betaFeedback

WorksPhysics and Astronomy

A short note on the eigenvalues of regular Hall-plates

Abstract

This note is about uniform, plane, singly connected, regular Hall-plates with an arbitrary number of contacts exposed to a uniform magnetic field of arbitrary strength. In practice, the regular symmetry is the most common one. If the Hall-plates are mapped conformally to the unit disk, regular symmetry means that all contacts are equally large and all contact spacings are equally large, yet the contact spacing may be different from the size of the contacts. The indefinite conductance matrices of such Hall-plates are circulant matrices, whose complex eigenvalues are computed in closed form.

Definitions

In electrically linear Hall-plates the currents into the terminals are linear combinations of the potentials at the terminals. Suppose that the Hall-plate has NN terminals. We group all NN currents to a vector I\boldsymbol{I} and all NN voltages to a vector V\boldsymbol{V}. Then we express the linear combination as a matrix multiplication I=iGV\boldsymbol{I} = \mathrm{i}\boldsymbol{G}\boldsymbol{V}, with the indefinite conductance matrix iG\mathrm{i}\boldsymbol{G}. Thereby, the reference potential (ground node) is arbitrary – none of the terminals needs to be grounded. If we ground the ℓ\ell-th terminal, we write I=GV\boldsymbol{I} = \boldsymbol{G}\boldsymbol{V}, where we delete the ℓ\ell-th current and voltage in I,V\boldsymbol{I},\boldsymbol{V}, respectively, and we delete the ℓ\ell-th row and column in iG\mathrm{i}\boldsymbol{G} to get G\boldsymbol{G}. The indefinite conductance matrix is positive semi-definite – its determinant vanishes and therefore it cannot be inverted. The definite conductance matrix of any passive (= dissipative) system is positive definite, its determinant is positive, and an inverse exists: V=RI\boldsymbol{V} = \boldsymbol{R}\boldsymbol{I} with the resistance matrix R=G−1\boldsymbol{R} = \boldsymbol{G}^{-1}.

In the presence of the Hall-effect Ohm’s law is

E=ρJ−ρμHJ×(Bznz),(1)\boldsymbol{E} = \rho\boldsymbol{J} - \rho\mu_H\boldsymbol{J} \times(B_z\boldsymbol{n}_z), \tag*{(1)}

with the electric field E\boldsymbol{E}, the current density J\boldsymbol{J}, the unit vector nz\boldsymbol{n}_z orthogonal to the plane Hall-plate, with the zz-component of the magnetic flux density BzB_z, and with the Hall-mobility μH\mu_H. Thus, E\boldsymbol{E} and J\boldsymbol{J} are not colinear – there is the Hall-angle θH\theta_H in-between, with tan⁡(θH)=μHBz\tan(\theta_H) = \mu_H B_z. We use the following abbreviations,

Rsheet=ρtH,Rsq=Rsheetcos⁡(θH),(2)R_{\mathrm{sheet}} = \frac{\rho}{t_H}, \qquad R_{\mathrm{sq}} = \frac{R_{\mathrm{sheet}}}{\cos(\theta_H)}, \tag*{(2)}

whereby tHt_H is the thickness of the Hall-plate. If we have a square plate with two contacts fully covering two opposite edges, the resistance between these contacts is the sheet resistance RsheetR_{\mathrm{sheet}} at zero magnetic field, and it is the square resistance RsqR_{\mathrm{sq}} in the presence of a magnetic field1.

Regular Hall-plates

Suppose we have a regular boundary with NN contacts of equal size, the spacings between the contacts are all equal, yet they are not necessarily identical to the size of the contacts (see Figure 1). Suppose we connect an external circuit to

Examples of strictly regular and regular Hall-plates with four contacts

Figure 1. Examples of strictly regular and regular Hall-plates with four contacts.

