State Key Laboratory of Optoelectronic Materials and Technologies， Nanotechnology Research Center， School of Materials Science and Engineering， Sun Yat-sen University， Guangzhou 510275， China
Confined synthesis is a template-based method that precisely guides and controls the directional assembly or reaction of precursor molecules within the template cavities by virtue of the specific structures and physicochemical properties of hard templates. This approach enables accurate structural regulation of target products， expands the structural boundaries of synthesizable materials， and holds significant research value and application prospects in the field of one-dimensional carbon material synthesis. This review focuses on three typical hard templates including carbon nanotubes， boron nitride nanotubes and metal-organic frameworks， and systematically elaborates the technical routes and research progress of confined synthesis of one-dimensional carbon materials such as graphene nanoribbons， carbon chains， carbon nanotubes and polythiophene using these templates. In addition， the similarities and differences between confined synthesis and on-surface synthesis in the preparation of one-dimensional carbon materials are compared and analyzed， and the potential of molecular sieves as new templates for the confined synthesis of one-dimensional carbon materials is discussed. Finally， the advantages and existing problems of confined synthesis technology in the preparation of one-dimensional carbon materials are summarized， and the future research directions of this field are prospected， aiming to provide references for relevant research.
Contents
1 Introduction
2 Confined synthesis using carbon nanotubes
2.1 Carbon nanotube-confined synthesis of graphene nanoribbons
2.2 Confined synthesis of small-diameter carbon nanotubes in carbon nanotubes
2.3 Carbon nanotube-confined synthesis of carbon chains
2.4 Carbon nanotube-confined synthesis of carbides
3 Confined synthesis using boron nitride nanotubes
4 Confined synthesis using metal-organic frameworks
4.1 Metal-organic framework-confined synthesis of graphene nanoribbons
4.2 Metal-organic framework-confined synthesis of polyacene
4.3 Metal-organic framework-confined synthesis of one-dimensional polymers
Carbon atoms have 4 valence electrons and can form sp1, sp2, sp3 three types of hybrid orbitals. Different hybrid orbitals endow carbon materials with diverse microstructures, such as: sp3 hybridization forming a three-dimensional diamond structure, giving it ultra-high hardness; sp2 hybridization constitutes the two-dimensional planar structure of graphene, endowing it with excellent flexibility, ultra-high carrier mobility, and high thermal conductivity; sp1 hybridization forms the one-dimensional linear structure of carbon chains (Carbon chains, CC, also known as Carbyne), which is theoretically the carbon material with the greatest mechanical strength.
The research dimensions of carbon materials have gradually expanded from three-dimensional to two-dimensional and one-dimensional, rooted in the unique physicochemical properties exhibited by low-dimensional carbon materials. For instance: graphene is a semi-metal with no intrinsic bandgap, making it difficult to directly apply as a semiconductor device.[1]; when fabricated into quasi-one-dimensional graphene nanoribbons (GNRs) with widths ranging from sub-nanometers to several nanometers, controllable bandgaps can be introduced by regulating the width and edge structure of the nanoribbons, thereby endowing them with semiconductor characteristics.[2-3]; furthermore, rolling graphene into seamless quasi-one-dimensional tubular structures forms carbon nanotubes (CNTs). Depending on their chirality, one-third retain metallic characteristics, while two-thirds transform into semiconductors, with their bandgaps determined by chirality. Pure one-dimensional carbon chains are themselves semiconductor materials, with bandgaps that can be continuously tuned by regulating chain length; the longer the chain, the smaller the bandgap.[4].
Compared to three-dimensional and two-dimensional carbon materials, one-dimensional carbon materials combine excellent mechanical and thermal properties with tunable electrical properties, making them a research hotspot in the field of carbon materials in recent years. However, the precise synthesis of high-performance one-dimensional carbon materials still faces many challenges, such as preparing single-chirality carbon nanotubes, graphene nanoribbons with sub-nanometer widths and controllable edge structures, and carbon chains with controllable lengths and stable structures; these remain key issues urgently needing resolution in this field.
Confined synthesis is a novel material preparation technique developed based on the template method. It utilizes the spatial confinement effect and interfacial interactions of hard templates to guide the directional assembly and reaction of raw material molecules within the template channels. Among various hard templates, carbon nanotubes and boron nitride nanotubes (BNNTs), with their unique hollow crystalline cylindrical structures, have become common templates for the confined synthesis of one-dimensional/quasi-one-dimensional nanomaterials. These nanotubes possess high aspect ratios, which can induce molecules to fill the inner walls via capillary action, promoting the polymerization or self-assembly of precursor molecules inside the tubes to form one-dimensional carbon materials and polymers.[5]. Meanwhile, templates such as metal-organic frameworks (MOFs) and molecular sieves (including zeolites) also play important roles in confined synthesis. Especially those with one-dimensional channels, these materials feature uniform channel structures that can effectively guide molecules to polymerize and self-assemble along a single direction, making them ideal templates for synthesizing one-dimensional carbon materials. Among them, metal-organic frameworks offer high flexibility in structural design; by regulating the combination of organic ligands and metal ions, systematic adjustments to the size, shape, and surface functionality of nanochannels can be achieved. Through precursor encapsulation and confined polymerization within the channels, one-dimensional carbon materials and polymers can be efficiently prepared.[6]; molecular sieves, due to their excellent catalytic performance and uniform pore size distribution, have become important templates for the polymerization of precursor molecules, further enhancing the efficiency of confined synthesis and the regularity of the products.
In summary, templates such as carbon nanotubes, boron nitride nanotubes, metal-organic frameworks, and molecular sieves each possess unique structural and performance advantages in confined synthesis, collectively driving the synthesis and development of one-dimensional carbon materials. This article systematically explores the application advantages and limitations of the aforementioned templates in the confined synthesis of one-dimensional carbon materials, summarizes research progress in related preparation technologies, and outlines future development directions for this field.
2 Confined synthesis within carbon nanotubes
Since Kroto et al. in 1985[7]discovered C60, carbon allotropes have received widespread attention. In 1991, Iijima[8]discovered carbon nanotubes, further stimulating research enthusiasm for novel carbon allotropes. As a hollow one-dimensional nanostructure with diameters ranging from 0.4 to tens of nanometers, the unique spatial confinement effect of carbon nanotubes makes them ideal templates for the confined synthesis of one-dimensional atomic chains, quasi-one-dimensional nanowires, nanoribbons, and novel nanotubes. In 1993, Ajayan et al.[9]utilized capillary action to fill lead into carbon nanotubes, initiating research on confined synthesis within carbon nanotubes. In 1998, Smith et al.[10]filled C60into carbon nanotubes, forming a peapod structure. This was a significant study on carbon nanotubes as templates for small molecules, laying the foundation for subsequent synthesis of one-dimensional carbon materials and polymers using carbon nanotubes as templates. To date, multiple review articles have elaborated on the potential of carbon nanotubes as reaction templates for preparing low-dimensional materials[4-5].
