Introduction

Although conventional sensing devices generally use common secondary batteries, the time required to replace spent batteries and difficulties associated with their recycling have become problematic in recent years with the ongoing development of Internet of Things (IoT) devices and applications involving big data. Therefore, thermoelectric generators (TEGs) have attracted attention as maintenance-free power sources that generate electricity from temperature differences1,2,3,4,5,6,7,8. The application of TEGs to IoT devices requires operation at room temperature as well as passive cooling, necessitating good thermal contact with a heat source and efficient cooling performance. To realise good thermal contact, TEGs must have high bendability and stretchability with low thermal contact resistance because heat sources such as heat exhaust pipes, high-temperature pipes in plants, and the human body have curved surfaces. In addition, to achieve passive cooling for IoT applications, it is critical to reduce the cost of heat exchangers, given that the cost of heat exchangers in TEGs is a well-known problem9,10.

In recent years, numerous flexible TEGs have been proposed11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44. Flexibility includes factors such as bendability and stretchability, but the ease with which they are realised differs greatly. Although bendability can be realised easily by using a thin substrate14,15,16,24,30,31,34,40, realising stretchability is relatively more difficult. One approach to realising stretchability is to use elastomers, particularly stretchable silicone rubbers, as the substrate13,21,22,23,26,33,36,37,38,39,40. This approach can increase the contact area for curved heat sources; however, elastomers typically have low thermal conductivity and thus high thermal resistance. Consequently, achieving a large temperature difference within the thermoelectric (TE) elements is impossible with stretchable TEGs based on elastomer substrates. Another approach to realise stretchability is to use organic TE materials; however, these materials have lower TE performance compared with inorganic TE materials2,3,16,30,32,33,34,35.

We propose a new configuration for TEGs with the following advantages: simple fabrication based on “kirigami,” versatility in terms of the TE element shape, high TE output based on conformable thermal contact and cooling surfaces and stretchability. Specifically, we propose a pop-up kirigami-type TEG (hereafter, pop-up kirigami TEG) that uses a kirigami structure for the TEG substrate, as shown in Fig. 1a. The individual TE elements are conveniently mounted in a flat state, and the pop-up kirigami TEG generates the power in a pop-up state, giving it stretchability as well as conformable thermal contact and cooling surfaces. Although several stretchable kirigami-type TEGs capable of following the surfaces of curved heat sources have been reported18,19, they have less thermal contact area with the heat source surface, leading to high thermal resistance between the heat source surface and the TEG, thereby preventing the realisation of high output power.

Fig. 1: Schematic illustrations of the proposed pop-up kirigami TEG.
Fig. 1: Schematic illustrations of the proposed pop-up kirigami TEG.The alternative text for this image may have been generated using AI.
Full size image

a Schematic illustrations of the array-type pop-up kirigami TEG proposed in this paper. b Fabrication process of the pop-up kirigami TEG. c Structural analysis of pop-up deformation. d Photographs of the array-type pop-up kirigami TEG during bending deformation and stretching deformation.

In addition, most conventional TEGs can be categorised as either out-of-plane or in-plane1,2,3,4,5,6,7,8,9,10,11. Out-of-plane TEGs are usually used for bulk TE materials, while in-plane TEGs are commonly used for thin-film TE materials. Because the TE elements are mounted in a flat state, as previously mentioned, the pop-up kirigami TEG proposed in this paper imposes no restrictions on the shape of the TE element, making it possible to use both bulk and thin-film TE materials. When using thin-film or thin rod-shaped TE elements, the TE elements can be difficult to mount and may be easily damaged in an out-of-plane type TEG. In contrast, in a pop-up kirigami TEG, thin-film or thin rod-shaped TE elements can be deposited or mounted in a flat state and then used in a pop-up state. Furthermore, the three-dimensional substrate obtained after popping up the kirigami structure provides a stretchable cooling surface, making it possible to achieve passive cooling at a low cost. The pop-up kirigami structure reported in this paper provides bendability and stretchability solely by cutting the kirigami pattern, even for non-stretchable substrate materials such as metals. This is an advantage that neither out-of-plane TEGs nor in-plane TEGs possess.

The innovative pop-up kirigami TEG proposed in this paper exhibits both good flexibility and high performance. We investigated the effects of pop-up deformation of the pop-up kirigami structure, bending and stretching deformation, substrate size, TE element size and heat source shape on the power generation performance of the pop-up kirigami TEG. Finally, we demonstrate the usefulness of the fabricated pop-up kirigami TEG for monitoring changes in body temperature and achieving TE power generation.