this Hall-plate. If we keep the Hall-plate fixed and move the external circuit by one terminal, say clockwise, then the resulting potentials and currents into the external circuit must remain the same, because the Hall-plate is regular symmetric in shape and the applied magnetic field is parallel to the rotation axis. This means that the entries in the current and voltage vectors move one instance downwards, and the entries in the indefinite conductance matrix move one row down and one column to the right. Both the original and the new iG\mathrm{i}\boldsymbol{G} matrices must be identical. We can repeat this procedure NN times. The conclusion is that the elements along diagonals parallel to the main diagonal are identical, which gives the following patterns for iG\mathrm{i}\boldsymbol{G} and G\boldsymbol{G},

iG=(ABCDEEABCDDEABCCDEABBCDEA)⇒G=(ABCDEABCDEABCDEA).(3)\mathrm{i}\boldsymbol{G} = \begin{pmatrix} A & B & C & D & E \\ E & A & B & C & D \\ D & E & A & B & C \\ C & D & E & A & B \\ B & C & D & E & A \end{pmatrix} \Rightarrow \boldsymbol{G} = \begin{pmatrix} A & B & C & D \\ E & A & B & C \\ D & E & A & B \\ C & D & E & A \end{pmatrix}. \tag*{(3)}

iG\mathrm{i}\boldsymbol{G} is an N×NN \times N matrix, G\boldsymbol{G} is an (N−1)×(N−1)(N-1) \times(N-1) matrix. We get G\boldsymbol{G} from iG\mathrm{i}\boldsymbol{G} by deleting the row and column which corresponds to the grounded contact. G\boldsymbol{G} and iG\mathrm{i}\boldsymbol{G} are Toeplitz matrices, which have the property that Gℓ,m=gℓ−mG_{\ell,m} = g_{\ell-m} – it depends only on ℓ−m\ell-m. Moreover, iG\mathrm{i}\boldsymbol{G} is a circulant matrix with gℓ−m=gℓ−m+Ng_{\ell-m} = g_{\ell-m+N}. Third, the sum of entries per row in any indefinite conductance matrix has to vanish, because no currents flow if all contacts are at identical non-zero potential.

If we map a regular Hall-plate to the unit disk, its contact CmC_m ranges from azimuthal angle

αm=2πm−1N to βm=αm+χπN,(4)\alpha_m = 2\pi\frac{m-1}{N}\ \text{to}\ \beta_m = \alpha_m + \chi\frac{\pi}{N}, \tag*{(4)}

for χ∈(0,2)\chi\in(0,2) and m=1,2,…,Nm = 1,2,\ldots,N. Contacts are small for $\chi \to 0, whereas contact spacings are small for χ→2\chi\to2. The case χ=1\chi= 1 gives strictly regular Hall-plates, whose contacts are equally large as the contact spacings (see Figure 1).

We can write down the eigenvalues and eigenvectors of a real-valued circulant matrix2

iG=Q iΓQCwithiΓ=diag⁡(iγ1,iγ2,…,iγN),{}^{i}G = Q\,{}^{i}\Gamma Q^{C}\quad\text{with}\quad{}^{i}\Gamma= \operatorname{diag}({}^{i}\gamma_{1},{}^{i}\gamma_{2},\ldots,{}^{i}\gamma_{N}),
(Q)k,ℓ=1Nexp⁡(2πikℓN) ∀k,ℓ∈{1,2,…,N}.(5)(Q)_{k,\ell} = \frac{1}{\sqrt{N}}\exp\left(\frac{2\pi i k\ell}{N}\right)\ \forall k,\ell\in\{1,2,\ldots,N\}. \tag*{(5)}

QCQ^{C} is the conjugate complex of QQ, and i=−1i = \sqrt{-1}. iΓ{}^{i}\Gamma is a diagonal matrix, which comprises all eigenvalues iγℓ{}^{i}\gamma_{\ell} of iG{}^{i}G. It holds Q=QTQ = Q^{T} and QQC=QCQ=1QQ^{C} = Q^{C}Q = 1. The eigenvectors of iG{}^{i}G are the columns of QQ, which are coefficient vectors of the discrete Fourier transform (DFT). It generally holds

iγN−ℓ=iγℓC,(6){}^{i}\gamma_{N-\ell} = {}^{i}\gamma_{\ell}^{C}, \tag*{(6)}

for the eigenvalues of any real circulant matrix, where zCz^{C} is the conjugate complex of zz.

