Materials Horizons This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online COMMUNICATION View Journal | View Issue The magnetopyroelectric effect: heat-mediated magnetoelectricity in magnetic nanoparticle- Cite this: Mater. Horiz., 2023, Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. 10, 2627 ferroelectric polymer composites† Received 31st October 2022, Accepted 18th April 2023 Joaquin Llacer-Wintle, a Jan Renz,a Lukas Hertle,a Andrea Veciana, a Denis von Arx,a Jiang Wu,a Pere Bruna, b Marija Vukomanovic, c DOI: 10.1039/d2mh01361d Josep Puigmartı́-Luis, d Bradley J. Nelson, a Xiang-Zhong Chen *a and rsc.li/materials-horizons Salvador Pané *a Magnetoelectricity enables a solid-state material to generate electricity under magnetic fields. Most magnetoelectric composites are developed New concepts through a strain-mediated route by coupling piezoelectric and This work demonstrates a heat-mediated magnetoelectric effect. Being different from the strain-mediated magnetoelectric effect, in which the magnetostrictive phases. However, the limited availability of high- deformation of the magnetostrictive material under a magnetic field performance magnetostrictive components has become a constraint strains the piezoelectric material, heat-mediated magnetoelectricity for the development of novel magnetoelectric materials. Here, we relies on the cross-talk between a material that is capable to generate demonstrate that nanostructured composites of magnetic and pyro- heat under magnetic fields (e.g., magnetic hyperthermia effect) and a electric materials can generate electrical output, a phenomenon we pyroelectric material. This approach expands the scope of the materials capable of transforming magnetic fields into electrical signals, which can refer to as the magnetopyroelectric (MPE) effect, which is analogous to open up new avenues to various non-conventional applications of the magnetoelectric effect in strain-mediated composite multiferroics. magnetoelectric effects, especially in the biomedical fields. Our composite consists of magnetic iron oxide nanoparticles (IONPs) dispersed in a ferroelectric (and also pyroelectric) poly(vinylidene fluoride–trifluoroethylene) (P(VDF–TrFE)) matrix. Under a high- and ferroelectricity – which enables magnetoelectrics to generate frequency low-magnitude alternating magnetic field, the IONPs gen- electricity under magnetic fields.1–6 This ability makes magneto- erate heat through hysteresis loss, which stimulates the depolarization electrics appealing for wide-ranging applications, including sensing, process of the pyroelectric polymer. This magnetopyroelectric low energy consumption memory devices, wireless communi- approach creates a new opportunity to develop magnetoelectric cations, energy harvesting, cell sorting, cell and tissue electro- materials for a wide range of applications. stimulation, drug delivery, and catalysis.7–11 However, magnetoelectricity is a rare property in single-phase materials as ferromagnetism and ferroelectricity are usually exclusive to each other.12 A primary challenge in developing single-phase Introduction magnetoelectric multiferroics is to discover new compounds and to develop synthetic strategies that can be used for their Recent years have seen increasing research efforts to develop manufacturing. Additionally, magnetoelectricity is usually magnetoelectric multiferroics. These materials combine two revealed at very low temperatures, with only a few exceptions ferroic orderings – ferromagnetism (or antiferromagnetism) at (or close to) room temperature such as BiFeO3.13,14 a Multi-Scale Robotics Lab (MSRL), Institute of Robotics and Intelligent Systems (IRIS), ETH Zurich, CH-8092 Zurich, Switzerland. E-mail: chenxian@ethz.ch, vidalp@ethz.ch b Departament de Fı́sica, Universitat Politècnica de Catalunya, BarcelonaTech (UPC); Institut de Tècniques Energètiques (INTE); Barcelona Research Center in Multiscale Science and Engineering, Av. Eduard Maristany 16, 08019 Barcelona, Spain c Biomaterials group, Advanced Materials Department, Jožef Stefan Institute, 1000 Ljubljana, Slovenia d Departament de Ciència dels Materials i Quı́mica Fı́sica, Institut de Quı́mica Teòrica i Computacional, 08028 Barcelona, Spain; Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain † Electronic supplementary information (ESI) available: Additional TEM and Mössbauer characterization of IONPs, films with different loading of IONPs, additional SEM and SEM-EDX of the cross section of a film, XRD of films and IONPs, schematic of the measuring setup, effect of magnetic pulse period on MPE, schematic of the electrical circuit integration the MPE film, quadratic fit of the pyroelectric coefficient as a function of temperature, phase decay fitting of the temperature of a film, a simple model to simulate magnetopyroelectricity, calculated heat transfer coefficients. See DOI: https://doi.org/10.1039/d2mh01361d This journal is © The Royal Society of Chemistry 2023 Mater. Horiz., 2023, 10, 2627–2637 | 2627 View Article Online Communication Materials Horizons To overcome these limitations, magnetoelectrics can also be nanoparticles for heating is the vast availability of materials and engineered by creating strain-mediated multiferroic compo- shapes, along with the ability to use low-magnitude magnetic sites in which a piezoelectric material is intimately coupled fields.24,25 For example, FDA-approved iron oxide nanoparticles with a magnetostrictive material. In these composites, the have been extensively used in magnetic hyperthermia to selec- This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. deformation of the magnetostrictive material under a magnetic tively kill cancerous cells, thermally trigger the release of a drug field strains the piezoelectric material, resulting in a change of or as a contrast agent for magnetic resonance imaging.26–29 its electrical polarization.15 This strategy benefits from many The heat-mediated magnetoelectric approach opens up new more options in terms of available materials and their compo- possibilities for magnetoelectricity from magnetic nanoparticles sitions, enabling a plethora of architectures ranging from 