Robust 2D layered MXene matrix-boron carbide hybrid films for neutron radiation shielding

High-reliability electronics used in aerospace, automotive, nuclear, and medical environments face failures from external radiation. Neutrons are especially penetrating and harmful; shielding leverages scattering (e.g., hydrogenous materials) and capture mechanisms. Hydrogen-based scatterers require centimeter-scale thicknesses to thermalize 2 MeV neutrons, limiting form factors. Boron-containing materials, especially boron carbide (B₄C), are attractive absorbers due to the high thermal neutron capture cross section of ¹⁰B, high melting point, and low density. However, implementing high B loading with uniform distribution in practical, flexible formats is difficult: metal-matrix composites suffer from poor B solubility and heterogeneous structures; monolithic B₄C densification is challenging; polymer composites need surface modification or expensive functionalization; aqueous B₄C dispersions require high solids or dispersants. Two-dimensional MXenes (e.g., Ti₃C₂Tₓ) offer solution processability, mechanical robustness, and conductivity, but assembling them with guest particles into functional architectures remains challenging. This study addresses the question: can layered, solution-processed Ti₃C₂Tₓ MXene serve as a robust, flexible matrix to host high fractions of uniformly dispersed B₄C to achieve thin, scalable neutron-shielding films and paints with strong mechanical integrity and high neutron attenuation?

The paper reviews neutron shielding strategies, contrasting neutron moderation in hydrogenous media (water, concrete, polyethylene) with absorption by nuclei such as ¹⁰B. Among boron compounds (B₂O₃, BN, B₄C, elemental B), B₄C is widely used due to favorable physical properties and reduced secondary gamma emissions compared to impure B powders. Prior B₄C composites include borated stainless steels, B₄C-reinforced Al alloys, and polymer matrices; these often lack flexibility, exhibit heterogeneous distributions, limited B content, or require complex/expensive surface functionalization. MXenes have been demonstrated as strong, conductive films for EMI shielding and other applications, but incorporating fillers while maintaining ordered architectures is nontrivial. The authors position their work at the intersection of MXene assembly and B₄C-based neutron absorbers to overcome these limitations.

Materials synthesis and preparation:

• Ti₃AlC₂ (MAX phase) synthesis: TiC (99.5%, 2 µm), Ti (99.5%, 325 mesh), and Al (99.7%, 30 µm) powders were ball-milled at a 2:1:1 molar ratio for 24 h. The mixture was heated at 5 °C min⁻¹ to 1450 °C, held 2 h under Ar, then ground and sieved (325 mesh).
• Ti₃C₂Tₓ MXene synthesis: 2 g Ti₃AlC₂ was etched in a mixed solution of HF (4 mL, 48%), H₂O (12 mL), and HCl (24 mL, 36.5–38%) for 24 h at 35 °C to produce multilayer MXene. The mixture was washed by centrifugation at 3500 rpm for 5 min cycles until pH ~6. Li⁺ intercalation: 2 g LiCl in 40 mL DI water stirred with multilayer MXene for 12 h, followed by washing (3500 rpm, 5 min) until self-delamination occurred. Delaminated supernatant collected; MXene was further separated (3500 rpm, 30 min) and concentrated (12,000 rpm, 10 min).
• Nano-sized B₄C (n-B₄C) preparation: 2 g as-received B₄C dispersed in DI water, sonicated in ice bath for 2 h, centrifuged at 4000 rpm for 15 min to sediment large particles. Supernatant (<300 nm particles) collected and vacuum-dried at 60 °C for 12 h. Surface chemical state was modified to improve electrostatic dispersibility while maintaining crystallinity.

Hybrid dispersion and film/paint fabrication:

