Flexible Bi₂Te₃/PEDOT nanowire sandwich-like films towards high-performance wearable cross-plane thermoelectric generator and temperature sensor array

Abstract

It is of significance to develop high-performance organic/inorganic hybrid thermoelectric (TE) materials for flexible cross-plane TE devices. In this work, we synthesize a flexible and free-standing sandwich-like TE film by depositing Bi₂Te₃ nanocrystals into a conducting porous PEDOT nanowire scaffold. The mechanically robust Bi₂Te₃@PEDOT nanowire film exhibits a record Seebeck coefficient of 266.4 μV K⁻¹ with a corresponding power factor of 740.2 μW m⁻¹ K⁻² and a zT value of 0.27 at room temperature. Afterwards, we build new types of wearable cross-plane TE generator (TEG) and TE-powered temperature sensor array with a reduced number of electric leads in comparison to traditional designs. The TEG shows a maximum output power density of 3.48 μW cm⁻² at a temperature difference of 17.8 K (predicted output power density of 70.6 μW cm⁻² at ΔT = 80 K), outperforming most of the organic and organic/inorganic based ones. The conformal TE-powered temperature sensor array exhibits high and stable thermal sensitivity with a resolution of 0.05 K.

---

1 Introduction

Conducting polymers (CPs) with excellent mechanical flexibility and good thermoelectric (TE) properties, especially the poly(3,4-ethylenedioxythiophene) (PEDOT) family, have become promising flexible TE materials for wearable heat-to-electricity conversion applications.^1–4 However, doped conducting polymers usually have low Seebeck coefficients (S), which affects the figure of merit (zT = S²σT/κ) more significantly than other parameters including electrical conductivity (σ) and thermal conductivity (κ). Various approaches were developed to enhance their TE properties, such as secondary dopants,^5–9 organic mixed solution,^10–12 and sequential treatments,^13,14 which are the common methods for Seebeck coefficient and/or electrical conductivity enhancement of PEDOT derivatives. Nevertheless, the TE properties of conducting polymers, especially the Seebeck coefficient, are still much lower than that of their inorganic counterparts.

Organic/inorganic composites are promising in designing high-performance flexible TE materials and wearable TE devices by taking advantages of the high Seebeck coefficient in inorganic alloys¹⁵ as well as intrinsic flexibility and low thermal conductivity of CPs. Besides carbon nanocrystals,^16–18 inorganic semiconductors (Te, Bi₂Te₃, Cu₂Se, etc.)^19–31 are usually physically or chemically mixed with CPs to form binary or ternary composite TE films with enhanced TE properties. For instance, a spin-coated PEDOT:PSS/Bi₂Te₃ nanowire composite TE film exhibits a Seebeck coefficient of 45 μV K⁻¹ and power factor of 205 μW m⁻¹ K⁻².²¹ A PEDOT/Ag₂Se/CuAgSe composite film shows a high Seebeck coefficient of 121.8 μV K⁻¹ and power factor of 1603 μW m⁻¹ K⁻².³¹ Regarding wearable TE devices, it is usually difficult to directly utilize the out-of-plane temperature difference in thin film based in-plane TE devices. Moreover, it is indispensable to develop high-performance and versatile TE modules (e.g., power generator and sensor), which will lay the cornerstone of multifunctional wearable TEs.^32–39 Hence, rational design and construction of high-performance, flexible TE materials and devices are to be addressed for practical applications.

In this study, we report a type of sandwich-like Bi₂Te₃ nanocrystal/PEDOT nanowire TE film by simply thermally depositing Bi₂Te₃ nanocrystals into a porous PEDOT NW scaffold with further thermal-shock treatment. Taking advantages of high TE properties of an inorganic alloy and flexible conducting polymer nanowire scaffold, the composite film exhibits a comparatively high TE performance and flexibility. Utilizing the out-of-plane temperature difference, the resultant flexible cross-plane TE devices exhibit intriguing performances of power generation and temperature sensing, showing high potential in wearable electronics.

2 Results and discussion

2.1 Morphology characterization of p-type Bi₂Te₃@PEDOT NW films

As depicted in Fig. 1a, we simply prepared the Bi₂Te₃@PEDOT NW sandwich-like films by directly depositing Bi₂Te₃ powder equivalently on both surfaces of PEDOT NW scaffolds via thermal evaporation. Briefly, Bi₂Te₃ powder was fused to sublimate, and gaseous Bi₂Te₃ was transported from the evaporation source and re-nucleated on the surface of PEDOT NWs with tight interaction. Both surfaces of PEDOT NW scaffolds were equally deposited with Bi₂Te₃. After proper thermal-shock treatment with an input power of 41 W (3.5 A for 10 s), intensive and well-crystallized Bi₂Te₃ nanocrystals were formed.

Fig. 1 (a) Schematic illustration of the preparation process of Bi₂Te₃@PEDOT NW sandwich-like films. FESEM images of the surface view of (b) the PEDOT NW scaffold and (c and d) thermally deposited Bi₂Te₃@PEDOT NW films with 0.25 vol% and 21.5 vol% Bi₂Te₃. FESEM images of (e) the surface view and (f) the cross-sectional view with (g) EDS elemental mapping of the thermal-shocked 15 vol%-Bi₂Te₃@PEDOT NW film.

