Offering sustainable sources of electricity is one of the most important challenges of this century. Research areas in energy harvesting materials stem from this motivation, including thermoelectric1, photovoltaic2 and thermophotovoltaics3. Although we lack materials and devices capable of harvesting energy in the Joule range, pyroelectric materials that can convert electrical energy into periodic temperature changes are considered sensors4 and energy harvesters5,6,7. Here we have developed a macroscopic thermal energy harvester in the form of a multilayer capacitor made of 42 grams of lead scandium tantalate, producing 11.2 J of electrical energy per thermodynamic cycle. Each pyroelectric module can generate electrical energy density up to 4.43 J cm-3 per cycle. We also show that two such modules weighing 0.3 g are enough to continuously power autonomous energy harvesters with embedded microcontrollers and temperature sensors. Finally, we show that for a temperature range of 10 K, these multilayer capacitors can reach 40% Carnot efficiency. These properties are due to (1) ferroelectric phase change for high efficiency, (2) low leakage current to prevent losses, and (3) high breakdown voltage. These macroscopic, scalable and efficient pyroelectric power harvesters are reimagining thermoelectric power generation.
Compared to the spatial temperature gradient required for thermoelectric materials, energy harvesting of thermoelectric materials requires temperature cycling over time. This means a thermodynamic cycle, which is best described by the entropy (S)-temperature (T) diagram. Figure 1a shows a typical ST plot of a non-linear pyroelectric (NLP) material demonstrating a field-driven ferroelectric-paraelectric phase transition in scandium lead tantalate (PST). The blue and green sections of the cycle on the ST diagram correspond to the converted electrical energy in the Olson cycle (two isothermal and two isopole sections). Here we consider two cycles with the same electric field change (field on and off) and temperature change ΔT, albeit with different initial temperatures. The green cycle is not located in the phase transition region and thus has a much smaller area than the blue cycle located in the phase transition region. In the ST diagram, the larger the area, the greater the collected energy. Therefore, the phase transition must collect more energy. The need for large area cycling in NLP is very similar to the need for electrothermal applications9, 10, 11, 12 where PST multilayer capacitors (MLCs) and PVDF-based terpolymers have recently shown excellent reverse performance. cooling performance status in cycle 13,14,15,16. Therefore, we have identified PST MLCs of interest for thermal energy harvesting. These samples have been fully described in the methods and characterized in supplementary notes 1 (scanning electron microscopy), 2 (X-ray diffraction) and 3 (calorimetry).
a, Sketch of an entropy (S)-temperature (T) plot with electric field on and off applied to NLP materials showing phase transitions. Two energy collection cycles are shown in two different temperature zones. The blue and green cycles occur inside and outside the phase transition, respectively, and end in very different regions of the surface. b, two DE PST MLC unipolar rings, 1 mm thick, measured between 0 and 155 kV cm-1 at 20 °C and 90 °C, respectively, and the corresponding Olsen cycles. The letters ABCD refer to different states in the Olson cycle. A-B: MLCs were charged to 155 kV cm-1 at 20°C. B-C: MLC was maintained at 155 kV cm-1 and the temperature was raised to 90 °C. C-D: MLC discharges at 90°C. D-A: MLC chilled to 20°C in zero field. The blue area corresponds to the input power required to start the cycle. The orange area is the energy collected in one cycle. c, top panel, voltage (black) and current (red) versus time, tracked during the same Olson cycle as b. The two inserts represent the amplification of voltage and current at key points in the cycle. In the lower panel, the yellow and green curves represent the corresponding temperature and energy curves, respectively, for a 1 mm thick MLC. Energy is calculated from the current and voltage curves on the top panel. Negative energy corresponds to the collected energy. The steps corresponding to the capital letters in the four figures are the same as in the Olson cycle. The cycle AB’CD corresponds to the Stirling cycle (additional note 7).
where E and D are the electric field and the electric displacement field, respectively. Nd can be obtained indirectly from the DE circuit (Fig. 1b) or directly by starting a thermodynamic cycle. The most useful methods were described by Olsen in his pioneering work on collecting pyroelectric energy in the 1980s17.
On fig. 1b shows two monopolar DE loops of 1 mm thick PST-MLC specimens assembled at 20 °C and 90 °C, respectively, over a range of 0 to 155 kV cm-1 (600 V). These two cycles can be used to indirectly calculate the energy collected by the Olson cycle shown in Figure 1a. In fact, the Olsen cycle consists of two isofield branches (here, zero field in the DA branch and 155 kV cm-1 in the BC branch) and two isothermal branches (here, 20°С and 20°С in the AB branch). C in the CD branch) The energy collected during the cycle corresponds to the orange and blue regions (EdD integral). The collected energy Nd is the difference between input and output energy, i.e. only the orange area in fig. 1b. This particular Olson cycle gives an Nd energy density of 1.78 J cm-3. The Stirling cycle is an alternative to the Olson cycle (Supplementary Note 7). Because the constant charge stage (open circuit) is more easily reached, the energy density extracted from Fig. 1b (cycle AB’CD) reaches 1.25 J cm-3. This is only 70% of what the Olson cycle can collect, but simple harvesting equipment does it.