2D and pyroelectric materials. This approach enables the combi- layered superlattices and particulate composites to core–shell nation of materials that could help overcome the limitations of Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. architectures. Although strain-mediated multiferroic compo- strain-mediated magnetoelectricity and potentially unlock new sites have many potential applications, the availability of bio-related applications that were previously not feasible due to magnetostrictive materials with high strain and low toxicity is the toxicity of strain-mediated magnetoelectric materials.29 severely limited, which significantly restricts their potential Here we show that pyroelectric polymer composites contain- applications, particularly in the field of biology.16,17 ing magnetic iron oxide nanoparticles (IONPs) can generate Alternatively, the magnetoelectric effect could be attained electrical currents under alternating magnetic fields (Fig. 1). via a heat-mediated mechanism by combining materials that We refer to this phenomenon as the magnetopyroelectric (MPE) can be magnetically heated and pyroelectric materials that effect, which is analogous to the magnetoelectric effect in change their electrical polarization when exposed to temperature composite multiferroics with a coupling mechanism based on changes. Pyroelectric materials are widely used in various fields, heat instead of strain. We evaluate the impact of the alternating including commercial thermal imaging devices, thermal energy magnetic field pulse period on the MPE response, assess the harvesting, and electrochemical processes such as disinfection effect of the concentration of IONPs in the film, and characterize and pollutant degradation.18–20 Magnetocalorics are one type of the composite as a current source. A simulation is provided to material that can generate heat in the presence of a magnetic corroborate the experimental results. The magnetopyroelectric field under adiabatic conditions. However, triggering the mag- effect opens new opportunities for the use of pyroelectrics in netocaloric effect usually requires large magnetic fields – on the biomedicine, catalysis, and robotics, in which heat and electri- order of 1 to 5 T – and the library of magnetocaloric materials is city could play a synergistic role. mainly limited to Gd-based alloys, which are toxic in their non- chelated state.21–23 In contrast, magnetic nanoparticles, when exposed to an alternating magnetic field, can generate heat Results and discussion through Néel and/or Brown relaxation, which involves the rapid reorientation of their magnetic moments in response to the The IONPs implemented in the composite films are fabricated magnetic field. One of the main advantages of using magnetic via the thermal decomposition of iron oleate in eicosane. Fig. 1 The magnetopyroelectric effect is the combination of magnetic particle heating and pyroelectricity. An alternating magnetic field induces magnetic hysteresis and heats the magnetic nanoparticles causing a reversible depolarization of the pyroelectric material. 2628 | Mater. Horiz., 2023, 10, 2627–2637 This journal is © The Royal Society of Chemistry 2023 View Article Online Materials Horizons Communication Transmission electron microscopy (TEM) images (Fig. 2a and more appropriate, in accordance with the XRD results. The low Fig. S1, ESI†) show that the as-prepared particles are spherical uniformity in the shape of the hyperfine field distributions with a diameter of 21.8  1.8 nm (x%  SD, n = 50) and high indicates the presence of structural defects (Fig. S2, ESI†). Vibrat- uniformity. X-ray powder diffraction of the IONPs (Fig. S5, ESI†) ing sample magnetometry (Fig. 2b) shows that the IONPs have a This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. matches the peak pattern of magnetite and maghemite (PDF: 01- saturation magnetization (Ms) of 52 emu g1 and a coercive field 084-2782, 00-039-1346), and of hematite (PDF: 01-072-6229). (m0Hc) of 1.5 mT. The observed saturation magnetization values, Mössbauer spectra shows that 77 at% of Fe atoms are in a Fe3+ which are considerably lower than the bulk values of maghemite site, 22 at% are in a Fe2.5+ site and 1 at% of Fe atoms in a wüstite- (B70 emu g1) or magnetite (B90 emu g1), are consistent with type site. The ratio between Fe3+ and Fe2.5+ doesn’t match the the presence of structural defects typically associated with parti- typical occupancy of tetrahedral and octahedral iron sites in bulk cles synthesized via the thermal decomposition of iron oleate.30–32 magnetite [Fe3+]tet[Fe2+,Fe3+]octO4. This implies a level of oxidation Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. The heating power per unit of mass of IONPs, also denoted as and confirms that describing the samples as Fe3xO4 would be specific loss power (SLP), is calculated from the heating rate of a Fig. 2 Characterization of the composite films. (a) A TEM image (scale bar: 100 nm) of the iron oxide nanoparticles. (b) The magnetic hysteresis loop of the iron oxide nanoparticles. The inset shows the curve from 10 mT to 10 mT. (c) Magnetic heating of the iron oxide nanoparticles: the time-dependent temperature variation under an alternating magnetic field. (d) A typical composite film and (e) an SEM image of its cross section (scale bar: 1 mm). (f) X-ray diffraction patterns of P(VDF–TrFE) and composite films with different treatments. The dashed lines with pink and green color filling show the crystalline and amorphous deconvoluted peaks, correspondingly. (g) Pyroelectric coefficient as a function of temperature of films with different IONPs wt% and poling treatments, obtained by the Sharp-Garn method. For the poled samples, the dashed line represents the SE (n = 3). (h) Pyroelectric coefficient of the poled films with different IONPs wt% at the phase transition temperature. The error bars represent the SE (n = 3). This journal is © The Royal Society of Chemistry 2023 Mater. Horiz., 2023, 10, 2627–2637 | 2629 View Article Online Communication Materials Horizons dispersion of the IONPs in water (5 mg mL1) under an offset.44 The pyroelectric coefficient is then calculated as