betaFeedback

WorksPhysics and Astronomy

On the Fermi gas, the Sommerfeld fine structure constant, and the electron-electron scattering in conductors

Abstract

Electrical energy is considered as a fundamental parameter for inclusion in Fermi gas theory, in addition to thermal energy. It is argued that electrical energy can move some electrons to above the Fermi Level, providing free charges to carry the electrical current, even at absolute zero temperature. The Drude model, Ohm's law, quantum resistance, and the electrical resistivity due to electron-electron scattering appear naturally as a consequence of the theoretical description, which is based on the quantization of the angular momentum and the Fermi-Dirac distribution, considering total energy as εε = kB_BTT + Φ0Φ_0II. The electrical and magnetic forces acting on an electron are related to the ratio between the Fermi velocity and the speed of light and show that the electron motion is due to helical paths. Considering the center of mass description for the Bohr atom, it was possible to show that the magnetic force is related to the electrical force as FMF_M = (αα/ππ) FEF_E, which demonstrates that the electrons move in helical paths along the orbit. The helical motion naturally provides for quantization of the magnetic flux, the spin of the electron, and the first correction term of the anomalous magnetic moment. Applying the model to describe the electron-electron scattering allows prediction of the behavior of the electrical resistivity of many metals at low temperatures, which is in excellent agreement with empirical observations.

Introduction

Understanding electrical conductivity and how elementary particles interact during electrical transport is one of the goals of solid-state physics. Based on classical mechanics, Drude and Lorentz[1, 2, 3, 4, 5, 6] developed the first successful classical theory based on the fundamental idea in which electrical current is carried by free electrons moving randomly in a lattice of atoms. The Drude model was the first to use the idea of a free gas, non-interacting electrons that would move among the positive ion cores. This classical theory provided verification of the Ohm’s law and the Wiedemann-Franz law[7, 8], but it fails to explain several other experimental observations related to electron-electron, electron-impurity, and electron-phonon interactions. Furthermore, the statistical description of the electrical conductivity cannot take electron-electron scattering into account, which is relevant for describing the electrical transport properties in conventional metals, especially near zero temperature.

Besides the classical Drude model[5, 6], the quantum electron theory was developed by Sommerfeld[5, 6, 9, 10], in which quantum laws were applied to describe the electrical conductivity of metals, using the Pauli exclusion principle as a basis[5, 6]. According to Sommerfeld, conducting electrons move in a constant potential inside the metal and are considered as wave-like particles with quantized energy values.

In addition, the contribution of electron-electron scattering to electrical conductivity in metals has been studied for a long time[11, 12, 13]. In the late 1930s, Landau and Pomerantschusk[11], and Baber[12] identified electron-electron scattering and suggested that the collision between electrons of a metal should lead to T2T^2 electrical resistivity dependence. At sufficiently low temperatures, electrical resistivity (ρ\rho) follows the simple expression ρ∼AT2\rho\sim AT^2, where AA is an intrinsic property of the metal related to the Sommerfeld constant of the electronic heat capacity (γ\gamma) or the Fermi temperature (TFT_{\mathrm{F}}) of the metal[11, 12, 13]. Nowadays, this behavior has been frequently revisited[14, 15, 16, 17, 18, 19, 20, 21, 22] in order to properly describe electron-electron scattering and the electrical behavior of the metals, separating it from the impurities and lattice scattering mechanisms.

On the other hand, quantization of some electrical transport properties, such as quantum resistance[23, 24, 25] and Josephson[26, 27, 28, 29] effect, have received much theoretical interest[30, 31, 32]. The Josephson effect occurs when two superconducting phases are placed in proximity to each other, separated by a thin insulating barrier. The electrical quantization is expressed by the Josephson energy, given by ϵJ∼Φ0IC\epsilon_{\mathrm{J}} \sim\Phi_{0} I_{\mathrm{C}}, where Φ0\Phi_{0} is the magnetic flux quantum and ICI_{\mathrm{C}} is the superconducting electrical current crossing the junction[28]. Regarding the quantization of the electrical resistance, it was first reported by Klaus von Klitzing in 1980, who observed that the Hall effect becomes quantized when electrons confined in a two-dimensional system are subjected to an external magnetic field[23, 24, 25]. In such cases, the electrical quantization is expressed in terms of quantum resistance RQ=h/2e2R_{\mathrm{Q}} = h/2e^2, where hh is the Planck’s constant and ee is the elementary charge.

According to Fermi-Dirac statistics, electrons in a metal are distributed in quantized energy levels, where the topmost energy level occupied by electrons, at absolute zero temperature, is called the Fermi energy level, ϵF\epsilon_{\mathrm{F}} (ref. 5,6). All energy levels below ϵF\epsilon_{\mathrm{F}} are completely filled by electrons and all energy levels above are completely empty at zero temperature. Heating the metal above absolute zero, the electrons are thermally excited and move up to higher energy levels becoming free to conduct electricity. The distribution of electrons among various energy levels is given by the statistical function called Fermi-Dirac distribution, f(ϵ,T)f(\epsilon,T), which determines the probability of occupation to a given energy level under thermal equilibrium. Based on that, it is well-known that the thermal energy kBTk_B T is the important parameter which controls the number of electrons that move above to the Fermi level, leading to electrical conduction. This leads the following question: How do electrons carry electrical current at absolute zero?

To address this question, we propose that electrical energy is a fundamental parameter required to describe electron conduction at absolute zero. A small increase in electrical energy will move some electrons above the ϵF\epsilon_F. The associated motion differs drastically from thermal energy, in which electrons move randomly in the metal, since the electrical energy drives free electrons to move in the direction of the field.

Just to give an idea about the electrical energy-electrical current relation, let us consider the first orbit of the Bohr atom[33, 34], in which the orbital current is given by

I=eτB=evλ(1)I = \frac{e}{\tau_B} = \frac{ev}{\lambda} \tag*{(1)}

where τB\tau_B is the period of one turn around the proton and the orbital velocity is related with the de Broglie wavelength, given by λ=h/mv\lambda= h/mv, which results in a clear relationship between the electrical energy and the kinetic energy in the orbit, such as

ϵ=12mv2=h22mλ2=h2eevλ=Φ0I(2)\epsilon= \frac{1}{2}mv^2 = \frac{h^2}{2m\lambda^2} = \frac{h}{2e}\frac{ev}{\lambda} = \Phi_0 I \tag*{(2)}

where Φ0=h/2e\Phi_0 = h/2e is the magnetic quantum flux, providing a similar energy-current relation described in the Josephson effect[28].

Results and discussion

Fermi-Dirac distribution and electrical energy

As pointed out in the previous session, in this work we consider the possibility of electrical energy given by Φ0I\Phi_0 I being included in the Fermi gas theory and evaluate its consequences for the electrical properties.

In the case of the Fermi gas, this implies a new insight in which

ϵF=eVF=kBTF=12mvF2=Φ0IF(3)\epsilon_F = eV_F = k_B T_F = \frac{1}{2}mv_F^2 = \Phi_0 I_F \tag*{(3)}

where ϵF\epsilon_F is the Fermi energy, TFT_F is the Fermi temperature, vFv_F is the Fermi velocity, and IFI_F is the concept just introduced, hereafter called Fermi current. Just to give an example, Cu, with carrier density n=8.49×1028 m−3n = 8.49 \times10^{28}\ \mathrm{m}^{-3}, has vF=1.57×106 m/sv_F = 1.57 \times10^6\ \mathrm{m/s}, vF=1.57×106 m/sv_F = 1.57 \times10^6\ \mathrm{m/s} and TF=8.12×104 KT_F = 8.12 \times10^4\ \mathrm{K} (ref. [5]), resulting in IF=5.42×10−4 AI_F = 5.42 \times10^{-4}\ \mathrm{A}.

The important question that arises is: what excites electrons above Fermi level if no thermal energy is introduced in the Fermi gas at zero Kelvin? In fact, all the electrons have non-zero energy, but their energy has to do only with the kinetic and potential energy related to the quantized orbits around the nucleus of the atoms. Thus, another source of energy should be introduced in the gas in order to increase the energy of some electrons above the Fermi level at absolute zero temperature to make them free to carry the electrical current. One possibility is introducing electrical energy as mentioned above, although other forms of energy could be added as well, such as magnetic or photonic energy.

In order to see the implications of this possibility, let us start discussing what happens to the Fermi gas statistics, taking into account the electrical energy (eV=Φ0IeV = \Phi_{0}I) in the Fermi-Dirac distribution[5, 6, 35], as follows

f(ϵ,T,I)=1exp⁡(ϵ−ϵFkBT+Φ0I)+1.(4)f(\epsilon,T,I) = \frac{1}{\exp\left(\frac{\epsilon-\epsilon_{F}}{k_{B}T+\Phi_{0}I}\right)+1}. \tag*{(4)}

Figure 1 displays the density of states as a function of the energy at absolute zero temperature for an electrical energy Φ0I\Phi_{0}I, calculated using Equation (4).

