Skip to content

Nanodiamonds in Filtration and Adsorption Systems - Structure-Property Relationships, Immobilization Strategies, and Emerging Applications

FieldValue
JournalACS Applied Materials & Interfaces
Published2026-08-28
AuthorsCherppathu Sasidharan Shalumon, Štěpán Stehlík
DOI10.1021/acsami.6c11549

A hydrogenated detonation nanodiamond coating on quartz microfiber membranes reduced MS2 bacteriophage by at least 6.2 log10 units across pH 5 to 9. The review identifies this as the most directly filtration-relevant demonstration to date. Its main design rule is that the accessible nanodiamond interface is the functional unit.

Structure and termination determine aqueous behavior

Section titled “Structure and termination determine aqueous behavior”

The review separates detonation nanodiamonds (DNDs) from high-pressure high-temperature (HPHT) nanodiamonds because their structures produce different aqueous interfaces. DND primaries are typically 3 to 8 nm, with 300 to 350 m²/g specific surface area and about 2 at % nitrogen. HPHT particles commonly measure tens to more than 100 nm, with specific surface area at or below 90 m²/g and nitrogen from tens to hundreds of ppm.

Thermal oxidation in air commonly uses 420 to 450 °C for DNDs and 450 to 550 °C for HPHT particles. It creates hydroxyl, carbonyl, carboxyl, and related oxygen groups. Oxidized nanodiamonds then have negative zeta potentials that often fall below -30 mV near neutral pH.

Hydrogenation uses hydrogen-containing atmospheres or plasma and produces stable C-Hx surface bonds. Hydrogenated DNDs commonly reach +40 to +55 mV, while hydrogenated HPHT particles commonly reach +30 to +45 mV. Defect-rich, nitrogen-containing H-DNDs also show unusually strong hydration and structured interfacial water. HPHT H-NDs have negative electron affinity, whereas H-DNDs do not, despite similar hydrogen termination.

Positive H-DND colloids aggregated at ionic strengths of 10⁻³ M or higher, compared with 10⁻² M for negative O-DNDs. Identical hydrogen termination therefore does not imply identical charge, hydration, or salt tolerance.

The review separates removal from binding evidence

Section titled “The review separates removal from binding evidence”

The authors synthesize published studies from water treatment, biomedicine, and chromatography rather than report a new filtration experiment. They distinguish direct contaminant-removal data from observations that establish only interfacial binding. Biomedical binding in controlled buffers does not demonstrate contaminant capture in complex water.

Evidence includes batch isotherms, kinetic fits, potentiometric titration, XPS, infrared spectroscopy, DRIFTS, mass spectrometry, zeta potential, and flow filtration. Pseudo-second-order fits do not prove chemisorption, while Langmuir or Freundlich fits do not establish uniform adsorption sites.

The review treats common model fits as descriptive tools rather than mechanistic rankings. Head-to-head studies under identical water chemistry and flow conditions remain absent.

Adsorption results depend on contaminant and termination

Section titled “Adsorption results depend on contaminant and termination”

Oxidized DNDs capture multivalent cations through electrostatic attraction and carboxylate coordination. Pairs of adjacent deprotonated carboxylates chelate Cu²⁺, Zn²⁺, Cd²⁺, and Ni²⁺. Acid-treated DNDs showed low-concentration chemisorption energies of 8 to 12 kJ/mol, followed by mesopore physisorption at higher loading. Nitric acid treatment at 250 °C raised capacity above 100 μmol/g for every tested cation and retained sorption through at least 10 treatment cycles.

Hydrogenated DNDs behave differently because nitrogen-rich surfaces show positive electrokinetic response. Sulfate exchange capacity reached about 0.20 mmolC/g, with surface charge near 80 mC/m². This behavior weakened in DNDs containing 0.2 at % nitrogen and in HPHT material. Hydrogen termination alone therefore does not explain the anion-exchange behavior.

For PFAS, 254 nm irradiation of H-DND suspensions generated hydrated electrons that degraded PFOS into shorter fluorinated products. A 2026 preprint reported complete PFOA mineralization by hydrogenated HPHT nanodiamonds under simulated sunlight. That study was not peer reviewed, so its mechanism and performance require separate validation.

H-DNDs bound fibroblast growth factor 2 with nanomolar affinity in a 10⁵-fold excess of competing serum proteins. This result demonstrates selective binding, not filtration performance. No published study directly tested nanodiamond capture of microplastics or nanoplastics.

Benchmarks show strong metal capacities but no universal ranking

Section titled “Benchmarks show strong metal capacities but no universal ranking”

Activated carbon remains the industrial benchmark, with 500 to 2,000 m²/g area. Its reported heavy-metal capacities are typically 30 to 100 mg/g. Graphene oxide reaches about 700 m²/g, with reported capacities of 106 mg/g for Cd²⁺ at pH 6 and 46.6 mg/g for Cu²⁺.

Functionalized metal-organic frameworks reached 769 mg/g for Cu²⁺. An ethylenediamine-functionalized Zr-MOF reached 217 mg/g for Cd²⁺.

Oxidized DNDs gave 181 mg/g for Cu²⁺, 225 mg/g for Cd²⁺, 205 mg/g for Cr⁶⁺, and 192 mg/g for Fe³⁺ under comparable batch conditions. These values exceeded unmodified activated carbon and plain graphene oxide. They approached functionalized graphene oxide but remained below the highest metal-organic framework values.