alternating magnetic field (500 kHz, 20 mT). The resulting IAmp  sin f temperature increase of the magnetic fluid is around 2.5 1C pðT Þ ¼ ; (1) A  o  TAmp in 10 s, which corresponds to an SLP of 273 W g1 (Fig. 2c). Due This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. to the high sensitivity of SLP to the size, shape, composition, where IAmp, f, A, and TAmp are current amplitude, phase offset and aggregation state of the particles,24 commonly available between current and temperature and electrode area. While the IONPs tend to have smaller SLP values o100 W g1 for setup for this method is relatively simple, it allows for accurate magnetic fields of similar frequency and intensity.33–35 IONPs and continuous characterization of the pyroelectric coefficients with much higher SLP values 41000 W g1 have been obtained as a function of temperature, and it suppresses non-pyroelectric by fine-tuning the aforementioned parameters, but their repro- contributions caused by strain-activated piezoelectric and flexo- ducibility is often complicated, because the synthesis methods electric currents.44–46 Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. are highly sensitive and their underlying mechanisms not well Using the Sharp-Garn method, the pyroelectric coefficient of understood.36,37 nine films with different IONPs concentrations, including non- Composite films are prepared by mixing different amounts poled and de-poled pure P(VDF–TrFE) films, is measured of IONPs in the poly(vinylidene fluoride–trifluoroethylene) between 25 1C and 125 1C (Fig. 2g). Poled films with 0, 0.3, (P(VDF–TrFE)) polymer, depending on the desired weight per- 1.3, 2.6, 3.9, 5.1, and 11.9 wt% IONPs display pyroelectric centage (wt%). The obtained films are homogeneous in appear- coefficients that start at 20 mC K1 m2 for 25 1C and reach ance, flexible, and brown colored (Fig. 2d and Fig. S3, ESI†). a minimum of 90 mC K1 m2 at 100 1C. Beyond this Scanning electron microscopy (SEM) was performed to image minimum, which indicates the approximate phase transition the cross-section of seven different films, showing that they temperature of P(VDF–TrFE) from ferroelectric to paraelectric have an average thickness of 83  9 mm (x%  SD, n = 7). No large state,43 the pyroelectric coefficient tends to zero as the films IONP aggregates can be observed, indicating a good dispersion lose their spontaneous polarization, reaching 0 mC K1 m2 at of the particles in the polymer mixture (Fig. 2e and Fig. S4, 120 1C. Non-poled films and films that were subject to a ESI†). X-ray diffraction in y–2y mode was used to assess the complete temperature ramp (de-poled films) show negligible effect of annealing, poling, and the loading content of IONPs pyroelectric coefficients independent of the temperature, i.e., on the crystallinity order of the P(VDF–TrFE) (Fig. 2f and p(T) = 0. The difference between the pyroelectric coefficients of Fig. S5, ESI†). The peaks at 22.71 and 23.71 are background the poled films at the phase transition temperature is statisti- signals caused by the sample holder. The peak at 19.91 is cally insignificant (ANOVA, a = 0.001) regardless of the concen- assigned to the diffraction of (200) and (110) crystalline planes tration of IONPs in the films (Fig. 2h). in the ferroelectric b-phase, and the shoulder at lower angles is To prove that the described composite films of IONPs and associated with the amorphous phase.38–42 For quantitative P(VDF–TrFE) display magnetopyroelectric coupling, the tem- analysis, the crystalline and non-crystalline regions are decon- perature changes induced by a pulsed alternating magnetic voluted using Voigt distribution functions. The degree of field are correlated to the induced pyroelectric current between the crystallinity, Xc, is calculated as the ratio of crystalline area to electrodes of the films (closed-circuit). Briefly, the experimental the sum of crystalline and non-crystalline areas under the fitted setup consists of a magnetic coil to generate the alternating peaks. Annealing of a pure P(VDF–TrFE) film increases Xc from magnetic field, a thermal camera to monitor the temperature of 0.27 to 0.64. The remaining samples, poled or with IONPs, have the films, and an electrometer to measure the generated currents. similar Xc (0.50–0.63) compared with the annealed P(VDF– The films are framed by the edges and placed in the center of the TrFE). The peaks of the annealed samples show a slight shift coil with the electrodes parallel to the magnetic field to minimize towards lower angles related to an increase in interchain eddy currents (Fig. S6, ESI†). Upon the application of the pulsed spacing, possibly caused by gauche defects introduced during alternating magnetic field, the films should be heated by the the paraelectric to ferroelectric transition after cooling from hysteretic losses of the IONPs and cooled by natural convection annealing.43 Conversely, the peaks of the poled samples show a of air, with their temperature saturating once thermal equilibrium slight shift towards higher angles, indicating that the strong is reached. Because of the pyroelectric property of the films, these electric field promotes tighter packing of the polymer chains temperature changes should then induce an electrical current within the crystals.40 proportional to the temperature change rate (Fig. 1). The pyroelectric coefficient quantifies the coupling between Magnetopyroelectric coupling is demonstrated with three temperature and the spontaneous polarization of a certain simple experiments in which the temperature and the current ƒ! material and is defined as ~ p ¼ d PS  dT 1 . To characterize of a film is measured under a pulsed alternating magnetic field. the pyroelectric coefficient of a material, the change in sponta- In the first experiment, a non-poled (p = 0) composite film neous polarization is usually measured indirectly through (11.9 wt% IONPs) is measured as a control. As expected, the voltage or current signals. For example, in the Sharp-Garn temperature of the film increases sharply, accomplishing a method, the current generated by the sample is monitored maximum temperature increase (DT) of 45 1C