Lanthanum‑Doped Halloysite Nanotubes Significantly Boost CdS Photocatalytic Hydrogen Production

Background

Aluminosilicate minerals—kaolinite, zeolite, montmorillonite, and halloysite—are attractive catalyst supports due to their environmental benignity and abundance. Techniques such as polymer coating, carbon coating, and atomic doping have been employed to tailor their surface chemistry and improve catalytic performance. Metal‑ion incorporation can impart unique electronic and mechanical properties, opening avenues for catalysis, drug delivery, and lithium‑ion batteries. While density‑functional theory (DFT) has elucidated stability and electronic structures of nanostructured aluminosilicates, the precise mechanism of substitutional doping and its impact on electronic structure remain unclear.

To address this gap, we developed an efficient La‑doping strategy for halloysite nanotubes (HNT). La replaces Al in the alumina sheet via a dynamic equilibrium in saturated AlCl₃ solution, preserving the tubular morphology while subtly expanding interlamellar spacing. CdS, a visible‑light‑responsive semiconductor, was subsequently loaded onto La‑HNT, enabling us to probe the effect of La on photocatalytic water splitting.

Methods

Experimental Section

Materials Preparation

Halloysite nanotubes (HNT) were sourced from Hunan, China. All reagents were analytical grade. HNT were pretreated by emulsion dispersion, filtration, washing, and drying at 313 K for 8 h. La‑HNT was prepared by mixing 34.3 g AlCl₃ in 60 mL water to generate a supersaturated solution, then adding 3 mmol HNT and 6 mmol La(NO₃)₃·6H₂O dissolved in 5 mL water. The suspension was stirred, sonicated, and transferred to a 100 mL Teflon‑lined autoclave, then heated to 373 K for 48 h. After cooling, the product was filtered, washed, and dried at 353 K under vacuum. For comparison, acid‑treated HNT was obtained by refluxing 1 g HNT in 250 mL 6 M HCl at 373 K for 4 h.

CdS/La‑HNT was fabricated via successive ionic layer adsorption and reaction (SILAR). 3 mmol La‑HNT was alternately immersed in 0.5 M Cd(NO₃)₂ ethanol and 0.5 M Na₂S methanol solutions for 5 min each, repeating until ~11 wt % CdS loading was achieved.

Characterization

Phase purity was confirmed by XRD (Cu Kα, λ = 1.5418 Å). Morphology and size distribution were examined by TEM (JEOL JEM‑200CX, 200 kV) and SEM. FTIR spectra were recorded on a Nicolet 5700. Surface area and porosity were measured via nitrogen adsorption (BET). XPS (Thermo Fisher K‑Alpha) assessed elemental composition and chemical states. UV–vis diffuse reflectance spectra were obtained with a Shimadzu UV‑2400. Photoluminescence (PL) was recorded on a Hitachi H‑4500. Electrochemical measurements employed a three‑electrode cell with FTO working electrode, Pt‑black counter electrode, and SCE reference. Photocurrent and impedance were measured using a CHI‑660A workstation.

Photocatalytic Reaction

Water splitting was performed in a 300‑mL aqueous solution containing 0.1 M Na₂S and 0.1 M Na₂SO₃ under a 300‑W Xe lamp (135 mW cm⁻²). Hydrogen evolution was quantified by gas chromatography (Agilent 6890 N).

Computational Details

First‑principles DFT calculations were carried out with CASTEP, employing the LDA functional and ultrasoft pseudopotentials. A 400 eV plane‑wave cutoff and 3×3×1 k‑point mesh were used. Geometry optimizations converged to forces < 0.03 eV Å⁻¹. The La‑doped model was generated by substituting an Al atom in the alumina sheet of HNT with La.

Results and Discussion

The tubular morphology of HNT was preserved after La doping (Fig. 1), with an average La content of 4.2 %. XRD patterns (Fig. 2a) confirm that the halloysite phase remains intact, while a slight shift of the (001) peak indicates expanded interlamellar spacing. FTIR spectra (Fig. 2b) show shifts of the Al–OH and Al–OSi bands, confirming Al substitution by larger La ions. BET analysis (Fig. 2d) reveals a modest decrease in surface area (59 m² g⁻¹) and pore volume, accompanied by an increase in average pore diameter (25 nm), consistent with lattice expansion.

CdS nanoparticles (~5 nm) were uniformly anchored on both HNT and La‑HNT (Fig. 3). TEM images (Fig. 3d,f) confirm dense, homogeneous coverage. XPS (Fig. 5) confirms successful La incorporation and indicates a shift in Al binding energies, further evidence of substitution.

Optical studies (Fig. 4a) show that CdS/La‑HNT exhibits stronger visible‑light absorption and a slightly reduced band gap (2.25 eV) compared to CdS/HNT (2.31 eV). Photocatalytic hydrogen evolution (Fig. 4b) demonstrates that CdS/La‑HNT achieves 47.5 µmol h⁻¹—nearly double that of CdS/HNT (26.0 µmol h⁻¹). Transient photocurrent measurements (Fig. 4c) and impedance spectroscopy (Fig. 4d) reveal lower charge‑transfer resistance and higher photocurrent density for the La‑doped composite, indicating more efficient charge separation. PL spectra (Fig. 4e) corroborate this, showing reduced recombination in CdS/La‑HNT.

DFT results (Fig. 8) reveal that La introduces donor‑like 5d states near the conduction band minimum, facilitating electron transfer. Charge‑density difference maps show electron transfer from surrounding Al atoms to La, supporting the role of La as a charge‑transfer bridge between the support and CdS.

Conclusions

We have successfully incorporated La into halloysite nanotubes, inducing subtle structural expansion and creating reactive surface sites. This modification enhances CdS loading uniformity and dramatically improves photocatalytic hydrogen evolution under visible light. The strategy offers a generalizable route for doping aluminosilicate clays and tailoring their catalytic properties.