Carbon nanotubes possess a one-dimensional channel structure and can be classified into single-walled carbon nanotubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs) based on the number of layers. Their confinement of encapsulated guests relies primarily on electrostatic interactions generated by highly delocalized electrons, rather than coordination interactions, which allows the encapsulated guests to retain their intrinsic structures and properties to the greatest extent; meanwhile, the chemical stability of carbon nanotubes is also a crucial reason for their effectiveness as reaction templates.
The confinement effect of carbon nanotubes on encapsulated molecules and their interfacial interactions are the core factors determining the structure of confined synthesis products; these two aspects are mechanistically coupled, jointly regulating reaction pathways and product morphology. Regarding the confinement effect, the geometric constraints of the carbon nanotube channels reduce the degrees of freedom of precursor molecules from three-dimensional to quasi-one-dimensional. Molecular motion in the radial direction is strictly limited by the physical boundaries of the tube walls, forcing molecules to align and diffuse along the tube axis. When the channel inner diameter approaches the cross-sectional size of the precursor molecules, the orientational degrees of freedom are further reduced, energetically favoring specific molecular configurations, thereby suppressing multi-directional polymerization side reactions. Taking the synthesis of armchair graphene nanoribbons as an example, precursor molecules inside single-walled carbon nanotubes can only align in specific orientations due to channel size constraints, causing dehydrogenative cyclization reactions to occur solely along the molecular long axis, ultimately forming graphene nanoribbons with specific widths determined by the channel inner diameter.[11-14]. Regarding interfacial interactions, non-covalent interactions dominated by van der Waals forces and π-π stacking exist between the carbon atoms of the nanotube walls and the encapsulated molecules. For aromatic precursor molecules (such as polycyclic aromatic hydrocarbons), this π-π interaction causes the molecular planes to tend to orient parallel to the inner surface of the tube wall, further fixing the molecular arrangement and providing an ordered initial configuration for subsequent polymerization reactions.[15-16]. For carbon chains, the interfacial interaction between the tube wall and the carbon chain modulates the bond length alternation (BLA) amplitude of the carbon chain through charge transfer, thereby altering its bandgap. Both theoretical calculations and experiments indicate that the Raman characteristic frequency of carbon chains is closely related to the inner diameter and chirality of the host carbon nanotubes, confirming the substantial regulatory role of interfacial interactions on the electronic structure of carbon chains.[17]. In summary, the spatial confinement effect of carbon nanotubes determines the macroscopic morphology and width selectivity of the products, while the interfacial interactions between the tube walls and the encapsulands regulate the electronic structure and physical properties of the products at the microscopic level; the synergistic action of these two factors achieves precise control over the structure of one-dimensional carbon materials.
2.1 Confined synthesis of graphene nanoribbons within carbon nanotubes
Graphene nanoribbons, due to their unique electronic properties, can have their characteristics modified by altering their width, edge structure, and doping, thereby meeting application requirements in various fields. They represent an important research direction for one-dimensional carbon materials. Based on the type of edge structure, graphene nanoribbons can be classified into Armchair and Zig-zag types. Armchair graphene nanoribbons (AGNRs) are semiconductors, with bandgaps that can be tuned via width[2,18]. Both theoretical studies and experimental tests indicate that the bandgap of armchair graphene nanoribbons can be tuned from 0 to over 3 eV, with its magnitude correlating to the width[3]. Typically, armchair graphene nanoribbons are classified based on the number of dimer lines along the width direction (N). As shown in Figure 1a, if the number of dimer lines in the width direction is 7, it is denoted as a 7-armchair graphene nanoribbon (7-AGNR). Zig-zag graphene nanoribbons are theoretically metallic[1]; in practical measurements, their bandgap is often less than 0.1 eV[19]. Zig-zag graphene nanoribbons possess exotic edge magnetism[1-2,20-21] and spin transport properties[19,22-23]. Other types of graphene nanoribbons can be regarded as combinations of armchair and zig-zag segments, with properties intermediate between the two, mostly being semiconductor materials. In recent years, the field of graphene nanoribbons has achieved a series of significant advances, covering controlled preparation, device applications, and heterostructure construction. Utilizing a semiconductor AgTe monolayer intercalation method, electronic decoupling of graphene nanoribbons from silver substrates was realized, confirming that this approach can weaken interface interactions and restore their intrinsic electronic properties, providing an effective means for precisely characterizing the intrinsic electronic properties of carbon-based nanomaterials and their device applications[24]. Regarding heterostructure construction, graphene nanoribbon homojunctions and various vertical heterojunctions were precisely fabricated on copper substrates. By tuning the structures, diverse band alignments were achieved, offering effective strategies for fabricating high-performance nanodevices[25].
2.1.1 Confined synthesis of armchair graphene nanoribbons within carbon nanotubes
Surface synthesis is the traditional method for synthesizing armchair graphene nanoribbons[26-29]. Asshown in Figure 1a, 7-AGNRs were synthesized using 10,10′-dibromo-9,9′-bianthryl as precursor molecules on Au(111) through polymerization and dehydrogenative cyclization[30]. Raman spectroscopy, a common characterization technique for carbon materials, is also applicable to the characterization of graphene nanoribbons. Asshown in Figure 1b, graphene nanoribbons exhibit both the G peak typical of sp2 hybridized carbon materials (around 1600 cm-1), as well as a radial breathing-like mode (RBLM) similar to the breathing mode of carbon nanotubes. The width of the nanoribbons can be calculated using Equation (1) based on the peak position of the RBLM.
where:ωRBLM is the peak position of RBLM, cm-1;W is the width of AGNRs (excluding edge H), nm; a and b are constants,a=1667.9 $\sqrt[]{\mathrm{A}̊}∕\mathrm{c}\mathrm{m}$(1 Å=0.1 nm), b=-210.2 cm-1[31]. As shown in Figure 1b, the RBLM peak position is located at 396 cm-1, and the calculated width is 0.76 nm.
