{"id":"d554f5e1-f29c-4f20-984d-7b94405e7435","arxiv_id":"2507.08140","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Thulium atoms in solid argon show two nearly axial crystal-field sites, magnetic-electric transition character, and optically resolvable Zeeman shifts usable for magnetometry.","lead":"Researchers mapped the crystal-field energy levels of thulium atoms embedded in solid argon using high-resolution laser spectroscopy, identifying two stable trapping sites and resolving hyperfine and Zeeman structure. The same all-optical readout detects millitesla magnetic fields, a step toward nanoscale magnetometry in rare-gas hosts.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Site II lacks the independent hyperfine validation that supports Site I, and its observed 2F7/2 spacings deviate substantially from the single-O2^0 interval rule; the two-site axial-crystal-field claim is therefore not yet equally supported.","rationale":"I read the paper as a spectroscopy plus proof-of-principle magnetometry contribution. The central claim has three components: (1) two stable trapping sites each with a nearly axial crystal field and fitted beta_2^0; (2) magnetic-electric transition character; (3) all-optical detection of mT fields. Component (2) is reasonably supported by the 11-line-per-site counting, the presence of |Delta mJ| = 2 transitions, and the polarized spectroscopy in Sec. V, even though the M1+FD/M1+PQ decomposition is degenerate. Component (3) is directly demonstrated by the ratiometric Zeeman measurement in Fig. 4, with modest sensitivity. Component (1) is the load-bearing modeling claim. Site I is independently validated by the parameter-free hyperfine comparison using literature constants; the same validation is not shown for Site II. Moreover, the Site II fit residual in the upper part of the 2F7/2 manifold is large enough, roughly 0.31 cm^-1 at the 3-sigma level, that the simple axial model's level ordering and labels are not forcibly determined by the data. The paper's statement that higher-order terms only marginally improve agreement is plausible but not quantified, and if such terms are needed, the quoted beta_2^0 is an effective parameter rather than a direct measure of the axial crystal field. This does not invalidate the paper; it makes the two-site crystal-field determination conditional on the same kind of hyperfine check that was already performed for Site I. The reader's weakest-assumption statement pointed to the axial and spin-decoupled model generally; I partially agree but localize the weak point to Site II. I therefore do not move the verdict: it remains conditional.","tokens_in":10280,"tokens_out":16313,"duration_ms":176631,"concrete_test":"Record high-resolution fluorescence spectra of the narrowest annealed Site II lines, e.g. the |mJ|: 1/2 ↔ 1/2 or 1/2 ↔ 3/2 transitions assigned in Table I, and compute the predicted hyperfine splittings and relative amplitudes for Site II using the proposed mJ assignments and the literature 169Tm hyperfine constants cited as [37], exactly as was done for Site I in Fig. 2 and Sec. IV. If the predicted hyperfine pattern fails to match the observed spectrum within the linewidth, the Site II axial assignments and beta_2^0=-0.1433 would need revision; if it matches, the main missing validation for the two-site claim is supplied.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The weakest load-bearing point is not the existence of crystal-field splitting, but the assertion that both sites are separately described by a single axial O2^0 Hamiltonian with the quoted beta_2^0 values. The paper's strongest independent check, the parameter-free hyperfine comparison in Sec. IV, is presented only for Site I. Site II's level assignments come from the same combinatoric fit that produced the 'observed' levels, and the internal interval-rule check is only approximate. In Table I, the fitted 2F7/2 spacings for Site II are 1.98, 1.51, and 0.28(9) cm^-1, whereas the O2^0 model with beta_2^0=-0.1433 predicts 1.7839, 1.1893, and 0.5946 cm^-1; the top spacing is off by about 0.31 cm^-1, roughly 3 times the quoted uncertainty. That is a larger fractional error than anywhere in Site I, and it is precisely the spacing that would set the ordering and labels of the ±3/2 and ±1/2 doublets. If higher-rank crystal-field terms are invoked to repair Site II, the 'nearly axial, single O2^0' description and the quoted beta_2^0 become effective parameters whose physical interpretation is less secure. Since the abstract and conclusion assert that the crystal field is determined to be nearly axial in both sites, Site II's missing independent validation is the weakest load-bearing link.