• Stable MXene/n-B₄C (MB) colloids prepared by mixing negatively charged Ti₃C₂Tₓ flakes and n-B₄C, yielding ζ-potential ~ -38.2 mV; stable without aggregation for ≥7 days across 0–100 wt% n-B₄C.
• PVA binder addition: 20 wt% PVA (Mw 85,000–124,000; 99+% hydrolyzed) added as interlayer binder to MB solution to form MXene/B₄C/PVA (MBP) hybrid.
• Freestanding MBP films (vacuum-assisted filtration): With fixed MXene+n-B₄C mass of 50 mg, prepared MB 20/40/60 by mixing 4 mg mL⁻¹ MXene and 1 mg mL⁻¹ n-B₄C at volumes (MXene:n-B₄C) of (10:10), (7.5:20), and (5:30) mL, respectively. Added PVA solution to achieve 20 wt% PVA relative to solids; stirred 1 h. Filtered on polycarbonate membrane (0.1 µm pores), peeled off, dried ambiently.
• Painted MBP films (blade-coating): Concentrated paints prepared by centrifugation/redispersion to obtain 20 mg mL⁻¹ MXene ink. Mixed 5 mL of 20 mg mL⁻¹ MXene with 5 mL of 20 mg mL⁻¹ n-B₄C; added 1 mL of 100 mg mL⁻¹ PVA; shaken 15 min, sonicated 5 min; degassed under vacuum 1 h. Substrates (stainless steel, glass, nylon membrane) cleaned (acetone, IPA, DI), glass UVO-treated 15 min. Blade-coated and dried at ambient.

Characterization:

• Structural: XRD (Cu Kα, Rigaku D/MAX2500V/PC), SEM (Hitachi S-4800), TEM/SAED (FEI Tecnai G2 F20), AFM (Bruker Dimension, tapping), XPS (Thermo K-alpha), FTIR (Varian 670), DLS/ζ-potential (Malvern Nano ZS). BET surface area and BJH pore analysis via N₂ adsorption/desorption at 77 K (Micromeritics ASAP 2420) after degassing at 80 °C for 12 h.
• Mechanical: Tensile tests on freestanding strips (5×20 mm, gauge ~10 mm) using LS1 (Lloyd), 10 N load cell, 0.01 mm s⁻¹; ≥5 specimens/sample. Bending tests on MBP/nylon strips (15×40 mm) with deformation tester (CKMF-12P) at bending radii 5.9 and 2.8 mm; resistance monitoring; optional heat-treatment at 100 °C for 1 h.
• Neutron-shielding tests: Conducted at KRISS with an SP9 ³He proportional counter and ²⁴¹Am–Be source (1.227×10⁷ s⁻¹). Standard thermal neutron field with graphite piles; detector Cd-shielded except front. Measured total counts (R), epithermal counts with Cd front (Rₑ), and sample-transmitted counts (Sₜ, Sₑ). Thermal neutron permeability P = (Sₜ − Sₑ)/(R − Rₑ). Half-value layer (HVL) determined via repeated coatings and fitting; HVL = ln(2)/Σ.
• MCNP 6.1 simulations: Planar circular neutron source and planar square detector (both 2 cm), Maxwell–Boltzmann energy distribution at 311 K for thermalized ²⁴¹Am–Be. Geometry per Supplementary Fig. 22. Inputs: elemental composition, bulk density, thickness for MBP20/40/60. 10⁶ neutron histories; transmission probability computed. Simulations included natural-B and ¹⁰B-enriched B₄C scenarios (N-MBP, E-MBP).

