REVIEW 4 major objections 5 minor 16 references
Unusual magnetic order in Eu$_{10}$Hg$_{55}$
T0 review · 4 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash
Pith's one-line read Eu10Hg55 has a fragile magnetic ground state with a magnetization pole reversal, driven by mixed-valence europium on four crystallographic sites.
desk verdict A careful, honest study of a hard air-sensitive compound; the magnetic order is real, but the mixed-valence and ferrimagnetic interpretation is much weaker than the data can carry. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the cage-like noncentrosymmetric P6 structure of Eu10Hg55 with its 65-atom unit cell, four distinct Eu positions (Eu1–Eu4) and 17 mercury positions. The Eu sites have coordination numbers 14–16 and volumes ranging from 99.6 to 116.1 Å3, which the paper proposes host Eu2+ and Eu3+ in proportions set by the measured saturation moment. The magnetic mechanism is a weak ferromagnetic coupling between Eu2+ moments (positive Weiss temperature 6.4–11.7 K) with an additional weak antiferromagnetic inter-sublattice exchange; this two-sublattice ferrimagnet model explains the magnetization pole reversal and the two transitions.
What would settle it
A 151Eu Mössbauer spectrum of an air-free Eu10Hg55 sample showing only the Eu2+ isomer shift—or a single-crystal structure refinement giving the composition Eu8Hg57—would show that the moment deficit comes from stoichiometry, not from Eu3+, and so falsify the mixed-valence core of the paper.
Extended reading notes
Core claim
The paper reports that in Eu10Hg55 the europium moments, diluted in a large cage-like mercury network, order below T1 = 5.5 K and then undergo a spin reorientation at T2 = 4.3 K. The low-field zero-field-cooled and field-cooled magnetizations are opposite in sign and nearly symmetric, which the authors interpret as a magnetization pole reversal—the signature of a weak ferrimagnet formed by two antiferromagnetically coupled europium sublattices, each internally ordered. From the saturation moment at 7 T (6.3–6.8 Bohr magnetons per Eu, well below the 7 Bohr magnetons of pure Eu2+), they estimate an average Eu valence between 2.02 and 2.18, implying that roughly one in five to one in ten europi
Load-bearing premise
The mixed-valence interpretation assumes the reduced saturation moment comes from nonmagnetic Eu3+ ions rather than from off-stoichiometry or Eu/Hg mixing; that assumption is untested because air sensitivity prevented single-crystal diffraction and the only direct valence probe, HERFD XANES, was invalidated by oxidation.
Editorial extensions
If this is right
- If the mixed-valence assignment holds, Eu10Hg55 joins a small family of intermetallics where Eu2+ and Eu3+ coexist on distinct sites, and its low ordering temperature is set by the dilution of magnetic moments by nonmagnetic Eu3+.
- The magnetization pole reversal means the ground state is a weak ferrimagnet with at least two magnetic sublattices; modest fields (0.1–0.5 T) can switch between magnetic configurations, so the H–T phase diagram contains field-induced phases.
- Applied pressure decreases the saturation moment by about 5% at 12.8 GPa, consistent with pressure pushing europium toward the smaller Eu3+ state; this makes Eu10Hg55 a pressure-tunable mixed-valence system.
- Partial substitution of Eu by Ca or Sr, which form isostructural superconducting compounds, could continuously tune the magnetic ground state and possibly bring superconductivity into the same lattice.
- ESR shows a Korringa slope three times larger than in conventional Eu metals, indicating strong coupling of the local Eu2+ moments to conduction electrons, with implications for transport and heavy-fermion-like behavior.
Reading between the lines
- The explicit site assignment—Eu3+ on the smaller Eu1 site and magnetic Eu2+ on the larger sites—could be tested directly by 151Eu Mössbauer spectroscopy on an oxygen-free crystal; a single Eu2+ subspectrum would falsify it.
- If the pole reversal is intrinsic, the same two-sublattice signature should appear in isostructural compounds where one of the Eu sites is replaced by a magnetic rare earth; searching that family could locate other pole-reversing ferrimagnets.
