REVIEW 3 major objections 4 minor 17 references
Huge hole injection in tungsten dichalcogenide heterostructures without electric gating: a DFT study
T0 review · 3 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash
Pith's one-line read Periodic DFT shows that noble-gas fluorides placed behind a single hBN layer spontaneously remove up to 0.35 electrons per W atom from WS2 and WSe2 monolayers, producing hole densities comparable to electrostatic gating without any applied
desk verdict A serious DFT study with a concrete new prediction—0.23/0.35 h+ per W via chemical capacitor—but the headline numbers hang on a PDOS normalization that has not been validated against Bader or other charge partitioning. 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 mechanism is the 'chemical capacitor' architecture: a strong electron acceptor (XeF2 or KrF2) separated from the active TMD monolayer by a dielectric hBN spacer. The relevant electronic feature is type-III (broken-gap) band alignment, in which the acceptor's conduction band minimum lies below the donor's valence band maximum, so electrons spontaneously transfer from the TMD into the molecular layer. The hBN spacer blocks covalent contact and keeps the acceptor physisorbed, while the final amount of charge transfer is set by the self-consistent equilibration of electrochemical potentials, progressively opposed by the electrostatic potential of the separated charges and modulated by dielec
What would settle it
Measure Hall carrier density and Raman spectra of a WSe2/hBN/KrF2 stack: the chemical capacitor picture predicts a hole density near 3.7×10^14 cm^-2 with an intact Kr–F stretching signature; if the density is much higher, unstable, or accompanied by an etched interface, the observed doping would come from chemical reaction rather than the proposed non-contact mechanism.
Extended reading notes
Core claim
On the paper's own terms, the central claim is that spontaneous, contactless hole injection into tungsten dichalcogenides can be achieved at gating-level densities by placing a noble-gas fluoride electron acceptor behind a dielectric hBN spacer. Periodic DFT of large supercells gives hole concentrations up to 0.23 h+ per W atom in WS2|hBN|XeF2 and 0.35 h+ per W atom in WSe2|hBN|KrF2, comparable to the 0.44 h+/W obtained for WSe2 by electric gating but requiring no external bias and no covalent modification of the TMD. The hBN layer transfers less than 0.004 h+ per B atom, supporting its role as a nearly charge-neutral dielectric separator that blocks direct chemical contact while allowing el
Load-bearing premise
The entire contactless-doping claim rests on the assumption that XeF2 and KrF2 remain intact and only physisorb on hBN, acting purely as electron acceptors, instead of chemically reacting with or etching the TMD or hBN—a risk that the paper itself acknowledges by selecting low coverage to avoid 'unphysical covalent bonding' and that is underscored by its cited experiment where XeF2 vapor etches and p-dopes WSe2.
Editorial extensions
If this is right
- Hole densities comparable to electric gating (0.35 vs 0.44 h+/W) are achievable without external bias, offering a contactless, permanent doping route for TMD monolayers.
- The hBN separator stays nearly charge-neutral and prevents direct chemical reaction between the TMD and the acceptor, so the active layer is not covalently modified.
- The transferred charge is not simply proportional to the initial band offset: final carrier density is instead governed by self-consistent electrostatic feedback, meaning separator thickness and acceptor coverage are practical tuning knobs.
- If realized experimentally, these heterostructures could reach carrier concentrations where correlated electronic phases, such as superconductivity, may emerge in monolayer TMDs.
Reading between the lines
- This points toward a 'permanent gate' for 2D devices: carrier density set at fabrication time by selecting acceptor and spacer thickness, with no steady-state power draw and no gate electrode.
- Since coverage was shown to matter (1/3 coverage avoids covalent bonding), patterning or diluting the acceptor layer could yield spatial control of doping in a single chip.
- The same broken-gap screening logic should extend to electron doping using a strong reductant such as lithium, as the supplementary information explicitly hints, and to other acceptor molecules if noble-gas fluorides prove too reactive.
- The apparent insensitivity of final charge transfer to the initial band offset suggests a practical design rule: separator thickness and acceptor coverage may be more useful dials than acceptor strength alone when targeting a specific carrier density.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes that van der Waals heterostructures of a W-based TMD, an hBN spacer, and a noble-gas difluoride (XeF2 or KrF2) act as 'chemical capacitors' that spontaneously inject large hole densities into the TMD without external gating. Using SCAN+rVV10 DFT, the authors screen band alignments, identify WS2|hBN|XeF2 and WSe2|hBN|KrF2 as type-III systems, and then compute, via a PDOS-normalization scheme, hole concentrations of 0.23 h+/W and 0.35 h+/W respectively. They compare these values with literature doping methods and conclude that the approach offers a contactless, permanent route to high carrier densities in 2D semiconductors.
