WorksPhysics and Astronomy
Scaling laws of entanglement entropy in fractal dimensions
Abstract
Entanglement entropy scaling links geometry to quantum correlations and diagnoses quantum criticality. Fractal lattices extend this link to noninteger dimensions, where boundary and bulk dimensions are independent. It remains open which dimension governs entropy scaling and how strongly gap closing can enhance it. Nonlocal occupied-state projectors and the lack of translation symmetry obstruct standard methods. We overcome these obstacles and classify free-fermion entropy scaling with bounds set by the hopping across the subsystem boundary: an upper bound from its rank and the spectral gap, and a lower bound from how much it lowers the ground-state energy. With only negative-energy states filled and this energy lowering proportional to the boundary, the entanglement entropy obeys the boundary law under a uniform gap and satisfies when the gap closes no faster than algebraically. Here is the subsystem linear size and the boundary dimension. Half occupation of the chiral zero modes gives a mixed-state entropy with , where is the local zero-mode weight; an extensive thus makes follow the bulk dimension.
Introduction
**Introduction—**Entanglement entropy (EE) scaling links spatial geometry to quantum correlations and provides a probe of critical behavior [1, 2]. For example, logarithmic EE scaling [3] and the entanglement spectrum [4] have been used to diagnose critical topology in experiments. For a subsystem of linear size in a periodic -dimensional Euclidean lattice, the EE of gapped free fermions obeys the boundary-law bound , whereas a codimension-one Fermi surface gives [5, 6, 7]. The rigorous gapless result relies on Widom asymptotics, regular geometry, and translation symmetry [8]; localization instead restores a boundary law for disordered free fermions [9]. Beyond ground states, highly excited fermionic eigenstates and generic random states can exhibit volume-law entanglement [10, 11].
Fractal lattices have been realized experimentally [12, 13, 14] and offer a setting for exploring EE. Their self-similar structure can modify conventional boundary-law behavior [15, 16], but the resulting scaling cannot be inferred from the fractal dimension alone. The bulk scales as , while an entangling boundary can scale independently as , where and are the bulk and boundary dimensions. In particular, unlike on Euclidean lattices, need not equal [Fig. 1]. Field theory can treat noninteger dimensions through analytic continuation to spacetime dimension , allowing calculations of EE and related sphere free energies [17, 18].
However, this continuation does not incorporate self-similar lattices. Numerical correlation-matrix calculations can directly evaluate EE on finite fractal lattices [19, 20], but the rapid growth of the number of sites with generation limits the accessible sizes, so finite-size data cannot settle the asymptotic scaling. It therefore remains open which dimension governs EE scaling on fractal lattices and how gap closing can enhance it [Fig. 1b].
Answering these questions requires bounds on the occupied-state projector restricted to . The full projector can extend across the system even for finite-range hopping. Without translation symmetry, the standard Euclidean asymptotic methods do not directly apply [6, 8].
In this Letter, we overcome these obstacles and establish a classification of free-fermion entropy scaling on fractal lattices. We bound the entropy using only the hopping across the subsystem boundary and the energy gap. The upper bound scales with the number of independent hopping channels and grows only logarithmically as the gap closes. The lower bound grows with how much this hopping lowers the ground-state energy. Neither bound requires translation symmetry. With only negative-energy states filled and this energy lowering proportional to the boundary, these bounds give for a uniform occupied–unoccupied gap (L1) and for at most algebraic gap closure (L2), with size-independent . Separately, for the grand canonical state at zero temperature and chemical potential, the subsystem entropy obeys the gap-independent bound (L3), where is the local zero-mode weight. These results answer both questions: the boundary dimension governs the pure-state EE, algebraic gap closing adds at most one logarithm, and an extensive zero-mode weight makes the bulk dimension govern . We verify L1–L3 on the Sierpiński carpet and Vicsek lattices.

