REVIEW 2 major objections 5 minor 31 references
Line defects in infinite networks of resistors
T0 review · 2 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read This paper derives exact Green's-function formulas for the equivalent resistance between arbitrary nodes of an infinite square resistor lattice with periodic line defects, reducing the problem to one-dimensional integrals.
desk verdict A clean, exact Sherman-Morrison treatment of periodic line defects in resistor networks; the main formulas are new and sound, with only minor presentation issues to fix. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is the lattice Green's function in reciprocal space together with the Sherman-Morrison identity. For a defect invariant along one direction, Fourier transformation in that direction block-diagonalizes the Laplacian; each block is the perfect-lattice operator plus a rank-one perturbation of the type $|u\rangle\langle v|$, so the inverse (the Green's function) is updated by the algebraic Sherman-Morrison formula rather than by infinite sums. The alternating parallel defect, whose block operator is a sum of dyads rather than a single dyad, is handled by the Sherman-Morrison-Woodbury identity with an $n\times n$ matrix. This converts an infinite perturbation into a one-dimensional integral whose integrand is built from $G_0(k_x,y_1,y_2) = e^{-|y_1-y_2|s}/(2\sinh s)$, the exactly known propagator of the perfect strip, with $\cosh s = 2 - \cos k_x$.
What would settle it
Solve the full Kirchhoff equations numerically for a large finite square lattice with periodic boundary conditions and a tilted defect of slope $m/n$ in its original integer coordinates, then take the thermodynamic limit; if the resulting two-point resistances deviate from the one-dimensional integral obtained from Eqs. (69)-(74) beyond the finite-size extrapolation error, the relabeling step is not faithful. A simpler check is to verify that the modified Laplacian of Eq. (71) has the same bond set as the original tilted-defect network after relabeling.
Extended reading notes
Core claim
The central claim is that a periodic line defect in an infinite square resistor network leaves the problem exactly solvable in the same sense as the perfect lattice. Writing the modified resistors as a perturbation of the discrete Laplacian, the paper derives explicit Green's-function formulas—Eq. (29) for parallel defects, Eqs. (50)-(51) for perpendicular defects, Eq. (69) for alternating perpendicular defects, and Eq. (A.6) with the Woodbury identity for alternating parallel defects—and obtains the equivalent resistance as a one-dimensional Brillouin-zone integral. For special parameters (short-circuited or removed defect lines, or nodes on the defect) the integral collapses to finite combinations of known perfect-lattice resistances, giving exact numbers such as $R(1,0;0,0)=1+1/\pi$ for a removed parallel line and $R(0,1;0,1)=2/\pi$ for a removed perpendicular line. The method extends to tilted defects of rational slope $m/n$ by an integral whose kernel uses $\cos(k_x - m k_y/n)$, and to arbitrary lattices and complex impedances because only translational symmetry along one direction is needed.
Load-bearing premise
The load-bearing premise is that the coordinate relabeling used for tilted defects, which renames site $(x,y)$ as $(x, y - mx/n)$ so that the defect becomes horizontal, faithfully describes the original network even though the new vertical labels are not integers; if that mapping changes which bonds are connected, the tilted-defect results fail. The rank-one form of the perturbation also assumes the single-resistor substitution-current relation of Eq. (12).
Editorial extensions
If this is right
- For any of the treated defect geometries, the equivalent resistance between arbitrary nodes can be evaluated by a single one-dimensional integral, making boundary-adjacent resistances computable with elementary numerical quadrature rather than large linear solves.
- Special limits (removed lines with $g=1$, short-circuited lines with $g\to-\infty$, and halved resistances with $g=-1$) reduce the defect problem to finite combinations of perfect-lattice resistances, giving exact closed forms such as those in Tables 2 and 3.
- The alternating perpendicular defect interpolates between a full cut ($n=1$, resistance diverges) and a single missing resistor ($n\to\infty$, resistance tends to $R$), so the method quantifies how the delocalization of a defect changes transport.