Significance. If the reported charge transfer is robust, the work is significant: it proposes an experimentally plausible architecture that achieves hole densities comparable to electrostatic gating without an external bias or direct chemical functionalization of the TMD. The DFT setup is described carefully (SCAN+rVV10, dipole correction, large P1 supercells, dense k-mesh for PDOS), and the SI provides CIF structures before/after relaxation plus bond-length data, which is commendable. However, the central quantitative claim rests on a PDOS projection that is not cross-validated with any independent charge-partitioning method, and the 'contactless' interpretation depends on the inertness of NgF2 molecules that is not established beyond the relaxed geometry.
major comments (3)
- [Sec. 5, Eqs. (1)–(3)] The headline CT values (0.23 and 0.35 h+/W) are obtained exclusively from the PDOS normalization scheme described here. Because s_i is fitted to the isolated-layer PDOS and then applied to the heterostructure PDOS, apparent occupation changes can arise from rehybridization/spectral-weight redistribution without net interlayer charge transfer. No Bader analysis, charge-density-difference integration, or total-ΔN sum is reported; the hBN 'nearly charge neutral' statement is itself a PDOS result. This is load-bearing for the main claim. Please add an independent charge-partitioning cross-check and report the total charge balance. If Bader or another method gives materially different values, the central claim would need revision.
- [Sec. 3 and Sec. 6] The 'contactless' and 'no permanent chemical modification' characterization requires that NgF2 remain intact and physisorbed on hBN. The relaxed structures show intact molecules, but the cited experimental work [29] demonstrates that XeF2 vapor etches and p-dopes WSe2. The selection of ~1/3 coverage is justified only as avoiding 'unphysical covalent bonding' in the model, not as evidence of kinetic or thermodynamic stability against reaction at or through the hBN separator, at defects, or under realistic processing. Please either provide additional calculations (e.g., dissociation/reaction pathways or hBN permeation barriers) or explicitly moderate the permanence/contactless claim to the computed relaxed geometry.
- [Sec. 5 and Table 1] The derived areal carrier densities (2.6 and 3.7×10^14 cm^-2) depend directly on the assumed acceptor coverage (~1/3), which is a free parameter chosen by a heuristic. No coverage-dependence study or alternative-functional sensitivity check is reported. Since the paper itself notes that CT is not proportional to the initial band offset, a coverage/functional sweep would substantiate the quantitative 'up to' values and the claimed predictive power of the screening. At minimum, report CT for another coverage and a PBE or hybrid-functional spot check.
minor comments (4)
- [References] The chemical capacitor concept is introduced with citation [3], which is a TMD review; the relevant conceptual references appear to be [4] and [16]. Please correct the citation throughout, including the Conclusions.
- [Sec. 5, Eq. (3)] The integration limits in the ΔN_i formula are not defined. Please specify the energy window (e.g., occupied states up to the Fermi level) and state how the isolated-layer energies are aligned relative to the heterostructure.
- [Table 2 and Sec. 4] The alignment of VBM/CBM relative to the 'He 1s peak' is mentioned only in the table caption. Add a description in the main text of how this core-level alignment was performed, since the screening conclusions depend on it.
- [General / SI] Typos and language issues: 'occuring' in Sec. 5; 'preoptimized' in SI S1; 'the starting geometry reflects well the optimized TMD|hBN and NgF2 geometries' is awkward. Figures 1 and 2 are referenced but not visible in the submitted text; please ensure they are included.
Circularity Check
No significant circularity: the central charge-transfer values come from self-consistent full-heterostructure DFT, not from the screening offsets or fitted parameters.
full rationale
The paper's derivation chain is self-contained: isolated-material DFT band edges provide a screening criterion (type-III alignment), and the reported hole concentrations (0.23 h+/W in WS2|hBN|XeF2 and 0.35 h+/W in WSe2|hBN|KrF2) are obtained from separate periodic supercell DFT calculations of the full relaxed heterostructures. The charge transfer is not a restatement of the initial band offsets; the paper explicitly notes that 'the transferred charge is not proportional to the initial band offset determined for the isolated constituents.' The PDOS normalization used to estimate ΔN is a post-processing projection analysis, not a fit of any parameter to the target CT; the normalization factor is fixed using isolated layers and then applied unchanged to the heterostructure. The 'chemical capacitor' framing cites prior work by the same group (e.g., refs [4] and [16]), but the quantitative evidence in this paper is computed here and does not reduce to those citations. The main caveats—sensitivity of PDOS-derived charges to projection scheme and the possible reactivity of XeF2/KrF2—are correctness or robustness concerns, not circularity. No equation or construction makes the predicted CT equal to its input by definition.