Using the same precursor molecules and single-walled carbon nanotubes as templates, 7-AGNRs were also successfully synthesized (Figure 1c,1d)[32]. The interaction between 7-AGNRs and carbon nanotubes causes a shift of several wavenumbers in their Raman spectral peaks relative to surface-synthesized 7-AGNRs. Surface synthesis occurs on a two-dimensional scale; although introducing halogen groups can guide the direction of intermolecular polymerization, the orientation of molecules on the surface is not fully controllable, leading to side reactions that affect the purity and growth direction of graphene nanoribbons. When graphene nanoribbons grow synchronously in all directions until they contact each other, growth stops; therefore, surface-synthesized graphene nanoribbons are generally only tens of nanometers in length. In contrast, the one-dimensional channels of carbon nanotubes confine graphene nanoribbons to grow only along the tube axis, enabling the growth of ultra-long graphene nanoribbons. Furthermore, by controlling the aspect ratio of the precursor molecules, the orientation of molecules entering the carbon nanotubes can be regulated, thereby reducing intermolecular side reactions and improving the purity of the synthesized graphene nanoribbons.
Furthermore, adjacent graphene nanoribbons may also combine with each other to form wider nanoribbons during surface synthesis. Asshown in Figure 2a, using 4,4″-Dibromo-p-terphenyl (DBTP), Scanning Tunneling Microscopy (STM) tests revealed that 3-AGNRs were initially formed on Au(111). With increasing heat treatment temperature, random fusion of adjacent 3-AGNRs can form 6-, 9-, 12-, and 15-armchair graphene nanoribbons (Figure 2a,c)[34]. In contrast, when DBTP was filled into single-walled carbon nanotubes, only Raman spectral signals of 6-AGNRs were observed (Figure 2e)[12], with no signals from wider nanoribbons detected, indicating high purity of the synthesized 6-AGNRs. This is because the diameter of the carbon nanotubes restricts the fusion of graphene nanoribbons in the width direction.
Furthermore, by controlling the directional polymerization between molecules, the purity of target graphene nanoribbons can also be improved. For example, using halogenated derivatives of perylene as precursor molecules, theoretically, they can directionally polymerize into 5-AGNRs on an Au(111) surface (Figure 3a)[35]. However, in actual experiments, the deposition orientation of the molecules is inconsistent, leading to uncontrollable growth directions of the nanoribbons and shorter grown lengths (Figure 3bandFigure 3c). In confined synthesis, when using parent perylene without halogen groups as the precursor molecule, due to the small size of the molecules, their orientations within the carbon nanotubes vary, theoretically resulting in non-unique structures of the finally synthesized graphene nanoribbons (Figure 3d). When using a perylene dimer as the precursor molecule, the molecules can only enter the single-walled carbon nanotubes along their length direction, thus synthesizing high-purity 5-AGNRs (Figure 3e)[13].
More importantly, confined synthesis does not require precursor molecules matching the structural units of graphene nanoribbons; it can directly synthesize graphene nanoribbons from non-structural precursor molecules, breaking through the strict requirements for precursor structures in surface synthesis.[36-37]For example, ferrocene molecules have a "sandwich" structure, where an iron cation (Fe2+) is sandwiched between two parallel cyclopentadienyl anions (C5H5-). According to the principles of surface synthesis, this cannot form graphene nanoribbons composed of benzene rings. However, experiments found that when ferrocene molecules are filled into single-walled carbon nanotubes and decomposed during high-temperature heat treatment, they ultimately transform into a series of graphene nanoribbons with different widths. High-resolution electron microscopy (HRTEM) and Raman spectroscopy confirmed the presence of 6-AGNRs and 7-AGNRs in the samples (Fig. 4a~d)[38]. By changing the diameter of the single-walled carbon nanotubes, samples enriched with either 6-AGNRs or 7-AGNRs can be selectively synthesized (Fig. 4e~g)[14]. Traditional heat treatment usually employs muffle furnaces or tube furnaces, often taking several hours, whereas microwave heating can shorten the synthesis time to tens of seconds. Although the crystallinity of the product is slightly reduced, this provides a feasible pathway for low-cost, large-scale preparation of graphene nanoribbons.[39].
Confined synthesis also provides a facile method for preparing doped graphene nanoribbons: surface synthesis of doped graphene nanoribbons requires first synthesizing precursor molecules containing dopant elements, which involves difficult molecular design and synthesis; whereas in confined synthesis, doping of the product can be achieved simply by filling molecules containing dopant elements into carbon nanotubes. For example, asFigure 5ashows, using aminoferrocene or cyanoferrocene as precursor molecules filled into single-walled carbon nanotubes for confined synthesis can synthesize nitrogen-doped 6-AGNRs and 7-AGNRs[40]. AsFigure 5bshows, the Raman spectrum of nitrogen-doped 6-AGNRs exhibits a certain shift relative to the undoped sample, and the degree of shift is positively correlated with the nitrogen doping content. In addition, using deuterated p-terphenyl as a precursor molecule, deuterated 6-AGNRs can be synthesized in single-walled carbon nanotubes (Figure 5c)[12]. AsFigure 5dshows, the Raman peak located at 1244 cm-1belongs to the C—H in-plane vibration mode, while the C—H peak of fully deuterated 6-AGNRs shifts to 879 cm-1, and controllable deuteration of graphene nanoribbons can be achieved by adjusting the mixing ratio of deuterated and hydrogen-saturated precursors.
2.1.2 Confined synthesis of thio-zigzag graphene nanoribbons within carbon nanotubes
Zigzag graphene nanoribbons (zz-GNRs) are an important research type of graphene nanoribbons[19,41]. The surface synthesis method was the first to synthesize zigzag graphene nanoribbons, but it is often limited by precursors, resulting in wider widths. Due to the high energy and poor stability of zz-GNRs themselves, narrower zz-GNRs are more difficult to synthesize. By introducing sulfur elements to form thio-zigzag graphene nanoribbons, the energy state of the products can be effectively reduced and structural stability improved, becoming an important preparation strategy for narrow-width zigzag graphene nanoribbons.
The core of the confined synthesis method for synthesizing sulfur-doped zz-GNRs lies in selecting suitable sulfur-containing precursor molecules; currently, commonly used ones in research are C60molecules[42]or tetrathiafulvalene (TTF) molecules (Figure 6a)[43]. After filling the above two types of molecules into single-walled carbon nanotubes, through heat treatment (Figure 6b) or electron beam irradiation (Figure 6c), both methods can induce polymerization and structural transformation of the molecules, ultimately forming sulfur-doped zz-GNRs (Figure 6bandFigure 6c).