- Because only one direction of translational symmetry must survive, the same Sherman-Morrison construction applies to other two-dimensional lattices, three-dimensional lattices, and networks with complex impedances, opening the way to topolectrical-circuit boundaries.
Reading between the lines
- The tilted-defect integral likely inherits the logarithmic large-distance growth of the perfect lattice, so line defects of rational slope should not change the qualitative conductance scaling, only the prefactor; this is an extrapolation from the paper's asymptotic remarks.
- The same framework could be applied to semi-infinite or kinked defects by stitching together several line defects, although the paper notes that breaking translation invariance in two directions lies outside its method.
- A direct experimental check in an electrical-impedance network or a topolectrical circuit could measure two-point resistances near a line of modified capacitors or inductors and compare with Eq. (29) by sweeping frequency-dependent complex $g$.
- The equivalence between a short-circuited perpendicular defect and a halved parallel defect, expressed in Eqs. (56)-(57), suggests a general duality that may extend to alternating or tilted defects, though the paper only proves it for the specific cases in Section 4.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper develops an exact Green's-function framework for computing the equivalent resistance between arbitrary nodes in an infinite square resistor network containing a periodic line defect. The defect is represented as a rank-one or finite-rank perturbation of the lattice Laplacian, and the Sherman-Morrison/Woodbury identities are applied in a partially Fourier-transformed basis, reducing the problem to one-dimensional integrals. The cases treated are: horizontal defects on horizontal bonds (parallel, Section 3), horizontal defects on vertical bonds (perpendicular, Section 4), alternating perpendicular defects with period n (Section 5), tilted defects of rational slope (Section 6), and alternating parallel defects (Appendix A). The paper also derives closed-form special-case results, duality relations among configurations, and numerical examples. The central formulas are Eqs. (29), (50)-(51), (69), and (A.6).
Significance. If the framework is correct, this is a useful and nontrivial extension of the classical perfect-lattice resistor problem to extended line defects. The Sherman-Morrison reductions are internally consistent, the dualities (e.g., perpendicular g -> -infinity versus parallel g = -1; perpendicular g = 1/2 versus parallel g = 1) are physically motivated and cross-checked, and the special-case tables provide concrete, falsifiable values. The method is modular and plausibly transferable to other lattices and to networks with complex impedances, although those generalizations are only sketched. The strengths are the systematic derivation of integral representations, the exact treatment of the perturbation without truncation, and the independent numerical checks in Section 5.1. The main weaknesses are that some closed-form expressions are not stated unambiguously and that the tilted-defect reduction is too compressed; these issues require attention but do not appear to invalidate the central framework.
major comments (2)
- [Sec. 3.2, Eqs. (42)-(43)] As reproduced in the manuscript, the closed forms for R(1,0;0,0) and R(0,1;0,0) do not reproduce Table 2 or the special values in Eqs. (37) and (41). For example, using the second branch of Eq. (43) one obtains F_parallel(1) = 3/2, and Eq. (42) then gives R(1,0;0,0) = 0 and R(0,1;0,0) about 1.818, whereas Eq. (37) and Table 2 give 1 + 1/pi and 1 - 1/pi, respectively. Likewise, for g = -1 the displayed expression gives zero instead of 1/pi. Since Section 3.2 is a load-bearing claim about exact special-case evaluations, the formulas or the definition of F_parallel must be corrected and stated unambiguously so that Table 2 and Eqs. (37)/(41) are reproduced.