In addition, we directly measured the energy collected during the Olson cycle by energizing the PST MLC using a Linkam temperature control stage and a source meter (method). Figure 1c at the top and in the respective insets shows the current (red) and voltage (black) collected on the same 1 mm thick PST MLC as for the DE loop going through the same Olson cycle. The current and voltage make it possible to calculate the collected energy, and the curves are shown in fig. 1c, bottom (green) and temperature (yellow) throughout the cycle. The letters ABCD represent the same Olson cycle in Fig. 1. MLC charging occurs during the AB leg and is carried out at a low current (200 µA), so SourceMeter can properly control charging. The consequence of this constant initial current is that the voltage curve (black curve) is not linear due to the non-linear potential displacement field D PST (Fig. 1c, top inset). At the end of charging, 30 mJ of electrical energy is stored in the MLC (point B). The MLC then heats up and a negative current (and therefore a negative current) is produced while the voltage remains at 600 V. After 40 s, when the temperature reached a plateau of 90 °C, this current was compensated, although the step sample produced in the circuit an electrical power of 35 mJ during this isofield (second inset in Fig. 1c, top). The voltage on the MLC (branch CD) is then reduced, resulting in an additional 60 mJ of electrical work. The total output energy is 95 mJ. The collected energy is the difference between the input and output energy, which gives 95 – 30 = 65 mJ. This corresponds to an energy density of 1.84 J cm-3, which is very close to the Nd extracted from the DE ring. The reproducibility of this Olson cycle has been extensively tested (Supplementary Note 4). By further increasing voltage and temperature, we achieved 4.43 J cm-3 using Olsen cycles in a 0.5 mm thick PST MLC over a temperature range of 750 V (195 kV cm-1) and 175 °C (Supplementary Note 5). This is four times greater than the best performance reported in the literature for direct Olson cycles and was obtained on thin films of Pb(Mg,Nb)O3-PbTiO3 (PMN-PT) (1.06 J cm-3)18 (cm .Supplementary Table 1 for more values ​​in the literature). This performance has been reached owing to the very low leakage current of these MLCs (<10−7 A at 750 V and 180 °C, see details in Supplementary Note 6)—a crucial point mentioned by Smith et al.19—in contrast to the materials used in earlier studies17,20. This performance has been reached owing to the very low leakage current of these MLCs (<10−7 A at 750 V and 180 °C, see details in Supplementary Note 6)—a crucial point mentioned by Smith et al.19—in contrast to the materials used in earlier studies17,20. Эти характеристики были достигнуты благодаря очень низкому току утечки этих MLC (<10–7 А при 750 В и 180 °C, см. подробности в дополнительном примечании 6) — критический момент, упомянутый Смитом и др. 19 — в отличие от к материалам, использованным в более ранних исследованиях17,20. These characteristics were achieved due to the very low leakage current of these MLCs (<10–7 A at 750 V and 180 °C, see Supplementary Note 6 for details) – a critical point mentioned by Smith et al. 19 – in contrast to materials used in earlier studies17,20.由于这些MLC 的泄漏电流非常低（在750 V 和180 °C 时<10-7 A，请参见补充说明6 中的详细信息）——Smith 等人19 提到的关键点——相比之下，已经达到了这种性能到早期研究中使用的材料17,20。由于 这些 mlc 的 泄漏 非常 （在 在 在 750 V 和 180 ° C 时 <10-7 A ， 参见 补充 说明 6 中 详细 信息））））） — 等 人 19 提到 关键 关键 点 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下 相比之下，已经达到了这种性能到早期研究中使用的材料17.20。 Поскольку ток утечки этих MLC очень низкий (<10–7 А при 750 В и 180 °C, см. подробности в дополнительном примечании 6) — ключевой момент, упомянутый Смитом и др. 19 — для сравнения, были достигнуты эти характеристики. Since the leakage current of these MLCs is very low (<10–7 A at 750 V and 180 °C, see Supplementary Note 6 for details) – a key point mentioned by Smith et al. 19 – for comparison, these performances were achieved. to materials used in earlier studies 17,20.
The same conditions (600 V, 20–90 °C) applied to the Stirling cycle (Supplementary note 7). As expected from the results of the DE cycle, the yield was 41.0 mJ. One of the most striking features of Stirling cycles is their ability to amplify the initial voltage through the thermoelectric effect. We observed a voltage gain of up to 39 (from an initial voltage of 15 V to an end voltage of up to 590 V, see Supplementary Fig. 7.2).
