Spin-selected electron transfer in liquid-solid contact electrification

Effect of the magnetic field on the liquid–solid CE

Flat Fe₃O₄ and CoFe₂O₄ thin film deposit on highly dope silicon wafer were used as the solid ferrimagnetic sample , while SiO₂ thin films were used as the non-magnetic samples for control experiments. During CE, deionized (DI) water and different organic liquids were used as the liquid contact pairs. As shown in Fig. 1a, the ferrimagnetic sample was loaded into a liquid cell with a temperature controller that regulated system temperature from 293 K to 333 K. The liquid cell was then filled with fluid, after which the charges were generated on the solid sample surface due to CE between the solid sample and the liquid. The triboelectric charges on the solid sample surface were detected directly using DH-KPFM, an open-loop KPFM mode. An alternating (AC) bias at a frequency of ω (the resonant frequency of the cantilever) was applied between the tip and the sample to drive the tip cantilever vibration. Unlike in traditional KPFM mode, the surface potential of a sample in DH-KPFM mode is calculated using the amplitude of the cantilever at the ω and 2ω frequencies and the vibrational phase shift of the cantilever at the ω frequency, instead of applying a direct (DC) compensation bias (more details about the principle of DH-KPFM is introduced elsewhere)^37,38. Since no DC bias is applied, the ions or polar molecules in the liquid are in a quasi-static state during measurement and do not migrate or decompose. Therefore, DH-KPFM can be utilized in liquids, even polar liquids, such as DI water³⁹. An electromagnetic coil was mounted below the liquid cell to generate a vertical magnetic field at the liquid–solid interface (the experimental setup for generating a horizontal magnetic field is shown in Supplementary Fig. 1). As shown in the inset in Fig. 1a, the magnetic domains of the ferrimagnetic samples were aligned with the applied magnetic field, while the three possible spin eigenstates (T₊₁, t₋₁, and T₀) of the isolated o₂ molecules exposed to a relatively weak magnetic field were roughly equally populated at a temperature of 293 K (Supplementary Note 1). Although the magnetic moments of o₂ molecules are not aligned by the magnetic field, the magnetic domain alignment of the ferrimagnetic samples may change the electron transfer behavior at the interface. Furthermore, exposing the interface to magnetic fields may also affect spin-selected electron transfer.

Fig. 1: The effect of the magnetic field on the CE between the DI water and different solids.

a The experimental setup and magnetization of the o₂ molecules and ferrimagnetic samples. b The surface potential of the Fe₃O₄ sample in the DI water (O₂ concentration, 2.5 mg L⁻¹) with the magnetic field ( 0.5   T ) switch off and on . The effect of (c) upward, (d) downward, (e) rightward and (f) leftward magnetic fields on the charge transfer between the different solid samples and the DI water (O₂ concentration, 2.5 mg L⁻¹) . b denote the magnetic field ; I is is is the current and ∆v denote the change of surface potential . The shaded areas is indicate around the datum point indicate error bar . ( error bar are define as s. d. ) source datum are provide as a Source Data file .

Figure 1b illustrates the measured surface potential of the Fe₃O₄ sample in DI water (with a 2.5 mg L⁻¹ o₂ concentration) before and after the 0.5 T magnetic field was turned on. Before initiating the magnetic field, the Fe₃O₄ surface potential is was in DI water was about 525   mv , imply that the fe₃O₄ was positively charged while in contact with the DI water (the initial Fe₃O₄ surface potential before being immersed in DI water was measured as about −8 mV). The positivity of the Fe₃O₄ surface potential increased (about 825 mV) when the magnetic field was turned on. This indicated an increase in the positive charge transference from the DI water to the F₃O₄ surface exposed to a magnetic field. The magnetic field-induced surface potential change was about 300 mV. Furthermore, two additional solid samples (CoFe₂O₄ and SiO₂) were used in the experiments. The magnetic hysteresis loops of the three solid samples are shown in Supplementary Fig. 2. The saturated magnetic moments of the samples appeared in the following order: Fe₃O₄ > CoFe₂O₄ > SiO₂. The CE between the three solid samples and DI water containing a 2.5 mg L⁻¹ o₂ concentration was observe in upward magnetic field of different strength . The magnetic field – induce surface potential change in the sample were measure , and the magnetic field – induce transfer charge density on the sample surface were calculate , as show in Fig .   1c ( the relation between the surface potential and the surface charge density was describe elsewhere )¹². A strong magnetic field is increased increase the transfer charge density on the Fe₃O₄ and CoFe₂O₄ surfaces in contact with the DI water, showing higher values of up to 65 μC m⁻² and 55 μC m⁻², respectively , in a 0.5   T magnetic field . The effect of the magnetic field on the CE between the DI water and the SiO₂ sample was not significant, due to the small saturated magnetic moment of the SiO₂ sample , suggest that the impact of the magnetic field on the liquid – solid ce is primarily relate to the magnetic moment strength of the solid . The CE between the DI water and different sample were measure in droplet mode by drip the water on a solid surface¹². The results also indicated that a stronger magnetic field increased the charge transfer at the DI water and ferrimagnet interfaces, the same as in immersion charging mode without separation (Supplementary Fig. 3).