in 50 s of while its temperature is oscillated according to T(t) = Tamp alternating magnetic field. However, no pyroelectric current is sin(ot) + at + b, where Tamp, t, o, a, and b are temperature detected (Fig. 3a). In the second control experiment, a poled amplitude, time, angular frequency of oscillation, slope, and (p a 0) film of pure P(VDF–TrFE) is measured under the same 2630 | Mater. Horiz., 2023, 10, 2627–2637 This journal is © The Royal Society of Chemistry 2023 View Article Online Materials Horizons Communication This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. Fig. 3 Proof of magnetopyroelectric coupling. Upon the application of a pulsed alternating magnetic field (m0H = 20 mT, f = 500 kHz, Tpulse = 100 s), the temperature and the current are measured for (a) a non-poled P(VDF–TrFE) film with IONPs, (b) a poled P(VDF–TrFE) film and (c) a poled P(VDF–TrFE) film with IONPs in forward and (d) reversed orientation. conditions. In this case, the temperature of the film rises only origin (Fig. 4). The increasing error in the coefficients calculated by 10 1C, a small value caused by induction heating in the from the heating step at high temperatures and from the cooling electrodes. The corresponding measured current is small but step at low temperatures originates from the small signal-to- indicates a pyroelectric response from the poled P(VDF–TrFE) noise ratio of the measured current as the films reach thermal (Fig. 3b). In the last experiment, conducted under the same equilibrium and from the presence of eddy currents during the conditions, a poled (p a 0) composite film (11.9 wt% IONPs) is heating step. measured. Here, the high temperature change rate provided by The temperature and the current delivered by a magneto- the IONPs, combined with the pyroelectricity given by the poled pyroelectric film (11.9 wt% IONPs) is measured and simulated P(VDF–TrFE), results in significantly larger current peaks than for several pulse periods of the alternating magnetic field (Fig. 5a those in the previous two control experiments (Fig. 3c). These and Fig. S7, ESI†). The Supporting Information provides a results imply the presence of magnetopyroelectric coupling in detailed explanation of the model used for the simulation, which the composite, i.e., the alternating magnetic field induces a pyroelectric current through magnetic heating of the IONPs. To ensure that the electrical signal originates from the film, the measurements were carried with both forward and flipped film orientations (Fig. 3c and d). After the film is flipped, the measured current is switched in sign, corroborating that the magnetopyroelectric signal originates in the film. The pyro- electric coefficient of the film (11.9 wt% IONPs, T = 100 s) is calculated from its magnetopyroelectric response according to Ip pðT Þ ¼ ; (2) A  dT=dt where Ip, A and dT/dt denote the pyroelectric current, the area of the measuring electrodes and the temperature change rate, which is obtained by fitting the heating or cooling temperature curves to a two-phase exponential (R2 4 0.9995, SE o 0.16, Fig. 4 Comparison of the pyroelectric coefficients from the MPE mea- Table S2, ESI†) and calculating the corresponding derivatives. surements and from Sharp-Garn measurements. Pyroelectric coefficients The obtained pyroelectric coefficients are in good agreement are calculated from the heating and cooling steps of the MPE data (Fig. 3c with the coefficients calculated using the Sharp-Garn method, and d) according to eqn (2) and compared to pyroelectric coefficient further indicating that the measured currents are of pyroelectric obtained from the Sharp-Garn method. The dashed line indicates the SE. This journal is © The Royal Society of Chemistry 2023 Mater. Horiz., 2023, 10, 2627–2637 | 2631 View Article Online Communication Materials Horizons This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. Fig. 5 Characterization of magnetopyroelectricity. (a) Magnetopyroelectric response of a composite film (11.9 wt% IONPs) measured for different pulse periods and the corresponding simulations (Appendix S1, ESI†). (b) Temperature and (c) current for composite films with different loading content of IONPs. (d) Maximum attained temperature change rate and current as a function of the loading content of IONPs in the film. The line is a guide to the eye. (e) Current and (f) voltage delivered by a film (11.9 wt% IONPs) connected to different resistor loads. (g) Positive quadrant of the I–V plot of a film (11.9 wt% IONPs). The dashed line is a guide to the eye. relies on basic heat transfer principles and pyroelectricity. the decrease in heating power of the IONPs in the composite, Additionally, the model’s parameters are also described, and possibly caused by their immobilized state and the corres- its implementation is done using Python (Appendix S1, ESI†). ponding suppression of Brownian relaxation. The remaining Note that to obtain a good match between measured and parameters of the model were either taken from the literature48 simulated results, the simulated heating power of the IONPs in or measured experimentally (Appendix S1, Fig. S9, ESI†). The the films is set to 14% of that expected from the IONPs in water measured and simulated results show that the minimum and (SLP [W g1]IONPs [g]).47 This correction factor accounts for maximum attained temperatures within each cycle become 2632 | Mater. Horiz., 2023, 10, 2627–2637 This journal is © The Royal Society of Chemistry 2023 View Article Online Materials Horizons Communication closer to the average temperature as the pulse period is pyroelectric, and their coefficient was unaltered by the concen- shortened. This leads to smaller changes in the temperature tration of IONPs. Both effects, magnetic heating and pyroelectri- dependent pyroelectric coefficient between heating and cooling city, become coupled via heat in the composite films. As a result, steps and therefore more continuous and symmetric current we show that pyroelectric currents are triggered upon the applica- This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. peaks. A current offset, clearly visible for short pulse periods, is tion of alternating magnetic fields. The magnetopyroelectric caused by the difference between the heating and cooling rates coupling, here demonstrated with IONPs and P(VDF–TrFE), is as the average temperature of the film approaches thermal cost-effective, up-scalable, and versatile since the polymer compo- equilibrium. At thermal equilibrium, this offset tends to zero site can be easily processed and shaped as desired. This magne- and the current peaks become centered around 0 nA. Although topyroelectric approach opens a new avenue to develop the temperature variations within each cycle become undetect- magnetoelectric materials for a wide range of applications. Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. able for short pulse periods (o2.5 s), the simulations and the measured currents indicate that the heating and cooling rates remain undiminished for pulse periods down to 0.5 s. Experimental The temperature and current of films with 0, 2.6, 5.1, and 11.9 wt% IONPs are measured under pulses of alternating Synthesis of magnetic nanoparticles magnetic field (Tpulse = 25 s) to characterize the effect of particle Magnetic iron oxide nanoparticles were synthesized by thermal concentration on the magnetopyroelectric effect. Films with decomposition of iron oleate following an adapted protocol.51,52 To increased content of IONPs yield larger temperature change produce the iron oleate, FeCl36H2O (Sigma-Aldrich CAS: 10025-77- rates (Fig. 5b) and currents (Fig. 5c), both following a linear 1) (2.162 g, 8 mmol) and sodium oleate (abcr CAS: 143-19-1) (7.309 dependence with IONPs wt% (Fig. 3d). Again, the proportion- g, 24 mmol) were dissolved in a mixture of deionized water (12 mL), ality constant between maximum temperature change rate and ethanol (Thommen-Furler AG CAS: 64-17-5) (16 mL) and hexane current at room temperature is found to be 33.8  (Sigma-Aldrich CAS: 110-54-3) (28 mL) and were refluxed at 70 1C 5.2 mCK1m2 (n = 4), matching the previously reported pyro- for 4 hours with stirring. The resulting iron oleate was thoroughly electric coefficients of P(VDF–TrFE).39,40,45,49 washed with water and aged for 6 days in an open glass bottle, Pyroelectric materials act as current sources upon heating under atmospheric conditions, with stirring at 700 rpm. Once aged, or cooling. The behavior of a magnetopyroelectric film 7.43 mmol of iron oleate was mixed with sodium oleate, oleic acid (11.9 IONPs wt%) is assessed by measuring the current or (Sigma-Aldric 112-80-1) and eicosane (Sigma-Aldrich CAS: 112-95-8) voltage in a closed circuit consisting of the film and a resistor in a 100 mL three-neck round-bottom flask maintaining a molar of 0.01, 0.1, 1, or 10 MOhm (Fig. S8, ESI†). In this range, the ratio of 1 : 0.5 : 0.825 : 17.7 (iron oleate : sodium oleate : oleic acid : current delivered by the film is independent of the value of the eicosane). The flask openings were sealed, and the mixture was resistor, peaking at 25 nA during heating and 50 nA during homogenized at 50 1C and 700 rpm overnight. Then, while stirring cooling. In accordance with Ohm’s law V = IR, the voltage is at 50 1C, N2 atmosphere was introduced in the flask. After this, the proportional to the resistance, reaching 250 mV and 500 mV, thermal decomposition process was started: the mixture was correspondingly (Fig. 5e and f). In a more general approach, heated from 50 1C to 345 1C at 3 1C min1, then it was kept at the current delivered by the film is plotted as a function 345 1C for 30 min and finally it was removed from the heating of the voltage load given by a wide range of resistor values mantle and allowed to cool to 25 1C. To improve the magnetic (0–300 MOhm). For the given experimental conditions, properties of the obtained particles, the resulting mixture was this MPE film can supply a maximum average power output annealed in air at 150 1C for 2 hours (ramping time included) of B 120 nW, which corresponds to a current of 20 nA at a while stirring at 700 rpm. To remove the remaining organic voltage load of 6 V (Fig. 5g). However, the maximum output substances, the particles were washed with 10 : 1 acetone:hexane power is expected to increase if the properties of the applied and with 3 : 1 methanol:chloroform several times. The phase magnetic field (e.g., frequency, intensity or pulse period) and transfer from apolar (hexane, chlorophorm, etc.) to polar solvents the fabrication procedure of the films (e.g., electrode quality, (water, DMF) was carried by substituting the oleic acid on the poling voltage) are optimized. surface of the particles for 3, 4-dihydroxyhydrocinnamic acid The coupling between magnetic heating of IONPs and (DHCA) following a well-established ligand exchange protocol.53 pyroelectricity, which we call magnetopyroelectricity, has been Briefly, hydrophobic particles (100 mg) were transferred into THF demonstrated. First, the magnetic heating capability of IONPs (Sigma-Aldrich CAS: 109-99-9) (5 mL) in a flask, and DHCA (Sigma- fabricated via the thermal decomposition of iron oleate was Aldrich CAS: 1078-61-1) (250 mg) was dissolved in THF (30 mL) quantified by performing SLP measurements, yielding a heating in a different flask. Inert atmosphere (N2) was introduced to power of 273 W g1. Then, the pyroelectric performance of pure both mixtures through several short cycles of vacuum and N2. poled P(VDF–TrFE) was established to be  20 mCK1m2 at Then, the dispersion of particles was added dropwise to the room temperature via the Sharp-Garn method, matching values solution of DHCA, previously heated to 50 1C. The mixture was reported in literature.38,39,45,49,50 The functionality of each mate- left stirring thoroughly at 50 1C for 3–5 hours and then allowed to rial was conserved after being integrated into a composite film. cool slowly. Particles from the resulting dispersion were precipi- IONPs induced a temperature increase in the films of up to 45 1C tated with NaOH (Sigma-Aldrich CAS: 1310-73-2) (0.5–1 mL, 0.5 M) in 50 s under an alternating magnetic field. The films remained and subsequently washed with water until the pH was restored to 7. This journal is © The Royal Society of Chemistry 2023 Mater. Horiz., 2023, 10, 