REVIEW 4 major objections 6 minor 30 references
Tuning of electronic properties in highly lattice-mismatched epitaxial SmN
T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read Growth temperature tunes epitaxial SmN from a ferromagnetic insulator to a ferromagnetic metal, with no superconductivity down to 0.35 K.
desk verdict A solid experimental report showing substrate temperature tunes SmN from insulating to metallic ferromagnet on MgO, with an honest superconductivity null—just needs to address the CrN cap as a possible parallel conduction path. 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 knob is the density of nitrogen vacancies (N$_V$) in the rocksalt SmN lattice, controlled during growth by substrate temperature and nitrogen flux. Each vacancy releases electrons from the three Sm atoms it coordinates; Sm 4f states drop to the Fermi level and hybridize with Sm 5d and N 2p states, converting the material from a small-gap semiconductor (gap 4–8 meV) to a degenerate metal. The transport signatures — positive temperature coefficient of resistivity, $T^2$ resistivity below ordering, negative magnetoresistance — all track this carrier density.
What would settle it
Measure the resistivity and Hall effect of a SmN film grown without the CrN cap (or of a CrN-only film on MgO under identical growth conditions). If the CrN-only film shows comparable conduction or if the uncapped SmN film does not reproduce the metallic crossover and carrier densities, the claim of an intrinsic tuning of SmN's electronic phase fails.
Extended reading notes
Core claim
The paper shows that substrate temperature during molecular beam epitaxy is the dominant synthesis knob for the electronic phase of SmN on MgO(001). By raising substrate temperature from 576°C to 813°C at low nitrogen flux, with all other parameters fixed, the carrier density rises from 4.14×$10^{19}$ $cm^{-3}$ to 5.51×$10^{20}$ $cm^{-3}$ and the resistivity changes from semiconducting (negative temperature coefficient) to metallic (positive coefficient above 50 K), while the ferromagnetic transition near 27.5 K persists in all films. The authors interpret this as nitrogen-vacancy doping: higher temperatures and lower N flux increase nitrogen vacancies, releasing electrons into Sm 5d and N 2p states that hybridize with Sm 4f states. No superconductivity is observed down to 0.35 K in any film, including the most conductive, which the authors contrast with earlier reports of superconductivity in polycrystalline SmN and attribute to lower carrier density, disorder sensitivity of triplet pairing, strain, or quantum confinement in thin films.
Load-bearing premise
The measurements that show the insulator-to-metal crossover assume the 3 nm CrN capping layer is electrically passive; if CrN conducts in parallel, the reported resistivities and Hall carrier densities would not be intrinsic to SmN.
Editorial extensions
If this is right
- Epitaxial SmN on MgO(001) can host both insulating and metallic ferromagnetic states depending on growth temperature, enabling monolithic integration of distinct electronic phases without strain from dissimilar lattice-matched layers.
- Substrate temperature is a much stronger control than nitrogen flux for carrier density, so device-relevant doping can be set by one synthesis parameter while preserving crystallinity (FWHM around 0.39°).
- The absence of superconductivity down to 0.35 K in high-quality films places an experimental constraint: any pairing mechanism must be suppressed by the combination of lower carrier density, thin-film disorder, strain, or confinement, or must be a property of polycrystalline inhomogeneity rather than intrinsic SmN.
- Carrier concentration in these films is roughly an order of magnitude below the value reported for superconducting samples (~2×10^21 cm^-3), suggesting carrier density is the controlling parameter for superconductivity.
- The ferromagnetic transition remains robust (~27.5 K) across the insulator-metal crossover, so the magnetic order does not require metallicity.
Reading between the lines
- If the CrN cap is electrically passive, then the reported Hall carrier densities imply the metal-insulator crossover is intrinsic; a natural test is gating or annealing a single film to sweep carrier density continuously and check whether superconductivity emerges at higher density than the 5.5×10^20 cm^-3 reached here.
- The strain-free growth on a highly mismatched substrate via an interfacial Sm-O-N or Sm-O layer suggests a route to integrate rare-earth nitride phases with transition-metal nitride platforms; one could probe whether the interfacial layer also contributes a parallel conductive channel, which would affect the interpretation of low-temperature resistivity.
- The $T^2$ resistivity and negative magnetoresistance below $T_C$ are consistent with electron-magnon scattering in a ferromagnetic metal; measuring the anomalous Hall effect would test whether the orbital-dominated magnetic character reported elsewhere in SmN persists across this insulator-metal crossover.
