REVIEW 4 major objections 4 minor 51 references
Electronic structure reorganization in MPS3 via d-shell-selective alkali metal doping
T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read Alkali doping reorganizes CoPS3 into a metal while MnPS3 stays electronically intact.
desk verdict Systematic four-compound XPS/ARPES study with a plausible two-mechanism picture; the metal-reduction branch rests on XPS assignments that need charge-transfer multiplet analysis and error bars before they carry the weight. 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 explanatory object is the d-shell electronic configuration of the M2+ ion: half-filled 3d5 for Mn2+, and 3d6, 3d7, and 3d8 for Fe2+, Co2+, and Ni2+. The argument is carried by three observations: XPS core-level changes (a new low-binding-energy doublet and, for Co, a spin-orbit splitting that drops from about 16 eV to 15 eV) read as a formal oxidation-state reduction; ARPES band shifts and the appearance of dispersive states above the valence band maximum; and DFT+U charge-density-difference and Bader-charge calculations showing Li donates about 0.89 electrons, mostly to Co and S. Together these make d-shell filling the selector between ligand-centered and metal-centered doping.
What would settle it
Measure Co 2p XPS on the same Li/CoPS3 system with higher energy resolution and calibrated cobalt references (Co metal, CoO, CoPS3), reporting peak positions and widths with uncertainties; if the 15 eV splitting and the new doublet track surface alkali clusters rather than a bulk cobalt signal, the oxidation-state-reduction claim would collapse. A complementary check is Co L2,3 X-ray absorption before and after doping, which counts d-occupancy directly and would settle whether cobalt actually approaches a reduced, near-zero-valent configuration.
Extended reading notes
Core claim
On the paper's own terms, the discovery is that electron doping in MPS3 is governed by the electronic stability of the metal d-shell rather than by the host lattice alone. MnPS3 resists accepting electrons at the metal because the half-filled 3d5 configuration of Mn2+ is especially stable; instead the extra charge localizes on the P2S6 clusters, seen in a second S 2p doublet and almost no change in Mn 2p or the valence band. FePS3, CoPS3, and NiPS3 accept the electrons into the metal d orbitals, signaled by new lower-binding-energy M 2p components and reduced spin-orbit splittings. In CoPS3 the reorganization goes furthest: ARPES shows Co-derived bands shifting roughly 400 meV to higher binding energy and dispersive states emerging up to 1 eV above the pristine valence band maximum, which combined with DFT+U indicates a semiconductor-to-metal transition. The authors take these observations to establish a direct correlation between d-shell filling and doping response across the MPS3 family.
Load-bearing premise
The load-bearing assumption is that the new low-binding-energy XPS peak and the roughly 1 eV smaller Co 2p peak separation really mean the cobalt atoms in the doped crystal changed oxidation state, rather than being artifacts of the lithium-covered surface or of how the photoemission final state screens the core hole.
Editorial extensions
If this is right
- CoPS3 becomes metallic upon alkali-metal doping, so the same material can be switched between semiconducting and conducting behavior by choosing the alkali species and coverage.
- The doping response scales with the amount of charge transferred: cesium produces the same qualitative band changes as lithium but smaller shifts, making the electronic modification continuously tunable.
- MnPS3 can store donated electrons on its P2S6 ligands without changing its metal oxidation state, so it should remain electronically robust under doping.
- In FePS3 and NiPS3 charge transfer occurs but does not create dispersive metallic states, so only CoPS3 among the four compounds undergoes a strong low-energy band reorganization.
- Alkali-metal doping therefore offers a route to tailor electronic and magnetic functionality in 2D antiferromagnets by targeting compounds with less stable d-shell configurations.
Reading between the lines
- If the spin-orbit-splitting proxy is right, the cobalt in doped CoPS3 should be close to zero-valent; this could be checked directly with X-ray absorption at the Co L-edge, which counts d-occupancy without relying on peak fitting.
- The same d-shell-stability logic may predict doping behavior in other MPX3 or metal-halide 2D magnets: compounds whose d-count is neither half-filled nor closed should be the ones that accept electrons and reorganize.