Another distinguishing feature of these MLCs is that they are macroscopic objects large enough to collect energy in the joule range. Therefore, we constructed a prototype harvester (HARV1) using 28 MLC PST 1 mm thick, following the same parallel plate design described by Torello et al.14, in a 7×4 matrix as shown in Fig. The heat-carrying dielectric fluid in the manifold is displaced by a peristaltic pump between two reservoirs where the fluid temperature is kept constant (method). Collect up to 3.1 J using the Olson cycle described in fig. 2a, isothermal regions at 10°C and 125°C and isofield regions at 0 and 750 V (195 kV cm-1). This corresponds to an energy density of 3.14 J cm-3. Using this combine, measurements were taken under various conditions (Fig. 2b). Note that 1.8 J was obtained over a temperature range of 80 °C and a voltage of 600 V (155 kV cm-1). This is in good agreement with the previously mentioned 65 mJ for 1 mm thick PST MLC under the same conditions (28 × 65 = 1820 mJ).
a, Experimental setup of an assembled HARV1 prototype based on 28 MLC PSTs 1 mm thick (4 rows × 7 columns) running on Olson cycles. For each of the four cycle steps, temperature and voltage are provided in the prototype. The computer drives a peristaltic pump that circulates a dielectric fluid between the cold and hot reservoirs, two valves, and a power source. The computer also uses thermocouples to collect data on the voltage and current supplied to the prototype and the temperature of the combine from the power supply. b, Energy (color) collected by our 4×7 MLC prototype versus temperature range (X-axis) and voltage (Y-axis) in different experiments.
A larger version of the harvester (HARV2) with 60 PST MLC 1 mm thick and 160 PST MLC 0.5 mm thick (41.7 g active pyroelectric material) gave 11.2 J (Supplementary Note 8). In 1984, Olsen made an energy harvester based on 317 g of a tin-doped Pb(Zr,Ti)O3 compound capable of generating 6.23 J of electricity at a temperature of about 150 °C (ref. 21). For this combine, this is the only other value available in the joule range. It got just over half the value we achieved and nearly seven times the quality. This means that the energy density of HARV2 is 13 times higher.
The HARV1 cycle period is 57 seconds. This produced 54 mW of power with 4 rows of 7 columns of 1 mm thick MLC sets. To take it one step further, we built a third combine (HARV3) with a 0.5mm thick PST MLC and similar setup to HARV1 and HARV2 (Supplementary Note 9). We measured a thermalization time of 12.5 seconds. This corresponds to a cycle time of 25 s (Supplementary Fig. 9). The collected energy (47 mJ) gives an electrical power of 1.95 mW per MLC, which in turn allows us to imagine that HARV2 produces 0.55 W (approximately 1.95 mW × 280 PST MLC 0.5 mm thick). In addition, we simulated heat transfer using Finite Element Simulation (COMSOL, Supplementary Note 10 and Supplementary Tables 2–4) corresponding to the HARV1 experiments. Finite element modeling made it possible to predict power values ​​almost an order of magnitude higher (430 mW) for the same number of PST columns by thinning the MLC to 0.2 mm, using water as a coolant, and restoring the matrix to 7 rows. × 4 columns (in addition to , there were 960 mW when the tank was next to the combine, Supplementary Fig. 10b).
To demonstrate the usefulness of this collector, a Stirling cycle was applied to a stand-alone demonstrator consisting of only two 0.5 mm thick PST MLCs as heat collectors, a high voltage switch, a low voltage switch with storage capacitor, a DC/DC converter, a low power microcontroller, two thermocouples and boost converter (Supplementary Note 11). The circuit requires the storage capacitor to be initially charged at 9V and then runs autonomously while the temperature of the two MLCs ranges from -5°C to 85°C, here in cycles of 160 s (several cycles are shown in Supplementary Note 11). Remarkably, two MLCs weighing only 0.3g can autonomously control this large system. Another interesting feature is that the low voltage converter is capable of converting 400V to 10-15V with 79% efficiency (Supplementary Note 11 and Supplementary Figure 11.3).
