Direct ink writing of 3D thermoelectric architectures for fabrication of micro power generators

Fredrick Kim Ulsan National Institute of Science and Technology (UNIST) Seong Eun Yang Ulsan National Institute of Science and Technology (UNIST) Hyejin Ju Ulsan National Institute of Science and Technology (UNIST) Seungjun Choo Ulsan National Institute of Science and Technology (UNIST) Jungsoo Lee Ulsan National Institute of Science and Technology (UNIST) Gyeonghun Kim Ulsan National Institute of Science and Technology (UNIST) Soo-Ho Jung Korea Institute of Materials Science Suntae Kim Ulsan National Institute of Science and Technology (UNIST) Chaenyung Cha Ulsan National Institute of Science and Technology (UNIST) Kyung Tae Kim Korea Institute of Materials Science (KIMS) Sangjoon Ahn Ulsan National Institute of Science and Technology (UNIST) Han Gi Chae Ulsan National Institute of Science and Technology (UNIST) Jae Sung Son (  jsson@unist.ac.kr ) Ulsan National Institute of Science and Technology (UNIST)

The recent advancement in 3D printing technology has revolutionised the design and manufacturing of any arbitrary complex 3D structure 15,16 . Despite considerable progress in recent years, materials that can be used to print on a micrometre or nanometre scale are still mostly limited to photo-curable resins or their composites with inorganics, and printing is carried out by stereolithographic techniques 17 .
Alternatively, the extrusion-based direct writing process of colloid inks containing inorganic particles or precursors has attracted great interest because of its simplicity in processing, affordable equipment price, and a wider range of printable materials 18,19 . In this process, it is of paramount importance to optimise the viscoelastic properties of the applicable functional inks in a highly speci c range, in order to maintain structural integrity during printing and the functionality of the printed object 20,21 . The most conventional strategy of adding organic binders to secure viscoelasticity often leads to critical degradation of electrical or mechanical functionality of inorganic printed objects, especially TE materials, due to lowered sintering e ciency [22][23][24][25] . Recently, several approaches have been reported to fabricate TE modules by 3D printing of Bi 2 Te 3 , BiSbTe, SnSe, and skutterudites-based TE legs [22][23][24][25][26][27] . For example, inorganic ionic binders have been used to achieve moderate viscoelasticity in colloid inks and thereby realise the layer-wise deposition of 3D TE structures without degradation of TE performance 28,29 . However, the 3D printed TE structures developed thus far suffer from critically low resolution due to limited printability of inks and diminished functionality of the printed materials. Therefore, in order to seamlessly integrate into various microscale systems, it is imperative to attain high-resolution 3D printing of TE materials and modules.
In this study, we realised the 3D direct ink writing of (Bi, Sb) 2 (Te, Se) 3 -based p-and n-type inks to build arbitrarily shaped 3D TE architectures with high aspect ratios toward the fabrication of high-power μ-TEGs (Fig. 1a). (Bi, Sb) 2 (Te, Se) 3 -based alloys are regarded as the best TE materials for operations at temperatures less than 100 °C [30][31][32][33][34] , making them the most suitable candidates for power generation from low-grade heat dissipated from integrated systems. We precisely engineered various characteristics of the TE particles, such as overall size, size distribution, and surface charge states to achieve an extremely high viscoelasticity of the TE inks without using organic rheological modi ers. Based on this rheological design of inks, we demonstrated the direct 3D writing of the inks to print 3D laments in a dimension-controlled manner, fabricating highly performant 3D TE architectures such as arches and lattices standing on a substrate by computer-aided model (CAD)-based design. To show the feasibility of the current direct writing process, we fabricated μ-TEGs chipped with the direct written 3D TE legs on a substrate and characterised their power performance, thereby demonstrating that the proposed process is the most cost-effective and rapid method of manufacturing high-performance μ-TE modules.