Density of states as a function of the energy for a three-dimensional gas

Figure 1. Density of states as a function of the energy for a three-dimensional gas (blue line). The black line shows the density of occupied states at absolute zero temperature. At zero temperature electrons will be able to reach energy above the Fermi level (ϵF\epsilon_{F}) if an electrical energy (Φ0I\Phi_{0}I) is introduced in the Fermi gas. This behavior is represented by the red line, in which the number of electrons excited above the Fermi level (ϵ>ϵF\epsilon> \epsilon_{F}) is proportional to the applied electrical current (the curve was plotted using Φ0I=5%ϵF\Phi_{0}I = 5\% \epsilon_{F}). At T=0T = 0 and I=0I = 0, no electrons are above the Fermi level (black line) and their energies are related only to the kinetic and potential energies due to the quantized orbital motions.

In fact, the distribution is very similar to that considering the thermal energy instead, since Figure 1 was plotted basically exchanging the term kBTk_{\mathrm{B}}T by Φ0I\Phi_0 I in the Fermi-Dirac distribution (see, for instance, reference 5). However, one must keep in mind that the temperature in Figure 1 is strictly zero and the excitation of the electrons to above the Fermi level is only due to the increase of energy of the Fermi gas by the electrical energy. This has important consequences for understanding the statistical description of the electrical properties, as we will see below.

Let us start looking into the energy of the gas excited by electrical current near the Fermi level.

At T=0T = 0 K, one can calculate the energy for the electrons above the Fermi level (see reference 5) as

U=∫ϵF∞(ϵ−ϵF)D(ϵ) dϵ1exp⁡(ϵ−ϵFΦ0I)+1(5)U = \int_{\epsilon_F}^{\infty} (\epsilon- \epsilon_F) D(\epsilon)\,\mathrm{d}\epsilon\frac{1}{\exp\left(\frac{\epsilon- \epsilon_F}{\Phi_0 I}\right) + 1} \tag*{(5)}

or

U(ϵF)=D(ϵF)(Φ0I)2∫0∞xdx1exp⁡(x)+1.(6)U(\epsilon_F) = D(\epsilon_F)(\Phi_0 I)^2 \int_0^{\infty} x\mathrm{d}x \frac{1}{\exp(x) + 1}. \tag*{(6)}

after defining x=(ϵ−ϵF)/Φ0Ix = (\epsilon- \epsilon_F)/\Phi_0 I and considering the limit Φ0I≪ϵF\Phi_0 I \ll\epsilon_F.

Using

D(ϵF)=3N2ϵF(7)D(\epsilon_F) = \frac{3N}{2\epsilon_F} \tag*{(7)}

and

∫0∞xdx1exp⁡(x)+1=π212,(8)\int_0^{\infty} x\mathrm{d}x \frac{1}{\exp(x) + 1} = \frac{\pi^2}{12}, \tag*{(8)}

one can find

U(ϵF)=π28NϵF(Φ0I)2,(9)U(\epsilon_F) = \frac{\pi^2}{8}\frac{N}{\epsilon_F}(\Phi_0 I)^2, \tag*{(9)}

which provides the average of the energy for the electrons near the Fermi level (ϵ≳ϵF\epsilon\gtrsim\epsilon_F) such as

⟨U(ϵF)⟩=U(ϵF)N=π28(Φ0I)2ϵF.(10)\langle U(\epsilon_F) \rangle= \frac{U(\epsilon_F)}{N} = \frac{\pi^2}{8}\frac{(\Phi_0 I)^2}{\epsilon_F}. \tag*{(10)}

Considering

ϵF=12mvF2(11)\epsilon_F = \frac{1}{2}mv_F^2 \tag*{(11)}

and

⟨U⟩=12mvˉ2,(12)\langle U\rangle= \frac{1}{2}m\bar{v}^{2}, \tag*{(12)}

we obtain

vˉ=22πΦ0ImvF,(13)\bar{v} = \frac{\sqrt{2}}{2}\pi\frac{\Phi_{0}I}{mv_{F}}, \tag*{(13)}

where ⟨U⟩\langle U\rangle and vˉ\bar{v} are, respectively, the kinetic energy and the average velocity of the free electrons right above the Fermi level at zero temperature.

Drude model, Ohm’s laws, and quantum resistance

In Figure 2 we define the smallest electrical resistor for a conductor with carrier density nn, which has the cross section A=L2A=L^{2}, whose LL is established in the Fermi gas theory [5, 6, 35] as the side of a cube that contains one electron, calculated by L=n−1/3L=n^{-1/3}, and the mean free path (ll), which is related to the Fermi velocity and the relaxation time between collisions (l=vFτl=v_{F}\tau).

Schematic macroscopic sample of length d and cross section ht containing N charges

Figure 2. Schematic macroscopic sample of length dd and cross section htht containing NN charges. The magnification represents the quantum electrical resistance taken from any conducting material, which is defined with a cross section L2L^{2} and the mean free path ll. Equations (22) and (23) demonstrate that this resistance is an independent material whose value is given by quantum resistance, h/2e2h/2e^{2}. Zero temperature electrical resistivity (ρee\rho_{ee}) can be found connecting the quantum resistance along with Ohm’s law to the quantization of the angular momentum. The combination of parallel and series quantum resistances considering NN charges per volume (V=dhtV=dht) provides the macroscopic electrical resistance (RR) of the conductor.

Considering the quantum resistance of Figure 2, the first implication of Equation (13) is that the average electrical current predicted by the Drude model [5, 6, 36], given by

nevdA=I,(14)nev_{d}A=I, \tag*{(14)}

is recovered after defining the drift velocity as

vd≡2π(3π)1/3vˉ=(3π)1/3Φ0ImvF,(15)v_d \equiv\frac{\sqrt{2}}{\pi}\left(\frac{3}{\pi}\right)^{1/3}\bar{v} = \left(\frac{3}{\pi}\right)^{1/3}\frac{\Phi_0 I}{m v_F}, \tag*{(15)}

remembering that A=L2A = L^2 and

λ=2L=(3π)1/3hmvF,(16)\lambda= 2L = \left(\frac{3}{\pi}\right)^{1/3}\frac{h}{m v_F}, \tag*{(16)}

considering Fermi gas theory[5].

This definition allows us to write the microscopically drift velocity as

vd=LIe(17)v_d = L\frac{I}{e} \tag*{(17)}

where e/Ie/I defines the average time related to the electron crossing the distance LL.

In order to estimate the free number of carriers above the Fermi level due to a small electrical energy (Φ0I≪ϵF\Phi_0 I \ll\epsilon_F) we can write

ΔN=∫ϵF∞D(ϵ) dϵ1exp⁡(ϵ−ϵFΦ0I)+1(18)\Delta N = \int_{\epsilon_F}^{\infty} D(\epsilon)\,d\epsilon\frac{1}{\exp\left(\frac{\epsilon-\epsilon_F}{\Phi_0 I}\right)+1} \tag*{(18)}

providing

ΔN=D(ϵF)Φ0I∫0∞dxexp⁡(x)+1=3N2ϵFΦ0Iln⁡2(19)\Delta N = D(\epsilon_F)\Phi_0 I\int_0^\infty\frac{dx}{\exp(x)+1} = \frac{3N}{2\epsilon_F}\Phi_0 I\ln2 \tag*{(19)}

where x=(ϵ−ϵF)/Φ0Ix=(\epsilon-\epsilon_F)/\Phi_0 I and

∫0∞dx1exp⁡(x)+1=ln⁡2.(20)\int_0^\infty dx\frac{1}{\exp(x)+1}=\ln2. \tag*{(20)}

After some algebraic calculation taking into account Equations (11), (15), and (19) we can show that

ΔNvF=3(π3)1/3ln⁡2 Nvd=2.112 Nvd(21)\Delta N v_F = 3\left(\frac{\pi}{3}\right)^{1/3}\ln2\,N v_d = 2.112\,N v_d \tag*{(21)}

which demonstrates that the number of charges that carry the electrical current near the Fermi level is a very small fraction of the total electrons. For example, a 1 meter long 12-gauge copper wire (cross section = 3.31×10−6 m23.31 \times10^{-6}\ \mathrm{m}^2, n=8.49×1028 m−3n=8.49 \times10^{28}\ \mathrm{m}^{-3}) carrying an electrical current of 1 A has 2.81×10232.81 \times10^{23} electrons available (vd=2.22×10−5 m/sv_d=2.22 \times10^{-5}\ \mathrm{m/s}), but only 8.37×10128.37 \times10^{12} electrons carry the electrical current (vF=1.57×106 m/sv_F=1.57 \times10^6\ \mathrm{m/s}). The result in Equation (21) provides the equivalence of the Drude model with the Fermi gas theory[6, 35].