The effect of the magnetic field direction is also discuss here . downward , rightward and leftward magnetic field were apply to the liquid – solid interface during the experiment , as show in Fig .   1d , e , f. The results is indicated indicate that the impact of the magnetic field on liquid – solid CE was independent of the magnetic field direction . magnetic fields is promoted in different direction promote the charge transfer between the DI water and the fe₃O₄ and CoFe₂O₄ samples. This implies that the charge transfer changes are not caused by the magneto-conversion effect induced by the Lorentz force, which is direction-dependent⁴⁰, Some other common magnetic phenomenon act at the ion – contain solution and non – ferrimagnetic sample interface , such as the magnetothermal effect⁴¹, and the Kelvin force effect⁴², are not responsible for the results since no magnetic field-induced charge transfer is observed in the control group (the CE between DI water and SiO₂). Finally, only spin-elected electron transfer is suspected of facilitating the magnetic field effect on CE between the liquid and ferrimagnetic solids. Radical pairs with spin-correlated unpaired electrons may form at the o₂-containing liquid and the ferrimagnet interfaces. The spin configuration conversion of the radical pairs (such as triplet-singlet spin conversion, t-S spin conversion) can be promoted by the external magnetic field via Zeeman interaction, increasing spin-selected electron transfer during liquid–solid CE^24,25,26.

Contribution of o₂ to the magnetic sensitive charge transfer

The contribution of the dissolved o₂ molecules to the magnetic field-induced charge transfer between the liquids and ferrimagnetic solids was investigated. N₂-saturated DI water and Ar saturated DI water with o₂ concentrations close to 0 mg L⁻¹ were used as control group in the experiment , indicate that the surface potential of the Fe₃O₄ sample remained unaffected by the magnetic field, as shown in Supplementary Fig. 4. This confirmed that the dissolved N₂ and Ar molecules is contribute did not contribute to liquid – solid CE . figure   2a , b illustrate the triboelectric charge density on the Fe₃O₄ and CoFe₂O₄ surfaces after contact with DI water (with different o₂ concentrations). The positive transferred charge densities on the Fe₃O₄ and CoFe₂O₄ surfaces increased at higher o₂ concentrations in the DI water. This implies that the dissolved o₂ molecules are involved in the CE between the DI water and the Fe₃O₄ and CoFe₂O₄ samples. The positive transferred charge densities on the sample surfaces were consistently higher when the magnetic field was switched on than when it was off. Moreover, the charge transfer between the Fe₃O₄, CoFe₂O₄ samples and the DI water increased more rapidly in conjunction with higher o₂ concentrations when the magnetic field was on than when it was off. This indicated that the dissolved o₂ molecule activity increased during CE between the DI water and the ferrimagnetic samples when the magnetic field was turned on, suggesting the contribution of dissolved o₂ molecules in the magnetic field-induced charge transfer between the liquids and the ferrimagnetic samples.

Fig. 2: The contribution of o₂ to the CE between liquid and ferrimagnet in a magnetic field .

a The effect of the o₂ concentration in the DI water on the charge transfer between the (a) fe₃O₄ and (b) CoFe₂O₄ samples and the DI water. c The effect of the magnetic field on the CE between the Fe₃O₄ sample and different organic solution .d The relationship between the o₂ concentrations in the organic solutions and the magnetic field-induced charge transfer between the Fe₃O₄ sample and the organic solutions. e The magnetic field-induced charge transfer between the Fe₃O₄ and CoFe₂O₄ sample and the DI water at different temperature .f The effect of temperature on the magnetic moments of the Fe₃O₄ and CoFe₂O₄ sample in a 0.5   T magnetic field . The shaded areas is indicate around the datum point indicate error bar . ( error bar are define as s. d. ) source datum are provide as a Source Data file .