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a high-resolution laser spectroscopy study of the 1140 nm 2F7/2 → 2F5/2 transition of neutral thulium atoms implanted in solid argon. Pump-probe spectroscopy is used to separate two stable trapping sites, each showing 11 spectral lines. The authors model both sites with a single axial Stevens operator O2^0, fit the parameter beta_2^0 separately for each site, and assign mJ labels. For Site I, resolved hyperfine structure is compared with a parameter-free prediction based on hyperfine constants from Ref. [37], confirming the assignments. Selection rules and polarized spectra are interpreted as magnetic-electric character from forced-dipole and/or pseudoquadrupole admixtures. Zeeman-resolved spectra and a two-color fluorescence ratio are used to demonstrate all-optical detection of mT-level DC magnetic fields, with an estimated sensitivity of 330 muT/sqrt(Hz).","tokens_in":10530,"tokens_out":7653,"duration_ms":79312,"significance":"If the two-site axial crystal-field model is correct, this is the first site-resolved crystal-field and Zeeman characterization of Tm in a rare-gas matrix and the first demonstration of microwave-free all-optical magnetometry in a matrix-isolated atom. The paper's strengths are substantial: the Site I hyperfine comparison in Sec. IV is a genuine external check with no adjustable parameters; the measured ensemble linewidths of 129 MHz are impressively narrow; and the data are openly available. The significance is conditional, however, because the second site lacks the same independent validation and because the intensity decomposition underlying the 'magnetic-electric' conclusion is not uniquely determined.","major_comments":[{"comment":"The single-O2^0 model is not equally supported for Site II. The observed 2F7/2 spacings for Site II are 1.98, 1.51, and 0.28(9) cm^-1, while the fitted beta_2^0 = -0.1433 predicts 1.7839, 1.1893, and 0.5946 cm^-1; the uppermost spacing disagrees by about 0.31 cm^-1, roughly three times the quoted uncertainty. Because no independent hyperfine check is presented for Site II, the mJ assignments and beta_2^0 value for that site rest on the same combinatoric fit that generated the 'observed' levels. The abstract and conclusion state that the crystal field is nearly axial in both sites; this claim needs either additional Site II validation or a more cautious statement of the confidence in the Site II parameters.","section":"Section III, Table I"},{"comment":"The agreement between observed and predicted energies in Table I is partly self-consistent by construction. The 'observed' crystal-field levels are obtained from a combinatoric fit to the line positions, and the predicted levels are recomputed from beta_2^0 values fitted to those same observed energies. This circularity should be stated explicitly in the methods; the genuinely independent confirmation is the Site I hyperfine comparison in Sec. IV, not the energy-level agreement in Table I. The paper should clarify which comparisons are independent tests and which are fitting diagnostics.","section":"Section III and Table I"},{"comment":"The identification of the electric admixture is not unique. The low-power fluorescence fit yields a M1/FD mixture of 70/30 with T = 8.12 K, while the M1/PQ fit yields 86/14 with T = 6.99 K, and the text then concludes that a combination of M1 and PQ 'best explain' the selection rules without a quantitative model-comparison or reported uncertainties. This matters because the 'magnetic-electric' transition-type claim and the interpretation of the polarized spectra in Sec. V depend on the decomposition. The authors should either provide a model-selection criterion with uncertainties and a common temperature treatment, or state the conclusion as 'M1 with forced-dipole and/or pseudoquadrupole admixture'.","section":"Section IV and Figure 2"}],"minor_comments":[{"comment":"The text contains small typographical errors ('detection protols' in the Introduction and 'A single sampled held at base temperature' in Sec. II) that should be corrected.","section":"Sections I and II"},{"comment":"The DOI printed in the header, '10.1103/x5zx-ykff', appears malformed or incomplete; the correct publisher DOI should be used.","section":"Header/DOI"},{"comment":"The units of beta_2^0 are not stated in the text or Table I; since the predicted energies are in cm^-1, the values 0.3142 and -0.1433 should be explicitly labeled as cm^-1.","section":"Section