Furthermore, if a small electrical current is applied to the electrical resistance shown in the magnification of Figure 2, the drop in electric potential between two sides of the block over the electrical current must be connected to Equation (3) through

VI=VFIF=Φ0e=h2e2=RQ=12,909 ohm.(22)\frac{V}{I}=\frac{V_F}{I_F}=\frac{\Phi_0}{e}=\frac{h}{2e^2}=R_Q=12{,}909\ \text{ohm}. \tag*{(22)}

This value represents the quantum electrical resistance, in which the electrical current is carried by a single charge.

Considering the Ohm’s law[5, 6, 8, 37] for this resistance, we can write

ρee1Al=RQ(23)\rho_{ee}\frac{1}{A}l=R_Q \tag*{(23)}

where A=L2A=L^2 and ρee\rho_{ee} is the electrical resistivity of the conductor at T=0T=0, which must be related to the electron-electron scattering, such as will be demonstrated later on. In fact, ρee\rho_{ee} and LL are connected in such a way they provide the quantum resistance value (RQR_Q), no matter the conducting material.

Considering a macroscopic sample with NN charges (NN is comparable to the Avogadro’s number) per volume V=dhtV=dh t (see Figure 2 again), one can compute the parallel and series quantum resistances. Noting that ht=N2/3L2h t=N^{2/3}L^2 and d=N1/3ld=N^{1/3}l, the expected electrical resistance of a typical macroscopic sample is

R=ρeedht.(24)R=\rho_{ee}\frac{d}{h t}. \tag*{(24)}

Note that Ohm’s law and the quantum resistance are connected to the quantization of the angular momentum[34] through

mvB∗rB∗=ℏ=h2π,(25)m v_B^*r_B^*=\hbar=\frac{h}{2\pi}, \tag*{(25)}

where rB∗=λ/2πr_B^*=\lambda/2\pi and vB∗v_B^* are the orbital radius and orbital velocity of the electron in the equivalent Bohr atom, respectively. These are connected to the Fermi velocity

vF=h2πm(3π2n)1/3,(26)v_F=\frac{h}{2\pi m}(3\pi^2n)^{1/3}, \tag*{(26)}

or the wavelength

λ=2L=hmvF(3π)1/3=hmvB∗,(27)\lambda=2L=\frac{h}{m v_F}\left(\frac{3}{\pi}\right)^{1/3}=\frac{h}{m v_B^*}, \tag*{(27)}

providing

vB∗=(π3)1/3vF=(π3)1/3lτ=lτ∗=vF∗.(28)v_B^*=\left(\frac{\pi}{3}\right)^{1/3}v_F=\left(\frac{\pi}{3}\right)^{1/3}\frac{l}{\tau}=\frac{l}{\tau^*}=v_F^*. \tag*{(28)}

where λ\lambda is the de Broglie wavelength of the electron in the Fermi level of a particular conductor and τ∗≡(3/π)1/3τ\tau^*\equiv(3/\pi)^{1/3}\tau. This leads to a complete equivalence between vB∗v_B^* and vF∗v_F^*.

Taking (27) along with (23), noting that vF=l/τv_F=l/\tau, L3=n−1L^3=n^{-1}, and A=L2A=L^2, one can show

ρee=(π/3)1/3mne2τ=mne2τ∗,(29)\rho_{ee}=(\pi/3)^{1/3}\frac{m}{ne^2\tau}=\frac{m}{ne^2\tau^*}, \tag*{(29)}

which is the well-known expression for the electrical resistivity[8, 37]. This demonstrates that the microscopic electrical resistance, schematically shown in the Figure 2, is indeed related to the quantum resistance.

Forces acting on the electron and Sommerfeld fine structure constant

Imagine an electron near the Fermi level moving around the nucleus in the outermost orbit with Fermi velocity (vF∗=vB∗v_F^*=v_B^*), schematically shown in Figure 3(a). A small increase of electric energy will move it to above the Fermi level. The electron’s circular motion will be maintained, but a forward motion due to the force of the electric field will occur, as illustrated in Figure 3(b). As a result, an electrical force and a magnetic force act on the electron and its electrical dissipation is given by the Poynting vector, which is associated to the electrical resistance and Joule effect[8, 37].

Orbital motion and helical motion of an electron

Figure 3. (a) displays the orbital motion of an electron with equivalent Bohr radius rB∗=λ/2πr_B^*=\lambda/2\pi and equivalent Bohr velocity or Fermi velocity vB∗=vF∗v_B^*=v_F^*. (b) shows the helical motion of the electron as a result of the action of both electrical and magnetic forces.

Now, imagine an electron carrying the electrical current through a quantum resistance discussed in the previous session. The electric force (FEF_E) which makes it to move across the resistance must be given by

FE=e24πε0(2rB∗)2=π16e2ε0L2(30)F_E=\frac{e^2}{4\pi\varepsilon_0(2r_B^*)^2}=\frac{\pi}{16}\frac{e^2}{\varepsilon_0L^2} \tag*{(30)}

where 2rB∗2r_{B}^{*} is the diameter of the equivalent Bohr atom or the typical distance between two spheres in the classical collision theory[38], which is related to LL or λ\lambda by

2rB∗=2Lπ=λπ.(31)2r_{B}^{*} = \frac{2L}{\pi} = \frac{\lambda}{\pi}. \tag*{(31)}

As previously mentioned, the dissipation due to the Joule effect in the quantum resistance (RQR_{Q}) must be related to the electric power (WE∝RQIF∗2W_{E} \propto R_{Q}I_{F}^{*2}) (ref. [8, 37]). Thus, one can write the electric power for an electron at the Fermi level interacting with another charge as

WE=FEvF∗=π16e2ε0L2vF∗=Ch2e2IF∗2(32)W_{E} = F_{E}v_{F}^{*} = \frac{\pi}{16}\frac{e^{2}}{\varepsilon_{0}L^{2}}v_{F}^{*} = C\frac{h}{2e^{2}}I_{F}^{*2} \tag*{(32)}

which suggests that an electrical current carried out by one electron crossing the quantum resistance is quantized, where CC is constant for a particular conductor. Based on that and using Equation (27), it is easy to show that

WE=π16e28mε0h2ϵF∗vF∗=παλCϵF∗vF∗,(33)W_{E} = \frac{\pi}{16}\frac{e^{2}8m}{\varepsilon_{0}h^{2}}\epsilon_{F}^{*}v_{F}^{*} = \pi\frac{\alpha}{\lambda_{C}}\epsilon_{F}^{*}v_{F}^{*}, \tag*{(33)}
C=πvBvF∗=πvBvB∗,(34)C = \pi\frac{v_{B}}{v_{F}^{*}} = \pi\frac{v_{B}}{v_{B}^{*}}, \tag*{(34)}

and

FE=πvBvF∗2h2e2IF∗2,(35)F_{E} = \pi\frac{v_{B}}{v_{F}^{*2}}\frac{h}{2e^{2}}I_{F}^{*2}, \tag*{(35)}

after taking into account that

IF∗=ϵF∗Φ0=2eh12mvF∗2=evF∗λ,(36)I_{F}^{*} = \frac{\epsilon_{F}^{*}}{\Phi_{0}} = \frac{2e}{h}\frac{1}{2}mv_{F}^{*2} = \frac{ev_{F}^{*}}{\lambda}, \tag*{(36)}
α=e22ε0hc,(37)\alpha= \frac{e^{2}}{2\varepsilon_{0}hc}, \tag*{(37)}

and

λB=λCα(38)\lambda_{B} = \frac{\lambda_{C}}{\alpha} \tag*{(38)}

where IF∗=(π/3)2/3IFI_{F}^{*} = (\pi/3)^{2/3}I_{F}, ϵF∗=(π/3)2/3ϵF\epsilon_{F}^{*} = (\pi/3)^{2/3}\epsilon_{F}, and vF∗=(π/3)1/3vFv_{F}^{*} = (\pi/3)^{1/3}v_{F} (see Equations (3), (27), and (28)), λC\lambda_{C} is the Compton wavelength, and vBv_{B} and λB\lambda_{B} are the orbital velocity and wavelength of the electron in the first orbit of the Bohr atom, respectively.

Using Equations (28), (35), and (37), and remembering that 1/ε0c=μ0c1/\varepsilon_{0}c = \mu_{0}c, it is possible to show that

1πvB∗cFE=μ04cvF∗IF∗2=μ04lclIF∗2,(39)\frac{1}{\pi}\frac{v_B^*}{c}F_E=\frac{\mu_0}{4}\frac{c}{v_F^*}I_F^{*2}=\frac{\mu_0}{4}\frac{l_c}{l}I_F^{*2}, \tag*{(39)}

considering

lcl=cvF∗,(40)\frac{l_c}{l}=\frac{c}{v_F^*}, \tag*{(40)}

because

lcc=lvF∗=τ∗=(3/π)1/3τ,(41)\frac{l_c}{c}=\frac{l}{v_F^*}=\tau^*=(3/\pi)^{1/3}\tau, \tag*{(41)}

if one assumes lcl_c as the length which an electron would travel at the speed of light (lc=cτ∗l_c=c\tau^*), instead the mean free path ll with the equivalent Fermi velocity (vF∗v_F^*) during the same time between collisions (τ∗\tau^*).

