The Seebeck effect lies at the heart of thermoelectric textiles. When one side of a material is hotter than the other, charge carriers diffuse from the hot region to the cold one, generating a measurable voltage. The ratio of that voltage to the temperature gradient defines the Seebeck coefficient, the central parameter for any thermoelectric material. Because the human body maintains a temperature of roughly 37 degrees Celsius against an ambient surroundings that is often cooler, clothing that exploits this gradient could, in principle, trickle-charge sensors, health monitors, and other smart textile devices without batteries. Cotton is a natural substrate for such applications: it is the most widely used fabric in the fashion industry, comfortable against the skin, and compatible with established coating techniques.

The challenge has always been how to make cotton conductive without destroying its textile character. Carbon nanomaterials such as carbon nanotubes and carbon nanofibers offer promising electronic properties, but they are typically produced as solid powders that resist integration into fabrics. The standard workaround is to disperse them into inks, which can then be applied by screen printing, inkjet printing, spray coating, or dip coating. In this study, the researchers formulated aqueous inks containing two Pyrograf III carbon nanofiber grades, PR 24 LHT XT and PR 25 PS XT, both synthesized by chemical vapor deposition at 1,100 degrees Celsius but subjected to different post-synthesis thermal treatments: 1,500 degrees Celsius for the PR 24 grade and 700 degrees Celsius for the PR 25 grade. These treatments strip a polyaromatic carbon layer from the nanofiber surfaces, and they leave the two grades with distinct structural signatures.

Transmission electron microscopy revealed that both nanofibers share a double-walled structure enclosing a hollow tubular core, with an inner layer of compacted parallel graphene sheets oriented at an angle to the core axis and a more disordered outer layer. The PR 24 fibers, however, have smaller total diameters of 80 to 100 nanometers compared with 125 to 150 nanometers for PR 25, and Raman spectroscopy confirmed that PR 24 possesses higher crystallinity, showing narrower D and G bands and a detectable 2D mode. The fabrics were functionalized by repeated immersion in inks containing 1.6, 3.2, or 6.4 milligrams of nanofibers per milliliter of distilled water, stabilized with the surfactant sodium dodecybenzenesulfonate, followed by drying cycles and rinsing steps to remove surfactant residues. Scanning electron microscopy showed that at the lowest concentration the cotton weave remained largely visible, while at 6.4 milligrams per milliliter an entangled mat of nanofibers completely covered the fabric surface, with the fibers concentrated near the surface rather than penetrating the full weave.

Thermogravimetric analysis under nitrogen confirmed that all coated fabrics remained thermally stable up to 300 degrees Celsius, matching the behavior of uncoated cotton, and revealed that the fabrics coated with the most concentrated inks carried approximately 8 to 9 percent nanofiber content by weight. At the highest concentration, the residue percentages at 600 degrees Celsius converged to about 18 to 19 percent for both nanofiber types, indicating that the coating had reached saturation and become homogeneous. This saturation point proved decisive for the electrical performance of the resulting conductive textiles.

The thermoelectric measurements at 30 degrees Celsius told a nuanced story. Both nanofiber types, measured in agglomerated solid form, showed similar electrical conductivities of roughly 133 to 136 siemens per meter, values that represent a practical ceiling for textile applications using simple dispersion methods. The coated fabrics, however, diverged depending on ink concentration: at the lowest loading, fabrics coated with the thicker PR 25 fibers actually conducted better, reaching 5.1 siemens per meter, while at the highest loading the PR 24 fabrics took the lead, achieving 35.4 siemens per meter against 23.2 for PR 25. Both fabric types exhibited negative Seebeck coefficients, confirming n-type conduction, with PR 24-based textiles reaching about minus 6 microvolts per kelvin compared with minus 1 for PR 25. Intriguingly, the absolute Seebeck values of the fabrics exceeded those of the corresponding nanofibers, a phenomenon the researchers attribute to slight electron doping from adsorption between the graphitic planes and cellulose molecules, and possibly from residual surfactant.

Combining these parameters, the best fabric, coated with the highest concentration of PR 24 nanofibers, achieved a power factor of approximately 1.2 times 10 to the minus 3 microwatts per meter per square kelvin. Using a novel null-point scanning thermal microscopy method, the team measured the thermal conductivity of the coated fabrics at just 0.06 watts per meter per kelvin, only twice that of uncoated cotton and far below the values typical of carbon nanotube-filled polymer composites. This low thermal conductivity is a genuine advantage for thermoelectric applications, since it keeps heat from leaking across the gradient. The resulting figure of merit, zT, reached approximately 6.3 times 10 to the minus 6 at 30 degrees Celsius, modest compared with inorganic thermoelectrics but notable for a flexible textile and higher than values previously reported for the same nanofibers in melt-mixed polymer composites.