Fig. 1b shows the surface morphology of the as-synthesized PEDOT NW scaffold, which consists of highly ordered one-dimensional PEDOT NWs. Randomly overlapped PEDOT NWs endow the scaffold with a nanoporous structure, possessing advantages of light weight and excellent flexibility. With the high-vacuum thermal evaporation progress, as shown in Fig. 1c and S1,† Bi₂Te₃ powder first aggregates on the PEDOT NW surface and then penetrates the internal pores in the PEDOT NW scaffold. With increased deposition time, as shown in Fig. 1d, Bi₂Te₃ powder gradually fills the space between PEDOT NWs until the texture of the PEDOT NW scaffold is completely obscured, and then, the Bi₂Te₃ layer forms and becomes thicker on the PEDOT NW scaffold. Surface and cross-sectional FESEM images of the 15 vol%-Bi₂Te₃@PEDOT NW film after thermal-shock treatment are shown in Fig. 1e and f. In Fig. 1e, we can see that both large and small Bi₂Te₃ nanocrystals are formed and randomly distributed after thermal-shock treatment. Particle sizes of Bi₂Te₃ nanocrystals (Fig. S3†) are concentrated in the range of 100–250 nm. The introduction of highly dense nano/micro-structured Bi₂Te₃ nanocrystals is believed to provide fast carrier transport and improved electrical performance.^40–43 As shown in Fig. 1f, the Bi₂Te₃@PEDOT NW composite exhibits a sandwich-like assembled structure with a clear and tightly bonded interface between highly ordered stacked Bi₂Te₃ nanocrystals and the nanoporous PEDOT NW scaffold (Fig. S4†). The energy dispersive spectroscopy (EDS) elemental mapping of the whole cross-section (Fig. 1g) shows the distribution of characteristic elements (S for sulphur in PEDOT, Te, Sb, and Bi for tellurium, stibium, and bismuth in Bi₂Te₃). Additionally, individual EDS patterns for each characteristic element shown in Fig. S5† clearly show that Bi₂Te₃ penetrates into the PEDOT NW scaffold, which is consistent with the observation from FESEM images in Fig. S1.†

The crystallographic structure of thermal-shocked Bi₂Te₃ was characterized with the X-ray diffraction (XRD) patterns, as shown in Fig. 2. The Bi₂Te₃@PEDOT NW film exhibits sharp peaks with a narrow full width at half maximum (FWHM), indicating that Bi₂Te₃ nanocrystals possess high crystallinity after thermal-shock treatment are successfully deposited on the PEDOT NW scaffold. Thermal-shocked Bi₂Te₃ can be identified as (015) oriented Bi_0.4Sb_1.6Te₃ nanocolumns, which is a single-phase crystallographic structure according to the standard reference pattern (JCPDS no. 65-3674).^40,41 This is also verified by the detected atom ratio of FESEM-EDS elemental analysis in Table S2.† Compared with the characteristic diffraction peak 2θ = 25.6° of the as-synthesized PEDOT NW scaffold, the diffraction peak corresponding to PEDOT in the Bi₂Te₃@PEDOT NW film shifts to a lower diffraction angle (2θ = 22.3°). This probably originates from the thermal expansion of PEDOT chains during thermal shock treatment, resulting in enlarged spacing between conjugated chains and a shortened crystal correlation length along the π–π stacking axis.

Fig. 2 XRD patterns of the as-synthesized PEDOT NW scaffold and thermal-shocked Bi₂Te₃@PEDOT NW film. The orange one is the standard reference pattern of the (015) oriented Bi_0.4Sb_1.6Te₃ nanocolumns (JCPDS 65-3674).

2.2 Thermoelectric properties of p-type Bi₂Te₃@PEDOT NW films

For the purpose of optimizing the TE performance of p-type Bi₂Te₃@PEDOT NW films, the crucial TE properties including the Seebeck coefficient (S), electrical conductivity (σ), and power factor (PF) were investigated as a function of Bi₂Te₃ volume ratio (Table S3†).

As shown in Fig. 3a, both the S and σ display an almost upward trend with the increasing Bi₂Te₃ volume ratio. The pure PEDOT NW exhibits a small S of 22.4 μV K⁻¹, which is dramatically enhanced by further combination with Bi₂Te₃. The 15 vol%-Bi₂Te₃@PEDOT NW film shows a S of 266.4 μV K⁻¹, but this slightly declines to a S of 256.8 μV K⁻¹ in the 21.5 vol%-Bi₂Te₃@PEDOT NW film. A homogeneous and dense Bi₂Te₃ nanocrystal layer has much better TE performance than a PEDOT NW scaffold.⁴⁴ Theoretically, the higher the content of Bi₂Te₃, the better the TE performance of Bi₂Te₃@PEDOT NW films. However, compared with the 15 vol%-Bi₂Te₃@PEDOT NW film, the 21.5 vol%-Bi₂Te₃@PEDOT NW film suffers from a declining tendency. The reasons can be deduced as follows. For the 15 vol%-Bi₂Te₃@PEDOT NW film, a tighter interfacial contact can form between Bi₂Te₃ nanocrystals and PEDOT NWs during the thermal-shock process. Due to different work functions (φ) of Bi₂Te₃ nanocrystals and PEDOT NWs, as shown in Fig. S6† and 3b, numerous nanoscale boundaries generated between Bi₂Te₃ nanocrystals and PEDOT NWs may produce an effective energy filtering effect that can block the transport of low-energy carriers, leading to a higher mean transport energy (E_T) of the holes in the 15 vol%-Bi₂Te₃@PEDOT NW film. The S is thus enhanced because it is proportional to the difference between E_T and the Fermi energy (E_F), S ∝ |E_T − E_F|.^45,46 However, for the 21.5 vol%-Bi₂Te₃@PEDOT NW film, superficial Bi₂Te₃ nanocrystals tend to crack and exfoliate after the thermal-shock treatment according to the pre-experiment (Fig. S7†), which leads to an inferior interfacial contact state and a decline in the Seebeck coefficient of the 21.5 vol%-Bi₂Te₃@PEDOT NW film.

Fig. 3 (a) Seebeck coefficient, electrical conductivity, and power factor of p-type Bi₂Te₃@PEDOT NW films as a function of Bi₂Te₃ volume ratio. (b) Band structures of the PEDOT NW and Bi₂Te₃ nanocrystal with the band alignment at the interface. The energy barrier (E_barrier) scatters low-energy holes (blue ball) and selectively passes high-energy holes (red ball). Comparison of the (c) power factor and (d) zT at 300 K in this work with the data of organic/inorganic composite TE materials reported in the literature.