Contaminant identity, concentration, sorbent form, normalization, and protocol prevent a universal numerical ranking.

Accessible surface determines filtration performance

Section titled “Accessible surface determines filtration performance”

The review distinguishes exposed surface immobilization from embedding particles in the membrane bulk. Electrostatic deposition uses opposite particle and support charges. Electrophoretic deposition adds an electric field for conductive supports. Covalent grafting uses attachment chemistry.

Each method can preserve accessible surface, but each has a different stability profile. Electrostatic seeding layers remain fragile and may detach when pH, ionic strength, or solution composition changes. Carboxylated DNDs can form denser EDC/NHS amide bonds to APTES-treated titanium. However, attachment groups may consume adsorption sites.

The strongest filtration example used dip-coating on quartz microfiber membranes followed by high-vacuum thermal annealing. The coating retained a zeta potential of +45 mV at pH 7 and an isoelectric point near pH 11. It retained acid black 2 almost completely and showed no measurable charge loss over 30 pressurized flow cycles.

Against five commercial electropositive filters, the coated quartz membrane held 35 to 52 mV across the full pH range. Commercial filters held 5 to 30 mV. The nanodiamond-coated membrane also gave the highest dye retention under equivalent test conditions.

Embedded DNDs in PVDF, PES, and polysulfone can change flux, wettability, fouling, and strength. Most embedded surfaces remain inaccessible to feed water. No general method yet stabilizes H-DNDs on polymer membranes while preserving hydrogen termination.

Durability, release, and scale remain unresolved

Section titled “Durability, release, and scale remain unresolved”

Thirty flow cycles are the only quantitative durability benchmark for a nanodiamond filtration architecture. The study did not test backwashing, chemical cleaning, or extended operation. Practical backwash can reach twice the forward transmembrane pressure, while sodium hypochlorite can attack the interface.

Particle loss, peel strength, and critical shear stress have not been quantified. Natural organic matter can alter charge, aggregation, and mobility. It may compete for surface sites or form persistent organic coronas. Hydrogenated surfaces may oxidize during prolonged water exposure, and steam sterilization has reduced their positive zeta potential.

In Daphnia magna, 5 nm DNDs and 15 nm HPHT nanodiamonds were tested at 1, 10, and 50 μg/L for 24 hours. All caused oxidative responses, although their strength depended on size, charge, aggregation, and endpoint.

Industrial nanodiamond synthesis is established. Deaggregation, coating uniformity, cost, regeneration, and life-cycle energy remain unresolved. Scalable workflows must preserve deaggregated DNDs through surface treatment, deposition, and membrane fabrication.

Material selection follows surface function

Section titled “Material selection follows surface function”

The evidence favors oxidized DNDs for cationic metal capture because their carboxylate sites provide direct coordination. Hydrogenated DNDs suit anionic species and electropositive membranes, but their behavior depends on defects, nitrogen, hydration, and salt exposure.

Surface immobilization should be selected for selective capture. Embedding should be selected for membrane matrix modification. Device development should measure accessible area, particle release, regeneration, and charge stability under realistic water chemistry.

The current benchmark remains a thermally stabilized H-DND layer on an inorganic support, because stable polymer attachment without changing H-ND surface chemistry remains unresolved.

View Original Abstract

Nanodiamonds (NDs) combine the exceptional hardness, chemical durability, and thermal stability of bulk diamond with a high surface-to-volume ratio and widely tunable surface chemistry. These properties enable broad control over surface charge, polarity, hydration, and interfacial interactions, making NDs attractive for adsorption and water-treatment applications. Nevertheless, their practical potential in these fields remains substantially underdeveloped relative to the depth of accumulated physicochemical knowledge. This review critically examines known and filtration-relevant interactions between ND surfaces and dissolved species, and evaluates how these interactions could be exploited in adsorption and filtration technologies. We compare the two technologically relevant ND classes-detonation nanodiamonds (DNDs) and high-pressure high-temperature (HPHT) NDs-and analyze the consequences of surface termination, defect structure, nitrogen content, and colloidal state for zeta potential, hydration, and interfacial reactivity in aqueous environments. A particular focus is placed on hydrogenated NDs, which exhibit an unusual combination of strongly positive electrokinetic response, pronounced interfacial water structuring, and surface electronic effects associated with charge transfer and particle structure that distinguish them from oxidized NDs and other carbon nanomaterials. We systematically review documented interactions between ND surfaces and relevant pollutant classes: multivalent metal ions, charged organic molecules, PFAS, and biomolecules, drawing on literature from water treatment, biomedicine, and chromatography to establish the breadth of ND interfacial chemistry. A central argument of this review is that the functional unit in ND-based filtration is the accessible ND surface: immobilization strategies that preserve surface exposure-electrostatic deposition, covalent grafting, and thermally assisted stabilization on inorganic supports-are critically distinguished from embedded composite approaches where ND surfaces are mostly buried and rendered passive. We identify the translation of these immobilization strategies to polymer membrane supports as the principal unresolved engineering challenge, and outline the role of ND structural heterogeneity, long-term surface stability, and scalable deaggregation as key scientific priorities for realizing the full potential of NDs in next-generation filtration systems.