3D direct ink writing of TE materials
The system used for the extrusion-based 3D direct ink writing of TE materials (Fig. 1a) consists of a nozzle connected to a syringe containing viscoelastic TE ink, which is extruded by a pneumatic pressure controller that is moved in the x, y, and z axis according to a CAD-based pre-design. Considering the rheological properties of the inks and the TE properties of the sintered materials, p-type Bi 0.55 Sb 1.45 Te 3 with a Sb 2 Te 4 2--based chalcogenidometallate (ChaM) content of 25 wt% and n-type Bi 2 Te 2.7 Se 0.3 with a ChaM content of 10 wt% were chosen for 3D-printing experiments. The detailed rheological and TE properties are discussed in the following sections. Standing TE laments were vertically printed on a Si/SiO 2 substrate (Supplementary Videos 1). The laments exhibited excellent structural retention with smooth lateral surfaces in a single pass. Moreover, various printing parameters such as the dispensing pressure and nozzle diameter were controlled, and thereby, the diameters of TE laments were precisely controlled from 180 μm to 810 μm at the sintered states ( Fig. 1b and 1c). This shows the diverse applicability of the current process in multiple scales ranging from micrometres to millimetres. Optical microscopic (OM) images (Fig. 1f) revealed that the written laments with various diameters had the uniform line widths with a standard deviation less than 10 μm (Supplementary Fig. 1). Moreover, the maximum aspect ratios achieved by the written laments reached 4.9-9.4, depending on the diameter ( Fig. 1d and 1e). This capability to print highly anisotropic 3D TE laments by direct writing made it possible to build complex 3D architectures. As an example, Fig. 1g presents an arched architecture consisting of junctional p-and n-type TE legs fabricated by the direct writing process, where the uniformity in line width is not signi cantly changed at bent laments or junctions. Moreover, a 3D lattice was built by the layer-by-layer deposition of TE laments ( Fig. 1h and 1i, Supplementary Video 2). At the junctions, individual laments maintained their primary architectures without merging or thickening, which demonstrates the structural retention of the inks (Fig. 1j). These examples of complex 3D structures indicate the potential of using the proposed process to fabricate p-n junctions in 3D patterned TE semiconductors and 3D porous TE structures for controlled thermal transport.
When the as-printed structures were heat treated, the TE particles in the 3D structures were well consolidated by the sintering promotion effect of the ChaM additives, leading to the formation of effectively sintered grains in the microstructure and robust 3D structures at the macroscale 28 . The OM and scanning electron microscopy (SEM) images of the sintered laments in the printed 3D lattice ( Fig.  1k and 1l) revealed that the TE laments maintained their primary architecture. Moreover, the TE laments at the junction were well fused together without crack formation (Fig. 1m). The high-resolution SEM images of p-and n-type TE laments ( Supplementary Fig. 2) showed grain sizes of several tenths of micrometres, which further indicated that the grain growth was effective. Although signi cant volume shrinkage is unavoidable in 3D-printed structures, high reproducibility was obtained for all the samples in all directions. Typically, the volume contracted by 36% during sintering at 450 °C for 30 min; these sintering conditions are required to obtain a competitive TE performance of materials. Thus, we could design the nal 3D architecture according to the CAD-based pre-design.

Ink design and rheological properties
Controlling the viscoelastic properties of inks is of paramount importance for securing reliable ow through ne nozzles and ensuring the structural integrity of the printed lament. Our group previously demonstrated that the addition of ChaM inorganic ions provides the required viscoelasticity to TEparticle-containing colloidal inks, thereby facilitating the shape-conformable 3D printing of TE materials by layer-wise ink deposition 28 . However, the direct writing of TE inks without organic binders requires considerable improvement in the rheological properties to achieve smooth and reliable extrusion and faster structural recovery after printing. Hence, we attempted to optimally design the ink viscoelasticity by controlling the size, size distribution, and surface states of TE particles. Fig. 2a shows a schematic description of the strategy used in the current study: control of the TE particle (1) size and (2) size distribution, and (3) surface oxidation of particles to obtain a stable ink system using ChaM. Fig. 2b and 2c, and Supplementary Fig. 3 show the particle size distribution with a median particle size and the effect of the distribution on the dynamic viscosities (η') of the corresponding TE particle colloids, respectively. A smaller average particle size and narrower particle size distribution produced a higher η' in inks because the effective volume of a smaller particle system was greater. A wider distribution in the particle size may lead to aggregation, which can result in clogging during printing [35][36][37][38][39] . Accordingly, the shear stress dependencies of the storage (G') and loss (G") moduli of colloidal inks (Supplementary Fig.   4) were measured to evaluate the linear viscoelasticity region and phase stability, respectively. The smallest TE particle system (D median =4.378 μm) exhibited higher critical stress value (30.8 Pa) than those of the larger particle systems (D median =8.759 μm and 17.276 μm; 4.2 and 1.5 Pa, respectively), indicating the ability to sustain higher shear stress without losing the structural integrity of the ink. In addition, higher G" after critical stress may also suggest that the former system possesses superior phase stability as compared to those of latter cases 40 . It should be noted that regardless of particle size, the colloidal inks all demonstrated higher G' than G'' at a wide range of shear stress, indicating their highly viscous semisolid behaviour ( Supplementary Fig. 4).