2627–2637 | 2633 View Article Online Communication Materials Horizons The particles were stored in water (typically no longer than measurement was weighed and then mixed in melted eicosane, two weeks). which provided a solid pellet with the magnetic particles fixed in it at room temperature. Fabrication of pyroelectric films This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. Pyroelectric films were obtained by the annealing and poling of Transmission mössbauer spectroscopy drop-casted P(VDF–TrFE). First, it was established that given The Transmission Mössbauer Spectra (TMS) were obtained in the density (1.9 g cm3) and size of the substrate (glass slide, transmission mode at room temperature and pressure using a 26  76 mm), 0.37 g of P(VDF–TrFE) were needed to obtain B conventional spectrometer with constant acceleration with a 100 mm films. Then, P(VDF–TrFE) (70/30) (Piezotech FC30) 25 mCi 57Co radioactive source in a Rh matrix. The spectra (0.37 g) was dissolved in dimethylformamide (DMF) (Sigma- were recorded in a standard multichannel analyzer using a velocity range of 12.2 mm s1, and were subsequently fitted Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. Aldrich, CAS: 68-12-2) at a concentration of 10 wt%. The viscous mixture was poured on the substrate and dried in a ventilated with the NORMOS software. The samples were fitted using one oven at 60 1C for 4 3 hours. The obtained films were annealed or serveral distribution of hyperfine fields and one paramag- overnight at 120 1C. Then the films were peeled off from the netic singlet. The fitted parameters are the isomer shift (d), glass substrate and cut into pieces (13  38 mm). An electrode always expressed relative to the isomer shift of the bcc-Fe at (8  30 mm) was painted at the centre of each film with silver room temperature, the quadrupole splitting (D) and the hyper- paint (PELCO Colloidal Silver 16034). Finally, the films were fine magnetic field (BHF). In all cases the area expressed in % poled for 30 min at 120 1C under a corona electric field (13 kV, corresponds to the fraction of Fe atoms in a particular environ- needle-film distance 2 cm), in which the painted electrode was ment with respect the total amount of Fe atoms and the grounded while the exposed side of the film was facing the magnitudes inside parentheses are the standard deviation. biased needle. To ensure the film remained poled, the electric field was kept on as the films were cooled to room temperature. Specific loss power Pyroelectric films integrating magnetic nanoparticles were pre- In a typical specific loss power (SLP) measurement, a sample of pared following the same procedure, but the P(VDF–TrFE) powder IONPs dispersed in water (5 mg mL1) was placed in the center was dissolved in a dispersion of DMF with the corresponding of an alternating magnetic field ( f = 500 kHz, m0H = 20 mT) and weight of magnetic nanoparticles. Note that the IONPs, which had its temperature was monitored with an optic temperature been stored in water, were washed thoroughly with DMF before sensor (Osensa PRB-G40-01M-STM-MRI). The SLP was then the P(VDF–TrFE) was added. The mixture was homogenized calculated as without magnetic stirring to avoid particle aggregation. C dT SLP ¼ Transmission electron microscopy m dt The transmission electron microscopy (TEM) measurements where C is the specific heat capacity of the dispersion per unit were carried out on a FEI Talos F200X (Chem S/TEM) at 200 kV volume and m is the iron oxide concentration. The temperature equipped with an X-FEG emitter and a CETA camera (16 M pixel change rate was only considered between t = 4 s and t = 10 s to CMOS Camera). The synthesized nanoparticles were diluted in avoid the initial non-linear regime. It is assumed that the hexane and dropped on a carbon-coated 400 mesh TEM grid. system is adiabatic for the given time range and that C E The grid was dried under air. Cwater = 4.184 J K1 mL1.54 Scanning electron microscopy Pyroelectric coefficient The scanning electron microscopy (SEM) measurements were Pyroelectricity of the films was assessed using a custom-made performed on a Zeiss ULTRA 55 at 5 kV equipped with a FEG setup based on the Sharp-Garn method. Samples consisted of emitter and an Inlens SE detector. The films were fractured films (13  38 mm) to which an electrode (8  30 mm) was in liquid nitrogen to avoid shear deformation on the edges, painted on each side with silver paint. In a typical measurement, allowing us to image their true thickness. the sample was clamped on top of a thermoelectric plate and its electrodes were connected to a Keithley 6517A electrometer. The X-ray diffraction temperature of the sample, controlled by a PID feedback loop, was X-ray diffraction (XRD) measurements were performed on an varied sinusoidally with an amplitude, frequency, and slope of Empyrean diffractometer with a copper X-ray source (l = 1.5406 Å) 1 1C, 0.01 Hz and 7.2 1C h1, ensuring thermal equilibrium. The and a PIXcel detector (Malvern Panalytical). The measured mag- temperature and current readings were stored and processed in a netic particles were phase transferred into water and freeze-dried. Raspberry Pi. The pyroelectric coefficient as a function of the Films were measured without further treatment. temperature could be calculated as stated in eqn (1). Vibrating sample magnetometry Magnetopyroelectric characterization Magnetic characteristics were evaluated using a vibrating sample Magnetopyroelectric currents were measured with a custom- magnetometer (VSM) (MicroSense). To prepare the sample, the made setup. The samples, prepared in the same way as for powder of magnetic particles used for the X-ray diffraction pyroelectric coefficient measurements, were framed using acryl 2634 | Mater. Horiz., 2023, 10, 2627–2637 This journal is © The Royal Society of Chemistry 2023 View Article Online Materials Horizons Communication holders, which left the electrode area exposed to air convection. (No. CRSK-2_190451). JLL acknowledges his PhD grants in the The electrodes of the sample were contacted with two small framework of the project BeMAGIC funded by the European pieces of copper foil that protruded from each acryl holder and Union’s Horizon 2020 Research and Innovation Programme which were connected to the Keithley 6517A electrometer. under the Marie Skłodowska-Curie grant agreement No. 861145. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. The sample was then placed in the center of the induction coil PB acknowledges support from grant PID2020-112975GB-I00 (magneTherm, NanoTherics) and pulses of alternating funded by MCIN/AEI/10.13039/501100011033 and from grant magnetic fields ( f = 500 kHz, m0H = 20 mT, fpulse = 0.01–2 Hz) AGAUR 2017-SGR-42 funded by Generalitat de Catalunya. M. V. were applied to the sample to induce heating and allow cooling. acknowledges support from ARRS Grant N-0150. J. P.-L. acknowl- The generated pyroelectric current or voltage was measured edges support from the European Research Council Starting with the electrometer and the temperature was monitored with Grant microCrysFact (ERC-2015-STG No. 677020), the Swiss Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. an infrared camera (Fluke Ti Infrared Camera). Resistors, when National Science Foundation (project no. 200021_181988), and used, were introduced far from the induction coil. Note that grant PID2020-116612RB-C33 funded by MCIN/AEI/10.13039/ the copper foil and wires used to connect the sample to the 501100011033. electrometer were thin and fixed to a position parallel to the applied magnetic field, minimizing electrical noise and induc- tion heating. References 1 M. Fiebig, Revival of the magnetoelectric effect, J. Phys. D: I–V plot Appl. Phys., 2005, 38, R123–R152. The current and voltage output of a film (11.9 wt% IONPs) was 2 N. A. Spaldin and M. Fiebig, The Renaissance of Magneto- measured for load resistor values from 0 to 300 MOhm (smaller electric Multiferroics, Science, 2005, 309, 391–392. resistors were placed in series). The period of the pulsed 3 P. Martins and S. Lanceros-Méndez, Polymer-Based Magne- alternating magnetic field was 2.5 s. For each resistor load, toelectric Materials, Adv. Funct. Mater., 2013, 23, 3371–3385. the average (n = 100–120) of the current and the voltage was 4 X.-Z. Chen, et al., Hybrid Magnetoelectric Nanowires for calculated from the values ranging from 50 to 100 s. Nanorobotic Applications: Fabrication, Magnetoelectric Coupling, and Magnetically Assisted In Vitro Targeted Drug Statistical information Delivery, Adv. Mater., 2017, 29, 1605458. The pyroelectric coefficients obtained via the Sharp-Garn 5 A. Nicolenco, et al., Strain gradient mediated magnetoelec- method were measured in triplicate (different films, different tricity in Fe-Ga/P(VDF-TrFE) multiferroic bilayers integrated days). The error-bars of the I–V plot correspond to the SD (n = on silicon, Appl. Mater. Today, 2020, 19, 100579. 100–120). 6 X.-Z. Chen, et al., Magnetoelectric micromachines with wirelessly controlled navigation and functionality, Mater. Author contributions Horiz., 2016, 3, 113–118. 7 N. A. Spaldin and R. Ramesh, Advances in magnetoelectric JLL contributed to the design, data acquisition, data analysis, multiferroics, Nat. Mater., 2019, 18, 203–212. data interpretation and drafting of this work; JR contributed to 8 M. Bibes and A. Barthélémy, Towards a magnetoelectric the data acquisition and analysis; LH contributed to the devel- memory, Nat. Mater., 2008, 7, 425–426. opment of the magnetic nanoparticles; AV contributed to the 9 S. Kopyl, R. Surmenev, M. Surmeneva, Y. Fetisov and data acquisition; DVA contributed to the data interpretation; PB A. Kholkin, Magnetoelectric effect: principles and applica- contributed to the acquisition and interpretation of TMS; XC tions in biology and medicine– a review, Mater. Today Bio, contributed to the conception, data interpretation, funding 2021, 12, 100149. acquisition, and supervision; SP contributed to the conception, 10 M. Dong, et al., 3D-Printed Soft Magnetoelectric Microswim- funding acquisition, and supervision. All the authors contrib- mers for Delivery and Differentiation of Neuron-Like Cells, uted to the writing and revision of the manuscript. Adv. Funct. Mater., 2020, 30, 1910323. 11 D. Kim, et al., Magnetoelectric Effect in Hydrogen Harvest- Conflicts of interest ing: Magnetic Field as a Trigger of Catalytic Reactions, Adv. Mater., 2022, 34, 2110612. There are no conflicts to declare. 12 N. A. Hill, Why Are There so Few Magnetic Ferroelectrics?, J. Phys. Chem. B, 2000, 104, 6694–6709. 13 J. Wang, et al., Epitaxial BiFeO3 Multiferroic Thin Film Acknowledgements Heterostructures, Science, 2003, 299, 1719–1722. SP acknowledges funding from a Consolidator Grant from the 14 S. Picozzi and C. Ederer, First principles studies of multi- European Research Council (ERC) under the European Union’s ferroic materials, J. Phys.: Condens. Matter, 2009, 21, 303201. Horizon 2020 Research and Innovation Programme (Grant 15 Y. Wang, J. Hu, Y. Lin and C.-W. Nan, Multiferroic magneto- Agreement No. 771565) and Swiss National Science Foundation electric composite nanostructures, NPG Asia Mater., 2010, 2, (SNF) (No. 192012). X. C. is partially supported by SNF 61–68. This journal is © The Royal Society of Chemistry 2023 Mater. Horiz., 2023, 10, 2627–2637 | 2635 View Article Online Communication Materials Horizons 16 J. Wang and W. Gao, Nano/Microscale Motors: Biomedical 33 S. Ota, et al., Quantitation method of loss powers using Opportunities and Challenges, ACS Nano, 2012, 6, 5745–5751. commercial magnetic nanoparticles based on superpara- 17 F. Narita and M. Fox, A Review on Piezoelectric, Magnetos- magnetic behavior influenced by anisotropy for hyperther- trictive, and Magnetoelectric Materials and Device Technol- mia, J. Magn. Magn. Mater., 2021, 538, 168313. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. ogies for Energy Harvesting Applications, Adv. Eng. Mater., 34 M. Kallumadil, et al., Suitability of commercial colloids for 2018, 20, 1700743. magnetic hyperthermia, J. Magn. Magn. Mater., 2009, 321, 18 D. Zhang, H. Wu, C. R. Bowen and Y. Yang, Recent Advances 1509–1513. in Pyroelectric Materials and Applications, Small, 2021, 35 O. L. Lanier, et al., Evaluation of magnetic nanoparticles for 17, 2103960. magnetic fluid hyperthermia, Int. J. Hyperthermia, 2019, 36, 19 R. W. Whatmore, Pyroelectric devices and materials, Rep. 686–700. Prog. Phys., 1986, 49, 1335–1386. 