III"},{"comment":"The phrase 'calibration-free estimate' for the ratiometric field measurement should be qualified: the Monte Carlo simulations use fitted linewidths, amplitudes, and background, so the ratio is not calibration-free in an absolute sense; the authors should define precisely what they mean by calibration-free.","section":"Section V"}],"recommendation":"major_revision","confidential_remarks":"The experimental work appears careful and the open data and Site I hyperfine check are strong assets. My concerns are confined to overreach in the Site II axial-model claim and in the intensity decomposition; both are fixable with additional analysis or more cautious language. I do not see grounds for rejection, and the paper is a good fit for the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, this is a genuinely good experimental paper, with one real soft spot that the authors should fix. The new capabilities are real: pump-probe spectroscopy separates two trapping sites for Tm in solid Ar, the axial O2^0 crystal field with a single parameter per site accounts for most of the level structure, and the hyperfine comparison for Site I is parameter-free and convincing. The Zeeman and magnetometry demonstration is a credible proof of principle, even if the sensitivity is far from NV centers.\n\nThe strongest part is Sec. IV. Using hyperfine constants from Mishin et al. to predict the resolved hyperfine structure with no fit parameters is exactly the kind of independent check that makes an assignment trustworthy. That check lands for Site I, and it gives me confidence the overall picture is not a coincidence.\n\nNow the soft spots. The stress-test note is right: Site II does not get the same hyperfine validation, and the 2F7/2 spacings deviate from the single-parameter O2^0 model by more than the quoted uncertainties. The top spacing is 0.28(9) cm^-1 observed versus 0.59 predicted—roughly three times the uncertainty. That means the claim that both sites are 'nearly axial' is not equally supported. The authors should either provide a Site II hyperfine measurement, or explicitly weaken the abstract and conclusion to say the axial model is well established for Site I and approximate for Site II. Related, Table I labels the model energies as 'Predicted' when beta_2^0 was fit to exactly those levels. That is misleading; they should be called 'fit' or cross-validated. The transition-type conclusion is also softer than the text suggests: the M1+FD and M1+PQ fits are degenerate, and neither reproduces the relative amplitudes exactly. That is not fatal, but the 'magnetic-electric' claim should be stated as consistent with the data rather than determined.\n\nMinor: the data repository has no URL or hash, and the Monte Carlo parameters are not fully specified. Both are easily fixed.\n\nVerdict: this deserves a serious referee. It is a solid advance in matrix-isolated atom spectroscopy, and the hyperfine confirmation for Site I is the kind of result that should be in the literature. I would accept for review with the expectation of moderate revision, mainly to relabel the fits and to bring the Site II claims in line with the evidence.","headline":"Solid experimental paper on Tm:Ar spectroscopy; the axial-crystal-field story is convincing for Site I, but Site II needs either independent validation or a more cautious claim.","tokens_in":11186,"tokens_out":3004,"would_cite":true,"duration_ms":28273,"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":"Neutral thulium embedded in solid argon splits its 1140 nm transition into 22 sharp crystal-field lines from two stable sites, and the same lines resolve millitesla magnetic fields all-optically.","keywords":["thulium","matrix isolation","solid argon","crystal field","Stevens operator","hyperfine structure","Zeeman effect","optical magnetometry"],"falsifier":"Resolve the hyperfine structure of Site II (or of a single crystal axis) at sub-100 MHz resolution and compare the number of lines and relative intensities with the M1+PQ model; the appearance of a $|\\Delta m_J|=3$ transition, or a mismatch of the 9/5 interval rule beyond the fitted uncertainties, would rule out the purely axial single-parameter model.","tokens_in":9951,"feed_emoji":"🧲","tokens_out":7636,"duration_ms":78220,"temperature":0.7,"pith_summary":"This paper reports laser spectroscopy of the 1140 nm inner-shell transition of neutral thulium atoms embedded in a solid argon matrix. Pump-probe measurements separate the spectrum into two thermally stable trapping sites, each producing eleven crystal-field-split