Our previous research showed that the addition of ChaM is unavoidable for obtaining high-quality TE devices by sintering, where voids are lled and the particles are structurally transformed from molecular ions to the crystalline phase 28 . In the present study, we found that as-synthesised non-oxidised TE particles have weak negative surface charges (Fig. 2d). The inks composed of these non-oxidised TE particles are inherently unstable even in the presence of the counter ions of ethylenediammonium in the ChaM anions, which may screen the electrostatic surface charge on the TE particles ('screen effect') and induce irreversible aggregation 41 . To overcome this problem, controlled surface oxidation and the incorporation of ChaMs were carried out to investigate their effect on the viscoelasticities of various inks. The degree of oxidation of the TE particles was investigated by ζ-potential measurement, Fourier transform infrared (FTIR) spectroscopy, and X-ray photoelectron spectroscopy (XPS) (Fig. 2d and 2e, Supplementary Figs. 5 and 6). In both p-and n-type, the X-ray diffraction (XRD) patterns of the oxidised particles were identical to those of the non-oxidised ones, suggesting that the oxidised surface layers were very thin or amorphic ( Supplementary Fig. 7). However, the FTIR spectra of the oxidised particles clearly showed a peak indexed to OH stretching, which indicated surface oxidation, and this observation was further supported by the XPS spectra, which showed oxide peaks. More importantly, the surface charges of TE particles became neutral by the oxidation, minimizing the screen effect caused by the ChaM additives (Fig. 2d). Consequently, the viscoelastic properties of the oxidised TE particles containing ChaM showed dramatic enhancement in various rheological properties, whereas those of the nonoxidised TE particles containing ChaM worsened owing to the screening effect. The quality of the ChaMcontaining inks was assessed by the rheological properties: (1) shear stress dependency of G' (Fig. 2f,  Supplementary Fig. 8), (2) dynamic viscosity in the static state (Fig. 2g, Supplementary Fig. 8), (3) initial slope of the increase in G' as a function of the shear rate (Fig. 2h, top), and (4) yield stress (Fig. 2h,  bottom). In addition, the structure recovery of inks after deposition is one of the most critical factors to ensure shape retention. To evaluate the structure recovery, three interval thixotropy tests (3ITTs) were comparatively performed for the oxidised and non-oxidised particle-based inks with the 25 wt% ChaM content (Fig 2i). 3ITTs are designed to measure the rate and degree of elastic recovery (G') after inducing structural disruption via increasing shear stress. It is very important to note that the oxidised TE particlebased ink exhibited reduced structure disruption upon high shear stress and immediate elastic recovery, as compared to those of the non-oxidised particle ink, suggesting the strong inter-particle interaction. These intriguing results con rmed that the surface oxidation coupled with ChaM signi cantly improved the 3D printability, by attaining both shear thinning and fast elastic recovery that are rarely demonstrated for extremely high viscous uids, especially for those without organic binders. This combination of attractive viscoelastic properties is expected to realise a highly-desired eld of binder-free, all-inorganic inks for precise direct writing process.

TE properties of direct-written 3D structures
The remarkable sinterability observed in direct-written TE laments was clearly re ected in their excellent TE properties. The XRD spectra of both p-and n-type particles showed patterns corresponding to bulk Bi 0.5 Sb 1.5 Te 3 and Bi 2 Te 2.7 Se 0.3 with slight peak shifts, indicating the effective sintering of materials into a single phase (Fig. 3a). We varied the ChaM content for both p-type Bi 0.55 Sb 1.45 Te 3 and n-type Bi 2 Te 2.7 Se 0.3 particles, and the highest gure-of-merit (ZT) values were observed for 25% and 10% ChaM content for the p-and n-type particles, respectively ( Supplementary Fig. 11). Accordingly, the characterisations and the related discussion were described for the corresponding samples. The roomtemperature electrical conductivities and Seebeck coe cients of the direct-written TE laments were 607.7 S m −1 and 191.13 μV K −1 (p type) and 793.1 S m −1 and 111.7 μV K −1 (n type), respectively ( Fig. 3b  and 3c), which are comparable to the reported values for (Bi, Sb) 2 (Te,Se) 3 -based bulk alloys with corresponding compositions. Simultaneously, we comparatively characterised the TE properties of 3Dprinted p-and n-type TE bulk cuboids that were 12.7 mm wide and 1.0-2.0 mm thick and found that their electrical conductivities and Seebeck coe cients were identical to those of the TE laments within the measurement error range (10%). Thus, the homogeneous TE properties of the 3D-printed samples were independent of their dimensions, which demonstrated the expansibility of the current process to materials ranging from micrometre-scale single laments to centimetre-scale 3D bulks obtained by fusing layerwisely deposited TE laments without the loss of TE performance.