Finally, we evaluated the efficiency of these MLC modules in converting thermal energy into electrical energy. The quality factor η of efficiency is defined as the ratio of the density of the collected electrical energy Nd to the density of the supplied heat Qin (Supplementary note 12):
Figures 3a,b show the efficiency η and proportional efficiency ηr of the Olsen cycle, respectively, as a function of the temperature range of a 0.5 mm thick PST MLC. Both data sets are given for an electric field of 195 kV cm-1. The efficiency \(\this\) reaches 1.43%, which is equivalent to 18% of ηr. However, for a temperature range of 10 K from 25 °C to 35 °C, ηr reaches values ​​up to 40% (blue curve in Fig. 3b). This is twice the known value for NLP materials recorded in PMN-PT films (ηr = 19%) in the temperature range of 10 K and 300 kV cm-1 (Ref. 18). Temperature ranges below 10 K were not considered because the thermal hysteresis of the PST MLC is between 5 and 8 K. Recognition of the positive effect of phase transitions on efficiency is critical. In fact, the optimal values ​​of η and ηr are almost all obtained at the initial temperature Ti = 25°C in Figs. 3a,b. This is due to a close phase transition when no field is applied and the Curie temperature TC is around 20 °C in these MLCs (Supplementary note 13).
a,b, the efficiency η and the proportional efficiency of the Olson cycle (a)\({\eta }_{{\rm{r}}}=\eta /{\eta}_{{\rm{Carnot}} for the maximum electric by a field of 195 kV cm-1 and different initial temperatures Ti, }}\,\)(b) for the MPC PST 0.5 mm thick, depending on the temperature interval ΔTspan.
The latter observation has two important implications: (1) any effective cycling must begin at temperatures above TC for a field-induced phase transition (from paraelectric to ferroelectric) to occur; (2) these materials are more efficient at run times close to TC. Although large-scale efficiencies are shown in our experiments, the limited temperature range does not allow us to achieve large absolute efficiencies due to the Carnot limit (\(\Delta T/T\)). However, the excellent efficiency demonstrated by these PST MLCs justifies Olsen when he mentions that “an ideal class 20 regenerative thermoelectric motor operating at temperatures between 50 °C and 250 °C can have an efficiency of 30%”17. To reach these values ​​and test the concept, it would be useful to use doped PSTs with different TCs, as studied by Shebanov and Borman. They showed that TC in PST can vary from 3°C (Sb doping) to 33°C (Ti doping) 22 . Therefore, we hypothesize that next generation pyroelectric regenerators based on doped PST MLCs or other materials with a strong first order phase transition can compete with the best power harvesters.
In this study, we investigated MLCs made from PST. These devices consist of a series of Pt and PST electrodes, whereby several capacitors are connected in parallel. PST was chosen because it is an excellent EC material and therefore a potentially excellent NLP material. It exhibits a sharp first-order ferroelectric-paraelectric phase transition around 20 °C, indicating that its entropy changes are similar to those shown in Fig. 1. Similar MLCs have been fully described for EC13,14 devices. In this study, we used 10.4 × 7.2 × 1 mm³ and 10.4 × 7.2 × 0.5 mm³ MLCs. MLCs with a thickness of 1 mm and 0.5 mm were made from 19 and 9 layers of PST with a thickness of 38.6 µm, respectively. In both cases, the inner PST layer was placed between 2.05 µm thick platinum electrodes. The design of these MLCs assumes that 55% of the PSTs are active, corresponding to the part between the electrodes (Supplementary Note 1). The active electrode area was 48.7 mm2 (Supplementary Table 5). MLC PST was prepared by solid phase reaction and casting method. The details of the preparation process have been described in a previous article14. One of the differences between PST MLC and the previous article is the order of B-sites, which greatly affects the performance of EC in PST. The order of B-sites of PST MLC is 0.75 (Supplementary Note 2) obtained by sintering at 1400°C followed by hundreds of hours long annealing at 1000°C. For more information on PST MLC, see Supplementary Notes 1-3 and Supplementary Table 5.
The main concept of this study is based on the Olson cycle (Fig. 1). For such a cycle, we need a hot and cold reservoir and a power supply capable of monitoring and controlling the voltage and current in the various MLC modules. These direct cycles used two different configurations, namely (1) Linkam modules heating and cooling one MLC connected to a Keithley 2410 power source, and (2) three prototypes (HARV1, HARV2 and HARV3) in parallel with the same source energy. In the latter case, a dielectric fluid (silicone oil with a viscosity of 5 cP at 25°C, purchased from Sigma Aldrich) was used for heat exchange between the two reservoirs (hot and cold) and the MLC. The thermal reservoir consists of a glass container filled with dielectric fluid and placed on top of the thermal plate. Cold storage consists of a water bath with liquid tubes containing dielectric fluid in a large plastic container filled with water and ice. Two three-way pinch valves (purchased from Bio-Chem Fluidics) were placed at each end of the combine to properly switch fluid from one reservoir to another (Figure 2a). To ensure thermal equilibrium between the PST-MLC package and the coolant, the cycle period was extended until the inlet and outlet thermocouples (as close as possible to the PST-MLC package) showed the same temperature. The Python script manages and synchronizes all instruments (source meters, pumps, valves, and thermocouples) to run the correct Olson cycle, i.e. the coolant loop starts cycling through the PST stack after the source meter is charged so that they heat up at the desired applied voltage for given Olson cycle.