The temperature dependence of the electrical conductivity between 30 and 100 degrees Celsius revealed perhaps the most striking result. While the nanofibers themselves showed a positive temperature coefficient, with conductivity decreasing as temperature rose, the coated fabrics behaved in the opposite way, exhibiting a negative temperature coefficient. The best PR 24 fabric increased from 35.4 to 45.0 siemens per meter across the range, and the PR 25 fabric rose from 23.2 to 29.5. Fitting the data to a three-dimensional variable range hopping model showed that the fabrics conduct through thermally activated hopping of charge carriers across a random network of potential wells, with mean energy potentials of roughly 15 electronvolts, whereas the nanofibers alone conduct through thermally enhanced backscattering from impurities or structural defects. In other words, coating the fibers onto cotton fundamentally changes the transport mechanism.

The Seebeck coefficients of the fabrics also grew in absolute value with temperature, from minus 6.1 to minus 7.5 microvolts per kelvin for the lowest-concentration PR 24 fabric. Modeling this nonlinear behavior with an approach developed for doped multi-walled carbon nanotube mats suggested that the PR 24 fabrics are more likely to be negatively doped, consistent with electron donation from the cellulose fibers or the surfactant used in the ink formulation. The PR 25 fabrics, by contrast, retained a positive metallic term in the model, indicating a different doping character despite their similar structure.

Overall, the study demonstrates that despite meaningful structural differences between the two Pyrograf III grades, the resulting conductive textiles are remarkably similar in their thermoelectric properties, with the smaller-diameter, more crystalline PR 24 fibers holding a slight edge at high ink concentrations. The researchers suggest these n-type carbon nanofiber textiles could serve as reference materials for future studies combining them with other n-type organic or inorganic components, moving toward a new generation of flexible, wearable thermoelectric generators that draw power directly from the temperature difference between our bodies and the world around us.

Subject of Research: thermoelectric cotton textiles coated with Pyrograf III carbon nanofibers

Article Title: Comparative study of thermoelectricity in cotton fabrics coated by immersion with Pyrograf® III carbon nanofibers-based inks

Article References: Paleo, A. J., Abreu, J., Krause, B., González-Casal, S., Thomas, L., Cerqueira, M. F., Muñoz, E., & Pötschke, P. (2026). Comparative study of thermoelectricity in cotton fabrics coated by immersion with Pyrograf® III carbon nanofibers-based inks. Frontiers in Materials, 13, Article 1933448. https://doi.org/10.3389/fmats.2026.1933448

Image Credits: AI Generated

DOI: 10.3389/fmats.2026.1933448

Keywords: thermoelectric textiles, carbon nanofibers, cotton fabrics, n-type conduction, smart textiles, wearable energy harvesting, dip coating, Seebeck effect, Comparative, thermoelectricity, cotton, fabrics

The key insight rests on a piece of physics that has been known for nearly two centuries. A Peltier module pumps heat when current flows through it, but the same device also behaves as a thermoelectric generator in reverse: whenever a temperature difference exists across its junctions, the Seebeck effect produces a voltage proportional to that gradient. In principle, the module can serve simultaneously as thermal actuator and thermal sensor, a concept researchers call intrinsic or self-sensing. Similar ideas have transformed other fields, from sensorless motor drives that estimate rotor position from electrical signatures to soft robotic actuators that infer their own deformation. The thermoelectric community has explored pieces of this picture before, including closed-loop systems that used a single module as both sensor and heat pump, but practical obstacles have kept self-sensing thermoelectric control out of mainstream use.

The central obstacle is electrical masking. The voltage applied to drive active cooling is enormously larger than the millivolt-scale Seebeck signal the module generates, so the two cannot be read through the same electrical interface at the same time. During powered operation, the thermal state of the cold side effectively goes dark. The Thai team’s answer is a time-multiplexed scheme: a relay-based switching circuit alternates the module between actuation mode, in which it is connected to a programmable power supply and driven with pulse-width-modulated current, and sensing mode, in which the drive is completely disconnected and the module’s open-circuit Seebeck voltage is measured directly. The relay switches in about a millisecond, and the researchers wait twenty milliseconds after disconnection before sampling, allowing switching transients to settle before the thermal signal is captured.

