English

• 1. Introduction

  Frozen solutions are extensively found in environmental media such as sea ice, permafrost, glaciers, and snow on the Earth [1]. Frozen media significantly impact the natural transformation of environmental species and pollutants, and many studies have shown that frozen solutions accelerate various environmental reactions [2–4]. This unexpected behavior is primarily ascribed to the “freeze-concentration” effects in ice grain boundaries (IGBs), where the concentration of reactants can be enhanced by several orders of magnitude compared to their aqueous counterparts [5,6]. Additionally, the quasi-liquid layer (QLL), a water layer present on the surface of ice, can act as a space-confined reactor for chemical reactions. The water molecules on the ice surface possess unsaturated hydrogen bonds with higher bond energy, favoring the formation of ultra-thin QLL at the ice-air or ice-solid/liquid interface (Fig. S1 in Supporting information) [7,8]. Theoretical molecular dynamics simulations indicated the low connectivity structure of hydrogen-bond networks on the surface of ice, increasing the diffusion coefficient of water molecules at the QLL [7]. Consequently, the physiochemical characteristics of ice result in the unique behavior of various reactions in frozen solutions.

  The principle of ice chemistry can also be utilized to guide the preparation of novel functional materials [9–12]. The link between ice and material preparation dated back to the 1930s [13], when researchers observed distinct behaviors of materials during ice melting compared to bulk water. In the 1980s, Mahler and Bechtold [14] revealed that freezing process accelerated the silica fibers formation and affected the morphology of the resulting products. Tong et al. [15] developed the directional freezing technique to prepare anisotropic ice-agar composites. Another significant progress was the application of ice-templating methods for preparing porous and layered hybrid materials with improved mechanical response [16]. This approach later expanded to ice-assisted synthetic graphene oxide (GO)-based materials that have attracted wide attention, such as GO aerogels [17] and GO-based composites [18]. Recently, ice-assisted chemistry has been employed to synthesize atomic-dispersed materials with enhanced catalytic performance [19,20]. The growing interest in this field is reflected in a significant increase in annual publications related to ice-based material science, rising from 1 in 2000 to 53 in 2023, accompanied by a corresponding surge in citations (Fig. S2 in Supporting information). Since ice is low-cost, facile, easily handled, and environmentally friendly [21], ice-assisted synthesis has attracted much attention to various environmental applications [22–24]. For example, 3D-quasilayered polyamide nanofiltration membranes were prepared by ice-confined interfacial polymerization and achieved separation of Cl^-/SO₄^2- from water [25]. Moreover, ice-templated 3D graphene-based materials have served as effective adsorbents for removing dyes and heavy metals [26]. Considering these recent advances, it would be timely to provide a comprehensive summary and forward-looking perspective on the potential environmental application of ice-assisted synthesis material.

  In sharp contrast to current reviews on reaction principles of environmental ice chemistry [3,10,27], the present article aims to review the science of ice chemistry in material synthesis for environmental applications. Fig. 1 visually outlines the scope of the present review, summarizing novel ice-assisted methods in preparing functional materials and potential environmental applications. The new synthetic approaches will stimulate further interest among environmental researchers in exploring the development of environmental functional materials based on ice chemistry.

  Figure 1

  

  Figure 1.  Schematic of ice-assisted synthetic strategies and potential environmental applications.

  DownLoad: Full-Size Img PowerPoint

  2. Recent progress of ice chemistry in functional materials synthesis

  Although hundreds of materials have been synthesized with the assistance of ice chemistry, there is an absence of generalization about the underlying synthesis principles. Here, three critical strategies of ice-assisted synthesis are classified as follows: (1) Ice-confined interfacial synthesis, (2) ice-templating physical synthesis, and (3) ice-melting chemical synthesis.

  2.1 Ice-confined interfacial synthesis

  Crushed ice granules immersed in organic solvent create a confined interfacial environment for the preparation of 2D nano-functional materials. An early example of a large-scale and rapid synthesis of 2D metal oxides utilizing the ice granules-organic solvent interface is the preparation of ultra-thin TiO₂ nano-disks (Fig. 2a) [28]. The ultra-thin TiO₂ nano-disks were synthesized by freezing improved sol-gel reaction. Specifically, crushed ice granules were introduced into an organic solvent to initiate the reaction between titanium isopropoxide and the thin water layer of ice surfaces. Subsequently, the hydrolysis of titanium isopropoxide at the ice-organic solvent interfaces produced 2D TiO₂ nano-disks. The thickness of ice-synthesized TiO₂ nano-disks ranged from 0.5 nm to 1 nm, thinner than most reported 2D metal oxides. Compared to conventional TiO₂ nanoparticles [29–31], these ice-synthesized TiO₂ nano-disks exhibited a larger surface area (>400 m²/g) and a tendency to aggregate. The ice-assisted sol-gel method might be a generalizable strategy for synthesizing various 2D metal oxides that relies on the unmixable organic solvent/inorganic ice granules interface [32,33]. The precursors (usually as metal alkoxides, such as zirconium n-propoxide, titanium isopropoxide, zinc acetate, and tungsten(Ⅵ) ethoxide) in organic solvents were allowed for the hydrolysis and condensation at the ice granules/organic solvent interface, which might be a rational strategy for designing 2D metal oxides materials (e.g., zirconia, zinc oxide, and tungsten oxide) [32,34,35].

  Figure 2

  

  Figure 2.  (a) Freezing-improved sol-gel methods for the synthesis of 2D metal oxides. Reproduced with permission [28]. Copyright 2013, the Royal Society of Chemistry. (b) The synthesis procedures of 2D PANI on bulk ice surfaces. Reproduced with permission [38]. Copyright 2015, Wiley Publishing Group. (c) Schematic of the IC-PANF membrane at the ice/n-hexane interface. Copied with permission [25]. Copyright 2023, American Association for the Advanced of Science. (d) 2D materials synthesis within ice grain boundaries (IGBs). Reproduced with permission [43]. Copyright 2023, Wiley Publishing Group. (e) Schematic diagram of ice-assisted photochemical reduction method. Reproduced with permission [44]. Copyright 2017, Springer Nature. (f) Schematic diagram of ice-assisted photocatalytic synthesis method. Reproduced with permission [19]. Copyright 2019, Elsevier.

