Soft, Skin-Interfaced 3D Microfluidic Systems for High Performance Assessment of Sweat Rate, Cumulative Loss and Biochemical Content

Emerging classes of wearable microfluidic sensors of eccrine sweat enable non-invasive and real-time monitoring of health status in applications that range from sports to worker safety, environmental exposures, and medical care.


Author perspective

“The hardest person to measure has always been the one who barely sweats, and that’s often exactly who you want a reading from. Most of the effort in this work went into the side of the device nobody sees, the channels that pull sweat in sideways from the glands around the inlet instead of waiting on the few underneath it. Seven times more sample from the same 40 minutes is what makes the rest of the chemistry worth doing.”

Alexander J. Aranyosi, PhD · Chief Digital Officer, Epicore Biosystems

Study findings

Two problems have limited what a colorimetric sweat patch can do. A chemical assay that’s sensitive enough to read low concentrations saturates at high ones, so you have to choose which end of the range to serve. And people who don’t sweat much often don’t produce enough to fill the device at all, which is the group most likely to need a reading in the first place.

This study solves both with geometry rather than new chemistry. Each assay reservoir is shaped like a flower, with petals at five different depths from 200 to 1000 micrometres. Because colour intensity depends on how far light travels through the liquid, the deep petals read faint low concentrations and the shallow ones stay unsaturated at high concentrations. You read whichever petal hasn’t maxed out. On the skin-facing side, a set of radiating water-attracting channels wicks sweat sideways from glands around the inlet rather than waiting for the few directly beneath it.

The combination widens the working range substantially: chloride from 2.4 to 125 mM, xanthine from 0.23 to 500 micromolar, creatinine from 1.97 to 500 micromolar, all from one patch, alongside sweat rate and cumulative loss read from a separate tapered channel. On a subject with a confirmed low sweat rate, the collection channels gathered 5.80 microlitres of sweat over 40 minutes in a sauna against 0.85 microlitres for the same device without them, roughly seven times more.


Scientific highlights

One reservoir, five depths

Petals at 200 to 1000 micrometres let a single assay read both faint and strong signals, removing the usual trade-off between sensitivity and range.


Seven times more sweat from a light sweater

Water-attracting collection channels gathered 5.80 microlitres over 40 minutes against 0.85 microlitres without them, on a subject whose sweat rate was too low to be picked up by a standard clinical collector.


Three biomarkers plus sweat rate

Chloride for electrolyte balance, xanthine for purine metabolism, and creatinine for kidney function, measured alongside sweat rate and total loss.


Softer on the skin, and steadier because of it

A hollow-shell construction halves the stretch modulus, so that at a 20% stretch typical of on-skin use the reservoir volume was changed by only about 10%, compared to 30% for a typical design. This allows measurement to remain accurate while the skin moves.


Why this research matters

Sweat rates vary between people by ten to a hundred times or more. A device tuned for an average sweater fails at both ends, and the failure is silent: not enough sweat looks the same as no result. Fixing that at the collection interface, rather than asking the wearer to work harder, is what makes a single device usable across a real population.

The biomarker set matters too. Chloride is a well-established electrolyte marker for sweat electrolyte loss and hydration. Creatinine is a standard clinical indicator of kidney function, and the paper tracked it in sweat rising from 8 to 53 micromolar and back over 24 hours after a creatine supplement. Xanthine follows caffeine metabolism, peaking 3 to 6 hours after a coffee. Both are measurable in sweat at useful precision, which extends the utility of sweat patches for monitoring important health biomarkers.



How this connects to Epicore’s solutions

This work comes from the laboratory of Epicore co-founder John A. Rogers at Northwestern University, with Epicore’s Chief Digital Officer Alexander J. Aranyosi and CEO Roozbeh Ghaffari as co-authors. The colorimetric chemistry and passive microfluidic routing follow the same principles that Epicore engineers and manufactures at scale.

Epicore’s commercial platforms measure hydration and electrolytes today. What this study adds is engineering that widens the range of biomarker concentrations that a single patch can cover and pulls a usable sample from people who barely sweat, which enables the use of sweat as a biomarker in both clinical and workplace settings without requiring strenuous exercise or risky heat exposure. The kidney and metabolism markers point at where multi-analyte sensing goes next.

Connected Hydration

Real-time sweat and hydration sensing for industrial workers, with personalized intervention before heat stress sets in.


GX Sweat Patch

A wearable microfluidic patch that gives athletes sweat rate and electrolyte feedback to tailor a personalized action plan.


Discovery Patch

An FDA-registered collection system for multi-biomarker sweat analysis in clinical studies, the foundation for cortisol work.

FAQs


Sweat rate is how much sweat you produce per unit of skin area per unit of time. It differs between people because of gland density, fitness, heat acclimatisation, age, sex and body size, and it differs across the body on the same person. The spread is wide: this paper cites variation of ten to a hundred times or more. That’s why people need individualized hydration advice rather than generic recommendations.

Most collection devices pick up sweat only from the glands directly underneath their opening. If those glands are producing slowly, the device takes a long time to fill or never fills enough to give a valid reading. In this study a subject who consistently failed to fill a standard clinical collector produced a usable sample in 13 minutes once the device could draw sweat sideways from a wider area of skin.

Creatinine is the waste product left when muscles break down creatine. Because the kidneys clear it from the blood at a steady rate, its concentration is the standard clinical indicator of how well the kidneys are working. It appears in sweat too, at around 30 micromolar in healthy people, and much higher in people with severe kidney impairment.

Xanthine is part of purine metabolism, the pathway that ends in uric acid and, when it goes wrong, in gout. It’s also where caffeine goes after you drink it. In this study sweat xanthine was low before coffee, peaked 3 to 6 hours afterward, and returned to baseline only after 24 hours, which suggests fully clearing a dose of caffeine takes about a day.

A camera photographs the patch alongside a printed colour reference card. Software converts the image into a standard colour space that separates hue, which identifies which chemistry you’re looking at, from saturation, which tracks how much of the substance is present. Using a reference card and a colour space designed for this makes the reading independent of the lighting in the room.

Because colour intensity depends on how far light travels through the liquid, not just on concentration. A deeper chamber gives a stronger colour for the same concentration, which helps at the low end but saturates early at the high end. A shallow chamber does the opposite. Putting several depths in one reservoir means there’s always a region reading in its useful range.

Not as things stand. Sweat is easier and less invasive to collect, and several markers appear in it, but the relationship between a sweat concentration and the corresponding blood concentration has to be established separately for each marker. The authors of this study say directly that establishing those blood-to-sweat correlations for creatinine and xanthine is the next piece of work.

A valve with no moving parts, made purely from the shape of the channel. Liquid arriving at a sudden widening stops until enough pressure builds behind it to push through. By setting different widths, designers control the order in which chambers fill, so a patch can collect a timed sequence of samples using nothing but the pressure the sweat glands already generate.

It depends on the chemistry. In this study the chloride reaction is essentially immediate, while the enzyme-based xanthine and creatinine assays take longer to reach full colour, roughly 100 minutes for xanthine and between 5 and 60 minutes for creatinine. The authors note that speeding up the creatinine reaction is a target for future work.