The σ increases almost linearly from the pure PEDOT NW (19.1 S cm⁻¹) to the 21.5 vol%-Bi₂Te₃@PEDOT NW film (104.3 S cm⁻¹), showing a σ enhancement of ∼5.5 times. This elevation in σ could be explained based on the following aspects. As analyzed with XRD patterns, the Bi₂Te₃ nanocrystal layer with high σ becomes gradually continuous and dense, playing an important role in facilitating the electrical transport in the Bi₂Te₃@PEDOT NW film as Bi₂Te₃ content increases. The high Bi₂Te₃ content in the 21.5 vol%-Bi₂Te₃@PEDOT NW film contributes to higher carrier concentration than that of the 15 vol%-Bi₂Te₃@PEDOT NW film. Hence the electrical conductivity of the Bi₂Te₃@PEDOT NW film continuously increases with increasing Bi₂Te₃ content, which is different from the variation of the Seebeck coefficient.

The power factor (PF = S²σ) of Bi₂Te₃@PEDOT NW films embodies the synergistic effect of S and σ, reaching a peak PF value of 740.2 μW m⁻¹ K⁻² in the 15 vol%-Bi₂Te₃@PEDOT NW film, which is ∼32 times higher than that of the pristine PEDOT NW film.⁴⁷ Moreover, Seebeck coefficients and power factors of organic/inorganic TE composites in reported literature are summarized in Table S4† and Fig. 3c, illustrating that the TE properties of the 15 vol%-Bi₂Te₃@PEDOT NW film in our work stands among a competitive level.

Regarding the figure of merit zT, the in-plane thermal conductivity (k) of the 15 vol%-Bi₂Te₃@PEDOT NW film was measured to be ∼0.82 W m⁻¹ K⁻¹ at 300 K by the self-heating 3ω method.^48,49 At 300 K, the k of the PEDOT NW scaffold was measured to be 0.98 W m⁻¹ K⁻¹, while the k of thermal-shocked Bi₂Te₃ was measured to be 0.62 W m⁻¹ K⁻¹,⁴⁴ which is much lower than that of a typical Bi₂Te₃ crystal (1.35–1.65 W m⁻¹ K⁻¹)^50,51 due to numerous internal pores filled with low heat-conducting air. According to the principle of thermal resistance in parallel, the k value of the Bi₂Te₃@PEDOT NW film at 300 K would be lower than that of both components. On the other hand, numerous nanoscale boundaries generated between Bi₂Te₃ nanocrystals and PEDOT NWs serve as phonon scattering centers and thus further reduce the total κ for the Bi₂Te₃@PEDOT NW composite. Thus, the measured k of 0.82 W m⁻¹ K⁻¹ is reliable. The zT value of the 15 vol%-Bi₂Te₃@PEDOT NW film can be calculated to be ∼0.27, as shown in Fig. 3d, also standing in a competitive level among organic/inorganic composite TE materials reported in the literature.

2.3 Mechanical flexibility of p-type Bi₂Te₃@PEDOT NW films

Besides the TE properties, the flexibility of composite TE films is also crucial for practical wearable applications, which enables TE electronics to maintain stable output performance under mechanical deformation.^52,53 For mechanical flexibility assessment, the 15 vol%-Bi₂Te₃@PEDOT NW film in our work was tested by bending along glass tubes with different radii. The resistances (R) of the bent Bi₂Te₃@PEDOT NW films were measured and normalized using the original resistance (R₀) as a reference. The relative resistance variation of the 15 vol%-Bi₂Te₃@PEDOT NW film at different bending radii is shown in Fig. 4a. R/R₀ slightly increases with decreasing bending radius (r) from 25 mm to 7.5 mm. As the bending radius dramatically decreases to 3 mm, R/R₀ dramatically goes up, which can be attributed to the breakage of the relatively rigid Bi₂Te₃ nanocrystal layer (Fig. S8†). However, R increases by only 4% at a bending radius of 7.5 mm compared to R₀, indicating good flexibility of the 15 vol%-Bi₂Te₃@PEDOT NW film. In addition, the cyclic bending test was also conducted as shown in Fig. 4b. At a bending radius of 7.5 mm, the 15 vol%-Bi₂Te₃@PEDOT NW film retains almost the same original resistance value in the first 100 cycles but the value increases vigorously after that. But the resistance is maintained well even after 200 bending cycles when the bending radius increases up to 15 mm.

Fig. 4 Resistance changes of the 15 vol%-Bi₂Te₃@PEDOT NW film as a function of (a) bending radius and (b) bending cycle at bending radii of 7.5 mm and 15 mm. Insets show the diagram of the bending test and the photograph of the bent 15 vol%-Bi₂Te₃@PEDOT NW film. Comparison of the resistance change in the 15 vol%-Bi₂Te₃@PEDOT NW film with that in composite films in the literature as a function of (c) bending radius and (d) bending cycle. ΔR/R₀ was set within 5%.

A PEDOOT NW scaffold consisting of highly ordered PEDOT NWs has abundant nanovoids, which provides large specific surface area for Bi₂Te₃ deposition. An interlocked Bi₂Te₃/PEDOOT NW interfacial contact can be formed, contributing to good structural maintenance during dramatical deformation and repetitive bending cycles. As shown in Fig. 4c and d, the mechanical stability of the 15 vol%-Bi₂Te₃@PEDOT NW free-standing film in this work is at a high level,^54–66 especially compared with the composite TE films on flexible conductive supporting substrates.^57,60,62,63

2.4 Thermoelectric power generation performance of the cross-plane power generator

Based on TE performance of Bi₂Te₃@PEDOT NW films, we fabricated a cross-plane TE power generator (TEG) as shown in Scheme S3.† Both p-type and n-type 15 vol%-Bi₂Te₃@PEDOT NW films were selected as TE leg materials. The Seebeck coefficient and electrical conductivity of the n-type 15 vol%-Bi₂Te₃@PEDOT NW film were measured to be −158.1 μV K⁻¹ and 112 S cm⁻¹, respectively (Fig. S9†). Fig. 5a shows the diagram of the cross-plane TEG (lateral size of 14 mm × 6 mm and thickness of 5 mm), in which 16 TE legs (8 p–n pairs) were embedded in porous PDMS, electrically connected with Ga₇₅In₂₅, and encapsulated with a PDMS cover.