- Tuning the Eu:Hg ratio within the narrow homogeneity range should move the average valence and the two transition temperatures in a predictable direction, giving a synthesis-level test of the mixed-valence scenario.
- The strong Korringa slope implies the conduction electrons are coupled to the Eu moments; measuring transport in a substituted superconducting analogue could reveal whether magnetic fluctuations persist into the superconducting state.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports a detailed experimental study of the intermetallic compound Eu10Hg55, which crystallizes in a noncentrosymmetric structure with four distinct Eu sites and 17 Hg sites. Magnetization, specific heat, ESR, pressure, and HERFD XANES data are presented for air-sensitive single crystals. The authors claim that Eu may be present in both Eu2+ and Eu3+ states (average valence 2.02–2.18 per sample), that the compound orders magnetically below T1 = 5.5 K with a second transition at T2 = 4.3 K, and that the low-field magnetization shows a pole reversal reminiscent of a weak ferrimagnetic ground state. The evidence for magnetic order is supported by specific heat, magnetization isotherms, and ESR line broadening; however, the mixed-valence interpretation rests almost entirely on a reduced saturation moment relative to the Eu2+ free-ion value of 7 μB, with the only direct valence probe (XANES) invalidated by oxidation.
Significance. If the claims are established, Eu10Hg55 would be a rare example of an inhomogeneous mixed-valence Eu intermetallic with a fragile, possibly ferrimagnetic, ground state. The paper is valuable as a first characterization of a difficult, air-sensitive system, and the authors are transparent about measurement limitations. The strengths include the use of large single crystals, corroborating bulk probes (magnetization, specific heat, ESR), and explicit statements about what could and could not be measured. However, the central mixed-valence and ferrimagnetic conclusions are not yet conclusive, and at least one piece of data (magnetic entropy) appears to be in tension with the mixed-valence scenario. The significance would be much higher if a direct valence-sensitive probe (e.g., 151Eu Mössbauer on unoxidized samples) were provided.
major comments (4)
- The average Eu valence is extracted solely from M(7 T, 2 K) assuming Eu2+ contributes 7 μB and Eu3+ contributes 0. This decomposition is not unique: a reduced saturation moment could also arise from Eu/Hg mixing or off-stoichiometry, as the authors acknowledge in footnote 3. More importantly, the magnetic entropy shown in the inset of Fig. 2(c) is stated to reach Rln8 at T = 10 K per mole Eu. For a sample with average valence 2.18 (18% Eu3+), the expected magnetic entropy would be only 0.82 Rln8, a difference far larger than typical experimental scatter. The authors do not identify which sample was used for the specific-heat measurement or reconcile this apparent contradiction. If the specific-heat sample has a valence close to 2.0, that should be stated; if it is representative of the compound, the mixed-valence scenario with valence up to 2.18 is seriously weakened.
- The magnetization pole reversal is observed only at B = 0.005 T, in a regime where the authors state that absolute values are not precise due to remnant field in the magnet. While the zfc/fc symmetry is suggestive of a compensation effect, no hysteresis loops or quantitative model are presented. The interpretation as a 'weak ferrimagnetic ground state' relies on the same moment decomposition that is questioned above, and the phase diagram in Fig. 5(a) is explicitly labeled preliminary. The claim would be more convincing if the authors provided field-cooled magnetization at several fields with error assessment, or at least a clear statement of how a constant field offset would affect the sign change.
- The HERFD XANES measurement is invalidated by oxidation of the sample, as the authors concede: the sample decomposed into Hg and Eu2O3, and the fit gives 0.8 Eu3+. An oxidized sample would show Eu3+ regardless of whether the pristine sample contained any Eu3+. The sentence 'The XANES analysis indicates that both Eu2+ and Eu3+ are present' is therefore misleading, even with the caveat that the ratios are not quantitative. As presented, this section does not support the mixed-valence claim; it should be rephrased to state that the measurement was unsuccessful in determining the pristine valence.