- The absence of superconductivity even at 0.35 K in the most conductive film suggests that if triplet pairing is intrinsic to SmN, the clean thin-film limit actually suppresses it — the opposite of the usual expectation that disorder suppresses unconventional pairing more than conventional pairing — a point worth checking in other rare-earth nitride systems.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports molecular-beam epitaxy growth of 30 nm SmN films on MgO(001) with a 3 nm CrN cap, varying substrate temperature (576–813 °C) and nitrogen mass flow (0.1–2.0 sccm). The authors claim that these growth parameters tune the electronic properties from an insulating ferromagnetic state to a metallic ferromagnetic state, with room-temperature Hall carrier densities increasing from 4.14×10^19 to 5.51×10^20 cm^-3 as substrate temperature increases. Structural characterization shows highly crystalline, epitaxial films with a possible strained interfacial Sm-O-N layer. Resistivity measurements show a ferromagnetic transition near 27.5 K, negative magnetoresistance below TC, and, in the most conductive samples, no superconductivity down to 0.35 K. The paper discusses possible reasons for the absence of superconductivity, including carrier concentration, disorder, strain, and quantum confinement.
Significance. If the central claims hold, the paper establishes substrate temperature as a primary synthesis knob for controlling the electronic phase of epitaxial SmN and provides a useful experimental bound on superconductivity in high-quality thin films. The systematic series of 14 samples, the structural characterization (XRD, RSM, pole figures), and the transport and Hall measurements are valuable empirical contributions. The authors also explicitly connect their observations to the proposed triplet-pairing scenario, which gives the null result interpretive weight. The main weakness is that the transport conclusions rely on the uncharacterized electrical behavior of the CrN cap and a possible interfacial layer, and the superconductivity null is not yet fully documented in terms of which samples reached 0.35 K.
major comments (4)
- [Methods; Fig. 2(a)–(e)] The 3 nm CrN cap grown at 650 °C is implicitly assumed to be electrically passive, but the paper provides no resistivity or Hall data for CrN grown under the same conditions, no parallel-conduction correction, and no discussion of possible conduction through the CrN/SmN interface. Since the central insulator-to-metal tuning claim and the reported carrier densities (4.14×10^19–5.51×10^20 cm^-3) are extracted from transport and Hall measurements on capped films, a conductive cap or the strained interfacial layer seen in Fig. 1(d) could materially distort these results. Please provide a CrN-only control, a two-layer transport model, or at least an estimate of the cap's sheet conductance; the GaN-capped sample in Fig. 2(f) is a promising control but is not quantitatively compared.
- [Methods; Results and Discussion; Fig. 2(f)–(h)] The statement that no superconductivity is observed down to 0.35 K is broader than the measurements justify. The Methods say the PPMS He3 option was used only for selected samples, and the text does not state how many films were measured to 0.35 K or which growth conditions they correspond to. Fig. 2(f), cited in the superconductivity discussion, is the GaN-capped sample, not necessarily the most conductive CrN-capped sample. Please restrict the null claim to the specific samples measured and specify their growth parameters.
- [Fig. 2(e); Results and Discussion] Room-temperature Hall carrier densities are reported as a function of substrate temperature and nitrogen flow for 14 samples, but no error bars, measurement reproducibility, or Hall analysis assumptions are given. In a ferromagnetic, potentially multi-band semiconductor, the single-band Hall formula may not give the true carrier density. The comparison with the superconducting polycrystalline samples (n_i = 2×10^21 cm^-3) therefore needs an explicit statement of the Hall factor and its uncertainty.
- [Results and Discussion] The semiconducting samples are said to have small energy gaps of 4–8 meV in the paramagnetic state, but no equation, fitting procedure, or definition of the gap is provided. Because this estimate contributes to classifying the low-temperature films as insulating, the extraction method (e.g., activation energy from ln ρ versus 1/T) and the fitting range should be stated.
minor comments (6)
- [Fig. 1 caption; main text] The main text refers to phi scans as 'seen in Fig. 1(g)', but the Fig. 1 caption has no panel (g); the phi-scan panel appears to be (b). Please correct the cross-references and label all panels consistently.
- [Methods] The word 'Deaulaney' should be 'Delaunay' in the description of the contouring triangulation algorithm.
