Arxiv:1904.05774V5 [Physics.Class-Ph] 7 Jan 2021

The Electrodynamics of Free and Bound Charge Electricity Generators using Impressed Sources and the Modiﬁcation to Maxwell’s Equations

Michael E. Tobar,1, ∗ Ben T. McAllister,1 and Maxim Goryachev1 1ARC Centre of Excellence for Engineered Quantum Systems and ARC Centre of Excellence for Dark Matter Particle Physics, Department of Physics, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. (Dated: January 8, 2021) Electric generators convert external energy, such as mechanical, thermal, nuclear, chemical and so forth, into electricity and are the foundation of power station and energy harvesting opera- tion. Inevitably, the external source supplies a force per unit charge (commonly referred to as an impressed electric field) to free or bound charge, which produces AC electricity. In general, the external impressed force acts outside Maxwell’s equations and supplies a non-conservative electric action generating an oscillating electrodynamic degree of freedom. In this work we analyze the electrodynamics of ideal free and bound charge electricity generators by introducing a time dependent permanent polarization, which exists without any applied electric field, necessarily modifying the constitutive relations and essential to oscillate free or bound charge in a lossless way. For both cases, we show that Maxwell’s equations, and in particular Faraday’s law are modified, along with the required boundary conditions through the addition of an effective impressed magnetic current boundary source and the impressed electric field, related via the left hand rule. For the free charge case, we highlight the example of an electromagnetic generator based on Lorentz force, where the impressed force per unit charge that polarizes the conductor, comes from mechanical motion of free electrons due to the impressed velocity of the conductor relative to a stationary DC magnetic field. In contrast, the bound charge generator is simply an idealized permanently polarized bar electret, where the general case of a time dependent polarized electret is the underlying principle behind piezoelectric nano-generators. In the open circuit state, both bound and free charge electricity generators are equivalent to idealized Hertzian dipoles, with the open circuit voltage equal to the induced electromotive force (emf). Analyzing the short circuit responses, we show that the bound charge electricity generator has a capacitive source impedance. In contrast, we show for the ideal free charge AC electricity generator, the back emf from the inductance of the loop that defines the short circuit, directly cancels the source emf, so the voltage across the inductor is solely determined by the magnetic current boundary source. Thus, we determine the magnetic current boundary source best describes the output voltage of an AC generator, rather than the electric field.

I. INTRODUCTION or MEMS device [2, 3], which can also be configured as an AC generator capable of generating radio frequencies of The most common form of electricity generation con- 60 − 260MHz [4]. A more recent type of generator is the verts motive (or mechanical) energy into photonic or elec- nanogenerator, which uses special materials that convert tromagnetic energy described by Faraday’s law and is an external strain or motion to a time-dependent elec- known as an alternating current (AC) induction genera- tric polarization of a material independent of an applied tor. In general, the creation of photons or electricity via electric field[5, 6]. Such a device can convert mechanical Faraday’s law must come from another external energy energy to electricity via a piezoelectric or triboelectric source to drive the turbines into motion. The conserva- effect [7–11], or temperature variations to electricity via tion of energy is a fundamental law of physics, but only a pyroelectric effect [12–18]. applies to an isolated system. This law means that en- When considering the generation of electricity only ergy cannot be created or destroyed, but only transferred from the view point of the created electromagnetic de- from one form to another. For example, a nuclear reac- gree of freedom, the creation of electromagnetic energy tion where mass is converted into other forms of energy is a non-conservative process. Thus, when we consider through E = mc2. In this case, if the other form of en- arXiv:1904.05774v5 [physics.class-ph] 7 Jan 2021 the electrodynamics in isolation to the whole system, the ergy is electricity, then we can call this device a nuclear standard Maxwell’s equations must be made more gen- generator or battery. For example, a nuclear battery uses eral to take into account the non-conservative processes. the energy from the decay of a radioactive isotope to gen- This requires the ability to add the impressed forces into erate electricity and can produces large DC electric fields Maxwell’s equations, and we show in this work, that this and voltages of up to 10-100 kV [1]. The modern form of can be achieved by generalising the constitutive relations. the nuclear battery is a micro-electromechanical system The modification of the constitutive relations inevitably leads to a modification of Maxwell’s equations, in a similar way to what happens when we consider the difference between Maxwell’s equations in Matter and in Vacuum. ∗ [email protected] For example, in a dielectric medium, the applied electric 2

field polarizes the material, which causes an opposing in- µr = 1+χm0. The free and bound currents and charges in ternal electric force, which reduces the electric field in Maxwell’s equations given above are not source terms for the medium, which modifies Gauss’ Law. the fields and do not add energy into the system. They The standard Maxwell’s equation in differential form either propagate without loss due to the interaction with and in dielectric and magnetic media are in general given the electromagnetic fields, or can describe a dissipative by (SI units), (or resistive) system where electromagnetic energy is lost, usually by conversion to heat (in this case the electric and

∇~ · D~ = ρf , (1) magnetic ﬁeld phasors can become complex).

∂D~ ∇~ × H~ − = J~ , (2) ∂t f The non-electric energy sources (for example, nuclear energy as discussed previously) capable of transmitting energy and hence a force to electric charges are commonly ~ ~ ∇ · B = 0, (3) referred to as “impressed” sources of the field. Theoreti- cally, they can be represented either as an ideal current or voltage generator of electronic network theory [19]. Thus ∂B~ ∇~ × E~ + = 0, (4) in electrodynamics, an impressed electric current needs ∂t to be added as a current source and hence will mod- where ify Ampere’s law. In contrast, we show an impressed electric voltage needs to be added as a magnetic cur- −1 rent source, and hence will modify Faraday’s law. The D~ = 0E~ + P~ H~ = µ B~ − M,~ (5) 0 later does not mean that magnetic monopole particles and exists, but is a consistent way to model boundary value problems when considering the electrodynamics of a non- ρb = −∇ · P~ J~b = ∇ × M,~ (6) conservative electricity generator [20–22]. This technique is more generally known as the Compensation Theorem Here E~ is the electric field intensity, H~ is the magnetic [19, 23, 24]. We also note the impressed sources are not field intensity, D~ is the electric flux density, B~ is the mag- influenced by Maxwell’s equations because they represent creation of electromagnetic energy from an exter- netic flux density, J~f is the free electric current density, nal source. Two-potential theory is summarized in the J~ is the bound electric current density, ρ is the free elec- b f appendix, which is a common way to include the non- tric charge density, ρ is the bound electric charge density b conservative impressed terms in antenna theory. and 0 and µ0 are the permittivity and permeability of free space. When an electric field is applied to a dielectric material its molecules respond by forming microscopic electric dipoles, which causes a distribution of bound charge In this paper we analyse the electrodynamics of free and bound charge electricity generators using impressed dependent on the applied field E~ , which acts to reduce sources. Importantly, we have shown that the impressed the electric field in the dielectric. For a dielectric mate- sources modify the constitutive relations, which essen- rial the macroscopic polarization vector, P~ is related to tially are the same as the force balance equations be- the average microscopic bound charge density by eqn. tween the external energy source and the electrodynamic (6). In a similar way when a B~ -field is supplied to generated degree of freedom. We show that this mod- a magnetic material, magnetic moments of the atoms ~ ifies Maxwell’s equations with the addition of a free or align to cause a macroscopic magnetization M related bound charge macroscopic polarization, created without to the bound current given in eqn. (6). Thus in matter, an applied electric field. Because the vector curl of this Maxwell’s equations can be fully specified in terms of the polarization is non-zero, we have shown that Faraday’s ~ ~ ~ ~ E and B fields along with the auxiliary fields D and H law must be modified. We then apply this technique to given by eqns. (4), combined with the constitutive re- explain the physics of some free charge and bound charge lations given by eqns. (5). However, in order to fully electricity generators. In particular we find that the def- apply these equations the general relationships between inition of the voltage, Ve, of the electricity generator is the fields in eqns. (5) must be further specified, which i i best described by, Ve = −Im, where Im is the total effec- depend on the material properties. In general this can be tive impressed magnetic current at or near the boundary. quite complex, as materials may be anisotropic, magneto- This approach is similar to the approach recently under- electric, piezoelectric, and so forth. The simplest form taken to explain axion modifications to electrodynamics, of the constitutive relations represents isotropic and lin- where the axion mixes with a photon to acts as the exter- ~ ~ ~ −1 −1 ~ ear materials such that D = 0rE and H = µ0 µr B, nal impressed force, which converts axions into electricity where P~ = 0χe0E~ so r = 1 + χe0 and M~ = χm0H~ so [25–27]. 3

~ i The effective impressed field, Ee, is confined to the voltage source and does not exist outside it.