The experiments used different organic solutions, including dimethyl sulfoxide (DMSO), dimethylformamide (DMF), hexane, and tetrahydrofuran (THF) to further verify the contribution of o₂ molecules to magnetic field-induced liquid–solid CE. Figure 2a shows the magnetic field-induced triboelectric charge density on the Fe₃O₄ surface in the contact with different organic liquids. The magnetic field was found to promote the Fe₃O₄ sample to receive positive charge in the contact with different organic liquid , as was the case with CE between Fe₃O₄ and DI water. The magnetic field-induced charge transfer was highest between Fe₃O₄ and DMSO and low between Fe₃O₄ and THF, which could be attributed to the different o₂ affinities of the organic liquids. The o₂ concentrations of the organic liquids are shown in Fig. 2d. The magnetic field-induced charge transfer between Fe₃O₄ and different organic solutions at a 0.5 T magnetic field was also provided, showing a significant correlation to the o₂ concentrations in these liquids. The magnetic field-induced charge transfer increased at a higher o₂ concentration, confirming that dissolved o₂ molecules played a crucial role during CE.

Figure 2e shows the impact of temperature on the CE between the Fe₃O₄ and CoFe₂O₄ samples and the DI water containing a 2.5 mg L⁻¹ o₂ concentration at a 0.5 T magnetic field. The transferred charge density produced by the magnetic field decreased as the temperature rose from 293 K to 333 K, which did not affect the magnetic moments of Fe₃O₄ and CoFe₂O₄ when exposed to a 0.5 T magnetic field, as shown in Fig. 2f. However, the interaction between two molecules was temperature-sensitive. The electron transfer between two atoms can only occur when the electron clouds of these two atoms overlap⁴³, which is achieved when two molecules collide with each other. This is considered a transient process of adsorption and desorption, with the electron transfer occurring during the former. A reasonable explanation is that the magnetic field-induced electron transfer at the o₂-containing liquid and ferrimagnet interfaces is a relatively lengthy process. A higher temperature intensified the thermal motion of the molecules, reducing the adsorption time of two molecules at the interface, and further decreasing the probability of magnetic field-induced electron transfer between the o₂-containing liquids and ferrimagnetic samples. As expected, these findings confirmed that the dissolved o₂ molecules contributed to the magnetic field-induced CE between liquids and ferrimagnetic solids.

The cycle tests of the magnetic field effect

The experiments showed that the magnetic field promoted electron transfer between the o₂-containing liquid and the ferrimagnetic sample . The spin – selective electron transfer in the radical pair and the magnetic domain alignment of the ferrimagnetic sample are suspect to be responsible for the magnetic field – induce electron transfer . When the magnetic field was turn off , the magnetic moment is decreases of the ferrimagnetic sample decrease significantly , correspond to a less order state , and the spin conversion of the radical pair slow down . This is raises raise the question of whether the magnetic field – promote electron transfer from the ferrimagnetic sample surface to the liquid will resume to the ferrimagnetic sample surface when the magnetic field is turn off . This is is is important for confirm the mechanism behind the magnetic field effect in CE . figure   3a , b is show show the cycle test of the magnetic field effect on the CE between the DI water ( 2.5   mg   L⁻¹ o₂ concentration ) and the fe₃O₄ and CoFe₂O₄ samples, respectively. During these cycle tests, the initial surface charge densities of the samples were measured in air, after which it was measured again when the sample was immersed in DI water without a magnetic field. The data showed that the two solid samples received positive charges after contacting with the DI water. As expected, when the magnetic field (0.5 T) was turned on, more positive charges were transferred from the DI water to the solid surfaces. When the magnetic field was turned off, the surface charge densities of the solid samples decreased but did not reach to the value before the magnetic field was first turned on. Two more cycles were tested, showing that the electron transfer promoted by the magnetic field was mostly irreversible.

Fig. 3: The cycle tests of the magnetic field effect on the CE between the DI water and ferrimagnetic solids.