lines. The authors show that a single axial Stevens operator $O_2^0$ with a fitted parameter $\\beta_2^0$ reproduces the level spacings in both sites, and that the transition becomes magnetic-electric (M1 mixed with forced-dipole or pseudoquadrupole components) in the argon host. Hyperfine-resolved Zeeman shifts are used to demonstrate all-optical detection of millitesla magnetic fields with a DC sensitivity of $330\\ \\mu\\mathrm{T}/\\sqrt{\\mathrm{Hz}}$. This matters because matrix-isolated atoms are candidate nanoscale sensors and identical quantum emitters, and this is the first all-optical magnetometry in a rare-gas matrix that avoids microwave or radio-frequency radiation.","feed_headline":"Two trapping sites split thulium's 1140 nm line into 22 sharp lines","feed_subtitle":"Crystal-field fits and Zeeman-resolved hyperfine lines enable all-optical millitesla magnetometry in a solid argon matrix.","key_machinery":"The load-bearing object is the axial Stevens operator $O_2^0 = 3L_z^2 - L(L+1)$ acting with $L=3$ for the f-electron shell, whose single coefficient $\\beta_2^0$ sets the crystal-field splittings. This operator preserves $m_J$ as a good quantum number and yields the observed 9/5 scaling between the $J=7/2$ and $J=5/2$ manifolds and the interval rule. The paper combines this operator with pump-probe saturation spectroscopy to separate sites, with ab initio hyperfine structure (from Tm constants measured in an optical lattice) to confirm assignments, and with a model of M1, forced-dipole, and pseudoquadrupole amplitudes to identify the transition's magnetic-electric character.","core_discovery":"The central claim is that the $^2F_{7/2}\\to{}^2F_{5/2}$ ground spin-orbit transition of Tm in solid Ar is not a single line but a superposition of two site-specific crystal-field manifolds. In each site the degeneracy is fully lifted into Kramers doublets, producing eleven allowed lines (one $|\\Delta m_J|=3$ transition forbidden), and the measured level structure is reproduced by a purely axial crystal field with $\\beta_2^0=0.3142$ (Site I) and $-0.1433$ (Site II). The paper further claims that the host turns the gas-phase magnetic-dipole transition into a mixed magnetic-electric transition, most plausibly M1 with pseudoquadrupole (PQ) contributions plus some forced-dipole (FD) character, as inferred from selection rules, hyperfine amplitudes, and polarized spectroscopy. Ab initio hyperfine predictions using constants from optical-lattice measurements confirm the $m_J$ assignments for Site I. Finally, Zeeman shifts of the narrow hyperfine lines at fields of order 10 mT are resolved directly by fluorescence, giving a calibration-free ratiometric magnetometer with an estimated DC sensitivity of $330\\ \\mu\\mathrm{T}/\\sqrt{\\mathrm{Hz}}$.","pith_inferences":["If the magnetic-electric character comes mainly from the dielectric pseudoquadrupole channel, then the same transition in heavier or denser rare-gas hosts (neon, krypton) should show a systematically different M1/PQ ratio, offering a clean test of the dielectric-coupling mechanism.","The two-site structure suggests a route to monitor crystal growth and annealing by tracking the relative site populations; a future experiment could correlate the Site I/Site II fluorescence ratio with annealing temperature and time.","Because the Zeeman response depends on the angle between the field and crystal axis, an ensemble with random axes acts as a distributed set of vector probes; extracting the full field direction from the line shapes might be possible without isolating a single site.","The reported sensitivity is shot-noise limited at 330 $\\mu$T/$\\sqrt{\\mathrm{Hz}}$; cavity enhancement or optical pumping out of the lower $J=5/2$ level could plausibly reach the nT/$\\sqrt{\\mathrm{Hz}}$ class while retaining the all-optical, RF-free advantage."],"forward_implications":["If the axial crystal-field model is right, each trapping site's local symmetry is lower than tetrahedral, and a single parameter predicts all line positions to within the stated uncertainties.","The confirmed $m_J$ assignments and hyperfine structure make the 1140 nm line a usable spectroscopic probe of the argon matrix's local order and annealing state.","Because Zeeman shifts are resolved directly in fluorescence, DC magnetic fields in the millitesla range can be measured without applying microwaves or radio-frequency fields, a first for matrix-isolated atoms.","The measured 129 MHz ensemble linewidth, if reduced toward the 30 Hz transform limit by annealing and cooling, would improve the