- Because the measurements are surface-sensitive, a bulk transport or magnetometry measurement on intercalated crystals would test whether the metallic and magnetic changes survive beyond the deposition layer and actually affect spintronic device-relevant properties.
- The emerging metallic bands in CoPS3 could couple to the antiferromagnetic order through carrier-mediated exchange, potentially turning doping into a magnetic-phase switch; the paper stops short of measuring magnetism, so this is an open consequence.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports a combined ARPES, XPS, and DFT+U study of alkali-metal (Li and Cs) deposition on four MPS₃ compounds (M = Mn, Fe, Co, Ni). The authors identify two classes of doping response: MnPS₃, with its half-filled 3d⁵ shell, localizes added electrons on the P₂S₆ ligands with negligible Mn 2p core-level changes; FePS₃, CoPS₃, and NiPS₃ accept electrons into transition-metal d states, as inferred from new low-binding-energy M 2p doublets and reduced spin-orbit splittings. For CoPS₃, ARPES shows a ~400 meV shift of near-VBM Co-derived bands to higher binding energy and new dispersive states up to 1 eV above the pristine VBM, which, with supporting DFT+U, is interpreted as a semiconductor-to-metal transition. Cs doping produces similar but weaker effects, and formation-energy calculations suggest thermodynamic stability of intercalated Li. The central claim is a direct correlation between d-shell filling and doping response, establishing alkali-metal doping as a tunable route for 2D AFM semiconductors.
Significance. If the central interpretation is correct, the paper offers a useful design rule: the electron-accepting site in MPS₃ (metal d states versus P₂S₆ ligands) is determined by d-shell filling, and CoPS₃ in particular becomes a tunable metallic system. The study is systematic across four compounds, uses two dopants, combines three spectroscopic/computational approaches, and provides raw data in a repository. The Cs-doping experiment is a valuable control showing that the magnitude of the band-structure changes scales with charge transfer. However, the headline mechanism relies on core-level spectral interpretations that are not uniquely determined without multiplet calculations or complementary probes; the ARPES claim of metallicity rests on indirect evidence; and the assumption of bulk intercalation is not experimentally verified. These gaps currently limit the strength of the conclusions.
major comments (4)
- [XPS study (Figure 2, Table S3)] The assignment of the new low-binding-energy Co 2p doublet and the 16→15 eV spin-orbit splitting to a ground-state Co²⁺→Co⁰ transition is not unique. In correlated 3d compounds, a low-binding-energy component can arise from final-state screening rather than a distinct ground-state species, and metallic Co itself has a 2p spin-orbit splitting near 15 eV. The fits are presented without error bars, and no charge-transfer multiplet calculation is provided to separate ground-state d occupancy from final-state effects. The cited Frost et al. (1972) study documents empirical trends in Co complexes but does not establish the ground-state assignment in thiophosphates. This is load-bearing because the metal-reduction branch of the proposed two-mechanism picture depends on this interpretation. Please provide a quantitative multiplet/charge-transfer simulation or an independent d-occupancy-sensitive probe (e.g., Co L-edge XAS or XMCD), and report fit uncertainties.
- [ARPES study (Figure 5c, 5d)] The claim of a semiconductor-to-metal transition in CoPS₃ is inferred from spectral weight above the pristine VBM and from DFT+U band crossings, but the experimental data do not directly establish metallicity. The binding-energy scale is referenced to the pristine VBM rather than to the sample Fermi level, and no Fermi-edge step or transport measurement is presented. The momentum maps at 0.3 eV below and 0.5 eV above the pristine VBM show dispersive features, but they do not locate the Fermi level. Please either report the absolute energy of the new states relative to a calibrated Fermi level, provide transport or a clear spectral Fermi step, or temper the 'metallic' conclusion accordingly.
- [Methods (Sample preparation, Formation energy) and Results] The paper assumes that alkali atoms are intercalated into the van der Waals gap, but no experiment directly demonstrates bulk intercalation. The XPS and ARPES measurements are surface-sensitive, so the observed core-level and valence-band changes could arise from surface adsorption, alkali-induced surface decomposition, or cluster formation rather than bulk doping. The DFT models place Li in the interlayer space, but this does not confirm the experimental geometry. Please provide evidence for intercalation (e.g., Li 1s core-level binding energy analysis, XPS depth profiling, or a structural probe such as STM or LEED), or explicitly limit the claims to surface doping.