In a DC battery where external energy is converted to electromagnetic energy [29–31], the impressed electric Figure 1: Illustration of the electric field generated in a field allows the electrons to move in the opposite direc- free-charge ideal DC voltage source from an external tion to the electric field created by the electrons them- energy source. The associated delivered external force selves, even though the internal resistance inside an ideal ~ per unit charge, fS, supplies the energy to seperate (and free charge voltage source is near zero, with (ignoring hence polarize) impressed free surface charges, σi at the ~ ~ i ~ e fringing) E ≈ −Ee and thus ET ≈ 0 (these values de- axial boundaries. This is equivalent to an impressed pend highly on aspect ratio and hence the fringing field). ~ i ~ ~ i electric field, Ee = fS, or polarization, Pe , as given by An example of such a voltage source is illustrated in equations (7) and (8), and related to a magnetic current Fig.1. Given that the impressed force per unit charge at the radial boundary, I , by the left hand rule. ~ i menc Ee is non-conservative, there is an effective impressed The non-conservative nature means an electromotive magnetic current boundary source linked by the closed ~ i ~ force as calculate in eqn. (11), E = −Imenc = P Ee · dl, path, as shown in Harrington and Balanis [20, 21], which is generated, resulting in a voltage output that¸ can means for the DC case, drive an electric circuit. The separated free charges ~ ~ ~ i ~i ~ ~ ~ ~ ~i then generate a conservative dipole electric field, E. ∇ × Ee = −Jm, ∇ × E = 0, ∇ × ET = −Jm. (9) Then by Stokes’ theorem we can calculate the emf of the voltage generator, E, by,

II. ELECTRODYNAMICS OF THE E = −Ii = E~ i · d~l = E~ · d~l, (10) menc e T GENERATION OF ELECTRICITY FROM FREE ˛P ˛P CHARGE where the path encloses the effective impressed magnetic current at the radial boundary, given by, To analyse a free charge voltage generator at DC, or in the quasi-static limit, we can start with the equations Ii = J~i · d~a. (11) menc m given in advanced electrodynamics text books such as ˆS Griffiths [28], where he shows that the total force per unit Since E~ i will apply a force per unit charge to seperate charge, f~ involved in a free charge DC voltage source is e free charge in the system, an impressed free charge dis- given by, i tribution, ρe will be created so that, ~ ~ ~ f = E + fS. (7) ~ i ~ i i 0∇ · Ee = ∇ · Pe = −ρe, (12) Here f~ is the force per unit charge, which supplies the S effectively polarizing the free charge (creating a dipole), energy to seperate the charges and supply an electromo- and allowing the definition of a permanent free charge tive force (emf) from an external energy source. Follow- ~ i ~ i ~ polarization of, Pe = 0Ee. For the situation in Fig.1, ing this a resulting electric field, E, is produced by the the impressed free charges occur as surface charge at the separated charges. Harrington [20] presents essentially ends of the DC voltage source, given by, the same equation as Griffiths [28] for a general AC generator, but using different terminology. Harrington con- σ = P~ i.nˆ = ±σi . (13) siders the electric field more generally such that a total ± e e ~ ~ ~ field, ET ≡ f, which is consistent with the ET in the According to this definition, the ideal DC voltage source ~ i ~ appendix. Harrington also defines Ee ≡ fS as the source will only have impressed free charge, and thus 0∇ · E~ = i impressed electric field. Here to be consistent with Grif- ρe, as a result. The net result is the system has both fith [28] and the left hand rule for the relation between an electric vector and scalar potential and outside the ~ i magnetic current and electric field, we have defined Ee voltage source the electric field resembles a capacitor like in the opposite direction to Harrington, so that dipole even though it is modelled as a perfect conductor. The next step is to expand this technique to an AC ~ ~ ~ i ET = E + Ee, (8) electricity generator. 4

A. Time-Dependent Free Charge Electricity may be written as, Generation Qi Qi ~ eenc ~ i eenc E · d~a = and Ee · d~a = − , (23) The DC system described previously has no magnetic ‹S 0 ‹S 0 field component in the system and adding time dependence will potentially induce a magnetic field. First, ~ ~ d ~ ~ i we note that the impressed magnetic current boundary B · dl − µ00 (E + Ee) · d~a = 0 (24) source is divergenceless. This means from two potential ˛P dt ˆS i ~ ~ theory in the appendix, ρm = 0 and ∇ · B = 0. Follow- ing this we can implement the quasi-static approximation ~ and calculate the B~ -field from the modified Ampere’s B · d~a = 0, (25) ‹S law. To calculate this, we combine with the continuity ~i equation in eqn. (107) for the impressed free current, Je, d with eqn. (12) to obtain, E~ ·d~l+ B~ ·d~a = 0 and E~ i ·d~l = −Ii (26) e menc ˛P dt ˆS ˛P ∂P~ i ∂E~ i J~i = e = e . (14) e ∂t 0 ∂t We can also write eqn. (26) as,

Assuming a lossless system then any other free current ~ d i E~T · dl + B~ · d~a = −I (27) ~i menc besides the impressed current, Je, must be divergence ˛P dt ˆS free, so ∇ · J~f = 0 and ρf = 0 in eqn. (99) and (100) in the appendix, then in the time varying case, the modified and (23) as ~ Ampere’s law is given by, ∇~ × B~ − µ ∂ET = J~ , with 0 0 ∂t f ~ ~ ~ ET · d~a = 0, (28) a modified Gauss’ law of ∇ · ET = 0. The next step of ‹S the quasi-static approximation is to calculate the created E~ -field from the calculated B~ -field from Faraday’s law, In general from a circuit theory perspective, the struc- ~ tures under consideration are considered as electrically ∇ × E~ + ∂B = 0, then combining with eqn. (9), the ∂t small (quasi-static limit), no time delay exists between ~ ~ ∂B~ ~i modified Faraday’s law becomes, ∇ × ET + ∂t = −Jm, sources and the rest of the circuit and the only loss occurs ∂B~ through dissipation. From an antenna theory perspective here the ∂t term can be considered as a displacement magnetic current. The only part of Maxwell’s equations, these assumptions in general can be relaxed as time de- which remains unmodified in this case is the magnetic lays may be important. In this work we just consider the Gauss’ law. quasi-static limit relevant for antenna and circuit theory Thus, Maxwell’s equations in differential form may be in the limit that the structures are small compared to the ~ i ~ wavelength. written in terms of Ee and E, by, ρi ρi ∇~ · E~ = e and ∇~ · E~ i = − e , (15) e B. Ideal Cylindrical AC Free Charge Electricity 0 0 Generator ~ ~ i ~ ~ ∂E ~i ∂Ee ∇ × B − 0µ0 = µ0Je = 0µ0 , (16) ∂t ∂t In the following we will analyse the electrodynamics of ∇~ · B~ = 0, (17) an ideal zero resistance cylindrical AC voltage generator, ~ as shown in Fig.2. We recognise that the net charge ~ ~ ∂B ~ ~ i ~i ∇ × E + = 0 and ∇ × Ee = −Jm. (18) in the system is zero, and the charge will be present as ∂t i a surface charge density, ±σe, at the upper and lower boundaries of the cylinder. Furthermore, there will be a or E~ by T non-familiar boundary condition to be determined at the radial boundary, since E~ i must be contained within the ∇~ · E~T = 0, (19) e cylinder. ∂E~ ∇~ × B~ − µ T = 0, (20) For an ideal source in the quasi-static limit, we ignore 0 0 ∂t any source impedance, including inductance or capaci- ∇~ · B~ = 0, (21) tance, so the AC emf is similar to the DC case (eqn. 10), given by, ∂B~ ∇~ × E~ + = −J~i . (22) T ∂t m E = −Ii = E~ i · d~l, (29) menc e ˛P Note, we set J~f = 0, for the open circuit no-load case. Following this, the integral forms of Maxwell’s equations and the surface charge on the end faces caused by the 5

