Description
Evaporation remains a pervasive yet often underestimated source of error in microfluidic platforms, particularly as liquid volumes approach the nanoliter and sub-nanoliter scale. In life-science workflows such as sample preparation, biochemical assays, and thermal cycling, even modest evaporative losses can compromise volumetric accuracy, alter solute concentrations, and limit experiment duration. While evaporation has been extensively studied in open systems (e.g., droplets and thin films) and in simple confined geometries such as single microchannels, evaporation in realistic microfluidic circuits—comprising interconnected chambers, wells, and channels—remains poorly quantified and difficult to predict.
In this work, we present a combined experimental and theoretical investigation of evaporation in microfluidic circuits spanning simple single-path geometries to complex, multi-path networked designs. Using controlled heating experiments, we quantify evaporation rates of liquid plugs with volumes ranging from approximately 1 µL down to 100 nL. To directly probe vapor transport within the devices, we employ fluorescence-based thermo-sensitive coatings (TSCs) that enable spatial mapping of temperature and relative humidity inside the microfluidic chips during evaporation. These measurements provide a direct visualization of vapor gradients inside confined microfluidic circuits.
Guided by these observations, we develop a diffusion-limited evaporation model based on Fick’s law and a lumped-element framework. Each channel, chamber, or well is treated as a resistance to diffusion, allowing complex circuits to be reduced to equivalent series and parallel resistance networks, analogous to electrical circuits. This approach yields a single effective diffusion length that governs the evaporation rate, even for geometrically complex designs. Analytical expressions are derived for both uniform channels and non-uniform geometries, including pyramidal wells relevant to multi-well plate architectures.
Across all investigated designs, the measured evaporation rates are accurately predicted by the model, with deviations below 7% for most geometries. Importantly, we find that, within the isothermal and diffusion-limited regime explored here, evaporation is insensitive to contact-line position or pinning at chamber edges. Instead, the dominant control parameters are the global vapor-diffusion pathways defined by channel length, cross-section, depth, and auxiliary leakage routes. For larger chambers, localized condensation can transiently enhance vapor gradients and increase evaporation, an effect that is captured qualitatively by the resistance framework.
Beyond explaining evaporation losses, this study establishes practical design rules for engineering microfluidic circuits with predictable evaporative behavior. By framing evaporation as a network-level transport problem rather than a local interface phenomenon, the presented approach opens new opportunities for evaporation-aware microfluidic design, improved assay robustness, and controlled evaporation strategies in applications ranging from life-science automation and diagnostics to wearable and portable microfluidic devices.