Considering the right side of the equation (39) as the magnetic force (FMF_M), which measures the magnetic interaction between the electrical currents by two charges (see, for instance, references 33 and 34) within the distance of the order of the mean free path, one can write

FM=μ04lclIF∗2,(42)F_M=\frac{\mu_0}{4}\frac{l_c}{l}I_F^{*2}, \tag*{(42)}

providing

FM=1πβFE=1πvB∗cFE,(43)F_M=\frac{1}{\pi}\beta F_E=\frac{1}{\pi}\frac{v_B^*}{c}F_E, \tag*{(43)}

where

β=vB∗c=vF∗c.(44)\beta=\frac{v_B^*}{c}=\frac{v_F^*}{c}. \tag*{(44)}

It is interesting that if vB∗=vBv_B^*=v_B, the previous equation becomes

FM=απFE,(45)F_M=\frac{\alpha}{\pi}F_E, \tag*{(45)}

which shows that the magnetic force is related to the Sommerfeld fine structure constant[39, 40, 41, 42, 43] and is three orders of magnitude smaller than the electrical force for the first orbit of the Bohr atom.

Equation (43) seems to be a fundamental equation and confirms the hypothesis that the two forces act on the electron, one related to the electrical force, which could be either nondissipative (centripetal) or dissipative, and another related to the magnetic interaction, making the electron move in a helical motion instead of in a straightforward way.

In addition, if one considers the electrical and magnetic fluxes due to the electrical (FEF_E) and magnetic forces (FMF_M), respectively, crossing a closed surface area given by A=4π(2rB∗)2A = 4\pi(2r_B^*)^2, it is easy to show from Equations (30), (39), and (42) that

EA=FEe4π(2rB∗)2=eε0=ΦE(46)EA = \frac{F_E}{e}4\pi(2r_B^*)^2 = \frac{e}{\varepsilon_0} = \Phi_E \tag*{(46)}

and

BA=FMevB∗4π(2rB∗)2=μ0πce=4απh2e=4απΦB.(47)BA = \frac{F_M}{ev_B^*}4\pi(2r_B^*)^2 = \frac{\mu_0}{\pi}ce = \frac{4\alpha}{\pi}\frac{h}{2e} = \frac{4\alpha}{\pi}\Phi_B. \tag*{(47)}

After using (43), we can demonstrate that

ΦE=4αcΦB,(48)\Phi_E = 4\alpha c\Phi_B, \tag*{(48)}

which provides a direct connection between the electrical and magnetic fluxes with the definition of the Sommerfeld fine structure constant. Furthermore, Equation (48) is the same equation found recently for the energy distribution of an electromagnetic radiation, using a different approach[44].

Now, the questions that arise are: What is the physical meaning of the length lcl_c and what does it have to do with the electron motion? To address these questions, we will look at the Bohr atom.

Helical motion and consequences to the Bohr atom

In order to have a deeper insight about the lcl_c length and the helical motion, let us look at the Bohr atom carefully.

Remembering that vB/c=αv_B/c = \alpha for the first orbit of the Bohr atom and supposing that λB/lB\lambda_B/l_B has to do with the ratio between the Bohr radius (rB=λB/2πr_B = \lambda_B/2\pi) and a free electron traveling at the speed of light that must be proportional to α\alpha as well, then lB=λB/αl_B = \lambda_B/\alpha or lB=λC/α2l_B = \lambda_C/\alpha^2, since λC/λB=α\lambda_C/\lambda_B = \alpha (ref. [39, 40, 41, 42]).

As such, electrical and magnetic forces act on the electron but in a way the Poynting[8, 37] vector produces the conservation of the energy, since there is no dissipation in the orbits of the electrons in atoms. In terms of dissipation this is very different from the electrons’ motion above the Fermi level described in the previous session.

After spending a long time considering this picture, the only way we have found to take into account both forces simultaneously was to suppose that the electron travels in helical paths around the orbit of the Bohr atom. Figure 4(a) displays a schematic view of this helical motion. It considers N=18,778.865N = 18{,}778.865 (=1/α2= 1/\alpha^2) spirals per turn of the first orbit in the Bohr atom, leading to a type of toroidal solenoid[8, 37].

In fact, the number of spirals in the solenoid has to do with the ratio lB/λB=1/αl_{B}/\lambda_{B} = 1/\alpha or lB/λC=1/α2l_{B}/\lambda_{C} = 1/\alpha^{2}. Thus, considering α=1/137.03599911\alpha= 1/137.03599911 (ref. 43) and λC=2.462×10−12 m\lambda_{C} = 2.462 \times10^{-12}\ \mathrm{m}, one can obtain lB=4.556×10−8 ml_{B} = 4.556 \times10^{-8}\ \mathrm{m} for the first orbit of the Bohr atom.

It is interesting that the distance of one spiral (DD), shown in Figure 4(b), can be calculated using

Helical motion of the electron leading to a toroidal solenoid and paths of spirals

Figure 4. (a) Helical motion of the electron leading to a toroidal solenoid with 18,778.865 (N=1/α2N = 1/\alpha^{2}) spirals. In such a case, the fraction of the λB\lambda_{B} orbit with λC\lambda_{C} length has 137.036 spirals (= 1/α1/\alpha). (b) Path of one spiral with a radius rC≈λC/2πr_{C} \approx\lambda_{C}/2\pi and spacing between adjacent spirals p=λCsen⁡αp = \lambda_{C}\operatorname{sen}\alpha (for viewing purposes, the drawing is enlarged along vBv_{B} direction). Red and blue arrows indicate the velocity of the electron which is equal to the speed of light. The projection of cc onto the x-axis provides the Bohr velocity of the first orbit, i. e. vB=csen⁡α≈cαv_{B} = c\operatorname{sen}\alpha\approx c\alpha (black arrow). (c) Since λCsen⁡α≪(λC/2π)cos⁡α\lambda_{C}\operatorname{sen}\alpha\ll(\lambda_{C}/2\pi)\cos\alpha, the electrical current in the spiral (I=ec/λCI = ec/\lambda_{C}) produces a magnetic field given by Equation (59). The electrical current, the magnetic field, and the cross section of the spiral naturally provide the quantization of the magnetic flux and the electron spin given by equations (61) and (62), respectively. The forwarded helical motions for two electrons in clockwise and counterclockwise directions lead to spin up (B⃗up\vec{B}_{up}) or spin down (B⃗down\vec{B}_{down}), respectively.

D2=4π2rC2+p2(49)D^{2} = 4\pi^{2}r_{C}^{2} + p^{2} \tag*{(49)}

where rC=λC/2πcos⁡αr_{C} = \lambda_{C}/2\pi\cos\alpha is the radius and p=λCsen⁡αp = \lambda_{C}\operatorname{sen}\alpha is the spacing between adjacent spirals, which is very close to the classical electron radius^{34}, since sen⁡α≈α\operatorname{sen}\alpha\approx\alpha. Furthermore, independently of the α\alpha angle, the shape of the spiral provides D=λCD = \lambda_{C}, which suggests that this must be a universal behavior of electrons in atomic orbits. Furthermore, considering the lB=NλCl_{B} = N\lambda_{C} it is easy to show that

R∞=α22λC=12lB,(50)R_{\infty} = \frac{\alpha^2}{2\lambda_C} = \frac{1}{2l_B}, \tag*{(50)}

where R∞=8ε02h3c/me4R_{\infty} = 8\varepsilon_0^2 h^3 c/me^4 is the Rydberg constant for the infinite nuclear mass[45], which is clearly related to the length of the NN spirals, suggesting again that the number of spirals is a constant for the orbits of the electrons in the atoms.

Right after calculating both the number of spirals NN and the distance DD, we realized that the velocity of the electron in the first orbit of the Bohr atom is related to the projection onto the orbital direction due to the helical motion of the electron running at speed of light (see red, blue, and black arrows in Figure 4(b)). This suggests that electrons do not travel directly along the orbits but in helical paths at speed of light under a spiral radius given by rC≈λC/2πr_C \approx\lambda_C/2\pi, as shown in Figure 4, which seems to agree with Dirac’s predictions for the motion of electrons (see, for instance, references 46 and 47).

This description confirms the proportionality relationship between FMF_M and FEF_E described in the previous session, as can be seen below. Let us look carefully at the Bohr atom using the center of mass idea, taking into account the motion of both electron and proton, as displayed in Figure 5.

Bohr atom considering the center of mass

Figure 5. Bohr atom considering the center of mass. vev_e and vpv_p, rer_e and rpr_p, and IeI_e and IpI_p are the orbital velocities, the orbital radius, and electrical currents of the electron (e−e^-) and proton (p+p^+), respectively. FEF_E and FBF_B represent the electrical and magnetic forces due to the interaction of the charges and electrical currents of the electron and proton. Both forces are centripetal as predicted by Equations (56) and (57).