Figure 1. Entanglement entropy scaling in Euclidean and self-similar lattices. Subsystem has linear size and complement . a, On a periodic -dimensional lattice, gapped free-fermion ground states satisfy , while a codimension-one Fermi surface gives . b, A self-similar lattice with Hausdorff dimension . The red question mark denotes the exponent to be determined; square brackets indicate possible logarithmic enhancement. L1–L3 address this scaling problem through conditions on the cut, spectrum, and occupation.
**Entanglement entropy bounds—**For a pure number-conserving Gaussian state, let be its single-particle correlation projector and the projector onto all local orbitals in subsystem . The boundary between and its complement defines the cut, with . The eigenvalues of the restricted correlation matrix determine the EE [21, 22, 23]
Let be a single-particle Hamiltonian with and . Thus is the largest single-particle energy magnitude. The cut hopping matrix defines the energy gain . Let contain the geometric bonds crossing the cut, each counted once, and let . For distinct sites, define the density anticorrelation , where is the total particle-number operator at site and expectations are taken in the Gaussian state. Then
For pure negative filling, is the energy cost of removing cut hopping while keeping the state fixed. A boundary lower bound follows if either of the following averages has a positive, size-independent lower bound:
Here , and and are independent of system size; the energy route also requires a size-independent upper bound on . If , either condition yields
with independent of size. Neither lower bound requires a spectral gap.
The upper bound depends on the cut hopping rank and the spectral separation. If the occupied and unoccupied energies lie in and , respectively, the block satisfies a Sylvester equation with displacement rank at most . Applying the singular-value estimate of Beckermann and Townsend [24] gives
Here bounds the occupied–unoccupied separation from below at generation and may decrease with . Derivations are given in Secs. S1B–S1E of [25].
Eqs. (3), (6) provide the common basis of L1 and L2. For Hamiltonians with uniformly bounded and , either condition in Eq. (3) fixes a positive boundary lower bound; the gap dependence of Eq. (6) then controls the possible enhancement. A uniform gap gives a boundary law, while at most algebraic gap closure permits at most a logarithmic enhancement. L3 changes the zero-mode occupation and uses a separate comparison bound.
Models and cuts
**Models and cuts—**To realize these conditions, we consider two quadratic tight-binding models on finite bipartite graph approximants , with nearest-neighbor bonds joining opposite sublattices. The first is [19]
where creates a spinless fermion, is real, and includes both bond orientations. In the basis, the single-particle matrix has the chiral form and anticommutes with . Hence zero energies occur in pairs and [26]. For a nonempty nonzero spectrum, define
We call the nonzero spectrum uniformly gapped when and gapless when . Throughout, denotes the EE under pure negative filling, while denotes the mixed-state subsystem entropy under independent half-occupation of the zero modes. We set .
For an explicitly gapped reference, we use the two-orbital Hamiltonian
Here contains the - and -orbital operators, and and act in orbital space. The chiral operator of Eq. (9) is , which acts within each physical site. For every eigenvalue of the graph adjacency matrix, ; therefore guarantees a graph-independent half-filling gap of at least . The ground state fills the negative-energy band. We take and .
We test both Hamiltonians on graph approximants of the Sierpiński carpet and Vicsek fractal. At every iteration, each retained square is divided into a array. The Sierpiński carpet retains all eight subsquares except the center [Figure 2a], whereas the Vicsek fractal construction retains the center and four corners [Figure 2e]. At generation , we place one lattice site at the center of every smallest retained square and join neighboring centers—edge-sharing cells for the Sierpiński carpet and corner-sharing cells for the Vicsek fractal—to obtain the graphs in Figure 2b and Figure 2f. Taking the smallest-cell side as the unit length, the overall side length is , and the total site number is for the Sierpiński carpet and for the Vicsek lattice. Their Hausdorff dimensions are therefore and , respectively.