- [Methods (Density functional theory)] The Hubbard U values for Fe, Ni, Co, and Mn are derived by fitting pristine ARPES band structures (Figure S13). Using these same U values for the doped compounds to compute PDOS, Bader charges, and band structures assumes that U is unchanged by electron doping, and this transferability is not tested. This is not circular because the fitted values are not used to match the doping results, but the quantitative support for the doped mechanisms would be stronger if the sensitivity of the conclusions to U were demonstrated (e.g., a U scan for doped CoPS₃) or if an independent justification were provided.
minor comments (4)
- [Throughout] The notation is inconsistent: the text alternates between 'MPS3' and 'MPS₃' and between 'CoPS3' and 'CoPS₃'; please standardize the chemical formulas throughout.
- [Methods (Density functional theory)] The calculations assume a nonmagnetic configuration even though the compounds are antiferromagnets; a brief justification (e.g., the paramagnetic phase at 300 K and the good ARPES benchmark) and a note on the potential impact of spin ordering on the doped PDOS would be helpful.
- [XPS study and ARPES study] The reported shifts (400 meV in ARPES, about 1.0 eV in Co 2p spin-orbit splitting) appear without error bars; please report standard uncertainties for the fitted peak positions and splittings, ideally from repeated fits or Monte Carlo resampling.
- [Figure 5 caption] The caption says 'The left sides show the signal stemming from the clean surfaces, and the right side shows the signal from the lithium doped surfaces' but the figure panels appear to show only one side of each cut; please clarify the layout of the panels.
Circularity Check
No significant circularity: the doped DFT and XPS/ARPES interpretations are not constructed from the target doping results, and self-citations are not load-bearing.
full rationale
I walked the derivation chain: XPS core-level shifts are used to infer oxidation-state changes, ARPES band shifts are used to infer doping-induced electronic reorganization, and DFT+U is used to support the interpretation. No step reduces to its own inputs by construction. The Hubbard U parameters were fit to pristine band structures by comparing calculations to the undoped ARPES data (Methods, Figure S13), not to the doped XPS or ARPES data they are later used to interpret, so the doped DFT results are out-of-sample predictions rather than fitted inputs renamed as findings. The interpretation of the Co 2p spin-orbit splitting reduction from 16 eV to 15 eV as a ground-state oxidation-state change is an experimental assumption supported by an external reference (Frost et al.), and while it may be challenged on final-state screening or surface-metallization grounds, that is a correctness or interpretation risk, not circularity. The self-citations present in the reference list ([11], [12], [43], [44]) supply U parameters, sample-preparation details, and instrument references; none of them carries the central claim by itself, and the Fe U value from [11] is peripheral and obtained by the same pristine-band fitting procedure as the other U values. The cesium doping experiment provides an independent scaling check of the doping-level dependence. No fitted quantity is renamed as a prediction, no uniqueness theorem is imported from the authors' prior work, and no ansatz is smuggled in via citation. The central mechanism is therefore not circular within the definitions used here.