Results

Design and fabrication of the pop-up kirigami TEG

The design and fabrication of the pop-up kirigami TEG are illustrated in Fig. 1. The pop-up kirigami TEG proposed in this paper simultaneously solves three major problems pertaining to stretchable TEGs, namely, (i) the realisation of good thermal contact with a curved heat source, (ii) the presence of a heat sink on a stretchable TEG and (iii) the mounting of TE elements in a stretchable TEG. With respect to the first problem, two important factors for achieving good thermal contact with a curved high-temperature heat source are the contact area with the heat source and the thermal resistance of the substrate. Regarding the former, it is desirable to increase the contact area between the heat source and the TEG so as to obtain a large temperature difference within the TE elements. In the case of conventional TEGs, the substrate material is typically a hard ceramic plate, which leads to an extremely small contact area (point contact or line contact) when installed on a curved heat source2,7,8,10,11. Our pop-up kirigami TEG uses a polyimide/copper substrate that is not inherently stretchable. Although a polyimide layer was applied for insulation, it was extremely thin at only 50 µm, and its thermal resistance was lower than that of a typical elastomer. Regarding the second problem, although conventional TEGs require heat sinks that are considerably larger than the TEG itself, stretchable TEGs are expected to utilise stretchable heat sinks; however, there are very few examples of the successful implementation of this design strategy. In our pop-up kirigami TEG, the parts playing the role of heat sinks are located within the pop-up kirigami structure, such that a stretchable heat sink is already included in the design without the need for additional components. Finally, regarding the third problem concerning the mounting of TE elements in stretchable TEGs, typical TEGs are specifically designed for either thin-film or bulk TE elements20,38,39. By contrast, for our pop-up kirigami TEG, the flexibility of the entire TEG is achieved through the careful design of the TEG structure, such that the TE elements themselves do not need to be flexible. This greatly expands the range of TE materials and material shapes that can be used with the current design concept. In other words, stretchable TEGs can be realised using the high-performance and hard TE elements that have been developed in previous research.

Next, the deformation resistivity of the pop-up kirigami TEG mounted with the TE elements will be discussed. The pop-up kirigami TEG developed in this paper has a structure in which the TE elements are mounted on a sheet substrate with an L-shaped cut pattern (see Fig. 1a). Commercially available BiTe-based TE wafers with a thickness of 1 mm (Toshima Co., Ltd.; p-type: Bi0.3Sb1.7Te3; n-type: Bi2Te3 + Ru) were used. Here, the notation “+Ru” indicates a Ru additive. The Seebeck coefficients of the p-type and n-type TE elements were 152.5 and −157.5 μW/K, respectively. The detailed TE properties of the TE elements are presented in Supplementary Table 1. Upon bringing the left and right electrodes closer to each other, the pop-up kirigami structure deforms from a flat shape to a three-dimensional shape (see Fig. 1b). In contrast to previously proposed kirigami-type TEGs18,19, our pop-up kirigami TEG features an electrode surface that can achieve good thermal contact with the heat source surface, thus greatly reducing the thermal resistance between the pop-up kirigami TEG and the heat source. A simple example of a configuration that can generate electricity while ensuring flexibility is the bellows structure depicted in panel (i) of Fig. 1c. However, in the bellows structure, the radius of curvature at the apex is extremely small, resulting in high tensile stress in this region. This leads to plastic deformation of the substrate and ultimately to breakage. On the other hand, as illustrated in panel (ii) of Fig. 1c, if a substrate lacking sharp corners is bent with a large radius of curvature, the TE elements cannot be mounted because there are no flat areas. To address these problems, a pop-up kirigami structure with an additional beam in the middle was adopted, as shown in panel (iii) of Fig. 1c. This pop-up kirigami structure makes it possible to create a structure with flat areas despite the lack of sharp corners and a large radius of curvature. Even when the pop-up kirigami structure rises mostly in the height direction from the flat state, the angle at the edge of the cut line remains small at 23.5° at the maximum height. Therefore, no fracture occurs at the cut-line edges. A finite element analysis using COMSOL Multiphysics® was performed to analyse the strain concentration at the cut-line edges. The cut line width of the TEG was set to 0.1 mm, consistent with the fabricated TEG. A deformation was applied to reduce the distance between the two electrodes of the pop-up kirigami TEG from 37 mm to 1 mm. As a result, as shown in panel (v) of Fig. 1c, the maximum strain at the cut-line edge was 1.3%, which is below the failure strain. Therefore, the pop-up kirigami structure adopted in this paper was expected to be suitable for use in electronic devices. Furthermore, this pop-up kirigami structure possesses regions that do not deform even upon bending or stretching, such that it can be used with not only thin-film TE materials but also bulk TE materials. In addition to this high bendability and stretchability, the pop-up kirigami TEG was also anticipated to exhibit high performance in terms of heat transfer because the popped-up substrate has a structure that rises up considerably from the heat source surface. Therefore, no separate heat sinks are required. Although the hard TE elements do not bend upon pop-up deformation of the pop-up kirigami TEG, the substrate undergoes significant local bending, making the entire TEG resistant to bending and stretching. In addition, the pop-up kirigami TEG can change the shape of the contact surface to match the shape of the heat source. Because the pop-up kirigami TEG is deformable on curved surfaces and can achieve a conformable contact area with a heat source surface, it was envisaged to exhibit superior power generation performance to conventional non-flexible TEGs.

