Computational Facilitation of Large Scale Microfluidic Fuel Cell Architectures

Michel Takken 1 Robert Wille 1 2 · 2026

Read Computational Facilitation of Large Scale Microfluidic Fuel Cell Architectures by Michel Takken 1 Robert Wille 1 2 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-27755.txt.

Introduction

Takken and Wille speed up design of microfluidic fuel cell stacks. Full fluid simulations scale badly to large stacks. Their reduced-order method models one cell in detail. Stack behavior then builds from that single-cell model. Runtime drops sharply versus full CFD runs. Agreement with detailed results stays close. Scaling tests probe very large layouts. The tool supports practical stack development.

Catalog record

CREATORMichel Takken 1 Robert Wille 1 2
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE45145
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.27755

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Computational Facilitation of Large Scale Microfluidic Fuel Cell Architectures Thanks: Michel Takken 1 Robert Wille 1 2 Abstract Hydrogen fuel cells are a key technology in the transition toward carbon-neutral energy systems, offering clean power with water as the only byproduct. Microfluidic fuel cells, which operate at the microliter scale, are an emerging variant that offer fine control over fluid and thermal dynamics, along with compact, efficient designs. However, scaling these systems to meet practical power demands remains a major challenge—particularly due to the limitations of conventional simulation methods like Computational Fluid Dynamics (CFD), which are computationally expensive and scale poorly. In this work, we propose a reduced-order simulation method that models the behavior of individual microfluidic fuel cells and efficiently extends it to large scale stacks. This approach significantly reduces simulation time while maintaining close agreement with detailed CFD results. The method is validated, evaluated for scalability, and discussed in the context of ongoing advancements in microfluidic fuel cell fabrication. The obtained results demonstrate that this abstraction can support the design and development of scalable microfluidic fuel cell systems and, for the first time, the consideration of first macroscale instances of practical value. I Introduction In order to reduce the risks and impacts of climate change [ 1 ] , many economies have set carbon neutrality goals and developed strategies to achieve those [ 2 , 3 , 4 , 5 ] . Most of which have identified hydrogen and fuel cells as key energy technologies of the 21 st century [ 6 , 7 , 8 ] . Fuel cells are electrochemical cells that convert chemical reactants into electrical energy and byproducts. A schematic illustration of a hydrogen fuel cell system is provided in Fig. 1 , which shows the inputs to the system (i.e., the hydrogen and oxygen) in blue, and the outputs (i.e., electrical energy and water) in red. During operation, the fuel cell produces heat, which can optionally be managed using, e.g., a heat sink or a coolant combined with a thermal control subsystem. Furthermore, a hydrogen fuel cell generally contains an electrolyte that is necessary for the electrochemical reactions. The products of that reaction must be removed from the system, which is illustrated by the water removal subsystem. The advantages of the system are evident, as it produces electrical energy with water as its only byproduct. Unlike batteries, fuel cells do not require charging, but rather a replenishment of the chemical reactants, i.e., fuel. The use of fuel cells is versatile, ranging from mobile applications, such as aviation [ 9 ] , shipping [ 10 ] , and Unmanned Aerial Vehicles (UAVs, [ 11 ] ), to stationary applications, such as domestic power generation and Combined Heat and Power (CHP) production on residential blocks [ 12 , 13 ] . Common fuel cell types for these applications are the Proton Exchange Membrane Fuel Cell (PEMFC, [ 14 ] ) or the Solid Oxide Fuel Cell (SOFC, [ 15 ] ) . Fig. 1 : A schematic representation of a conventional air-breathing fuel cell system. Besides the fuel cell, the system is composed of various subsystems, including: A fuel and air intake subsystem, a water removal subsystem, and thermal control subsystem. However, before a large scale adoption of fuel cells is possible, many technological challenges still need to be overcome. These include challenges with flow and/or thermal control, flooding, or membranes that degrade over time [ 16 ] . In parallel, microfluidic devices (i.e., devices that operate fluids in the nano-/microliter scale) are being developed and have succesfully been employed for Point-of-Care (PoC, [ 17 ] ), Lab-on-Chip (LoC, [ 18 ] ), and Organ-on-Chip (OoC, [ 19 ] ) applications. The capability of microfluidic devices to manipulate fluids on a small scale allows for a better control of, e.g., flow and temperature, making them a promising platform for (membraneless) fuel cells. In fact, microfluidic fuel cells are actively being developed [ 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ] since their introduction [ 33 ] . Similar to conventional fuel cells, they are successfully being stacked into microfluidic fuel cell stacks, with up to 6 cells, to provide more power [ 34 , 35 , 36 , 37 , 38 , 39 , 40 ] . In addition, recent developments show efforts made to integrate micro-scale pumps on Micro-Electromechanical Systems (MEMS) [ 41 ] , purify water in electrochemical microfluidic devices [ 42 ] , and manage heat in the scale-up of microfluidic devices [ 43 ] . These form micro-scale equivalents of the supporting sub-systems (shown in Fig. 1 ) that are necessary to operate fuel cells efficiently. Overall, these developments indicate that microfluidic fuel cells are an emerging technology that show great potential and may play a role in achieving carbon neutrality. However, in order to meet the power production of their macrofluidic counterparts, a major challenge that still remains is scaling. Thus far, this was mainly a fabrication issue, but with the advances in micro-fabrication methods [ 44 ] , (rapid) 3-D printing techniques for microfluidic devices [ 45 , 46 , 47 , 48 ] and the advances in integrating electrodes in microfluidic devices [ 49 , 50 , 51 ] , substantial steps towards the realization of large scale microfluidic devices have been made. As a result, scaling fuel cell stacks has progressively become a design concern. To provide an intuition of the required scaling in practical purposes, Table I lists some examples of required stack sizes needed for different systems, considering a representative cell size of 0.25x2 mm and a peak power density of 100 mW/cm 2 [ 52 ] . It is obvious that, with stack sizes like that, the realization of microfluidic fuel cell stacks requires significant design capabilities. Simulation can significantly aid in the design of corresponding devices [ 53 ] and, in this regard, are key as they allow to explore different designs and check/validate them prior to fabrication. Unfortunately, state-of-the-art simulation methods for microfluidic fuel cells [ 54 , 23 , 55 ] , such as those using Computational Fluid Dynamics (CFD, [ 56 ] ) , are time-consuming and scale poorly [ 57 , 58 ] . This makes it practically impossible to simulate large scale fuel cell stacks. In an effort to address the current situation, this work proposes an alternate simulation approach that utilizes a higher level of abstraction than the state of the art and, by this, allows to efficiently simulate large scale microfluidic fuel cell stacks, using a reduced-order model while maintaining the desired accuracy. In the remainder of this work, we carefully review the necessary background and governing equations for a representative microfluidic fuel cell, in Section II . Afterwards, the abstract simulation method is explained in a two-fold fashion: first the abstraction is explained for a single cell in Section III , based on which the abstraction is extended to a stack in Section IV . Subsequently, the abstraction is validated against simulation results from CFD in Section V , followed by a scaling evaluation in Section VI

Read the complete text