Of course, a sensor that only works intermittently cannot by itself sustain tight temperature regulation. The team’s second innovation fills the blind intervals with a digital twin. Using MATLAB and Simscape, they built a reduced-order electrothermal model that captures Peltier heat pumping, Joule heating, conductive heat transfer, and environmental thermal loading of the aluminum thermal mass attached to the module’s cold side. While the module is busy cooling, the digital twin propagates a prediction of the cold-side temperature. When the sensing window opens, the measured Seebeck voltage, converted to a temperature through an experimentally calibrated relationship, is compared against the twin’s prediction, and the resulting error drives a supervisory proportional correction that adjusts the next actuation cycle. The architecture is essentially a predictor-corrector loop operating under intermittent observability, a structure familiar from navigation systems that fuse inertial dead reckoning with periodic GPS fixes.

The experimental platform itself was designed with unusual care for thermal repeatability. The thermoelectric module was clamped between an aluminum block, which acts as the cold-side thermal load, and a heat sink immersed in an insulated ice bath held near zero degrees Celsius, approximating a constant hot-side boundary condition. A long-wave infrared thermal camera mounted above the assembly captured the spatial evolution of cooling, revealing the localized cold region and the gradual propagation of thermal gradients through the aluminum. Reference thermistors were installed for validation only; crucially, they were never used as the feedback signal in the self-sensing experiments, so the intrinsic Seebeck measurements had to stand on their own.

The characterization results are striking. Across pulse-width-modulated duty cycles from ten to ninety percent, the measured Seebeck voltage tracked the temperature difference across the module in an approximately linear fashion, with fitted slopes that remained broadly similar across operating conditions. The practical relationship did, however, exhibit an operating-condition-dependent offset, which the team attributes to non-ideal heat rejection, thermal accumulation near the hot side, Joule-heating effects, and residual electrical conditions after switching. Rather than treating this as a defect, the researchers built the offset into their sensing model as a duty-cycle-dependent term. The pooled fitted relationship achieved a coefficient of determination of 0.998, with a root-mean-square error of about 0.02 volts, demonstrating that even a deliberately imperfect, parasitic-laden signal can carry usable thermal-state information when properly calibrated.

Cyclic testing confirmed that the intermittent sensing does not destabilize cooling. The module was driven for ten seconds and then interrupted for a one-second sensing window, repeated over twenty-five cycles, and the overall cooling trajectory remained smooth and stable, a consequence of the aluminum mass’s thermal inertia and the system’s slow first-order dynamics. The duty cycle of the scheme corresponds to roughly ninety-one percent available actuation time, a modest sacrifice for continuous thermal observability. In single-event tests, the module was cooled to approximately minus five, minus ten, and minus fifteen degrees Celsius before actuation was cut, and in each case a clear Seebeck voltage appeared, decaying in step with the collapsing thermal gradient, exactly as the underlying physics predicts.

Closed-loop regulation experiments then put the full framework to the test against a reference temperature of minus sixteen degrees Celsius. The self-sensing controller reached its rise time in 44.61 seconds, faster than the 54.84 seconds achieved by a conventional continuous thermistor-feedback proportional controller running under identical thermal and electrical conditions. The trade-off was honest and the authors report it plainly: the faster initial response came with a settling time of 150 seconds and six percent overshoot, compared with 97.64 seconds and zero overshoot for the baseline. Against a fixed-duty open-loop response, however, the advantage was unambiguous. The open-loop system drifted steadily away from the reference as thermal accumulation and non-ideal heat rejection took their toll, while the model-assisted self-sensing loop held convergence near the target throughout operation, with the Seebeck-based corrections compensating for real-world disturbances the digital twin could not anticipate.

The authors are careful to frame the work as a feasibility demonstration rather than a finished product. The comparison baseline used proportional action only, without the integral and derivative terms a tuned PID controller would employ, and the study covered a single module at a single reference temperature without repeated-trial statistics or energy-consumption measurements. Every thermoelectric module has its own effective Seebeck coefficient, electrical resistance, and thermal conductance, so the sensing relationship and model parameters would need recalibration for any new device, and extending the approach to multi-module systems would raise questions of coordinated switching and thermal coupling between units. Long-term stability under thousands of cycles also remains untested.

Even with those caveats, the implications are considerable. Precision thermal management is becoming a bottleneck in fields where every cubic millimeter and every wire counts: cryo-cooled photonics, lab-on-chip diagnostics, satellite payloads, and the dense power electronics of electric vehicles. A cooling element that reports its own temperature, needing nothing more than a relay, an analog-to-digital converter, and a lightweight model running on a microcontroller, could strip cost and complexity out of such systems while improving reliability by eliminating a class of failure-prone components. The Chulalongkorn team’s predictor-corrector architecture shows one rigorous path to that future, and it suggests a broader principle worth watching: as models grow cheap enough to run alongside every actuator, the sensors we bolt on may increasingly give way to the signals our hardware was quietly producing all along.