2.1.3 Confined synthesis of other types of graphene nanoribbons within carbon nanotubes
Other types of graphene nanoribbons, such as Chevron-type GNRs and Cove-type GNRs, can be regarded as combinations of segments from armchair and zigzag graphene nanoribbons; therefore, their properties lie between the two, and they are generally also semiconductor materials. Asshown in Figure 7b,7c, coronene is a fully fused polycyclic aromatic hydrocarbon, and its one-dimensional polymer can be converted into graphene nanoribbons with irregular edges[44]. However, theoretical calculations indicate that when coronene molecules are filled into carbon nanotubes, the most likely arrangement is actually longitudinal stacking (Figure 7a)[45], which is obviously unfavorable for forming graphene nanoribbons, thus leading to limited lengths of the synthesized graphene nanoribbons[46]. Fujihara et al.[47]attempted to first convert coronene molecules into their dimers before filling them into carbon nanotubes to grow graphene nanoribbons, which improved the structure and length of the formed graphene nanoribbons to some extent (Figure 7d).
Compared with surface synthesis methods, confined synthesis offers advantages in structural controllability, reaction selectivity, and product size regulation. Surface synthesis is performed on metal substrates (such as Au(111)), where precursor molecules are influenced by substrate symmetry and molecule-substrate interactions, diffusing in multiple directions; nanoribbons grow synchronously in different directions until they contact each other and stop, typically reaching lengths on the order of tens of nanometers. In contrast, carbon nanotube channels restrict molecular motion to the one-dimensional tube axis direction, allowing the resulting GNRs to reach lengths on the micrometer scale.[11,13,38]. Furthermore, in surface synthesis, random lateral fusion can occur between adjacent GNRs, forming mixed products with widths of 3N(N=1, 2, 3, …).[34]; in confined synthesis, the physical constraint of the channel diameter prevents the formation of products exceeding the channel dimensions. For example, DBTP synthesizes only 6-AGNRs without 9-AGNRs within SWCNTs of specific inner diameters.[12]. In surface synthesis, the GNR width is primarily determined by the precursor molecular structure; changing the width requires redesigning the synthetic precursor, involving complex molecular synthesis steps. In confined synthesis, while keeping the precursor unchanged, the product width can be continuously regulated by systematically adjusting the template channel diameter. For instance, using ferrocene as a precursor and selecting SWCNTs with different inner diameters allows for the selective preparation of products enriched in either 6-AGNRs or 7-AGNRs.[14], providing an independent degree of freedom for external control of product size beyond precursor design. These advantages make the confined synthesis method one of the optimal routes for high-purity, large-scale synthesis of graphene nanoribbons with specific widths.
2.2 Confined synthesis of small-diameter carbon nanotubes within carbon nanotubes
Double-walled carbon nanotubes (DWCNTs) are multi-walled carbon nanotubes with the fewest layers, combining the small diameter and high flexibility of single-walled carbon nanotubes with the high chemical stability of multi-walled carbon nanotubes. They have important applications in fields such as nanoelectronic devices and composite materials, and their precise fabrication is a key research focus for one-dimensional carbon materials.[48-51].
In 2001, studies reported that filling C60into single-walled carbon nanotubes as a carbon source, followed by heat treatment at 1100 °C, could convert C60into inner-layer carbon nanotubes, ultimately forming double-walled carbon nanotubes (Figure 8a); Raman spectroscopy could clearly detect the breathing modes of the inner small-diameter carbon nanotubes (Figure 8b)[52]. However, this method has limitations; even with complete filling of C60, the provided carbon source is difficult to form inner-layer carbon nanotubes equal in length to the outer-layer carbon nanotubes. Subsequent research[53]found that various molecules such as ferrocene can serve as carbon sources to grow inner-layer carbon nanotubes confined within carbon nanotubes, thereby preparing double-walled carbon nanotubes.
Recent studies[53]found that without adding an external carbon source, the conversion from single-walled carbon nanotubes to double-walled carbon nanotubes can be achieved solely through high-temperature heat treatment (Figure 8c). Raman spectroscopy studies revealed that single-walled carbon nanotubes are unstable at high temperatures of 1500 °C; their diameters gradually shrink as the temperature increases. During the structural reconstruction process, excess carbon atoms can serve as a carbon source to form inner-layer carbon nanotubes inside the tubes (Figure 8d). This method does not require filling with precursor molecules. The diameter of the initial single-walled carbon nanotubes is tunable, enabling the preparation of double-walled carbon nanotubes with ultra-small inner diameters. Furthermore, high-temperature heat treatment improves the crystallinity of the inner-layer carbon nanotubes, offering characteristics of both high yield and high quality, with a broader scope of application compared to the fullerene filling method.
2.3 Carbon nanotube-confined synthesis of carbon chains
Carbyne is a one-dimensional linear carbon allotrope formed by carbon atoms with sp¹ hybridization. Theoretically, it is a semiconductor material whose band gap decreases as the chain length increases. In terms of mechanical properties, its tensile strength along the length direction is twice that of graphene, making it the carbon material with the best known mechanical properties.[4,54-55]. The chemical bond structure of carbyne is mainly divided into two categories: one consists of alternating single and triple bonds between carbon atoms; the other consists of consecutive double bonds between carbon atoms. Theoretical studies indicate that carbyne with consecutive double bonds has high energy and poor stability, making it difficult to synthesize and exist stably. Therefore, carbyne with alternating single-triple bonds has become the core direction of carbyne research.
Organic synthesis is a traditional method for preparing carbon chains, but the products exhibit poor stability and short lengths, typically requiring chemical groups to be introduced at the molecular ends for capping (polyynes). Currently, the longest carbon chain synthesized organically contains only 68 carbon atoms.[56]. The minimum diameter of carbon nanotubes can be less than 1 nm, and their one-dimensional channel spatial confinement effect enables the confined synthesis of single-atom-wide carbon chains. Furthermore, the interfacial interaction between the carbon nanotube walls and the carbon chains can effectively regulate the physical properties of the chains while simultaneously protecting them and enhancing their structural stability. Compared with end-capped carbon chains produced by organic synthesis, confined carbon chains exhibit significantly improved stability and allow for the preparation of ultra-long carbon chains nearly 1 μm in length containing over 6000 carbon atoms, thereby breaking through the length limitations of organic synthesis methods.[15-16,57].