Using the well-known description of the center of mass for the Bohr atom, the following equations can be found[34]:

vere=vprp=e22ε0h(re+rp)=1re+rp,(51)\frac{v_e}{r_e} = \frac{v_p}{r_p} = \frac{e^2}{2\varepsilon_0 h(r_e+r_p)} = \frac{1}{r_e+r_p}, \tag*{(51)}
Ie=eve2πre,(52)I_e = e\frac{v_e}{2\pi r_e}, \tag*{(52)}
Ip=evp2πrp,(53)I_p = e\frac{v_p}{2\pi r_p}, \tag*{(53)}
IT=e(ve+vp)2π(re+rp),(54)I_T = e\frac{(v_e+v_p)}{2\pi(r_e+r_p)}, \tag*{(54)}
Ie=Ip=IT,(55)I_e = I_p = I_T, \tag*{(55)}

and

ve+vpc=α=e22ε0hc(56)\frac{v_e+v_p}{c}=\alpha=\frac{e^2}{2\varepsilon_0hc} \tag*{(56)}

where vev_e and vpv_p, rer_e and rpr_p, and IeI_e and IpI_p are the orbital velocities, the orbital radius, and the electrical currents of the electron and proton, respectively.

A careful inspection of the electrical currents (see Figure 5 again) shows that a magnetic force must exist in the same direction as the electrical force, similarly to two parallel wires carrying the electrical currents IeI_e and IpI_p (ref. 8,37). This magnetic force per length unity (lBl_B) must be given by

FMlB=μ02π(r+R)IeIp.(57)\frac{F_M}{l_B}=\frac{\mu_0}{2\pi(r+R)}I_eI_p. \tag*{(57)}

Once we use lB=λB/α=λC/α2l_B=\lambda_B/\alpha=\lambda_C/\alpha^2, along with Equations (51) to (56), considering the limit r+R≈rr+R\approx r, the immediate consequence is that

FM=απFE(58)F_M=\frac{\alpha}{\pi}F_E \tag*{(58)}

where FEF_E is the classic electrical force between the electron and the proton, which perfectly agrees with the prediction given in Equations (43) to (45), after considering vF∗=vB∗=vBv_F^*=v_B^*=v_B.

In order to validate this picture, we have looked into the relationship between the spiral cross section and the helical electrical current along with the magnetic fields produced by it. This should provide information about the magnetic flux and must agree with the well-known magnetic moment of the electron, i. e. the electron spin.

The helical motion of the electron must be related to a centripetal force given by

Fc=mc2rc=m2c3ℏ=ecB(59)F_c=\frac{mc^2}{r_c}=\frac{m^2c^3}{\hbar}=ecB \tag*{(59)}

which provides the strong magnetic field

B=m2c2eℏ=4.413×109 T.(60)B = \frac{m^{2}c^{2}}{e\hbar} = 4.413 \times10^{9}\ \mathrm{T}. \tag*{(60)}

It should be noted that this centripetal force relates with the electrical force exactly as

Fc=1α3FE(61)F_{\mathrm{c}} = \frac{1}{\alpha^{3}}F_{\mathrm{E}} \tag*{(61)}

which is extremely intense compared to the charge attraction between electron and proton (6 orders higher). However, no matter the intensity of the magnetic field and FcF_{\mathrm{c}}, they have no influence on the proton, since the magnetic field is enclosed inside the toroidal solenoid predicted in Figure 4(a).

Even more interesting is the fact that one can use the centripetal force (FE=mvB∗2/rB∗F_{\mathrm{E}} = mv_{B}^{*2}/r_{B}^{*}) for the orbit of the equivalent Bohr atom in Equation (61), in which α=vB/c\alpha= v_{B}/c must be exchanged by the factor β=vB∗/c\beta= v_{B}^{*}/c. This directly provides the quantization of the angular momentum (mvB∗rB∗=ℏmv_{B}^{*}r_{B}^{*} = \hbar) as a consequence, suggesting again that the helical motion is a universal behavior of electrons in the orbits of atoms.

Making pp in Figure 5(b) equal to zero, since λCsin⁡α≪(λC/2π)cos⁡α\lambda_{\mathrm{C}}\sin\alpha\ll(\lambda_{\mathrm{C}}/2\pi)\cos\alpha, the magnetic flux per spiral of radius rC=λC/2πr_{\mathrm{C}} = \lambda_{\mathrm{C}}/2\pi is given by

Φ=BS=m2c2eℏπ(λC2π)2=h2e=Φ0(62)\Phi= BS = \frac{m^{2}c^{2}}{e\hbar}\pi\left(\frac{\lambda_{\mathrm{C}}}{2\pi}\right)^{2} = \frac{h}{2e} = \Phi_{0} \tag*{(62)}

where S=π(λC/2π)2S = \pi(\lambda_{\mathrm{C}}/2\pi)^{2} is the cross section of one spiral, which provides exactly one quantum flux per spiral.

Furthermore, computing the magnetic moment (μm\mu_{\mathrm{m}}) for each spiral allows us to directly obtain the electron spin (μS\mu_{\mathrm{S}}), as follows

μm=IS=ecλCπ(λC2π)2=eℏ2m=μS.(63)\mu_{\mathrm{m}} = IS = e\frac{c}{\lambda_{\mathrm{C}}}\pi\left(\frac{\lambda_{\mathrm{C}}}{2\pi}\right)^{2} = \frac{e\hbar}{2m} = \mu_{\mathrm{S}}. \tag*{(63)}

It is interesting to observe that the motion of the electron either in the clockwise or counterclockwise directions (see red and blue arrows in Figure 4(b) and (Figure 4c)), running towards the orbit, naturally provides both up and down spins for electrons, which is in agreement with what was observed in the Stern-Gerlach experiment[7, 48, 49]. In addition, these two electrons running the helical paths tend to cancel the total magnetic field inside the toroidal solenoid (see Figure 4(c) again), which must be the reason for the electron spin pairing predicted in the Pauli exclusion principle for a particular orbital[7].

On the other hand, if we look at the toroidal solenoid (Figure 4(a)), we can define its impedance as

L=NΦ0I=Φ0α2λCec=1α2h2e2hmc2=1α2RQhmc2(64)\mathcal{L}=\frac{N\Phi_0}{I}=\frac{\Phi_0}{\alpha^2}\frac{\lambda_C}{ec}=\frac{1}{\alpha^2}\frac{h}{2e^2}\frac{h}{mc^2}=\frac{1}{\alpha^2}R_Q\frac{h}{mc^2} \tag*{(64)}

providing the period of one turn in the solenoid given by

τB=LRQ=1α2hmc2(65)\tau_B=\frac{\mathcal{L}}{R_Q}=\frac{1}{\alpha^2}\frac{h}{mc^2} \tag*{(65)}

which is the period for the electron to make one turn in the first orbit of the Bohr atom.

It should be noted that while τB\tau_B has to do with the lBl_B distance (= λC/α2\lambda_C/\alpha^2) or the number NN of spirals (= 1/α21/\alpha^2), the time constant τC\tau_C has to do with the fundamental cyclotron frequency of the electron traveling at the speed of light in one spiral turn, which is clearly related to the time constant (τZ=τC/2=h/2mc2\tau_Z=\tau_C/2=h/2mc^2) discussed in zitterbewegung physics[46, 50, 51, 52].

Finally, we briefly consider the distortion of the circular spirals due to the small variation in the distance between electrons and protons, i.e. the Bohr radius, caused by the electrical force on the spirals (see Figure 4(a) again). In the point nearest to the proton, the electron is under the highest electrical force than in the farthest point, in which it is under the influence of the smallest force. The small difference in the electrical force with regard to the average value (rBr_B) expected in the Bohr first orbit happen in all points of each spiral in the helical motion, providing elliptical spirals instead of circular ones. Thus, considering the semi-minor and the semi-major axes of an elliptical spiral as b=λC/2πb=\lambda_C/2\pi and a=(1+α/2π)ba=(1+\alpha/2\pi)b, respectively, the elliptical cross section (S=πabS=\pi ab) leads to

μS ellipticalμS=IπabIπ(λC/2π)2=ab=(1+α2π)(66)\frac{\mu_{\mathrm{S\ elliptical}}}{\mu_{\mathrm{S}}}=\frac{I\pi ab}{I\pi(\lambda_C/2\pi)^2}=\frac{a}{b}=\left(1+\frac{\alpha}{2\pi}\right) \tag*{(66)}

or

μS elliptical=ge2=(1+α2π)μS(67)\mu_{\mathrm{S\ elliptical}}=\frac{g_e}{2}=\left(1+\frac{\alpha}{2\pi}\right)\mu_{\mathrm{S}} \tag*{(67)}

which corresponds to the first correction term of the anomalous magnetic moment predicted by Schwinger[53], where geg_e is the g-factor related to the electron gyromagnetic ratio[54].