- [Sec. 6, Eqs. (71)-(73)] The tilted-defect reduction is stated very tersely. The relabeling |x,y> -> |x, y - m x/n> maps the lattice to coordinates with non-integer y labels, and the manuscript does not explicitly show that the perturbed links in the new basis are exactly the alternating perpendicular defect of Eq. (60), nor that the sheared plane waves in Eq. (72) form a complete basis over the required Brillouin zone. Since the tilted-defect results and their use of Eq. (69) depend entirely on this equivalence, this missing justification is load-bearing and should be supplied. The statement that the relabeling breaks down at n = 0 should also be reconciled with the claim that Appendix A covers this case.
minor comments (5)
- [Eq. (18)] The block operator L_1(k_x) should read 2g(1 - cos k_x)|0><0|; as printed the factor 2g is missing, which conflicts with Eq. (22) and Eq. (24).
- [Sec. 3.1] In the enumerated list of special g values, item (iii) says g = 1 corresponds to 'short-circuited' resistors along the defect line; the following paragraph and Eq. (34) correctly state that g = 1 corresponds to r -> infinity, i.e., removed resistors. The list entry should be corrected.
- [Sec. 4.2, Eq. (58)] At g = 1/2, the displayed expression for R(0,1;1,1) contains terms that diverge as (1-2g)^-2; since Table 3 gives a finite value at this point, the cancellation (which relies on F_perp(1/2) = 1/2) should be stated explicitly so readers are not left with an apparent pole.
- [Sec. 5.1 and Table 4] The numerical values in Table 4 are quoted to three decimals without error estimates; please specify the numerical accuracy used in the integration and in the independent finite-removal check.
- [Abstract and Conclusions] The statements that the method 'readily extends' to other lattice geometries, three dimensions, and general complex impedances are plausible but are not demonstrated in the manuscript; please label these as outlook or provide the explicit generalization for at least one additional lattice.
Circularity Check
No significant circularity: the Sherman–Morrison derivation is self-contained; same-group citations appear only as standard ingredients or as a numerical consistency check.
full rationale
The derivation chain is genuinely first-principles rather than circular. The perfect-lattice Green's function (Eq. 10) is a standard external result, and the perturbed Green's functions in Eqs. (24), (48), (69), and (A.6) are obtained by applying the Sherman–Morrison/Woodbury identities to explicit block operators derived from Ohm's and Kirchhoff's laws via Eq. (12). No target resistance is assumed as an input, and no fitted parameter is renamed as a prediction. The special-g reductions (e.g., Eqs. (36), (54), and the dualities in Eqs. (56)–(57)) follow by algebraic identities and symmetry/image arguments, and they reduce to the known perfect-lattice values from Table 1. The tilted-defect treatment in Section 6 is an exact relabeling: the map |x,y> -> |x, y - m x/n> is bijective, and Eq. (72) is the exact Fourier kernel of the relabeled Laplacian, with a determinant-one shear in momentum space; no approximation or hidden assumption is introduced. The n=0 case is explicitly acknowledged and deferred to Appendix A. The only same-group citations are the known single-resistor perturbation [22], the general perturbation framework [23] used for an independent numerical cross-check in Section 5.1, and a remark about the invertibility of the Laplacian; none of these is load-bearing for the central claim. Under the stated rules, these self-citations do not constitute circularity, and the score therefore is minimal.
Assumptions & free parameters
assumptions (5)
- standard math Sherman-Morrison and Sherman-Morrison-Woodbury identities are valid for the block operators used here.
- standard math The perfect square lattice Green's function G0(kx,y1,y2) = e^(-|y1-y2|s)/(2 sinh s) with cosh s = 2 - cos kx correctly represents the infinite lattice.
- domain assumption A changed resistor can be represented exactly by injected currents delta I = (r - R)/(r R) times the endpoint potential difference.
- domain assumption For tilted defects, the relabeling |x,y> -> |x, y - m x/n> is a valid basis transformation even though the new y labels are non-integer.
- domain assumption The perturbation preserves translational symmetry along at least one direction, so a partial Fourier transform diagonalizes the problem.