estimated $330\\ \\mu\\mathrm{T}/\\sqrt{\\mathrm{Hz}}$ sensitivity by orders of magnitude.","The same pump-probe method can be extended to other thulium-doped rare-gas solids and to selective bleaching, potentially isolating a single crystal axis for vector magnetometry."],"supporting_citations":[{"why":"This earlier result establishes that the 1140 nm transition of Tm in rare-gas crystals can be narrow, motivating high-resolution spectroscopy.","marker":"[17]"},{"why":"This prior work provides the high-resolution spectrum of Tm implanted in solid noble gas crystals that the present paper extends with pump-probe site separation.","marker":"[18]"},{"why":"This measurement supplies the ab initio hyperfine constants used to confirm the $m_J$ assignments in Site I.","marker":"[37]"},{"why":"This study identifies the gas-phase 1140 nm transition and clock properties, used to assign each site to neutral Tm.","marker":"[15]"},{"why":"This reference introduces the Stevens operator-equivalent formalism that underlies the axial crystal-field model.","marker":"[31]"},{"why":"Judd-Ofelt theory from these papers is used to model forced-dipole admixtures in the magnetic-electric transition.","marker":"[38, 39]"},{"why":"These works supply the pseudoquadrupole transition mechanism that the paper invokes to explain the M1+PQ selection rules and hypersensitive lanthanide intensities.","marker":"[40, 41]"},{"why":"This reference gives the local-field dielectric correction used to account for the shortened excited-state lifetime in argon.","marker":"[44]"}],"fun_headline_variants":["Thulium in argon: 22 sharp lines, two sites, one magnetometer","Two-site crystal field splits thulium line into 22","Zeeman shifts in thulium-doped argon enable mT sensing","All-optical DC magnetometry using thulium in argon","Crystal field and Zeeman effect map thulium in argon"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that each trapping site's local environment is accurately described by a single axial Stevens term, with the f-electron spin decoupled from the crystal; if the true site symmetry is lower or spin coupling is significant, the fitted parameters, level assignments, and hyperfine comparison all collapse.","fun_headline_variants_meta":{"raw":{"variants":["Thulium in argon: 22 sharp lines, two sites, one magnetometer","Two-site crystal field splits thulium line into 22","Zeeman shifts in thulium-doped argon enable mT sensing","All-optical DC magnetometry using thulium in argon","Crystal field and Zeeman effect map thulium in argon"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000818,"raw_usage":{"total_tokens":3627,"prompt_tokens":1035,"completion_tokens":2592,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":651,"completion_tokens_details":{"reasoning_tokens":2499}},"tokens_in":651,"tokens_out":2592,"duration_ms":17550,"temperature":1.0,"reasoning_tokens":2499,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:27:49.711759+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Resolve the hyperfine structure of Site II (or of a single crystal axis) at sub-100 MHz resolution and compare the number of lines and relative intensities with the M1+PQ model; the appearance of a $|\\Delta m_J|=3$ transition, or a mismatch of the 9/5 interval rule beyond the fitted uncertainties, would rule out the purely axial single-parameter model.","supporting_citations":[{"cited_title":"Golovizin, E","cited_arxiv_id":null,"evidence_quote":"This earlier result establishes that the 1140 nm transition of Tm in rare-gas crystals can be narrow, motivating high-resolution spectroscopy."},{"cited_title":"Ishikawa, A","cited_arxiv_id":null,"evidence_quote":"This prior work provides the high-resolution spectrum of Tm implanted in solid noble gas crystals that the present paper extends with pump-probe site separation."},{"cited_title":"Crepin-Gilbert and A","cited_arxiv_id":null,"evidence_quote":"This measurement supplies the ab initio hyperfine constants used to confirm the $m_J$ assignments in Site I."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This study identifies the gas-phase 1140 nm transition and clock properties, used to assign each site to neutral Tm."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference introduces the Stevens operator-equivalent formalism that underlies the axial crystal-field model."},{"cited_title":"G¨ orller-Walrand and K","cited_arxiv_id":null,"evidence_quote":"This reference gives the local-field dielectric correction used to account for the shortened excited-state lifetime in argon."}],"review_version":1}