Fig. 5 (a) Schematic illustration of the electrical series connection of 16 TE legs and operating mechanism when the top side is exposed to a heat source. (b) Output power density as a function of external resistance at different applied temperature differences. (c) Output voltage density and output power density as a function of circuit current at different applied temperature differences. (d) Comparison of the normalized output power density (specific power) of the cross-plane TEG in our work with the data reported in the literature including that of organic based, organic/inorganic based, and inorganic based TEGs integrated in textile or non-textile substrates.

As a given temperature difference (ΔT) is applied along the out-of-plane direction of the cross-plane TEG, the open-circuit voltage (V_oc) is linear to the applied ΔT as demonstrated in Fig. S11a.†V_oc values can be theoretically estimated from the equation V_oc = (|S_p| + |S_n|) × ΔT × N, where N is the number of p–n pairs. The measured V_oc can reach 54.7 mV at a ΔT of 17.8 K, which is close to the theoretically calculated value (60.5 mV at a ΔT of 17.8 K) (Explanatory note 1 and Table S5†). Fig. 5b and c show the output power density and output voltage density of the cross-plane TEG at different ΔT values, respectively. The maximum output power density P_max occurs as the internal resistance (r) is equal to the external resistance (R).^44,67 The P_max achieves a peak value of 3.48 μW cm⁻² at a ΔT of 17.8 K as R = 250 Ω. As shown in Fig. 5c, the output voltage also increases as the applied ΔT increases, showing a voltage density up to 65.1 mV cm⁻² at ΔT = 17.8 K. The circuit current is negatively linear to the output voltage, achieving a circuit current of 208.8 μA at ΔT = 17.8 K. A higher output voltage and power can be further realized by operating the cross-plane TEG at larger ΔT and preparing large-area cross-plane TEGs with more pairs of p-type and n-type TE legs (Explanatory note 2 and Table S6†). Provided that a ΔT of 80 K is applied, a corresponding P_max of 70.6 μW cm⁻² can be obtained (Fig. S11b†).

As shown in Fig. 5d, we normalized the reported output powers of various cross-plane TEGs^39,67–76 by dividing the area and the applied ΔT in unit of nW cm⁻² K⁻². Our cross-plane TEG shows a specific power of 10.7 nW cm⁻² K⁻², which is much higher than that of most organic and organic/inorganic based TEGs, and even some inorganic based TEGs. The higher power output performance is believed to originate from careful optimization of the TE properties of Bi₂Te₃@PEDOT NW films and the proper design of TEG. Moreover, we also calculated the density of TE legs in these cross-plane TEGs. The density of TE legs of our cross-plane TEG is up to 19 pieces per cm², which is just inferior to that of the doped-CNTY based TEG⁷⁵ and stands in a highly competitive level.

2.5 Thermoelectric-powered temperature sensor array

Due to the ultrahigh Seebeck coefficient, we also investigated the temperature sensing performance of the p-type 15 vol%-Bi₂Te₃@PEDOT NW film based sensor. As shown in Fig. S12a,† each TE leg, which is electrically connected with external copper electrodes and encapsulated with a PDMS cover, can act as an independent TE-powered temperature sensor. The thermovoltage (V_therm) can be used to assess the temperature-sensing response. Fig. S12b† presents the real-time V_therm of the sensor at different temperature differences within 1 K. Notably, even a small ΔT of 0.05 K is clearly detected with a V_therm of 13 μV. Fig. S12c and d† show the good stability of the sensor at different large temperature differences, with respect to the heating time and heating–cooling cycles. The stable and reproducible thermoelectric conversion enables sensitive detection of the TE-powered temperature sensor.

According to the aforementioned temperature sensing performance of one TE leg, a flexible cross-plane temperature sensor array consisting of 16 TE legs (16 sensing units) was fabricated. Fig. 6a shows the schematic diagram of the cross-plane temperature sensor array. The ends of all TE legs on the measured side were separately connected using conducting Ga₇₅In₂₅ with external copper electrodes (a₁–a₁₆). As for the other ends of all TE legs on the reference side, they were all covered and connected by one thin Ga₇₅In₂₅ layer instead of being separately connected with 16 individual Ga₇₅In₂₅ conducting leads, which was further connected with an external copper electrode (b). This design is different from others in the literature, resulting in significantly reduced numbers of involved electrodes. V_therm between a_n (n = 1, 2, 3…16) and b can be captured one by one when a ΔT is applied along the out-of-plane direction of the cross-plane temperature sensor array. As shown in Fig. 6b, the sensor array shows excellent flexibility. It can be bent and twisted, indicating its good conformability with curved surfaces, e.g. human skin.

Fig. 6 (a) Schematic diagram and (b) optical photographs showing mechanical flexibility, including bending and twisting deformation of the cross-plane temperature sensor array. (c and d) Demonstration of the temperature-sensing performance on curved surfaces by attaching the sensor array to bottles with different radii containing hot and cold water.