The temperature dependence of TE properties at 300-525 K was characterised using 3D-printed TE bulk cuboids because of the di culty involved in measuring a micro-scale single lament. The electrical conductivities of the p-and n-type samples decreased with increasing temperatures, indicating the degenerate behaviour of semiconductors. Moreover, the Seebeck coe cients of both the samples generally increased with increasing temperatures. The Seebeck coe cient of the p-type samples was 196.6 μV K −1 at room temperature with a peak value of 216.3 μV K −1 at 423 K, whereas that of the n-type samples was 110.1-132.0 μV K −1 over the entire measured temperature range (Fig. 3d). The p-and n-type samples had carrier mobilities of 197.6 and 81.7 cm 2 V −1 s −1 , respectively, characterised by the Hall effect measurement. These high carrier mobilities are comparable to the reported values of (Bi, Sb) 2 (Te, Se) 3 -based bulk alloys with corresponding compositions and are responsible for the high electrical conductivity of the 3D printed samples 33,42 . The carrier concentrations of the p-and n-type samples were 2.1 10 19 and 6.3 10 19 cm −3 , respectively. Since the Seebeck coe cient is reciprocally proportional to the carrier concentration, the relatively low Seebeck coe cient of the n-type sample compared with the corresponding bulk values is attributed to the high carrier concentration 33 . Thus, the p-and n-type samples had high TE power factors of 2.6 and 1.0 mW m −1 K −2 , respectively, at 300 K ( Supplementary  Fig. 12).
The temperature-dependent thermal conductivity of the p-type and n-type samples was 0.74-1.15 W m -1 K -1 over the entire measured temperature range (Fig. 3e), which was signi cantly reduced in comparison to 1.5-2.5 W m -1 K -1 of bulk Bi 2 Te 3 -based materials 42 . This reduction in the thermal conductivities is attributed to phonon scattering at the pore sites of the printed TE materials, which had relative densities of 82% and 72% for the p-and n-type samples, respectively. Because the as-printed samples were sintered under pressure-less conditions without pre-compaction, 20%-30% porosity was unavoidable. These  Supplementary Fig. 13).
These promising electrical and thermal properties led to markedly high ZT values for the p-and n-type 3Dprinted samples. The p-and n-type 3D-printed samples had ZT values of 0.92 and 0.37, respectively, at room temperature (Fig. 3f). Furthermore, the maximum ZT values of 1.03 and 0.50 were achieved at 350 and 420 K for the p-and n-type TE materials, respectively. These values are comparable to those obtained for typical Bi 2 Te 3 and BiSbTe ingots (ZT ≈ 0.8-1.0) 48 , and enhanced from the reported value of 0.9 for the 3D printed BiSbTe materials 28 . To the best of our knowledge, the ZT values reported in this paper are one of the highest for TE materials produced from TE inks or pastes [22][23][24]27 .
Fabrication and power measurement of μ-TEG When combined with traditional patterning processes, our ink writing technique can facilitate the heterogeneous integration of TE legs into patterned electrode arrays, thereby making it possible to fabricate μ-TEGs in a rapid, direct, and cost-effective manner. Moreover, the shape engineerability of the current process makes it possible to design TE legs for optimised thermal transfer, maximising the temperature gradient and resulting output power of μ-TEGs. We fabricated a μ-TEG composed of 3D-printed vertical TE laments with a diameter of 350 μm and a height of 1400 μm. The TE leg thickness of the fabricated μ-TEG is at least an order of magnitude greater than that of previously reported μ-TEGs prepared by the traditional MEMS, hot-pressing and dicing, and screen-printing processes 1,6,10,49,50 . Moreover, the p-and n-type TE legs were directly printed on a Si/SiO 2 substrate using pre-patterned Ag electrodes (width: 500 µm), which were fabricated by screen printing with a shadow mask ( Fig. 4a and 4b, and Supplementary Video 3). The printed TE legs were electrically connected using Ag adhesive. The top of the module was further covered with a water-captured hydrogel cooler for maintaining a temperature difference between the hot and cold sides. Polyacrylamide hydrogel, with a high amount of strongly bound water within the polymeric matrix, prevented premature water evaporation and allowed the sustained maintenance of water-cooling capability at a wide range of temperature.