Assumptions & free parameters
free parameters (2)
- Acceptor coverage fraction =
~1/3 of B atoms
- Strain threshold for supercell matching =
<1%
assumptions (4)
- domain assumption SCAN+rVV10 DFT accurately describes band alignments and charge transfer in TMD/hBN/NgF2 heterostructures
- domain assumption Band-edge alignment referenced to the He 1s core level predicts the direction and rough magnitude of charge transfer
- domain assumption The PDOS normalization method yields quantitative per-layer charges
- domain assumption hBN acts as an inert dielectric separator preventing chemical reaction between TMD and noble-gas fluoride
Cite this review
Pith. "Pith review of Huge hole injection in tungsten dichalcogenide heterostructures without electric gating: a DFT study." pith.science (2026). https://pith.science/paper/7BHN6S2L
@misc{pith2026260720032,
author = {Pith},
title = {Pith review of: Huge hole injection in tungsten dichalcogenide heterostructures without electric gating: a DFT study},
year = {2026},
howpublished = {\url{https://pith.science/paper/7BHN6S2L}},
note = {Machine review of arXiv:2607.20032}
}
read the original abstract
Van der Waals heterostructures based on transition metal dichalcogenides, TMDs, provide a versatile platform for tailoring electronic properties through interlayer charge transfer, CT. Precise control of CT is essential because it directly determines the electronic structure and carrier concentration in atomically thin materials. Recently, the concept of a chemical capacitor has been proposed as a route to achieving exceptionally high carrier densities through CT across insulating separator layers. Here, we extend this concept to van der Waals heterostructures by investigating TMD hBN OX, oxidizer, systems using density functional theory, DFT. Following the screening of candidate TMDs and electron acceptors, XeF2 and KrF2 were identified as suitable acceptors exhibiting type III broken gap band alignment with WS2 and WSe2, respectively. Periodic DFT calculations of large supercells reveal CT corresponding to hole concentrations of up to 0.23 h+ and 0.35 h+ per W atom in WS2 hBN XeF2 and WSe2 hBN KrF2 heterostructures, respectively. The resulting charge redistribution demonstrates that noble gas fluorides provide an efficient route for noncontact engineering of carrier density in TMD heterostructures, offering a new strategy for tuning correlated electronic phases in two dimensional materials.
Figures
Reference graph
Works this paper leans on
-
[61]
Kohulak O and Martonak R 2022 New High-Pressure Phases of MoSe2 and MoTe2 Phys. Rev. B 95, 054105
2022
-
[62]
Science Jul 29;353(6298):aac9439
Novoselov K S, Mishchenko A, Carvalho A and Castro Neto A H 2016 2D materials and van der Waals heterostructures. Science Jul 29;353(6298):aac9439
2016
-
[63]
Nat Rev Mater 1, 16042
Liu Y et al 2016 Van der Waals heterostructures and devices. Nat Rev Mater 1, 16042
2016
-
[64]
Hu Z et al 2022 Hernández-Martínez PL, et al Interfacial charge and energy transfer in van der Waals heterojunctions. InfoMat. 4(3):e12290
2022
-
[65]
Annual Review Physical Chemistry
Sangwan V K and Hersam M C 2018 Electronic Transport in Two-Dimensional Materials. Annual Review Physical Chemistry. 69:299-325
2018
-
[66]
Nature Nanotech 9, 682–686
Hong X et al 2014 Ultrafast charge transfer in atomically thin MoS2/WS2 heterostructures. Nature Nanotech 9, 682–686
2014
-
[67]
Kresse G and Hafner J 1993 Ab initio molecular dynamics for liquid metals Physical review B 47,
1993
-
[68]
Kresse G and Furthmüller J 1996 Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set Computational Materials Science 6,
1996
Show all 17 references
-
[69]
Kresse G and Furthmüller J 1996 Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set Physical review B 54, 11169
1996
-
[70]
Kresse G and Hafner J 1994 Norm-conserving and ultrasoft pseudopotentials for first-row and transition elements Phys.: Condens. Matt. 6,
1994
-
[71]
Blöchl P E 1994 Projector augmented-wave method Physical review B 50, 17953
1994
-
[72]
Kresse G and Joubert D 1999 From ultrasoft pseudopotentials to the projector augmented-wave method Physical review B 59,
1999
-
[73]
Sun J, Ruzsinszky A and Perdew J P 2015 Strongly constrained and appropriately normed semilocal density functional Phys. Rev. Lett. 115 036402
2015
-
[74]
Sabatini R, Gorni T and de Gironcoli S 2013 Nonlocal van der Waals density functional made simple and efficient Phys. Rev. B 87 041108(R)
2013
-
[75]
Bengtsson L 1999 Dipole correction for surface supercell calculations Phys. Rev. B 59 12301–12304
1999
-
[76]
Jezierski D et al 2022 Charge doping to flat AgF2 monolayers in a chemical capacitor setup. Phys. Chem. Chem. Phys. 2022; 24 (26): 15705–15717
2022
-
[77]
Lazić P 2015 CellMatch: Combining two unit cells into a common supercell with minimal strain Computer Physics Communications 197, 324-334. Huge hole injection in tungsten dichalcogenide heterostructures without electric gating: a DFT study Dawid Ciszewski1* and Wojciech Grocha...
2015
Reviewed August 1, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.