Unexpectedly efficient ion desorption of graphene-based materials

The study addresses the challenge of desorbing strongly adsorbed ions from high-performance graphene-based sorbents, a critical step in adsorption separation technologies used for extraction, concentration, and purification. Graphene-based materials exhibit strong cation adsorption due to their atomically thin, aromatic surfaces and cation-π interactions, enabling applications such as ion sieving and salt crystallization at sub-saturation concentrations. However, conventional ion desorption requires large volumes of concentrated acids or bases (e.g., 0.1–0.2 M HCl or NaOH) and long contact times (1–2 hours), which are inefficient and resource-intensive. The authors propose using small amounts of Al3+ to rapidly and efficiently desorb divalent ions (e.g., Co2+, Mn2+, Sr2+) from magnetite-graphene oxide (M-GO), hypothesizing that stronger hydrated cation-π interactions of Al3+ with graphene will drive ion substitution and enable efficient desorption and potential enrichment of target ions such as radioactive 60Co.

Background literature emphasizes adsorption separation as an effective approach across energy and environmental applications, and highlights the strong ion-surface interactions on graphene-based materials arising from cation-π interactions. Prior works show graphene oxide and its magnetic composites effectively adsorb multivalent metal ions (e.g., Co2+, Mn2+, Sr2+, Cu2+, Cd2+, Cr3+, Pb2+) and enable ion sieving via interlayer spacing control. Conventional regeneration of adsorbents typically relies on high-concentration acids/bases with significant reagent use and time. Theoretical and experimental reports support hydrated cation-π interactions at graphitic interfaces and ion enrichment on hydrophobic carbon surfaces. These studies motivate exploring alternative, milder desorption strategies leveraging differential adsorption strengths among ions, particularly the potentially stronger interaction of Al3+ compared to mono- and divalent metal ions.

Materials and synthesis: Graphene oxide (GO) was synthesized from natural graphite using a modified Hummers method. Magnetite-graphene oxide (M-GO) was prepared by chemical coprecipitation of Fe3+ and Fe2+ to deposit magnetic iron oxide nanoparticles on GO under alkaline conditions. Characterization included TEM, XRD, Raman spectroscopy, XPS, and VSM.

Radioactive 60Co enrichment experiment: Step 1: 60 mg M-GO was added to 300 mL of solution containing 15 Bq/L 60Co and 1.0 mg/L Co2+. Step 2: Mixture stirred at 298 K for 5 min, then magnetically separated; the 60Co- and Co2+-loaded M-GO (60Co@M-GO) was redispersed in deionized water to 30 mL. Step 3: 60 µL Al3+ stock solution was added to reach 20 mg/L Al3+, stirred at 298 K for 5 min, then separated by magnetic separation and filtration. Radioactivity in filtrates was measured by a high-purity germanium γ spectrometer (GEM-100).

Desorption kinetics for Co2+, Mn2+, Sr2+: 200 mg M-GO was added to 200 mL of 10 mg/L Co2+, Mn2+, or Sr2+ solutions, stirred at 298 K for 125 min to reach adsorption equilibrium. A negligible volume (400 µL) of concentrated Al3+ solution was then added to achieve 10 mg/L Al3+, and stirring continued at 298 K for another 125 min. At intervals from 0 to 250 min, 5 mL aliquots were withdrawn, filtered, and residual ion concentrations measured. Adsorption capacities qt and equilibrium capacities qe were calculated (per Supplementary Note 1). Desorption kinetics were analyzed via the pseudo-second-order model using Al3+ adsorption during substitution: t/qt = 1/(k2 qe^2) + t/qe.