• Stable, homogeneous MXene/B₄C colloids achieved across 0–100 wt% n-B₄C, ζ-potential ≈ -38.2 mV; no aggregation/sedimentation for ≥7 days.
• Layered, ordered MBP films formed with B₄C contents 20–60 wt%; B, C, Ti uniformly distributed (EDS mapping). Raman showed B₄C bands (~30 and ~1100 cm⁻¹) increasing with B₄C fraction.
• Mechanical properties: Ti₃C₂Tₓ/20 wt% PVA film tensile strength 119.38 ± 11.29 MPa; failure strain 5.26 ± 0.36%. Strength decreases with increasing B₄C, but MBP20 and MBP40 retain good properties, outperforming a commercial resin-based shielding material (Mirrobor™) in strength at lower thickness. MBP60 weaker due to reduced flake alignment/packing.
• Porosity: Type IV N₂ isotherms with mesoporous hysteresis. BET surface area increased from 23.8 m² g⁻¹ (MBP20) to 45.6 m² g⁻¹ (MBP60). BJH pore volume increased from 0.11 to 0.21 cm³ g⁻¹ (MBP20→MBP60), pore sizes ~2–60 nm.
• Painted films: Blade-coated MBP40 formed uniform, adherent coatings on stainless steel, glass, and nylon fabric; thickness scales nearly linearly with number of coats and is uniform over ~25×50 mm areas (e.g., t_avg ≈ 113.1 ± 1.3 µm across 10 points). Good adhesion (tape test) and maintained substrate mechanical strength.
• Flexibility/durability: Painted MBP/nylon withstood ≥20,000 bending cycles at radii 5.9 and 2.8 mm with minor resistance changes; heat treatment at 100 °C for 1 h improved flexibility (likely PVA chain rearrangement over Tg ≈ 80 °C). Surfaces remained crack-free.
• Neutron shielding: High macroscopic cross-section comparable between freestanding and painted films of the same composition (reported ≈ 85.02 ± 9.32 cm² for large-area painted MBP). HVL values: ≈ 40 µm for filtrated MBP and ≈ 140 µm for painted MBP, markedly lower than other boron-containing structural materials. Absorption capacity 39.8% for neutrons from a ²⁴¹Am–Be source.
• Simulations: MCNP results agreed with experiments. Natural-B MBP (N-MBP) showed reduced macroscopic cross-sections relative to enriched-B (E-MBP), consistent with lower ¹¹B absorption. Fitted absorbance vs thickness of N-MBP comparable/slightly better than conventional composites, supporting feasibility with natural B₄C.
• Practical metrics: Freestanding MBP density ~2.0 g cm⁻³ within 20–60 wt% B₄C; thermal stability up to ~180 °C with <5% weight loss aside from PVA decomposition. Films are ultrathin, lightweight (can rest on dandelion tips) and scalable to several hundred cm².

The study demonstrates that Ti₃C₂Tₓ MXene provides a robust, flexible, layered matrix capable of hosting high loadings (up to 60 wt%) of uniformly dispersed nano-B₄C, overcoming traditional challenges of agglomeration, poor interfacial bonding, and limited processability in metal or polymer matrices. The resulting freestanding and painted MBP films achieve anomalously high neutron attenuation at ultralow thicknesses (HVL ≈ 40–140 µm) while maintaining mechanical flexibility and durability under repeated bending. Structural uniformity across thicknesses is supported by agreement between experimental HVL fits and the theoretical HVL = ln(2)/Σ relation, as well as by MCNP simulations for both natural and ¹⁰B-enriched B₄C. The ability to blade-coat large, uniform, adherent films on diverse substrates (stainless steel, glass, nylon) with minimal resistance change under mechanical deformation highlights practical applicability for lightweight, wearable, and conformal neutron shielding. The combination of scalable solution processing, high boron capacity, and mechanical robustness addresses the initial design goals of integrating high B₄C content into ultrathin, flexible, structurally stable shields.

This work introduces MXene/B₄C/PVA (MBP) hybrid films and paints that integrate high B₄C content (20–60 wt%) into an ordered, layered Ti₃C₂Tₓ matrix, yielding thin, flexible, and scalable neutron shields. Key contributions include: (i) a stable MXene–B₄C colloidal platform enabling homogeneous dispersion and high loading; (ii) freestanding and blade-coated films with uniform microstructures, strong substrate adhesion, and excellent bending durability; (iii) high neutron attenuation with ultralow half-value layers (~40 µm filtrated; ~140 µm painted), validated by experiments and MCNP simulations; and (iv) straightforward, scalable fabrication without high pressure or complex functionalization, compatible with large-area and curved surfaces. Future work could enhance thermal resilience and long-term stability by replacing PVA with higher-temperature polymers (e.g., polyimide), optimizing interfacial chemistry for MBP60 and above to recover mechanical strength, and exploring integration into multifunctional MXene coatings (e.g., combining EMI and neutron shielding) and wearable textile platforms.

• Mechanical properties degrade at the highest B₄C loading (MBP60) due to reduced MXene flake alignment and packing density, weakening interflake interactions.
• Thermal stability is limited by the PVA binder; noticeable decomposition above ~180 °C suggests the need for higher-temperature polymers for harsh environments.
• Vacuum filtration, while producing highly aligned films, is less practical for complex geometries; the paint approach mitigates this but yields a higher HVL (~140 µm) than filtrated films.
• Use of natural B₄C results in lower macroscopic cross-sections compared to ¹⁰B-enriched B₄C due to the low absorption of ¹¹B, although performance remains competitive with conventional composites.