How to compute the eigenvalues of iG{}^{i}G

We use Ref. 2. Their equation (17) reads in our notation

∑m=1NBk,mϕm=∑m=1N−Ak,mψmcos⁡(θH)for k∈{1,…,N−1},(7)\sum_{m=1}^{N} B_{k,m}\phi_{m} = \sum_{m=1}^{N}\frac{-A_{k,m}\psi_{m}}{\cos(\theta_{H})}\quad\text{for }k\in\{1,\ldots,N-1\}, \tag*{(7)}

whereby ϕm\phi_{m} is the potential on CmC_{m} (= the mm-th peripheral contact), and ψm\psi_{m} is the stream function on the insulating peripheral segment between CmC_{m} and Cm+1C_{m+1}, with CN=C0C_{N}=C_{0}. In Ref. 2 the matrices A,BA,B are defined as

Ak,m=∫βmαm+1h(τ) dτsin⁡((τ−βk)/2)sin⁡((τ−βN)/2),A_{k,m} = \int_{\beta_{m}}^{\alpha_{m+1}}\frac{h(\tau)\,\mathrm{d}\tau}{\sin((\tau-\beta_{k})/2)\sin((\tau-\beta_{N})/2)},
Bk,m=∫αmβmh(τ) dτsin⁡((τ−βk)/2)sin⁡((τ−βN)/2),(8)B_{k,m} = \int_{\alpha_{m}}^{\beta_{m}}\frac{h(\tau)\,\mathrm{d}\tau}{\sin((\tau-\beta_{k})/2)\sin((\tau-\beta_{N})/2)}, \tag*{(8)}

with

h(τ)=∏ℓ=1N[sin⁡((τ−βℓ)/2)sin⁡((τ−αℓ)/2)]1/2+θH/π.h(\tau)=\prod_{\ell=1}^{N}\left[\frac{\sin((\tau-\beta_{\ell})/2)}{\sin((\tau-\alpha_{\ell})/2)}\right]^{1/2+\theta_{H}/\pi}.

Note that in (7) the sums extend over NN terms, whereas in Ref. 2 they had only N−1N-1 terms. Hence, in contrast to Ref. 2, we do not require ϕN=0\phi_{N}=0 and ψN=0\psi_{N}=0. This enables us to work with the indefinite conductance matrix, which is circulant, and therefore it lends for a simpler mathematical treatment than the definite conductance matrix. Therefore, our matrices A,BA,B have NN columns and N−1N-1 rows. Equation (7) must hold, if we add an arbitrary constant to all ϕk\phi_{k} or to all ψk\psi_{k}. Consequently it must hold

∑m=1NAk,m=0and∑m=1NBk,m=0for k∈{1,…,N−1}.(9)\sum_{m=1}^{N}A_{k,m}=0\quad\text{and}\quad\sum_{m=1}^{N}B_{k,m}=0\quad\text{for }k\in\{1,\ldots,N-1\}. \tag*{(9)}

Multiplying the left equation of (9) with ψN/cos⁡(θH)\psi_{N}/\cos(\theta_{H}) and adding it to (7) gives

∑m=1NBk,mϕm=∑m=1N−Ak,m(ψm−ψN)cos⁡(θH)for k∈{1,…,N−1}.(10)\sum_{m=1}^{N}B_{k,m}\phi_{m}=\sum_{m=1}^{N}\frac{-A_{k,m}(\psi_{m}-\psi_{N})}{\cos(\theta_{H})}\quad\text{for }k\in\{1,\ldots,N-1\}. \tag*{(10)}

The currents and the stream function are linked via (8) in Ref. 2 (see also (18) in Ref. 3),

Ik=ψk−ψk−1Rsheet.(11)I_{k}=\frac{\psi_{k}-\psi_{k-1}}{R_{\mathrm{sheet}}}. \tag*{(11)}

This means

I1=ψ1−ψNRsheetI1+I2=ψ2−ψNRsheet⋮I1+⋯+IN−1=ψN−1−ψNRsheetIN=0\begin{aligned} I_{1} &= \frac{\psi_{1}-\psi_{N}}{R_{\mathrm{sheet}}}\\ I_{1}+I_{2} &= \frac{\psi_{2}-\psi_{N}}{R_{\mathrm{sheet}}}\\ &\vdots\\ I_{1}+\cdots+I_{N-1} &= \frac{\psi_{N-1}-\psi_{N}}{R_{\mathrm{sheet}}}\\ I_{N} &= 0 \end{aligned}
L1I=1Rsheet(ψ1−ψNψ2−ψN⋮ψN−1−ψN0),L_{1}I=\frac{1}{R_{\mathrm{sheet}}} \begin{pmatrix} \psi_{1}-\psi_{N}\\ \psi_{2}-\psi_{N}\\ \vdots\\ \psi_{N-1}-\psi_{N}\\ 0 \end{pmatrix},

with

L1=(100⋯0110⋯0111⋯0⋮⋮⋮⋱⋮111⋯1).(12)L_{1}= \begin{pmatrix} 1&0&0&\cdots&0\\ 1&1&0&\cdots&0\\ 1&1&1&\cdots&0\\ \vdots&\vdots&\vdots&\ddots&\vdots\\ 1&1&1&\cdots&1 \end{pmatrix}. \tag*{(12)}