- The ~5% decrease in M at 7 T and 10 K under 12.8 GPa is interpreted as partial suppression of Eu2+ to Eu3+. However, the authors state that the magnetization at 10 K 'did not reach saturation', so comparing M at a single field is not a reliable measure of the moment. Pressure can also modify exchange interactions and lattice parameters, affecting the magnetization through mechanisms other than valence change. This piece of evidence is too indirect to support the valence-pressure connection and should be presented as preliminary speculation unless additional fields/temperatures or a proper Curie-Weiss analysis under pressure are shown.
minor comments (5)
- [Abstract vs. §2 Fig. 3] The abstract gives T1 = 5.5 K and T2 = 4.3 K, while the low-field magnetization data are described as having features at T1 ≈ 4.98 K and T2 = 4.71 K. Please clarify whether these differences are due to the applied field or to different samples, and define the zero-field transition temperatures explicitly.
- [Fig. 2(c) inset] The inset axis label appears to read 'SM (J/molEuK2)', which is not a valid unit for entropy. This should be corrected to J/(mol Eu K).
- [§2, specific heat] Please specify which sample (Sample 1–4) was used for the specific-heat measurement, since the mixed-valence analysis reports sample-to-sample variation.
- [Fig. 5(b)] The XANES spectrum is dominated by oxidation products. It would be useful to show the spectrum of an intentionally oxidized reference (e.g., Eu2O3) for comparison, or to remove the quantitative peak fit entirely, as it could be misinterpreted.
- [§2, ESR] The ESR intensity is said to be 'qualitatively similar' to the bulk susceptibility. A quantitative comparison of the ESR intensity (or integrated area) with the bulk χ(T) would be a useful check on whether a fraction of Eu is ESR-silent (Eu3+).
Circularity Check
Mixed-valence claim is the saturation-moment deficit relabeled; off-stoichiometry alternative is excluded via the authors' own in-prep data, so the Eu3+ scenario is partially circular.
-
fitted input called prediction
[Analysis of crystal structure and magnetic properties, Fig. 2(a) discussion, p. 6]
"Based on the value of μ(H = 7 T, T= 2 K), it is possible to estimate the ratio of Eu3+ to Eu2+ for each of the Eu10Hg55 samples, as summarized in Fig. 2(a). The average Eu valence, listed for each of the samples, ranges from 2.02 (pink) to 2.18 (orange)."
The 'average Eu valence' is defined by the saturation-moment deficit under the assumption that only Eu3+ is nonmagnetic: with Eu2+ fixed at 7 μB and Eu3+ at 0, any deficit maps one-to-one onto a valence between 2 and 3. Thus the valence number is not independent evidence of mixed valence; it is the measured M(7 T) value relabeled. The conclusion 'europium in Eu10Hg55 might be present in two valence states' (abstract) is therefore a restatement of the input deficit, not a separate finding. The HERFD XANES, which could have provided an independent check, was invalidated by oxidation ('the peak corresponding to Eu3+ is dominating... sample decomposed'), so the loop is unbroken.
-
self citation load bearing
[Footnote 3 (p. 6) and Fig. 1(a) caption]
"However, to achieve a full 7 μB moment, a stoichiometry of Eu8Hg57 is needed (for Sample 1, for example), which would mean that the homogeneity range of this phase is far beyond that reported for other R/A10−xHg55+x phases – see Fig. 1(a). For the rest of the analysis, we therefore use the Eu10Hg55 formula, since the magnetic behavior of this phase is driven by the ratio of the Eu3+ to Eu2+ species rather than by the ratio of Eu to Hg."