From the integral equations (23)→(26) it is straightforward to derive the boundary conditions. Here subscript “in” refers to inside the ideal generator and subscript “out” refers to outside the generator, while the subscript “⊥” refers to the perpendicular components of the ﬁeld with respect to a surface and the subscript “k” refers to the parallel components of the ﬁelds with respect to i σe the surface. We also note that E~in = − zˆ, and that 0 there is no electric surface current (only volume current). Thus, the boundary conditions on the axial surfaces for the ideal voltage source become,

⊥ Eout = 0, (33)

~ ⊥ ~ ⊥ Bout = Bin = 0, (34) and the boundary conditions on the radial surface gives,

i ~ k i ~ i σe Eout = −~κm × nˆ = −Ee = − z,ˆ (35) 0

Figure 2: Above, schematic of the cylindrical ideal free ~ k ~ k Bout = Bin = 0, (36) charge AC electric generator of axial length, de and radius re, with associated field and source terms. For Applying the radial boundary condition, eqn.(35), the ideal AC generator of terminal voltage, Ve, the i ~ k σe i gives Eout = − zˆ. This means the electric field just impressed electric current density, J~ , flows up and 0 e outside the generator has maximum value on the radial down the cylindrical axis as the terminal charges, ±σi , e boundary, despite having infinite conductance, similar to oscillate in polarity. Below, the active near field of a fringing in a capacitor. To calculate the magnetic field Hertzian dipole antenna resembles that of a fringing in the system we can start inside the voltage source and field in a capacitor[32], and is equivalent to the above apply the modified Ampere’s law given by eqn. (24). free charge AC electric generator. Effectively we find that the B~ -field caused by the displacement current produced by the time varying E~ -field is suppressed by the B~ -field caused by the time vary im- ~ i ~ i impressed field, Ee, can be calculated to be, pressed Ee-field (or impressed electrical current), so is small within the electricity generator depending on the i ~ i ~ i ~ i i σe = Pe · nˆ = 0Ee · n,ˆ where Pe = σez.ˆ (30) aspect ratio. Thus, with no load circuit attached to the generator, the solution is similar to that of a capacitor Heren ˆ is the normal to the surface, which is equal toz ˆ like dipole, even though we consider an ideal conductiv- on the top surface and -ˆz on the bottom surface. Then, ity. In reality the generator will have an internal resis- from eqn.(10) the emf generated in the quasi-static limit tance, so E~T will be non-zero, and there will be finite is calculated to be, field to match on the axial boundary. Assuming a harmonic surface charge density of the σi d i e e i jω0t E = Eede = = Ve, (31) form, σe = σe0 e , then the terminal voltage may be 0 determined to be, which is similar to a voltage across a capacitor, we label σ d jω0t e0 e jω0t this the free charge terminal voltage, Ve. Ve(t) = Ve0 e = e . (37) 0 Next we consider the magnetic surface current per unit length, which will be apparent at the radial boundary, Likewise, the current inside the voltage source may be (r = re) of the generator. This surface magnetic current determined to be, will determine the parallel boundary condition, and can i jω0t 2 jω0t be calculated from equation (29), using the left hand rule Ie(t) = Ie0 e = jω0σe0 πre e . (38) to be, The ratio of the terminal voltage to internal current can 2 i Ve 1 0πre i σe be calculated to be Ii = jω C where Ceff = d ~κ = − φ.ˆ (32) e 0 eff e m i π 0 is similar to a capacitance and Ve lags Ie by 2 . Note, 6 this is not a capacitance, but just the phase relationship between the current and voltage of the generator, an active component where the impressed force main- tains the charge separation with no power dissipation, so the internal current and terminal voltage must be out π of phase (power factor of 2 for no load). Such current and emf generation with no load will create the near field of a Hertzian dipole antenna, which is reactive and exists as stored energy, acting very much like the field of a dynamically charging and discharging capacitor of C , eff Figure 3: Exaggerated short circuit of an AC with the dipole ‘ends’ acting as plates giving a fringing free-charge voltage generator. Any short circuit will capacitance. To dissipate or radiate power an effective form a loop, which will enclose the impressed magnetic resistive component in the model must be added, which current on the boundary, J~i (t). Initially, the electrons would represent far field radiation or resistive dissipation m ~ i within the generator. will flow in the loop driven by, Ee. However, a back emf will be generated, which directly opposes this field, so ~ ~ i ~ ET = Ee + E = 0, and there is no net force on the electrons. The net result is that the voltage supplied by the generator, Ve, which exists across the effective C. Short circuit response of the ideal free charge inductance, will be determined uniquely by the enclosed dΦB AC generator magnetic current, given by, Ve = −Imenc = dt dIsc dL = L dt + Isc dt , were ΦB is the linked magnetic flux of the loop, and Isc is the short circuit current. The Effectively the calculation in the previous section cal- inductance and current flow will be determined by culated the open circuit voltage of the AC generator. In geometry of the loop by implementing Ampere’s law. reality the generator has a source impedance, which will be designed to have minimal effect, usually represented by a Thevenin or Norton equivalent circuit depending if configured as a voltage or current source. To calculate equation for a voltage across the inductor, where the Norton equivalent impedance, the short circuit cur- dΦB dIsc dL rent needs to be calculated and J~f in eqn. (20) cannot Ve = = L + Isc . (41) be set to zero and has to be reinstated on the right hand dt dt dt side. For the ideal system, assuming a perfect conductor, In this case, the free current flow in the loop is determined the impedance will actually depend on how the voltage by the inductance of the loop, calculable from the imple- terminals are short circuited. Inevitably the short circuit mentation of Ampere’s law with J~f of non-zero value. forms a current loop as highlighted in Fig.3 and thus Even though E~T = 0, the voltage across the short circuit will have an inductance which will limit the current flow is defined uniquely by the magnetic current boundary for an AC signal. At DC the finite conductivity of the source, which best describes the output voltage of an AC generator and the metal in the short circuit, will give generator, rather than the total electric field. an effective resistance, which will limit the current flow. Combined with this inductance, the resistance will define This concept also allows interpretation of Faraday ex- the time constant of the short circuit. periments, which either rely on a Coulomb force from the separation of charges opposing the impressed force (open If we consider an idealized short circuit for the time- circuit case), or a back emf, from a Faraday force caused dependent case, with just the equivalent inductance of by the time rate of change of magnetic flux through a the loop, then as shown in Fig.3, the net force on the closed loop, which also opposes the impressed force (short electrons will be zero, as a back emf will be produced circuit case). ~ ~ i from a Faraday induced electric field of, E = −Ee, so that E~T = 0. Because they have opposing values of curl, Faraday’s law becomes, D. Example: A cylindrical ideal conductor ∂B~ oscillating in a DC magnetic field − J~ = , (39) m ∂t In this example we use the impressed source technique so that the voltage across the loop is given by, to analyse the emf induced in a cylindrical ideal conductor due to an external motional kinetic energy or applied i dΦB Ve = −I = , (40) force in the quasi-static limit. The conductor generates menc dt electricity from its motion due to the impressed Lorentz ~ i where ΦB is the magnetic flux. This is basically the force, Fe , acting on free charge carriers in the bar as 7 shown in Fig.4 and given by,

