Small-diameter bamboo is a promising raw material for high-performance composites used in outdoor construction. This study aimed to compare the raw-bamboo characteristics of Bambusa textilis (BT) and Bambusa emeiensis (BE), evaluate processing conditions for BT composites, and assess mold resistance and dimensional stability. BE provided a reference for raw-material characterization and a selected flexural comparison. The tested BT material had a denser vascular-bundle distribution and higher isolated-fiber tensile properties. At a target density of 1.35 g/cm³ and 10% phenol-formaldehyde (PF) loading, BT composites reached a flexural strength of 324 MPa, compared with 234 MPa for BE, a flexural modulus of 33 GPa, and a tensile strength of 358 MPa. Thickness swelling was 3.3% and water absorption approximately 16% after 18 days of room-temperature immersion. In a separate BT series at 1.20 g/cm³ and 20% PF, 21% nominal CMIT/MIT formulation retention prevented visible Botryodiplodia theobromae growth on both evaluated faces and Aspergillus niger growth on the front face; side-face A. niger coverage was 35% at day 28. These results support further development of high-strength BT composites with improved mold resistance.

Raw lignocellulosic biomass offers a low-input route for nutrient capture, yet its internal tissue heterogeneity is rarely considered when adsorption behavior is interpreted. Here, Giant Juncao (Pennisetum giganteum; also reported as Cenchrus fungigraminus) straw was separated into rind and pith to determine how native tissue architecture and cell-wall chemistry relate to ammonium (NH₄⁺) and phosphate adsorption. The pith occupied 72.33% of stem volume but only 26.85% of dry mass and had a much lower basic density than the rind (0.072 vs. 0.570 g·cm⁻³). It also showed a larger N₂-accessible specific surface area (2.67 vs. 1.13 m²·g⁻¹), greater pore volume (0.006 vs. 0.001 cm³·g⁻¹), higher pectin content (20.58 vs. 14.81 mg·g⁻¹), and lower cellulose crystallinity (48.28% vs. 57.30%). These contrasts were accompanied by tissue-dependent NH₄⁺ uptake: the rind and pith adsorbed 4.19 ± 1.02 and 10.36 ± 1.54 mg·g⁻¹, respectively, whereas whole straw reached 13.80 ± 0.23 mg·g⁻¹ under the same comparison conditions. Phosphate uptake was less tissue-sensitive (19.57–22.40 mg·g⁻¹). More than 90% of the 24 h uptake was reached within 60 min. Nonlinear least-squares fitting, evaluated by multiple criteria (R², SSE, RMSE, χ², or AIC), identified the pseudo-second-order kinetic model and the Langmuir–Freundlich isotherm as the best empirical descriptions of the measured data. Spectroscopic changes were consistent with participation of oxygen-containing cell-wall groups in nutrient retention, but did not uniquely resolve molecular binding pathways. The results show that tissue-scale architecture and heterogeneous lignocellulosic chemistry jointly shape nutrient uptake in minimally processed Giant Juncao and support further evaluation of the unseparated feedstock in real wastewater and nutrient-reuse scenarios.