The module resistance was 92 Ω at room temperature. Upon heating, the hot-side temperature gradually increased from room temperature to 121.8 °C, whereas the cold-side temperature remained at less than 48 °C (Supplementary Fig. 14). Accordingly, the temperature difference across the μ-TEG increased up to 82.9 °C, which is an order of magnitude higher than the typical temperature gradient observed in previously reported μ-TEGs upon heating 6,11,50 . The creation of such a high temperature difference in the current system is attributed to the high aspect ratio of the 3D-printed TE legs. As the temperature difference increased, the output voltage increased almost linearly and the output power increased quadratically ( Fig. 4c and 4d), achieving a maximum output voltage of 42.4 mV and a power of 2.8 μW at a temperature difference of 82.9 °C. Moreover, the maximum power density reached 479.0 μW cm -2 , which is su cient to run advanced wireless sensor networks [3][4][5]7,51 . These results clearly demonstrate the practicability of 3D direct writing for manufacturing high-performance μ-TEGs that can be integrated into electronic systems.

Conclusions
This study demonstrated direct ink writing using highly viscoelastic TE inks for designing micro-scale geometries of 3D TE materials and power generators. By optimising the particle size, size distribution, and surface states of ink materials, we developed all-inorganic TE inks with extremely high viscoelasticity, which is suitable for 3D direct writing. This particle design strategy can be extended to diverse classes of functional materials to synthesise 3D-printable inks without using organic rheological modi ers, which could facilitate the conservation of the primary properties of particles in 3D-printed structures, as indicated by the high ZT values obtained for direct-written TE laments. When combined with traditional lithographic processes, our direct writing approach can facilitate the integration of μ-TE modules into various emerging electronic systems, in which the modules can function as power generators for energyautonomous systems or as Peltier coolers for heat management. We consider the proposed approach to be an important step toward achieving a broad technological adaption of 3D-printable materials in electronic device manufacturing.
Synthesis of (Bi, Sb) 2 (Te, Se) 3  The Sb 2 Te 4 solution was stirred for over 24 h until the particles fully dissolved to produce a dark purple colour. Anti-solvent of acetonitrile was added to the solution with the volume ratio of 8:1, which is followed by centrifugation at 6615 g for 10 min to precipitate out the precursor. The precipitate was dried under vacuum for 30 minutes to produce Sb 2 Te 4 ChaM powder. A mixture of 4 g of the TE powder and desired amount of the ChaM were dispersed in 4 g of glycerol, and the solution was mixed with a planetary centrifugal mixer (ARM-100, Thinky) for 2 h to fully homogenise the ink.

Rheological properties of the ink
The rheological properties of TE inks were measured using a rotational rheometer (Haake MARS III, Thermo Scienti c) equipped with a coaxial cylinder geometry at 25 °C. The stress sweep tests were conducted over the range of 0.005 -300 Pa at a frequency of 1 rad s −1 and the frequency sweep tests were carried out before and after stress sweep test at a constant stress of 1 Pa to assess the phase stability of ink. The three-interval thixotropy tests (3ITTs) at various stresses also conducted as reported elsewhere 40 . The 1st interval was conducted under constant shear stress of 0.1 Pa (linear viscoelastic range) for 120 s, the 2nd interval was done at various stresses (1, 5, 10, 50, and 200 Pa) for 120 s. The 3rd interval was attempted at a shear stress of 0.1 Pa for 300 s in order to evaluate the structural recovery of ink after deposition during 3D printing.
Direct 3D writing process Direct writing was executed using an in-house pneumatic extrusion-based 3D printer with a nozzle connected to an ink reservoir, a pressure controller, a compressor unit and a three-axis stage with a stepper motor. Ink was loaded in a 5 ml syringe barrel (Saejong) with a metal needle of inner diameters ranging from 80 to 310 μm and deposited in a z-axis at 0.25 mm s -1 . The pressure control unit was adjusted to extrude a high aspect ratio lament.