Characterization of the pop-up kirigami TEG

Evaluating the performance of the pop-up kirigami TEG is equivalent to evaluating the temperature difference that occurs in the TE elements. The design of the pop-up kirigami TEG causes the TE elements to be raised upward in the height direction, and the popped-up substrate to play the role of a heat sink. Therefore, in this section, the results of experiments to examine the effects of raising the TE elements and the area of the upper cooling surface on the power generation are described.

First, to clarify the relationship between the raising up of the TE elements and performance, the output power of pop-up kirigami TEGs composed of a single pair of TE elements was measured while changing the pop-up height of the kirigami TEG, Hdev as defined in Fig. 2a. Compressing the substrate by sliding the right side to the left as illustrated in Fig. 2b led to displacement in the height direction. Because the pop-up kirigami TEG is compressed rather than stretched, the installation area can be reduced. A high-thermal-conductivity grease (Shin-Etsu Silicones, G-747) was applied between the aluminium base plate and the heater. The temperature of the heater was set to 100 °C. During the measurements, the temperature of the room was set at approximately 22 °C. The I–V characteristics of the pop-up kirigami TEG were determined by measuring the output current while sweeping the applied voltage using a source meter (Keithley Instruments, 2614B). In addition, the aspect ratios of the TE elements were varied while maintaining a constant volume to clarify the influence of the length and width of the TE elements on heat radiation. By keeping the volume of the TE element constant, we could compare the amount of TE power generation at the same TE material cost. Therefore, we compared the maximum output power when the dimensions of the TE element were changed at constant volume. The examined dimensions of each TE element (length × width × thickness) were 8 × 1.5 × 1, 6 × 2 × 1, 4 × 3 × 1, and 2 × 6 × 1 mm3. We defined the aspect ratio as the length divided by the width of the TE elements, LTE/WTE. The experimental open-circuit voltage and output power with respect to the pop-up height of the kirigami TEG are plotted in Fig. 2c. The pop-up configuration led to significant improvements in both the open-circuit voltage and output power. However, for the TE elements with high aspect ratios, the slope of the output power was low. This was ascribed to the increased electrical resistance of the TE elements. In the case of a high aspect ratio, a large temperature difference occurred, leading to a higher open-circuit voltage, but the small cross-sectional area of the TE element also resulted in increased electrical resistance. Consequently, although the open-circuit voltage increased with increasing aspect ratio, the output power did not. These findings demonstrate the importance of not only generating a large temperature difference in the TE elements but also considering the electrical resistance of the elements for realizing high performance. In this work, the highest output power was achieved for the pop-up kirigami TEG using TE elements with dimensions of 4 × 3 × 1 mm3.

Fig. 2: Maximum output with respect to heating temperature for each TEG.
Fig. 2: Maximum output with respect to heating temperature for each TEG.The alternative text for this image may have been generated using AI.
Full size image

a Geometry of the pop-up kirigami TEG. b Pop-up deformation of the pop-up kirigami TEG. c Experimental results upon varying the pop-up height of the kirigami TEG and TE element length and width (constant thickness). d Experimental results upon varying the pop-up height of the upper kirigami TEG and TE element length and thickness (constant width). e Electrical resistance with respect to the shape of TE element. f Experimental results upon varying the width of the upper cooling surface. g Comparison of output power between pop-up kirigami TEG and conventional flexible TEGs.