  DownLoad: Full-Size Img PowerPoint

  Bulk ice surface facilitates the growth of 2D conducting polymers due to the presence of ultrathin QLL. At the early stage, the oxidative polymerization of aniline by ammonium peroxydisulfate produced polyaniline (PANI) after 20 days of reactions in ice (at −24 ℃) [36]. Later, Ma et al. [37] prepared PANI microflakes by freezing mixed reagents (aniline:FeCl₃:H₄SiW₁₂O₄ = 6:7:1) at −10 ℃ for 10 days. However, the synthesis of PANI in IGBs requires a long reaction time and has an unsatisfactory 2D structure. Park’s group developed a quick synthesis method of 2D PANI nanosheets on ice surfaces [38], where the reaction between aniline and ammonium peroxydisulfate (the molar ratio of 8:3) can produce 2D PANI nanosheets with a thickness of ~30 nm within just 3 min reaction. Notably, ice surfaces appeared to lower nucleation barrier for the aniline polymerization, enabling efficient polymer formation (Fig. 2b). The resulting 2D PANI nanosheets exhibited a conductivity of 35 S/cm, a few orders of magnitude higher than most HCl-doped PANI materials. Moreover, these 2D PANI nanosheets could be further developed to support Pt nanocrystals at the air-water interface [39]. Interestingly, PANI nanosheets grown on the highly crystalline pure ice surface exhibited higher conductivity than polyethylene glycol (PEG)-modified ice, as the crystallinity of ice decreased with the addition of PEG [40]. More efficient charge transport in highly crystalline ice surfaces resulted from the polaron delocalization and the enhancement of the order degree of backbones. To further extending the versatility of ice surfaces, researchers developed another 2D sheet-like poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) conducting films (i.e., 2D sheet-like PEDOT:PSS) by dropping the precursor solution on ice surfaces (−20 ℃) [41]. In conclusion, ice surfaces provided an ideal template to induce the linked network of monomers in 2D directions for polymerization, which is difficult to achieve in conventional liquid solutions.

  Recently, a novel ice-confined polyamide nano-filtration membrane (IC-PANF) with a highly ionized 3D-quasilayered structure has been fabricated at the unmixed bulk ice/organic solvent interface, formed between the frozen m-phenylenediamine (MPD)-water solution and unfrozen trimesoyl chloride (TMC)/n-hexane solution (Fig. 2c) [25]. The interfacial polymerization reaction mechanism at a confined ice-organic solvent interface during the ice-water phase transition process was clarified using molecular dynamics (MD) simulations. Initially, TMC molecules were anchored onto the MPD-ice surface due to high adsorption energy in the initial stage (below the melting temperature). Then the melting of MPD-ice induced the controllable polymerization with TMC molecules due to the slow diffusion efficiency and high binding free energy between MPD and ice. After the MPD phase further increased to ambient temperature, the reaction rate between MPD and TMC decreased and continuously created new layers during the gaps and channels of the previously formed IC-PANF. At this point, the release of dissolved gases from ice further contributes to generating the internal microporous-layered structure of IC-PANF. In contrast, at ambient temperature, the diffusion rate of monomers cannot be effectively controlled, resulting in insufficient crosslink and hindering the formation of polyamide nano-filtration membrane. This novel ice-confined interfacial strategy may be extended to synthesize diverse advanced materials based on liquid-liquid interface synthesis.

  Ice grain boundaries (IGBs) contain confined space and concentrated reactants during the freezing process, supporting the in-situ chemical synthesis of 2D materials (Fig. 2d). Zhang et al. [42] reported that freezing a commercial PEDOT:PSS suspension induces the squeezing of PEDOT:PSS grains within IGBs, disrupting the hydrogen bonds and electrical attraction. This process facilitates the untangling of PEDOT: PSS chains and the alignment of 2D PEDOT:PSS sheets along the ice crystal growth direction. Here, the directional freezing technique enables the adjustment of the assembly direction of 2D PEDOT:PSS sheets thereby expanding the applications of PEDOT:PSS polymer. Most recently, Peng et al. [43] reported a large-scale fabrication method of 2D PEDOT sheets in IGBs. The monomer (3,4-ethylenedioxythiophene, EDOT), encapsulated in micelles of sodium dodecyl sulfate (SDS), was “squeezed” into IGBs and underwent oxidative polymerization with ammonium persulfate (APS). The resulting PEDOT maintains a 2D sheet structure with a thickness of ~83 nm, which exhibited higher crystallinity and better morphology than the PEDOT powders synthesized by conventional methods.

  Integrating other methods to initiate synthetic reactions within ice-confined space further broadens the scope of ice-assisted synthesis. Generally, the nucleus growth of reactant precursors within IGBs is prevented due to the limited ice lattice space, thereby benefiting the synthesis of atomic-dispersed materials. For instance, a frozen chloroplatinic acid (H₂PtCl₆) solution can be transformed into a Pt single-atom solution under ultraviolet (UV) irradiation through an ice-assisted photochemical reduction method (Fig. 2e) [44]. Upon melting, the resulting Pt (or Au, Ag) single-atom solution was further absorbed onto a series of substrate surfaces to form single-atomic catalysts, while the formed single-atom catalysts were unstable. One improvement is the ice-assisted in-situ photocatalytic reduction strategy: A mixed frozen solution of H₂PtCl₆ and photocatalysts exposed to light allows direct reduce the H₂PtCl₆ to Pt single-atoms on the substrate surface through photogenerated electrons, thus forming a robust metal-support interaction (Fig. 2f) [19]. In this process, photogenerated electrons from a frozen electron-deficient g-C₃N₄ solution capture Pt single atoms, stabilizing them within a Pt-C structure. This method enhances both the stability and functionality of the resulting single-atomic catalysts.

  2.2 Ice-templating physical synthesis

  Frozen processes can control the structure, morphology, and dimensions of functional materials [45]. The slow growth of ice crystals during the freezing process would repel solute particles or molecules into the IGBs, resulting in the assembly process within a confined space for preformed materials [3]. The directional freeze-casting method creates a confined space during the anisotropic growth of ice crystals, guiding the synthesis of hierarchical micro- and nanostructural materials (Fig. 3a) [46]. Under the unidirectional freeze-casting technique, where the freezing front extends in one direction, the carbon nanotubes (CNTs) and chitosan mixture would transform into electro-conductive foam [46]. Similarly, silica, titania, silica-alumina, titania-silica, and carbon materials with porous microfibers and micro-honeycomb structures could be prepared by the unidirectional freezing followed by freeze-drying [46]. Furthermore, the hierarchical and anisotropic silver nanowire aerogel is fabricated under a bidirectional freeze-casting technique that froze from the side to the center and from the bottom to the top (Fig. 3b) [47]. Besides, Fang et al. [48] reported that the dispersed metal-cyano colloidal precursor suspension during anisotropic-growing ice crystals can be assembled into the 2D metal-cyano nanosheets with millimeter lateral dimensions.

  Figure 3

  

  Figure 3.  (a) Unidirectional freeze-casting technique. (b) Bidirectional freeze-casting technique. Copied with permission [47]. Copyright 2019, American Chemical Society. (c) Ice-templating assisted exfoliation strategy. Reproduced with permission [50]. Copyright 2020, Wiley Publishing Group. (d) Schematic illustration of ice-aid transfer and ice-stamp process. Copied with permission [51]. Copyright 2023, Wiley Publishing Group. (e) Schematic of constructing 2D MOF nanoparticle monolayer via ice-templating-assisted self-assembly method. Copied with permission [53]. Copyright 2023, American Chemical Society. (f) Schematic illustration of ice-templating-assisted H₂-reduction method for producing diverse metallic co-catalysts. Copied with permission [20]. Copyright 2024, Wiley Publishing Group.