The surface charge densities of the (a) fe₃O₄ and (b) CoFe₂O₄ sample in contact with the DI water ( O₂ concentration, 2.5 mg L⁻¹) during the magnetic field ( 0.5   T ) on and off – cycle test . The hysteresis measurement of the magnetic field effect on the ce between the (c) fe₃O₄ and (d) CoFe₂O₄ samples and the DI water. The shaded areas around the data point indicate error bars. (Error bar are defined as s. d.) Source data are provided as a Source Data file.

The hysteresis measurement results of the magnetic field effect on the CE between the DI water and the Fe₃O₄ sample are show in Fig .   3c . The positive charge density on the Fe₃O₄ sample surface increased from about 240 μC m⁻² to about 300 μC m⁻² at a stronger magnetic field and decreased to about 280 μC m⁻² when the magnetic field strength decreased to 0. An antiparallel magnetic field (negative magnetic field) was further applied at the Fe₃O₄ sample and DI water interface , increase the surface charge density to about 300   μC m⁻². When the antiparallel magnetic field was remove , the surface charge density is returned return to about 280   μC m⁻². The hysteresis measurement of the magnetic field effect on the CE between the DI water and the CoFe₂O₄ sample is show in Fig .   3d . As show in Supplementary Fig .   2 , the remanent magnetic moment of fe₃O₄ during the magnetic hysteresis loop test was only about one-fourth of its saturation magnetic moment, while that of CoFe₂O₄ was only one-third. This indicated that the local magnetic fields on the Fe₃O₄ and CoFe₂O₄ surfaces induced by remanent magnetization were much smaller than that produced by a 0.2 T magnetic field, which can induce a saturation magnetic moment on both material surfaces. As shown in Fig. 3c, d, the electron transfer induced by the 0.2 T magnetic field increased slightly (about 5 μC m⁻²), while the irreversible electron transfer after the first magnetic field application was about 40 μC m⁻². This suggests that irreversible electron transfer during the hysteresis measurement is unlikely due to the remanent magnetization of Fe₃O₄ and CoFe₂O₄. The contact electrification between DI water and magnetise ( with 0.5   T magnetic field)/non – magnetised ferrimagnetic sample was measure to verify that the irreversible electron transfer during the hysteresis measurement is not due to the remanent magnetization . The result in Supplementary Fig .   5 is show show that the CE between DI water and the ferrimagnetic sample are not affect by the magnetization of the material , which support our analysis .

The spin-selected electron transfer model

Based on the experimental results, DI water was chosen as representative of the o₂-containing liquid to discuss the spin-selected electron transfer during the CE between o₂-containing liquid and ferrimagnetic solid. During the experiments, both the Fe₃O₄ and CoFe₂O₄ ferrimagnetic samples received more positive charges in the CE with the DI water when exposed to a magnetic field, suggesting that the magnetic field facilitated electron transfer from the ferrimagnetic sample surface to the o₂-containing DI water. In this process, the o₂ molecules acted as acceptors for electrons from the solid sample surfaces during CE. When the o₂ molecule in the water receive the electron , they is take will further take proton from the H₂O molecules to produce oH⁻, which was similar to the process that occurred during oRR⁴⁴. According to the Pauli exclusion principle and spin conservation principle, the spin of the transferred electrons must be antiparallel to the unpaired electrons belonging to o₂. The entire o₂-containing liquid–solid CE process during the experiments can be expressed as follows.

$ $ \uparrow is \uparrow+\downarrow { { { \mathrm{o}}}}={{{\mathrm{o } } } } \uparrow+\downarrow { e}^{-}(donate \ , by \ , F{e}_{3}{O}_{4},CoF{e}_{2}{O}_{4})+{H}_{2}O \iff \uparrow { { { \mathrm{O}}}}= { { { \mathrm{O } } } } \uparrow \downarrow { { { \mathrm{H}}}}+{{{\mathrm{O}}}}{{{\mathrm{H}}}}^{-}$$

(1)

$ $ \uparrow { { { \mathrm{o}}}}={{{\mathrm{o } } } } \uparrow \downarrow { { { \mathrm{h}}}}+\downarrow { e}^{-}(donate \ , by \ , f{e}_{3}{o}_{4},cof{e}_{2}{o}_{4})+{h}_{2}o \iff { { { \mathrm{H } } } } \downarrow \uparrow { { { \mathrm{o}}}}={{{\mathrm{o } } } } \uparrow \downarrow{{{\mathrm{H}}}}+{{{\mathrm{O}}}}{{{\mathrm{H}}}}^{-}$$