Alternatively, we have confirmed these direct measurements of collected energy with indirect methods. These indirect methods are based on electric displacement (D) – electric field (E) field loops collected at different temperatures, and by calculating the area between two DE loops, one can accurately estimate how much energy can be collected, as shown in the figure. in figure 2. .1b. These DE loops are also collected using Keithley source meters.
Twenty-eight 1 mm thick PST MLCs were assembled in a 4-row, 7-column parallel plate structure according to the design described in the reference. 14. The fluid gap between PST-MLC rows is 0.75mm. This is achieved by adding strips of double-sided tape as liquid spacers around the edges of the PST MLC. The PST MLC is electrically connected in parallel with a silver epoxy bridge in contact with the electrode leads. After that, wires were glued with silver epoxy resin to each side of the electrode terminals for connection to the power supply. Finally, insert the entire structure into the polyolefin hose. The latter is glued to the fluid tube to ensure proper sealing. Finally, 0.25 mm thick K-type thermocouples were built into each end of the PST-MLC structure to monitor the inlet and outlet liquid temperatures. To do this, the hose must first be perforated. After installing the thermocouple, apply the same adhesive as before between the thermocouple hose and wire to restore the seal.
Eight separate prototypes were built, four of which had 40 0.5 mm thick MLC PSTs distributed as parallel plates with 5 columns and 8 rows, and the remaining four had 15 1 mm thick MLC PSTs each. in 3-column × 5-row parallel plate structure. The total number of PST MLCs used was 220 (160 0.5 mm thick and 60 PST MLC 1 mm thick). We call these two subunits HARV2_160 and HARV2_60. The liquid gap in the prototype HARV2_160 consists of two double-sided tapes 0.25 mm thick with a wire 0.25 mm thick between them. For the HARV2_60 prototype, we repeated the same procedure, but using 0.38 mm thick wire. For symmetry, HARV2_160 and HARV2_60 have their own fluid circuits, pumps, valves and cold side (Supplementary Note 8). Two HARV2 units share a heat reservoir, a 3 liter container (30 cm x 20 cm x 5 cm) on two hot plates with rotating magnets. All eight individual prototypes are electrically connected in parallel. The HARV2_160 and HARV2_60 subunits work simultaneously in the Olson cycle resulting in an energy harvest of 11.2 J.
Place 0.5mm thick PST MLC into polyolefin hose with double sided tape and wire on both sides to create space for liquid to flow. Due to its small size, the prototype was placed next to a hot or cold reservoir valve, minimizing cycle times.
In PST MLC, a constant electric field is applied by applying a constant voltage to the heating branch. As a result, a negative thermal current is generated and energy is stored. After heating the PST MLC, the field is removed (V = 0), and the energy stored in it is returned back to the source counter, which corresponds to one more contribution of the collected energy. Finally, with a voltage V = 0 applied, the MLC PSTs are cooled to their initial temperature so that the cycle can start again. At this stage, energy is not collected. We ran the Olsen cycle using a Keithley 2410 SourceMeter, charging the PST MLC from a voltage source and setting the current match to the appropriate value so that enough points were collected during the charging phase for reliable energy calculations.
In Stirling cycles, PST MLCs were charged in voltage source mode at an initial electric field value (initial voltage Vi > 0), a desired compliance current so that the charging step takes around 1 s (and enough points are gathered for a reliable calculation of the energy) and cold temperature. In Stirling cycles, PST MLCs were charged in voltage source mode at an initial electric field value (initial voltage Vi > 0), a desired compliance current so that the charging step takes around 1 s (and enough points are gathered for a reliable calculation of the energy) and cold temperature. В циклах Стирлинга PST MLC заряжались в режиме источника напряжения при начальном значении электрического поля (начальное напряжение Vi > 0), желаемом податливом токе, так что этап зарядки занимает около 1 с (и набирается достаточное количество точек для надежного расчета энергия) и холодная температура. In the Stirling PST MLC cycles, they were charged in the voltage source mode at the initial value of the electric field (initial voltage Vi > 0), the desired yield current, so that the charging stage takes about 1 s (and a sufficient number of points are collected for a reliable energy calculation) and cold temperature.在斯特林循环中，PST MLC 在电压源模式下以初始电场值（初始电压Vi > 0）充电，所需的顺应电流使得充电步骤大约需要1 秒（并且收集了足够的点以可靠地计算能量）和低温。 In the master cycle, the PST MLC is charged at the initial electric field value (initial voltage Vi > 0) in the voltage source mode, so that the required compliance current takes about 1 second for the charging step (and we collected enough points to reliably calculate (energy) and low temperature. В цикле Стирлинга PST MLC заряжается в режиме источника напряжения с начальным значением электрического поля (начальное напряжение Vi > 0), требуемый ток податливости таков, что этап зарядки занимает около 1 с (и набирается достаточное количество точек, чтобы надежно рассчитать энергию) и низкие температуры. In the Stirling cycle, the PST MLC is charged in the voltage source mode with an initial value of the electric field (initial voltage Vi > 0), the required compliance current is such that the charging stage takes about 1 s (and a sufficient number of points are collected to reliably calculate the energy) and low temperatures . Before the PST MLC heats up, open the circuit by applying a matching current of I = 0 mA (the minimum matching current that our measuring source can handle is 10 nA). As a result, a charge remains in the PST of the MJK, and the voltage increases as the sample heats up. No energy is collected in arm BC because I = 0 mA. After reaching a high temperature, the voltage in the MLT FT increases (in some cases more than 30 times, see additional fig. 7.2), the MLK FT is discharged (V = 0), and electrical energy is stored in them for the same as they be the initial charge. The same current correspondence is returned to the meter-source. Due to voltage gain, the stored energy at high temperature is higher than what was provided at the beginning of the cycle. Consequently, energy is obtained by converting heat into electricity.