Next, we performed similar experiments to investigate the open-circuit voltage and output power with respect to the pop-up height of the pop-up kirigami TEG when the width and volume of the TE elements were kept constant and the length and thickness were varied, as shown in Fig. 2d. The examined dimensions of each TE element were 8 × 1 × 1.5, 6 × 1 × 2, 4 × 1 × 3 and 2 × 1 × 6 mm3. In this experiment, aspect ratio was the length divided by the width of the TE elements, LTE/WTE. The experimental results are presented in Fig. 2d. The pop-up height resulted in significant improvements in both the open-circuit voltage and output power, just as in the experiment shown in Fig. 2c. However, increasing the thickness of the TE elements resulted in decreases in both the open-circuit voltage and output power.

As shown in Fig. 2e, we measured the electrical resistance of the pop-up kirigami TEG and compared it with the resistance calculated from its geometry and electrical conductivity. In both cases where either the width or the thickness is kept constant, a shorter element length resulted in a larger deviation between the calculated and experimental resistances. This deviation may be attributed to current concentration. On the other hand, when the element length is long, the experimental resistances agree well with the calculated resistances, indicating that the effect of the contact resistance is negligible in our case.

Based on these experimental results, the dimensions of the TE elements for maximising the power generation were fixed at 4 × 3 × 1 mm3. Next, the heat radiation performance of the upper cooling surface was evaluated by varying its area, as shown in Fig. 2f. The width of the upper cooling surface (Wtop) was varied from 1 to 13 mm in increments of 3 mm, and the open-circuit voltage, output power and power density were plotted with respect to Wtop. The power density was the output power per installation area of the pop-up kirigami TEG. In addition, because the dimensions of the TE elements mounted on each device remained the same (i.e., the electrical resistance of the TE elements was constant), the tendencies of the increases in open-circuit voltage and output power with respect to the increase in Wtop were not markedly different. In our case, the pop-up kirigami TEG with a Wtop of 7 mm exhibited a 50% higher open-circuit voltage and output power compared with the pop-up kirigami TEG with a Wtop of 1 mm. However, as Wtop increased, the installation area increased, and the power density decreased. We chose Wtop = 7 mm for subsequent experiments for the output power. Figure 2g shows the output performance, which was also used in a previous study42,43,44,45 and is defined as the output power per temperature difference for both the pop-up kirigami TEG and the previously proposed flexible TEGs. In the graph in Fig. 2g, r is the minimum radius of curvature that each TEG can be bent, and εt is the maximum tensile strain of each TEG that can be stretched. The output performance of the pop-up kirigami TEG was the highest among the previously reported flexible TEGs and was more than ten times greater than that of the best-performing one. Compared with previous TEGs, the pop-up kirigami TEG had a larger contact area with the heat source surface and a heat dissipation surface, which resulted in the highest power output.

Evaluation of pop-up kirigami TEG during deformation

In this section, the output characteristics of an array-type pop-up kirigami TEG upon bending and stretching are evaluated. In addition, the resistance changes during repeated bending deformation and repeated stretching deformation are described. A pop-up kirigami TEG featuring 12 pairs of TE elements with dimensions of 4 × 3 × 1 mm3 mounted in series was used in these experiments. A schematic diagram and photo of the array-type pop-up kirigami TEG before pop-up deformation are shown in Supplementary Fig. 1. The temperature of the heater was increased from 40 °C to 100 °C at 30 °C intervals in each experiment.

First, the array-type pop-up kirigami TEG was positioned on a flat heat source, and the measured output characteristics are plotted in Fig. 3a. At a heat source temperature of 100 °C, the open-circuit voltage was 144 mV and the maximum output power was 12.4 mW. This open-circuit voltage for the array-type pop-up kirigami TEG is in good agreement with the value expected from the open-circuit voltage of 12 mV from a single pair of TE elements, as obtained in the experiment shown in Fig. 2f (12 × 12 mV = 144 mV), thus confirming that the same temperature difference occurred even in the TE elements having an array configuration.

Fig. 3: Power generation characteristics and stability of the pop-up kirigami TEG during various deformations.
Fig. 3: Power generation characteristics and stability of the pop-up kirigami TEG during various deformations.The alternative text for this image may have been generated using AI.
Full size image

a Power generation characteristics when attached to a flat heat source surface. b Power generation characteristics when attached to a curved heat source surface. c Power generation characteristics when attached to a flat heat source surface with a stretching deformation. d Resistance measured at each radius of curvature. e Resistance measured at each tensile strain. f (i) Experimental setup to measure the resistance change, (ii) resistance change during the cyclic bending test and (iii) resistance change during the cyclic stretching test.