36 J.-H. Lee, et al., Exchange-coupled magnetic nanoparticles for Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. 20 C. R. Bowen, et al., Pyroelectric materials and devices for efficient heat induction, Nat. Nanotechnol., 2011, 6, 418–422. energy harvesting applications, Energy Environ. Sci., 2014, 7, 37 J. Muro-Cruces, et al., Precise Size Control of the Growth of 3836–3856. Fe3O4 Nanocubes over a Wide Size Range Using a Rationally 21 V. Franco, et al., Magnetocaloric effect: From materials research Designed One-Pot Synthesis, ACS Nano, 2019, 13, 7716–7728. to refrigeration devices, Prog. Mater. Sci., 2018, 93, 112–232. 38 J. F. Legrand, Structure and ferroelectric properties of 22 J. Ramalho, M. Ramalho, M. Jay, L. M. Burke and P(VDF-TrFE) copolymers, Ferroelectrics, 1989, 91, 303–317. R. C. Semelka, Gadolinium toxicity and treatment, Magn. 39 A. Aliane, et al., Impact of crystallization on ferro-, piezo- Reson. Imaging, 2016, 34, 1394–1398. and pyro-electric characteristics in thin film P(VDF–TrFE), 23 E. Akbas, F. Unal and D. Yuzbasioglu, Cellular toxicities of Org. Electron., 2015, 25, 92–98. gadolinium-based contrast agents used in magnetic reso- 40 C. Y. B. Ng, W. C. Gan, T. S. Velayutham, B. T. Goh and nance imaging, J. Appl. Toxicol., 2022, 1–15. R. Hashim, Structural control of the dielectric, pyroelectric 24 J. Carrey, B. Mehdaoui and M. Respaud, Simple models for and ferroelectric properties of poly(vinylidene fluoride-co- dynamic hysteresis loop calculations of magnetic single- trifluoroethylene) thin films, Phys. Chem. Chem. Phys., 2020, domain nanoparticles: Application to magnetic hyperther- 22(4), 2414–2423. mia optimization, J. Appl. Phys., 2011, 109, 083921. 41 Y. Li, et al., Investigation on in-situ sprayed, annealed and 25 R. Chen, M. G. Christiansen and P. Anikeeva, Maximizing corona poled PVDF-TrFE coatings for guided wave-based Hysteretic Losses in Magnetic Ferrite Nanoparticles via structural health monitoring: From crystallization to piezo- Model-Driven Synthesis and Materials Optimization, ACS electricity, Mater. Des., 2021, 199, 109415. Nano, 2013, 7, 8990–9000. 42 Y. Tajitsu, H. Ogura, A. Chiba and T. Furukawa, Investiga- 26 S. Laurent, S. Dutz, U. O. Häfeli and M. Mahmoudi, Mag- tion of Switching Characteristics of Vinylidene Fluoride/ netic fluid hyperthermia: Focus on superparamagnetic iron Trifluoroethylene Copolymers in Relation to Their Struc- oxide nanoparticles, Adv. Colloid Interface Sci., 2011, 166, tures, Jpn. J. Appl. Phys., 1987, 26, 554–560. 8–23. 43 R. Gregorio Jr. and M. M. Botta, Effect of crystallization 27 C. A. Quinto, P. Mohindra, S. Tong and G. Bao, Multi- temperature on the phase transitions of P(VDF/TrFE) copo- functional superparamagnetic iron oxide nanoparticles for lymers, J. Polym. Sci., Part B: Polym. Phys., 1998, 36, 403–414. combined chemotherapy and hyperthermia cancer treat- 44 L. E. Garn and E. J. Sharp, Use of low-frequency sinusoidal ment, Nanoscale, 2015, 7, 12728–12736. temperature waves to separate pyroelectric currents from 28 I. Hilger and W. A. Kaiser, Iron oxide-based nanostructures nonpyroelectric currents. Part I. Theory, J. Appl. Phys., 1982, for MRI and magnetic hyperthermia, Nanomedicine, 2012, 7, 53, 8974–8979. 1443–1459. 45 S. Jachalke, et al., How to measure the pyroelectric coeffi- 29 E. Hannachi and Y. Slimani, Advanced Progress in Magneto- cient?, Appl. Phys. Rev., 2017, 4, 021303. electric Multiferroic Composites: Fundamentals, Applica- 46 I. Lubomirsky and O. Stafsudd, Invited Review Article: tions, and Toxicity, in Handbook of Magnetic Hybrid Practical guide for pyroelectric measurements, Rev. Sci. Nanoalloys and their Nanocomposites, ed. Thomas, S. and Instrum., 2012, 83, 051101. Rezazadeh Nochehdehi, A., Springer International Publish- 47 M. Avolio, et al., In-gel study of the effect of magnetic ing, 2022, pp. 351–385, DOI: 10.1007/978-3-030-90948-2_52. nanoparticles immobilization on their heating efficiency 30 D. Cabrera, et al., Unraveling viscosity effects on the hyster- for application in Magnetic Fluid Hyperthermia, J. Magn. esis losses of magnetic nanocubes, Nanoscale, 2017, 9, Magn. Mater., 2019, 471, 504–512. 5094–5101. 48 C. Y. Iguchi, W. N. dos Santos and R. Gregorio, Determina- 31 A. Lak, S. Disch and P. Bender, Embracing Defects and tion of thermal properties of pyroelectric polymers, copoly- Disorder in Magnetic Nanoparticles, Adv. Sci., 2021, 8, 2002682. mers and blends by the laser flash technique, Polym. Test., 32 Z. Chen, Size and Shape Controllable Synthesis of Monodis- 2007, 26, 788–792. perse Iron Oxide Nanoparticles by Thermal Decomposition of 49 T. Tansel, Effect of electric field assisted crystallisation of Iron Oleate Complex, Synth. React. Inorg., Met.-Org., Nano-Met. PVDF-TrFE and their functional properties, Sens. Actuators, Chem., 2012, 42, 1040–1046. A, 2021, 332, 113059. 2636 | Mater. Horiz., 2023, 10, 2627–2637 This journal is © The Royal Society of Chemistry 2023 View Article Online Materials Horizons Communication 50 Y. Wada and R. Hayakawa, A Model Theory of Piezo- and 53 Y. Liu, et al., Facile Surface Functionalization of Hydropho- Pyro-electricity of Poly (vinylidene fluoride) Electret, Ferro- bic Magnetic Nanoparticles, J. Am. Chem. Soc., 2014, 136, electrics, 1981, 32, 115–118. 12552–12555. 51 E. Wetterskog, et al., Precise control over shape and size of 54 S.-Y. Wang, S. Huang and D.-A. Borca-Tasciuc, Potential This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. iron oxide nanocrystals suitable for assembly into ordered Sources of Errors in Measuring and Evaluating the particle arrays, Sci. Technol. Adv. Mater., 2014, 15, 055010. Specific Loss Power of Magnetic Nanoparticles in an Alter- 52 J. Park, et al., Ultra-large-scale syntheses of monodisperse nating Magnetic Field, IEEE Trans. Magn., 2013, 49, nanocrystals, Nat. Mater., 2004, 3, 891–895. 255–262. Open Access Article. Published on 18 April 2023. Downloaded on 12/5/2023 12:14:32 PM. This journal is © The Royal Society of Chemistry 2023 Mater. Horiz., 2023, 10, 2627–2637 | 2637