We used a Keithley 2410 SourceMeter to monitor the voltage and current applied to the PST MLC. The corresponding energy is calculated by integrating the product of voltage and current read by Keithley’s source meter, \ (E = {\int }_{0}^{\tau }{I}_({\rm {meas))}\left(t\ right){V}_{{\rm{meas}}}(t)\), where τ is the period of the period. On our energy curve, positive energy values ​​mean the energy we have to give to the MLC PST, and negative values ​​mean the energy we extract from them and therefore the energy received. The relative power for a given collection cycle is determined by dividing the collected energy by the period τ of the entire cycle.
All data are presented in the main text or in additional information. Letters and requests for materials should be directed to the source of the AT or ED data provided with this article.
Ando Junior, OH, Maran, ALO & Henao, NC A review of the development and applications of thermoelectric microgenerators for energy harvesting. Ando Junior, OH, Maran, ALO & Henao, NC A review of the development and applications of thermoelectric microgenerators for energy harvesting. Ando Junior, Ohio, Maran, ALO and Henao, NC Overview of the development and application of thermoelectric microgenerators for energy harvesting. Ando Junior, OH, Maran, ALO & Henao, NC 回顾用于能量收集的热电微型发电机的开发和应用。 Ando Junior, OH, Maran, ALO & Henao, NC Ando Junior, Ohio, Maran, ALO, and Henao, NC are considering the development and application of thermoelectric microgenerators for energy harvesting. resume. support. Energy Rev. 91, 376–393 (2018).
Polman, A., Knight, M., Garnett, EC, Ehrler, B. & Sinke, WC Photovoltaic materials: present efficiencies and future challenges. Polman, A., Knight, M., Garnett, EC, Ehrler, B. & Sinke, WC Photovoltaic materials: present efficiencies and future challenges. Polman, A., Knight, M., Garnett, E.K., Ehrler, B. and Sinke, V.K. Photovoltaic materials: current performance and future challenges. Polman, A., Knight, M., Garnett, EC, Ehrler, B. & Sinke, WC 光伏材料：目前的效率和未来的挑战。 Polman, A., Knight, M., Garnett, EC, Ehrler, B. & Sinke, WC Solar materials: current efficiency and future challenges. Polman, A., Knight, M., Garnett, E.K., Ehrler, B. and Sinke, V.K. Photovoltaic materials: current performance and future challenges. Science 352, aad4424 (2016).
Song, K., Zhao, R., Wang, ZL & Yang, Y. Conjuncted pyro-piezoelectric effect for self-powered simultaneous temperature and pressure sensing. Song, K., Zhao, R., Wang, ZL & Yang, Y. Conjunct pyro-piezoelectric effect for self-powered simultaneous temperature and pressure sensing. Song K., Zhao R., Wang Z.L. and Yan Yu. Combined pyropiezoelectric effect for autonomous simultaneous measurement of temperature and pressure. Song, K., Zhao, R., Wang, ZL & Yang, Y. 用于自供电同时温度和压力传感的联合热压电效应。 Song, K., Zhao, R., Wang, ZL & Yang, Y. For self-powering at the same time as temperature and pressure. Song K., Zhao R., Wang Z.L. and Yan Yu. Combined thermopiezoelectric effect for autonomous simultaneous measurement of temperature and pressure. Forward. alma mater 31, 1902831 (2019).