- [Results and Discussion] The phrase 'in agreements with previous estimates' should be 'in agreement with previous estimates', and 'we did not observed' should be 'we did not observe'.
- [Fig. 2(f)] The text says 'As can be seen from Figs. 2(f), no sign of a superconducting state is observed down to 0.35 K', but the panel shown is a GaN-capped sample. Please clarify in the caption and text which sample is being displayed and whether it is representative of the most conductive films.
- [Abstract; main text] The notation SmN_x is introduced but then the text reverts to SmN without stating the measured nitrogen stoichiometry or its uncertainty. Please state whether x was quantified or is used only as a formal notation.
- [Results and Discussion] The carrier mobility values (25.4 and 52.68 cm^2/V-s) are reported without units typeset consistently; use cm^2 V^-1 s^-1 and state the measurement temperature (presumably room temperature).
Circularity Check
No circularity: the paper's central claims are direct transport and Hall measurements, with no fitted parameter or self-citation chain serving as the derivation.
full rationale
I find no circular derivation in this manuscript. The central claims—carrier density increasing from 4.14×10^19 to 5.51×10^20 cm^-3 with substrate temperature, resistivity evolving from negative to positive temperature coefficient, and the absence of superconductivity down to 0.35 K—are empirical transport and magnetotransport observations, not outputs of a fitted model. No parameter is fitted to a subset of data and then used to predict a closely related quantity; the carrier density map is an interpolation of measured Hall data, and the insulator-to-metal classification is read directly from the resistivity curves. The interpretation that nitrogen-vacancy concentration increases with substrate temperature is a qualitative mechanism, not a fitted input used to generate the measured values. The only self-citation, Ref. [17], supplies the prior synthesis procedure and is methodological, not load-bearing for the electrical-tuning or superconductivity-null conclusions. The reviewer concern about the CrN capping layer's possible parallel conduction is a measurement-interpretation risk rather than a circularity, because the paper does not derive SmN transport from CrN properties. Accordingly, the derivation chain is self-contained and no circular step can be exhibited.
Assumptions & free parameters
free parameters (3)
- rho_0 (residual resistivity) =
46.7 micro-ohm-cm
- A (T^2 coefficient) =
3.2e-8 ohm-cm/K^2
- Energy gap (paramagnetic state) =
4 - 8 meV
assumptions (6)
- domain assumption The 3 nm CrN capping layer does not contribute significantly to the measured Hall and resistivity signals.
- domain assumption Hall measurements in van der Pauw geometry yield the bulk carrier density of the SmN film.
- domain assumption The anomaly in d rho/dT at about 27.5 K marks the Curie temperature of SmN.
- domain assumption Higher substrate temperature increases nitrogen vacancies, which release electrons into the conduction band.
- domain assumption The interfacial layer (Sm-O-N or strained Sm-O) plus misfit dislocations relax the lattice mismatch so the SmN film is unstrained.
- domain assumption Absence of zero resistivity down to 0.35 K implies no superconducting transition in the measured samples.
Cite this review
Pith. "Pith review of Tuning of electronic properties in highly lattice-mismatched epitaxial SmN." pith.science (2026). https://pith.science/paper/YFHK2WKN
@misc{pith2026250706372,
author = {Pith},
title = {Pith review of: Tuning of electronic properties in highly lattice-mismatched epitaxial SmN},
year = {2026},
howpublished = {\url{https://pith.science/paper/YFHK2WKN}},
note = {Machine review of arXiv:2507.06372}
}
read the original abstract
We demonstrate that the electronic properties of epitaxial SmN thin films can be effectively tuned during growth by controlling the synthesis parameters. By carefully adjusting these parameters, we are able to drive SmN from an insulating ferromagnetic state to a ferromagnetic metallic state. However, no signatures of previously reported superconductivity were observed down to 0.35 K, even in the most conductive samples. We discuss possible scenarios for the absence of superconductivity in these films and examine implications for the underlying pairing mechanism in this material. These findings open a new pathway for the epitaxial engineering of multifunctional materials, enabling the monolithic integration of diverse electronic phases, such as ferromagnetism and metallicity, without the lattice mismatch and strain typically associated with heteroepitaxial growth of dissimilar materials.
