{"id":"0aaa4aa6-35d3-4a05-b7ea-eaea04860a06","arxiv_id":"2501.05296","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"EuCu2As2 is argued to crystallize in the locally non-centrosymmetric P4/nmm structure, with a phonon anomaly near 165 K that may signal an electronic density wave instability.","lead":"Raman and density-functional calculations on the pnictide EuCu2As2 indicate that its true local structure is the non-centrosymmetric P4/nmm form rather than the widely assumed I4/mmm form. The same data show an unexpected hardening of the main phonon near 165 K, which the authors interpret as a possible electronic density wave instability.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Bulk structural reassignment rests on a single U=7 eV phonon calculation and an acknowledged surface/bulk ambiguity; without a U/functional scan and bulk-sensitive confirmation, P4/nmm is plausible but not established.","rationale":"The reader's weakest assumption correctly identifies the DFT+U phonon calculation and the surface/bulk ambiguity as the hinge of the structural claim. My stress-test confirms that this is the most load-bearing point: the experimental observation of seven Raman peaks is fully compatible with a surface P4/nmm layer on a bulk I4/mmm crystal, and the authors themselves state that XRD cannot distinguish the two space groups. The only bulk-sensitive evidence is therefore the phonon calculation, which is reported for a single Hubbard U and a single magnetic state. Because the soft mode in I4/mmm involves in-plane Cu/As displacements, while U is applied to localized Eu 4f states, it is not obvious that the instability should be insensitive to U; however, it is also not obviously wrong. The correct response is not to reject the paper, but to require the missing calculation before accepting the structural reassignment as conclusive. This matches the CONDITIONAL verdict already given, so I recommend no change to that verdict. The density-wave interpretation at 165 K is explicitly speculative and the authors admit the absence of folded modes or amplitude modes; it therefore does not independently change the verdict. The concrete U-scan test is the single check most likely to settle whether the structural claim has a real soft spot or survives the scrutiny.","tokens_in":15372,"tokens_out":8234,"duration_ms":87569,"concrete_test":"Recompute the I4/mmm and P4/nmm phonon dispersions with the same PBEsol+U Dudarev scheme for U = 0, 3, 5, 7, and 9 eV, in both nonmagnetic and AFM states, and report E(P4/nmm) - E(I4/mmm). If the I4/mmm imaginary E_g mode disappears for U < 7 or in the nonmagnetic state, the claimed bulk instability is an artifact and the structural conclusion requires a bulk-sensitive confirmation; if the imaginary mode persists across all U values and magnetic states while P4/nmm remains lower in energy, the DFT support for the central claim survives this specific attack.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing premise is that the seven Raman bands are first-order bulk zone-center modes of a P4/nmm phase and that bulk I4/mmm is dynamically unstable. In §3.2, this is supported only by PBEsol+U with U = 7 eV in an AFM-ordered calculation. Table 3 lists only AFM I4/mmm frequencies, and the paper never reports the total-energy difference between P4/nmm and I4/mmm. The imaginary E_g mode at -50.07i could depend on the Hubbard U imposed on the Eu 4f states or on the magnetic constraint, and no U-dependence, functional-dependence, or magnetic-state-dependence check is provided. The authors explicitly acknowledge that Raman is a near-surface probe and that 'purely from experiments it is possible to envisage a situation where the surface belongs to the P4/nmm phase and the bulk belongs to the I4/mmm phase'; their only answer is the same bulk DFT calculation whose parameter choice is untested. If the I4/mmm soft mode is an artifact of U = 7 eV or of the AFM constraint, the central structural claim loses its only bulk support, and the subsequent interpretation of the 165 K anomaly as a density-wave signature in a locally non-centrosymmetric bulk compound is correspondingly weakened.