Sebald, G., Pruvost, S. & Guyomar, D. Energy harvesting based on Ericsson pyroelectric cycles in a relaxor ferroelectric ceramic. Sebald, G., Pruvost, S. & Guyomar, D. Energy harvesting based on Ericsson pyroelectric cycles in a relaxor ferroelectric ceramic. Sebald G., Prouvost S. and Guyomar D. Energy harvesting based on pyroelectric Ericsson cycles in relaxor ferroelectric ceramics. Sebald G., Prouvost S. and Guyomar D. Energy harvesting in relaxor ferroelectric ceramics based on Ericsson pyroelectric cycling. Smart alma mater. structure. 17, 15012 (2007).
Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Next-generation electrocaloric and pyroelectric materials for solid-state electrothermal energy interconversion. Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Next-generation electrocaloric and pyroelectric materials for solid-state electrothermal energy interconversion. Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Электрокалорические и пироэлектрические материалы следующего поколения для взаимного преобразования твердотельной электротермической энергии. Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Next generation electrocaloric and pyroelectric materials for solid state electrothermal energy interconversion. Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW 用于固态电热能相互转换的下一代电热和热释电材料。 Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Электрокалорические и пироэлектрические материалы следующего поколения для взаимного преобразования твердотельной электротермической энергии. Alpay, SP, Mantese, J., Trolier-Mckinstry, S., Zhang, Q. & Whatmore, RW Next generation electrocaloric and pyroelectric materials for solid state electrothermal energy interconversion. Lady Bull. 39, 1099–1109 (2014).
Zhang, K., Wang, Y., Wang, ZL & Yang, Y. Standard and figure-of-merit for quantifying the performance of pyroelectric nanogenerators. Zhang, K., Wang, Y., Wang, ZL & Yang, Y. Standard and figure-of-merit for quantifying the performance of pyroelectric nanogenerators. Zhang, K., Wang, Y., Wang, Z.L. and Yang, Yu. A standard and quality score for quantifying the performance of pyroelectric nanogenerators. Zhang, K., Wang, Y., Wang, ZL & Yang, Y. 用于量化热释电纳米发电机性能的标准和品质因数。 Zhang, K., Wang, Y., Wang, ZL & Yang, Y. Zhang, K., Wang, Y., Wang, Z.L. and Yang, Yu. Criteria and performance measures for quantifying the performance of a pyroelectric nanogenerator. Nano Energy 55, 534–540 (2019).
Crossley, S., Nair, B., Whatmore, RW, Moya, X. & Mathur, ND Electrocaloric cooling cycles in lead scandium tantalate with true regeneration via field variation. Crossley, S., Nair, B., Whatmore, RW, Moya, X. & Mathur, ND Electrocaloric cooling cycles in lead scandium tantalate with true regeneration via field variation. Crossley, S., Nair, B., Watmore, R.W., Moya, X. and Mathur, N.D. Electrocaloric cooling cycles in lead-scandium tantalate with true regeneration by means of field modification. Crossley, S., Nair, B., Whatmore, RW, Moya, X. & Mathur, ND 钽酸钪铅的电热冷却循环，通过场变化实现真正的再生。 Crossley, S., Nair, B., Whatmore, RW, Moya, X. & Mathur, ND. Tantalum酸钪钪钪钪钪钪钪钪电求的电池水水水水水气水在电影在在线电影。 Crossley, S., Nair, B., Watmore, R.W., Moya, X. and Mathur, N.D. An electrothermal cooling cycle of scandium-lead tantalate for true regeneration through field reversal. physics Rev. X 9, 41002 (2019).
Moya, X., Kar-Narayan, S. & Mathur, ND Caloric materials near ferroic phase transitions. Moya, X., Kar-Narayan, S. & Mathur, ND Caloric materials near ferroic phase transitions. Moya, X., Kar-Narayan, S. and Mathur, N. D. Caloric materials near ferroid phase transitions. Moya, X., Kar-Narayan, S. & Mathur, ND 铁质相变附近的热量材料。 Moya, X., Kar-Narayan, S. & Mathur, ND Thermal materials near ferrous metallurgy. Moya, X., Kar-Narayan, S. and Mathur, N. D. Thermal materials near iron phase transitions. Nat. alma mater 13, 439–450 (2014).
Moya, X. & Mathur, ND Caloric materials for cooling and heating. Moya, X. & Mathur, ND Caloric materials for cooling and heating. Moya, X. and Mathur, N. D. Thermal materials for cooling and heating. Moya, X. & Mathur, ND 用于冷却和加热的热量材料。 Moya, X. & Mathur, ND Thermal materials for cooling and heating. Moya X. and Mathur N. D. Thermal materials for cooling and heating. Science 370, 797–803 (2020).