~ i i ~ ~ i Fe = q~v (t) × BDC = qEe(t). (42) For this case the impressed force per unit charge comes from a driven oscillating mechanical degree of freedom interacting with a DC magnetic ﬁeld (DC photonic degree of freedom). This in turn creates an oscillating AC photonic degree of freedom characterised by an oscillating emf, E(t), and given by,

i i E(t) = Ee(t)de = deBDC v (t) (43) Figure 4: A cylindrical ideal conductor (orange) of i Assuming simple harmonic motion, ~x(t) = x0 sin(ω0t)ˆx, time-dependent velocity, ~v (t), moving in thex ˆ then the impressed velocity vector may be written as, direction within a DC magnetic field of B~ DC orientated in they ˆ direction (black crosses). An emf is generated i ~v (t) = ω0x0 cos(ω0t)ˆx. (44) through the induced Lorentz force per unit charge, F~ i e = E~ i = ~vi(t) × B~ = v (t)B zˆ in thez ˆ direction, Then given that B~ = B yˆ, the impressed Lorentz q e DC 0 DC DC DC which is impressed on the free charges in the conductor. force per unit charge becomes, The separation of charges due to the impressed Lorentz ~ i force creates a reverse electric field, E~ (green), with an Ee(t) = ω0BDC x0 cos(ω0t)ˆz, (45) i effective magnetic current boundary source, ~κm creating an emf of, (magenta).

E(t) = ω0BDC dex0 cos(ω0t). (46) DC conductivity, which will dominate at low frequencies, The Lorentz force also drives the motion of the charges this will be a source of damping as a component of, F~B(t) and hence the impressed internal electrical current within would be in phase with the velocity, which also means a the cylinder of, component of the current and voltage would be in phase. At higher frequencies the resistance will increase due to ~i 2 2 Ie(t) = −ω00πre BDC x0 sin(ω0t)ˆz. (47) the skin effect, and the cylinder will have an inductance. It is not our intention to analyse these effects in this pa- Note the current lags the voltage by π as expected. The 2 per, which will contribute to the source impedance of a oscillating separation of ± charges creates an electric real voltage source or generator. As discussed previously, field, which opposes the impressed force so the net force the cylindrical generator will also emit an E~ (t)-field out- on the oscillating charges is zero, which means there is no side, which will be in the form of the near field of an work done in driving the internal electrical current. This oscillating Hertzian dipole. means that there is only reactive power in the electrical degree of freedom as there is no load to dissipate active power. However, as a consequence of this oscillating III. GENERATION OF ELECTRICITY FROM current in the DC magnetic field, there is an additional BOUND CHARGE Lorentz force acting on the mass of the cylinder, F~B(t), which is given by, A. The DC Electret F~ (t) = d I~i(t) × B~ , (48) B e e DC A DC bound charge voltage source is essentially a due to the mechanical motion. This force is in the same bar electret. The ideal bar electret exhibits a perma- direction, but out of phase with the velocity, given by, nent electrical dipolar field as shown in Fig.5, due to an i impressed macroscopic polarization, Pb , and has over- ~ 2 2 all charge neutrality[33]. Some common ways to impress FB(t) = ω00VeBDC x0 sin(ω0t)ˆx, (49) a polarization and make an electret, is to heat a polar where Ve is the volume of the conductor. In the ideal dielectric material under the influence of a large elec- case there is no work done unless a load is attached to tric field (thermo electret) [34] or through piezoelectric- the generator. Of course the analysis ignores any real ity (piezo electret) [7]. For the former, once cooled and material effects, such as finite conductivity, skin effect, removed from the electric field a net polarization will be kinetic inductance etc., which occur at high frequencies. maintained, while the later is maintained through the application of strain. Other forms of electrets include the Thus, in the real case the oscillating conductor has an magneto-electret and the magneto active electret [35] as effective impedance, including a source resistance due to well as ferroelectric electrets [36–38]. The electret thus 8

a linear magnetic material with a permanent magnetization M~ S. The vector M~ S is in actual fact an impressed source as energy needs to be added by an external force to permanently magnetize the material, and we may iden- ~i ~ ~ tify an impressed bound current, Jb = ∇ × MS so that ~ ~ ~i Ampere’s law becomes ∇ × B = µ0µrJb where the effective impressed bound current at the radial boundary of the bar magnet sources the magnetic field. Thus, the Figure 5: From left to right, 3D sketch of the E~ , P~ i and fields in the bar magnet could also be represented in Fig.5 b with the following substitutions B~ → D~ , H~ → E~ and D~ fields inside and outside a cylindrical bar electret M~ → P~ i. Now we can identify how static Maxwell’s (reproduced from the solution manual of [28]). S b equations need to be generalised to be able to describe Assuming the impressed polarization, P~ i, is constant b a system with a permanent polarization as an impressed within the electret and along the cylindrical z−axis of source, that is to take the curl of eqn. (52) and combine the bar in the positive direction. Note, P~ i is only b with eqn. (51) to give ∇~ × D~ = ∇~ × P~ i. One can iden- defined inside the electret. b tify an impressed magnetic bound (subscript mb) current,

J~mb, boundary source at the radial boundary of the electret, from, becomes a bound charge voltage source and is useful for supplying DC bias reducing the requirement for high ex- i i ∇~ × P~ = −0rJ~ (53) ternal DC voltages [39, 40], and can supply a current b mb and be discharged in a similar way to a battery [41]. In so that a modified Faraday’s law becomes, fact batteries with solid electrolytes are not too dissim- ~ ~ ~i ilar to a DC electret. For example, when a ferroelectric ∇ × D = −0rJmb (left hand rule). (54) material becomes permanently polarized, it undergoes a phase change, which corresponds to the crystal structure There is no free charge in this system so the divergence breaking a certain symmetry under phase transition [42]. of D~ is zero. However, the divergence of E~ will be non Likewise a rechargeable battery with a solid electrolyte zero due to the separation of bound charge, which has undergoes induced symmetry breaking of the electrolyte in general two components, ρ , due to the dielectric structure, such as polyhedron distortions upon charge bχe susceptibility of the material and, ρb , the bound charge and discharge [43]. Some solid batteries even have elec- P i ~ i trolyte of perovskite structure [44, 45], similar to the driven by the AC impressed polarization vector, Pe (t), so structure of common ferroelectric materials. that, The bar electret as shown in Fig.5 is the electrostatic 0∇~ · E~ = ρbχ + ρb , (55) analogue of the bar magnet. Here we assume a lossless e P i dielectric media so ideally there is no free charge or cur- and rent in the system and hence they are set to zero, in this case Maxwell’s equations, (1)→(5), become ∇~ · E~ = −∇~ · P~ i = ρ . (56) 0 r b bP i ∇~ · D~ = 0, (50) In a similar way to the free charge voltage source, the ∇~ × E~ = 0, (51) forces in the bound charge system may be defined using equation (8) [20, 28]. Thus, the total force per unit where charge, E~T acting on the bound charges is given by, ~ ~ ~ ~ i ~ ~ i D = 0E + 0χeE + Pb = 0rE + Pb , (52) ~ ~ ~ i ET = E + Eb, (57) for a linear dielectric material with a permanent polar- Accordingly the impressed electric field acting on bound ~ i ~ ization, Pb , which is independent of the electric field, E. ~ i charge, Eb, may be identified to be related to the perma- ~ i nent source polarization Pb by,

1. Impressing a Source Term into static Maxwell’s ~ i ~ i Eb = Pb /(0r), (58) Equations to Describe an Electret and then if we multiply equation (57) through by the Equations (50)→(52) seem incomplete, as there is a permittivity, 0r, we obtain exactly eqn. (52) (where lack of a source term on the right hand side of the equa- D~ = 0rE~T ), the well known constitutive relation be- tions. For the analogue bar magnet, Maxwell’s equations tween the D~ -field, E~ -field and P~ -field in the electret. give ∇~ · B~ = 0, ∇~ × H~ = 0 and ∇~ × B~ = µ0µr∇~ × M~ S for Thus, equation (57), which balances the forces in the 9 voltage source is essentially on the same footing as a con- impedance and not the emf. However, in a general sys- stitutive relationship between fields given by eqn. (52). tem one might expect a numeric calculation is necessary In this case the modified static Faraday’s law may be if the polarization and electric fields cannot be assumed written as, constant, similar to a bar magnet.