2.3.1 Confined synthesis of carbon chains in multi-walled carbon nanotubes
The confined synthesis of carbon chains in carbon nanotubes was first achieved in multi-walled carbon nanotubes[58]: By introducing hydrogen partial pressure in an inert gas, one-dimensional carbon chain structures were discovered for the first time inside multi-walled carbon nanotubes in the cathode deposits after arc discharge of graphite rods. The product lengths ranged from a few nanometers to tens of nanometers, with the longest carbon chains containing over 100 carbon atoms. Electron microscopy characterization confirmed that the carbon chains are located at the center of the multi-walled carbon nanotubes, forming a carbon chain@carbon nanotube structure[59]; first-principles calculations indicate that this structure can effectively enhance the conductivity of carbon nanotubes[58,60-61], providing new ideas for the preparation of carbon-based conductive materials.
2.3.2 Confined synthesis of carbon chains in double-walled carbon nanotubes
Unlike the preparation of multi-walled carbon nanotubes, since the arc discharge method for preparing double-walled carbon nanotubes requires metal catalysts, it cannot be used to directly synthesize carbon chains within double-walled carbon nanotubes. Asshown in Figure 9a, to obtain ultra-long carbon chains, scientists first attempted filling double-walled carbon nanotubes with short polyynes (C10H2), followed by heat treatment at 800 °C to fuse multiple polyynes inside a single double-walled carbon nanotube into long carbon chains. Asshown in Figure 9b, the Raman characteristic peak of polyynes containing 10 carbon atoms is located at 2066 cm-1; after heat treatment, the Raman characteristic peak shifted to 1831 cm-1. Since the Raman peak of carbon chains decreases as the chain length increases, this confirms the formation of long carbon chains[62]. However, the yield of carbon chains obtained by this method is low, and the chain length remains limited to tens of nanometers; therefore, new methods for synthesizing high-yield carbon chains need to be explored.
Recently, researchers achieved the preparation of ultra-long carbon chains inside double-walled carbon nanotubes without external carbon sources: small-diameter double-walled carbon nanotubes with an inner diameter of approximately 0.9 nm were subjected to high-vacuum heat treatment at 1460 °C, successfully growing ultra-long carbon chains approximately 1 μm in length and containing over 6000 carbon atoms inside the tubes (Figure 9c)[63]. Asshown in Figure 9d, the Raman peak of the carbon chain is higher than the G peak of the carbon nanotube, indicating a high yield of the carbon chain. In this method, the carbon source for growing the carbon chain mainly comes from two aspects: first, amorphous carbon remaining in the carbon nanotubes; second, excess carbon atoms generated from the structural reconstruction of carbon nanotubes at high temperatures. Subsequent research found that selective heating of individual double-walled carbon nanotubes using lasers can also form carbon chains inside the tubes, and controllable synthesis of carbon chains can be achieved by regulating the diameter of the double-walled carbon nanotubes[64]. Property studies indicate that the bandgap of the carbon chain is related to the chirality of the enclosing carbon nanotube; therefore, precise adjustment of the carbon chain bandgap can be achieved by regulating the chirality of the carbon nanotube[17].
When using single-walled carbon nanotubes of different diameters to prepare carbon chains via high-temperature heat treatment without providing an additional carbon source, the single-walled carbon nanotubes actually first transform into double-walled carbon nanotubes. The carbon chains ultimately grow inside these double-walled carbon nanotubes, and the physical properties of the carbon chains are related to the chirality of the inner tubes of the formed double-walled carbon nanotubes.[64-66]. However, due to the limited carbon source provided by the single-walled carbon nanotubes, the yield of carbon chains is low. To improve the yield of carbon chains, it is necessary to introduce an additional carbon source.
Fullerenes (such as C60 and C70), due to their sub-nanometer size, can be filled into single-walled carbon nanotubes to form peapod structures[67]. At high temperatures, fullerene molecules fuse within the carbon nanotubes, eventually forming inner carbon nanotubes; their diameter is independent of the fullerene type but correlates with the diameter of the single-walled carbon nanotubes[68]. As shown in Figure 10a, continuing to increase the heat treatment temperature results in the formation of carbon chains within double-walled carbon nanotubes; compared to samples without fullerene filling, the yield of carbon chains increased by approximately fourfold[69]. In addition to polyynes and fullerenes, many organic molecules can actually serve as precursor molecules, because their pyrolysis at high temperatures can provide additional carbon sources for the formation of inner carbon nanotubes and carbon chains. For example, using ethanol as a precursor molecule increased the yield of carbon chains by approximately twofold[70].
Carbon sources can also be introduced on the exterior of carbon nanotubes. Asshown in Figure 10c, surfactants are wrapped around the carbon nanotubes; under high temperatures, the surfactants decompose, allowing some carbon atoms to enter the carbon nanotube lattice, providing a carbon source for carbon chain formation and increasing the yield by an average of approximately 1.5 times. Furthermore, to promote the reconstruction of carbon nanotubes, defects are introduced onto the surface of double-walled carbon nanotubes via oxidation. These defective double-walled carbon nanotubes reconstruct under high temperatures, facilitating the formation of double-walled carbon nanotubes with smaller diameters. This process generates excess carbon atoms, thereby providing additional carbon sources for carbon chain formation.
Based on the above research, the researchers finally proposed a three-step method to synthesize carbon chains with high yield (Figure 10c): first, fill single-walled carbon nanotubes with fullerenes internally and wrap them with surfactants externally; second, convert the single-walled carbon nanotubes into double-walled carbon nanotubes under vacuum at 1300 °C, then perform heat treatment in air at 500 °C to oxidize and introduce surface defects; finally, conduct heat treatment under vacuum at 1560 °C to obtain carbon chains@double-walled carbon nanotubes. Ultimately, the yield of carbon chains increased by more than 30 times compared to direct high-temperature heat treatment of single-walled carbon nanotubes.
In addition to improving the yield of carbon chains, using precursor molecules has another important function: isotopically labeled or heteroatom-doped carbon chains can be synthesized by regulating the elements within the precursor molecules. For example, using 50% isotopically labeled ethanol as a precursor molecule, the isotopic labeling content of carbon atoms in the carbon chains was calculated to be 4.6%~11.9% based on the characteristic peak shifts in Raman spectra.[70]; whereas after using approximately 81% isotopically labeled C60, the isotopic content of the prepared carbon chains reached 28% (Figure 10b)[69]. Furthermore, when using C59N (i.e., one C atom in C60 replaced by an N atom) as the precursor molecule, the characteristic peaks of the carbon chains were observed to shift to lower wavenumbers by 2~4 cm-1, thus suggesting the preparation of nitrogen-doped carbon chains[71].