To evaluate the stability of the pop-up kirigami TEG during bending deformation, the output characteristics when the array-type pop-up kirigami TEG was attached to a curved heat source were measured (Fig. 3b). In this measurement, a semi-cylindrical aluminium block with a radius of 37.5 mm was prepared as a curved heat source. The open-circuit voltage was 143 mV and the maximum output power was 11.7 mW. Thus, essentially the same open-circuit voltage and output power were obtained for the flat and curved heat sources. In addition, to evaluate the stability of the pop-up kirigami TEG during stretching deformation, the output characteristics when the stretched array-type pop-up kirigami TEG was attached to a flat heat source were measured (Fig. 3c). The tensile strain in this experiment was 40%. The open-circuit voltage was 143 mV and the maximum output power was 11.1 mW, which are again very similar to the values obtained for the flat heat source. When the pop-up kirigami TEG was not stretched, the electrical resistance of the TEG was 415.9 mΩ, and the maximum output was obtained at a voltage of 75.0 mV. In contrast, when the pop-up kirigami TEG was stretched by 40%, the electrical resistance of the TEG was 439.4 mΩ, and the maximum output was obtained at a voltage of 71.2 mV. The change in electrical resistance was approximately 5.0%. In addition, the change in voltage was 5.0%, which led to a 10% change in output power. However, the voltage change when the TEG was stretched by 40% of 5% indicated greater stability than previous studies18,35,37. By stretching the pop-up kirigami TEG, the voltage at the maximum output power point decreased by 5.0%, which means that the temperature difference inside the TE element decreased by 5.0%. These results demonstrate that the array-type pop-up kirigami TEG exhibited stable output and the temperature difference in the TE elements was maintained irrespective of the occurrence of bending and stretching. In addition, to evaluate the minimum radius of curvature and the maximum tensile strain that the TEG can withstand, electrical resistance was measured during both bending and stretching deformation. For the bending test, as shown in Fig. 3d, the radius of curvature was continuously varied from 600 mm to 30 mm while monitoring the resistance. Additional measurements were conducted at small radii of curvature of 0.1 mm and 15 mm). In all cases, the resistance change was less than 2%. For stretching test, as shown in Fig. 3e, the TEG was stretched to a tensile strain of 70%, and the resistance change was less than 1%. These results indicate that the Pop-up TEG exhibits has high mechanical durability under both bending and stretching deformations.

Next, the pop-up kirigami TEG was subjected to a cyclic bending test. The structural change in the array-type pop-up kirigami TEG during bending is shown in Video S1. The measured internal resistance during repeated bending to a radius of curvature of 30 mm is plotted in panel (ii) of Fig. 3f, with the detailed changes shown in the inset. It can be seen that the internal resistance of the array-type pop-up kirigami TEG did not increase over the course of the experiment.

Subsequently, a cyclic stretching test was performed. The structural change of the array-type pop-up kirigami TEG during stretching is shown in Video S2. The measured internal resistance during repeated stretching to a tensile strain of 40% is presented in panel (iii) of Fig. 3f, with the detailed changes shown in the inset. The internal resistance of the array-type pop-up kirigami TEG did not increase with increasing number of cycles, similar to the results of the cyclic bending test. Repeated bending and stretching tests showed that the resistance change due to deformation was about 0.2%. No drift in resistance change was observed after 1000 deformations, and the instantaneous resistance change was about 1%. This resistance change was caused by contact between the substrates.

The lack of change in the internal resistance observed during the cyclic bending and stretching tests indicates that there was no damage to the soldered mounted parts, TE element parts or TEG substrate. In other words, the areas where the TE elements were mounted remained flat even when the pop-up kirigami TEG was subjected to bending or stretching deformation. Therefore, the fabricated pop-up kirigami TEG is expected to prove useful for collecting energy with high efficiency from environmental heat sources, leading to broad application prospects compared with conventional TEGs.

Practical demonstration of the pop-up kirigami TEG

To evaluate the practical utility of our pop-up kirigami TEG, we attached the fabricated array-type pop-up kirigami TEG to a truncated-cone-shaped kettle to evaluate the possibility of realising power generation from a curved surface. The array-type pop-up kirigami TEG was connected to a source meter and the generated voltage was measured for 1 h. Panel (i) of Fig. 4a shows a photograph of the experimental setup, and panel (ii) shows the voltage with respect to the elapsed time. The sharp peak occurring at approximately 5 min corresponds to boiling water being poured into the kettle. The generated voltage of approximately 150 mV is in accordance with that expected based on the previous experimental results. Therefore, the pop-up kirigami TEG is expected to be useful in situations such as camping and disaster response where there is no access to a power source.