This step rejects the off-stoichiometry alternative to Eu3+ dilution by appealing to the Eu10+xHg55-x homogeneity range shown in Fig. 1(a). That range is labeled in the figure as 'current work' and 'Nixon et al., in prep. (2025)', i.e., the authors' own unpublished data—not an external literature value. Moreover, the final sentence simply asserts that the behavior is driven by the Eu3+/Eu2+ ratio, which is exactly the claim to be established; the possibility that the moment deficit is due to Eu vacancies or Hg/Eu mixing is dismissed by definition rather than by a measured Eu:Hg ratio. The paper elsewhere admits no definitive composition analysis was possible.
full rationale
The paper is primarily an experimental characterization of a new compound. The two magnetic transitions (T1 = 5.5 K, T2 = 4.3 K), the magnetization pole reversal at low field, and the specific-heat anomaly are direct observations; they are not circular. The circularity concerns only the mixed-valence interpretation. The average valence is computed from M(7 T) by assuming that any shortfall from 7 μB/Eu is due to nonmagnetic Eu3+; it is therefore a one-to-one rename of the moment deficit. The conclusion that both Eu2+ and Eu3+ are present then re-imports this renamed quantity as evidence. The only independent valence probe, HERFD XANES, was invalidated by sample oxidation, and the authors explicitly state that no definitive Eu:Hg ratio could be obtained. The alternative explanation—off-stoichiometry or Eu/Hg site mixing reducing the average Eu moment—is rejected using a homogeneity-range plot whose Eu portion is labeled 'current work' / 'in prep.' by the same group, and then dismissed by the assertion that the behavior is driven by the Eu3+/Eu2+ ratio. That assertion is the conclusion under test, making the argument circular at that point. The magnetic order/pole-reversal narrative does not depend on the valence decomposition and retains independent content; hence the score is 6 rather than 8.
Assumptions & free parameters
free parameters (5)
- Average Eu valence per sample =
2.02 to 2.18 (sample dependent)
- Effective magnetic moment meff =
7.24 to 7.85 Bohr magnetons per Eu
- Weiss temperature theta_W =
6.4 to 11.7 K
- Korringa slope b =
3 mT/K
- XANES Eu2+/Eu3+ ratio =
0.2 Eu2+ and 0.8 Eu3+
assumptions (6)
- domain assumption Eu2+ has L=0, S=7/2, with a theoretical saturation moment of 7 Bohr magnetons and effective moment of 7.94 Bohr magnetons.
- domain assumption Eu3+ is nonmagnetic with J=0 and is ESR-silent.
- domain assumption The crystal structure of Eu10Hg55 is correctly described by the prior single-crystal refinement with four Eu sites and 17 Hg sites, including a partially occupied Eu4 position.
- domain assumption Eu3+, being smaller than Eu2+, occupies the smaller Eu1 site, giving a nominal Eu3+ to Eu2+ ratio near 3:7.
- domain assumption The low-field zfc/fc magnetization reversal is an intrinsic magnetic property of Eu10Hg55 rather than an artifact of remnant field or sample degradation.
- domain assumption The magnetic samples remained stoichiometric and unoxidized inside sealed holders during magnetization and specific heat measurements.
Cite this review
Pith. "Pith review of Unusual magnetic order in Eu$_{10}$Hg$_{55}$." pith.science (2026). https://pith.science/paper/JO3JSEVE
@misc{pith2026250907589,
author = {Pith},
title = {Pith review of: Unusual magnetic order in Eu$_10$Hg$_55$},
year = {2026},
howpublished = {\url{https://pith.science/paper/JO3JSEVE}},
note = {Machine review of arXiv:2509.07589}
}
abstract
In solid-state compounds, the valence of europium can sometimes be mixed -- which is especially favored in structures with several positions for the europium atoms. In this work, we study the Eu-based intermetallic noncentrosymmetric system Eu$_{10}$Hg$_{55}$ which has 65 atoms per unit cell and 4 distinct crystallographic positions for europium and 17 positions for mercury. Our detailed analysis of magnetism of large single crystals suggests that europium in Eu$_{10}$Hg$_{55}$ might be present in two valence states, resulting in a fragile magnetic ground state. Due to the cage-like structure with a large distance between the Eu atoms, those atoms are weakly ferromagnetically coupled and Eu$_{10}$Hg$_{55}$ orders at low temperatures, below $T_{1} = 5.5$ K, with a subsequent spin re-orientation at $T_{2} = 4.3$ K. There is no sign of magnetic frustration. Interestingly, the magnetic ordering of europium sub-lattices results in a magnetization pole reversal with a weak ferrimagnetic ground state. Additional magnetic phases can be induced by application of a modest external magnetic field.