Figure 2. L1: Negative filling with a uniform nonzero-energy gap. Panels a–c and e–g show generation . a,e, Sierpiński carpet and Vicsek fractal geometries. b,f, Corresponding lattices with cuts I (orange), II (blue), and III (green). c,g, Spectra of Eq. (9) with and ; occupied (empty) states are black (gray). d,h, EE for the same cuts, with matching colors. Points show in d and in h. Dashed lines are power-law fits over and , respectively.
With the lower-left corner as the origin and the site coordinates, define three subsystems: , ,
and . Their boundaries are labeled I–III in Figure 2b and Figure 2f. All three linear sizes are proportional to ; we use , since fixed rescalings leave the exponents unchanged.
Negative-filling entropy with a uniform nonzero-energy gap
Consider any Hamiltonian with uniformly bounded and . A uniform gap turns off, in Eq. (6), into . Either lower-bound condition in Eq. (3) then gives the boundary law . When , it follows that .
Under these conditions, for finite-range Hamiltonians on bounded-degree graphs with a fixed number of orbitals per site, convergence of the boundary content and local boundary-pattern frequencies further gives
The coefficient conditions and proof are given in Secs. S1 G–S1 H of [25].
For a concrete realization, the two-orbital model in Eq. (9) supplies both the gap condition above: for fixed , and , where is a generation-independent upper bound on the vertex degree. It therefore obeys the boundary law, providing an analytical basis for the gapped Sierpiński carpet scaling previously inferred numerically [19]. The model verification is given in Sec. S1 F of [25].
Figure 2d gives fitted exponents for Sierpiński carpet cuts I–III, consistent with @page 3/6 starts_mid=0 ends_mid=1 , where . The Vicsek fractal fits are , consistent with [Fig. 2h]: the two bottleneck cuts approach saturation, while the third grows with its boundary. These results show how boundary geometry selects the EE exponent for the same gapped Hamiltonian. Further model tests and fit diagnostics are given in Secs. S1 I–S1 K of [25].
L2: Negative-filling entropy under gap closure— Retain the uniform bounds on and , but let with for fixed and . This condition includes the algebraic gap scaling familiar from finite-size scaling at continuous quantum transitions [27]. Combining Eq. (6) with either condition in Eq. (3) yields
for sufficiently large generations, with size-independent . These model-independent bounds constrain the allowed scaling but leave the leading logarithmic factor undetermined; the derivation is given in Sec. S2 A of [25].
The single-orbital model in Eq. (7) on the uniform Sierpiński carpet provides a sharper application. We prove and both conditions in Eq. (3) for cuts I–III. Combining exact dimer elimination with local correlation estimates and convergent boundary-pattern frequencies establishes the asymptotic boundary law
Here is independent of generation and depends on the cut; the boundary exponents are . Thus the gapless carpet realizes a boundary law within the general L2 bounds, with a definite leading coefficient and no multiplicative logarithmic enhancement. Dimer elimination maps the carpet onto a hierarchy of even cycles. Each cycle contributes at most a logarithm of its length, but the number of long cycles crossing the cut per boundary bond decreases as a power of their length, so the total remains proportional to the boundary. Proofs are given in Secs. S2 B–S2 D of [25].
Figure 3a shows the pure negative-filling spectrum. Fits to the EE in Figure 3b over give finite-size effective exponents for cuts I–III. The excess over comes from a negative subleading correction to the boundary law, whose relative size falls by about half per generation. The asymptotic boundary exponents and the absence of a leading logarithm follow from the analytical argument above. Detailed finite-size and fitting-window analyses are given in Sec. S2 E of [25].

Figure 3. L2: Negative filling on the gapless Sierpiński carpet. a, Spectrum of Eq. (6) at . Negative-energy states (black) are filled; zero- and positive-energy states (gray) are empty. b, EE for cuts I–III of Figure 2b, with matching colors. Points show ; dashed lines are power-law fits over .