As stated in Explanatory note 3,† the temperature detected by each sensing unit can be calculated using T_m = 3.846V_them + T₀ + 0.533, where T₀ is the temperature of the reference side, e.g. room temperature, and T_m is the temperature of the detected point. Hence the temperature mapping was diagramed. First, we characterized the hot point detection of the cross-plane sensor array by covering different sensing units with a human finger. The corresponding temperature mappings are shown in Fig. S13,† which clearly demonstrate the temperature distributions, thus realizing the temperature detection and location. Moreover, practical temperature detection and location on curved surfaces is demonstrated. Fig. S14† shows the conformance and flexibility of the cross-plane sensor array. It can even be closely attached to a tube with a small radius of 4 mm. Then the cross-plane sensor array was attached to vials with different radii containing hot and cold water, where the ambient temperature is a T₀ of 21.8 °C. Fig. 6c shows the temperatures detected by 8 sensing units when the lower half of the sensor array is attached to a vial (radius of 20 mm) containing hot water (50 °C). Likewise, as shown in Fig. 6d, the temperatures of 8 cold points can be observed when the lower half of the sensor array is attached to a vial (radius of 10 mm) containing cold water (4 °C), indicating accurate temperature response and great potential in wearable temperature monitoring applications of the sensor array.

3 Conclusion

Herein a sandwich-like Bi₂Te₃ nanocrystal/PEDOT nanowire composite film is constructed through a facile thermal deposition process with further thermal-shock treatment. Taking advantage of high thermoelectric properties of inorganic alloy and the flexible polymer nanowire scaffold, the organic/inorganic composite film exhibits a comparatively high thermoelectric performance and mechanical flexibility, in which the optimal power factor is 740.2 μW m⁻¹ K⁻² (corresponding zT of 0.27) with an optimal Seebeck coefficient of 266.4 μV K⁻¹ at room temperature. Cross-plane thermoelectric devices based on the aforementioned composite film exhibit good performances of thermoelectric power generation and temperature sensing. Utilizing the out-of-plane temperature difference, the cross-plane thermoelectric generator shows a maximum power density of 3.48 μW cm⁻² at a temperature difference of 17.8 K outperforming most organic and organic/inorganic based cross-plane TEGs, while the conformal thermoelectric-powered temperature sensor array exhibits high and stable thermal sensitivity with a resolution of 0.05 K, laying the foundation for wearable power generation and temperature monitoring applications.

4 Experimental

4.1 Chemicals and materials

PEDOT nanowires (PEDOT NWs) were synthesized in our lab. P-type and n-type bismuth telluride (Bi₂Te₃, 99.99%) powder were provided by Changsha Santech Materials Co., Ltd (China). Poly(dimethylsiloxane) (PDMS) and the curing agent (Sylgard™ 184) were purchased from Dow Corning Corporation. N-Hexane was bought from Sinopharm Chemical Reagent Co., Ltd (China). Liquid metal (Ga₇₅In₂₅ with a resistivity of 29.4 μΩ cm and thermal conductivity of 26.43 W m⁻¹ K⁻¹) was purchased from Dongguan Dingguan Metal Technology Co., Ltd (China). Polylactic acid (PLA) tubes. All chemicals were of analytical grade and used as received.

4.2 Preparation of Bi₂Te₃@PEDOT NW sandwich-like films

Flexible Bi₂Te₃@PEDOT NW sandwich-like films were prepared as follows. Free-standing PEDOT NW scaffolds were prepared as previously reported.^77,78 P- and n-type Bi₂Te₃ powder were respectively evenly and compactly deposited in/onto PEDOT NW scaffolds through the high-vacuum thermal evaporation process (Fig. 1a). That is, Bi₂Te₃ powder was molten to sublimate, and gaseous Bi₂Te₃ was transported from the evaporation source and re-nucleated on the surface of PEDOT NWs, forming tight interaction. The two surfaces of PEDOT NW scaffolds were equally deposited with Bi₂Te₃. Thus, Bi₂Te₃@PEDOT NW sandwich-like films were obtained. Diverse contents of Bi₂Te₃ on PEDOT NW scaffolds were controlled by varying the evaporation time. The thickness of the thermally deposited Bi₂Te₃ layer can be recorded using a FTM-V film thickness monitor.

To obtain an alloyed Bi₂Te₃ nanocrystal for better electrical conductivity, a thermal-shock treatment using intense current was applied. Bi₂Te₃@PEDOT NW films were heated between two pieces of carbon paper (Toray, TGP-H-060, 62 mm in length, 12 mm in width and 0.19 mm in thickness) at a current of 3.5 A for 10 s under an Ar atmosphere (supply power was ∼37 W) in a glovebox (<1 ppm H₂O and <1 ppm O₂). As shown in Fig. S2,† the temperature of the carbon paper surface is ∼220 °C.

4.3 Characterization of Bi₂Te₃@PEDOT NW films

Surface and cross-sectional morphologies of Bi₂Te₃@PEDOT NW films were observed by using a field-emission scanning electron microscope (FESEM, Hitachi SU8010) with energy dispersive spectroscopy (EDS) elemental mapping analysis. The crystallographic structure of Bi₂Te₃@PEDOT NW films was examined by using an X-ray diffractometer (XRD, Bruker D8 Advance) using Cu Kα (λ = 1.5406 Å) radiation at 40 kV and 200 mA.

4.4 Thermoelectric property measurements of Bi₂Te₃@PEDOT NW films

The room-temperature thermoelectric properties of Bi₂Te₃@PEDOT NW films, including the Seebeck coefficient, electrical conductivity, and thermal conductivity, were measured by employing a self-built test system according to our previous work.^47,77 The Seebeck coefficient was calculated by using the formula S ∼ −ΔV/ΔT. The thermovoltage ΔV was recorded by using a Keithley 2182A and the temperature difference ΔT was calculated from the measured temperature data by using a Fotric 226 thermal infrared camera (Scheme S1†). The electrical conductivity was measured by using two Keithley 2400 current source meters using a standard van der Pauw (VDP) method (Scheme S2a†). Thicknesses of all samples were precisely determined from cross-sectional FESEM images, and at least five different locations were measured for each sample. The in-plane thermal conductivity was measured by the self-heating 3ω method (Scheme S2b and c†).^48,49 The variable-temperature thermal conductivity test system consists of an dewar (East Changing Technology Co. Ltd) with liquid nitrogen as the coolant and a TC-202 temperature controller.