Assumptions & free parameters
free parameters (4)
- Hubbard U for Fe =
1.9 eV
- Hubbard U for Ni =
4.5 eV
- Hubbard U for Co =
4.6 eV
- Hubbard U for Mn =
5.0 eV
assumptions (4)
- domain assumption DFT+U with PBE and PAW accurately describes the relevant band structure of MPS3 compounds
- domain assumption The nonmagnetic (NM) approximation is adequate for comparing with room-temperature paramagnetic ARPES data
- domain assumption A single Li in a 2x2x2 supercell donates essentially one full electron, as used in the DFT doping model
- domain assumption XPS spin-orbit splitting changes can be mapped directly to formal oxidation state changes
Cite this review
Pith. "Pith review of Electronic structure reorganization in MPS3 via d-shell-selective alkali metal doping." pith.science (2026). https://pith.science/paper/WX6XZRSF
@misc{pith2026250601527,
author = {Pith},
title = {Pith review of: Electronic structure reorganization in MPS3 via d-shell-selective alkali metal doping},
year = {2026},
howpublished = {\url{https://pith.science/paper/WX6XZRSF}},
note = {Machine review of arXiv:2506.01527}
}
read the original abstract
Semiconducting two-dimensional (2D) antiferromagnetic (AFM) transition-metal thiophosphates (MPS3) offer promising opportunities for spintronic applications due to their highly tunable electronic properties. While alloying and intercalation have been shown to modulate ground states, the role of d-shell filling in governing these transitions remains insufficiently understood. Here, we investigate electron doping effects in MPS3 using angle-resolved photoemission spectroscopy (ARPES), X-ray photoelectron spectroscopy (XPS), and density functional theory (DFT+U). Lithium and cesium deposition are employed to induce doping across different MPS3 compounds. We identify two distinct doping mechanisms: in MnPS3, electrons are primarily donated to the P2S6 ligand clusters, with negligible Mn 2p core-level shifts and no major changes in the valence band. In contrast, FePS3, CoPS3, and NiPS3 exhibit clear reductions in transition-metal oxidation states, with a 1.0 eV reduction in spin-orbit splitting for Co upon doping. ARPES on CoPS3 reveals a 400 meV shift of Co-derived bands towards higher binding energies and new dispersive states up to 1 eV above the valence band maximum, indicating metallic behavior. These results establish a direct correlation between d-shell filling and doping response, highlighting alkali metal doping as a tunable route to tailor the electronic and magnetic properties of 2D AFM semiconductors for spintronic applications.
Reference graph
Works this paper leans on
-
[1]
K. S. Burch, D. Mandrus, J.-G. Park, Nature 2018, 563, 47
2018
- [2]
-
[3]
L. Sternemann, D. M. Janas, J. E. Nitschke, K. Schiller, T. Willershausen, L. Becker, A. Isaeva, G. Zamborlini, S. Tappertzhofen, M. Cinchetti, 2D Mater. 2025, DOI 10.1088/2053- 1583/add7ea
doi:10.1088/2053- 2025
-
[4]
J. Yang, Y. Zhou, Q. Guo, Y. Dedkov, E. Voloshina, RSC Adv. 2020, 10, 851
work page 2020
-
[5]
P. Wang, F. Lian, R. Du, X. Cai, S. Bao, Y. Han, J. Xiao, K. Watanabe, T. Taniguchi, J. Wen, H. Yang, A. S. Mayorov, L. Wang, G. Yu, Appl. Phys. Lett. 2024, 124, 012406
work page 2024
- [6]
- [7]
-
[8]
J. M. Pereira, D. Tezze, M. Ormaza, L. E. Hueso, M. Gobbi, Adv. Phys. Res. 2023, 2, DOI 10.1002/apxr.202200084
Show all 51 references
-
[9]
Y. Liu, Z. Wang, G. Hu, X. Chen, K. Xu, Y. Guo, Y. Xie, C. Wu, Precis. Chem. 2025, 3, 51
2025
-
[10]
Mertens, D
F. Mertens, D. Mönkebüscher, U. Parlak, C. Boix-Constant, S. Mañas-Valero, M. Matzer, R. Adhikari, A. Bonanni, E. Coronado, A. M. Kalashnikova, D. Bossini, M. Cinchetti, Adv. Ma- ter. 2023, 35, e2208355
2023
-
[11]
J. E. Nitschke, D. L. Esteras, M. Gutnikov, K. Schiller, S. Mañas-Valero, E. Coronado, M. Stupar, G. Zamborlini, S. Ponzoni, J. J. Baldoví, M. Cinchetti, Mater. Today Electron. 2023, 6, 100061
2023
-
[12]
J. E. Nitschke, L. Sternemann, M. Gutnikov, K. Schiller, E. Coronado, A. Omar, G. Zambor- lini, C. Saraceno, M. Stupar, A. M. Ruiz, D. L. Esteras, J. J. Baldoví, F. Anders, M. Cinchetti, Newton 2025, 100019
2025
-
[13]
Strasdas, B
J. Strasdas, B. Pestka, M. Rybak, A. K. Budniak, N. Leuth, H. Boban, V. Feyer, I. Cojocariu, D. Baranowski, J. Avila, P. Dudin, A. Bostwick, C. Jozwiak, E. Rotenberg, C. Autieri, Y. Amouyal, L. Plucinski, E. Lifshitz, M. Birowska, M. Morgenstern, Nano Lett. 2023, 23, 10342
2023
-
[14]
Pestka, J
B. Pestka, J. Strasdas, G. Bihlmayer, A. K. Budniak, M. Liebmann, N. Leuth, H. Boban, V. Feyer, I. Cojocariu, D. Baranowski, S. Mearini, Y. Amouyal, L. Waldecker, B. Beschoten, C. Stampfer, L. Plucinski, E. Lifshitz, P. Kratzer, M. Morgenstern, ACS Nano 2024, DOI 10.1021/acsna...