Electron-electron scattering, Metal-Insulator transition, and electrical resistivity T2T^2 behavior

Considering Equations (23) and (29) together it is possible to write down

ρeen=h2e21Ll=mτ∗e2.(68)\rho_{ee}^n=\frac{h}{2e^2}\frac{1}{Ll}=\frac{m}{\tau^*e^2}. \tag*{(68)}

Due to the SrTiO3−δ_{3-\delta} to show metal-like behavior with very low charge carrier density almost temperature independent (see, for instance, figure 3(a) of reference 55) this compound along with conventional metals seem to be the best examples for this study.

Considering the conservation of the angular momentum, we can write

mvB∗rB∗=mvBλB2π=ℏ=h2π(69)m v_B^{*} r_B^{*} = m v_B \frac{\lambda_B}{2\pi} = \hbar= \frac{h}{2\pi} \tag*{(69)}

or

(π3)1/3mvFλ=mvBλB=mcλC=h(70)\left(\frac{\pi}{3}\right)^{1/3}m v_F \lambda= m v_B \lambda_B = m c \lambda_C = h \tag*{(70)}

which means there is an equivalent Bohr radius given by 2πrB∗=λ=2L2\pi r_B^{*} = \lambda= 2L for the electrons at the Fermi level of each metal, in which the atom is seen as an electron travelling at an orbit connected to a single proton at the center of the atom. This seems to be supported by Gauss’ law in which any atom can be seen as a unique proton at the center of the atom inside the volume just below the equivalent Bohr radius, i.e.i.e. an atom with one electron and one proton or the equivalent Bohr atom. This is supported by a description of the Metal-Insulator Transition (MIT) predicted by Mott[13] and discussed in detail by Edwards et al.[56] and Behnia[57].

Putting vF=l/τv_F = l/\tau, vB=lB/τ∗v_B = l_B/\tau^{*}, and c=lc/τ∗c = l_c/\tau^{*} into (70) and remembering that τ=(π/3)1/3τ∗\tau= (\pi/3)^{1/3}\tau^{*}, implies that

2Ll=2LBlB=λClc=hτ∗m=(3π)1/3hτm(71)2Ll = 2L_B l_B = \lambda_C l_c = \frac{h\tau^{*}}{m} = \left(\frac{3}{\pi}\right)^{1/3}\frac{h\tau}{m} \tag*{(71)}

which is in agreement with (29) and (67). It should be noted that the smaller the equivalent Bohr radius is, the larger the mean free path is.

Such as

lB=λBα=λCα2(72)l_B = \frac{\lambda_B}{\alpha} = \frac{\lambda_C}{\alpha^2} \tag*{(72)}

the length lcl_c can be calculated based on

λClc=λB2α=λC2α3(73)\lambda_C l_c = \frac{\lambda_B^2}{\alpha} = \frac{\lambda_C^2}{\alpha^3} \tag*{(73)}

which implies

lc=λCα3=6.228×10−6 m(74)l_c = \frac{\lambda_C}{\alpha^3} = 6.228 \mathbin{\times} 10^{-6}\ \mathrm{m} \tag*{(74)}

that is the largest mean free path for the electron traveling at the speed of light (vF∗=cv_F^{*} = c) in the conductor supposedly with the highest electrical conductivity, which has charge density n=5.603×1035 m−3n = 5.603 \mathbin{\times} 10^{35}\ \mathrm{m}^{-3} and the maximum Fermi current of IF∗=19.80 AI_F^{*} = 19.80\ \mathrm{A} (ref. [46, 47, 58, 59, 60]).

Furthermore, the time constant between two electron-electron collisions is given by

τ∗=1α3λCc=1α3hmc2=2.082×10−14 s(75)\tau^* = \frac{1}{\alpha^3}\frac{\lambda_C}{c} = \frac{1}{\alpha^3}\frac{h}{mc^2} = 2.082 \times10^{-14}\ \mathrm{s} \tag*{(75)}

or

τ=(π/3)1/3τ∗=2.114×10−14 s(76)\tau= (\pi/3)^{1/3}\tau^* = 2.114 \times10^{-14}\ \mathrm{s} \tag*{(76)}

which is independent of the electrical conductor. It is noteworthy that this relaxation time allows a metal with large carrier density, i.e. small LL and many charges close to each other, to have a very large mean free path. This has to do with the collision cross section (σS\sigma_S) (ref. 38) given by

σS=1nl(77)\sigma_S = \frac{1}{nl} \tag*{(77)}

After using Equations (70) and (72), one can show that

σS=2αλB2L4(78)\sigma_S = \frac{2\alpha}{\lambda_B^2}L^4 \tag*{(78)}

leading to a typical collision radius for the electron-electron scattering which is proportional to rB∗2/rBr_B^{*2}/r_B.

Additionally, Equation (78) demonstrates that the larger the carrier density is, the smaller LL and σS\sigma_S are, reducing electron-electron scattering, favoring electrical conduction. This is in agreement with the recent discussion by Behnia17.

Considering Equations (71) and (73), Equation (68) can be rewritten as

ρeen=h2e21Ll=h2e22αλB2=mτ∗e2=1.705×1021 ohm/m2.(79)\rho_{ee} n = \frac{h}{2e^2}\frac{1}{Ll} = \frac{h}{2e^2}\frac{2\alpha}{\lambda_B^2} = \frac{m}{\tau^*e^2} = 1.705 \times10^{21}\ \mathrm{ohm/m^2}. \tag*{(79)}

This suggests that the mean free path of a metal has to do with the helical motion of the electron at the speed of light. Furthermore, this brings a clear insight on the mean free path of the electron. In fact, Equations (71), (73), and (79) suggest that all conducting electrons travel a mean free path proportional to lcl_c (l=lcλC/λl = l_c\lambda_C/\lambda) through helical paths at the speed of light, during the time τ∗\tau^* between two collisions.

Another interesting experimental aspect has to do with the MIT predicted by Mott13, in which the electrical resistance RR must be equal to the quantum resistance at the transition and the critical carrier density and the equivalent Bohr radius should show a universal behavior in which nc1/3rB∗=constantn_c^{1/3}r_B^* = \text{constant} (see, for instance, (reference 57)).

As predicted in the Equation (71), the smallest mean free path ll must be equal to λC\lambda_C, which is related to the longest equivalent Bohr radius rB∗=lc/2πr_B^* = l_c/2\pi (ref. 32,33). For this atom, the critical charge density is

nc=1Lc3=(2lc)3=3.312×1016 m−3.(80)n_c = \frac{1}{L_c^3} = \left(\frac{2}{l_c}\right)^3 = 3.312 \times10^{16}\ \mathrm{m}^{-3}. \tag*{(80)}

Thus, Equations (74) and (79) provide

ρc=h2e22αλB2(lc2)3=h2e2λc4α6=51,848 ohm.m(81)\rho_c = \frac{h}{2e^2}\frac{2\alpha}{\lambda_B^2}\left(\frac{l_c}{2}\right)^3 = \frac{h}{2e^2}\frac{\lambda_c}{4\alpha^6} = 51{,}848\ \mathrm{ohm.m} \tag*{(81)}

which allows us to determine the electrical resistance at the MIT as

Rc=ρclL2=h2e22αλB2(lc/2)3λc(lc/2)2=h2e2=RQ=12,909 Ω,(82)R_c = \rho_c\frac{l}{L^2} = \frac{h}{2e^2}\frac{2\alpha}{\lambda_B^2}(l_c/2)^3\frac{\lambda_c}{(l_c/2)^2} = \frac{h}{2e^2} = R_Q = 12{,}909\ \Omega, \tag*{(82)}

which precisely agrees with the expected quantum resistance at the transition.

Furthermore, considering lc=2πrB∗l_c = 2\pi r_B^*, it is easy to show that

nc1/3rB∗=1/π=0.318(83)n_c^{1/3}r_B^* = 1/\pi= 0.318 \tag*{(83)}

which agrees, at least within the experimental error (nc1/3rB∗=0.26±0.05n_c^{1/3}r_B^* = 0.26 \pm0.05), with the result for the MIT in many metallic systems reported by Edwards et al.56.

Finally, let us to discuss the electron-electron scattering and T2T^2 behavior of the electrical resistivity at low temperatures.

From Equation (33) it is possible to write the repulsion energy between electron-electron as

U=FEL=παλcϵF∗L(84)U = F_E L = \pi\frac{\alpha}{\lambda_c}\epsilon_F^*L \tag*{(84)}

which, corrected for the volume of the one cell, provides

ULl=π(π/3)2/3ϵF1λBL2l=π(π/3)2/3ϵF1λB1RQρee(85)U\frac{L}{l} = \pi(\pi/3)^{2/3}\epsilon_F\frac{1}{\lambda_B}\frac{L^2}{l} = \pi(\pi/3)^{2/3}\epsilon_F\frac{1}{\lambda_B}\frac{1}{R_Q}\rho_{ee} \tag*{(85)}

after using Equations (23) and (68).