Figures
Figures from the paper (2 more)
Reference graph
Works this paper leans on
-
[1]
(1) Engel, S.; Gießelmann, E. C. J.; P¨ ottgen, R.; Janka, O. Trivalent europium – a scarce case in intermetallics. Rev. Inorg. Chem. 2023, 43, 571–582. (2) Lawrence, J. M.; Riseborough, P. S.; Parks, R. D. Valence fluctuation phenomena. Rep. Prog. Phys. 1981, 44, 1–84. (3) Weill, D. F.; Drake, M. J. Europium Anomaly in Plagioclase Feldspar: Experimental ...
work page 2023
-
[3]
ACS Org. Inorg. 17 Au 2023, 3, 143–150. (43) Nixon, R.; Prots, Y.; Krnel, M.; Zaremba, N.; Schmidt, M.; Akselrud, L.; Grin, Y.; Svanidze, E. Superconductivity of non-centrosymmetric Ca 10Hg55 and Sr 10Hg55. in preparation 2025, (44) Nixon, R.; Prots, Y.; Krnel, M.; Zaremba, N.; Pavlosiuk, O.; Schmidt, M.; Akselrud, L.; Grin, Y.; Svanidze, E. Superconducti...
work page 2023
-
[9]
(58) Tkachuk, A. V.; Mar, A. Alkaline-earth metal mercury intermetallics A 11−xHg54+x (A = Ca, Sr). Inorg. Chem. 2008, 47, 1313–1318. (59) Buschow, K.; van Steenwijk, F. Magnetic properties and M¨ ossbauer effect of some Eu–Cd compounds. Physica B+C 1976, 85, 122–126. (60) Nakamura, A.; Hiranaka, Y.; Uejo, T.; Takeuchi, T.; Honda, F.; Harima, H.; Matsub- ...
work page 2008
-
[11]
J. Phys. Soc. Jpn. 2014, 83, 074714. (61) Fisk, Z.; Stewart, G. R.; Willis, J. O.; Ott, H. R.; Hulliger, F. Low-temperature prop- erties of the heavy-fermion system uranium-cadmium UCd
work page 2014
-
[12]
Phys. Rev. B 1984, 30,
work page 1984
-
[13]
Inorg. Chem. 2022, 61, 19695– 19701. (63) Cafasso, F. A.; Feder, H. M.; Gruen, D. M. Electronic Configuration and Valency of Uranium in UCd11, A Magnetically Dilute Alloy. J. Chem. Phys. 1963, 38, 1256–1256. 19 (64) Misiuk, A.; Mulak, J.; Czopnik, A. Magnetic properties of uranium compounds of low uranium contents: UCu 5, U 2Zn17 and UCd
work page 2022
-
[14]
Bull. Acad. Pol. Sci., Ser. Sci. Chim. 1973, 21, 487–493. (65) Jeitschko, W.; Jaberg, B. Neue Verbindungen mit Zr 2Fe12P7-Struktur und Verfeiner- gung der Kristallstrukturen von Er2Co12P7 und Er2Ni12P7. Z. Anorg. Allg. Chem. 1980, 467, 95–104. (66) Reehuis, M.; Jeitschko, W. Magnetic properties of lanthanoid iron and cobalt phos- phides with Zr 2Fe12P7 ty...
work page 1973
-
[34]
(92) Glatzel, P.; Weng, T.-C.; Kvashnina, K.; Swarbrick, J.; Sikora, M.; Gallo, E.; Smolent- sev, N.; Mori, R. A. Reflections on hard X-ray photon-in/photon-out spectroscopy for electronic structure studies. Journal of Electron Spectroscopy and Related Phenomena 2013, 188, 17–25. (93) Scheinost, A. C.; Claussner, J.; Exner, J.; Feig, M.; Findeisen, S.; He...