  DownLoad: Full-Size Img PowerPoint

  Small ice crystals between layers of material act as self-sacrificing templates to effectively regulate the interlamellar spacing (Fig. 3c). This approach proves particularly suitable for the sheet-like exfoliation of bulk materials or the structural reconstruction of 3D hierarchical materials. Zhang et al. [49] prepared the dispersion by mixing black phosphorous (BP) powder into the N-methyl-2pyrrolidone (NMP). And few-layer black phosphorous (BP) nanosheets were easily exfoliated from the frozen dispersion by ultrasonic treatment, accompanied by the melting of NMP crystals. Remarkably, the freeze-ultrasonication method has a 75% yield of a few-layer BP nanosheet, which showed a lateral size ranging from 50 nm to 3 µm with around 6 layers. Later, Zhang et al. [50] achieved the reconstruction of wet Ti₃C₂T_x/CNTs film by freezing the water molecules within Ti₃C₂T_x/CNTs interlayer into small ice granules. Subsequently, the ice granules within the Ti₃C₂T_x/CNTs interlayer acted as a self-sacrificing template and then were removed by a freeze-drying process to form a 3D porous Ti₃C₂T_x/CNTs film with wider ion transport channels. The frozen water or organic crystals can be intercalated into the material to extend the distance of the material interlayer, facilitating the generation of the material nano-sheets. Removing frozen crystals through ultrasonication or sublimation effectively exfoliates the nano-sheet structural materials

  Ice surfaces can also act as removable hard templates for transferring 2D materials from a growth substrate to a target substrate, which helps solve the potential issue of organic solvent pollution in traditional 2D materials transfer techniques [51]. For instance, in a technique involving MoS₂ flakes, the original growth substrate covered with MoS₂ flake was flipped over on the aqueous target substrate (Si/SiO₂) to form a sandwich structure [51]. After freezing, the original growth substrate can be removed, leaving the ice layer containing the MoS₂ flake on the target substrate (Si/SiO₂) (Fig. 3d). Moreover, the MoS₂ flake can be transferred between substrates via a similar ice-stamp transfer strategy. The interaction between ice surfaces and 2D sample, derived from the high-density hydrogen bonds, is stronger than the van der Waals Force between ice and original substrates. Additionally, during the ice-stamp process, some contaminants on 2D materials would remain on the ice surface due to the differences in adhesion, thereby achieving an ice-cleaning process. Overall, the transfer and cleaning of 2D materials can be realized via the controlled adhesion of ice.

  Ice-template synthesis is also an improved method for the self-assembled arrangement of nanoparticles. Compared to the evaporation-induced interfacial assembly technique [52], freezing monodispersed colloidal metal-organic framework (MOF) nanoparticles allows for their orderly self-assembly within IGBs independent of external forces to form a high quantity of 2D superstructures [53]. Song et al. [53] fabricated a 2D layered MOF with a quasi-ordered array superstructure by freezing colloidal MOF nanoparticles. Colloidal MOF nanoparticles, such as ZIF-8, UiO-66, and MIL-88, can self-assemble within IGBs to form mono- or bi-layer MOF superstructures. Subsequently, the as-prepared MOFs with superstructures via ice-templating methods are subjected to pyrolysis to produce 2D layered hollow carbon nanoparticles (Fig. 3e). Similarly, Song et al. [54] also prepared a well-performed 2D carbon polyhedron superstructure catalyst via an etching-assembly-pyrolysis procedure. First, colloidal ZIF-8 nanoparticles are etched by phytic acid to introduce phosphorous and nitrogen elements and create mesoporous structures for enhancing the intrinsic catalytic activity. Then, the treated ZIF-8 nanoparticles can be assembled into a 2D polyhedron superstructure via an ice-templating self-assembly strategy. The ZIF-8 superstructure precursor transformed into N,P co-doped 2D metal-free carbon polyhedron array superstructure via further pyrolysis, which exhibits ordered arrangement, engineered porous structure, and high activity for oxygen reduction reaction (ORR).

  Additionally, ice-templating physical pretreatment can be effectively coupled with other post-treat measures, such as reduction and pyrolysis, to prevent the aggregation of single atoms. Generally, conventional photodeposition [55] and impregnation [56] methods for synthesizing metal-photocatalyst substrate composites often suffer from the issue of inevitable wasting of metal cation precursors in bulk solution. To address this, an ice-templating-assisted H₂ reduction method was developed. This approach involves blending metal cation precursors (Mⁿ⁺, where M represents Ni, Ru, Pt) with photocatalyst substrates (e.g., Ga doped-La₅Ti₂Cu_0.9Ag_0.1O₇S₅, TiO₂, CdS, and g-C₃N₄), followed by ultrasonic dispersion, freezing, freeze-drying, and calcination (in a mixed Ar/H₂ atmosphere), resulting in well-dispersed atomic cluster co-catalysts (Fig. 3f) [20]. Mⁿ⁺ confined within IGBs can be homogeneously localized on the substrate surface, effectively avoiding the wasting of metal precursor in bulk solution and achieving high atomic efficiency.

  2.3 Ice-melting chemical synthesis

  Reaction energy barriers and nucleation processes seriously control synthetic reactions of materials. Generally, conventional liquid-phase synthesis methods, despite improvements such as surfactant-mediated approaches and microfluidic engineering, often lead to fast nucleation kinetics that adversely affect crystal structure uniformity [57,58]. In contrast, the melting of frozen precursor solutions supports a controlled release of reactants, offering a new way to prepare materials with a uniform size (Fig. 4a). A typical example of the ice-melted strategy has been reported by Wei et al. (Fig. 4b) [59], where atomically dispersed silver (Ag) was prepared via ultra-slowly releasing Ag⁺ from frozen AgNO₃ solution into a reductive NaBH₄ solution at ~0 ℃. The size of ice-melting-synthesized Ag nanoparticles did not exceed 0.5 nm, significantly smaller than those produced in conventional liquid-phase processes [60]. This methodology is subsequently extended to synthesize atomically dispersed metals, including cobalt, nickel, copper, rhodium, ruthenium, palladium, osmium, iridium, platinum, and gold (Fig. 4c). Ice-melted processes can also synthesize organic semiconductor nanostructures (OSNs). During natural melting processes, frozen organic molecules gradually release into a sodium dodecyl sulfate solution, forming OSNs with crystalline structures instead of amorphous structures [61]. Likewise, various composite catalysts, including Cu/TiO₂ and Ce-Cu/TiO₂, prepared by the ice-melted method, show small crystalline sizes, enhanced particle uniformity, and well-catalytic performances compared to conventional methods [62,63]. In conclusion, the easy-handled ice-melted process introduces an innovative approach to redesigning conventional synthesis, including co-precipitation, hydrolysis, and displacement reactions.