Materials characterisation XRD patterns were collected by using X'pert Pro (PANalytical) with a Cu K X-ray source (wavelength of 1.5418 nm), operating at 40 kV and 30 mA equipped with an X'Celerator detector. The microstructure was characterised by using a eld-effect SEM (Nova-NanoSEM230, FEI and S-4800, Hitachi High-Technologies) operated at 10 kV. The optical microscopic images were obtained using an Olympus BX51M. The CCD images were obtained by a CCD camera (MicroPublisher 5.0 RTV, QImaging). Particle size distributions of the TE powders were determined with Laser particle size analyser (LS13 320, Beckman Coulter). ζ-potential of the particles was determined by a Zetasizer (Nano ZS, Malvern). The surface functional groups of the particles were obtained using a FT-IR spectrometer (Varian 670/620) in attenuated total re ectance (ATR) mode over the scanning range of 650-4000 cm −1 . XPS measurements were performed by an X-ray photoelectron spectrometer (K-alpha, Thermo Scienti c).

TE properties of 3D-printed samples
The room-temperature electrical conductivity of the laments was measured using the sheet resistance of the samples by a four-point method (Keithley 2400 multimeter controlled Lab trace 2.0 software, Keithley Instrument, Inc.) The room-temperature Seebeck coe cienct was obtained by using home-built set-up where the temperature gradient and the open circuit voltage were measured using a Keithley 2400 sourcemeter. Seebeck coe cient was calculated by the slope of the measured voltages versus the applied temperature differences ranging from 1 K to 15 K, which was measured by two T-type thermocouples.
The temperature-dependent electrical conductivity and the Seebeck coe cients of the bulk materials were measured simultaneously under Ar atmosphere in the temperature range from 300 K to 525 K using thermal analyzer (SBA458 Nemesis, Netzsch, Germany). A typical sample for measurement had cuboid shape with the dimension of 10×10×2 mm 3 . The thermal conductivity (κ) was calculated from the relationship κ = ρC p D, where ρ is the density, C p is the speci c heat capacity and D is the thermal diffusivity. The thermal diffusivity was measured in the temperature range 300 K to 525 K by using laser ash analyzer (LFA-457, Netzsch, Germany). The speci c heat capacity was determined using the rule of mixtures with the previously reported speci c heat capacity values of Bi 2 Te 3 and Sb 2 Te 3 52,53 . The density was estimated by measuring the volume and weight of the cuboid shape. The carrier concentrations and mobilities were measured at room temperature by using a Hall measurement system (HMS-8400, Lake Shore) with the magnetic eld of 10T.
Fabrication and output power characterization of the μ-TEG Ag electrodes were screen-printed on a SiO 2 wafer using Ag-containing adhesive (Pyro-Duct 597-A, Aremco, USA) as illustrated in Supplementary Fig. 15. On top of the Ag electrodes, 3 pairs of n-type and ptype TE laments were extruded out of a metal needle with the inner diameter of 210 μm. The printed laments were initially dried at 110 °C on a hot plate for 5 minutes, and annealed at 450 °C for 30 minutes in a tube furnace owing a gas mixture of 15% H 2 and 85% Ar. Each pair of TE laments was electrically connected in series and thermally in parallel by bridging Ag-containing adhesive on top of the annealed laments. To measure the temperature difference across each pair of TE laments, two T-type thermocouples were attached to the surface of the SiO 2 wafer (hot side) and on top of the Ag electrodes (cold side). The ceramic heater (10 mm×10 mm) was used as a heat source and the cold side temperature was maintained below 48 o C by attaching polyacrylamide (PAAm) hydrogel on top of the μ-TEG. The PAAm hydrogel was separately fabricated by radical polymerisation of 10 % (w/v) acrylamide and 0.5 % (w/v) N,N'-methylenebisacrylamide, with 0.2 % (w/v) ammonium persulfate and 0.2 % (w/v) N,N,N′,N′-tetramethylethylenediamine as co-initiators 54 . For the measurement of power generation, the TEG was connected to a Keithley 2400 at each temperature differences and the maximum power output (P) was calculated using P=V 2 /4R. Thermal silicone grease (YYTG-201, Youngyiel) was applied between the heater, and μ-TEG to minimise thermal contact resistance.

Data availability
The data that support the plots within this paper and other ndings of this study are available from the corresponding author upon reasonable request.