Subject of Research: Intrinsic Seebeck self-sensing and model-assisted closed-loop temperature regulation of a thermoelectric Peltier cooling module

Article Title: Intrinsic Seebeck feedback for thermoelectric temperature regulation using a Peltier module

Article References: Chungsangsatiporn, W., Ajavakom, N., Chancharoen, R., & Phanomchoeng, G. (2026). Intrinsic Seebeck feedback for thermoelectric temperature regulation using a Peltier module. Results in Engineering, 32, Article 113446. https://doi.org/10.1016/j.rineng.2026.113446

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113446

Keywords: thermoelectric cooling, Peltier module, Seebeck effect, self-sensing, digital twin, temperature regulation, predictor-corrector control, intermittent observability, embedded thermal management, electrothermal modeling, sensorless control, Intrinsic

The two clusters at the heart of this work are chemical cousins with a fascinating structural relationship. M44, formally Au24Ag20 and protected by twenty-four 4-tert-butylphenylacetylide ligands and two chlorides, consists of a hollow icosahedral Au12 kernel wrapped in a fullerene-like Ag20 inner shell and capped by twelve exterior gold atoms. M81, with the formula Au43Ag38, is essentially a dimer of M44: two Au12 units fuse through an additional gold atom to form an Au25 kernel, while the two Ag20 shells merge with the loss of two silver atoms to create an Ag38 shell, all further wrapped by eighteen gold atoms and ligands. This face-fusion architecture is distinct from the side-on biicosahedral fusion seen in related clusters such as Au20Ag32 and Au23. Kohn-Sham molecular orbital analysis reveals that M44 behaves as a superatom resembling Au25, whereas M81 shows molecular-like character reminiscent of Au38, giving the researchers a matched pair in which electronic structure alone dictates whether a cluster leans toward photoluminescence or photothermal behavior.

The synthetic route is elegant in its simplicity. A mixture of chloro(dimethylsulfide)gold(I) and silver acetate dissolved in dichloromethane and methanol is combined with the deprotonated acetylene ligand, forming a reddish-yellow Au-Ag-alkynyl polymeric intermediate within minutes. Reduction with tert-butylamine borane, followed by reaction in the dark for one to three days, yields the clusters. Rather than relying on reaction timing and direct crystallization as in earlier work, the team employed thin-layer chromatography to separate the two products, which appear as distinct bands and were unambiguously identified by ultraviolet-visible-near-infrared absorption spectroscopy. Reaction time biases the outcome: quenching at roughly twenty-four hours favors M44, while extending to thirty-six or seventy-two hours promotes M81 growth. This post-synthetic purification strategy makes both clusters accessible from a single pot, a practical advantage for scaling up.

Optically, the monomer-to-dimer transformation produces dramatic consequences. M44 shows absorption peaks at 365 nanometers, convoluted features at 450 and 505 nanometers, and a broad band at 640 nanometers with tailing into the near-infrared. M81 displays additional bands at 640 and 765 nanometers and extends its absorption tail into the NIR-II window beyond 1000 nanometers. Most strikingly, the dimerization boosts the molar extinction coefficient at the long-wavelength band by roughly sixty times, from about 850 to 50,000 per molar per centimeter, consistent with the stronger transition dipole of the rod-like M81 structure. Both clusters emit photoluminescence in the NIR-II region, peaking near 1080 nanometers for M44 and 1320 nanometers for M81, the latter being among the rarest emission depths achieved by atomically precise clusters, previously matched only by gold quantum rods.

The photoluminescence quantum yields, however, are deliberately low: 3 percent for M44 and 1.5 percent for M81 in deuterated chloroform solution. Herein lies the counterintuitive design principle behind the work. Strong electron-phonon coupling quenches radiative emission and instead channels the absorbed photon energy into lattice vibrations, releasing it as heat. Transient absorption spectroscopy confirmed the underlying excited-state dynamics, with approximately one-picosecond components attributed to internal conversion and intersystem crossing, and longer nanosecond components matching the emission lifetimes of roughly 300 nanoseconds for M81. The long-lived triplet states even enable singlet oxygen generation in M44, as confirmed by the diagnostic 1275-nanometer singlet-oxygen phosphorescence signal under oxygen purging, while M81’s emission energy of 0.89 electron volts falls just below the 0.97-electron-volt gap needed for triplet-to-singlet oxygen energy transfer, so no such signal appears.