2024 doi
-
[15]
Dedkov, Y
Y. Dedkov, Y. Guo, E. Voloshina, Electron. Struct. 2023, 5, 043001
2023
-
[16]
Autieri, G
C. Autieri, G. Cuono, C. Noce, M. Rybak, K. M. Kotur, C. E. Agrapidis, K. Wohlfeld, M. Birowska, J. Phys. Chem. C 2022, 126, 6791
2022
-
[17]
Harchol, S
A. Harchol, S. Zuri, E. Ritov, F. Horani, M. Rybak, T. Woźniak, A. Eyal, Y. Amouyal, M. Birowska, E. Lifshitz, 2D Mater. 2024, 11, 035010
2024
-
[18]
Chakraborty, A
N. Chakraborty, A. Harchol, A. Abu-Hariri, R. K. Yadav, M. Dawod, D. P. Gaonkar, K. Sharma, A. Eyal, Y. Amouyal, D. Naveh, E. Lifshitz, arXiv 2024, DOI 10.48550/arxiv.2410.11264
2024 doi
-
[20]
Basnet, T
R. Basnet, T. Patel, J. Wang, D. Upreti, S. K. Chhetri, G. Acharya, M. R. U. Nabi, J. Sakon, J. Hu, Adv. Electron. Mater. 2024, DOI 10.1002/aelm.202300738
2024 doi
-
[21]
Basnet, D
R. Basnet, D. Ford, K. TenBarge, J. Lochala, J. Hu, J. Phys.: Condens. Matter 2022, 34, 434002
2022
-
[22]
Basnet, A
R. Basnet, A. Wegner, K. Pandey, S. Storment, J. Hu, Phys. Rev. Mater. 2021, 5, 064413
2021
-
[23]
Upreti, R
D. Upreti, R. Basnet, M. M. Sharma, S. K. Chhetri, G. Acharya, M. R. U. Nabi, J. Sakon, B. Da, M. Mortazavi, J. Hu, J. Phys.: Condens. Matter 2025, 37, 135805
2025
-
[24]
X. Li, X. Wu, J. Yang, J. Am. Chem. Soc. 2014, 136, 11065
2014
-
[25]
Clement, L
R. Clement, L. Lomas, J. P. Audiere, Chem. Mater. 1990, 2, 641
1990
-
[26]
D. Chen, C. Wang, C. Peng, Phys. Chem. Chem. Phys. 2024, 26, 8436
2024
-
[27]
Y. Cao, Q. Tan, Y. Guo, C. G. Vieira, M. S. C. Mazzon, J. Laverock, N. Russo, H. Gao, C. Jozwiak, A. Bostwick, E. Rotenberg, J. Guo, M. Yi, M. J. S. Matos, X. Ling, K. E. Smith, arXiv 2024
2024
-
[28]
X. Wu, Z. Shen, W. Xiao, J. Yang, C. Song, J. Mater. Sci.: Mater. Electron. 2022, 33, 1871
2022
-
[29]
R. P. Gupta, S. K. Sen, Phys. Rev. B 1974, 10, 71
1974
-
[30]
R. P. Gupta, S. K. Sen, Phys. Rev. B 1975, 12, 15
1975
-
[31]
Brotons-Alcazár, R
I. Brotons-Alcazár, R. Torres-Cavanillas, M. Morant-Giner, M. Cvik, S. Mañas-Valero, A. Forment-Aliaga, E. Coronado, ChemRxiv 2021, DOI 10.26434/chemrxiv-2021-lph4p
2021 doi
-
[32]
A. P. Grosvenor, B. A. Kobe, M. C. Biesinger, N. S. McIntyre, Surf. Interface Anal. 2004, 36, 1564