work page 2013
Show all 16 references
-
[85]
Magnetic properties and 151Eu M¨ ossbauer effects of mixed valence europium copper sulfide, Eu 2CuS3
(16) Furuuchi, F.; Wakeshima, M.; Hinatsu, Y. Magnetic properties and 151Eu M¨ ossbauer effects of mixed valence europium copper sulfide, Eu 2CuS3. J. Solid State Chem. 2004, 177, 3853–3858. (17) Jakubcov´ a, P.; Johrendt, D.; Sebastian, C. P.; Rayaprol, S.; P¨ ottgen, R. Stru...
2004
-
[106]
W.; Qi, X.-L.; Zhang, S.-C
(30) K¨ onig, M.; Wiedmann, S.; Br¨ une, C.; Roth, A.; Buhmann, H.; Molenkamp, L. W.; Qi, X.-L.; Zhang, S.-C. Quantum Spin Hall Insulator State in HgTe Quantum Wells. 16 Science 2007, 318, 766–770. (31) Virot, F. m. c.; Hayn, R.; Richter, M.; van den Brink, J. Engineering Topo...
2007
-
[118]
(24) Onnes, H. K. The Superconductivity of Mercury. Comm. Phys. Lab. Univ., Leiden 1911, 120b, 122–124. (25) Pelloquin, D.; Hervieu, M.; Michel, C.; Tendeloo, G. V.; Maignan, A.; Raveau, B. A 94 K Hg-based superconductor with a ”1212” structure Hg 0.5Bi0.5Sr2Ca1−xRxCu2O6+δ (R ...
1911
-
[1052]
Complex cubic metallides AM ∼6 (A = Ca, Sr; M = Zn, Cd, Hg)
(54) Schwarz, M.; Wendorff, M.; R¨ ohr, C. Complex cubic metallides AM ∼6 (A = Ca, Sr; M = Zn, Cd, Hg). Synthesis, crystal chemistry and chemical bonding. Z. Kristallogr. Cryst. Mater. 2017, 232, 515–541. (55) Tambornino, F.; Sappl, J.; Pultar, F.; Cong, T. M.; H¨ ubner, S.; G...
2017
-
[1970]
(87) Barnes, S. E. Theory of electron spin resonance of magnetic ions in metals. Adv. Phys. 1981, 30, 801–938. (88) Yannello, V.; Guillou, F.; Yaroslavtsev, A. A.; Tener, Z. P.; Wilhelm, F.; Yaresko, A. N.; Molodtsov, S. L.; Scherz, A.; Rogalev, A.; Shatruk, M. Revisiting bond...
1981
-
[2006]
Empirical way for finding new uranium-based heavy-fermion ma- terials
(39) Svanidze, E.; Amon, A.; Borth, R.; Prots, Y.; Schmidt, M.; Nicklas, M.; Leithe- Jasper, A.; Grin, Y. Empirical way for finding new uranium-based heavy-fermion ma- terials. Phys. Rev. B 2019, 99, 220403. (40) Prots, Y.; Krnel, M.; Grin, Y.; Svanidze, E. Superconductivity i...
2019
-
[2007]
A review of the synthesis and properties of Hg-containing superconductors, focused on bulk materials and thin films
(28) Bryntse, I. A review of the synthesis and properties of Hg-containing superconductors, focused on bulk materials and thin films. Studies of High Temperature Superconductors 1997, 1–26. (29) Tresca, C.; Profeta, G.; Marini, G.; Bachelet, G. B.; Sanna, A.; Calandra, M.; Boe...
1997
-
[2023]
WinCSD: Software package for crystallographic calculations (Version 4)
(97) Akselrud, L.; Grin, Y. WinCSD: Software package for crystallographic calculations (Version 4). J. Appl. Crystallogr. 2014, 47, 803–805. (98) Joshi, J.; Bhat, S. On the analysis of broad Dysonian electron paramagnetic resonance spectra. J. Magn. Mag. Res. 2004, 168, 284–28...
2014
Reviewed August 4, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.