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports Raman spectra, powder XRD, magnetic susceptibility, and DFT-based lattice dynamics for EuCu2As2. The central claim is that the compound does not adopt the accepted ThCr2Si2-type I4/mmm structure but instead realizes the CaBe2Ge2-type P4/nmm structure: seven Raman peaks are observed above 100 cm−1, which is more than the four Raman-active modes expected for I4/mmm, and DFT phonon calculations find an imaginary mode for I4/mmm while P4/nmm is dynamically stable. Temperature-dependent Raman measurements show a deviation of the A1g mode frequency from an anharmonic model below about 165 K, which the authors interpret as a possible signature of a subtle electronic density-wave instability.","tokens_in":15653,"tokens_out":4601,"duration_ms":47751,"significance":"If the structural reassignment is correct, the paper is significant because it challenges the accepted I4/mmm classification of ACu2As2 compounds and adds EuCu2As2 to the small family of locally non-centrosymmetric 122 systems that host density-wave-like instabilities. The work has clear strengths: the DFT phonon calculations are independent first-principles results rather than fits to the Raman peaks, the XRD ambiguity between the two space groups is explicitly acknowledged, the Raman spectra are carefully analyzed, and the authors openly state the limits of their evidence, including the surface-sensitivity of Raman and the absence of CDW signatures. These strengths make the central claim plausible, but the bulk structural conclusion currently rests on a narrow set of computational parameters and the density-wave interpretation is speculative.","major_comments":[{"comment":"The bulk structural claim rests on the DFT prediction that I4/mmm is dynamically unstable, but this is shown for a single computational setup: PBEsol+U with U = 7 eV and an AFM-ordered state. The imaginary Eg mode at -50.07i in Table 3 could depend on the Hubbard U applied to the Eu 4f states or on the magnetic constraint, yet no U-dependence, exchange-correlation functional dependence, or magnetic-order dependence is reported for the phonon instability. Because the authors state, correctly, that Raman cannot distinguish a P4/nmm surface from a P4/nmm bulk, this DFT calculation is the only bulk-sensitive support for the structural reassignment. Without a robustness check of the soft mode, the conclusion that 'the whole structure belongs to the P4/nmm phase' is not fully supported.","section":"§2, §3.2, Table 3"},{"comment":"The seven observed peaks are assigned to specific irreducible representations (Eg, B1g, A1g) solely by comparing measured frequencies with the calculated P4/nmm spectrum; no polarization-resolved Raman measurements are reported. As a result, the symmetry labels attached to the modes, which are later used to discuss the displacement pattern of the anomalous A1g mode, are not independently verified. In particular, the assignments of P1 (calculated 106.7 cm−1, observed 115 cm−1) and P2 (calculated 137.4 cm−1, observed 126 cm−1) show discrepancies of 8–11 cm−1, and possible contributions from two-phonon or defect-related scattering are not discussed. Polarization-resolved measurements on an oriented single crystal would substantially strengthen the mode assignment and the structural conclusion.","section":"§3.2, Fig. 4, Fig. 6"},{"comment":"The phonon anomaly is quantified as the deviation from an anharmonic model fitted to Eq. (1) over a range that is chosen after inspecting the data: the fits are performed between 300 K and 165 K (or 160 K, according to the Fig. 7 caption) and then extrapolated to 80 K. The authors do not report the fitted values of ω0, A, and B, nor any goodness-of-fit measure or residual analysis, so it is difficult to judge whether the extrapolated baseline is reliable. A reproducible assessment would require reporting the fit parameters and uncertainties and testing how the inferred anomaly temperature changes when the fitting window is varied.","section":"§3.3, Eq. (1), Fig. 7"},{"comment":"The interpretation of the 165 K phonon hardening as a 'subtle electronic density wave instability' is presented without the characteristic signatures that would distinguish a density wave from other electronic or magnetic couplings; the authors themselves note that amplitude modes, zone-folded modes, and two-phonon modes were not observed. The susceptibility deviation in Fig. 3(b) is small, and its fit range is also defined relative to 165 K. The phonon anomaly itself is well documented, but the density-wave conclusion remains one of several possibilities. The abstract and summary should be tempered unless additional evidence, such as superlattice reflections, transport anomalies, or heat-capacity features, is provided.","section":"§3.3, §4"}],"minor_comments":[{"comment":"There is an inconsistency in the author list header ('V aitheeswaran' should be 'Vaitheeswaran') and the text contains typos such as 'N'eel', 'focussing', and 'contrary to that what is expected'. The manuscript would benefit from a careful proofreading pass.","section":"Throughout"},{"comment":"The description of the supercell used for the force-constant calculation, '2√2×2√2×1 conventional cells', is difficult to parse; the authors should clarify the exact supercell geometry and the number of atoms used.","section":"§2"},{"comment":"Equation (1) uses 'K_B' and later 'KBT' in the exponentials; the Boltzmann constant should be denoted consistently (typically k_B), and the temperature dependence should be written unambiguously.","section":"§3.3, Eq. (1)"},{"comment":"The contour plot overlaid in Fig. 3(b) is not clearly described in the caption; the reader cannot tell which color scale corresponds to the Raman shift evolution. A separate panel or a clearer caption would help.","section":"§3.1, Fig. 3(b)"},{"comment":"The notation 'Theo. (NM)' and 'Theo. (AFM)' is defined in the table, but the differences between non-magnetic and antiferromagnetic results are not discussed in the text; a brief comment on why the AFM frequencies were used for comparison would be useful.","section":"§3.2, Table 2"}],"recommendation":"major_revision","confidential_remarks":"The paper's central structural claim is plausible but would be greatly strengthened by a U-dependence study of the I4/mmm phonon instability and, ideally, a bulk-sensitive structural probe or polarization-resolved Raman on a single crystal. If the authors cannot provide these, they should revise the abstract and conclusions to present P4/nmm as a strong possibility rather than an established bulk structure. The density-wave interpretation is even less constrained and should be framed more cautiously."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi [colleague],\n\nThe thing to know about arXiv:2501.05296 is that it contains the first Raman and lattice-dynamics study of EuCu2As2, and it argues that the accepted I4/mmm structure is wrong: the compound is actually P4/nmm (locally non-centrosymmetric). The claim is plausible, the data are clearly presented, and the paper is honest about its limits, but the structural assignment is not airtight.\n\nWhat is genuinely new: a clean 80 K Raman spectrum with seven well-resolved peaks above 100 cm^-1. I4/mmm allows only four Raman-active modes, so the mode count alone is a strong experimental argument that the surface region is not I4/mmm. The DFT phonon calculations are independent first-principles results, not fit to the Raman peaks, and the P4/nmm frequencies match the measured ones to within about 8 cm^-1 on average, with the 235 cm^-1 mode exceeding the highest I4/mmm mode (about 200 cm^-1) altogether. The I4/mmm phonon dispersion shows an imaginary Eg mode, making that structure dynamically unstable at the level of theory used. The authors also correctly note that powder XRD cannot distinguish the two space groups, and they explicitly flag the possibility that a surface P4/nmm phase coexists with an I4/mmm bulk; their only counter is the bulk DFT.\n\nThe soft spots are real but not fatal. First, the DFT stability argument is a one-point check: PBEsol+U with U=7 eV in an AFM state, with no U-dependence, functional-dependence, or magnetic-state test, and no total-energy comparison between the two space groups. If the soft mode is an artifact of U=7, the bulk support for P4/nmm largely collapses. Second, the mode assignment relies on frequency matching only; there is no polarization-resolved Raman to confirm the symmetries. Third, the 80 K spectrum vs. the stated 100-300 K Linkam range is a small inconsistency. Fourth, the 165 K phonon anomaly is certainly real—the A1g mode hardens well beyond the anharmonic fit, and other modes follow—but the density-wave interpretation is supported mainly by a very small deviation in the magnetic susceptibility and an analogy to TMDs. To the authors' credit, they list the missing signatures (amplitude modes, zone-folded modes) and explicitly do not claim a confirmed CDW. The citation pattern is fine: earlier EuCu2As2 work is properly attributed, and the comparisons to SrPt2As2 and CeRh2As2 are apt.\n\nThis paper deserves a serious referee. The experimental observation is new, the mode-count argument is strong, and the authors have been appropriately cautious. A referee should ask for the U/functional scan and a bulk-sensitive check of the odd reflections; those could solidify or overturn the structural claim. I would send it to peer review rather than desk reject.","headline":"First Raman study of EuCu2As2 makes a plausible but not airtight case for P4/nmm over I4/mmm, and reports a real 165 K phonon anomaly whose density-wave origin is speculative.","tokens_in":16223,"tokens_out":9003,"would_cite":true,"duration_ms":83368,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["78.30.-j","63.20.-e","71.45.Lr"],"model":"deepseek-v4-flash","headline":"This paper argues that EuCu2As2 realizes the locally non-centrosymmetric P4/nmm structure instead of the accepted I4/mmm structure, and that anomalous hardening of the strongest phonon mode below 165 K signals a subtle electronic density…","keywords":["EuCu2As2","Raman spectroscopy","P4/nmm structure","I4/mmm structure","phonon anomaly","lattice dynamics","charge density wave","122 pnictides"],"falsifier":"Refining single-crystal X-ray or neutron diffraction of EuCu2As2 and hunting specifically for weak reflections with odd (hkl) index sums — forbidden by I4/mmm but allowed by P4/nmm — would decide the structure question directly, as would repeating the phonon calculation over a range of U values and functionals to see whether the I4/mmm imaginary modes survive the choice of method. For the 165 K anomaly, resolving the CDW fingerprints the paper could not see — an amplitude mode, zone-folded modes, or two-phonon features appearing below 165 K — would confirm the density wave, while a bulk probe (specific heat, resistivity, thermal expansion) showing no feature at 165 K would weigh against the electronic-instability interpretation.","tokens_in":15182,"feed_emoji":"⚛️","tokens_out":13071,"duration_ms":112585,"temperature":0.7,"pith_summary":"This paper argues that the copper pnictide EuCu2As2 has been assigned the wrong crystal structure since its discovery: Raman spectroscopy shows seven well-defined optical phonon peaks above 100 $cm^{-1}$, whereas the accepted ThCr2Si2-type I4/mmm structure admits only four Raman-active modes, and density-functional lattice dynamics find I4/mmm dynamically unstable while the locally non-centrosymmetric CaBe2Ge2-type P4/nmm structure is stable and reproduces the measured frequencies. The authors take this as strong evidence that EuCu2As2 actually realizes P4/nmm, a structure rare among the copper-arsenide 122 family. They further report that the strongest mode, the A1g phonon near 232 $cm^{-1}$, hardens below about 165 K far beyond the fitted anharmonic decay, with smaller anomalies in four other modes and a slight deviation in the $Eu^{{2+}}$ Curie–Weiss susceptibility. They interpret this as a strong signature of a subtle electronic density wave instability that couples to out-of-plane vibrations of the interlayer Cu(2) and As(1) atoms, while conceding that the usual CDW fingerprints (amplitude, zone-folded, or two-phonon modes) have not been resolved. If correct, the work puts EuCu2As2 in a short list of 122 pnictides with local inversion-symmetry breaking and points to a rare density-wave phase in this family.","feed_headline":"Seven Raman peaks challenge EuCu2As2's accepted structure","feed_subtitle":"Count and lattice dynamics favor a non-centrosymmetric P4/nmm phase; a 165 K phonon hardening hints at a density wave.","key_machinery":"The argument is carried by three coupled tools. Mode counting: for I4/mmm the Γ-point optical modes decompose into A1g + 2A2u + B1g + 2Eu + 2Eg, giving four Raman-active bands (A1g, B1g, and two Eg doublets), whereas P4/nmm gives ten Raman-active bands (3A1g + 2B1g + 5Eg); observing seven distinct peaks is therefore a symmetry-level contradiction of the accepted structure. Lattice dynamics: DFT+U (PBEsol, U = 7 eV on the Eu 4f states) phonon dispersions computed by the supercell force-constant method yield imaginary soft modes for I4/mmm — the hallmark of a dynamically unstable structure — and a purely real, stable spectrum for P4/nmm whose frequencies near the Γ point track the measured modes. Anharmonic model: the temperature dependence of each mode is fit to the Balkanski–Wallis–Haro three- and four-phonon decay formula between 165 K and 300 K and extrapolated down; the area between the extrapolated curve and the data below 165 K quantifies the anomaly, which is largest for the out-of-plane A1g mode and decreases monotonically from P7 to P1, tying the instability to the interlayer Cu(2)–As(1) vibrations of the A1g mode.","core_discovery":"Stated on the paper's own terms, the discovery is that the accepted I4/mmm picture of EuCu2As2 is wrong and that the consequences are observable: seven first-order Raman bands — at approximately 115, 126, 150, 170, 201, 224, and 235 $cm^{-1}$ — exceed what I4/mmm symmetry permits and match, in count, symmetry labels (Eg, B1g, A1g, B1g, Eg, Eg, A1g), and frequency, the Γ-point phonons of the locally non-centrosymmetric P4/nmm structure computed with DFT+U. The same calculation shows I4/mmm to be dynamically unstable, with imaginary soft modes from in-plane Cu and As vibrations, and P4/nmm to be stable, with a phonon spectrum whose highest modes are dominated by Cu and As motion while Eu vibrations sit below about 150 $cm^{-1}$. On cooling, the A1g mode at ~232 $cm^{-1}$ departs from the anharmonic (three- and four-phonon decay) curve below 165 K, abruptly hardening while narrowing faster than the anharmonic fit; four other modes show smaller versions of the same deviation. The authors propose that an electronic density wave instability, analogous to the CDW states seen in Raman studies of transition-metal dichalcogenides and kagome metals and in the P4/nmm 122 compound SrPt2As2, gaps part of the Fermi surface and stiffens the phonons, particularly the out-of-plane A1g vibration of the interlayer Cu(2) and As(1) atoms. They explicitly stop short of claiming conclusive proof: no amplitude modes, zone-folded modes, or two-phonon features were observed, and more detailed studies are called for.","pith_inferences":["If the P4/nmm assignment survives bulk-sensitive probes, then the 'collapsed tetragonal' picture of Cu-based 122 pnictides may have been describing a dynamical average: with I4/mmm phonons carrying imaginary soft modes, the accepted structure could be an unstable parent from which a locally non-centrosymmetric ground state emerges.","A targeted bulk probe — specific heat, resistivity, or thermal expansion across 165 K — is the cheapest test of the density-wave reading: a small anomaly located at 165 K would strengthen it, while a clean null result would push the hardening toward a magnetic-fluctuation origin tied to the Eu sublattice.","Because the experiment used a polycrystalline sample and a single 532 nm laser line, repeating the measurements on single crystals with polarization analysis and with different laser wavelengths (varying penetration depth) could confirm that all seven modes are bulk first-order Raman phonons and fix their symmetries unambiguously."],"forward_implications":["EuCu2As2 would join SrPt2As2 and CeRh2As2 as a 122 pnictide realizing the locally non-centrosymmetric CaBe2Ge2-type P4/nmm structure, breaking the rule that copper-arsenide 122 compounds take the ThCr2Si2-type I4/mmm form.","The seven-peak Raman pattern, anchored by the intense ~232 cm^-1 A1g mode, becomes a quick experimental fingerprint for P4/nmm order in other copper-based 122 compounds.","Below 165 K, the hardening of the A1g mode and its accelerated linewidth narrowing imply a partial gap on the Fermi surface and a reduced electron–phonon coupling — the behavior expected of a subtle charge density wave whose amplitude mode has so far escaped detection.","The near-degeneracy of the computed magnetic ground states and the small non-compensated moments induced on Cu(2) and As(1) near the Eu layer suggest that pressure or magnetic field could tune between density-wave and magnetic orders, as already seen for the Eu spin arrangement."],"supporting_citations":[{"why":"Prior study that assigned EuCu2As2 to the I4/mmm structure and characterized the Eu ordering near 14–17 K; the accepted assignment this paper overturns.","marker":"[21]"},{"why":"Raman plus ARPES study of BaCu2As2 reporting extra Raman peaks inconsistent with bulk I4/mmm and attributing them to surface reconstruction; supplies the precedent and the surface-state scenario the paper must rebut.","marker":"[17]"},{"why":"Parlinski–Li–Kawazoe direct supercell method used to obtain the interatomic force constants and phonon dispersions that show I4/mmm dynamically unstable and P4/nmm stable.","marker":"[34]"},{"why":"Dudarev DFT+U scheme, with U = 7 eV on the Eu 4f states, used in all structural and dynamical calculations.","marker":"[29]"},{"why":"Supplies the PBEsol exchange-correlation functional on which the lattice parameters, phonons, and stability comparison rest.","marker":"[28]"},{"why":"Balkanski–Wallis–Haro three- and four-phonon anharmonic relation that defines the expected temperature dependence; the 165 K deviation is measured against it.","marker":"[45]"},{"why":"NMR evidence for a density wave in the P4/nmm 122 compound SrPt2As2, the comparator the authors invoke to interpret the 165 K anomaly as an electronic density wave.","marker":"[52]"},{"why":"Raman study of a kagome metal showing A1g phonon hardening on entering the CDW state; the pattern the A1g behavior below 165 K is matched against.","marker":"[51]"},{"why":"Single-crystal study of EuCu2As2 establishing the Eu magnetic structure and ruling out linearly dispersing topological states, used to exclude magnon and Dirac-interband explanations of the anomaly.","marker":"[23]"}],"fun_headline_variants":["Seven Raman peaks expose EuCu2As2's hidden non-centrosymmetry","EuCu2As2: phonon hardening at 165 K hints at density wave","Raman and DFT dethrone I4/mmm for EuCu2As2, favor P4/nmm","Strong A1g anomaly in EuCu2As2 below 165 K points to CDW"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The structural conclusion rests on the assumption that the DFT+U phonon calculation (PBEsol with U = 7 eV on the Eu 4f states) correctly predicts I4/mmm to be dynamically unstable, and that the seven Raman bands are intrinsic first-order modes of bulk EuCu2As2 rather than a near-surface artifact; the authors themselves note that Raman probes only the surface and that a P4/nmm surface sitting on an I4/mmm bulk cannot be excluded from experiment alone, and they report no test of how the result depends on the chosen U value or exchange-correlation functional.","fun_headline_variants_meta":{"raw":{"variants":["Seven Raman peaks expose EuCu2As2's hidden non-centrosymmetry","EuCu2As2: phonon hardening at 165 K hints at density wave","Raman and DFT dethrone I4/mmm for EuCu2As2, favor P4/nmm","Strong A1g anomaly in EuCu2As2 below 165 K points to CDW"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000528,"raw_usage":{"total_tokens":2673,"prompt_tokens":1201,"completion_tokens":1472,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":817,"completion_tokens_details":{"reasoning_tokens":1374}},"tokens_in":817,"tokens_out":1472,"duration_ms":10799,"temperature":1.0,"reasoning_tokens":1374,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:14:23.083986+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Refining single-crystal X-ray or neutron diffraction of EuCu2As2 and hunting specifically for weak reflections with odd (hkl) index sums — forbidden by I4/mmm but allowed by P4/nmm — would decide the structure question directly, as would repeating the phonon calculation over a range of U values and functionals to see whether the I4/mmm imaginary modes survive the choice of method. For the 165 K anomaly, resolving the CDW fingerprints the paper could not see — an amplitude mode, zone-folded modes, or two-phonon features appearing below 165 K — would confirm the density wave, while a bulk probe (specific heat, resistivity, thermal expansion) showing no feature at 165 K would weigh against the electronic-instability interpretation.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior study that assigned EuCu2As2 to the I4/mmm structure and characterized the Eu ordering near 14–17 K; the accepted assignment this paper overturns."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Raman plus ARPES study of BaCu2As2 reporting extra Raman peaks inconsistent with bulk I4/mmm and attributing them to surface reconstruction; supplies the precedent and the surface-state scenario the paper must rebut."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Parlinski–Li–Kawazoe direct supercell method used to obtain the interatomic force constants and phonon dispersions that show I4/mmm dynamically unstable and P4/nmm stable."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Balkanski–Wallis–Haro three- and four-phonon anharmonic relation that defines the expected temperature dependence; the 165 K deviation is measured against it."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"NMR evidence for a density wave in the P4/nmm 122 compound SrPt2As2, the comparator the authors invoke to interpret the 165 K anomaly as an electronic density wave."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Single-crystal study of EuCu2As2 establishing the Eu magnetic structure and ruling out linearly dispersing topological states, used to exclude magnon and Dirac-interband explanations of the anomaly."}],"review_version":1}