L3: Mixed-state entropy from half-occupied zero modes— The third part of the classification changes the occupation of exact zero modes. At fixed generation, the zero-temperature grand canonical state at zero chemical potential independently occupies each exact zero mode with probability , forming a statistical mixture of many-body states with different particle numbers. With and , its correlation operator is . Define the local zero-mode weight . For any Hamiltonian
with site-local chiral symmetry, the pure negative-filling EE and mixed-state subsystem entropy satisfy
This bound holds at every finite generation, independently of the spectral gap. It classifies the mixed entropy according to the relative growth of and :
The term is the local weight of thermodynamic occupation entropy; the derivation is given in Sec. S3 A of [25]. When and , the last regime yields
where is independent of generation but may depend on the cut [Sec. S3 D of [25]. For the nearest-neighbor Vicsek fractal in Figure 4, recursive counting gives and , with for cuts I–III. A tree spectral bound gives , verifying both conditions and fixing . Fits over give finite-size effective exponents [Figure 4b], below the analytically established bulk exponent . Proofs, fit diagnostics, and further model tests are given in Secs. S3 E–S3 G of [25].

Figure 4. L3: Mixed half filling on the Vicsek lattice. a, Spectrum of Eq. (6) at . Negative-energy states (black) are filled, zero modes (dark gray) are independently occupied with probability , and positive-energy states (light gray) are empty. b, Mixed-state subsystem entropy for cuts I–III of Figure 2f, with matching colors. Points show ; dashed lines are power-law fits over .
**Conclusion—**Our classification provides an analytical framework for comparing entropy scaling across fractal lattices, cuts, and particle occupations. By relating nonlocal entropy to boundary couplings, spectral gaps, and local zero-mode weight, we derive scaling laws across free-fermion Hamiltonians without relying on translation symmetry. This approach establishes infinite-size behavior beyond numerically accessible generations, providing a physical basis for interpreting finite-size entropy data.
Future work should identify the conditions under which gap closure produces a leading logarithmic enhancement of entanglement entropy. For mixed states, a further goal is to quantify quantum correlations independently of thermodynamic occupation entropy. Nonbipartite models without chiral symmetry offer a direct test of the role of symmetry in entropy scaling [28]. Random fractals provide a setting to study how localization competes with cut geometry [20], while multifractal geometries raise the question of whether a single boundary exponent suffices to describe the entropy. In three-dimensional fractals, localization transitions open further questions about how the spatial structure of critical eigenstates shapes EE and whether it produces distinct scaling laws [29].
**Acknowledgments—**This work is supported by Singapore National Research Foundation Thematic Competitive Research Program under Grant No. NRF-T-CRP-2025-0004, and Singapore Ministry of Education Academic Research Fund Tier 2 under Grant No. MOE-T2EP50123-0007, and Tier 1 Grant RG176/25. Baile Zhang acknowledges support from Nanyang Technological University. Yihao Yang is supported by Key Research and Development Program of the Ministry of Science and Technology under Grant No. 2022YFA1404900, 2022YFA1405200, National Natural Science Foundation of China (NSFC) under Grants U25D8017, and Class D for Young Scientific Research Peak Creation No. K20250230.
We used Qiushi Discovery Engine [30] and OpenAI Codex to assist mathematical derivations.
**Data availability—**The numerical codes and data in this study have been deposited in the Nanyang Technological University database under accession code [insert upon publication].
tests on the Sierpiński carpet, Vicsek fractal, right-isosceles Sierpiński gasket lattice, and equilateral Sierpiński gasket lattice. The Supplemental Material also includes Refs. [31, 32, 33, 34, 35, 36, 37, 38, 39, 40].