In fact, we noticed that UL/lUL/l must be equal to the level of thermal energy per charge carrier (N=1N = 1) due to the Joule effect produced by the electron-electron scattering, which allows us to write

ULl=ϵFπ(π/3)2/31λB1RQρee=CeT=(γT)T=π2kB22EFT2(86)U\frac{L}{l} = \epsilon_F\pi(\pi/3)^{2/3}\frac{1}{\lambda_B}\frac{1}{R_Q}\rho_{ee} = C_eT = (\gamma T)T = \frac{\pi^2 k_B^2}{2E_F}T^2 \tag*{(86)}

where Ce=γTC_e = \gamma T is the electronic heat capacity, providing

ρee=π22(π/3)2/3λBRQπkB2ϵF2T2=QT2TF2=AT2(87)\rho_{ee} = \frac{\pi^2}{2}(\pi/3)^{2/3}\lambda_B\frac{R_Q}{\pi}\frac{k_B^2}{\epsilon_F^2}T^2 = Q\frac{T^2}{T_F^2} = AT^2 \tag*{(87)}

where

Q=π22(3π)2/3λBRQπ=6.539×10−6 Ohm⋅m(88)Q = \frac{\pi^{2}}{2}\left(\frac{3}{\pi}\right)^{2/3}\lambda_{B}\frac{R_{Q}}{\pi} = 6.539 \times10^{-6}\ \text{Ohm}\cdot\text{m} \tag*{(88)}

and

A=QTF2=6.539×10−6TF2 Ohm⋅m/K2(89)A = \frac{Q}{T_{F}^{2}} = \frac{6.539 \times10^{-6}}{T_{F}^{2}}\ \text{Ohm}\cdot\text{m}/\text{K}^{2} \tag*{(89)}

or

A=3.885×10−15γ2 Ohm⋅m/K2.(90)A = 3.885 \times10^{-15}\gamma^{2}\ \text{Ohm}\cdot\text{m}/\text{K}^{2}. \tag*{(90)}

Once we have found the expected behavior for the electron-electron scattering, it can be carefully compared with the experimental data for the pre-factor AA versus TFT_{F} available in the literature. Figure 6 displays data for many conventional and non-conventional conductors taken from the several references[14, 15, 16, 17, 18, 19, 20, 21, 22].

A pre-factor as a function of the Fermi temperature

Figure 6. AA pre-factor as a function of the Fermi temperature. Black and green circle symbols are related to the classical metals and SrTiO3−δ_{3-\delta}, respectively (mostly reviewed by Behnia and co-authors[16, 17, 18, 19]). Blue circle symbols are metals with very high Fermi temperature, whose AA pre-factor is not available experimentally. In such cases, points were calculated using Equation (89). Yellow circle symbols are due to weak magnetic correlated Yb-based compounds, reported by Tsujii et al.[22]. Open symbols are related to non-conventional metals, such as high temperature superconductors, strongly correlated systems, and others. Dashed red and blue lines represent the experimental observations by Rice[14] and Kadowaki-Woods[21], respectively. Black line represents the behavior given by Equation (89) and is in excellent agreement with the empirical prediction by Behnia and co-authors[16, 17, 18]. Insert compares the experimental results (dashed red line) by Rice[14] with the prediction (black line) by Equation (90).

Data for the most conducting materials (blue circle symbols), which have very high Fermi temperatures and very small AA, were calculated using Equation (89) and TFT_F available in the literature[5, 6], and are displayed in Figure 6 only for comparison purposes. Black and green circle symbols are related to the single metallic conductors. Yellow circle symbols are due to weak magnetic correlated Yb-based compounds[22].

Dashed red and blue lines are the behaviors reported by Rice[14] and Kadowaki-Woods[21]. Black line is the prediction given by Equation (87), which fits well with the experimental data in 8 orders of magnitude of the AA parameter. The insert displays a comparison of the result by Rice[14] with the behavior predicted by Equation (89), which shows excellent agreement.

Open symbols are related to non-conventional conductors, in which carrier densities are temperature dependent. Due to that, they are supposedly not able to be described by the model reported in this work. The comparison of the results for the conventional metals, SrTiO3−δ_{3-\delta}, and Yb-based compounds with the non-conventional conductors suggests that the strong magnetic correlation, high temperature superconductivity, and other effects should play an important role on electron-electron scattering, which deviate it from the expected behavior given by the model reported here. One possibility to describe this difference is to take into account the magnetic energy (ϵM=μB\epsilon_M = \mu B) in the Fermi-Dirac distribution.

Finally, the results are in excellent agreement with the recent observations by Behnia and co-authors[16, 17, 18], where pre-factor AA in Equation (88) can be rewritten as

A=ℏe2(kBϵF)2ℓquad(91)A = \frac{\hbar}{e^2}\left(\frac{k_B}{\epsilon_F}\right)^2 \ell_{\mathrm{quad}} \tag*{(91)}

where

ℏe2=RQπ=4,109 ohm(92)\frac{\hbar}{e^2} = \frac{R_Q}{\pi} = 4{,}109\ \mathrm{ohm} \tag*{(92)}

and

ℓquad=π22(3π)2/3λB=1.591×10−9 m,(93)\ell_{\mathrm{quad}} = \frac{\pi^2}{2}\left(\frac{3}{\pi}\right)^{2/3}\lambda_B = 1.591 \times10^{-9}\ \mathrm{m}, \tag*{(93)}

which is the same as the empirical prediction reported in references 16 to 18.

Conclusion

In this work we report arguments to consider electrical energy as an important parameter to describe the motion of electrons near the Fermi level at zero temperature. Thermal energy has been added to the electrical energy into the Fermi-Dirac distribution, which allows us to establish equations to predict the behavior of the electrical resistivity of many metals at low temperatures. As a consequence, Ohm’s law, the Drude model, and the quantization of electrical resistance have derived naturally from quantization of the angular momentum. The description allowed to conciliate the Drude model with Fermi gas theory.

The proportionality ratio between magnetic and electrical forces predicted the helical motion of the electrons and allowed a better understanding of the mean free path. Furthermore, it was possible to show that the Joule effect is directly related to electrical and magnetic forces, which, as far as we know, has never been reported so far.

Applying the considerations to the Bohr atom, it was shown that the helical motion of the electron is related to the Sommerfeld fine structure constant. This has been confirmed by the natural appearance of the magnetic flux quantization and electron spin, as well as the anomalous behavior of the magnetic moment predicted by Schwinger.

The comparison of the model with electrical transport properties of several metallic conductors provided an excellent agreement and a better understanding of the electron-electron scattering mechanism at low temperatures. The model predicts the behavior of pre-factor AA of the quadratic temperature electrical resistivity dependence at low temperatures, due to electron-electron scattering in conductors, in a wide range (8 orders of magnitude). The model is in excellent agreement with previous results by Rice and with the recent observations by Behnia and co-authors.

This work opens a new scenario for the description of electron motion, as well as introduces new insights into the quantum mechanics and relativistic behavior of the electrons. The dependence of the model with the Sommerfeld fine structure constant suggests that this work must be of interest for several areas, such as solid-state physics, particle physics, and quantum electrodynamic theory.

Acknowledgements

The authors are thankful to J. J. Neumeier for many valuable comments. L. M. S. Alves is post-doc at EEL-USP (Proc. 21.1.419.88.8). F. S. Oliveira is post-doc at UNICAMP (FAPESP 2021/03298-7). M. S. da Luz is CNPq fellow (Proc. 311394/2021-3).