Fig. 4: Demonstration of power generation using the fabricated array-type pop-up kirigami TEG with a heated kettle and a wireless transmission circuit.
Fig. 4: Demonstration of power generation using the fabricated array-type pop-up kirigami TEG with a heated kettle and a wireless transmission circuit.The alternative text for this image may have been generated using AI.
Full size image

a Demonstration of TE power generation using the surface of a kettle as a high-temperature heat source. b Demonstration using a wireless transmission circuit. (i) Pop-up kirigami TEG and wireless transmission circuit attached to a human arm. (ii) Photograph of the front side of the wireless circuit. (iii) Photograph of the back side of wireless circuit. (iv) Wireless transmission results for hand temperature powered by the pop-up kirigami TEG.

Next, an additional demonstration was performed using a wireless transmission circuit. We connected our array-type pop-up kirigami TEG to a wireless transmission circuit equipped with a temperature sensor and successfully demonstrated the wireless transmission of the temperature of a person’s wrist. The wireless transmission circuit consisted of a boost transformer (LPR6235-752SMR; Coilcraft), a capacitor (470 μF), an RF circuit (TWELITE RED; Mono Wireless Inc.) and a temperature sensor, as shown in panels (ii) and (iii) of Fig. 4b. The wireless transmission circuit requires 76 μW (19 mV, 4.0 mA) to charge the on-board 470 μF capacitor. The average power supplied by the pop-up kirigami TEG attached to human skin was 117 μW (25 mV, 4.7 mA). Wireless transmission occurred when the 470 μF capacitor in the wireless transmission circuit was charged to a voltage of 3.3 V. Therefore, even if the amount of power supplied from the pop-up kirigami TEG to the wireless transmission circuit exceeded 76 μW, a charging time was required to charge the capacitor. The array-type pop-up kirigami TEG was attached to the person’s arm using thermally conductive grease. The temperature of the wrist was first measured while the wearer was sitting on a chair, and again while they were performing squats. The body posture while squatting is shown in Video S3. The temperature data were wirelessly transmitted to a computer equipped with a USB wireless receiver.

Panel (iv) of Fig. 4b shows the temperature measurements over the course of the experiment after approximately 20 min of charging. The pop-up kirigami TEG charged the of 470 µF capacitor on the wireless transmission circuit board. Therefore, the initial charge took more than 15 min. After the initial charge, the capacitor only needed to recharge the amount of charge used. Overall, the temperature data were successfully transmitted wirelessly every 4 min on average for a total of approximately 40 min. This frequency of wireless transmission is sufficient to capture changes in body temperature. The wirelessly transmitted temperature was considered reasonable because it was close to normal body temperature as well as the value actually measured using a thermocouple. In addition, a rise in body temperature of approximately 1 °C was observed after the wearer shifted from the sitting state to the squatting state, confirming the successful detection of the increase in body temperature due to exercise. These results highlight the usefulness of the pop-up kirigami TEG developed in this work.

Discussion

In this paper, a pop-up kirigami structure was applied as a TEG to realise high stretchability and high conformability to a curved heat source, resulting in high power generation performance. The pop-up kirigami structure used in this work has the following advantages: (i) bendability and stretchability, (ii) convenient mounting of both bulk and thin-film TE elements, (iii) facile fabrication due to the possibility of mounting the TE elements on a flat substrate, (iv) a thermal contact surface that minimises thermal resistance between the TEG and the heat source and (v) a cooling surface that conforms to the deformation of the TEG, thereby avoiding the need for expensive additional heat sinks.

The evaluation of the power generation performance of the pop-up deformation of the pop-up kirigami TEGs revealed that the output power dramatically improved upon popping up the kirigami TEGs, regardless of the aspect ratio of the TE element. For the same volume of TE elements, the length of the TE elements contributed more than the thickness and width to improving the power generation performance, although a suitable aspect ratio exists. Although current crowding effects may occur depending on the dimensions of the TE element, the effect was small because the TE element ends were mounted with solder. In addition, the evaluation of the area of the upper cooling surface revealed that the output power gradually increased with increasing area of the upper cooling surface but did not contribute as much. These results indicate that the pop-up height of the pop-up kirigami TEG and the aspect ratio of the TE element contribute substantially to the power generation performance, whereas the area of the upper cooling surface does not contribute as much. In other words, these results demonstrate that the electrical resistance and thermal resistance of the TE elements contribute significantly more to the power generation performance compared with the heat dissipation from the upper cooling surface. In our case, the maximum power generation of 1.13 mW per pair of TE elements was obtained by mounting TE elements with dimensions of 4 × 3 × 1 mm3 (LTE × WTE × TTE), popping up the kirigami TEG to a height of 10 mm and giving the upper cooling surface a width of 7 mm.