References
- [1]G. Vidal, J. I. Latorre, E. Rico, and A. Kitaev, Entanglement in quantum critical phenomena, Phys. Rev. Lett. 90, 227902 (2003).pubmed.ncbi.nlm.nih.gov/12857342
- [2]J. Eisert, M. Cramer, and M. B. Plenio, Colloquium: Area laws for the entanglement entropy, Rev. Mod. Phys. 82, 277 (2010).DOI
- [3]Z. Cheng, X. Zhang, X.-J. Yu, L. Zhou, J. Gong, and B. Zhang, Observation of critical topological phase transition, Nature 658, 358 (2026).DOI
- [4]Z.-K. Lin, L.-W. Wang, Z.-L. Kong, Y. Zhou, H.-Q. Lin, X. Yu, S. Zhang, and J.-H. Jiang, Experimental observation of critical topology, Nature 658, 365 (2026).DOI
- [5]M. M. Wolf, Violation of the entropic area law for fermions, Phys. Rev. Lett. 96, 010404 (2006).pubmed.ncbi.nlm.nih.gov/16486427
- [6]D. Gioev and I. Klich, Entanglement entropy of fermions in any dimension and the Widom conjecture, Phys. Rev. Lett. 96, 100503 (2006).pubmed.ncbi.nlm.nih.gov/16605718
- [7]B. Swingle, Entanglement entropy and the Fermi surface, Phys. Rev. Lett. 105, 050502 (2010).pubmed.ncbi.nlm.nih.gov/20867902
- [8]H. Leschke, A. V. Sobolev, and W. Spitzer, Scaling of Rényi entanglement entropies of the free Fermi-gas ground state: A rigorous proof, Phys. Rev. Lett. 112, 160403 (2014).
- [9]A. Elgart, L. Pastur, and M. Shcherbina, Large block properties of the entanglement entropy of free disordered fermions, J. Stat. Phys. 166, 1092 (2017).DOI
- [10]D. N. Page, Average entropy of a subsystem, Phys. Rev. Lett. 71, 1291 (1993).DOI
- [11]L. Vidmar, L. Hackl, E. Bianchi, and M. Rigol, Entanglement entropy of eigenstates of quadratic fermionic hamiltonians, Phys. Rev. Lett. 119, 020601 (2017).DOI
- [12]S. N. Kempkes, M. R. Slot, S. E. Freeney, S. J. M. Zevenhuizen, D. Vanmaekelbergh, I. Swart, and C. Morais Smith, Design and characterization of electrons in a fractal geometry, Nat. Phys. 15, 127 (2019).DOI
- [13]X.-Y. Xu, X.-W. Wang, D.-Y. Chen, C. Morais Smith, and X.-M. Jin, Quantum transport in fractal networks, Nat. Photonics 15, 703 (2021).DOI
- [14]T. Biesenthal, L. J. Maczewsky, Z. Yang, M. Kremer, M. Segev, A. Szameit, and M. Heinrich, Fractal photonic topological insulators, Science 376, 1114 (2022).
- [15]A. Hamma, D. A. Lidar, and S. Severini, Entanglement and area law with a fractal boundary in a topologically ordered phase, Phys. Rev. A 81, 010102(R) (2010).DOI
- [16]X. Li, M. C. Jha, and A. E. B. Nielsen, Laughlin topology on fractal lattices without area-law entanglement, Phys. Rev. B 105, 085152 (2022).DOI
- [17]M. A. Metlitski, C. A. Fuertes, and S. Sachdev, Entanglement entropy in the O(N) model, Phys. Rev. B 80, 115122 (2009).
- [18]L. Fei, S. Giombi, I. R. Klebanov, and G. Tarnopolsky, Generalized F-theorem and the ε expansion, J. High Energy Phys. 2015 (12), 155.
- [19]Y. Zhou and P. Ye, Entanglement fractalization, Phys. Rev. Research 6, 043145 (2024).DOI
- [20]A. Kosior and J. Wang, Entanglement entropy of free fermions on random fractal lattices (2026), arXiv:2607.05611 [cond-mat.stat-mech].arxiv.org/abs/2607.05611
- [21]I. Peschel, Calculation of reduced density matrices from correlation functions, J. Phys. A: Math. Gen. 36, L205 (2003).DOI
- [22]I. Peschel and V. Eisler, Reduced density matrices and entanglement entropy in free lattice models, J. Phys. A: Math. Theor. 42, 504003 (2009).DOI
- [23]L. Zhou, Entanglement spectrum and entropy in Floquet topological matter, Phys. Rev. Research 4, 043164 (2022).DOI
- [24]B. Beckermann and A. Townsend, On the singular values of matrices with displacement structure, SIAM J. Matrix Anal. Appl. 38, 1227 (2017).arxiv.org/abs/1609.09494
- [25]See Supplementary Materials at [URL will be inserted by publisher] for the conditions and proofs of the entropy bounds and scaling laws in L1–L3, spectral-gap and zero-mode analyses, and numerical methods, fitting procedures, and additional tests on the Sierpiński carpet, Vicsek fractal, right-isosceles Sierpiński gasket lattice, and equilateral Sierpiński gasket lattice. The Supplemental Material also includes Refs. [31–40].
- [26]M. Mui, S. A. Trugman, and E. Abrahams, Unusual properties of midband states in systems with off-diagonal disorder, Phys. Rev. B 49, 3190 (1994).
- [27]M. Campostrini, A. Pelissetto, and E. Vicari, Finite-size scaling at quantum transitions, Phys. Rev. B 89, 094516 (2014).DOI
- [28]M. Brzezińska, A. M. Cook, and T. Neupert, Topology in the Sierpiński-Hofstadter problem, Phys. Rev. B 98, 205116 (2018).DOI
- [29]T. Li, X. Tang, S. Liu, and H. Hu, Anderson transition and mobility edges in a family of 3D fractal lattices (2026), arXiv:2605.17953 [cond-mat.dis-nn].arxiv.org/abs/2605.17953
- [30]S. Yang, F. Chen, R. Zhao, J. Wu, Y. Wang, H. Luo, N. Han, Q. Chen, Y. Hu, W. Li, M. Li, H. Chen, and Y. Yang, End-to-end autonomous scientific discovery on a real optical platform (2026), arXiv:2604.27092.arxiv.org/abs/2604.27092
- [31]A. S. Lewis, Derivatives of spectral functions, Math. Oper. Res. 21, 576 (1996).DOI
- [32]R. Bhatia, Matrix Analysis, Graduate Texts in Mathematics, Vol. 169 (Springer, New York, 1997).
- [33]I. Klich, An elementary derivation of Levitov’s formula, in Quantum Noise in Mesoscopic Physics, NATO Science Series, Vol. 97, edited by Y. V. Nazarov (Springer, Dordrecht, 2003) pp. 397–402.DOI
- [34]I. Klich, Lower entropy bounds and particle number fluctuations in a Fermi sea, J. Phys. A: Math. Gen. 39, L285 (2006).
- [35]N. J. Higham, Functions of Matrices: Theory and Computation (Society for Industrial and Applied Mathematics, Philadelphia, 2008).
- [36]A. E. Brouwer and W. H. Haemers, Spectra of Graphs, Universitex (Springer, New York, 2012).
- [37]M. Benzi, P. Boito, and N. Razouk, Decay properties of spectral projectors with applications to electronic structure, SIAM Rev. 55, 3 (2013).arxiv.org/abs/1203.3953
- [38]L. N. Trefethen, Approximation Theory and Approximation Practice (Society for Industrial and Applied Mathematics, Philadelphia, 2013).
- [39]J. M. Conde-Alonso, A. M. González-Pérez, J. Parcet, and E. Tablate, Schur multipliers in Schatten-von Neumann classes, Ann. Math. 198, 1229 (2023).
- [40]X. Li, X. Liang, and Y. Xue, Geodesics of right isosceles Sierpiński gaskets and their relatives, Chaos Solitons Fractals 192, 115937 (2025).DOI