Computational methods: Density functional theory calculations employed the M06-2X functional with Def2-SVP basis set for geometry optimization, frequency analysis (no imaginary frequencies), and energy evaluation. Both low-spin and high-spin states were considered for Co2+ and Mn2+. Adsorption energies were computed as ΔE = E(X@G) − EG − EX, where X denotes cation or hydrated cation and G is graphene. Partial charges were obtained via NBO analysis; electron density differences were analyzed with Multiwfn and visualized with VMD. Calculations examined hydrated ions Al3+-(H2O)6, Co2+-(H2O)6, Mn2+-(H2O)6, and Sr2+-(H2O)6 on graphene.

• Very low concentrations of Al3+ enable rapid and efficient desorption of divalent ions from M-GO. Compared to conventional acid/base desorption, Al3+ usage is reduced by at least a factor of 250.
• Desorption efficiencies upon Al3+ addition (10 mg/L) reached 99.9 ± 0.1% for Co2+, 97.0 ± 2.1% for Mn2+, and 98.3 ± 2.6% for Sr2+. Desorption occurred rapidly, with adsorption/desorption transitions observed within about 1 minute.
• Radioactive 60Co enrichment: After adsorption and transfer to 30 mL, adding 20 mg/L Al3+ yielded a final solution with 124.7 ± 4.1 Bq/L, a roughly 10-fold reduction in volume compared to the initial 300 mL at 15 Bq/L. The post-adsorption supernatant after Step 2 showed only 1.8 ± 0.4 Bq/L, confirming effective removal of 60Co by M-GO.
• DFT results: Hydrated ions adsorb stably on graphene with distances 2.38–2.73 Å. Calculated adsorption energies are approximately −80 kcal/mol for Co2+-(H2O)6, Mn2+-(H2O)6, and Sr2+-(H2O)6, whereas Al3+-(H2O)6 exhibits a much stronger adsorption energy of −139 kcal/mol (~75% larger magnitude). Al3+ shows the greatest reduction in cation partial charge and greatest increase in electron density on graphene, indicating stronger hydrated cation-π interaction.
• Predicted adsorption probability ratio at 300 K: PAl/PCo ≈ 1.03 × 10^42, implying Al3+ overwhelmingly outcompetes Co2+ for adsorption sites, consistent with rapid Al3+-induced desorption of Co2+.
• M-GO maintains adsorption efficiency and magnetic properties and is easily recycled and reused due to Al3+ hydrolysis behavior under alkaline conditions.

The findings validate the hypothesis that stronger hydrated cation-π interactions of Al3+ with graphene can drive displacement of previously adsorbed divalent ions, enabling rapid desorption at very low reagent levels. The experimental kinetics show near-complete desorption of Co2+, Mn2+, and Sr2+ upon low-dose Al3+ addition, while DFT confirms that hydrated Al3+ binds much more strongly to graphene than divalent ions, explaining the observed ion-substitution mechanism. The approach drastically reduces chemical consumption and time compared to conventional acid/base regeneration, addressing a key bottleneck in adsorption-based separations. The demonstrated enrichment of radioactive 60Co illustrates practical utility in concentrating target ions and reducing waste volume. Given that monovalent and common divalent ions generally have weaker adsorption than Al3+, the method is likely applicable to a broader set of ions across energy, environmental, and materials contexts, offering a scalable, reusable platform for selective ion enrichment and desorption.

The study introduces a facile, efficient desorption strategy for graphene-based adsorbents using low concentrations of Al3+ to displace strongly adsorbed divalent ions via hydrated cation-π interactions. It achieves near-quantitative desorption (≈97–100%), rapid kinetics (∼1 min), and significant reductions in reagent use relative to traditional acid/base methods. The method enables effective enrichment of radioactive 60Co with a 10-fold volume reduction and supports easy recycling and reuse of M-GO without performance loss. DFT calculations rationalize the mechanism by showing substantially stronger adsorption of hydrated Al3+ on graphene than that of Co2+, Mn2+, and Sr2+, and predict overwhelming adsorption preference for Al3+. Future research could explore broader ion systems (including various mono- and multivalent ions), optimization of operational conditions, integration with membrane or column processes, and application to real waste streams and complex matrices.