Furthermore, we measured the output characteristics for an array-type pop-up kirigami TEG having 12 pairs of TE elements attached to flat and curved heat sources. The generated output power remained the same for both heat sources. This performance was maintained even when the array-type pop-up kirigami TEG was stretched to a tensile strain of 40%. These experimental results demonstrate that stable output power generation could be successfully accomplished irrespective of the shape of the heat source surface. In other words, the pop-up kirigami TEG was able to display high performance without the need for dedicated heat sinks, even when attached to heat sources of various shapes. Table 1 shows the results of comparing existing flexible TEG devices with the proposed pop-up kirigami TEG, in terms of output, cooling method and stretchability. In TE power generation, the most commonly used TE materials are Bi and Te, which are rare and expensive. Therefore, a TEG that can produce a large output power with a small amount of TE elements is desirable; thus, in Table 1, the output power per TE element volume is included. The open-circuit voltage of the TEG, VTEG is proportional to the temperature difference across the TE element, ΔTTEG and can be expressed as

$${V}_{\mathrm{TEG}}=({S}_{{\rm{p}}}-{S}_{{\rm{n}}}){\Delta T}_{\mathrm{TE}{\rm{G}}}$$
(1)

where Sp and Sn are the Seebeck coefficients of the p-type and n-type TE element, respectively. As shown in Fig. 3a, the current-voltage (I–V) characteristics of the TEG are on straight lines; therefore, from the Eq. 1, the maximum output power is given by

$$P=\frac{{{V}_{\mathrm{TEG}}}^{2}}{4{R}_{\mathrm{TEG}}}=\frac{{\left({S}_{{\rm{p}}}-{S}_{{\rm{n}}}\right)}^{2}\,{\Delta {T}_{\mathrm{TEG}}}^{2}}{4{R}_{\mathrm{TEG}}}$$
(2)

where RTEG is the total electrical resistance of the TEG. Furthermore, ΔTTEG is proportional to the temperature difference between the temperature of the heat source (i.e., the hot side), Th and the temperature of the ambient air (i.e., the hot side), Tc. Therefore, by defining ΔT = Th − Tc, ΔTTEG can be expressed as

$$\Delta {T}_{\mathrm{TEG}}=\frac{{R}_{\theta \mathrm{TEG}}}{{R}_{\theta \mathrm{TEG}}+{R}_{\theta \mathrm{ex}}}\Delta T$$
(3)

where \({R}_{\theta \mathrm{TEG}}\) is the thermal resistance of the TEG and \({R}_{\theta \mathrm{ex}}\) is the total thermal resistance of the external component, which is basically the sum of the thermal contact resistance between the heat source and the TEG and the thermal resistance of the heat sink and the ambient air. From the Eqs. 2 and 3, P is shown as

$$P=\frac{{\left({S}_{{\rm{p}}}-{S}_{{\rm{n}}}\right)}^{2}\,}{4{R}_{\mathrm{TEG}}}{\left(\frac{{R}_{\theta \mathrm{TEG}}}{{R}_{\theta \mathrm{TEG}}+{R}_{\theta \mathrm{ex}}}\right)}^{2}{\Delta T}^{2}$$
(4)
Table 1 Comparison of power generation performance of bendable and/or stretchable TEGs
Full size table

In the Eq. 4 the coefficient of \({\Delta T}^{2}\) depends on the material properties, and electrical and thermal design of the TEG, and can therefore be regarded as an intrinsic property of the device, independent of the installation environment. From this reason, we defined the output performance as follows:

$${\text{output}}\,{\text{performance}}\,[{\rm{\mu }}{\rm{W}}/{{\rm{K}}}^{2}]:P/{\Delta T}^{2}=\frac{{\left({S}_{{\rm{p}}}-{S}_{{\rm{n}}}\right)}^{2}\,}{4{R}_{{\text{TEG}}}}{\left(\frac{{R}_{\theta {\text{TEG}}}}{{R}_{\theta {\text{TEG}}}+{R}_{\theta {\text{ex}}}}\right)}^{2}$$
(5)

As shown in Table 1, our TEG exhibits a constant output performance of 190 × 10−3 μW/K2 despite variations in Th of 40 °C and 100 °C. This confirms that the output performance, defined as \(P/{\Delta T}^{2}\), is a suitable metric for comparing devices operating under different environmental conditions, such as varying heat source temperatures. Our pop-up kirigami TEG values were calculated from the experimental results of the array-type pop-up kirigami TEG having 12 pairs of TE elements. It can be seen that the pop-up kirigami TEG had markedly improved output and output density per pair of TE elements, even though it does not use additional heat sinks and is stretchable.

