Wafer-scale lateral self-assembly of mosaic Ti3C2Tx MXene monolayer films

Summary¶

Mojtabavi et al. demonstrate wafer-scale self-assembly of Ti\(_3\)C\(_2\)T\(_x\) MXene monolayers into mosaic flake films with high coverage, using a liquid/liquid interfacial route that can transfer films to Si or glass substrates for device-oriented characterization. The experimental core combines controlled etching/dispersion of MXene from Ti\(_3\)AlC\(_2\) precursors (including MILD-type processing as described in Methods) with interfacial trapping that laterally orders flakes without requiring specialized Langmuir barrier hardware in the same way as classic LB workflows. ReaxFF molecular dynamics supplies an atomistic explanation for why MXene–solvent interactions stabilize flakes at an immiscible solvent interface, providing interaction energy trends that rationalize experimental yield and coverage sensitivities. Adri C. T. van Duin is a coauthor, connecting the study to the group’s interface modeling expertise.

Methods¶

Experiments (MXene + wafer-scale films). Ti\(_3\)C\(_2\)T\(_x\) MXene is prepared and suspended using MILD-type or related etch/dispersion protocols detailed in ACS Nano (MAX precursor Ti\(_3\)AlC\(_2\), etc.). A liquid–liquid interfacial self-assembly route (refined from prior work cited in the article) condenses monolayer-rich flakes and enables transfer to Si or glass substrates. Characterization on 3-inch (and scale-up) wafers includes optical/reflectance images, AFM for thickness, TEM/SEM morphology, Raman and ellipsometry mapping, and sheet resistance vs number of layer-by-layer depositions. The abstract reports <10% void area in optimized monolayer films by the cited microscopy.

MD application (ReaxFF molecular dynamics, interface support). The ACS Nano main text and Table S2 describe ReaxFF molecular dynamics (the MD engine—e.g. LAMMPS-class integration as in related ReaxFF distributions—is specified in the Supporting Information). A 21.56 × 26.67 Å² Ti₃C₂-type MXene slab is equilibrated in bulk water under NVT at 500 K to form –OH edge terminations, then desolvated. Ternary cells (chloroform / methanol / water, 1:8 by volume), 43.15 × 46.47 × 16.87 Å³, place edge-terminated MXene at the water↔chloroform interface or in a single phase; a separate orthorhombic “edge-free” MXene tile (26.25 × 30.30 Å in xy; z ≈ 40 Å per solvent) separates immiscible solvents for energy sampling. For the edge-free configurations: NPT equilibration 100 ps (P = 1 atm, T = 300 K, Berendsen barostat/thermostat, damping 5000 and 100 fs, respectively) followed by NVT 500 ps; running-average potential energy uses a ~150 ps window. A longer NPT (1 atm, 300 K) run of 1 ns with the same Berendsen controls is used in the discussion of methanol diffusion to the –OH-terminated MXene (Figure S12 / main text at p. 630). 3D PBC in all cells. Shear, shock, electric field, replica, metadynamics — N/A for the documented benchmarks.

Static QM. N/A (no standalone DFT production section; ReaxFF parameter file per SI).

FF reparameterization. N/A (existing ReaxFF; SI Table S2).

Findings¶

Outcomes and mechanisms. Interfacial assembly yields mosaic monolayer Ti\(_3\)C\(_2\)T\(_x\) films with high coverage and controllable bilayer/trilayer build-up, with optical and sheet-resistance metrics consistent with layer count. ReaxFF MD is presented as an interfacial stabilization mechanism: the flake–solvent energetics favor trapping at the immiscible solvent interface, the prerequisite in the paper’s account for lateral ordering.

Comparisons. The article contrasts this barrier-free interface method with more instrument-heavy Langmuir approaches and with prior air–water versions that can suffer from convection/variable thickness during deposition; numerical comparisons should be read from the primary PDF and figures.

Limitations of the MD layer. The reactive MD is qualitative/energetic support for interface trapping and reorganization, not a full device transport model; T\(_x\) termination heterogeneity remains a process variable.

Limitations¶

ReaxFF does not provide electronic band structures or defect physics at DFT accuracy; MXene surface terminations (T\(_x\)) vary with processing. Wafer-scale metrics include microscopy void fractions that are not atomistically resolved here.

Pair with 2D battery and conductive film pages carefully—MXene applications span energy storage and EMI shielding, but this article’s evidence is assembly + MD interface stabilization.

Relevance to group¶

Combines ReaxFF interface modeling with experimental MXene assembly (van Duin co-author).

Citations and evidence anchors¶

ACS Nano 15, 625–636 (2021); DOI 10.1021/acsnano.0c06393 — Materials and Methods and Results.