evaporation through (single or multiple) pores.


[1] J. Bart et al., A microfluidic high resolution NMR flow probe, J. Am. Chem. Soc. 131 (2009) 5014-5015; DOI: 10.1021/ja900389x

[2] H. Zhang et al., In-line sample concentration by evaporation through porous hollow fibers and micromachined membranes embedded in microfluidic devices, Electrophoresis 37 (2016) 463-471; DOI: 10.1002/elps.201500285

Contact information: Han Gardeniers; Email: