{"id":"860b47f2-92d2-4613-bca8-b0e018c2939e","arxiv_id":"1908.02187","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"DFT calculations predict K2SiI6 and Rb2SiI6 as stable, lead-free double perovskites with promising photovoltaic absorption.","lead":"The authors use computer simulations to predict a new family of silicon-based, lead-free double perovskite materials that could absorb sunlight in solar cells. Two of the three proposed compounds are calculated to be stable and have favorable optical properties, making them candidates for more sustainable photovoltaics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Thermodynamic stability is tested only against ASiI3 + AI3; the natural binary decomposition 2AI + SiI4 is omitted, so the positive ΔH values do not yet establish synthesizability.","rationale":"The reader's weakest assumption already flags the single-channel stability check and the absence of competing-phase screening. My concern is more specific: Equation 4 uses an atypical product set and misses the standard binary decomposition into 2AI + SiI4. This is not merely a completeness nit because SiI4 and alkali iodides are common, thermodynamically strong phases; if 2AI + SiI4 lies lower in energy, the predicted compounds would not be synthesizable as bulk phases, however attractive their computed band structures are. The abstract's band-gap and stability claims are also stated without the parity-forbidden transition caveat and the admitted Cs2SiI6 dynamic instability, but those are secondary. The decomposition-channel omission alone makes the paper conditional: the central photovoltaic promise depends on the compounds actually forming, and the current evidence does not test the most natural competing products. I would keep the reader's CONDITIONAL verdict, with the added condition that the 2AI + SiI4 channel be evaluated before the stability claim is accepted.","tokens_in":11632,"tokens_out":7928,"duration_ms":92516,"concrete_test":"Using the same PBEsol settings as Table III, optimize Fm3m A2SiI6, rocksalt AI, and the experimental SiI4 crystal structure, then evaluate ΔH = 2E(AI) + E(SiI4) − E(A2SiI6). Also place A2SiI6, AI, SiI4, ASiI3, AI3, and elemental references on a convex hull. If any decomposition involving 2AI + SiI4 is exothermic (negative ΔH), the thermodynamic-stability conclusion in Section III is invalid. Repeat at HSE06 for the most favorable channel to check functional sensitivity.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Equation 4 defines the only thermodynamic stability check as ΔH = E(ASiI3) + E(AI3) − E(A2SiI6). For vacancy-ordered double perovskites A2B(IV)X6, the conventional and chemically conservative competing phase set is 2AX + BX4, here 2AI + SiI4; SiI4 is a known stable molecular solid and KI/RbI/CsI are very stable ionic crystals. The chosen products ASiI3 + AI3 are not the oxidation-state-preserving binaries and appear to involve a redox decomposition (Si(IV) to Si(II) and I(−I) to I3−). A positive ΔH against this single, nonstandard channel (Table III: 41–76 meV/atom) says nothing about stability against 2AI + SiI4. If that channel is exothermic, K2SiI6 and Rb2SiI6 are metastable at best, and the central 'robust, high-temperature stable' claim, from which the photovoltaic promise follows, collapses. The paper's other stability evidence (elastic constants, 400 K MD for K and Rb) does not detect thermodynamic instability with respect to known phases; MD only probes kinetic barriers near the assumed structure. Cs2SiI6 is already admitted to be dynamically unstable at 400/300 K, so the family-level stability claim is also narrower than the abstract states. The missing decomposition channel is therefore the most load-bearing unverified premise.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses first-principles DFT (PBE/PBEsol for structure, HSE06 for electronic and optical properties) to predict that the vacancy-ordered double perovskites A2SiI6 (A = K, Rb, Cs) have direct band gaps of 0.84–1.15 eV, small electron effective masses, visible-region absorption, and favorable thermodynamic and dynamic stability, and it proposes them as lead-free, earth-abundant absorber candidates for perovskite solar cells. The manuscript also reports elastic constants, decomposition enthalpies against a single product set, 400 K molecular dynamics for K and Rb, Bader/Mulliken charges, and a parity analysis showing that the fundamental direct gap is dipole-forbidden with allowed optical transitions at 1.57–1.73 eV. The central claim is that K2SiI6 and Rb2SiI6, in particular, are synthesizable and promising photovoltaic absorbers.","tokens_in":11911,"tokens_out":4538,"duration_ms":44208,"significance":"If the stability and optical-gap claims survive scrutiny, the paper would identify two lead-free, earth-abundant vacancy-ordered double perovskites with allowed optical gaps (1.57–1.73 eV) in a useful range, small electron effective masses (0.17–0.23 m0), and visible absorption; the explicit parity analysis is a useful caution against screening on fundamental gaps alone. The manuscript is genuinely predictive rather than fitted: no parameters are tuned to target properties, HSE06 and the Shockley–Queisser limit are adopted from standard external benchmarks, and the band gaps, masses, and absorption spectra are computed. However, the significance is conditional on two load-bearing issues: the thermodynamic stability is tested only against a nonstandard decomposition product set, and the abstract's headline band-gap range misrepresents the optically relevant gaps. The paper also transparently reports that Cs2SiI6 is dynamically unstable, which narrows the family-level stability claim but is a point in favor of the authors' honesty.","major_comments":[{"comment":"The headline 'suitable band gaps of 0.84–1.15 eV' in the Abstract, and the screening discussion in §III that 'we here have screened three candidates,' report the fundamental direct band gaps, but the manuscript itself shows these transitions are parity-forbidden (Fig. 5 and the text 'the direct transition from VBM to CBM is forbidden'). The dipole-allowed optical transitions are 1.57 eV (K2SiI6), 1.63 eV (Rb2SiI6), and 1.73 eV (Cs2SiI6). Since the Shockley–Queisser analysis, the 'wide photon absorption in the visible range' claim, and the comparison to MAPbI3 all depend on the optically allowed gap, the Abstract and the screening rationale must be rephrased around the allowed gaps; the current text overstates how close these materials are to the 0.9–1.5 eV ideal range.","section":"Abstract and §III (band gaps)"},{"comment":"The thermodynamic stability test is defined only as ΔH = E(ASiI3) + E(AI3) − E(A2SiI6) in Eq. (4). This competing product set is nonstandard and appears to involve a redox decomposition (Si(IV) to Si(II), I(−I) to triiodide), rather than the oxidation-state-preserving binaries 2AI + SiI4. SiI4 is a known stable molecular solid and KI, RbI, and CsI are very stable ionic crystals, so the natural decomposition channel is 2AI + SiI4. The positive ΔH values in Table III (41–76 meV/atom) therefore do not establish stability with respect to known phases, and the 'robust high-temperature stability' claim collapses if the 2AI + SiI4 channel is exothermic. The authors should compute ΔH for 2AI + SiI4 (and preferably other competing products) before claiming synthesizability.","section":"§III, Eq. (4) and Table III"},{"comment":"The Abstract's 'good stability at high temperature' is not supported for the whole A2SiI6 family: §III reports that Cs2SiI6 is dynamically unstable at both 400 K and 300 K (Fig. S4), leaving only K2SiI6 and Rb2SiI6 dynamically stable (Fig. 6). The Conclusion partially acknowledges this, but the Abstract and the opening of the Conclusion ('These three DPs exhibit excellent ... stable properties') do not. The claims should be restricted to the two stable members, and the Si–I covalency analysis (Table S2) should be presented as a prediction for why Cs2SiI6 fails rather than as a minor exception.","section":"Abstract, §III, and §IV (family-level stability)"},{"comment":"The entire prediction is predicated on the assumption that these compounds crystallize in the vacancy-ordered Fm3m double-perovskite structure (Fig. 1(a)). No search over competing polymorphs or alternative ordered vacancy arrangements is reported, and the only decomposition channel tested is Eq. (4). If the true ground-state structure of K2SiI6 or Rb2SiI6 is not the assumed Fm3m prototype, the computed electronic, optical, and stability properties do not apply. The authors should either scan plausible competing structures or explicitly state this limitation as a caveat on the 'first-principles prediction' claim.","section":"§III (structural assumption)"}],"minor_comments":[{"comment":"The first sentence is ungrammatical: 'Despite the exceeding 23% photovoltaic efficiency achieved in organic-inorganic hybrid perovskite solar cells obtaining' should be rephrased, and 'the stable materials with desirable band gap are rare and are highly desired' can be tightened.","section":"Abstract"},{"comment":"In the paragraph introducing ε(ω), the phrase 'where ε1(ω) and ε2(ω) are the real and imaginary parts in several' is incomplete; it should state 'respectively' or otherwise finish the definition.","section":"§III (dielectric function)"},{"comment":"The column header 'mental Si I' should read 'elemental Si I', and the layout of the Bader and Mulliken charge columns is confusing because the elemental labels are repeated without clear separation between the two charge analyses.","section":"Table III"},{"comment":"Reference [3] (Battaglia et al.) lists a title that appears to belong to reference [4] (Stranks et al.); the citation titles should be checked against the actual sources. Reference [41] is cited as a supplemental-material pointer but appears in the main reference list and should be formatted as such.","section":"References"},{"comment":"There are several typographical errors: 'DPS' should be 'DPs' in the Conclusion, 'PCSs' should be 'PSCs', and in the Introduction 'some of which a portion of materials' is awkwardly phrased.","section":"Throughout (typos)"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this is the first computational study I know of for A2SiI6 vacancy-ordered double perovskites, and the body is more careful than the abstract. The authors run standard HSE06 calculations on K2SiI6, Rb2SiI6, and Cs2SiI6, get direct gaps, estimate effective masses and absorption, and—usefully—recognize that the VBM-to-CBM transition at Γ is parity forbidden and compute the allowed optical gaps (1.57–1.73 eV). That is real work, and the conclusion that K2SiI6 is the best optical candidate is consistent with their own numbers.\n\nNow the soft spots.\n\nThe abstract quotes 0.84–1.15 eV as the suitable band-gap range. Those are the fundamental direct gaps, but the paper's own parity analysis shows those transitions are forbidden; the optically allowed gaps are 1.57–1.73 eV. For a photovoltaic absorber the allowed gaps are the relevant quantities, so the abstract misleads. The body text is mostly fine, but the abstract should be corrected.\n\nThe bigger problem is stability. The thermodynamic check in Eq. 4 tests only A2SiI6 -> ASiI3 + AI3. For vacancy-ordered A2B(IV)X6 perovskites, the conventional competing set is 2AI + SiI4, and SiI4 is a known stable molecular solid. That channel is never computed. A positive enthalpy against ASiI3 + AI3 (41–76 meV/atom) therefore says very little about synthesizability. This is load-bearing because the paper's headline claim is “robust high-performance” and “good stability at high temperature.” Elastic constants do not detect thermodynamic instability; the MD runs are short and start from the assumed Fm3m structure. The paper also admits Cs2SiI6 is dynamically unstable at 300/400 K, so the family-level stability claim is narrower than the abstract implies.\n\nThe citation pattern is normal, the method is standard, and no parameters are fitted to target properties. This is a genuine prediction, not a retrofitted one.\n\nNet: a plausible candidate-family screen, not an established material. The electronic and optical calculations are useful, and the parity-forbidden insight is a good catch. But the thermodynamic evidence is insufficient to support the “robust” claim, and the abstract needs to report the allowed gaps. I would send it to a serious referee with the expectation of major revision, and I would not cite it as a stable photovoltaic family until the 2AI + SiI4 channel and, ideally, a polymorph search are addressed.","headline":"A genuine first-principles screen of a new Si-based double-perovskite family, but the abstract oversells the band gaps and the stability test omits the obvious competing phases.","tokens_in":12447,"tokens_out":3301,"would_cite":false,"duration_ms":35675,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["71.20.-b","78.20.Bh","78.20.Ci","71.15.Mb"],"model":"deepseek-v4-flash","headline":"First-principles calculations predict that three silicon-based double perovskites—K$_2$SiI$_6$, Rb$_2$SiI$_6$ and Cs$_2$SiI$_6$—are direct-gap, lead-free solar absorbers.","keywords":["double perovskite","lead-free absorber","silicon halide","vacancy-ordered perovskite","photovoltaic material","hybrid density functional","band structure","effective mass"],"falsifier":"Synthesize K$_2$SiI$_6$ or Rb$_2$SiI$_6$ and measure the crystal structure, band gap, and optical absorption. If X-ray diffraction shows a structure other than $Fm\\bar{3}m$, if the measured gap is far outside the predicted 0.7–1.2 eV range, or if the strong absorption onset is far from the predicted 1.57–1.73 eV allowed transition, the central prediction fails.","tokens_in":11451,"feed_emoji":"☀️","tokens_out":20145,"duration_ms":175925,"temperature":0.7,"pith_summary":"This paper predicts a new family of lead-free photovoltaic absorbers: the silicon-based vacancy-ordered double perovskites K$_2$SiI$_6$, Rb$_2$SiI$_6$ and Cs$_2$SiI$_6$. Hybrid density-functional calculations give direct band gaps of 0.84–1.15 eV (0.71–0.99 eV with spin-orbit coupling), small electron effective masses of 0.17–0.23 $m_0$, and broad absorption in the visible range. The potassium and rubidium compounds are also predicted to be mechanically stable, thermodynamically stable against decomposition into ASiI$_3$ + AI$_3$, and dynamically stable in 400 K simulations; Cs$_2$SiI$_6$ is found dynamically unstable at those temperatures. If the predictions hold, the family offers a nontoxic, silicon-based route to perovskite solar cells, with K$_2$SiI$_6$ identified as the best optical performer.","feed_headline":"Three silicon double perovskites predicted as lead-free absorbers","feed_subtitle":"K2SiI6 and Rb2SiI6 have direct band gaps near 1 eV and survive 400 K, so they are promising synthesis targets.","key_machinery":"The central object is the vacancy-ordered double-perovskite structure in space group $Fm\\bar{3}m$, formula A$_2$BX$_6$, in which half the B-site octahedra are empty and the remaining [BX$_6$]$^{2-}$ octahedra are isolated, with A cations occupying the voids between them. The argument is carried by the electronic structure of the [SiI$_6$]$^{2-}$ unit: Si-$3s$ and I-$5p$ states hybridize into a wide, isolated lower conduction band and I-$5p$-dominated valence bands, and the A-site cation sets the lattice size that tunes the gap. The calculations use a screened hybrid density functional plus spin-orbit corrections for band gaps, effective masses, and optical spectra, and they use parity and transition dipole moments at the zone centre to identify which band-to-band transitions are optically allowed. Stability is established from decomposition enthalpies, elastic constants, and finite-temperature molecular-dynamics trajectories.","core_discovery":"The central claim is that A$_2$SiI$_6$ (A = K, Rb, Cs) form a previously unstudied class of inorganic double perovskites with photovoltaic-grade electronic structure. All three are direct-gap semiconductors, with hybrid-functional gaps of 0.84 eV (K), 0.96 eV (Rb) and 1.15 eV (Cs), and the gap rises with A-site cation size and with halogen electronegativity across the wider A$_2$SiX$_6$ series. The lower conduction band is an isolated, dispersive band built from Si-$3s$ and I-$5p$ states, giving electron effective masses of 0.17–0.23 $m_0$ while holes are heavier, which the authors read as making these materials n-type-oriented absorbers. A parity analysis at the zone centre shows the direct transition between the valence-band maximum and the conduction-band minimum is dipole-forbidden, so the optical absorption is carried by an allowed transition from a lower valence band, with dipole-allowed gaps of 1.57–1.73 eV. Positive decomposition enthalpies (41–76 meV per atom), elastic constants satisfying the cubic stability criteria, and 400 K molecular-dynamics stability for K$_2$SiI$_6$ and Rb$_2$SiI$_6$ underpin the stability claim.","pith_inferences":["Because the predictions are benchmarked against other vacancy-ordered double perovskites rather than against any measured silicon double perovskite, the absolute band gaps carry a calibration uncertainty that a single optical measurement would resolve.","The parity-forbidden direct edge implies the nominal band gap is not the absorption onset: devices would absorb through the allowed transition at 1.57–1.73 eV, so sub-gap absorption and defect behaviour deserve direct study.","The monotonic gap trends across cation and halogen suggest that alloying, for example K/Rb mixing or I/Br mixing, could tune the gap across the 0.8–1.2 eV range, possibly even stabilizing the cesium member.","The effective-mass asymmetry points toward n-type transport, but the paper does not compute dopability or defect levels, so the actual carrier polarity in a working device could differ."],"forward_implications":["If K$_2$SiI$_6$ and Rb$_2$SiI$_6$ can be synthesized, they are concrete lead-free, silicon-based candidates for perovskite solar-cell absorbers.","The predicted gaps fall in the 0.9–1.2 eV rear-cell window used for perovskite-perovskite tandem devices, and K$_2$SiI$_6$'s allowed optical transition at 1.57 eV sits close to the single-junction detailed-balance optimum.","The small electron effective masses imply fast electron transport, so devices built from these absorbers should be designed to collect electrons preferentially.","Cs$_2$SiI$_6$'s predicted dynamic instability at 300–400 K means experimental effort should concentrate on the potassium and rubidium members.","A successful synthesis of any member would give experimenters a benchmark for the hybrid-functional prediction and open the wider A$_2$SiX$_6$ family to compositional tuning."],"supporting_citations":[{"why":"Establishes Cs2SnI6 and Cs2TeI6 as vacancy-ordered double-perovskite semiconductors, providing the structure type and the defect-tolerance context that motivate A2SiI6.","marker":"[28]"},{"why":"Shows that an isostructural iodide, Cs2SnI6, can be synthesized and used as a hole conductor, supporting the experimental feasibility of silicon-based iodides.","marker":"[29]"},{"why":"Provides the titanium-based vacancy-ordered double perovskites with tunable 1.0–1.8 eV gaps, used as the benchmark for why the hybrid-functional method is trusted for this family.","marker":"[30]"},{"why":"Supplies the screened hybrid density functional used for all electronic-structure and optical predictions in the paper.","marker":"[36]"}],"fun_headline_variants":["Silicon double perovskites predicted as stable lead-free absorbers","New Si-based double perovskites promising for efficient solar cells","Earth-abundant Si double perovskites with ideal band gaps for photovoltaics","Computational prediction of stable A2SiI6 double perovskites for solar cells","Lead-free Si double perovskites show stability and photovoltaic-grade gaps"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The predictions assume the compounds crystallize in the cubic vacancy-ordered double-perovskite structure and that their only relevant decomposition is into ASiI$_3$ plus AI$_3$; no competing crystal structures or other decomposition products are tested.","fun_headline_variants_meta":{"raw":{"variants":["Silicon double perovskites predicted as stable lead-free absorbers","New Si-based double perovskites promising for efficient solar cells","Earth-abundant Si double perovskites with ideal band gaps for photovoltaics","Computational prediction of stable A2SiI6 double perovskites for solar cells","Lead-free Si double perovskites show stability and photovoltaic-grade gaps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000244,"raw_usage":{"total_tokens":1563,"prompt_tokens":1007,"completion_tokens":556,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":623,"completion_tokens_details":{"reasoning_tokens":464}},"tokens_in":623,"tokens_out":556,"duration_ms":6301,"temperature":1.0,"reasoning_tokens":464,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:51:18.508631+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Synthesize K$_2$SiI$_6$ or Rb$_2$SiI$_6$ and measure the crystal structure, band gap, and optical absorption. If X-ray diffraction shows a structure other than $Fm\\bar{3}m$, if the measured gap is far outside the predicted 0.7–1.2 eV range, or if the strong absorption onset is far from the predicted 1.57–1.73 eV allowed transition, the central prediction fails.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes Cs2SnI6 and Cs2TeI6 as vacancy-ordered double-perovskite semiconductors, providing the structure type and the defect-tolerance context that motivate A2SiI6."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that an isostructural iodide, Cs2SnI6, can be synthesized and used as a hole conductor, supporting the experimental feasibility of silicon-based iodides."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the titanium-based vacancy-ordered double perovskites with tunable 1.0–1.8 eV gaps, used as the benchmark for why the hybrid-functional method is trusted for this family."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the screened hybrid density functional used for all electronic-structure and optical predictions in the paper."}],"review_version":1}