2.3.3 Confined synthesis of carbon chains in single-walled carbon nanotubes
Although some studies in the previous section initially used single-walled carbon nanotubes, they ultimately transformed into double-walled carbon nanotubes, meaning the carbon chains were still formed within the double-walled carbon nanotubes.[68-69,72]. Theoretical calculations indicate that the confined growth of carbon chains inside single-walled carbon nanotubes possesses unique physical significance: carbon chains can modulate the physical properties of single-walled carbon nanotubes. For instance, both carbon chains and (8,0) single-walled carbon nanotubes are semiconductor materials with different band gaps, yet the composite structure of carbon chain@single-walled carbon nanotube formed by them exhibits metallic behavior.[73]. Further research confirms that regardless of whether the single-walled carbon nanotube is metallic or semiconducting, the carbon chain@single-walled carbon nanotube composite structure is always metallic.[74], this characteristic endows it with significant application potential in the field of nano-electronic devices, making the synthesis of carbon chains inside single-walled carbon nanotubes a research hotspot.
Single-walled carbon nanotubes become structurally unstable at temperatures exceeding 1000 °C; therefore, traditional arc discharge methods and high-temperature high-vacuum heat treatment methods are unsuitable for growing carbon chains inside them. Researchers have begun seeking alternative methods to achieve confined synthesis of carbon chains within single-walled carbon nanotubes. For instance, after ultrasonic treatment of a suspension of carbon chains@double-walled carbon nanotubes, the carbon chains@inner carbon nanotubes are extracted from the outer layers of the double-walled carbon nanotubes, and then separated via gradient density centrifugation to obtain carbon chains@single-walled carbon nanotubes.[75]. However, this method yields low productivity, and ultrasonic treatment can also damage the structure of the carbon chains; therefore, other methods need to be developed to improve the yield.
Since polyacetylene molecules can fuse together at 800 °C, this method is also applicable for obtaining carbon chains@single-walled carbon nanotubes by fusing polyacetylene molecules inside single-walled carbon nanotubes[76]. Asshown in Figure 11a, placing polyacetylene and open-ended single-walled carbon nanotubes in a hydrothermal reactor at 80 °C effectively fills the polyacetylene into the single-walled carbon nanotubes. After suction filtration to form a film, annealing in an argon atmosphere at 700 °C fuses the polyacetylene into carbon chains. Raman spectroscopy shows (Figure 11b) that the characteristic peak corresponding to polyacetylene (approximately 2200 cm-1) disappears after annealing, while the characteristic peak corresponding to carbon chains appears (approximately 1860 cm-1), indicating that the polyacetylene has fused into carbon chains[77]. However, the structure of polyacetylene itself is not sufficiently stable at room temperature; therefore, the high-pressure environment at 80 °C destroys some of the polyacetylene, thereby reducing the filling rate. To improve the filling rate of polyacetylene, subsequent studies found that if single-walled carbon nanotubes with a slightly larger diameter (0.9 nm) are used, polyacetylene can spontaneously diffuse into the single-walled carbon nanotubes under conditions below room temperature (4 °C). After converting into carbon chains at an annealing temperature below 900 °C, the Raman spectra exhibit richer structures, indicating that polyacetylene can fuse within single-walled carbon nanotubes of different diameters, and its length has a certain distribution range (Figures 11c~e)[78].
To overcome the challenge of low carbon chain yield caused by the inherent structural instability of polyynes, scientists have recently begun to fill other precursor molecules and then convert them into carbon chains inside single-walled carbon nanotubes. For example, during the process of dispersing single-walled carbon nanotubes using surfactants, it was found that surfactants not only wrap around the surface of the nanotubes, but some molecules also enter the interior (Figure 12a); these molecules are converted into carbon chains after heat treatment at 400 °C, with a yield higher than that achieved using polyyne precursors[79]. Meanwhile, our research group discovered that various aromatic hydrocarbon molecules, when used as precursors and annealed at 500 °C, are converted into carbon chains within single-walled carbon nanotubes[80], and this method exhibits strong scalability; by substituting different aromatic hydrocarbon precursors, it holds promise for preparing confined carbon chains with isotopic labeling or heteroatom doping.
2.4 Confined synthesis of carbides in carbon nanotubes
Carbides are binary compounds formed by carbon and elements with lower electronegativity (mainly metals and some metalloids), in which carbon typically exists as an anion or quasi-anion. Carbides can be primarily classified into covalent and metallic types, with metallic carbides serving as important functional materials in the field of catalysis. Taking non-noble metal molybdenum carbide (MoC) as an example, its electronic structure is highly similar to that of the noble metal platinum, making it a class of high-performance electrocatalysts for the hydrogen evolution reaction. However, carbides are chemically relatively active and are highly prone to undergo oxidation in air, leading to a significant decline or even complete loss of catalytic activity. In contrast, carbon nanotubes possess excellent chemical stability and can provide effective protection for carbides confined within them, enabling the encapsulated carbides to maintain stable catalytic performance.[81].
Strategies such as the molten salt method and pyrolysis can be used to prepare carbide@carbon nanotube composite structures[82-85], but in these methods, carbides and carbon nanotubes are generated simultaneously, resulting in final products that typically consist of small-sized carbide particles and multi-walled carbon nanotubes with structural defects (Figure 13aandFigure 13b). Similarly, by doping metal precursors into the graphite anode of an arc discharge system, yttrium carbide@multi-walled carbon nanotube composites can be directly prepared[86]. Furthermore, mixing multi-walled carbon nanotubes with metal oxides or metal halides allows them to react during annealing, ultimately producing various carbide nanorod structures such as TiC, NbC, FeC, and SiC[87].
Regarding research on the confined synthesis of carbides within the inner cavity of carbon nanotubes, current mainstream preparation methods are primarily solution-based and vapor transport methods. Among these, the vapor transport method offers advantages of high filling efficiency and broad material applicability, making it more suitable for the confined preparation of metallic carbides.[81,88].Figure 13candFigure 13dAs shown in , by utilizing vapor transport technology to fill precursor molecules into the cavities of single-walled carbon nanotubes, followed by subsequent high-temperature annealing treatment, the controlled synthesis of MoC@single-walled carbon nanotube composite structures can be achieved.[81].[89].