  Figure 4

  

  Figure 4.  (a) Schematic of the conventional liquid-phase synthesis method. (b) Schematic of ice-melted synthesis method for atomically dispersed metals. (c) Periodic table of the elements and STEM images for atomically dispersed metals using the ice-melted method. (a-c) Reproduced with permission [59]. Copyright 2018, Wiley Publishing Group.

  DownLoad: Full-Size Img PowerPoint

  3. Potential environmental applications of ice-assisted functional materials

  3.1 Membrane-based separation

  The membrane separation technology has been extensively applied in hydrocarbon separation, seawater desalination, hydrogen generation from seawater, and trace pollutants removal [64–67]. Membrane materials are the core of membrane applications, and their thickness and porous structure are closely related to permeability and selectivity in engineering [65,68]. Ice chemistry provides unique advantages, such as a unique reaction interface, confined space, and solid-liquid phase transition, for preparing polymer or inorganic membrane materials.

  The pre-melted QLLs on ice surfaces can alter reaction kinetics to facilitate the synthesis of dimension-controlled and hierarchical porous membrane materials [25,59]. Additionally, the growth of ice dendrites and the controllable solid-liquid transition further regulate the porous structure, benefiting the formation of 3D consecutive channels. As early as 1970, Miller [69] reported an “ice sandwiched” semipermeable membrane, where solution, ice, and pure water were contained in left, middle, and right chambers with the common walls being rigid filters, for producing potable water from brackish water (Fig. S3 in Supporting information). Although this early membrane device did not use ice-assisted membrane materials, the ice in the middle chamber played a crucial role in hindering solute transport during the freezing process and producing purified water during the thawing process.

  Ice-assisted polymer membrane materials have recently been developed and shown significant application potential. Among these, the polyamide membrane materials, notably utilized in reverse osmosis and nanofiltration membranes for seawater desalination, are generally prepared by the interface polymerization (IP) method. Conventional IP method for preparing polyamide membrane is achieved at a liquid-liquid interface, where aqueous amine reacts with dissolved acyl chloride in an organic solvent [70]. However, the phase transition from ice to water at the ice/organic interface facilitated the formation of 3D-quasilayered PA nanofiltration membranes (IC-PANF) [59]. Compared to the conventional PA nanofiltration membranes, the IC-PANF membrane synthesized by Zhang et al. [25] exhibited a ~4-fold increase in water permeance and ~6- to 10-fold enhancement in Cl^-/SO₄^2- separation. Additionally, IC-PANF shows efficient salt rejection (around 42.5%−99.6% for various salts) associated with highly ionization of surface R-COOH and R-NH₃ compared to the conventional PANF membrane (Fig. 5a). Meanwhile, it provides a fast water channel based on the inter-connected, 3D-quasilayered, and high ionization structure and can further be used to separate the co-ion Cl^-/SO₄^2- for resource recovery (Fig. 5b). Except for the PA membrane, many 2D materials, such as PNAI sheets [38], graphene oxide (GO) [71], metal-organic frameworks (MOFs) [53], and MXene [72], have been fabricated via ice chemistry to achieve high porosity or layered structures, which could be anticipated with promising development prospects in separated membranes.

  Figure 5

  

  Figure 5.  (a) Schematic of the high-density ionized internal behavior of IC-PANF. (b) The water performance, co-ion Cl^-/SO₄^2- selectivity, and NaSO₄ rejection of IC-PANF membrane. (a, b) Copied with permission [25]. Copyright 2023, American Association for the Advanced of Science. (c) Schematic of the ice-assisted synthesis of porous 2D PANI nanosheets. (d) Rejection rate of the PTFE membrane and pPN membrane for SARS-Cov-2. (c, d) Copied with permission [73]. Copyright 2022, the Royal Society of Chemistry. (e) Schematic illustration of the fabrication of RC@PVA membranes. (f) The performance of RC@PVA membrane. (e, f) Copied with permission [74]. Copyright 2024, Elsevier.

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. 5c, Park’s group [73] prepares a 2D porous conducting polymer membrane by combining ice-assisted and particle-templated methods for virus sterilization. The 2D porous PANI sheets were grown on the surface of ice-containing polystyrene (PS) particles, with the porosity adjustable by varying the PS particles size. Then, the PANI can be further integrated into the commercial hydrophilic polytetrafluoroethylene (PTFE) membrane to form the nanofiltration membrane (pPN). The pPN could successfully capture the negatively charged SARS-Cov-2 virus due to its electrophoretic role. The pPN with a pore size of ~100 nm (pPN-100) achieves a rejection rate of >90% SARS-Cov-2 and exhibits high flux across a wide pressure range, superior to the efficiency of the supporting PTFE membrane (<70% rejection rate) (Fig. 5d). Moreover, during the virus filtration process, the color of pPN-100 transitioned from green to blue, then reverted to green again after washing with water, indicating its potential as a real-time sensing capability for virus adsorption.

  Recently, a supramolecular cellulose membrane (RC@PVA) with a dual-network and controllable 3D interconnected porous structure has been developed based on a “freezing-dissolution-regeneration” strategy [74]. The cellulose and polyvinyl alcohol (PVA) blending solution was frozen in the model to form an ice template. Then, the ice template was dipped in a regeneration bath (tannic acid and acetic acid mixture solution) to initiate the self-assembly of the reconstruction of cellulose. The solvent and non-solvent are exchanged during the ice template dissolving process (Fig. 5e). The as-prepared RC@PVA membrane exhibits good performance for purifying different oil/water mixture (filter flux values of 840.76 L m^-2 h^-1 for lubricating oil, 802.55 L m^-2 h^-1 for diesel oil, 764.33 L m^-2 h^-1 for sunflower oil, 878.98 L m^-2 h^-1 for cyclohexane oil, and 898.10 L m^-2 h^-1 for petroleum oil), emulsified oil, and methylene blue/oil/water mixture due to its superhydrophilic and underwater superhydrophobic properties (Fig. 5f). Furthermore, the introduction of supramolecular fiber enhanced the mechanical strength, which improved the reusability and durability (~99.0% purification efficiency of the lubricating-in-water emulsion after 10 separation cycles) of the RC@PTA membrane. The “ice-dissolution-regeneration” process contributed to the regeneration of supramolecular fiber to construct 3D interconnected porous structure.

  3.2 Environmental catalysis

  Ice-assisted synthesis strategies provide relatively mild preparation and cost-effective conditions for preparing environmental functional catalysts. Given the diversity of ice templating, variations in ice usage, as well as the combination of ice-assisted chemistry with new synthetic methodologies, the environmental functional catalyst based on metal or metal-free composites can be effectively constructed, promoting their efficiency, stability, and selectivity for environmental applications (Fig. S4 in Supporting information).

  Metal oxides based on ice-assisted synthesis show environmental catalytic potential. For instance, tungsten oxide (WO₃) nanosheets produced at the ice-organic solvent interface resulted in a thickness of <10 nm and a large lateral size [75]. These WO₃ nanosheets showed 100% As(Ⅲ) oxidation (~400 µmol/L) within 60 min by coupling with H₂O₂, outperforming both commercial WO₃ and WO₃ nanoparticles. The superior catalytic performance of ice-assisted synthetic WO₃ was probably ascribed to its nanosheet-like structure. Additionally, an ordered porous TiO₂ film derived from the removal of ice crystal was used in dye-sensitized solar cells, which showed higher photo-current density and lower loading density of dye compared to those produced by the conventional doctor blading method [76]. Furthermore, ice-interfacial synthesized ultra-thin TiO₂ has been developed as an anode in Li-ion batteries [28], and these TiO₂ nanosheets also show potential as photocatalyst or photocatalytic substrates for pollutants degradation, owing to their high surface area and ultra-thin structure [77].

  Besides, metal-free materials based on ice-assisted synthesis also serve as environmental functional catalysts. 2D polyarylamines (2DPA) are synthesized at the interface of the ice phase and toluene solution through C—N coupling [78]. The polymerization of 2DPA at the ice interface influences the distribution and valence of N-dopant, enhancing catalytic activity for water splitting during electrochemical hydrogen evolution reaction (HER). The BP nanosheets, synthesized by ice-assisted exfoliation, can be coupled with graphitic carbon nitride (g-C₃N₄) to form a 2D/2D heterostructure photocatalyst [49]. The resulting BP/g-C₃N₄ exhibited an excellent photocatalytic H₂ production rate of 384.17 µmol g^-1 h^-1 under λ > 420 nm light irradiation, superior to that of many precious metal-loaded photocatalysts via liquid-phase synthesis.

  Composites derived from ice-assisted assembly or synthesis have promising applications in the field of environmental catalysis. 3D macroporous TiO₂-chitosan scaffolds, prepared using the unidirectional freezing method, have demonstrated their role as reusable substrates for photocatalytic degradation of dyes [79]. Similarly, a highly aligned macroporous TiO₂/chitosan/rGO composite fabricated by the freeze casting method is well-suited for photocatalytic substrate, owing to its adjustable pore sizes and improved mechanical strength [80]. Besides, metals (Pt, Pd, and Ag)/3D-graphene nanocomposites produced by the freeze casting methods have been employed in catalyzing the reduction of 4-nitrophenol and methylene blue [81]. Additionally, the Ce-Cu/TiO₂ catalyst synthesized by the ice-melted method (i.e., Ce-Cu/TiO₂-ice) exhibited better structure and performance compared to its conventionally synthesized counterpart (i.e., Ce-Cu/TiO₂-con). Specifically, the Ce-Cu/TiO₂-ice possessed a large surface area (67.88 m²/g vs. 53.03 m²/g), abundant active sites (Cu²⁺/(Cu⁺ + Cu²⁺) ratio, 49.01% vs. 38.22%), small particle size (82.97 nm vs. 154.80 nm), and increased surface acidity, corresponding to 80% NO_x conversion at 250–375 ℃ (exceeding that of Ce-Cu/TiO₂-con by 20%) and nearly 100% N₂ selectivity [63]. Metal single-atom catalysts can also be developed using the ice-assisted method. For instance, a platinum single atom-based electrocatalyst was synthesized by irradiating frozen precursor solution, achieving excellent HER performance [82]. A photocatalytic substrate (electron-deficient g-C₃N₄) could be photoexcited within the ice, inducing the deposition of Pt single atoms (PtSA). The PtSA via ice-assisted in-situ photocatalytic method showed enhanced performances in H₂-evolution, attributing to its high-density of Pt-loading (0.35 mg/m²) [19]. Recently, atomic clusters co-catalysts (e.g., Ni, Rh, Ru, and Pt) were prepared on various photocatalytic substrates (e.g., TiO₂, CdS, and g-C₃N₄) by combing freeze and Ar/H₂ reduction [20]. The resulting Ru single atoms and clusters cocatalyst exhibited excellent photocatalytic H₂ evolution (578 µmol/h) under visible light irradiation.

  3.3 Pollutants adsorption

  Ice-assisted synthetic methods allow fabricating materials with tailored porosity content, serving as adsorbents for removing specific pollutants. Generally, the porous structure yields a larger specific surface area and abundant contact sites, enhancing the interaction with pollutants. The formation of a loose porous structure within the adsorbent can occur during the ice crystal growth, ice template removal, and solid-liquid transformation processes [25]. In addition, the adsorption capacity also depends on the functional groups of adsorbent, which offers interaction sites for pollutant anchoring [84]. Ice-assisted methods also facilitate the decoration of one material onto another to introduce additional functional groups [26]. Hence, ice-assisted strategies are suitable for developing novel adsorbents for environmental remediation (Fig. 6a).

  Figure 6

  

  Figure 6.  (a) The relationship between ice chemistry engineering and adsorbents. (b) The ice-assisted synthesis mechanism of GO aerogel. (c) The adsorption capacity of Cd²⁺, Ni²⁺, phenol, and bisphenol A by GO aerogel. (b, c) Reproduced with permission [17]. Copyright 2021, Elsevier. (d) Schematic of chitosan-CNT composite by freeze casting method. (e) The adsorption capacity of bilirubin by chitosan-CNT composite. (d, e) Reproduced with permission [83], Copyright 2015, Wiley Publishing Group.

  DownLoad: Full-Size Img PowerPoint

  The synthesis of hierarchically 3D porous absorbents via ice chemistry engineering has been reported. Pure 3D graphene oxide (GO) aerogels are constructed via the freezing and freeze-drying processes, where frozen water molecules between the 2D GO sheets occupy the partial material space, subsequently leading to the creation of a porous structure during freeze-drying through ice crystal sublimation. Huong et al. [17] have reported that ice-assisted synthetic GO aerogel exhibited high adsorption capacity for heavy metal ions (Cd²⁺, Ni²⁺) and phenolic compounds (phenol and bisphenol A), ascribed to the formation of 3D porous structure during the ice crystals growth (Figs. 6b and c). Due to its large contact area and short diffusion pathway, GO aerogel has also been widely applied in oil and water separation, CO₂ adsorption, and dye adsorption [85–87]. Notably, the introduction of other materials into pure GO material provides effective functional groups (such as hydroxyl groups and amino groups) and additional pores. Various 3D assembly hybrid materials, such as GO/chitosan, carbon nanotube/chitosan, and GO/kaolinite/poly(vinyl alcohol), synthesized by freeze-casting technique, efficiently adsorb Pb²⁺, Hg²⁺, bilirubin, and ciprofloxacin (Figs. 6d and e) [18,83,88]. Furthermore, a composite material derived from the frozen solution of MOFs and a cross-linking agent is also reported as an effective adsorbent due to its highly exposed surface area and enhanced mass transport [89,90]. These findings indicate that ice chemistry can guide the synthesis of the adsorbents for removing specific pollutants.

  4. Summary

  This review describes the ice-assisted synthetic strategies of functional materials and their potential environmental applications in recent years. As either reactants or nanoreactors, the ice crystals or freezing processes have displayed controllable reaction kinetics and thermodynamics by providing interface or confined space environments in synthesizing functional materials. Ice-assisted synthetic strategies are mainly categorized into three approaches: Ice-confined interfacial, ice-templated physical, and ice-melted chemical synthesis. Their potential environmental applications (including but not limited to membrane-based separation, environmental catalysis, and pollutant adsorption) are also illustrated in detail by discussing, comparing, and summarizing most examples. Despite the fruitful history of ice chemistry material, many challenges and opportunities should be emphasized.

  (1) Firstly, most works have reported the ice-assisted synthesis methods of materials and related applications in environmental engineering. It is necessary to reveal the kinetics and thermodynamic mechanisms of ice chemistry in synthesizing materials by combining in-situ characterization and molecular dynamics simulations, providing deeper insights into the synthetic reaction occurring on ice surface, between ice crystals, or during the melting of ice.

  (2) Then, more advanced materials fabrication methods, including but not limited to hydrothermal, pyrolysis, laser, and 3D printing, should be integrated to enhance the versatility and applicability of ice-assisted synthesis.

  (3) Additionally, the effects of key freezing parameters (e.g., freezing direction/temperature/rate, pH of the frozen solution, and temperature gradient), interfacial environmental conditions, and ice-melting rates, on the ice-assisted synthetic process should be further investigated to improve the reproducibility and uniformity of as-prepared functional materials.

  (4) Last but not least, more experimental studies about the application potentials and performances of ice-assisted synthetic materials under an authentic environmental matrix should be developed to drive the practical (or industrial) tests of ice-assisted synthesis in environmental engineering.

  CRediT authorship contribution statement

  Yaohua Wu: Writing – review & editing, Writing – original draft, Investigation. Yihong Chen: Visualization, Investigation, Formal analysis. Juanshan Du: Supervision, Project administration, Methodology, Investigation, Conceptualization. Huazhe Wang: Validation, Software, Resources, Investigation. Chuchu Chen: Writing – review & editing, Visualization. Wenrui Jia: Visualization, Investigation. Yongqi Liang: Methodology, Investigation. Qinglian Wu: Data curation. Wan-Qian Guo: Validation, Project administration, Funding acquisition, Conceptualization.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by National Natural Science Foundation of China (Nos. 52170030, 52200049), China Postdoctoral Science Foundation (No. 2022TQ0089), State Key Laboratory of Urban-rural Water Resource and Environment (Harbin Institute of Technology) (No. 2024TS28), Young Scientist Studio of Harbin Institute of Technology, and Fundamental Research Funds for the Central Universities.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.110944.

  
  
• 1. [1]

     T. Bartels-Rausch, Nature 494 (2013) 27–29. doi: 10.1038/494027a

  2. [2]

     J.S. Du, K. Kim, D.W. Min, et al., Environ. Sci. Technol. 56 (2022) 1007–1016. doi: 10.1021/acs.est.1c06747

  3. [3]

     R. O’Concubhair, J.R. Sodeau, Acc. Chem. Res. 46 (2013) 2716–2724. doi: 10.1021/ar400114e

  4. [4]

     J. Du, Y. Hu, K. Kim, et al., Environ. Sci. Technol. 57 (2023) 20272–20281. doi: 10.1021/acs.est.3c07278

  5. [5]

     J. Du, K. Kim, S. Son, et al., Environ. Sci. Technol. 57 (2023) 5317–5326. doi: 10.1021/acs.est.3c00604

  6. [6]

     Y. Wu, H. Wang, J. Du, et al., Environ. Sci. Technol. 58 (2024) 17948–17958. doi: 10.1021/acs.est.4c07493

  7. [7]

     T. Kling, F. Kling, D. Donadio, J. Phys. Chem. C 122 (2018) 24780–24787. doi: 10.1021/acs.jpcc.8b07724

  8. [8]

     L.A. Wilen, J.S. Wettlaufer, M. Elbaum, et al., Phys. Rev. B 52 (1995) 12426–12433.

  9. [9]

     J.S. Yeon, N. Gupta, P. Bhattacharya, et al., Adv. Funct. Mater. 32 (2022) 2112509.

  10. [10]

      K. Kim, M.J. Park, Nanoscale 12 (2020) 14320–14338. doi: 10.1039/d0nr02624g

  11. [11]

      D. Barpuzary, K. Kim, M.J. Park, J. Polym. Sci., Part B: Polym. Phys. 57 (2019) 1169–1176. doi: 10.1002/polb.24777

  12. [12]

      S.W. Wu, C.Q. Zhu, Z.Y. He, et al., Nat. Commun. 8 (2017) 15154.

  13. [13]

      E.G. Hammerschmidt, Ind. Eng. Chem. 26 (1934) 851–855. doi: 10.1021/ie50296a010

  14. [14]

      W. Mahler, M.F. Bechtold, Nature 285 (1980) 27–28. doi: 10.1038/285027a0

  15. [15]

      H.M. Tong, I. Noda, C.C. Gryte, Colloid. Polym. Sci. 262 (1984) 589–595.

  16. [16]

      E. Munch, M.E. Launey, D.H. Alsem, et al., Science 322 (2008) 1516–1520. doi: 10.1126/science.1164865

  17. [17]

      L. Huong, D.B. Thinh, T.H. Tu, et al., Surf. Interfaces 26 (2021) 101309.

  18. [18]

      X.H. Huang, J. Tian, Y.W. Li, et al., Langmuir 36 (2020) 10895–10904. doi: 10.1021/acs.langmuir.0c00654

  19. [19]

      P. Zhou, F. Lv, N. Li, et al., Nano Energy 56 (2019) 127–137.

  20. [20]

      H. Wang, F. Wang, S. Zhang, et al., Adv. Mater. 36 (2024) 2400764.

  21. [21]

      G. Zheng, X. Kang, H. Ye, et al., Chin. Chem. Lett. 35 (2024) 108817.

  22. [22]

      N. Song, S.Y. Ren, Y. Zhang, et al., Adv. Funct. Mater. 32 (2022) 2204751.

  23. [23]

      Z.G. Zhu, L.L. Zhong, Y. Wang, et al., Chin. Chem. Lett. 31 (2020) 2619–2622.

  24. [24]

      Z.X. Wei, B. Ding, H. Dou, et al., Chin. Chem. Lett. 30 (2019) 2110–2122.

  25. [25]

      Y.Q. Zhang, H. Wang, J. Guo, et al., Science 382 (2023) 202–206.

  26. [26]

      K.C. Lai, L.Y. Lee, B.Y.Z. Hiew, et al., J. Environ. Sci. 79 (2019) 174–199.

  27. [27]

      H. Kang, Acc. Chem. Res. 38 (2005) 893–900. doi: 10.1021/ar0501471

  28. [28]

      G. Kim, C. Jo, W. Kim, et al., Energy Environ. Sci. 6 (2013) 2932–2938. doi: 10.1039/c3ee41880d

  29. [29]

      G. Colón, M.C. Hidalgo, J.A. Navıo, Catal. Today 76 (2002) 91–101. ´

  30. [30]

      Y. Li, X. Sun, H. Li, et al., Powder Technol. 194 (2009) 149–152.

  31. [31]

      M. Hussain, R. Ceccarelli, D.L. Marchisio, et al., Chem. Eng. J. 157 (2010) 45–51.

  32. [32]

      B. Tyagi, K. Sidhpuria, B. Shaik, et al., Ind. Eng. Chem. Res. 45 (2006) 8643–8650. doi: 10.1021/ie060519p

  33. [33]

      G.J. Owens, R.K. Singh, F. Foroutan, et al., Prog. Mater Sci. 77 (2016) 1–79.

  34. [34]

      D. Bokov, A.T. Jalil, S. Chupradit, et al., Adv. Mater. Sci. Eng. 2021 (2021) 5102014.

  35. [35]

      J.N. Hasnidawani, H.N. Azlina, H. Norita, et al., Procedia Chem. 19 (2016) 211–216.

  36. [36]

      E.N. Konyushenko, J. Stejskal, M. Trchova, et al., Synth. Met. 158 (2008) 927–933.

  37. [37]

      H.Y. Ma, Y. Gao, Y.H. Li, et al., J. Phys. Chem. C 113 (2009) 9047–9052. doi: 10.1021/jp8112683

  38. [38]

      I.Y. Choi, J. Lee, H. Ahn, et al., Angew. Chem. Int. Ed. 54 (2015) 10497–10501. doi: 10.1002/anie.201503332

  39. [39]

      K. Kim, H. Ahn, M.J. Park, ACS Appl. Mater. Interfaces 9 (2017) 30278–30282. doi: 10.1021/acsami.7b10821

  40. [40]

      K. Kim, B. Kim, K. Kim, et al., Macromol. Rapid Commun. 42 (2021) 2100565.

  41. [41]

      D. Barpuzary, K. Kim, M.J. Park, ACS Nano 13 (2019) 3953–3963. doi: 10.1021/acsnano.8b07294

  42. [42]

      J. Zhang, X. Fan, X. Meng, et al., ACS Nano 15 (2021) 8870–8882. doi: 10.1021/acsnano.1c01459

  43. [43]

      X.M. Peng, J. Zhang, J. Zhou, et al., Angew. Chem. Int. Ed. 62 (2023) e202301940.

  44. [44]

      H. Wei, K. Huang, D. Wang, et al., Nat. Commun. 8 (2017) 1490.

  45. [45]

      S. Deville, Scr. Mater. 147 (2018) 119–124.

  46. [46]

      W. Tanthapanichakoon, H. Tamon, K. Nakagawa, et al., Eng. J. 17 (2013) 1–8. doi: 10.4186/ej.2013.17.3.1

  47. [47]

      L. Zhang, X. Liu, A. Deb, et al., ACS Sustainable Chem. Eng. 7 (2019) 19910–19917. doi: 10.1021/acssuschemeng.9b05413

  48. [48]

      Z.W. Fang, S.S. Tang, Z.Q. Wang, et al., ACS Cent. Sci. 8 (2022) 627–635. doi: 10.1021/acscentsci.2c00252

  49. [49]

      Q.Z. Zhang, S.Y. Huang, J.J. Deng, et al., Adv. Funct. Mater. 29 (2019) 1902486.

  50. [50]

      P. Zhang, Q.Z. Zhu, R.A. Soomro, et al., Adv. Funct. Mater. 30 (2020) 2000922.

  51. [51]

      H. Liu, Q.H. Thi, P. Man, et al., Adv. Mater. 35 (2023) 2210503.

  52. [52]

      G. Cossio, E.T. Yu, Nano Lett. 20 (2020) 5090–5096. doi: 10.1021/acs.nanolett.0c01277

  53. [53]

      Y. Song, X. Song, X. Wang, et al., J. Am. Chem. Soc. 144 (2022) 17457–17467. doi: 10.1021/jacs.2c06109

  54. [54]

      X. Song, Y. Song, X. Li, et al., Small 20 (2024) 2305459.

  55. [55]

      K. Wenderich, G. Mul, Chem. Rev. 116 (2016) 14587–14619. doi: 10.1021/acs.chemrev.6b00327

  56. [56]

      S. Yang, J. Kim, Y.J. Tak, et al., Angew. Chem. Int. Ed. 55 (2016) 2058–2062. doi: 10.1002/anie.201509241

  57. [57]

      Y.L. Xiao, X.C. Fang, W. Deng, et al., Adv. Mater. Interfaces 7 (2020) 1901753.

  58. [58]

      W. Deng, J.S. Jie, X.Z. Xu, et al., Adv. Mater. 32 (2020) 1908340.

  59. [59]

      H.H. Wei, K. Huang, L. Zhang, et al., Angew. Chem. Int. Ed. 57 (2018) 3354–3359. doi: 10.1002/anie.201711128

  60. [60]

      B. He, J.J. Tan, K.Y. Liew, et al., J. Mol. Catal. A: Chem. 221 (2004) 121–126.

  61. [61]

      Y.Z. Chen, D.J. Jiang, Y.J. Zhang, et al., J. Mater. Chem. A 8 (2020) 25275–25282. doi: 10.1039/d0ta09502h

  62. [62]

      X.M. Wang, X.B. Chen, L.M. Ye, et al., Mol. Catal. 497 (2020) 111225.

  63. [63]

      C. Shuang, Z. Haixiang, W. Xueman, et al., J. Chem. Technol. Biotechnol. 98 (2023) 506–515. doi: 10.1002/jctb.7266

  64. [64]

      M.R. Chowdhury, J. Steffes, B.D. Huey, et al., Science 361 (2018) 682–685. doi: 10.1126/science.aar2122

  65. [65]

      S. Li, R. Dong, V.E. Musteata, et al., Science 377 (2022) 1555–1561. doi: 10.1126/science.abq0598

  66. [66]

      H. Xie, Z. Zhao, T. Liu, et al., Nature 612 (2022) 673–678. doi: 10.1038/s41586-022-05379-5

  67. [67]

      S.P. Dharupaneedi, S.K. Nataraj, M. Nadagouda, et al., Sep. Purif. Technol. 210 (2019) 850–866.

  68. [68]

      H.B. Park, J. Kamcev, L.M. Robeson, et al., Science 356 (2017) eaab0530.

  69. [69]

      R.D. Miller, Science 169 (1970) 584–585. doi: 10.1126/science.169.3945.584

  70. [70]

      G. Zhao, H. Gao, Z. Qu, et al., Nat. Commun. 14 (2023) 7624. doi: 10.1021/acsnano.3c00074

  71. [71]

      S.N. Schiffres, S. Harish, S. Maruyama, et al., ACS Nano 7 (2013) 11183–11189. doi: 10.1021/nn404935m

  72. [72]

      J. Wang, D. Jiang, M. Zhang, et al., J. Mater. Chem. A 11 (2023) 1419–1429. doi: 10.1039/d2ta08176h

  73. [73]

      K. Kim, J. Min, M. Lee, et al., Nanoscale 14 (2022) 17157–17162. doi: 10.1039/d2nr04479j

  74. [74]

      M. Xia, J. Gao, P. Xu, et al., Chem. Eng. J. 485 (2024) 150023.

  75. [75]

      A.Q.K. Nguyen, K. Kim, Y.Y. Ahn, et al., Sci. Total Environ. 865 (2023) 161104.

  76. [76]

      X.X. Wang, B.C. Yang, K.C. Zhou, et al., Curr. Appl. Phys. 15 (2015) 662–668.

  77. [77]

      K. Nakata, A. Fujishima, J. Photochem. Photobiol. C 13 (2012) 169–189.

  78. [78]

      M. Cheng, M. Li, Chem. Eur. J. 25 (2019) 13860–13864. doi: 10.1002/chem.201902587

  79. [79]

      C. Suwanchawalit, A.J. Patil, R.K. Kumar, et al., J. Mater. Chem. 19 (2009) 8478–8483. doi: 10.1039/b912698h

  80. [80]

      C. Chen, Y. Zhang, J. Zeng, et al., Appl. Surf. Sci. 424 (2017) 170–176.

  81. [81]

      P. Kumar Sahoo, B. Panigrahy, D. Thakur, et al., New J. Chem. 41 (2017) 7861–7869.

  82. [82]

      Z.Y. Sun, Y.Q. Yang, C.H. Fang, et al., Small 18 (2022) 2203422.

  83. [83]

      A. Ouyang, Q.M. Gong, J. Liang, Adv. Eng. Mater. 17 (2015) 460–466. doi: 10.1002/adem.201400250

  84. [84]

      C.H. Tessmer, R.D. Vidic, L.J. Uranowski, Environ. Sci. Technol. 31 (1997) 1872–1878.

  85. [85]

      Y.B. Cheng, P. Xu, W. Zeng, et al., J. Environ. Chem. Eng. 5 (2017) 1957–1963.

  86. [86]

      J.J. Wu, X.Y. Qiu, S.X. Chen, J. Mater. Sci. 57 (2022) 1727–1737. doi: 10.1007/s10853-021-06768-3

  87. [87]

      S. Tang, D.S. Xia, Y. Yao, et al., J. Colloid Interface Sci. 554 (2019) 682–691.

  88. [88]

      Y.Q. He, N.N. Zhang, X.D. Wang, Chin. Chem. Lett. 22 (2011) 859–862.

  89. [89]

      Q.S. Fu, L. Zhang, H.F. Zhang, et al., Environ. Res. 186 (2020) 109608.

  90. [90]

      L. Wen, X.D. Chen, C. Chen, et al., Arabian J. Chem. 13 (2020) 5669–5678. doi: 10.1016/j.arabjc.2020.04.007

• 

  Figure 1  Schematic of ice-assisted synthetic strategies and potential environmental applications.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Freezing-improved sol-gel methods for the synthesis of 2D metal oxides. Reproduced with permission [28]. Copyright 2013, the Royal Society of Chemistry. (b) The synthesis procedures of 2D PANI on bulk ice surfaces. Reproduced with permission [38]. Copyright 2015, Wiley Publishing Group. (c) Schematic of the IC-PANF membrane at the ice/n-hexane interface. Copied with permission [25]. Copyright 2023, American Association for the Advanced of Science. (d) 2D materials synthesis within ice grain boundaries (IGBs). Reproduced with permission [43]. Copyright 2023, Wiley Publishing Group. (e) Schematic diagram of ice-assisted photochemical reduction method. Reproduced with permission [44]. Copyright 2017, Springer Nature. (f) Schematic diagram of ice-assisted photocatalytic synthesis method. Reproduced with permission [19]. Copyright 2019, Elsevier.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Unidirectional freeze-casting technique. (b) Bidirectional freeze-casting technique. Copied with permission [47]. Copyright 2019, American Chemical Society. (c) Ice-templating assisted exfoliation strategy. Reproduced with permission [50]. Copyright 2020, Wiley Publishing Group. (d) Schematic illustration of ice-aid transfer and ice-stamp process. Copied with permission [51]. Copyright 2023, Wiley Publishing Group. (e) Schematic of constructing 2D MOF nanoparticle monolayer via ice-templating-assisted self-assembly method. Copied with permission [53]. Copyright 2023, American Chemical Society. (f) Schematic illustration of ice-templating-assisted H₂-reduction method for producing diverse metallic co-catalysts. Copied with permission [20]. Copyright 2024, Wiley Publishing Group.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Schematic of the conventional liquid-phase synthesis method. (b) Schematic of ice-melted synthesis method for atomically dispersed metals. (c) Periodic table of the elements and STEM images for atomically dispersed metals using the ice-melted method. (a-c) Reproduced with permission [59]. Copyright 2018, Wiley Publishing Group.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Schematic of the high-density ionized internal behavior of IC-PANF. (b) The water performance, co-ion Cl^-/SO₄^2- selectivity, and NaSO₄ rejection of IC-PANF membrane. (a, b) Copied with permission [25]. Copyright 2023, American Association for the Advanced of Science. (c) Schematic of the ice-assisted synthesis of porous 2D PANI nanosheets. (d) Rejection rate of the PTFE membrane and pPN membrane for SARS-Cov-2. (c, d) Copied with permission [73]. Copyright 2022, the Royal Society of Chemistry. (e) Schematic illustration of the fabrication of RC@PVA membranes. (f) The performance of RC@PVA membrane. (e, f) Copied with permission [74]. Copyright 2024, Elsevier.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) The relationship between ice chemistry engineering and adsorbents. (b) The ice-assisted synthesis mechanism of GO aerogel. (c) The adsorption capacity of Cd²⁺, Ni²⁺, phenol, and bisphenol A by GO aerogel. (b, c) Reproduced with permission [17]. Copyright 2021, Elsevier. (d) Schematic of chitosan-CNT composite by freeze casting method. (e) The adsorption capacity of bilirubin by chitosan-CNT composite. (d, e) Reproduced with permission [83], Copyright 2015, Wiley Publishing Group.

  下载: 全尺寸图片 幻灯片
•