Torelló, A. & Defay, E. Electrocaloric coolers: a review. Torelló, A. & Defay, E. Electrocaloric coolers: a review. Torello, A. and Defay, E. Electrocaloric chillers: a review. Torelló, A. & Defay, E. 电热冷却器：评论。 Torelló, A. & Defay, E. 电热冷却器：评论。 Torello, A. and Defay, E. Electrothermal coolers: a review. Advanced. electronic. alma mater. 8. 2101031 (2022).
Nuchokgwe, Y. et al. Enormous energy efficiency of electrocaloric material in highly ordered scandium-scandium-lead. National communicate. 12, 3298 (2021).
Nair, B. et al. The electrothermal effect of oxide multilayer capacitors is large over a wide temperature range. Nature 575, 468–472 (2019).
Torello, A. et al. Huge temperature range in electrothermal regenerators. Science 370, 125–129 (2020).
Wang, Y. et al. High performance solid state electrothermal cooling system. Science 370, 129–133 (2020).
Meng, Y. et al. Cascade electrothermal cooling device for large temperature rise. National Energy 5, 996–1002 (2020).
Olsen, RB & Brown, DD High efficieincy direct conversion of heat to electrical energy-related pyroelectric measurements. Olsen, RB & Brown, DD High efficiency direct conversion of heat to electrical energy-related pyroelectric measurements. Olsen, R.B. and Brown, D.D. Highly efficient direct conversion of heat into electrical energy associated with pyroelectric measurements. Olsen, RB & Brown, DD 高效直接将热量转换为电能相关的热释电测量。 Olsen, RB & Brown, DD Olsen, R.B. and Brown, D.D. Efficient direct conversion of heat to electricity associated with pyroelectric measurements. Ferroelectrics 40, 17–27 (1982).
Pandya, S. et al. Energy and power density in thin relaxor ferroelectric films. National alma mater. https://doi.org/10.1038/s41563-018-0059-8 (2018).
Smith, AN & Hanrahan, BM Cascaded pyroelectric conversion: optimizing the ferroelectric phase transition and electrical losses. Smith, AN & Hanrahan, BM Cascaded pyroelectric conversion: optimizing the ferroelectric phase transition and electrical losses. Smith, A.N. and Hanrahan, B.M. Cascaded pyroelectric conversion: ferroelectric phase transition and electrical loss optimization. Smith, AN & Hanrahan, BM 级联热释电转换：优化铁电相变和电损耗。 Smith, AN & Hanrahan, BM Smith, A.N. and Hanrahan, B.M. Cascaded pyroelectric conversion: optimization of ferroelectric phase transitions and electrical losses. J. Application. physics. 128, 24103 (2020).
Hoch, S.R. The use of ferroelectric materials to convert thermal energy into electricity. process. IEEE 51, 838–845 (1963).
Olsen, RB, Bruno, DA, Briscoe, JM & Dullea, J. Cascaded pyroelectric energy converter. Olsen, RB, Bruno, DA, Briscoe, JM & Dullea, J. Cascaded pyroelectric energy converter. Olsen, R.B., Bruno, D.A., Briscoe, J.M. and Dullea, J. Cascade Pyroelectric Power Converter. Olsen, RB, Bruno, DA, Briscoe, JM & Dullea, J. 级联热释电能量转换器。 Olsen, RB, Bruno, DA, Briscoe, JM & Dullea, J. 级联热释电能量转换器。 Olsen, R.B., Bruno, D.A., Briscoe, J.M. and Dullea, J. Cascaded pyroelectric power converters. Ferroelectrics 59, 205–219 (1984).
Shebanov, L. & Borman, K. On lead-scandium tantalate solid solutions with high electrocaloric effect. Shebanov, L. & Borman, K. On lead-scandium tantalate solid solutions with high electrocaloric effect. Shebanov L. and Borman K. On solid solutions of lead-scandium tantalate with a high electrocaloric effect. Shebanov, L. & Borman, K. 关于具有高电热效应的钪铅钪固溶体。 Shebanov, L. & Borman, K. Shebanov L. and Borman K. On scandium-lead-scandium solid solutions with a high electrocaloric effect. Ferroelectrics 127, 143–148 (1992).
We thank N. Furusawa, Y. Inoue, and K. Honda for their help in creating the MLC. PL, AT, YN, AA, JL, UP, VK, OB and ED Thanks to the Luxembourg National Research Foundation (FNR) for supporting this work through CAMELHEAT C17/MS/11703691/Defay, MASSENA PRIDE/15/10935404/Defay- Siebentritt, THERMODIMAT C20/MS/14718071/Defay and BRIDGES2021/MS/16282302/CECOHA/Defay.
Department of Materials Research and Technology, Luxembourg Institute of Technology (LIST), Belvoir, Luxembourg