Inserting (12) into (10) gives in matrix form

Bϕ=−RsqAL1I=−RsqAL1 iGϕ.(13)B\phi=-R_{\mathrm{sq}}AL_{1}I=-R_{\mathrm{sq}}AL_{1}\,{}^{i}G\phi. \tag*{(13)}

This equation is valid for arbitrary voltage vectors ϕ\phi, therefore we can skip ϕ\phi on both sides of (13). Next we introduce the matrix

1^=(010⋯0001⋯0⋮⋮⋮⋱⋮000⋯1100⋯0).(14)\hat{1}= \begin{pmatrix} 0&1&0&\cdots&0\\ 0&0&1&\cdots&0\\ \vdots&\vdots&\vdots&\ddots&\vdots\\ 0&0&0&\cdots&1\\ 1&0&0&\cdots&0 \end{pmatrix}. \tag*{(14)}

1^\hat{1} is a circulant N×NN\times N matrix, which is obtained from the unit matrix 11 by shifting all entries up once, in a rolling way. It holds 1^−1=1^T\hat{1}^{-1}=\hat{1}^{T}, det⁡(1−1^)=0\det(1-\hat{1})=0, and det⁡(1−1^T)=0\det(1-\hat{1}^{T})=0. Multiplying (13) from right with (1−1^T)(1-\hat{1}^{T}) gives

B(1−1^T)=−RsqAL1 iG(1−1^T)=−RsqAL1(1−1^T) iG,(15)\begin{aligned} B(1-\hat{1}^{T})=-R_{\mathrm{sq}}AL_{1}\,{}^{i}G(1-\hat{1}^{T}) \\ &=-R_{\mathrm{sq}}AL_{1}(1-\hat{1}^{T})\,{}^{i}G, \tag*{(15)} \end{aligned}

whereby we used the fact that iG{}^{i}G and (1−1^T)(1-\hat{1}^{T}) are circulant matrices, and therefore they commute. Writing out the product in detail gives AL1(1−I^T)=A L_1(1-\hat{I}^{\mathrm{T}})=

A(10⋯011⋱011⋱⋮⋮⋮⋱⋱11⋯1)(10⋯0−1−11⋱000−1⋱00⋮⋱⋱1000⋱−11)=A(10⋯0−101⋱0−10⋱⋱⋱⋮0⋱⋱1−100⋯00)=A,(16)\begin{aligned} A\begin{pmatrix} 1 & 0 & \cdots& 0 \\ 1 & 1 & \ddots& 0 \\ 1 & 1 & \ddots& \vdots\\ \vdots& \vdots& \ddots& \ddots\\ 1 & 1 & \cdots& 1 \end{pmatrix}\begin{pmatrix} 1 & 0 & \cdots& 0 & -1 \\ -1 & 1 & \ddots& 0 & 0 \\ 0 & -1 & \ddots& 0 & 0 \\ \vdots& \ddots& \ddots& 1 & 0 \\ 0 & 0 & \ddots& -1 & 1 \end{pmatrix} \\ &= A\begin{pmatrix} 1 & 0 & \cdots& 0 & -1 \\ 0 & 1 & \ddots& 0 & -1 \\ 0 & \ddots& \ddots& \ddots& \vdots\\ 0 & \ddots& \ddots& 1 & -1 \\ 0 & 0 & \cdots& 0 & 0 \end{pmatrix}=A, \tag*{(16)} \end{aligned}

whereby we used (9) in the last equality of (16). Equation (16) is remarkable, because AL1(1−I^T)=AA L_1(1-\hat{I}^{\mathrm{T}})=A even though L1(1−I^T)≠1L_1(1-\hat{I}^{\mathrm{T}})\ne1. Inserting (16) into (15) gives

B(1−I^T)=−RsqAiGB(1-\hat{I}^{\mathrm{T}})=-R_{\mathrm{sq}}A^{\mathrm{i}}G

with (5) ⇒B(1−I^T)Q=−RsqAQiΓ\Rightarrow B(1-\hat{I}^{\mathrm{T}})Q=-R_{\mathrm{sq}}AQ^{\mathrm{i}}\Gamma.