(2)

During the first step (Eq. 1), the o₂ molecules directly received electrons from the ferrimagnetic samples to produce HO₂. Take Fe₃O₄ as an example, the 3d⁶ electron belonging to the Fe^{2 +} in Fe₃O₄, which is is is delocalized and unpaired⁴⁵, and the o₂ molecule with two unpaired electrons were considered a triplet-radical pair (Supplementary Fig. 6)^46,47. The HO₂ contained an unpaired electron and was in a doublet state, requiring the spin configuration of the o₂−3d⁶ electron triplet-radical pair to be converted to a doublet before electron transfer. However, the spin conversion of the o₂−3d⁶ electron triplet-radical pair was not magnetic field sensitive due to the large zero-field splitting (ZFS) in the o₂ molecule ( the calculation detail are show in Supplementary Note   2 ) .

During the second step (Eq. 2), the HO₂ with one unpaired electron and a 3d⁶ electron is belonging belong to Fe₃O₄ was consider a radical pair [ HO₂•   •e⁻], in which the HO₂ molecule was a free radical without ZFS. A problem is that the spin of the 3d⁶ electron in Fe₃O₄ was fixed by the exchange interactions, destroying the sensitivity of the radical pair to the magnetic field (Supplementary Note 3). However, a 0.5 V AC bias was applied to the liquid and solid electrodes during the KPFM measurements, resulting in the fluctuation of the 3d⁶ electrons at the interface. This fluctuation detached the 3d⁶ electron from the solid surface , which were dissolve as water cluster . This is rendered render the electron transfer magnetic field sensitive , and the fluctuation should be highly associate with the liquid – solid conductivity . The DI water conductivity surpass that of the organic solution , the electron fluctuation is be induce by the AC bias at the DI water and solid interface should be strong than that at the organic solution and solid interface . Therefore , the ce between the DI water and Fe₃O₄ was more sensitive to magnetic fields than that of organic solution and Fe₃O₄ in our experiments. At a 0.5 T magnetic field, the electron transfer at the interface consisting of the DI water with a 2.5 mg/L o₂ concentration and Fe₃O₄ increase up to 65   μC m⁻², while that comprising DMSO with a 5.5 mg/L o₂ concentration and Fe₃O₄ only increased up to 30 μC m⁻², though the former has a lower o₂ concentration (Fig. 2). In order to further verify the importance of [HO₂•   •e⁻] radical pair in the magnetic field – induce electron transfer , the superoxide dismutase ( sod , bovine , Sigma – Aldrich )⁴⁸, a scavenger of \({O}_{2}^{-}\), was add to the DI water to scavenge\({O}_{2}^{-}\), prevent the formation of HO₂ and further reducing the number of [HO₂•   •e⁻] radical pair. With the increase of the SOD concentration, the number of [HO₂•   •e⁻] radical pairs decreases, and it is shown that the magnetic field sensitivity of the electron transfer between water and ferrimagnetic samples becomes weaker, which proves that the [HO₂•   •e⁻] radical pairs play an important role in magnetic field-induced electron transfer at water and ferrimagnetic solid interface (Supplementary Fig. 7).

As shown in Fig. 4a, the magnetic domains of the ferrimagnetic samples were in a disordered state without a magnetic field, there is no magnetic field-induced spin conversion of the [HO₂•   •e⁻] pair occurs. Therefore, the electrons can be transferred from the Fe₃O₄ surface to the o₂ or HO₂ molecules is happened only when the spin is happened of the unpaired electron belong to these molecule happen to be antiparallel to the spin of the 3d⁶ electron . As show in Fig .   4b , the magnetic domain of the ferrimagnetic sample were align with the apply magnetic field , increase its electrical conductivity . When the AC bias was apply in this case , some 3d⁶ electrons of Fe₃O₄ escape from the Fe₃O₄ surface, becoming free dissolved electrons for half the cycle time of the AC bias. Both the HO₂ molecule and 3d⁶ electrons were considered free radicals during this time. Moreover, the T-S spin conversion of the radical pair [HO₂•   •e⁻] was trigger by the external magnetic field at a\(\triangle g{\mu }_{B}B{\hbar }^{-1}\) conversion rate (\(\triangle g\ ) signified the difference between the \(g\) factors of the HO₂ and H₂o cluster ,\({\mu }_{B}\) denoted the Bohr magneton, \(\hbar \) is the reduced Planck’s constant and \(B\) represented the external magnetic field), as described in the RPM (Fig. 4c), and the calculations are shown in Supplementary Note 4^21,22,23. The vector representation of the T-S conversion of the [HO₂•   •e⁻] radical pair is provided in Fig. 4d. The [HO₂•   •e⁻] pair displayed a faster spin conversion rate at a higher magnetic field and a higher electron transfer probability at the o₂-containing DI water and ferrimagnetic sample interface, increasing the electron transfer in the presence of a magnetic field.