4.5 Fabrication of the cross-plane thermoelectric generator

Porous PDMS (Fig. S10†) was adopted as supporting substrate in the cross-plane thermoelectric generator (TEG). P- and n-type 15 vol%-Bi₂Te₃@PEDOT NW films were cut into strips (2 mm × 23.5 μm × 5 mm). After thermal-shock treatment, as shown in Scheme S3,† p- and n-type TE legs were alternately fixed between two PDMS layers using a thin layer of semi-solid PDMS as the binder. The thickness of each PDMS layer and the space between two TE legs are 2 mm. The final cross-plane TEG with a thickness of 5 mm was obtained consisting of 16 TE legs (8 p–n pairs), which are electrically connected in series using liquid metal (Ga₇₅In₂₅) instead of conventional metal wires. For the sake of wearable safety and stability, Ga₇₅In₂₅ conducting wires were encapsulated with a transparent PDMS cover.

4.6 Thermoelectric power generation performance of the cross-plane TEG

A temperature difference is generated by placing the cross-plane TEG on a polyimide heating film (PI heater) with the other side exposed to air. The output voltage was recorded by using a Keithley 2182A. The output power was calculated by using the formula , where U is the open-circuit voltage, r is the internal resistance of the cross-plane TEG, and R is the external resistance which is altered with a variable resistor. The equivalent circuit of the cross-plane TEG supplying power is depicted in Scheme S4,† where V_output and I are the voltage on the external resistance and the circuit current.

4.7 Fabrication of the cross-plane temperature sensor array

For convenience and high conversion performance, only the p-type 15 vol%-Bi₂Te₃@PEDOT NW film was used in the cross-plane temperature sensor array. Similar to the aforementioned cross-plane TEG, p-type TE legs were embedded into a flexible porous PDMS supporting substrate. As for electrical connection, it is worth mentioning that all ends of TE legs on the same side of the device were integrated by a thin Ga₇₅In₂₅ layer, which was connected with an external copper electrode and ensures uniform temperature distribution of the reference side during the temperature monitoring. Likewise, conducting Ga₇₅In₂₅ wires were separately painted connecting the other ends of TE legs with copper external electrodes. Ga₇₅In₂₅ was also encapsulated with a transparent PDMS cover.

4.8 Practical application demonstration of the cross-plane temperature sensor array

To demonstrate temperature monitoring performance of the cross-plane temperature sensor array, the side covered with the whole Ga₇₅In₂₅ layer of the cross-plane temperature sensor array was placed in air, and the other side was attached to the surfaces of objects to be measured or monitored. When a certain ΔT is applied between the top and bottom ends of the cross-plane temperature sensor array, the thermovoltage (V_them) of each TE leg can be measured.

Data availability

The data supporting our findings in this work are available from the corresponding author K. Z. upon reasonable request.

Author contributions

K. Zhang conceived the idea and supervised the project. M. Du designed the experiments, prepared all PEDOT NW-based samples, conducted all characterization studies, measurements and data analysis, and drew all figures. K. Zhang and M. Du prepared the initial manuscript. All the authors discussed the results and contributed to reviewing the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are thankful for financial supports from the National Natural Science Foundation of China (51973034) and the Fundamental Research Funds for the Central Universities (2232022G01 and 19D110106).