2004
-
[33]
E. S. Ilton, J. E. Post, P. J. Heaney, F. T. Ling, S. N. Kerisit, Appl. Surf. Sci. 2016, 366, 475
2016
-
[34]
Lu, ACS Nano 2024, 18, 13973
Q. Lu, ACS Nano 2024, 18, 13973
2024
-
[35]
Yu. G. Borod’ko, S. I. Vetchinkin, S. L. Zimont, I. N. Ivleva, Yu. M. Shul’ga, Chem. Phys. Lett. 1976, 42, 264
1976
-
[36]
F. A. Stevie, C. L. Donley, J. Vac. Sci. Technol. A: Vac., Surf., Films 2020, 38, 063204
2020
-
[37]
D. C. Frost, C. A. McDowell, I. S. Woolsey, Chem. Phys. Lett. 1972, 17, 320
1972
-
[38]
Moser, J
S. Moser, J. Electron Spectrosc. Relat. Phenom. 2017, 214, 29
2017
-
[39]
M. C. Lemme, D. Akinwande, C. Huyghebaert, C. Stampfer, Nat. Commun. 2022, 13, 1392
2022
-
[40]
Jiang, Q
X. Jiang, Q. Liu, J. Xing, N. Liu, Y. Guo, Z. Liu, J. Zhao, Appl. Phys. Rev. 2021, 8, 031305
2021
-
[41]
Schmid, H
M. Schmid, H. Steinrück, J. M. Gottfried, Surf. Interface Anal. 2014, 46, 505
2014
-
[42]
Ponzoni, F
S. Ponzoni, F. Paßlack, M. Stupar, D. M. Janas, G. Zamborlini, M. Cinchetti, Adv. Phys. Res. 2023, 2, DOI 10.1002/apxr.202200016
2023 doi
-
[43]
D. M. Janas, A. Windischbacher, M. S. Arndt, M. Gutnikov, L. Sternemann, D. Gutnikov, T. Willershausen, J. E. Nitschke, K. Schiller, D. Baranowski, V. Feyer, I. Cojocariu, K. Dave, P. Puschnig, M. Stupar, S. Ponzoni, M. Cinchetti, G. Zamborlini, Inorg. Chim. Acta 2023, 557, 121705
2023
-
[44]
K. J. Schiller, L. Sternemann, M. Stupar, A. Omar, M. Hoffmann, J. E. Nitschke, V. Mischke, D. M. Janas, S. Ponzoni, G. Zamborlini, C. J. Saraceno, M. Cinchetti, Sci. Rep. 2025, 15, 3611
2025
-
[45]
Hohenberg, W
P. Hohenberg, W. Kohn, Phys. Rev. 1964, 136, B864
1964
-
[46]
W. Kohn, L. J. Sham, Phys. Rev. 1965, 140, A1133
1965
-
[47]
Kresse, J
G. Kresse, J. Furthmüller, Comput. Mater. Sci. 1996, 6, 15
1996
-
[48]
Kresse, D
G. Kresse, D. Joubert, Phys. Rev. B 1998, 59, 1758
1998
-
[49]
J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865
1996
-
[50]
E. H. Lieb, Phys. Rev. Lett. 1989, 62, 1201
1989
-
[51]
Q. Pei, Y. Song, X. Wang, J. Zou, W. Mi, Sci. Rep. 2017, 7, 9504
2017
-
[52]
Bhumla, M
P. Bhumla, M. Kumar, S. Bhattacharya, Nanoscale Adv. 2020, 3, 575
2020
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.