Cite this review
Pith. "Pith review of Line defects in infinite networks of resistors." pith.science (2026). https://pith.science/paper/TPNWPCYH
@misc{pith2026250908445,
author = {Pith},
title = {Pith review of: Line defects in infinite networks of resistors},
year = {2026},
howpublished = {\url{https://pith.science/paper/TPNWPCYH}},
note = {Machine review of arXiv:2509.08445}
}
read the original abstract
We study infinite resistor networks perturbed by line defects, in which the resistances are periodically modified along a single line. Using the Sherman-Morrison identity applied to the reciprocal-space representation of the lattice Green's function, we develop a general analytical framework for computing the equivalent resistance between arbitrary nodes. The resulting expression is a one-dimensional integral that is evaluated exactly in special cases. While our analysis is carried out for the square lattice, the method readily extends to other lattice geometries and networks with general impedances. Therefore, this framework is useful for studying the boundary behavior of topolectrical circuits, which serve as classical analogs of topological insulators.
Figures
Figures from the paper (6 more)
Reference graph
Works this paper leans on
-
[1]
der Pol B V and Bremmer H 1950Operational Calculus Based on the Two-Sided Laplace Integral (USA: AMS Chelsea Publishing)
-
[2]
Aitchison R E 1964American Journal of Physics32566–566 ISSN 0002-9505 (Preprint https://pubs.aip.org/aapt/ajp/article-pdf/32/7/566/11448403/566_1_online.pdf) URLhttps://doi.org/10.1119/1.1970777
-
[3]
Gn¨ adig P, Honyek G and Riley K F 2001200 Puzzling Physics Problems: With Hints and Solutions (Cambridge University Press)
-
[4]
Cserti J 2000American Journal of Physics68896–906 ISSN 0002-9505 (Preprinthttps://pubs. aip.org/aapt/ajp/article-pdf/68/10/896/7529081/896_1_online.pdf) URLhttps:// doi.org/10.1119/1.1285881
-
[5]
Bhattacharjee S and Ramola K 2023Phys. Rev. E108(4) 044148 URLhttps://link.aps.org/ doi/10.1103/PhysRevE.108.044148
-
[6]
Cserti J, Sz´ echenyi G and D´ avid G 2011Journal of Physics A: Mathematical and Theoretical44 215201 URLhttps://dx.doi.org/10.1088/1751-8113/44/21/215201
-
[7]
Owaidat M, Asad J and Tan Z Z 2019Results in Physics121621–1627 ISSN 2211-3797 URL https://www.sciencedirect.com/science/article/pii/S2211379718321557
-
[8]
Asad J H 2013Journal of Statistical Physics1501177–1182 ISSN 1572-9613 URLhttps: //doi.org/10.1007/s10955-013-0716-x
Show all 31 references
-
[9]
Tan Z Z 2022The European Physical Journal Plus137546 ISSN 2190-5444 URLhttps: //doi.org/10.1140/epjp/s13360-022-02750-3
-
[10]
Essam J W and Wu F Y 2008Journal of Physics A: Mathematical and Theoretical42025205 URLhttps://dx.doi.org/10.1088/1751-8113/42/2/025205
-
[11]
Izmailian N S and Huang M C 2010Phys. Rev. E82(1) 011125 URLhttps://link.aps.org/ doi/10.1103/PhysRevE.82.011125
-
[12]
Doyle P G and Snell J L 1984Random Walks and Electric Networks(Washington, DC: Mathematical Association of America)
-
[13]
org/10.1007/BF01046996
Tetali P 1991Journal of Theoretical Probability4101–109 ISSN 1572-9230 URLhttps://doi. org/10.1007/BF01046996