Systematic photothermal measurements across varying concentrations and optical power densities revealed a crucial and often overlooked trade-off: photothermal conversion efficiency and maximum temperature rise are counteracting parameters. At a modest optical power density of 0.5 watts per square centimeter and low absorbance, M81 achieved its record 91 percent efficiency with a temperature rise of 11.2 degrees Celsius. Pushing the power density to 2.5 watts per square centimeter dropped the efficiency to 38 percent but drove the temperature rise to 35.4 degrees Celsius, reaching a solution temperature of nearly 60 degrees. At higher absorbance, M81 heated toluene to 96.8 degrees Celsius. M44 followed similar trends with a peak efficiency of 69 percent and temperature rises up to 58 degrees. Under 1064-nanometer excitation deep in the NIR-II window, M81 still produced a measurable temperature rise despite weak extinction there, one of the first demonstrations of direct NIR-II photothermy from atomically precise clusters.

The solid-state results are where the numbers become genuinely startling. Thin films of the clusters embedded in a polymethyl methacrylate matrix, irradiated at 2.5 watts per square centimeter, reached temperature increases of 226 degrees Celsius for M44 and 155 degrees for M81. Such extreme heating in a solid film is precisely what is needed for energy-harvesting devices, since converting light into electricity requires a large, localized temperature gradient rather than the rapid dissipation that occurs in dilute solutions. Crucially, ultraviolet-visible-near-infrared spectra recorded before and after irradiation in all three media, organic solution, aqueous suspension, and solid film, showed no discernible change, confirming that both clusters survive months of ambient storage and repeated intense heating without degradation. The team also encapsulated both clusters in the FDA-approved poloxamer Pluronic F-127, producing water-soluble supraparticles around 150 nanometers in diameter with photothermal efficiencies near 50 percent in water, a regime well suited to photothermal tumor therapy.

The culmination of the study is the first thermoelectric generator application of atomically precise nanoclusters. The researchers deposited a layer of clusters onto the hot side of a commercial bismuth telluride thermoelectric generator and irradiated it with the same 808-nanometer laser, deliberately omitting any external cooling on the cold side to demonstrate electricity generation under minimal, device-integrated conditions. The Seebeck effect, in which a temperature gradient across a thermoelectric material drives directional charge-carrier diffusion, converted the cluster-generated heat into stable open-circuit voltages of 146 millivolts for M44 and 127 millivolts for M81, sustained over 900 seconds of continuous irradiation. Cycling experiments showed the devices could be switched on and off repeatedly without measurable degradation, and control experiments on bare thermoelectric generators produced no detectable voltage, confirming that the clusters themselves are responsible for the effect. Thermal paste at the interface yielded comparable voltages, indicating that direct cluster deposition already provides efficient thermal coupling.

The broader implications extend well beyond the laboratory laser experiments. More than half of the total solar radiation arrives in the near-infrared, yet conventional solar energy harvesting captures predominantly the ultraviolet and visible portions. The absorption profiles of both clusters overlap substantially with the solar spectrum, suggesting that direct sunlight could in principle drive the same photo-thermo-electric cascade. Encapsulating the clusters within heat-resistant carbon or silica shells could allow operation at even higher temperatures under focused solar illumination. Perhaps most conceptually significant is the reframing of electron-phonon coupling, typically dismissed as an unwanted quenching mechanism, into a deliberate design principle for near-infrared photonic energy harvesting. Key challenges remain, including broadening the absorption further into the shortwave infrared, improving thermal coupling, and demonstrating long-term operational stability, but this work establishes atomically precise nanoclusters as functional device components rather than mere solution-phase chromophores, and points toward a future in which quantum-sized metal clusters help reclaim the invisible half of sunlight.

Subject of Research: Near-infrared photothermal conversion and photo-thermo-electric energy generation using atomically precise bimetallic gold-silver nanoclusters

Article Title: Near‐Unity Photothermal Conversion in Bimetallic M44 and M81 Nanoclusters Enables NIR‐Driven Photo‐Thermo‐Electricity Generation

Article References: Sardar, A., Mazumder, A., Ji, W., He, G., Wang, Y., Luo, L., Chen, S., & Jin, R. (2026). Near‐Unity Photothermal Conversion in Bimetallic M 44 and M 81 Nanoclusters Enables NIR‐Driven Photo‐Thermo‐Electricity Generation. Advanced Science, Article e78061. https://doi.org/10.1002/advs.78061

Image Credits: AI Generated

DOI: 10.1002/advs.78061

Keywords: atomically precise nanoclusters, gold-silver nanoclusters, photothermal conversion, near-infrared, thermoelectric generator, Seebeck effect, NIR-II photoluminescence, solar energy harvesting, electron-phonon coupling, bismuth telluride, photothermal therapy, nanomaterials