Figures
Reference graph
Works this paper leans on
- [1]
-
[2]
E. M. Anton, S. Granville, A. Engel, S. V. Chong, M. Governale, U. Z¨ ulicke, A. G. Moghaddam, H. J. Trodahl, F. Natali, S. V´ ezian, and B. J. Ruck, Super- conductivity in the ferromagnetic semiconductor samar- ium nitride, Physical Review B 94, 10.1103/Phys- RevB.94.024106 (2016)
doi:10.1103/phys- 2016
-
[3]
R. Moon and W. Koehler, Magnetic properties of SmN, Journal of Magnetism and Magnetic Materials 14, 265 (1979)
work page 1979
-
[4]
A. Preston, S. Granville, D. Housden, B. Ludbrook, B. Ruck, H. Trodahl, A. Bittar, G. Williams, J. Downes, A. DeMasi, et al., Comparison between experiment and calculated band structures for DyN and SmN, Physical Review B—Condensed Matter and Materials Physics 76, 245120 (2007)
work page 2007
-
[5]
W. F. Holmes-Hewett, F. H. Ullstad, B. J. Ruck, F. Na- tali, and H. J. Trodahl, Anomalous hall effect in Smn: Influence of orbital magnetism and 4f -band conduction, Physical Review B 98, 10.1103/PhysRevB.98.235201 (2018)
-
[6]
W. F. Holmes-Hewett, R. G. Buckley, B. J. Ruck, F. Na- tali, and H. J. Trodahl, Optical spectroscopy of SmN: Locating the 4f conduction band, Physical Review B 99, 10.1103/PhysRevB.99.205131 (2019)
-
[7]
Azeem, On the optical energy gap of SmN, Chinese Journal of Physics 56, 1925 (2018)
M. Azeem, On the optical energy gap of SmN, Chinese Journal of Physics 56, 1925 (2018)
work page 2018
-
[8]
M. Azeem, Quantitative measure of nitrogen vacancy related effects in SmN and EuN, Advances in Nat- ural Sciences: Nanoscience and Nanotechnology 10, 10.1088/2043-6254/ab007d (2019)
Show all 30 references
-
[9]
Natali, B
F. Natali, B. Ludbrook, J. Galipaud, N. Plank, S. Granville, A. Preston, B. L. Do, J. Richter, I. Farrell, R. Reeves, S. Durbin, J. Trodahl, and B. Ruck, Epitaxial growth and properties of GdN, EuN and SmN thin films, Physica Status Solidi (C) Current Topics in Solid State Phy...
2012
-
[10]
E. M. Anton, B. J. Ruck, C. Meyer, F. Natali, H. War- ring, F. Wilhelm, A. Rogalev, V. N. Antonov, and H. J. Trodahl, Spin/orbit moment imbalance in the near-zero moment ferromagnetic semiconductor SmN, Physical Re- view B - Condensed Matter and Materials Physics 87, 10.1103/P...
2013 doi
-
[11]
J. F. McNulty, E. M. Anton, B. J. Ruck, F. Natali, H. Warring, F. Wilhelm, A. Rogalev, M. M. Soares, N. B. Brookes, and H. J. Trodahl, Twisted phase of the orbital- dominant ferromagnet smn in a GdN/SmN heterostruc- ture, Physical Review B - Condensed Matter and Mate- rials Ph...