References

1. Z. Fan, P. Li, D. Du and J. Ouyang, Adv. Energy Mater., 2017, 7, 1602116.
2. Z. Fan, D. Du, H. Yao and J. Ouyang, ACS Appl. Mater. Interfaces, 2017, 9, 11732–11738.
3. C. L. Hui Shi, Q. Jiang and J. Xu, Adv. Electron. Mater., 2015, 1, 0282.
4. H. Chen, Y. Chen, H. Zhang, W. Cao, C. Fang, Y. Zhou, Z. Xiao, J. Shi, W. Chen, J. Liu and W. Hong, Chin. Chem. Lett., 2022, 33, 523–526.
5. O. Bubnova, M. Berggren and X. Crispin, J. Am. Chem. Soc., 2012, 134, 16456–16459.
6. G. H. Kim, L. Shao, K. Zhang and K. P. Pipe, Nat. Mater., 2013, 12, 719–723.
7. Y. Jia, X. Li, F. Jiang, C. Li, T. Wang, Q. Jiang, J. Hou and J. Xu, J. Polym. Sci., Part B: Polym. Phys., 2017, 55, 1738–1744.
8. S. Zhang, Z. Fan, X. Wang, Z. Zhang and J. Ouyang, J. Mater. Chem. A, 2018, 6, 7080–7087.
9. O. Bubnova, Z. U. Khan, A. Malti, S. Braun, M. Fahlman, M. Berggren and X. Crispin, Nat. Mater., 2011, 10, 429–433.
10. J. Luo, D. Billep, T. Waechtler, T. Otto, M. Toader, O. Gordan, E. Sheremet, J. Martin, M. Hietschold and D. R. T. Zahn, J. Mater. Chem. A, 2013, 1, 7576–7583.
11. T. A. Yemata, Y. Zheng, A. K. K. Kyaw, X. Wang, J. Song, W. S. Chin and J. Xu, Front. Chem., 2020, 7, 870.
12. S. Zhang, Z. Fan, X. Wang, Z. Zhang and J. Ouyang, J. Mater. Chem. A, 2018, 6, 7080–7087.
13. S. Xu, M. Hong, X. Shi, Y. Wang, L. Ge, Y. Bai, L. Wang, M. Dargusch, J. Zou and Z. Chen, Chem. Mater., 2019, 31, 5238–5244.
14. A. K. K. Kyaw, T. A. Yemata, X. Wang, S. L. Lim, W. S. Chin, K. Hippalgaonkar and J. Xu, Macromol. Mater. Eng., 2018, 303, 1700429.
15. P. Li, X. Ai, Q. Zhang, S. Gu, L. Wang and W. Jiang, Chin. Chem. Lett., 2021, 32, 811–815.
16. J. Xie, J. Wang, Q. Yu, N. A. Khoso and J. Zhao, J. Textil. Res., 2018, 39, 50–55.
17. L. Liang, J. Fan, M. Wang, G. Chen and G. Sun, Compos. Sci. Technol., 2020, 187, 107948.
18. L. Liang, X. Wang, M. Wang, Z. Liu, G. Chen and G. Sun, Compos. Commun., 2021, 25, 100701.
19. S. Panigrahy and B. Kandasubramanian, Eur. Polym. J., 2020, 132, 109726.
20. Y. Lu, Y. Ding, Y. Qiu, K. Cai, Q. Yao, H. Song, L. Tong, J. He and L. Chen, ACS Appl. Mater. Interfaces, 2019, 11, 12819–12829.
21. W. S. Kim, G. Anoop, I. S. Jeong, H. J. Lee, H. B. Kim, S. H. Kim, G. W. Goo, H. Lee, H. J. Lee, C. Kim, J. H. Lee, B. S. Mun, J. W. Park, E. Lee and J. Y. Jo, Nano Energy, 2020, 67, 10427.
22. N. E. Coates, S. K. Yee, B. McCulloch, K. C. See, A. Majumdar, R. A. Segalman and J. J. Urban, Adv. Mater., 2013, 25, 1629–1633.
23. Z. Tian, H. Liu, N. Wang, Y. Liu and X. Zhang, J. Mater. Sci.: Mater. Electron., 2018, 29, 17367–17373.
24. J. Xiong, L. Wang, J. Xu, C. Liu, W. Zhou, H. Shi, Q. Jiang and F. Jiang, J. Mater. Sci.: Mater. Electron., 2016, 27, 1769–1776.
25. Y. H. K. Eun Jin Bae, K.-S. Jang, C. Lee, S. Y. Cho, E. J. Bae, Y. H. Kang, K. S. Jang, C. Lee and S. Y. Cho, Nanoscale, 2016, 8, 10885–10890.
26. W. Thongkham, C. Lertsatitthanakorn, K. Jiramitmongkon, K. Tantisantisom, T. Boonkoom, M. Jitpukdee, K. Sinthiptharakoon, A. Klamchuen, M. Liangruksa and P. Khanchaitit, ACS Appl. Mater. Interfaces, 2019, 11, 6624–6633.
27. B. Zhang, J. Sun, H. E. Katz, F. Fang and R. L. Opila, ACS Appl. Mater. Interfaces, 2010, 2, 3170–3178.
28. Q. Jiang, C. Liu, H. Song, J. Xu and Z. Zhu, Int. J. Electrochem. Sci., 2014, 9, 7540–7551.
29. Y. Du, K. F. Cai, S. Chen, P. Cizek and T. Lin, ACS Appl. Mater. Interfaces, 2014, 6, 5735–5743.
30. A. Rahman, A. Umar and H. Othman, Phys. E Low-dimens. Syst. Nanostruct., 2015, 66, 293–298.
31. Y. Lu, Y. Qiu, K. Cai, X. Li, M. Gao, C. Jiang and J. He, Mater. Today Phys., 2020, 14, 100223.
32. L. Huang, J. Chen, Z. Yu and D. Tang, Anal. Chem., 2020, 5, 2545–2554.
33. C. J. An, Y. H. Kang, H. Song, Y. Jeong and Y. C. Song, ACS Appl. Energy Mater., 2018, 2, 1093–1101.
34. F. Zhang, Y. Zang, D. Huang, C. Di and D. Zhu, Nat. Commun., 2015, 6, 8356.
35. C. K. Jeong, K. I. Park, J. H. Son, G. T. Hwang, S. H. Lee, D. Y. Park, H. E. Lee, H. K. Lee, M. Byun and K. J. Lee, Energy Environ. Sci., 2014, 7, 4035–4043.
36. L. E. Bell, Science, 2008, 321, 1457–1461.
37. K. Chatterjee and T. K. Ghosh, Molecules, 2021, 26, 3154.
38. M. Massetti, F. Jiao, A. J. Ferguson, D. Zhao, K. Wijeratne, A. Würger, J. L. Blackburn, X. Crispin and S. Fabiano, Chem. Rev., 2021, 121, 12465–12547.
39. C. Zheng, L. Xiang, W. Jin, H. Shen, W. Zhao, F. Zhang, C. Di and D. Zhu, Adv. Mater. Technol., 2019, 4, 1900247.