-
[14]
Chandra A K, Raghavan P, Ruzzo W L, Smolensky R and Tiwari P 1996computational complexity 6312–340 ISSN 1420-8954 URLhttps://doi.org/10.1007/BF01270385 Line defects in infinite networks of resistors28
-
[15]
Boksiner J and Leath P L 2003Phys. Rev. E67(6) 066610 URLhttps://link.aps.org/doi/ 10.1103/PhysRevE.67.066610
-
[16]
Gavrilchenko T and Katifori E 2019Phys. Rev. E99(1) 012321 URLhttps://link.aps.org/ doi/10.1103/PhysRevE.99.012321
-
[17]
Wang Y, Yang F, Zhang Z and Zhao Y 2021The Journal of Physical Chemistry C12516334–16342 ISSN 1932-7447 URLhttps://doi.org/10.1021/acs.jpcc.1c04832
1932 doi
-
[18]
Molenaar C G, Leusink D P, Wang X L and Brinkman A 2014Superconductor Science and Technology27104003 URLhttps://dx.doi.org/10.1088/0953-2048/27/10/104003
-
[19]
Lee C H, Imhof S, Berger C, Bayer F, Brehm J, Molenkamp L W, Kiessling T and Thomale R 2018Communications Physics139 ISSN 2399-3650 URLhttps://doi.org/10.1038/ s42005-018-0035-2
-
[20]
Dong J, Juriˇ ci´ c V and Roy B 2021Phys. Rev. Res.3(2) 023056 URLhttps://link.aps.org/ doi/10.1103/PhysRevResearch.3.023056
-
[21]
0265293/20503221/021503_1_5.0265293.pdf) URLhttps://doi.org/10.1063/5.0265293
Sahin H, Jalil M B A and Lee C H 2025APL Electronic Devices1021503 ISSN 2995-8423 (Preprinthttps://pubs.aip.org/aip/aed/article-pdf/doi/10.1063/5. 0265293/20503221/021503_1_5.0265293.pdf) URLhttps://doi.org/10.1063/5.0265293
-
[22]
Cserti J, D´ avid G and Pir´ oth A 2002American Journal of Physics70153–159 ISSN 0002- 9505 (Preprinthttps://pubs.aip.org/aapt/ajp/article-pdf/70/2/153/7421027/153_1_ online.pdf) URLhttps://doi.org/10.1119/1.1419104
-
[23]
Cserti J and D´ avid G 2025 General theory of perturbation of infinite resistor networks (Preprint 2506.18654) URLhttps://arxiv.org/abs/2506.18654
2025
-
[24]
Press W H, Teukolsky S A, Vetterling W T and Flannery B P 2007Numerical Recipes 3rd Edition: The Art of Scientific Computing3rd ed (USA: Cambridge University Press) ISBN 0521880688
-
[25]
1214/aoms/1177730946
Guttman L 1946The Annals of Mathematical Statistics17336–343 URLhttps://doi.org/10. 1214/aoms/1177730946
-
[26]
42, Statistical Research Group(Princeton Univ.) p 4
Woodbury M A 1950 Inverting modified matricesMemorandum Rept. 42, Statistical Research Group(Princeton Univ.) p 4
1950
-
[27]
Sherman J and Morrison W J 1950The Annals of Mathematical Statistics21124–127 URL https://doi.org/10.1214/aoms/1177729893
-
[28]
Venezian G 1994American Journal of Physics621000–1004 ISSN 0002-9505 (Preprinthttps: //pubs.aip.org/aapt/ajp/article-pdf/62/11/1000/11816228/1000_1_online.pdf) URL https://doi.org/10.1119/1.17696
-
[29]
Atkinson D and van Steenwijk F J 1999American Journal of Physics67486–492 ISSN 0002- 9505 (Preprinthttps://pubs.aip.org/aapt/ajp/article-pdf/67/6/486/7529301/486_1_ online.pdf) URLhttps://doi.org/10.1119/1.19311
-
[30]
Bairamkulov R and Friedman E G 2019IEEE Transactions on Circuits and Systems I: Regular Papers664368–4376
-
[31]
Bairamkulov R and Friedman E G 2020IEEE Transactions on Circuits and Systems I: Regular Papers673224–3233
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.