3 Confined synthesis of boron nitride nanotubes
Boron nitride nanotubes are one-dimensional hollow nanomaterials formed by alternating B and N atoms connected via covalent bonds. Their structure is similar to that of carbon nanotubes, and they can be classified into single-walled and multi-walled boron nitride nanotubes based on the number of layers. Currently, the preparation of high-purity single-walled boron nitride nanotubes is challenging, so research and applications primarily focus on multi-walled boron nitride nanotubes. Compared to carbon nanotubes, although it is difficult for boron nitride nanotubes to reach sub-nanometer diameters, they possess superior thermal stability and chemical inertness[90-92], maintaining structural integrity at even higher temperatures, making them suitable as templates for confined synthesis under high-temperature conditions.
Carbon nanotubes exhibit numerous chiralities, being either metallic or semiconducting, which significantly affects the characterization of physical properties of materials synthesized within them. In contrast, boron nitride nanotubes possess a bandgap as high as 5.5 eV and are transparent in the visible light range, enabling real-time tracking of precursor molecule assembly and reaction processes inside the tubes via optical methods; this provides a visual means for studying confined synthesis mechanisms. Furthermore, after synthesis is complete, it facilitates the characterization of the optical properties of the products. As insulating materials, boron nitride nanotubes exert no shielding effect on the electrical properties of synthesized products within the tubes, allowing direct characterization of the product's electrical performance and avoiding interference from the intrinsic electrical properties of carbon nanotubes during testing. Additionally, the π-electron density of boron nitride nanotubes is lower than that of carbon nanotubes; consequently, their π-π interactions with internal aromatic molecules are weaker. This not only favors the filling and diffusion of precursor molecules within the tubes but also facilitates the extraction of products from the tubes after synthesis, yielding pure-phase one-dimensional carbon materials. Theoretical calculations reveal that when hexabenzocoronene is used as a precursor encapsulated in boron nitride nanotubes, molecular polymerization can form graphene nanoribbons (Figure 14a)[93]; experimentally, graphene nanoribbons synthesized under confinement within boron nitride nanotubes have also been observed (Figure 14b)[94]. Recently, by filling 1,2,4-trichlorobenzene (TCB) into boron nitride nanotubes, graphene nanoribbons were also observed in samples subjected to high-temperature annealing (Figure 14c and Figure 14d)[95], although their specific structures still require further characterization and confirmation.
Currently, the preparation of small-diameter boron nitride nanotubes is challenging, limiting their widespread application in confined synthesis; however, owing to their advantages such as electrical insulation, high thermal stability, optical transparency, and ease of product extraction, confined synthesis using boron nitride nanotubes is expected to become an important technical means for preparing high-purity, high-performance one-dimensional/quasi-one-dimensional carbon materials. With breakthroughs in the preparation technology of small-diameter boron nitride nanotubes, their applications in the field of confined synthesis will be further expanded.
4 Confined synthesis in metal-organic frameworks
Metal-organic frameworks are crystalline porous materials formed by the self-assembly of metal ions/clusters and organic ligands through coordination bonds[96]. By selecting different metal clusters and organic ligands, the pore size, shape, and surface chemical properties can be precisely tuned. In particular, a class of metal-organic frameworks possesses one-dimensional non-interconnected channels with diameters distributed between 0.3 and 5.0 nm, such as MOF‑74 and MIL‑53, which can serve as templates for the confined synthesis of one-dimensional materials[97]. Furthermore, the metal-organic framework templates can be removed under relatively mild conditions to obtain pure-phase one-dimensional materials[97-100]. However, due to the generally limited thermal stability of metal-organic frameworks, with decomposition temperatures often below 400 °C, they cannot withstand the high temperatures required for preparing one-dimensional carbon materials. Therefore, developing metal-organic frameworks with higher thermal stability or developing precursor molecules capable of polymerization and dehydrogenative cyclization at lower temperatures represents a frontier research direction in this field.
4.1 Metal-Organic Framework Confined Synthesis of Graphene Nanoribbons
Metal-organic frameworks with highly ordered, tunable one-dimensional pores can precisely match the pore size to the width of target graphene nanoribbons. Through physical confinement effects, they strictly control the growth direction of precursor molecules. Meanwhile, by regulating metal nodes and organic linkers, precise design of pore shape and chemical environment is achieved, providing a structurally controllable template for the confined synthesis of graphene nanoribbons.[101].
In early studies, pyrolyzing metal-organic frameworks in an inert gas could first convert them into one-dimensional carbon nanorods, which, after exfoliation, yielded graphene nanoribbons; however, the products prepared by this method exhibited large widths, multiple layers, non-unique structures, and unclear edges, showing a significant deviation from precisely defined graphene nanoribbons.[102]. Recently, researchers synthesized a metal-organic framework possessing one-dimensional square channels with dimensions of 6.9 Å × 6.9 Å (1 Å = 0.1 nm). Asshown in Figure 15a, using perylene as the precursor molecule, due to the confinement of the channel dimensions, the molecules could only enter the one-dimensional channels along their length direction. After heat treatment at 400 °C, the molecules polymerized and underwent dehydrogenative cyclization to form 5-armchair graphene nanoribbons.[103]. Raman spectroscopy revealed distinct breathing-like modes, C—H peaks, and G peaks (Figure 15bandFigure 15c), indicating good crystallinity of the graphene nanoribbons.
4.2 Metal-Organic Framework Confined Synthesis of Polyacenes
Polyacene is a quasi-one-dimensional carbon material formed by the linear fusion of benzene rings. Due to its long-range π-conjugated system, it possesses excellent semiconductor characteristics and conductivity, making it an important research subject in fields such as organic electronic devices and new energy materials.[104]. Polyacene can be regarded as a 2-zigzag graphene nanoribbon, representing the narrowest zigzag graphene nanoribbon. The bandgap of polyacene decreases as the number of fused benzene rings increases, and its properties can be tuned by controlling the chain length. However, in traditional methods, when the number of benzene rings reaches or exceeds five, the synthesis difficulty of polyacene increases significantly. Previously, surface chemical synthesis successfully produced polyacene containing 12 benzene rings on Au(111) (Figure 16a and Figure 16b)[105]. Using metal-organic frameworks as templates for confined synthesis can effectively overcome the length limitations of polyacene, enabling the preparation of ultra-long polyacene. As shown in Figure 16c, using 2,6-bis(bromomethyl)naphthalene/anthracene as precursor molecules filled into the aforementioned metal-organic framework used for preparing 5-armchair graphene nanoribbons, followed by heat treatment at 250 °C, ultra-long polyacene was successfully prepared. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry tests indicated that the prepared polyacene contains up to several dozen benzene rings (Figure 15c)[106], far exceeding the length of polyacene synthesized by conventional methods.