~ ~ ~i ∇ × ET = −Jmb, (59) which is equivalent to eqn. (54). Thus as emphasized in the appendix, the DC voltage source has two components B. AC Generation of Electricity with Time Varying Bound Charge of force per unit charge in the system. The impressed ex- ~ i ternal force per unit charge, Eb, with an electric vector potential, which generates an emf, and applies the force Next, we generalise the concept of the electret to a time ~ ~ i to seperate the ± charges, and the electric field, E, which varying impressed permanent polarization, Pb (t), along is sourced by the separated ± bound charges, which ex- the lines to what was undertaken in the work of Zhong hibits a scalar electric potential. Lin Wang [6]. This system describes an AC electricity generator based on oscillating bound charge. The start- Revisiting the magnetic analogue of the DC electret (a ing point is to consider Maxwell’s equations for a linear ~ ~ ~i permanent magnet), where, ∇ × B = µ0µrJb, introduces isotropic dielectric media with a time varying permanent a bound (or Amperian) current [46]. In this example the polarization and an impressed magnetic current bound- bound current is an effective impressed electrical current, ary source, in a similar way to the static version discussed and there is no real current flow, especially this is high- previously (also see appendix). First we consider the con- lighted by the fact that a permanent magnet is usually tributions to Gauss’ Law for the three fields, D~ (t), E~ (t) not a good conductor. The collective alignment of spins, ~ i which creates this bound current occurs due to an im- and Pb (t), which is modified in the same way as the DC pressed force magnetising the material (in this case a case, magnetomotive force). This bound current is treated ex- ρb ~ ~ ~ ~ P i ~ ~ i actly like a real current when analysing macroscopic elec- ∇ · D = 0, ∇ · E = and ∇ · P = −ρb i . (60) b P trodynamic equations, like in magnetic circuit analysis. 0 r For example, the bound electrical current in a bar mag- Next, Ampere’s law is modified along the lines as first net acts as a source term for the fields of the macroscopic suggested by Wang [6], magnetic dipole. In a similar way, the magnetic current defined and introduced in eqns. (53), (54) and (59) is ∂E~ ∂P~ i ∇~ × B~ − µ − µ b = 0. (61) not a real magnetic current, and of course it cannot be 0 0 r ∂t 0 ∂t as monopoles as far as we know do not exist. This term is an effective magnetic current that defines the boundary The magnetic Gauss’ law remains unmodified, while conditions, and in analogy can be used when analysing Faraday’s law is modified so, macroscopic electrodynamic equations, like in electric cir- ~ cuit analysis. Similarly, the bound magnetic current in 1 ~ ~ ∂B ~i ∇ × D + = −Jmb, (62) a bar electret acts as a source term for the fields of the 0r ∂t macroscopic electric dipole. Another interesting point is that the free charge voltage source discussed previously with can be also thought of as a macroscopic dipole, with a ~ ~ ~ ∂B ~ ~ i ~i free charge polarization, with a similar magnetic current ∇ × E + = 0 and ∇ × Pb = −0rJmb. (63) boundary source. ∂t To calculate the fields in the DC electret, a numer- The last term in equation (62) is the impressed magnetic ical calculation is necessary, similar to that of a cylin- current, which acts as a source of the system, to create drical bar magnet. However, to calculate the circuit an AC voltage output. The modified Maxwell’s equations properties of the voltage source one just need to im- with impressed sources may also be written in terms of ~ ~ ~ i plement the integral form of equation (59), which gives the total force per unit charge, ET = E + Eb, for the E = −Ii = 1 P~ i · d~l independent of the electric time-dependent electret (also see the appendix) as, mb 0r P b field since E~ ·d~l ¸= 0. The other parameter that is nec- P ∇~ · E~ = 0 (64) essary to calculate¸ is the source impedance, which will T just be the capacitance of the electret with the source ∂E~ ∇~ × B~ − µ T = 0 (65) polarization removed. To first order this capacitance 0 0 r ∂t ~ may be calculated assuming a constant E field within ∇~ · B~ = 0 (66) the dielectric, which ignores fringing, and assumes an as- ∂B~ pect ratio of a thin polarized plate. Nevertheless, the ∇~ × E~ + = −J~i , (67) calculation uncertainties in this case is with the source T ∂t mb 10

Here the ∂B~ term in eqn.(62) and (67) can be identified P~ k ∂t E~ k = E~ k , E~ ik = bin = −~κi × n,ˆ (79) as the magnetic displacement current. In this system, the out in bin mb in ~i effective magnetic current source term, Jmb exists on the radial boundary of the electret, and drives the impressed In the following we show how the values of the elec- ~ i tromagnetic fields, output voltage and magnetic current electric filed, Eb by the left hand rule and also sets the boundary condition for the parallel components of the may be calculated with the aid of the constitutive rela- fields on the radial boundary. tions and boundary conditions that define an ideal electret.

C. Boundary Conditions 1. Axial Boundaries The boundary conditions of the ﬁelds on the normal and parallel surfaces of the electret can be calculated Assuming the impressed polarization is constant of the ~ i i i from the integral version of equations (60)→(63), which form Pb = Pb zˆ = 0rEbzˆ, at the top and bottom axial are given by, boundaries, as shown in Fig.6, we can use eqn.(74) to ~ ~ i obtain a relationship between E and Eb, where D~ · d~a = 0, (68) ‹ ~ ~ ~ i S 0Eout · nˆ = 0r(Ein + Eb) · nˆ (80)

i To take the analysis a step further one really needs a nu- Qb E~ · d~a = P i , P~ i · d~a = −Qi , (69) meric solution for the general case, as another equation is b bP i ‹S 0r ‹S needed to solve the problem. In general, E~out is non zero as highlighted in Fig.5, and E~in points in the opposite i i direction to E~out and E~ and in general |E~ | ≥ |E~in|. In ~ d b b B~ · dl − µ0 D~ · d~a = 0 (70) ~ i ˛ dt ˆ our example we have assumed a constant Pb along the P S z-axis, this is actually an approximation for a thin plate, where we ignore fringing, which means E~out ≈ 0 so that ~ ~ i B~ · d~a = 0, (71) Ein ≈ −Eb. This limit gives as a simple way of calculat- ‹S ing the Thevenin equivalent circuit for an electret. Matching this condition gives the following relation between vectors at the axial boundary, 1 ~ ~ d ~ i D · dl + B · d~a = −Imb, (72) 0r ˛P dt ˆS ~ ~ ~ i Eout = 0 and Ein = −Eb. (81) ~ d 1 so that D = 0 above and below the axial boundaries. To E~ ·d~l+ B~ ·d~a = 0, P~ i ·d~l = −Ii (73) calculate the impressed bound surface current, ±σi at ˛ dt ˆ ˛ b mb b P S 0 r P the upper and lower axial end faces respectively, we use From these integral equations it is straightforward i ~ i i to derive the modiﬁed boundary conditions given in σb = Pb · nˆ = 0rEb, (82) (74)→(79). Here subscript “in” refers to inside the bar electret and subscript “out” refers to outside the bar elec- asn ˆ =z ˆ, while on the lower axial boundary the value is tret., negative asn ˆ = −zˆ. The polarization current density may be calculated from the time rate of change of eqn. (82) to be, J~P = ⊥ ⊥ ∂P~ i ∂Ei D = D , (74) b b out in ∂t = 0r ∂t zˆ, and if it has a cross sectional area of Ab the eﬀective polarization current through the voltage source is given by, E⊥ − E⊥ = σ , Ei⊥ = P i⊥ = σ (75) out out in in bP i in bin bin bP i i i i ∂Pb ∂σb ∂Eb IP = Ab = Ab = Ab0r (83) ~ k ~ k ∂t ∂t ∂t Bout = Bin, (76)