It holds

(BI^TQ)j,m=∑k=1N∑ℓ=1NBj,kδ1+(ℓ mod N)Qℓ,m=∑ℓ=1NBj,1+(ℓ mod N)Qℓ,m=∑k=0N−1Bj,k+1Qk,m.(17)\begin{aligned} (B\hat{I}^{\mathrm{T}}Q)_{j,m}=\sum_{k=1}^{N}\sum_{\ell=1}^{N}B_{j,k}\delta_{1+(\ell\ {\rm mod}\ N)}Q_{\ell,m} \\ &=\sum_{\ell=1}^{N}B_{j,1+(\ell\ {\rm mod}\ N)}Q_{\ell,m}=\sum_{k=0}^{N-1}B_{j,k+1}Q_{k,m}. \tag*{(17)} \end{aligned}

Inserting this into (17) gives

iγmRsq∑k=1NAj,kQk,m=−∑k=1NBj,k(Qk,m−Qk−1,m).(18)\mathrm{i}\gamma_m R_{\mathrm{sq}}\sum_{k=1}^{N}A_{j,k}Q_{k,m}=-\sum_{k=1}^{N}B_{j,k}(Q_{k,m}-Q_{k-1,m}). \tag*{(18)}
iγm=−2χiRsq(2−χ)exp⁡(−iπmN)sin⁡(πmN)×∫01(1+icot⁡(π2χτ))exp⁡(iπχmNτ)[cot⁡(π2χ(1−τ))−cot⁡(π2χ)](1/2+θH/π) dτ∫01(1−icot⁡(π2(2−χ)τ))exp⁡(−iπ(2−χ)mNτ)[cot⁡(π2χ)−cot⁡(π2(χ+τ(2−χ)))](1/2+θH/π) dτ.(19)\begin{aligned} \mathrm{i}\gamma_m=\frac{-2\chi\mathrm{i}}{R_{\mathrm{sq}}(2-\chi)}\exp\left(-\mathrm{i}\pi\frac{m}{N}\right)\sin\left(\frac{\pi m}{N}\right) \\ &\times\frac{\int_0^1(1+\mathrm{i}\cot(\frac{\pi}{2}\chi\tau))\exp(\mathrm{i}\pi\chi\frac{m}{N}\tau)\left[\cot(\frac{\pi}{2}\chi(1-\tau))-\cot(\frac{\pi}{2}\chi)\right]^{(1/2+\theta_{\mathrm{H}}/\pi)}\,\mathrm{d}\tau}{\int_0^1(1-\mathrm{i}\cot(\frac{\pi}{2}(2-\chi)\tau))\exp(-\mathrm{i}\pi(2-\chi)\frac{m}{N}\tau)\left[\cot(\frac{\pi}{2}\chi)-\cot(\frac{\pi}{2}(\chi+\tau(2-\chi)))\right]^{(1/2+\theta_{\mathrm{H}}/\pi)}\,\mathrm{d}\tau}. \tag*{(19)} \end{aligned}

The properties of these eigenvalues will be discussed elsewhere. They can be used in (5) to compute the conductance matrix, which gives all electrical properties of the Hall-plate (impedances, response to magnetic field, thermal noise, signal-to-thermal-noise ratio over power).

Inserting (4) into (8) and using the product formula3 gives

h(τ)=sin⁡(χπ2)∣cot⁡(χπ2)−cot⁡(Nτ2)∣(1/2+θH/π).(20)h(\tau)=\sin\left(\frac{\chi\pi}{2}\right)\left|\cot\left(\frac{\chi\pi}{2}\right)-\cot\left(\frac{N\tau}{2}\right)\right|^{(1/2+\theta_{\mathrm{H}}/\pi)}. \tag*{(20)}

Back-inserting this into (8) gives after some manipulation

Ak,m=−2N(2−χ)sin⁡(χπ2)×∫0π/2∣cot⁡(χπ2)−cot⁡(χπ2+z(2−χ))∣(1/2+θH/π)sin⁡(((2−χ)z/N+πm/N))dzsin⁡((2−χ)z/N+π(m−k)/N).(21)\begin{aligned} A_{k,m}=-\frac{2}{N}(2-\chi)\sin\left(\frac{\chi\pi}{2}\right) \\ &\times\int_0^{\pi/2}\frac{\left|\cot\left(\frac{\chi\pi}{2}\right)-\cot\left(\frac{\chi\pi}{2}+z(2-\chi)\right)\right|^{(1/2+\theta_{\mathrm{H}}/\pi)}}{\sin(((2-\chi)z/N+\pi m/N))} \frac{\mathrm{d}z}{\sin((2-\chi)z/N+\pi(m-k)/N)}. \tag*{(21)} \end{aligned}

and

Bk,m=−2Nχsin⁡(χπ2)×∫0π/2∣cot⁡(χπ2)−cot⁡(χz)∣(1/2+θH/π)sin⁡(χ(2z−π)/(2N)+πm/N)dzsin⁡(χ(2z−π)/(2N)+π(m−k)/N).(22)\begin{aligned} B_{k,m}=-\frac{2}{N}\chi\sin\left(\frac{\chi\pi}{2}\right) \\ &\times\int_0^{\pi/2}\frac{\left|\cot\left(\frac{\chi\pi}{2}\right)-\cot(\chi z)\right|^{(1/2+\theta_{\mathrm{H}}/\pi)}}{\sin(\chi(2z-\pi)/(2N)+\pi m/N)} \frac{\mathrm{d}z}{\sin(\chi(2z-\pi)/(2N)+\pi(m-k)/N)}. \tag*{(22)} \end{aligned}

We insert (21), (21) into (19) and use the following identity (see Appendix)

∑k=1Nexp⁡(2πikm/N)sin⁡(x/N+πk/N)sin⁡(x/N+π(k−j)/N)=2Nsin⁡(πjm/N)sin⁡(πj/N)(icot⁡(x)−1)exp⁡(imπ−2xN).(23)\sum_{k=1}^{N}\frac{\exp(2\pi\mathrm{i}km/N)}{\sin(x/N+\pi k/N)\sin(x/N+\pi(k-j)/N)} =2N\frac{\sin(\pi jm/N)}{\sin(\pi j/N)}(\mathrm{i}\cot(x)-1)\exp\left(\mathrm{i}m\frac{\pi-2x}{N}\right). \tag*{(23)}

Finally, this gives the complex eigenvalues iγm,m∈{1,2,…,N}\mathrm{i}\gamma_{m,m}\in\{1,2,\ldots,N\} of the indefinite conductance matrix iG^{\mathrm{i}}G of a regular symmetric Hall-plate with NN contacts as a function of the Hall-angle θH\theta_{\mathrm{H}} and of the contact size χ\chi,

How to compute the sum in (23)

We compute the sum

∑k=1Nexp⁡(2πikm/N)sin⁡(x/N+πk/N)sin⁡(x/N+π(k−j)/N),(24)\sum_{k=1}^{N}\frac{\exp(2\pi\mathrm{i}km/N)}{\sin(x/N+\pi k/N)\sin(x/N+\pi(k-j)/N)}, \tag*{(24)}

for N≥2N\ge2, j∈{1,2,…,N−1}j\in\{1,2,\ldots,N-1\}, and integer mm. First we replace k=Nk=N by k=0k=0, then we express the sines by complex exponentials,

∑k=0N−1(2i)2exp⁡(2πikm/N)exp⁡(i(x/N+πk/N))exp⁡(i(x/N+π(k−j)/N))×[1−exp⁡(−2ix+πkN)]−1×[1−exp⁡(−2ix+π(k−j)N)]−1.(25)\sum_{k=0}^{N-1}\frac{(2\mathrm{i})^2\exp(2\pi\mathrm{i}km/N)}{\exp(\mathrm{i}(x/N+\pi k/N))\exp(\mathrm{i}(x/N+\pi(k-j)/N))} \times\left[1-\exp\left(-2\mathrm{i}\frac{x+\pi k}{N}\right)\right]^{-1} \times\left[1-\exp\left(-2\mathrm{i}\frac{x+\pi(k-j)}{N}\right)\right]^{-1}. \tag*{(25)}

Then we make twice a Taylor series expansion of the type [1−z]−1=1+z+z2+z3+⋯[1-z]^{-1}=1+z+z^2+z^3+\cdots. This gives