Finally, we successfully demonstrated the versatility of the fabricated pop-up kirigami TEG under two different scenarios. In the first scenario, the pop-up kirigami TEG was attached to the arm of a person to supply power to a wireless transmission circuit for monitoring body temperature, thus realising maintenance-free monitoring of a biophysical parameter. This may lead to useful clinical applications, such as in medical facilities where patients must be continuously monitored. In the second scenario, the pop-up kirigami TEG was used to generate power from the heat emitted by a kettle of hot water, with potential applications in driving electronic devices even in places and situations lacking power sources, such as during natural disasters and other emergencies.

Therefore, the fabricated pop-up kirigami TEG is expected to prove useful for collecting energy with high efficiency from environmental heat sources, leading to broad application prospects compared with conventional TEGs.

In this study, we successfully demonstrated the ability of a substrate with a kirigami structure to obtain both flexibility and high performance. The fabricated pop-up kirigami TEG with 12 TE element pairs generated stable power outputs of 12.4 and 11.7 mW for flat and curved heat sources, respectively. Furthermore, attachment of the pop-up kirigami TEG to a human arm enabled wireless body temperature monitoring. Thus, a flexible and high-performance TEG was realized solely through this innovative substrate design.

Methods

Fabrication of TE elements

Commercially available BiTe-based TE wafers with a thickness of 1 mm (Toshima Co. Ltd., p-type: Bi0.3Sb1.7Te3, n-type: Bi2Te3 + Ru) were used to prepare the TE elements for mounting on the pop-up kirigami TEGs. No electrodes were attached, and the wafer surfaces were not mirror polished. The wafers were cut to the desired sizes using a wire saw (New Metals and Chemicals Co. Ltd., CS-411) fitted with a diamond wire of 250 μm diameter. In general, samples for cutting with a wire saw are fixed to a bisque plate using wax, which melts when heated and hardens when cooled. However, fixing porous TE materials with wax causes the wax to remain inside the material. As such, double-sided tape was used in this work. The cut TE elements were ultrasonically cleaned with pure water. Next, a seed layer was formed on the surface for joining with the electrodes using an ultrasonic soldering machine (Kuroda Techno Co. Ltd., USM-5). A special solder (Kuroda Techno Co. Ltd., 168936) was used as the material for the seed layer, and eutectic solder (Taiyo Electric Ind. Co. Ltd., SD-61) was used for the areas connected to the electrodes.

Fabrication of the pop-up kirigami TEG

First, as illustrated in panel (i) of Fig. 1b, polyimide tape of 50 μm thickness was attached to a 40-μm-thick copper sheet. The polyimide tape acted as an insulating layer, while the copper sheet served as the electrodes and wiring. Next, as depicted in panel (ii) of Fig. 1b, the prepared polyimide–copper substrate was processed into a kirigami pattern with a UV laser (OPI Corporation, OLMUV-335-5A-K). Subsequently, as shown in panel (iii) of Fig. 1b, the copper layer was peeled off at the areas where the TE elements were to be mounted to expose the polyimide tape for insulation. Then, as shown in panel (iv) of Fig. 1b, the TE elements were soldered to the positions where the polyimide was exposed. Prior to soldering the TE elements to the copper sheet, an adhesive layer was applied to the surface of the TE elements using an ultrasonic soldering device (Kuroda Techno Co. Ltd., USM-560). This process removed the oxide layer from the TE element surface and improved the adhesion performance between the TE element surface and solder. Finally, as illustrated in panel (v) of Fig. 1b, by moving the left and right sides of the substrate closer to each other, a pop-up kirigami TEG mounted with a pair of TE elements was realised. This movement causes the central part of the substrate to move upward in the height direction, creating a large temperature gradient in the TE elements. The fabricated pop-up kirigami TEG was affixed to an aluminium base plate using an adhesive with a high thermal conductivity of 1.7 W/mK (Cemedine Co. Ltd., SX-1008). In addition, high-thermal-conductivity grease (Shin-Etsu Silicone, G-747) was applied between the aluminium base plate and the heater when measuring the performance of the pop-up kirigami TEG. The heating temperature was adjusted to 40 °C, 70 °C or 100 °C using a Peltier temperature controller (VICS, VTH1.8K-70S). The room temperature during the measurements was set to approximately 22 °C. The I–V characteristics were obtained by sweeping the applied voltage using a source meter (Keithley Instruments, 2614B).