⊥ ⊥ 2. Radial Boundaries Bout = Bin. (77)

On the radial boundary we can determine that the im- ~ k ~ k i Dout − Din = −0r~κmb × n,ˆ (78) pressed surface magnetic current density from equation 11

be calculated from,

i ~ i ~ E = −Imb = Eb · dl, (86) ˛P

i ~i where Imb = S Jm · d~a, is the enclosed magnetic current. Deﬁning´ the length of the voltage source as db, the induced emf is given by,

i i σbdb E = Ebdb = (87) 0r

Note, J~D = 0 is an approximation, and the value will depend on the aspect ratio, thus for any finite electret there will be in fact a finite D~ field (and hence finite E~T field) above the electret, as shown in Fig.5. For these Figure 6: Top left and right show respectively the cases a small magnetic field will be generated within the Norton and Thevenin equivalent circuits, for the ideal electret due to the time dependence. AC bar electret of cross sectional area Ab and length db Now assuming the charge density oscillates harmon- i jω0 shown underneath. The associated impressed force per ically such that σb = σb0e , we can calculate the ~ i Thevenin equivalent voltage, where the open circuit volt- unit charge Eb(t) supplies the force to seperate the i age across the electret is equivalent to the emf calculated bound charges σb resulting in an electret with a time ~ i i in eqn. (87) to give, varying permanent polarisation of Pb (t) = Pb (t)ˆz, which can be modelled as an impressed magnetic current σb0db boundary source, ~κi (t). The Thevenin equivalent V = ejω0t (88) mb b circuit is pictorially shown with the open circuit 0 r i voltage, Vb(t) = E(t) = −Imb, and effective source Also by short circuiting the electret, a free charge will impedance, Zb, which is calculated in the text and is oscillate in the short circuit wire equivalent to the po- equivalent to the capacitance of the electret without a larization current, as the E~in field is shorted to zero, the permanent polarization. Norton equivalent current source can be calculated to be,

∂σi I = A b = jω A ejω0t. (89) b b ∂t 0 b (79) to be, Ignoring any small resistive or inductive eﬀects, the σi source impedance may be calculated to be, ~κi = −Ei φˆ = − b φ,ˆ (84) m bin 0r Vb 1 0rAb Zb = = Cb = . (90) and the electric ﬁelds, E~ to be, Ib jω0Cb db

i Note, the impedance can also be calculated by setting ~ ~ i σb Ein = Eout = −Ebzˆ = − zˆ (85) the voltage source to zero (or permanent polarization, 0r i Pb = 0), and calculating the capacitance of the left over This is of similar form to the free charge voltage source, dielectric, which is equivalent to eqn. (90). This system i i σb is a Hertzian dipole with similar characteristics to the with σ ≡ , because both have E~T = 0 above and e r below the axial boundary. This also means that the total free charge system. displacement current, J~D = 0, above and below the axial boundary and hence there will be no B~ ﬁeld within the voltage source. D. Example: A piezoelectric nano generator (PENG)

One way to generate the time varying electret as discussed is through piezoelectricity, which has become an 3. Equivalent Circuit important way to undertake energy harvesting [47]. In this example we consider a direct piezoelectric eﬀect where polarization in certain materials can also be in- To calculate the emf we need to integrate around the duced by mechanical loads as shown in Fig.7. An exter- ∂B~ radial boundary, with ∂t = 0 and therefore the emf may nal impressed load (force, stress or strain) if time har- 12

derived in [6]. This calculation highlights, when the terminals are short circuited, the free current that flows will ∂P i bz be equal to the polarization current, IP = Ab ∂t , driven by the induced emf, E(t) as indicated by the Norton and Thevenin equivalent circuits shown in Fig.6. Note it is generally assumed that the external strains do not sig- nificantly perturb the dimensions[6], so to fist order the source impedance shown in Fig.6 is constant. Also, any inductive or resistive short circuit properties will be small compared to the source capacitance. Typical parameters for an AC electret based on vibra- tional energy include; 1) Based on charged resonantly 2 Figure 7: Exaggerated diagram of an impressed vibrating cantilevers with vibrations of 0.1g(1m/s ), can harvest up to 30µW per gram of mobile mass, with a mechanical load, F~ i(t), distorting a cubic structure source capacitance between 1 − 8pF [52]; 2) A vibration- ~ i (left) to induce an impressed polarization, Pb (t) (right), driven polymer energy harvester[53, 54], obtained output which is the principle of PENG. power as large as 100µW at 30Hz and 0.15g acceleration. 3) The flexible triboelectric generator[10] has attained an electrical output peak voltage of 3.3V and current of monic, will cause a time harmonic variation of impressed 0.6mA with a peak power density of 10.4mW/cm3. For ~ i polarization, Pb (t). Thus, the piezoelectric effect is ex- such devices, output impedances are capacitive and typ- plained through the coupling of electric and elastic phe- ically of the order of 100 MΩ. nomena, and thus in general is a tensorial theory combining continuum mechanics and electrodynamics leading to many complex effects beyond the scope of our discus- IV. CONCLUSION sion, see standard text books such as [47–49]. For the purposes of this work, we assume a basic mathematical We have explored the electrodynamics of bound and formulation[50], where dimensional effects are small and free charge electricity generators and voltage sources. the process is linear and in one dimension along the z- The external input to the system was represented by an axis. For this ideal example the constitutive relations impressed force per unit charge, which converts the exter- are, nal energy into electromagnetic energy and may be considered as a non-conservative electric field vector, or emf P i = d × T = d × c × S = e × S (91) bz pe pe pe pe per unit length, with an electric vector potential. The i source term is necessarily impressed into Maxwell’s equa- Dz = Pb + 0rEz. (92) z tions as an effective magnetic current boundary source, Here, dpe is the piezoelectric strain coefficient, T is the which sources the resulting charge distribution and emf stress to which piezoelectric material is subjected, cpe is produced by the generator, resulting in a modification the elastic constant relating the generated stress, T , and of Faraday’s law, the constitutive relations and hence the applied strain, S,(T = cpe ×S) spe is the compliance Maxwell’s equations. coefficient, which relates the deformation produced by the application of a stress (S = spe × T ), and epe is the piezoelectric stress constant. In this case equation (92) is ACKNOWLEDGEMENTS of the same form as eqn. (52), where the magnitude and time dependence of the electret polarization is dependent This work was funded by the Australian Research on the external mechanical load. Council Centre of Excellence for Engineered Quantum A general theory for potential and fields has been de- Systems, CE170100009 and Centre of Excellence for Dark veloped for PENG in [6, 51], which focuses on the genera- Matter Particle Physics, CE200100008. We also thank ~ ∂Dz ∂Ez Professor David Griffiths for allowing the reproduction tion of the displacement current, JD = ∂t = 0r ∂t + ∂P i of his figures and we thank Professor Ian McArthur for bz . However, in the no-load situation, with a flat as- ∂t his analysis and comments on the manuscript. pect ratio, we have shown in the last section that J~D ≈ 0, P i d i bz b σbdb with the open circuit voltage, Voc = E = = , 0r 0r calculable from the modified Faraday’s law, given by REFERENCES equations (86) and (87), which is consistent with that 13

[1] J. H. Coleman, “Radioisotopic high-potential, low- Materials 27, 2340–2347 (2015). current sources,” Nucleonics 11, 42–25 (1953). [17] J.-H. Lee, K.Y. Lee, M.K. Gupta, T.Y. Kim, D.-Y. Lee, [2] Amit Lal and James Blanchard, “Nuclear batteries the J. Oh, C. Ryu, W.J. Yoo, C.-Y. Kang, S.-J. Yoon, J.-B. daintiest dynamos,” IEEE Spectrum September, 36 – Yoo, and S.-W. Kim, “Highly stretchable piezoelectric- 41 (2004). pyroelectric hybrid nanogenerator,” Advanced Materials [3] Hui Li, Amit Lal, James Blanchard, and Douglass Hen- 26, 765–769 (2014), cited By 290. derson, “Self-reciprocating radioisotope-powered can- [18] T. Park, J. Na, B. Kim, Y. Kim, H. Shin, and E. Kim, tilever,” Journal of Applied Physics 92, 1122–1127 “Photothermally activated pyroelectric polymer films for (2002), https://doi.org/10.1063/1.1479755. harvesting of solar heat with a hybrid energy cell struc- [4] H. Li, A Lal, J. Blanchard, and D. Henderson, “Self- ture,” ACS Nano 9, 11830–11839 (2015), cited By 46. reciprocating radioisotope-powered cantilever,” in Trans- [19] B. D. Popovic, “Electromagnetic field theorems,” IEE ducers ’01 Eurosensors XV, edited by E. Obermeier Proceedings A - Physical Science, Measurement and In- (2001). strumentation, Management and Education - Reviews [5] Zhong Lin Wang, Tao Jiang, and Liang Xu, “Toward the 128, 47–63 (1981). blue energy dream by triboelectric nanogenerator net- [20] Roger E. Harrington, Introduction to Electromagnetic works,” Nano Energy 39, 9 – 23 (2017). Engineering, 2nd ed. (Dover Publications, Inc., 31 East [6] Zhong Lin Wang, “On maxwell’s displacement current 2nd Street, Mineola, NY 11501, 2012). for energy and sensors: the origin of nanogenerators,” [21] Constantine A Balanis, Advanced Engineering Electro- Materials Today 20, 74 – 82 (2017). magnetics (John Wiley,, 2012). [7] G. M. Sessler, P. Pondrom, and X. Zhang, “Stacked [22] Edward Conrad Jordan and Kieth G. Balmain, “Electro- and folded piezoelectrets for vibration-based energy magnetic waves and radiating systems,” (Prentice Hall, harvesting,” Phase Transitions 89, 667–677 (2016), Inc., 1968) Chap. 13, 2nd ed. https://doi.org/10.1080/01411594.2016.1202408. [23] G. D.Monteath, “Application of the compensation the- [8] Zhong Lin Wang and Jinhui Song, “Piezoelectric nano- orem to certain radiation and propagation problems,” generators based on zinc oxide nanowire arrays,” Science Proc. IEE 98 Pt. IV, 23–30 (1951). 312, 242–246 (2006). [24] Martin Stumpf, Electromagnetic Reciprocity in Antenna [9] Rusen Yang, Yong Qin, Liming Dai, and Zhong Lin Theory, edited by Tariq Samad (IEEE Press, 2018). Wang, “Power generation with laterally packaged piezo- [25] Michael E. Tobar, Ben T. McAllister, and Maxim Gory- electric fine wires,” Nature Nanotechnology 4, 34–39 achev, “Modified axion electrodynamics as impressed (2009). electromagnetic sources through oscillating background [10] Feng-Ru Fan, Zhong-Qun Tian, and Zhong Lin Wang, polarization and magnetization,” Physics of the Dark “Flexible triboelectric generator,” Nano Energy 1, 328 – Universe 26, 100339 (2019). 334 (2012). [26] Michael E. Tobar, Ben T. McAllister, and Maxim Gory- [11] Zhong Lin Wang, “Triboelectric nanogenerators as achev, “Broadband electrical action sensing techniques new energy technology for self-powered systems with conducting wires for low-mass dark matter axion de- and as active mechanical and chemical sensors,” tection,” Physics of the Dark Universe 30, 100624 (2020). ACS Nano 7, 9533–9557 (2013), pMID: 24079963, [27] ChunJun Cao and Ariel Zhitnitsky, “Axion detection https://doi.org/10.1021/nn404614z. via topological casimir effect,” Phys. Rev. D 96, 015013 [12] Hao Xue, Quan Yang, Dingyi Wang, Weijian Luo, Wen- (2017). qian Wang, Mushun Lin, Dingli Liang, and Qiming Luo, [28] David J Griffiths, Introduction to Electrodynamics, 3rd “A wearable pyroelectric nanogenerator and self-powered ed. (Prentice Hall, Upper Saddle River, New Jersey breathing sensor,” Nano Energy 38, 147 – 154 (2017). 07458, 1999). [13] Y. Yang, W. Guo, K.C. Pradel, G. Zhu, Y. Zhou, [29] Dana Roberts, “How batteries work: A gravitational ana- Y. Zhang, Y. Hu, L. Lin, and Z.L. Wang, “Pyro- log,” American Journal of Physics 51, 829–831 (1983), electric nanogenerators for harvesting thermoelectric en- https://doi.org/10.1119/1.13128. ergy,” Nano Letters 12, 2833–2838 (2012), cited By 384. [30] Ralph Baierlein, “The elusive chemical potential,” [14] Y.J. Ko, D.Y. Kim, S.S. Won, C.W. Ahn, I.W. Kim, American Journal of Physics 69, 423–434 (2001), A.I. Kingon, S.-H. Kim, J.-H. Ko, and J.H. Jung, “Flex- https://doi.org/10.1119/1.1336839. ible pb(zr0.52ti0.48)o3 films for a hybrid piezoelectric- [31] Wayne M. Saslow, “Voltaic cells for physicists: pyroelectric nanogenerator under harsh environments,” Two surface pumps and an internal resistance,” ACS Applied Materials and Interfaces 8, 6504–6511 American Journal of Physics 67, 574–583 (1999), (2016), cited By 32. https://doi.org/10.1119/1.19327. [15] Y. Yang, J.H. Jung, B.K. Yun, F. Zhang, K.C. Pradel, [32] Sean Victor Hum, Ideal (Hertzian) Dipole, Vol. ece422 W. Guo, and Z.L. Wang, “Flexible pyroelectric nanogen- (University of Toronto, 2020). erators using a composite structure of lead-free knbo3 [33] G. M. Sessler, Electrets (Springer-Verlag, New York, nanowires,” Advanced Materials 24, 5357–5362 (2012), Berlin, Heidelberg, 1987). cited By 145. [34] Oleg D. Jefimenko and David K. Walker, “Elec- [16] Y. Zi, L. Lin, J. Wang, S. Wang, J. Chen, X. Fan, trets,” The Physics Teacher 18, 651–659 (1980), P.-K. Yang, F. Yi, and Z.L. Wang, “Triboelectric- https://doi.org/10.1119/1.2340651. pyroelectric-piezoelectric hybrid cell for high-efficiency [35] G J Monkman, D Sindersberger, A Diermeier, and energy-harvesting and self-powered sensing,” Advanced N Prem, “The magnetoactive electret,” Smart Materials 14

and Structures 26, 075010 (2017). [52] S Boisseau, G Despesse, T Ricart, E Defay, and [36] Haruhiko Asanuma, Hiroyuki Oguchi, Motoaki Hara, A Sylvestre, “Cantilever-based electret energy har- Ryo Yoshida, and Hiroki Kuwano, “Ferroelectric dipole vesters,” Smart Materials and Structures 20, 105013 electrets for output power enhancement in electrostatic (2011). vibration energy harvesters,” Applied Physics Letters [53] Y. Suzuki, “Electret based vibration energy harvester for 103, 162901 (2013), https://doi.org/10.1063/1.4824831. sensor network,” in 2015 Transducers - 2015 18th In- [37] Ingrid Graz and Axel Mellinger, “Polymer electrets and ternational Conference on Solid-State Sensors, Actuators ferroelectrets as eaps: Fundamentals,” in Electromechan- and Microsystems (TRANSDUCERS) (2015) pp. 43–46. ically Active Polymers: A Concise Reference, edited [54] Kimiaki Kashiwagi, Kuniko Okano, Tatsuya Miyajima, by Federico Carpi (Springer International Publishing, Yoichi Sera, Noriko Tanabe, Yoshitomi Morizawa, and Cham, 2016) pp. 1–10. Yuji Suzuki, “Nano-cluster-enhanced high-performance [38] Chaoying Wan and Christopher Rhys Bowen, perfluoro-polymer electrets for energy harvesting,” Jour- “Multiscale-structuring of polyvinylidene fluoride nal of Micromechanics and Microengineering 21, 125016 for energy harvesting: the impact of molecular-, micro- (2011). and macro-structure,” J. Mater. Chem. A 5, 3091–3128 [55] A. N. Kudryavtsev and S. I. Trashkeev, “Formalism of (2017). two potentials for the numerical solution of maxwell’s [39] Chikako Sano, Manabu Ataka, Gen Hashiguchi, and equations,” Computational Mathematics and Mathemat- Hiroshi Toshiyoshi, “An electret-augmented low-voltage ical Physics 53, 1653–1663 (2013). mems electrostatic out-of-plane actuator for acoustic [56] N. Cabibbo and E. Ferrari, “Quantum electrodynamics transducer applications,” Micromachines 11, 267 (2020). with dirac monopoles,” Il Nuovo Cimento (1955-1965) [40] C. Jean-Mistral, T. Vu Cong, and A. Sylvestre, 23, 1147–1154 (1962). “Advances for dielectric elastomer generators: Re- [57] Ole Keller, “Electrodynamics with magnetic monopoles: placement of high voltage supply by electret,” Photon wave mechanical theory,” Phys. Rev. A 98, Applied Physics Letters 101, 162901 (2012), 052112 (2018). https://doi.org/10.1063/1.4761949. [58] Andreas Asker, Axion Electrodynamics and Measurable [41] Bernhard Gross and R. J. de Moraes, “Polarization of the Effects in Topological Insulators (Kaerstads University, electret,” The Journal of Chemical Physics 37, 710–713 2018). (1962), https://doi.org/10.1063/1.1733151. [59] Michael E. Tobar, Raymond Y. Chiao, and Maxim Gory- [42] Zijun C. Zhao, Maxim Goryachev, Jerzy Krupka, and achev, “Dual aharanov-bohm berry phase due to the gen- Michael E. Tobar, “Emergence of dielectric anisotropy eration of electricity through permanent bound and free of crystalline strontium titanate due to temperature- charge polarization,” arXiv:2101.00945 [physics.class-ph] dependent phase transitions,” arXiv:2008.07088 [cond- (2021). mat.mtrl-sci] (2020). [43] X.Z. Liu, Z.X. Tang, Q.H. Li, Q.H. Zhang, X.Q. Yu, and L. Gu, “Symmetry-induced emergent electrochemi- cal properties for rechargeable batteries,” Cell Reports V. APPENDIX A: TWO POTENTIAL Physical Science 1, 100066 (2020). FORMULATION [44] Henghui Xu, Po-Hsiu Chien, Jianjian Shi, Yutao Li, Nan Wu, Yuanyue Liu, Yan-Yan Hu, and John B. Goodenough, “High-performance all-solid- The general two potential formulation of impressed state batteries enabled by salt bonding to perovskite current and fields has been discussed in detail in standard in poly(ethylene oxide),” Proceedings of the Na- text books on Electrical Engineering [20–22, 55]. The tional Academy of Sciences 116, 18815–18821 (2019), two potential formulation is used in electrodynamics to https://www.pnas.org/content/116/38/18815.full.pdf. model electricity generation in circuit and antenna the- [45] Shahab Ahmad, Chandramohan George, David J. ory, when there is conversion of external energy into elec- Beesley, Jeremy J. Baumberg, and Michael De Volder, tromagnetic energy through non conservative processes “Photo-rechargeable organo-halide perovskite batteries,” as discussed in the main body of this paper. It has also Nano Letters, Nano Letters 18, 1856–1862 (2018). been used to describe duality in electrodynamics and ax- [46] C Birch, “The amperian current model of magnetisation ion electrodynamics[25, 56–58], and was recently applied and the prolate spheroid,” European Journal of Physics to electricity generation [59]. 6, 180–182 (1985). [47] Alper Erturk and Daniel J. Inman, PIEZOELECTRIC Using superposition, we can consider the electric and ENERGY HARVESTING (John Wiley and Sons, Ltd, magnetic current sources separately. So setting the mag- 2011). netic sources to zero, the electric and magnetic fields may [48] T Ikeda, Fundamentals of Piezoelectricity (Oxford Uni- be written in terms of the magnetic vector potential, A~, versity Press, New York, 1996). and the electric scalar potential, φ, [49] Jiashi Yang, An Introduction to the Theory of Piezoelec- tricity (Springer Nature Switzerland, 2018). E~ = −∇φ − ∂A~ A ∂t (93) [50] R.S. Dahiya and M. Valle, Robotic Tactile Sensing ~ ~ (Springer Science+Business Media Dordrecht, 2013). BA = ∇ × A. [51] Zhong Lin Wang, “On the first principle theory of nanogenerators from maxwell’s equations,” Nano Energy 68, Then by setting the electric sources to zero the electric 104272 (2020). and magnetic fields may be written in terms of the electric vector potential, C~ , and the magnetic scalar poten- 15 tial, φm, arate Maxwell’s equations into two parts given by,

~ 1 ~ ~ ~ i EC = − ∇ × C 0∇ · EA = ρe + ρf , (99) 0 (94) ~ ∂C~ BC = −µ0∇φm − µ0 ∂E~ ∂t ∇~ × B~ − µ A = µ (J~i + J~ ), (100) A 0 0 ∂t 0 e f The total electric and magnetic fields may be calcu- ~ ~ lated using the principle of superposition and are given ∇ · BA = 0, (101) by [20, 21]; ∂B~ ∇~ × E~ + A = 0, (102) A ∂t ~ ∂A 1 for the electric sources, and E~T ot = E~A + E~C = −∇φ − − ∇ × C~ (95) ∂t 0 ∇~ · E~C = 0, (103) ~ ∂C~ ~ ~ ∂EC B~ = B~ + B~ = −µ ∇φ − µ + ∇ × A.~ (96) ∇ × BC − 0µ0 = 0, (104) T ot A C 0 m 0 ∂t ∂t i ∇~ · B~C = ρ , (105) Considering the electric field given by equation (95), in m ∂B~ the quasi-static limit we can ignore the time-dependent ∇~ × E~ + C = −J~i , (106) terms and the main source terms are due to the charge C ∂t m distributions defined by the electric charge and the effec- for the magnetic sources. Here, the impressed sources, tive magnetic current, with the electric vector potential i ~i i ~i given by [20, 21], ρe, Je, ρm and Jm, in equations (99)→(106) can only exist due to an impressed external source exciting the sys- ~i 0 0 tem. Also, in general, there may be some free charge and 0 J ~r , t C~ (~r, t) = m d3~r0. (97) current in the system, ρ and J~ respectively. Because 4π ˆ |~r − ~r0| f f Ω magnetic monopoles do not exist, the effective magnetic ~i and the electric scalar potential given by, current, Jm, and the effective magnetic charge terms, i ρm can only exist as impressed sources. Equations (103) 0 0 1 ρ ~r , t 3 0 to (106) are the dual representation of equations (99) φ(~r, t) = 0 d ~r . (98) to (102). Due to the conservation of charge, impressed 4π0 ˆ |~r − ~r | Ω source currents and charges must satisfy the continuity Here C~ and φ at point ~r and time t is calculated from equations, magnetic current and charge distribution at distant posi- i 0 0 0 ∂ρe ~ ~i ∂ρf ~ ~ tion ~r at an earlier time t = t− ~r − ~r /c (known as the = −∇ · Je, and = −∇ · Jf (107) retarded time). The location ~r0 is a source point within ∂t ∂t volume Ω that contains the magnetic current distribution. The integration variable, d3~r0, is a volume element i ∂ρm i around position r0. In a similar way the magnetic po- = −∇~ · J~ , (108) ∂t m tentials may be written in terms of the electric current density and magnetic charge density, however, as such which completes Maxwell’s equations with impressed work focuses on voltage sources, we do not give the for- sources, which describe how electromagnetic energy or mulae but refer the reader to Refs [20, 21]. electricity can be generated from an external impressed The two potential formulation also means we can sep- energy source.