40. B. Poudel, Q. Hao, Y. Ma, A. Minnich, B. Yu, X. Yan, D. Wang, A. Muto, D. Vashaee, X. Chen, J. Liu, M. S. Dresselhaus, G. Chen and Z. Ren, Science, 2008, 320, 634–638.
41. S. Jimenez, J. G. Perez, T. M. Tritt, S. Zhu, J. L. Sosa-Sanchez, J. Martinez-Juarez and O. López, Energy Convers. Manag., 2014, 87, 868–873.
42. P. Y. Lee, J. Hao, T. Y. Chao, J. Y. Huang, H. L. Hsieh and H. C. Hsu, J. Electron. Mater., 2014, 43, 1718–1725.
43. X. Lu, Q. Zheng, S. Gu, R. Guo, L. Su, J. Wang, Z. Zhou, Y. Fan, W. Jiang and L. Wang, Chin. Chem. Lett., 2020, 31, 880–884.
44. Y. Zheng, X. Han, J. Yang, Y. Jing, X. Chen, Q. Li, T. Zhang, G. Li, H. Zhu and H. Zhao, Energy Environ. Sci., 2022, 15, 2374–2385.
45. O. Bubnova and X. Crispin, Energy Environ. Sci., 2012, 5, 9345–9362.
46. G. Kim and K. P. Pipe, Phys. Rev. B, 2016, 86, 6709–6717.
47. J. Zhang, K. Zhang, F. Xu, S. Wang and Y. Qiu, Compos. B Eng., 2018, 136, 234–240.
48. Z. L. Wang, D. W. Tang and W. G. Zhang, J. Phys. D: Appl. Phys., 2007, 15, 4686–4690.
49. Y. Sun, L. Qiu, L. Tang, H. Geng, H. Wang, F. Zhang, D. Huang, W. Xu, P. Yue and Y. S. Guan, Adv. Mater., 2016, 28, 3351–3358.
50. D. Park, S. Park, K. Jeong, H. S. Jeong, J. Y. Song and M. Cho, Sci. Rep., 2016, 6, 19132.
51. Q. He, W. Zhang, X. Liu and H. Song, Mod. Phys. Lett. B, 2022, 36, 28–29.
52. B. Russ, A. Glaudell, J. J. Urban, M. L. Chabinyc and R. A. Segalman, Nat. Rev. Mater., 2016, 1, 1–14.
53. C. Zheng, L. Xiang, W. Jin, H. Shen, W. Zhao, F. Zhang, C. a. Di and D. Zhu, Adv. Mater. Technol., 2019, 4, 1900247.
54. H. W. Ju, J. K. Sun and B. J. Cho, Energy, 2014, 73, 506–512.
55. C. Ou, A. L. Sangle, A. Datta, Q. Jing, T. Busolo, T. Chalklen, V. Narayan and S. K. Narayan, ACS Appl. Mater. Interfaces, 2018, 19580–19587.
56. L. Wang, Z. Zhang, L. Geng, T. Yuan, Y. Liu, J. Guo, L. Fang, J. Qiu and S. Wang, Energy Environ. Sci., 2018, 11, 1307–1317.
57. B. Wu, Y. Guo, C. Hou, Q. Zhang, Y. Li and H. Wang, Adv. Funct. Mater., 2019, 29, 1900304.
58. Y. Ding, Y. Qiu, K. Cai, Q. Yao, S. Chen, L. Chen and J. He, Nat. Commun., 2019, 10, 841.
59. D. Kong, W. Zhu, Z. Guo and Y. Deng, Energy, 2019, 175, 292–299.
60. K. Buks, J. Andzane, L. Bugovecka, M. V. Katkov, K. Smits, O. Starkova, J. Katkevics, A. Bērziņš, L. Brauna and V. Voikiva, Adv. Mater. Interfac., 2022, 9, 2200318.
61. J. Shi, X. Wu, X. Geng, L. Hu, F. Liu, W. Ao and C. Zhang, Chem. Eng. J., 2023, 458, 141450.
62. X. Chen, L. Feng, P. Yu, C. Liu, J. Lan, Y. Lin and X. Yang, ACS Appl. Mater. Interfaces, 2021, 13, 5451–5459.
63. C. Liu, D. L. Shan, Z. H. Shen, G. K. Ren and C. W. Nan, Nano Energy, 2021, 106380.
64. B. Xin, L. Wang, A. L. Febvrier, A. Elsukova, B. Paul, N. Solin and P. Eklund, ACS Omega, 2022, 7, 23988–23994.
65. J. Qin, Y. Du, Q. Meng and Q. Ke, Compos. Commun., 2023, 38, 101467.
66. D. Zheng, H. Jin, Y. Liao and P. Ji, Mater. Lett., 2023, 134019.
67. X. Yang and K. Zhang, ACS Appl. Mater. Interfaces, 2022, 14, 44704–44712.
68. M. K. Kim, M. S. Kim, S. Lee, C. Kim and Y. J. Kim, Smart Mater. Struct., 2014, 23, 105002.
69. J. Choi, Y. Jung, S. J. Yang, J. Y. Oh, J. Oh, K. Jo, J. G. Son, S. E. Moon, C. R. Park and H. Kim, ACS Nano, 2017, 11, 7608–7614.
70. M. Ito, T. Koizumi, H. Kojima, T. Saito and M. Nakamura, J. Mater. Chem. A, 2017, 5, 12068–12072.
71. Y. Shi, Y. Wang, D. Mei, B. Feng and Z. Chen, IEEE Rob. Autom. Lett., 2018, 3, 373–378.
72. T. Lee, K. T. Park, B. C. Ku and H. Kim, Nanoscale, 2019, 11, 16919–16927.
73. N. Wen, Z. Fan, S. Yang, Y. Zhao, T. Cong, S. Xu, H. Zhang, J. Wang, H. Huang and C. Li, Nano Energy, 2020, 78, 105361.
74. H. Xu, Y. Guo, B. Wu, C. Hou, Q. Zhang, Y. Li and H. Wang, ACS Appl. Mater. Interfaces, 2020, 12, 33297–33304.
75. Y. Zheng, Q. Zhang, W. Jin, Y. Jing, X. Chen, X. Han, Q. Bao, Y. Liu, X. Wang, S. Wang, Y. Qiu, C. Di and K. Zhang, J. Mater. Chem. A, 2020, 8, 2984–2994.
76. K. T. Park, T. Lee, Y. Ko, Y. S. Cho, C. R. Park and H. Kim, ACS Appl. Mater. Interfaces, 2021, 13, 6257–6264.
77. M. Du, X. Chen and K. Zhang, ACS Appl. Energy Mater., 2021, 4, 4070–4080.
78. X. Chen, W. Shi and K. Zhang, ACS Appl. Mater. Interfaces, 2020, 12, 34451–34461.

---

Footnote

† Electronic supplementary information (ESI) available: Supplementary schemes, figures, explanatory notes, and tables. See DOI: https://doi.org/10.1039/d3ta02876c

---