2015 doi
-
[12]
J. F. McNulty, K. Temst, M. J. V. Bael, A. Vantomme, and E. M. Anton, Epitaxial growth of (100)-oriented SmN directly on (100)Si substrates, Physical Review Ma- terials 5, 10.1103/PhysRevMaterials.5.113404 (2021)
2021 doi
-
[13]
J. R. Chan, S. V´ ezian, J. Trodahl, M. A. Khalfioui, B. Damilano, and F. Natali, Temperature-induced four- fold-on-six-fold symmetric heteroepitaxy, rocksalt SmN on hexagonal AlN, Crystal Growth and Design 16, 6454 (2016)
2016
-
[14]
V´ ezian, B
S. V´ ezian, B. Damilano, F. Natali, M. A. Khalfioui, and J. Massies, AlN interlayer to improve the epitaxial growth of SmN on GaN (0001), Journal of Crystal Growth 450, 22 (2016)
2016
-
[15]
Anton, E
E.-M. Anton, E. Trewick, W. F. Holmes-Hewett, J. R. Chan, J. F. McNulty, T. Butler, B. J. Ruck, and F. Na- tali, Growth of epitaxial (100)-oriented rare-earth ni- trides on (100)LaAlO3, Applied Physics Letters 123, 10.1063/5.0186522 (2023)
2023 doi
-
[16]
Mel´ endez-Sans, V
A. Mel´ endez-Sans, V. Pereira, C. Chang, C.-Y. Kuo, C. Chen, L. Tjeng, and S. Altendorf, Influence of nitro- gen stoichiometry and the role of Sm 5 d states in SmN thin films, Physical Review B 110, 045120 (2024)
2024
-
[17]
K. D. Vallejo, Z. E. Cresswell, V. Buturlim, B. S. Newell, K. Gofryk, and B. J. May, Synthesis of samarium ni- tride thin films on magnesium oxide (001) substrates us- ing molecular beam epitaxy, Crystals 14, 765 (2024)
2024
-
[18]
Xie, S.-Y
Y. Xie, S.-Y. Zhang, Y. Yin, N. Zheng, A. Ali, M. Younis, S. Ruan, and Y.-J. Zeng, Emerging ferromagnetic mate- rials for electrical spin injection: towards semiconductor spintronics, npj Spintronics 3, 10 (2025)
2025
-
[19]
Akinaga and H
H. Akinaga and H. Ohno, Semiconductor spintronics, IEEE Transactions on nanotechnology 1, 19 (2002)
2002
-
[20]
Dietl, H
T. Dietl, H. Ohno, and F. Matsukura, Ferromagnetic semiconductor heterostructures for spintronics, IEEE transactions on electron devices 54, 945 (2007)
2007
-
[21]
Pot, Magnetic Devices Using Rare-Earth Nitrides, Ph.D
C. Pot, Magnetic Devices Using Rare-Earth Nitrides, Ph.D. thesis, Open Access Te Herenga Waka-Victoria University of Wellington (2024)
2024
-
[22]
Shao and E
D.-F. Shao and E. Y. Tsymbal, Antiferromagnetic tunnel junctions for spintronics, npj Spintronics 2, 13 (2024)
2024
-
[23]
T. L. Felmlee and L. Eyring, Ternary system samarium- nitrogen-oxygen and the question of lower oxides of samarium, Inorganic Chemistry 7, 660 (1968)
1968
-
[24]
M. D. Nordstrom, T. A. Garrett, P. Reddy, J. McElear- ney, J. R. Rushing, K. D. Vallejo, K. Mukherjee, K. A. Grossklaus, T. E. Vandervelde, and P. J. Simmonds, Di- rect integration of GaSb with GaAs (111) A using inter- facial misfit arrays, Crystal Growth & Design 23, 8670 (2023)
2023
-
[25]
Zhang, M
Q. Zhang, M. Hilse, W. Auker, J. Gray, and S. Law, Growth conditions and interfacial misfit array in snte (111) films grown on inp (111) a substrates by molec- ular beam epitaxy, ACS Applied Materials & Interfaces 16, 48598 (2024)
2024
-
[26]
Meyer, B
C. Meyer, B. J. Ruck, J. Zhong, S. Granville, A. R. Preston, G. V. Williams, and H. J. Trodahl, Near- zero-moment ferromagnetism in the semiconductor smn, Physical Review B - Condensed Matter and Materials Physics 78, 10.1103/PhysRevB.78.174406 (2008)
2008 doi
-
[27]
Holmes-Hewett, K
W. Holmes-Hewett, K. Van Koughnet, J. Miller, E. Trewick, B. Ruck, H. Trodahl, and R. Buckley, Indi- cations of a ferromagnetic quantum critical point in smn 6 1-δ, Scientific Reports 13, 19775 (2023)
2023
-
[28]
S. Y. Karpov, R. Talalaev, Y. N. Makarov, N. Grand- jean, J. Massies, and B. Damilano, Surface kinetics of gan evaporation and growth by molecular-beam epitaxy, Surface science 450, 191 (2000)
2000
-
[29]
R. J. Mart ´ ın-Palma and J. Mart ´ ınez-Duart,Nanotechnol- ogy for microelectronics and photonics(Elsevier, 2017)
2017
-
[30]
S. B. Gupta, K. Suresh, and A. Nigam, Magnetocaloric and magnetotransport properties in RRhSn (R= Tb-Tm) series, Journal of Applied Physics 112, 103909 (2012)
2012
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.