∑ℓ=0∞∑q=0∞−4exp⁡(−2ix1+ℓ+qN)exp⁡(iπj1+2qN)∑k=0N−1exp⁡(2πikm−1−ℓ−qN).(26)\sum_{\ell=0}^{\infty}\sum_{q=0}^{\infty}-4\exp\left(-2\mathrm{i}x\frac{1+\ell+q}{N}\right)\exp\left(\mathrm{i}\pi j\frac{1+2q}{N}\right) \sum_{k=0}^{N-1}\exp\left(2\pi\mathrm{i}k\frac{m-1-\ell-q}{N}\right). \tag*{(26)}

The sum over index kk vanishes, except for m−1−ℓ−q=−pNm-1-\ell-q=-pN with p=0,1,2,…p=0,1,2,\ldots, hence it equals Nδq,m−1−ℓ+pNN\delta_{q,m-1-\ell+pN}. We re-arrange the double sum over indices ℓ,q\ell,q as a double sum over indices a,ba,b with a=ℓ−q,b=ℓ+qa=\ell-q,b=\ell+q (see Figure 2). With

Transformation of indices for a double sum

Figure 2. Transformation of indices for a double sum.

b=m−1+pNb=m-1+pN it holds

∑p=0∞−4Nexp⁡[i(πj−2x)m+pNN]∑a=−(m−1+pN)Δa=2m−1+pNexp⁡(−iπjaN),(27)\sum_{p=0}^{\infty}-4N\exp\left[\mathrm{i}(\pi j-2x)\frac{m+pN}{N}\right] \sum_{\substack{a=-(m-1+pN)\\ \Delta a=2}}^{m-1+pN}\exp\left(\frac{-\mathrm{i}\pi ja}{N}\right), \tag*{(27)}

whereby the index aa is incremented by Δa=2\Delta a=2. We replace the index aa by cc via a=2c−m+1−pNa=2c-m+1-pN. This gives

∑a=−(m−1+pN)Δa=2m−1+pNexp⁡(−iπjaN)=exp⁡(iπjm−1+pNN)∑c=0m−1+pNexp⁡(−2πijNc)=(−1)jpsin⁡(πjm/N)sin⁡(πj/N).(28)\begin{aligned} \sum_{\substack{a=-(m-1+pN)\\ \Delta a=2}}^{m-1+pN}\exp\left(\frac{-\mathrm{i}\pi ja}{N}\right) =\exp\left(\mathrm{i}\pi j\frac{m-1+pN}{N}\right) \sum_{c=0}^{m-1+pN}\exp\left(-2\pi\mathrm{i}\frac{j}{N}c\right) \\ &=(-1)^{jp}\frac{\sin(\pi jm/N)}{\sin(\pi j/N)}. \tag*{(28)} \end{aligned}

Inserting (A.5) into (A.4) and computing the infinite sum over index pp finally gives

2Nsin⁡(πjm/N)sin⁡(πj/N)(icot⁡(x)−1)exp⁡(imjπ−2xN).(29)2N\frac{\sin(\pi jm/N)}{\sin(\pi j/N)}(\mathrm{i}\cot(x)-1)\exp\left(\mathrm{i}m\frac{j\pi-2x}{N}\right). \tag*{(29)}

References

  1. [1]1H.J. Lippmann, F. Kuhrt, "Der Geometrieeinfluss auf den transversalen magnetischen Widerstandseffekt bei rechteckförmigen Halbleiterplatten," Zeitschrift für Naturforschung A 13, 462 (1958). (in German)
  2. [2]2D. Homontcovschi, B.T. Murray, "Explicit resistance matrix for a Hall disk with multiple peripheral contacts: Application to a van der Pauw type method for extended contacts," Sensors and Actuators A: Physical 294, 1 (2019).
  3. [3]3U. Ausscherlner, "The classical Hall effect in multiply-connected plane regions part I: topologies with stream function," Journal of Applied Mathematics and Physics 7(9), 1968 (2019).
  4. [4]4R.M. Gray, "Toeplitz and circulant matrices: A review," Foundations and Trends in Communications and Information Theory 3(2), 155 (2006).DOI
  5. [5]5I.M. Ryshik, I.S. Gradstein, Tables of Series, Products, and Integrals, (VEB Deutscher Verlag der Wissenschaften Berlin, 1957), page 34, equation 1.392.

Paper details

vlnc:W_u4r2e7za
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