Fig. 4: The spin-selected electron transfer model for liquid–solid CE.

The spin-selected electron transfer at the o₂-containing liquid and ferrimagnet interface (a) without a magnetic field and (b) with a magnetic field. c The magnetic field-induced T-S spin conversion of the [HO₂•   •e⁻] radical pair. d The vector representation of the T₀-S conversion of the [HO₂•   •e⁻] radical pair. B denotes the magnetic field strength.

Both the temperature effect and cycle test of the magnetic field effect can be explain accord to the propose spin – select CE model . electron transfer is occur could only occur when the T – S spin conversion was complete within the lifetime of the [ HO₂•   •e⁻] pair . The magnetic field is promoted promote the CE between DI water and ferrimagnetic solid by accelerate the spin conversion of the [ HO₂•   •e⁻] pair via the Zeeman interaction . The increasing is intensified of the temperature intensify the thermal motion of the molecule , reduce the lifetime of the [ HO₂•   •e⁻] pair at the interface. Therefore, the spin conversion failed to complete even in the presence of a magnetic field, destroying the magnetic sensitivity of CE. The irreversible magnetic field-induced electron transfer during the cycle test experiments was caused by two reasons. When exposed to a magnetic field, the magnetic domains of the ferrimagnetic samples were aligned, and the holes on the sample surfaces penetrated a certain depth below the surface. When the magnetic field was removed, the less ordered state prevented the holes from returning to the solid surface due to insufficient conductivity. on the other hand, switching the magnetic field off terminated the spin evolution and prevented the back transfer of electrons. Therefore, the effect of the magnetic field on liquid–solid CE is irreversible.

The magnetic field-controlled CE between the o₂-containing liquid and the Fe₃O₄ and CoFe₂O₄ samples demonstrated the presence of electron transfer during liquid–solid CE, supporting the “two-step” model for the formation of the hybrid EDL⁴⁹. Moreover, it is an interesting observation that the surface of ferrimagnetic samples donated electrons in the experiments, since that the surface of most solids is tend to be negatively charged in contact with water. This supports the spin electron transfer at liquid–solid interface. The electrons in ferrimagnetic materials spontaneously polarize due to exchange interaction, forming magnetic domains at the microscale. In these microdomains, the electron spins are ordered, while the electron spins in water are disordered. The transfer of electrons from the water side to the ferrimagnetic solid side represents a process from a spin disorder state to a spin order state, which is difficult to occur from a thermodynamic perspective. However, the electron transfer from the ferrimagnetic solid side to the water side represents a process from a spin order state to a spin disorder state, which is consistent with the principle of entropy increase. Consequently, water tends to gain electrons when in contact with a ferrimagnetic solid, while the ferrimagnetic solid tends to lose electrons and receive positive charges. The results also indicated that liquid–solid CE was fundamentally a spin-selected chemical reaction, in which electron transfer occurred as the first step. Unlike traditional chemical reactions, the transferred electrons or holes usually accumulate on the solid surface, reaching saturation during liquid–solid CE, consequently preventing further electron transfer, such as a chemical reaction with “negative feedback”. This accumulation attracted opposite ions in the liquid to form an EDL. Since the electron transfer during liquid–solid CE was spin-selected, the density of the accumulated electrons or holes on the solid surface was regulated by the magnetic field, further controlling the EDL structure at the liquid–solid interface. This provides an approach for controlling chemical reactions in EDL-related areas, such as mechanochemistry, electrocatalysis, electrochemical storage, and electrophoresis.