REVIEW 3 major objections 5 minor 32 references
Layer-Number-Controlled Symmetry Breaking and Surface-State Transport in Rhombohedral Graphene Multilayers
T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read Rhombohedral graphene stacks of four to six layers show that layer number controls which insulating or metallic phase appears at charge neutrality: the field for the layer-polarized transition rises with thickness, and six-layer samples…
desk verdict Genuinely new layer-dependent phase diagram in rhombohedral graphene, with a plausible but not fully proven surface-state interpretation for r-6LG; worth a serious referee, but the biggest gap is ruling out a classical Drude origin for the B = 0 resistive peak. 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 load-bearing object is the set of low-energy surface states in rhombohedral (ABC-stacked) multilayer graphene, whose kinetic energy scales as $E_k\propto p^N$; in thicker stacks these states carry an increasing share of the low-energy density of states and screen the external displacement field before it reaches interior layers. The argument is carried by a self-consistent electrostatic tight-binding model, with a discrete Gauss-law relation coupling layer potentials to layer charge densities and an out-of-plane dielectric constant; it yields layer-resolved potentials, a screening factor $S = 1 - \Delta U_{SC}/\Delta U_{unscreened}$, and the critical field for the layer-polarized transition. In transport, the decisive signatures are adjacent-gate-controlled Landau levels and the fully lifted 24-fold degeneracy of the zeroth Landau level in hexalayer graphene, whose valley–layer locking allows the authors to assign each quantum-Hall state a flavor (filling factor, spin, valley, layer).
What would settle it
Measure the same r-6LG semimetal phase in devices with mobility comparable to the best r-4LG and r-5LG devices and with contacts swapped or made symmetric: if the adjacent-gate-controlled peak and Landau levels persist, the screening interpretation is supported; if they vanish or track the contact asymmetry, the claim fails. A quantitative version is to fit $\rho_{xx}(D)$ at $B=0$ in r-6LG with the two-band Drude model using measured carrier densities and mobilities as functions of $D$; if the resistive peak is fully reproduced by the density–mobility competition alone, no additional surface-screening mechanism is needed.
Extended reading notes
Core claim
The central claim is that layer number $N$ tunes the balance between electron-electron interactions and interlayer screening in rhombohedral multilayer graphene, producing two unconventional effects. First, the critical displacement field $D_c$ for the semimetal-to-layer-polarized-insulator transition increases monotonically from tetralayer to hexalayer (and beyond), contradicting the simplest unscreened two-surface-band model in which the gap should grow with thickness. Second, in hexalayer graphene a surface-state-dominated transport regime appears in the semimetal phase: a resistivity peak and, in a magnetic field, Landau levels that are controlled almost entirely by the gate on the same side as the surface state. At high field the 24-fold degenerate zeroth Landau level of the six-layer system is fully resolved, with valley locked to layer, and states with counter-propagating edge modes on the two surfaces become dissipative—attributed to interlayer backscattering. The paper explains both effects with self-consistent tight-binding calculations in which the external field is progressively screened by charge redistribution into surface-derived bands, with a screening factor exceeding 0.94 for six layers.
Load-bearing premise
The strongest assumption is that the contrast between six-layer and thinner devices—the adjacent-gate-controlled resistive peak and Landau levels—reflects intrinsic layer-number-dependent screening, not device-to-device differences in mobility, contacts, or inhomogeneity; the six-layer device has notably lower mobility, and a Drude competition between carrier density and mobility can already produce a resistivity peak.
Editorial extensions
If this is right
- Layer number becomes a design knob: increasing $N$ suppresses the layer-antiferromagnetic phase beyond six layers and pushes the layer-polarized-insulator transition to higher displacement fields, so phase diagrams can be tailored by choosing thickness.
- Hexalayer graphene's decoupled surface states act as a gate-tunable electron–hole bilayer: each surface's Landau levels respond to the adjacent gate, allowing selective population of one surface.
- At high field, interlayer backscattering between counter-propagating edge modes can be switched on or off by the displacement field, giving electrically controlled dissipative versus quantized Hall states.
- The high-field Landau-level crossing patterns provide a reliable way to identify the layer number of a rhombohedral graphene sample.
- The screening-driven rise in $D_c$ means thicker rhombohedral stacks require larger external fields to reach layer-polarized insulating states, a practical constraint and opportunity for device design.
Reading between the lines
- The same screening mechanism would predict that even thicker rhombohedral stacks (seven, eight layers) show yet stronger surface decoupling; the paper's own data hint that the layer-polarized onset moves to fields beyond the measured range, which is a direct testable extension.
- If adjacent-gate control is intrinsic, then local gates could address individual surfaces independently, opening a route to electrically writable surface-state devices and valley-selective contacts without twist engineering.
- A cleaner test of the screening story would be a direct measurement of the layer-resolved electrostatic potential (for example by scanning probe) across $N=4,5,6$ at fixed displacement field, compared against the calculated screening factor.
- The dissipative states from interlayer backscattering suggest that at fractional fillings the decoupled surfaces may host interlayer coherent or fractional states, though that goes beyond the data presented here.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports dual-gated electrical transport measurements on rhombohedral graphene with four, five, and six layers (plus one eight-layer device in the Supplementary Information) and maps the zero-magnetic-field phases at charge neutrality as a function of displacement field. It claims that the critical displacement field for the layer-antiferromagnet to semimetal transition is approximately constant from tetralayer to hexalayer graphene, whereas the critical displacement field for the semimetal to layer-polarized-insulator transition increases with layer number, and it attributes this to enhanced interlayer screening quantified by a self-consistent tight-binding calculation. The paper further claims that hexalayer graphene shows surface-state-dominated transport, evidenced by a resistive peak and Landau levels that respond selectively to the adjacent gate, with high-field quantum Hall data interpreted through layer-valley-locked Landau levels and dissipative states possibly arising from interlayer backscattering.
Significance. If the claims hold, the paper makes a valuable contribution by establishing layer number as a practical tuning parameter for correlated and topological phases in rhombohedral multilayer graphene. The experimental strengths include multi-temperature phase-boundary measurements, data from multiple pentalayer devices, and a self-consistent electrostatic tight-binding calculation that uses standard Slonczewski-Weiss-McClure parameters and is not fitted to the measured critical fields. The high-field Landau-level crossing patterns are also compared with explicit schematics, which is a useful diagnostic for layer counting. The main weakness is that the surface-state interpretation of the zero-field resistive peak in r-6LG is not quantitatively separated from a classical Drude mechanism that is acknowledged in the Supplementary Information; because that peak is central to the abstract and to the screening narrative, this ambiguity is load-bearing.
major comments (3)
- [Sec. II.B, Fig. 3, SI Sec. 7] The resistive peak in the semimetal phase of r-6LG is presented as evidence for surface-state-dominated transport, but SI Sec. 7 states that within a simple Drude picture the competition between decreasing carrier density and increasing mobility with displacement field naturally leads to a resistivity maximum. The manuscript does not provide a quantitative calculation of the Drude rho_xx(D) using the measured n(D), p(D), mu_n(D), and mu_p(D) extracted in SI Sec. 4, nor does it demonstrate that the observed peak position, width, or magnitude disagrees with that classical prediction. Without this separation, the B = 0 peak cannot be uniquely attributed to surface states; indeed, the same mechanism could produce a layer-number-dependent peak because SI Sec. 4 reports that r-6LG has much larger carrier densities and substantially lower mobilities than r-4LG and r-5LG. This point is load-bearing for the abstract claim of surface-state-dominated transport.
- [Sec. II.B, Fig. 3b-d] The single-gate-controlled Landau-level features are demonstrated in one r-6LG device only, and the contrast with r-4LG and r-5LG devices is not controlled for device quality. SI Sec. 4 reports that r-6LG has substantially lower mobility than the thinner devices, so gate-specific contact effects or local inhomogeneity in this particular device could in principle produce similar-looking features. A quantitative comparison of device quality across the devices used for the contrast (for example, mobility, contact resistance, and residual charge inhomogeneity), or data from an additional r-6LG device, is needed to support the layer-number-dependent interpretation.
- [Sec. II.A, Fig. 2c] The central scaling trend of the semimetal-to-LPI critical field with layer number is based on one r-4LG and one r-6LG device in the main text, with additional r-5LG devices in the Supplementary Information, and no uncertainty estimates are given for the critical-field values. Because the phase boundaries are identified from crossing points of temperature-dependent resistivity traces, the authors should provide error bars or an explicit statement of the systematic uncertainty in identifying these crossings for each layer number; the significance of the claimed monotonic trend depends on these uncertainties.
minor comments (5)
- [SI Sec. 4] The sentence reporting carrier densities of approximately 10^12, 10^10, and 10^11 should state the units (cm^-2) and should include the fitting uncertainty in the extracted densities and mobilities.
- [References 18 and 39] References 18 and 39 contain placeholder DOIs (10.1103/chsq-ndzs and 10.1103/j1zf-v3j5) that appear not to be valid Crossref identifiers; these should be corrected before publication.
- [Fig. 3c,d] The line cuts in Fig. 3c,d are not fully self-explanatory; the figure should state which color corresponds to which magnetic field and which gate is swept, along with the fixed values of the other gate.
- [SI Sec. 2] The statement that LPI states are not observed in r-8LG should be presented in Fig. 2c as an upper limit or an unmeasured boundary rather than as a data point, so that the phase diagram is not misleading.
- [Methods, Eqs. (5)-(6)] The notation n_b and n_t is used for gate charge densities while n_j denotes layer charge densities; this is understandable but potentially confusing, and a brief definition of the subscript convention would improve clarity.
Circularity Check
No significant circularity; the central claims are experimental and the screening calculation uses fixed literature parameters rather than fitted targets.
full rationale
The central observations—layer-dependent Dc values, the emergence of resistive peaks and gate-controlled Landau levels in r-6LG—are measured transport data, not outputs of the theory. The self-consistent electrostatic tight-binding calculation (Methods Eqs. 1–7) uses standard Slonczewski-Weiss-McClure parameters and a fixed out-of-plane dielectric constant, and the calculated Dc shown in the inset of Fig. 2c is a model output compared with, rather than fitted to, the experimental Dc values. The resistive peak in the semimetal phase is explained in SI Sec. 7 via a Drude competition between density and mobility, but the paper does not label that explanation as a prediction of the model; the surface-state interpretation is an inference from the gate dependence of the peak and Landau levels, not a tautological restatement of the model's inputs. Self-citations to prior reports of layer-antiferromagnetic states in rhombohedral tetralayer and pentalayer graphene support sample identification but are not load-bearing for the new layer-number-dependent phase-boundary trend. No equation is shown to reduce to its own input, and no fitted parameter is renamed as a prediction. Therefore the derivation chain is self-contained and no circular step can be identified and quoted.
Assumptions & free parameters
assumptions (5)
- domain assumption Layer number and rhombohedral stacking are correctly identified by optical contrast and near-field infrared microscopy, and the specific devices are free of moiré alignment with the hBN encapsulant.
- domain assumption The phase boundaries at charge neutrality can be extracted from temperature-independent crossing points in the D-dependent resistivity curves.
- domain assumption The self-consistent tight-binding model with Slonczewski-Weiss-McClure parameters and out-of-plane dielectric constant ε⊥=4 is an adequate description of the electrostatics and band structure of rhombohedral multilayer graphene.
- domain assumption In the r-6LG surface-state interpretation, the resistive peak and the adjacent-gate-controlled Landau levels originate from intrinsic surface states and strong interlayer screening rather than from device-specific artifacts such as contact resistance or inhomogeneous carrier density.
- standard math The Landau level assignment in Fig. 4 assumes valley-layer locking and the 4N-fold degeneracy of the lowest Landau level in rhombohedral multilayer graphene, as given by chiral decomposition theory.
Cite this review
Pith. "Pith review of Layer-Number-Controlled Symmetry Breaking and Surface-State Transport in Rhombohedral Graphene Multilayers." pith.science (2026). https://pith.science/paper/4IOAOSER
@misc{pith2026260811450,
author = {Pith},
title = {Pith review of: Layer-Number-Controlled Symmetry Breaking and Surface-State Transport in Rhombohedral Graphene Multilayers},
year = {2026},
howpublished = {\url{https://pith.science/paper/4IOAOSER}},
note = {Machine review of arXiv:2608.11450}
}
abstract
Rhombohedral multilayer graphene hosts layer-polarized flat bands, providing an intriguing platform for correlated and topological electronic states; however, the role of layer number in governing symmetry breaking and surface screening remains elusive. Here we prepare rhombohedral graphene multilayers and systematically conduct electrical transport measurements. We uncover an unconventional layer dependence of phase transitions: the critical displacement field (D$_{c}$) for the layer-antiferromagnetic (LAF)-to-semimetal transitions remains constant across tetralayer to hexalayer graphene, whereas the D$_{c}$ for semimetal-to-layer-polarized-insulator (LPI) transition increases with layer number, defying unscreened Coulomb interaction models. In hexalayer graphene, surface-state-dominated transport emerges, with Landau levels (LLs) and resistive peaks selectively controlled by adjacent gates, a signature of strong interlayer screening absent in thinner stacks. High magnetic fields reveal valley-layer-locked LLs and dissipative states possibly from interlayer backscattering, highlighting the presence of decoupled surface states. Our findings establish layer number as a key tuning knob for engineering correlated and topological phases in rhombohedral graphene multilayers.
Figures
Reference graph
Works this paper leans on
-
[1]
LAF stability vs. layer thickness: While Coulomb interactions strengthen with N, the characteristic displacement-field scale associated with the LAF phase reaches a maximum at N = 6 and 7 and decreases in thicker layers. This behavior suggests that band reconstruction (e.g., trigonal warping) outweighs interaction effects beyond a critical thickness
-
[2]
Screening-band-overlapping-engineered LPI: The increase in Dc for the LPI transitions with N underscores the combined role of finite band overlap and enhanced electronic screening, with the latter increasingly associated with surface-derived low energy states in thicker multilayers
-
[3]
Surface states in the semimetal phase: Surface states tuned predominantly by the adjacent gate voltage are observed , which are likely responsible to the enhance d screening in the semimetal phase of r-6LG. In addition, t he gate -tunable surface states in r-6LG provide a platform for designer quantum Hall bilayers, where interlayer backscattering can be ...
-
[4]
Figure 1b shows a typical r-6LG device (r-6LG-D1) in a Hall bar geometry achieved by standard e-beam lithography, reactive ion etching and metal deposition. Figure 1. Rhombohedral 4, 5, 6-layer graphene. a, Schematic of the dual-gated devices. b, Optical image of the r-6LG device in Hall bar geometry. Scale bar, 3 μm. c-e, Schematic structures of r-4LG (c...
-
[5]
Sample fabrication Graphene and hBN flakes are mechanically exfoliated onto SiO2 (285 nm)/Si substrates, and the layer numbers are identified by optical contrast and atomic force microscope ( AFM) topographic characterizations. The rhombohedral graphene domain is identified using scanning near-field infrared microscope and isolated in-situ using the AFM c...
-
[6]
Transport measurements The devices are measured in a 1.5 K-base-temperature Oxford variable -temperature insert system. Stanford Research Systems SR830 /860 and Guangzhou Sine Scientific Instrument OE1201 lock-in amplifiers with an alternating current of 5 nA-10 nA at a frequency of 17.77 Hz in combination with a 100 M Ω resistor are used to measure the r...
-
[7]
Self-consistent electrostatic tight-binding calculations We evaluate the self -consistent electrostatic tight -binding model in a low -energy continuum form expanded around the 𝐾 valley. In the layer -sublattice basis 𝛹𝐤= (𝜓𝐴1,𝐤,𝜓𝐵1,𝐤,𝜓𝐴2,𝐤,𝜓𝐵2,𝐤,…,𝜓𝐴𝑁,𝐤,𝜓𝐵𝑁,𝐤) 𝑇 , the single-particle Hamiltonian is written as 𝐻̂=∑𝛹𝐤 † 𝐤 ℋ𝐾 (𝑁)(𝐤)𝛹𝐤. (1) Here 𝐤=(𝑘𝑥,𝑘𝑦) is...
arXiv 2007
-
[8]
Additional colormaps of r-5LG, r-6LG, and r-8LG devices at B = 0T
Show all 32 references
-
[9]
D-dependent 𝝆𝐱𝐱 of two additional r-5LG and a r-8LG for n = 0 at varied temperatures
-
[10]
Temperature dependent behavior of LAF phase for r-4-6LG and r-8LG
-
[11]
Analysis on magneto-transport in the semimetal phase
-
[12]
Layer-number and displacement-field-dependent LPI gap
-
[13]
Quantitative Analysis of Electrostatic Screening in Rhombohedral Multilayer Graphene
-
[14]
Resistive peak in the semimetal phase of r-6LG
-
[15]
Isospin polarized metals in r-6LG
-
[16]
Absence of single-gate-controlled behaviors in r-4LG and r-5LG
-
[17]
Crossover of the single-gate-controlled and dual-gate-controlled behaviors
-
[18]
Dissipative and non-dissipative quantum Hall states at B = 12T
-
[19]
LL crossing patterns of r-6LG and r-8LG at high magnetic fields
-
[20]
In addition to devices shown in Figure 1, more devices are fabricated and characterized (Fig
Additional colormaps of r-5LG, r-6LG, and r-8LG devices at B = 0T. In addition to devices shown in Figure 1, more devices are fabricated and characterized (Fig. S1). For the example of r-5LG (Fig. S1(a), S1(b), and F ig. 1(g)), all the devices have shown consistent critical Ds...
-
[21]
Figure S1
are routinely taken for confirmation the absence of moiré-related resistance peaks. Figure S1. Additional colormaps of r-5LG (a and b), r-6LG (c and e), and r-8LG (d and g) devices at B = 0T. Zoomed-in maps (a-d); Large-range maps (e-g)
-
[22]
In addition to the data of pentalayer sample ( r-5LG-D2) shown in F ig
D-dependent 𝝆𝐱𝐱 of two additional r-5LG and a r-8LG for n = 0 at varied temperatures. In addition to the data of pentalayer sample ( r-5LG-D2) shown in F ig. 2 (a), the other two pentalayer devices also exhibit critical D values (Fig. S2), that agree well with existing data, w...
-
[23]
Temperature dependent behavior of LAF phase for r-4-6LG and r-8LG. To understand better the layer -dependence of LAF states, temperature dependent transport behaviors at the charge neutrality n = 0 of r-4-6LG and r-8LG are discussed in this section. Owing to the correlated nat...
-
[24]
Analysis on magneto-transport in the semimetal phase Positive magnetoresistances are observed in semimetal phases in r-4-6LG. For in -depth understanding of layer-dependent magneto-transport in the semimetal phase , we extract electron and hole densities and their mobilities f...
-
[25]
As shown in F ig
Layer-number and displacement-field-dependent LPI gap Gap size Δ of the LPI state can be extracted from the fitting of linear regime of the Arrhenius plot. As shown in F ig. S5, for 4L to 6L, Δ all exhibits linear dependence on D, expected for tunable nature of LPI by out-of-p...
-
[26]
Quantitative Analysis of Electrostatic Screening in Rhombohedral Multilayer Graphene To further elucidate the origin of the enhanced screening discussed in the main text, we analyze the self-consistent electrostatic response of rhombohedral multilayer graphene under an applied...
-
[27]
For the understanding of this peak, both carrier densities and mobilities of electrons and holes need to be taken into consideration
Resistive peak in the semimetal phase of r-6LG Interestingly, a resistance peak is observed as a function of D, in the SM phase. For the understanding of this peak, both carrier densities and mobilities of electrons and holes need to be taken into consideration. Within the SM ...
-
[28]
SVP”: spin -and-valley-polarized state. “SP
Isospin polarized metals in r-6LG In contrast to the semimetal phases where single-gate dependent resistive features are observed (Fig. S9), spin-and-valley-polarized phase (SVP), spin-polarized phase (SP), and unpolarized phase (UP) are observed in r-6LG at relatively higher ...
-
[29]
For example, at B = 1T (Fig
Absence of single-gate-controlled behaviors in r-4LG and r-5LG In contrast to the r-6LG, surface-state-dominated electron transport behaviors are not observed in r-4LG and r-5LG at different magnetic-field regimes. For example, at B = 1T (Fig. S11(a)), although Landau Level s ...
-
[30]
As increasing the magnetic field, a crossover between the two regimes is expected to be observed
Crossover of the single-gate-controlled and dual-gate-controlled behaviors At B = 4T and B = 12T, single-gate-controlled LL features and dual-gate-controlled quantum Hall states are observed, respectively. As increasing the magnetic field, a crossover between the two regimes i...
-
[31]
0/3K” state is non -dissipative, whereas its neighboring “−1K′/4K
Dissipative and non-dissipative quantum Hall states at B = 12T Dissipative and non-dissipative quantum Hall states are observed in both Rxx (Fig. S13(a)) and Rxy data (Fig. S13(b)). Indeed, in the interlayer backscattering regime, Rxy does not correspond to the total filling f...
-
[32]
For r-6LG and r-8LG, the total filling factors of their lowest LL should be ±12, and ±16, respectively
LL crossing patterns of r-6LG and r-8LG at high magnetic fields For a N layer rhombohedral graphene, along with the valley and spin degrees of freedom, the zeroth LL contains additional quantum numbers associated with the LL orbital index 𝑛=0,1,2…N-1.7-9 The zero-energy LL is ...
2025
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.