References

  1. [1]Drude., P. Zur Elektronentheorie der Metalle. Ann. Phys. 306, 566–613 (1900).
  2. [2]Drude, P. Drude, P. (1900). Zur Elektronentheorie der Metalle; II. Teil. Galvanomagnetische und thermomagnetische Effecte. Ann. Phys. 308, 369–402 (1900).DOI
  3. [3]Scheffler, M., Dressel, M., Jourdan, M. & Adrian, H. Extremely slow Drude relaxation of correlated electrons. Nature 438, 1135–1137 (2005).pubmed.ncbi.nlm.nih.gov/16372004
  4. [4]Lorentz, H. A. The Theory of Electrons and Its Applications to the Phenomena of Light and Radiant Heat. (Cosimo Classics, 2007).
  5. [5]Kittel, C. Introduction to Solid State Physics, 8th Edition. (John Wiley & Sons, 2005).
  6. [6]Ashcroft, N. & Mermin, N. D. Solid State Physics. (Rinehart and Winston, 1976).
  7. [7]Griffiths, D. J. Introduction to quantum mechanics I. (Prentice Hall, Inc., 1995).
  8. [8]Jackson, J. D. Introduction to electrodynamics, 3rd Edition. (Wiley, 1998).
  9. [9]Sommerfeld, A. Zur Elektronentheorie der Metalle auf Grund der Fermischen Statistik - I. Teil: Allgemeines, Strömungs- und Austrittsvorgänge. Zeitschrift für Phys. 47, 1–32 (1928).
  10. [10]Eckerta, M. How Sommerfeld extended Bohr’s model of the atom (1913-1916). Eur. Phys. J. H 39, 141–156 (2014).DOI
  11. [11]Landau, L. & Pomerantschusk, I. Über Eigenschaften der metalle bei sehrniedrigen Temperaturen. Phys. Z. Sowjet 10, 649–653 (1936).
  12. [12]Baber, W. G. & Wills, H. H. The Contribution to the Electrical Resistance of Metals from Collisions Between Electrons. Proc. R. Soc. A 158, 383–396 (1936).DOI
  13. [13]Mott, N. F. Metal Insulator Transition, 2nd edition. (Taylor and Francis, 1990).
  14. [14]Rice, M. J. Electron-Electron Scattering in Transition Metals. Phys. Rev. Lett. 20, 1439–1441 (1968).
  15. [15]Ishizuka, H. & Levitov, L. Wide-range T² resistivity and umklapp scattering in moiré graphene. New J. Phys. 24, 1–9 (2022).DOI
  16. [16]Wang, J. et al. T-square resistivity without Umklapp scattering in dilute metallic Bi₂O₂Se. Nat. Commun. 11, 1–8 (2020).DOI
  17. [17]Behnia, K. On the Origin and the Amplitude of T-Square Resistivity in Fermi Liquids. Ann. Phys. 534, 1–12 (2022).DOI
  18. [18]Lin, X., Fauqué, B. & Behnia, K. Scalable T² resistivity in a small single-component Fermi surface. Science. 349, 2–6 (2015).DOI
  19. [19]Jaoui, A. et al. Departure from the Wiedemann–Franz law in WP2 driven by mismatch in T-square resistivity prefactors. npj Quantum Mater. 3, 1–7 (2018).DOI
  20. [20]Kaveh, M. & Wiser, N. Electron-electron scattering in conducting materials. Adv. Phys. 33, 257–372 (1984).
  21. [21]Kadowaki, K. & Woods, S. B. Universal Relationship of the Resistivity and Specific Heat in Heavy-Fermion Compounds. Solid State Commun. 58, 507–509 (1986).DOI
  22. [22]Tsujii, N., Yoshimura, K. & Kosuge, K. Deviation from the Kadowaki-Woods relation in Yb-based intermediate-valence systems. J. Phys. Condens. Matter 15, 1993–2003 (2003).DOI
  23. [23]Klitzing, K. V., Dorda, G. & Pepper, M. New method for high-accuracy determination of the fine-structure constant based on quantized Hall resistance. Phys. Rev. Lett. 45, 494–497 (1980).
  24. [24]von Klitzing, K. Quantized Hall Effect. Phys. B Phys. Condens. Matter C At. Mol. Plasma Physics, Opt. 126 B-C, 242–249 (1984).
  25. [25]Sarma, S. Das & Pinczuk, A. Perspectives in Quantum Hall Effects. (Wiley, 1997).
  26. [26]Josephson, B. D. Possible new effects in superconductive tunnelling. Phys. Lett. 1, 251–253 (1962).DOI
  27. [27]Ambegaokar, V. & Baratoff, A. Tunneling between superconductors. Phys. Rev. Lett. 11, 104 (1963).
  28. [28]Josephson, B. D. The Discovery of Tunnelling Supercurrents. Proc. IEEE 62, 838–841 (1974).pubmed.ncbi.nlm.nih.gov/17755018
  29. [29]Martinis, J. M., Devoret, M. H. & Clarke, J. Quantum Josephson junction circuits and the dawn of artificial atoms. Nat. Phys. 16, 234–237 (2020).DOI
  30. [30]Li, T. et al. Quantum anomalous Hall effect from intertwined moiré bands. Nature 600, 641–646 (2021).DOI
  31. [31]Du, Z. Z., Wang, C. M., Sun, H. P., Lu, H. Z. & Xie, X. C. Quantum theory of the nonlinear Hall effect. Nat. Commun. 12, (2021).DOI
  32. [32]Sukhatme, K., Mukharsky, Y., Chui, T. & Pearson, D. Observation of the ideal josephson effect in superfluid 4He. Nature 411, 280–283 (2001).DOI
  33. [33]Bohr, N. Philosophical Magazine Series 6 I . On the constitution of atoms and molecules. Philos. Mag. 26, 37 (1913).
  34. [34]Eisberg, R. Quantum Physics of Atoms, Molecules, Solids, Nuclei, and Particles, Second Edition. (John Wiley & Sons, 1985).
  35. [35]Ibach, H. & Luh, H. Solid-State Physics, 4th Edition. (Springer-Verlag, 2009).
  36. [36]Feynman, R. P., Leighton, R. B. & Sands, M. The Feynman Lectures on physics - Volume-2. (Addison-Wesley Publishing Company, 1964).
  37. [37]Griffiths, D. J. Introduction to electrodynamics, 3rd Edition. (Prentice Hall, 1999).
  38. [38]Krasnov, A. E., Yu Ryzhkova, D. & Vagin, V. A. Methodology for identifying gaseous constituents of the atmosphere. J. Phys. Conf. Ser. 2091, (2021).DOI
  39. [39]Morel, L., Yao, Z., Cladé, P. & Guellati-Khélifa, S. Determination of the fine-structure constant with an accuracy of 81 parts per trillion. Nature 588, 61–65 (2020).DOI
  40. [40]Bouchendira, R., Cladé, P., Guellati-Khélifa, S., Nez, F. & Biraben, F. New determination of the fine structure constant and test of the quantum electrodynamics. Phys. Rev. Lett. 106, 7–10 (2011).DOI
  41. [41]Parker, R. H., Yu, C., Zhong, W., Estey, B. & Müller, H. Measurement of the fine-structure constant as a test of the Standard Model. Science 360, 191–195 (2018).DOI
  42. [42]Hanneke, D., Fogwell, S. & Gabrielse, G. New measurement of the electron magnetic moment and the fine structure constant. Phys. Rev. Lett. 100, 1–4 (2008).DOI
  43. [43]2019 CODATA Value: Fine-structure constant”. The NIST Reference on Constants, Units, and Uncertainty. NIST. Retrieved 2022-05-07.
  44. [44]Duarte, C. D. A. A simple geometry-based model for the photon. Int. J. Geom. Methods Mod. Phys. 15, 1–13 (2018).DOI
  45. [45]Suto, K. Theoretical and Experimental Values for the Rydberg Constant Do Not Match. J. Appl. Math. Phys. 09, 1993–2003 (2021).DOI
  46. [46]Di Tommaso, A. O. & Vassallo, G. Electron structure, ultra-dense hydrogen and low energy nuclear reactions. J. Condens. Matter Nucl. Sci. 29, 525–547 (2019).
  47. [47]Celani, F., Di Tommaso, A. O. & Vassallo, G. The zitterbewegung interpretation of quantum mechanics as theoretical framework for ultra-dense deuterium and low energy nuclear reactions. J. Condens. Matter Nucl. Sci. 24, 32–41 (2017).DOI
  48. [48]Stern, O. A way towards the experimental examination of spatial quantisation in a magnetic field. Z. Phys., D At. Mol. Clust. 10, 114–116 (1988).DOI
  49. [49]Weinert, F. Wrong theory-Right experiment: The significance of the Stern-Gerlach experiments. Stud. Hist. Philos. Mod. Phys. 26, 75–86 (1995).DOI
  50. [50]Merali, Z. Quivering ions pass quantum test. Nature (2010). https://doi.org/10.1038/news.2010.2.
  51. [51]Hestenes, D. The zitterbewegung interpretation of quantum mechanics. Found. Phys. 20, 1213–1232 (1990).DOI
  52. [52]Gerritsma, R. et al. Quantum simulation of the Dirac equation. Nature 463, 68–71 (2010).DOI
  53. [53]Schwinger, J. Quantum electrodynamics. III. The electromagnetic properties of the electron-radiative corrections to scattering. Phys. Rev. 76, 790–817 (1949).
  54. [54]Odom, B., Hanneke, D., D’Urso, B. & Gabrielse, G. New measurement of the electron magnetic moment using a one-electron quantum cyclotron. Phys. Rev. Lett. 97, 6–9 (2006).pubmed.ncbi.nlm.nih.gov/16907490
  55. [56]Edwards, P. P., Ramakrishnan, T. V. & Rao, C. N. R. The Metal - Nonmetal Transition : A Global Perspective. J. Phys. Chem. 99, 5228–5239 (1995).
  56. [57]Behnia, K. On mobility of electrons in a shallow Fermi sea over a rough seafloor. J. Phys. Condens. Matter 27, 375501 (2015).DOI
  57. [58]We have named this metal as Luzonium in honor to the word light in Portuguese, but in the course of this work we have noticed that it has similarities with the so-called as Feynmanium atom (Z = 137) by some authors. See references 59 and 60 too.
  58. [59]Chandrasekhar, S. On Stars, their evolution and their stability. Rev. Mod. Phys., 56, 2, Part I (1984).pubmed.ncbi.nlm.nih.gov/17821494
  59. [60]Bezverkhniy, V. & Bezverkhniy, V. The Speed of Light and the Number of Chemical Elements. SSRN Electron. J. 1–8 (2020) doi:10.2139/ssrn.3709471.DOI

Paper details

vlnc:W_zzk765z7
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents