{"id":"061a48e4-cf81-44c7-96c7-c95ed8f2e254","arxiv_id":"2505.01252","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":0.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"MnBi2Te4 and its relatives are the best-studied intrinsic magnetic topological insulators, and this review maps their properties, open puzzles, and device prospects.","lead":"This paper is a short review of five years of research on the MnBi2Te4 family of intrinsic magnetic topological insulators. It consolidates the experimental and theoretical state of the art for a materials class that could enable dissipationless electronics and axion electrodynamics.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Mn-Bi intermixing explanation for the Dirac-gap puzzle rests on top-SL-only DFT and is explicitly conceded by the authors to need more realistic models; this makes the review's roadmap conditional, though not its primary claim.","rationale":"Read in good faith, this is a competent anniversary review by leading contributors. It explicitly identifies its scope, presents contradictory data, and flags unresolved questions. The central claim (first intrinsic MTI) is supported by independent magnetic characterization, neutron diffraction, transport (QAH, zero-plateau QAH), and multiple ARPES studies; the review does not manufacture a new result. I considered whether the sample-dependent gap contradicts the AFM TI identification, but the paper handles this by presenting it as an open problem, and the AFM TI state itself is not disproved by a gapless surface in defected samples. The only truly load-bearing soft spot is the mechanism assigned to the gap puzzle. The authors' own sentence limits the intermixing scenario, and the conflicting STM correlation studies mean the review's outlook is more speculative than the prose around it implies. Because the limitation is disclosed and the primary claim does not depend on resolving the gap puzzle, I would not change the ACCEPT verdict, though the authors could usefully soften the 'suppressing intermixing' statement in the Outlook. This is a minor editorial adjustment rather than a correctness failure, hence UNCHANGED.","tokens_in":21964,"tokens_out":4662,"duration_ms":51602,"concrete_test":"Recompute the (0001) surface Dirac gap with slab supercells containing Mn-Bi antisites distributed over the top two or three septuple layers (and, separately, with a realistic random concentration), following the methods of Refs. [67,68], and compare the resulting gap distribution with the laser-ARPES compilation in Fig. 2h. If the gap remains large when deeper layers are intermixed, or if an alternative structural relaxation is needed to close it, the intermixing-dominated scenario is falsified and the review's roadmap should be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The weakest load-bearing step is the review's treatment of the sample-dependent Dirac gap, the stated 'most important open problem.' The favored explanation (Mn-Bi intermixing) is supported almost entirely by two DFT studies [67,68] that place antisites only in the topmost septuple layer, leaving deeper layers pristine; the paper itself concedes: 'the calculations using more realistic structural models are needed to clarify which mechanism actually takes place in the experimental samples.' The uncertainty is compounded by conflicting STM results: Ref. [70] finds gap absence correlated with high Mn_Bi defect density, while Ref. [71] finds no correlation with individual Mn_Bi/Bi_Mn defects and instead reports nanometer-scale gap fluctuations. The Outlook then uses the intermixing scenario to claim that suppressing intermixing 'could hopefully allow getting rid of the Dirac point gap issue,' which is more conditional than the surrounding narrative suggests. This does not weaken the primary claim that MnBi2Te4 is the first intrinsic MTI—that claim rests on magnetic, transport, and photoemission evidence independent of the gap mechanism—but it does mean the review's proposed resolution of its own central open problem is not yet anchored.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This short review marks the fifth anniversary of the discovery of the MnBi2Te4 family of intrinsic magnetic topological insulators. The authors first introduce the quantum anomalous Hall effect and the topological magnetoelectric effect, then describe the limitations of magnetically doped topological insulators that motivated the search for intrinsic compounds. They summarize the structural, magnetic, and electronic properties of MnBi2Te4, including the predicted and measured antiferromagnetic topological insulator state, and then devote considerable attention to the sample-dependent Dirac point gap, reviewing candidate magnetic and structural mechanisms and the possible role of Mn-Bi antisite defects. The review proceeds to the broader family, covering MnBi2Te4\\cdot nBi2Te3 compounds, MnSb2Te4, MnBi2Se4, and doped variants, and then to the two-dimensional limit, where even-layer and odd-layer films host zero-plateau quantum anomalous Hall, layer Hall, nonlinear Hall, and quantum anomalous Hall effects. An outlook lists further directions and open problems.","tokens_in":22150,"tokens_out":5127,"duration_ms":58724,"significance":"Provided the citations faithfully represent the large body of external work, this is a valuable and well-structured review. Its strengths are the extensive and current reference list, the compact reproductions of key data, and an unusually candid treatment of controversies. In particular, the authors explicitly state that both DFT studies of Mn-Bi intermixing model only the topmost septuple layer and that more realistic models are needed, and they report the conflicting STM conclusions of Refs. [70] and [71]. These caveats give the reader an accurate picture of the Dirac-gap problem. The central claim that MnBi2Te4 is the first intrinsic magnetic topological insulator is supported by magnetic, transport, and photoemission evidence that does not depend on the unresolved gap mechanism, so the main assertion is robust. The manuscript contains no derivations or free parameters; as a review, its quality criterion is fidelity to the literature, which appears carefully handled.","major_comments":[],"minor_comments":[{"comment":"Given that both DFT studies [67,68] introduce intermixing only in the topmost septuple layer, as the manuscript itself concedes, and that the STM evidence is conflicting, the Outlook sentence stating that suppressing Mn-Bi intermixing \"could hopefully allow getting rid of the Dirac point gap issue\" should be explicitly conditional, e.g., \"if the intermixing scenario is correct.\" As written, it conveys more confidence than the cited evidence supports.","section":"Outlook; Mn-Bi intermixing and its possible impact on the Dirac point gap"},{"comment":"In the third paragraph, \"magnetolectric properties\" should read \"magnetoelectric properties.\"","section":"Introduction"},{"comment":"In the paragraph beginning \"In this context, an important comment should be made,\" the phrase \"As far as the the sixth atomic layer\" contains a duplicated \"the.\"","section":"Mn-Bi intermixing and its possible impact on the Dirac point gap"},{"comment":"The discussion of the persistence of the Dirac-point gap above TNéel is balanced, but it could briefly restate that a structural origin is not excluded; the current text lists magnetic explanations without explicitly noting that the gap persistence is also consistent with a nonmagnetic contribution to the gap.","section":"MnBi2Te4 surface electronic structure above TNéel"}],"recommendation":"minor_revision","confidential_remarks":"The authors are among the discoverers of MnBi2Te4 and cite their own prior work frequently; this is natural for a discoverer-written review and does not by itself raise concerns. The review's reliability depends on faithful representation of a large external literature that I cannot independently verify, so it would be prudent for the editor to ensure that at least one referee with independent expertise in ARPES and STM of this material checks the representation of competing claims about the Dirac-point gap. The manuscript's scope is appropriate for a topical review venue."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a competent review of the MnBi2Te4 family, written by the group that discovered it. Do not go in expecting new results: it is explicitly a short review, and its value is organizational and bibliographic. The central claim — that MnBi2Te4 is the first intrinsic magnetic topological insulator — is well supported and rests on magnetic, transport, and photoemission evidence that is independent of the unresolved gap puzzle.\n\nWhat the paper does well: it is clearly organized, covers the field's development from the 2019 prediction through the latest transport and spectroscopy results, and is genuinely honest about the open problems. I particularly appreciated the compilation of laser-ARPES Dirac gap values in Fig. 2h and the balanced presentation of conflicting STM studies: one correlating gap absence with Mn_Bi density, the other finding no such correlation and instead seeing nanometer-scale fluctuations. The authors also explicitly concede, in the intermixing section, that 'the calculations using more realistic structural models are needed to clarify which mechanism actually takes place.' That kind of candor is not universal in this field.\n\nThe soft spots are real but not fatal. The favored explanation for the sample-dependent Dirac gap — Mn-Bi intermixing — rests on two DFT studies that introduce antisites only in the topmost septuple layer. The authors acknowledge this, but the Outlook then says suppressing intermixing 'could hopefully allow getting rid of the Dirac point gap issue,' which is more optimistic than the evidence warrants. The review could do a better job separating the robust claim (MnBi2Te4 is an intrinsic MTI) from the speculative roadmap (intermixing is the dominant gap-killer). The conflicting STM results are presented fairly, but the review does not try to reconcile them, which is fine for a short review. The heavy self-citation is normal for a discoverer-written review and does not bother me.\n\nWho is this for? A newcomer wanting a map of the MnBi2Te4 literature, or an established researcher needing a quick update. It delivers what it promises. It deserves serious peer review and publication as a review article, with minor suggestions to temper the outlook and explicitly flag which explanations are still under test.\n\nBottom line: accept it, but ask the authors to tighten the gap between what is known and what is hoped.","headline":"A solid, honest review of the MnBi2Te4 family by the people who discovered it; the central claim holds, but the Dirac-gap story is more conditional than the outlook suggests.","tokens_in":22659,"tokens_out":1139,"would_cite":true,"duration_ms":13657,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A layered antiferromagnet becomes the first intrinsic magnetic topological insulator, and its family is a tunable home for quantized Hall and axion phenomena.","keywords":["MnBi2Te4","intrinsic magnetic topological insulator","antiferromagnetic topological insulator","quantum anomalous Hall effect","axion insulator","Dirac point gap","Mn-Bi intermixing","van der Waals layered materials"],"falsifier":"Take a MnBi2Te4(0001) sample whose Mn-Bi intermixing has been suppressed and verified by depth-resolved atom-probe tomography or cross-sectional electron microscopy, and measure its Dirac point gap with laser-ARPES. If the gap remains small or vanishing instead of recovering to the predicted tens of meV, the antisite-dominated explanation would be ruled out; alternatively, a realistic supercell calculation with intermixing in all septuple layers that still keeps a large gap would falsify the deep-antisite mechanism.","tokens_in":21786,"feed_emoji":"🧲","tokens_out":11857,"duration_ms":111486,"temperature":0.7,"pith_summary":"This short review consolidates five years of evidence that MnBi2Te4 is the first intrinsic magnetic topological insulator: a stoichiometric, undoped layered antiferromagnet in which band topology and magnetism come from the same crystal. It argues that the MnBi2Te4 family is a tunable platform for the quantum anomalous Hall effect, the axion insulator state, the layer Hall effect, and nonlinear Hall transport, and that the main unresolved obstacle is the sample-dependent Dirac point gap on the (0001) surface. The stakes are practical: intrinsic magnetic topological insulators could carry dissipationless chiral edge currents and quantized magnetoelectric responses without the random-dopant disorder that has kept magnetically doped topological insulators at millikelvin temperatures. The review also narrows the gap puzzle to a leading physical suspect, Mn-Bi antisite defects, while explicitly flagging that the supporting calculations need more realistic models.","feed_headline":"MnBi2Te4: first intrinsic magnetic topological insulator","feed_subtitle":"A van der Waals antiferromagnet hosts quantized Hall physics; the remaining puzzle is the sample-dependent surface gap.","key_machinery":"The load-bearing object is the septuple-layer building block of MnBi2Te4: a Te-Bi-Te-Mn-Te-Bi-Te stack in which Mn moments order ferromagnetically within each layer and antiferromagnetically between adjacent blocks, giving A-type antiferromagnetism. The symmetry that makes the material a topological insulator is $S = \\Theta T_{1/2}$, the product of time reversal and half a lattice translation; surfaces that respect it remain gapless, while the S-breaking (0001) surface acquires a magnetic gap. The second piece of machinery is the Mn-Bi antisite defect: Mn atoms sitting on Bi sites carry moments antiparallel to the central Mn layer, and because the topological surface state is localized near the Bi layers, these antisites directly attack the Dirac gap. In thin films, the parity of the number of septuple layers controls the Chern number through the $P\\Theta$ symmetry (inversion times time reversal): odd-layer films allow $C\\neq 0$ and even-layer films enforce $C = 0$, which produces the zero-plateau quantum anomalous Hall state.","core_discovery":"The review's central claim is that MnBi2Te4 — a van der Waals crystal made of septuple layers with the Te-Bi-Te-Mn-Te-Bi-Te sequence, Mn moments of about $4.6\\,\\mu_B$ ordered ferromagnetically in each layer and antiferromagnetically between layers below about 25 K — realizes the long-predicted antiferromagnetic topological insulator state. Thanks to the combined symmetry $S = \\Theta T_{1/2}$ (time reversal followed by half a lattice translation), the bulk is a $Z_2 = 1$ insulator while the natural (0001) cleavage surface should host a topological surface state with a gap of tens of meV at the Dirac point. The review takes magnetic, transport, and photoemission data on bulk crystals and molecular-beam-epitaxy films as confirming this state, making MnBi2Te4 the first intrinsic magnetic topological insulator. It then surveys the family: Sb- and Se-substituted variants, Bi2Te3-intercalated compounds, and thickness-controlled films in which odd numbers of septuple layers give a zero-field quantum anomalous Hall effect with Chern number $C = 1$ while even numbers give a zero-plateau state relevant to the axion insulator, along with layer Hall and nonlinear Hall effects. The paper identifies the sample-dependent Dirac point gap — anywhere from roughly zero to tens of meV across different laser-ARPES measurements — as the most important open problem, and weighs Mn-Bi intermixing as the leading explanation, with the caveat that existing DFT studies introduce intermixing only in the topmost septuple layer.","pith_inferences":["Extension: the review's caveat implies a sharper experiment than any it reports — map the Dirac gap against the density of Mn-Bi antisites in the sixth atomic layer from the surface, the defects predicted to act most strongly but nearly invisible to scanning tunneling microscopy; current data mostly track second-layer antisites.","Extension: the $T^*\\approx 12.5$ K order-disorder transition of the Mn-Bi antisite sublattice offers a clean test: temperature-dependent laser-ARPES crossing $T^*$ on a sample with known $T^*$ should show a kink in the Dirac gap if antiparallel antisite moments are what close it.","Extension: if the reported growth of the average gap from about 26 to 44 meV in a 1 T field is general, moderate magnetic fields could serve as a stopgap for device operation before defect-free growth matures.","Extension: the contrasting ferromagnetic coupling of Mn antisites in MnBi2Se4-based heterostructures suggests selenium substitution could remove the gap-killing antisite mechanism entirely, at the cost of a different magnetic ground state."],"forward_implications":["The intrinsic character of MnBi2Te4 removes the random-dopant disorder that suppressed the Dirac gap in doped topological insulators, so the quantum anomalous Hall and axion effects can be pursued in a stoichiometric material.","If Mn-Bi intermixing is the main gap-killer, suppressing antisite defects by growth or capping should restore a large surface gap and raise the zero-field quantum anomalous Hall effect well above its current 1.5 K observation temperature.","The even-odd thickness rule makes film thickness a topological switch: odd septuple layers give $C=1$ quantum anomalous Hall behavior and even layers give the zero-plateau axion state, consistent with observed axion, layer Hall, and nonlinear Hall transport.","The persistence of the Dirac gap above the Néel temperature implies short-range magnetic order governs the surface electronic structure up to roughly 50-60 K, so explanations of the gap must involve fluctuating magnetism, not only static antiferromagnetic order."],"supporting_citations":[{"why":"First prediction and bulk experimental confirmation of MnBi2Te4 as an antiferromagnetic topological insulator.","marker":"[26]"},{"why":"Supplies the Z2 classification of antiferromagnetic topological insulators based on the S=ΘT1/2 symmetry that MnBi2Te4 is claimed to realize.","marker":"[38]"},{"why":"Extends the prediction to the wider MnBi2Te4 family and to topological axion states.","marker":"[28]"},{"why":"Reports molecular-beam-epitaxy-grown ultrathin films showing the intrinsic magnetic topological insulator state.","marker":"[29]"},{"why":"Predicts the thickness-dependent Chern numbers, quantum anomalous Hall effect, and zero-plateau state in two-dimensional MnBi2Te4.","marker":"[30]"},{"why":"DFT calculation showing Mn-Bi antisites with opposite moments reduce or close the Dirac point gap.","marker":"[67]"},{"why":"Alternative DFT mechanism in which intermixing pushes the surface state toward the subsurface septuple layer, shrinking the magnetic gap.","marker":"[68]"},{"why":"High-field magnetization evidence that Mn-Bi antisites order antiparallel to the central Mn layer, the basis of the gap-reduction scenario.","marker":"[45]"},{"why":"First observation of the zero-plateau quantum anomalous Hall (axion insulator) state in even-layered MnBi2Te4 flakes.","marker":"[125]"},{"why":"First observation of the zero-field quantum anomalous Hall effect in a five-septuple-layer MnBi2Te4 flake.","marker":"[141]"}],"fun_headline_variants":["MnBi2Te4: intrinsic magnetic topological insulator","MnBi2Te4: the antiferromagnetic topological insulator","MnBi2Te4 films: zero-field quantum anomalous Hall","MnBi2Te4: five years of magnetic topology","The intrinsic magnet that hosts topological electrons"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that Mn-Bi intermixing in the topmost septuple layer of the DFT slabs captures the real defect distribution that controls the Dirac gap; the review itself concedes that more realistic structural models are needed to know which mechanism operates in actual samples.","fun_headline_variants_meta":{"raw":{"variants":["MnBi2Te4: intrinsic magnetic topological insulator","MnBi2Te4: the antiferromagnetic topological insulator","MnBi2Te4 films: zero-field quantum anomalous Hall","MnBi2Te4: five years of magnetic topology","The intrinsic magnet that hosts topological electrons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000259,"raw_usage":{"total_tokens":1616,"prompt_tokens":1009,"completion_tokens":607,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":625,"completion_tokens_details":{"reasoning_tokens":529}},"tokens_in":625,"tokens_out":607,"duration_ms":6334,"temperature":1.0,"reasoning_tokens":529,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:21:41.945545+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a MnBi2Te4(0001) sample whose Mn-Bi intermixing has been suppressed and verified by depth-resolved atom-probe tomography or cross-sectional electron microscopy, and measure its Dirac point gap with laser-ARPES. If the gap remains small or vanishing instead of recovering to the predicted tens of meV, the antisite-dominated explanation would be ruled out; alternatively, a realistic supercell calculation with intermixing in all septuple layers that still keeps a large gap would falsify the deep-antisite mechanism.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"First observation of the zero-plateau quantum anomalous Hall (axion insulator) state in even-layered MnBi2Te4 flakes."}],"review_version":1}