4.3 One-dimensional polymers synthesized within metal-organic framework confinement
Polythiophene is a typical conjugated polymer with unique photoluminescent properties and high conductivity, offering broad potential applications in organic electronic devices and nonlinear optics. Unsubstituted polythiophenes prepared by traditional methods are insoluble in common solvents, making it difficult to achieve ordered arrangement of molecular chains; whereas substituted polythiophenes, although processable, suffer from shortened conjugation lengths due to the introduction of substituents, which reduces their optical and electrical performance. Confined synthesis provides an effective approach to address this issue, enabling the preparation of unsubstituted polythiophenes within the one-dimensional channels of metal-organic frameworks to achieve ordered molecular chain alignment while preserving the complete conjugated system, thereby enhancing product performance.[107]. For example, as shown in Figure 17a and 17b, unsubstituted and substituted polythiophenes can be synthesized separately within metal-organic frameworks using different precursor molecules[108]. Based on the characteristic peaks in the Raman spectra, it is inferred that both contain more than 10 structural units (Figure 17b, 17d). More importantly, when using chiral metal-organic frameworks as templates, chiral polythiophenes can be prepared through spatial confinement and chiral induction effects (Figure 17e and Figure 17f)[109], providing a new method for the preparation of chiral conjugated polymers.
In addition to polythiophene, metal-organic frameworks can also be used for the confined synthesis of various one-dimensional polymer polymers, achieving ordered arrangement of product molecular chains and enhancing polymer performance. For example, styrene can polymerize within the pores of metal-organic frameworks to form polystyrene (Figure 17g)[110], and divinylbenzene can avoid the formation of cross-linked structures under confined conditions, generating discrete linear polymers. Currently, researchers have prepared polydivinylbenzene through metal-organic framework confined synthesis (Figure 17h)[111], polyacetylene[112-120], and a series of other one-dimensional polymers, confirming the universality and effectiveness of this method in the preparation of one-dimensional polymers.
5 Zeolite-confined synthesis
Molecular sieves are a class of porous aluminosilicate or aluminophosphate materials with regular crystalline channel structures. Their channel sizes are uniform and fall within the molecular scale range, enabling precise molecular screening and assembly through spatial confinement effects, making them traditional catalytic and separation materials. Among them, one-dimensional molecular sieves have channels that extend only in a single direction without interconnecting with other channels. Precursor molecules can only undergo one-dimensional diffusion and reaction along the channel axis, effectively guiding molecules to polymerize in a single direction to form ordered one-dimensional nanomaterials and polymer chains, serving as novel hard templates for the confined synthesis of one-dimensional carbon materials. Compared to metal-organic frameworks, molecular sieves possess extremely high thermal stability, making them suitable for confined synthesis under higher temperature conditions. More importantly, molecular sieves can be dissolved under mild weakly alkaline conditions, allowing the template to be easily removed after preparation to obtain pure-phase one-dimensional carbon materials. Additionally, acidic sites within the molecular sieve channels can catalyze certain reactions.[121-126].
Research on the confined synthesis of carbon materials within zeolites began relatively early. In 1997, Kyotani et al.[127]successfully synthesized a porous carbon structure with an amorphous configuration and a specific surface area as high as 2000 m²/g within zeolite channels. In subsequent studies, ultra-small diameter carbon nanotubes with a diameter of approximately 0.4 nm were successfully synthesized in AFI-type zeolites possessing one-dimensional channels (Figure 18a). After removing the zeolite template, due to the structural instability of the small-diameter carbon nanotubes, they underwent self-assembly to form multilayer graphene nanoribbons (Figure 18b~d)[128]. Recent studies[129]have found that various aromatic molecules can be converted into 3-armchair graphene nanoribbons within zeolites such as SAPO-11, ZSM-12, and mordenite. Furthermore, organic molecules can also undergo confined polymerization within zeolite channels to form one-dimensional polymer chains. For instance, methyl methacrylate (MMA) molecules are confined within zeolite channels to polymerize into polymethyl methacrylate (PMMA)[130-132]. Compared with PMMA products obtained via traditional polymerization, PMMA obtained through confined polymerization exhibits a higher degree of polymerization[132].
6 Summary and Outlook
The above four types of templates exhibit significant differences in pore size, interface interaction strength, thermal stability, and applicable reaction types, exerting a decisive influence on the product structure. Carbon nanotube pores range from 0.4 nm to tens of nanometers in diameter; the interaction between wall carbon atoms and encapsulated molecules is dominated by van der Waals forces and π-π stacking, resulting in strong interfacial interactions and extremely high thermal stability, withstanding temperatures above 1200 °C in vacuum, making them suitable for high-temperature polymerization and pyrolytic carbonization. Precise selection of single-walled carbon nanotube diameters can confine the product width to specific values; for instance, diameters of 0.7–0.9 nm are suitable for synthesizing one-dimensional carbon chains, while diameters between 1.2–1.3 nm selectively yield 6-AGNRs[14,38]. In comparison, boron nitride nanotubes are often multi-walled with diameters typically greater than 2 nm, making it difficult to prepare sub-nanometer samples. The polarity of the B—N bonds in the tube walls leads to weaker π-π interactions with encapsulated molecules compared to carbon nanotubes; however, their thermal stability is equally excellent, withstanding temperatures above 1800 °C in vacuum. Their electrical insulation and optical transparency make them suitable for in situ optical characterization without interfering with the optical signals of the products[90-92]. Unlike the first two templates, metal-organic frameworks possess more uniform pore sizes, which can be precisely tuned between 0.3 and 5.0 nm through the combination of ligands and metal nodes. Chemical functional sites provided on the inner pore walls (such as chiral sites) can exert additional influence on polymerization stereoregularity; however, their thermal decomposition temperatures are typically below 400 °C, limiting their application to mild conditions such as oxidative polymerization, radical polymerization, and [2+2] photopolymerization[97-101,116]. Zeolite pore sizes are uniform and at the molecular scale (0.3–1.3 nm), exhibiting good thermal stability, with high-silica zeolites withstanding temperatures above 600 °C in vacuum. Furthermore, acidic sites within the pores can catalytically activate precursor molecules entering the channels, promoting polymerization between internal molecules or molecular isomerization.
Confined synthesis relies on the spatial confinement effect and interfacial interactions of hard templates to effectively guide the reaction pathways of precursor molecules, achieving precise control over the dimensionality, width, edge structure, and even chirality of the products. In polymerization reactions, it can effectively suppress intermolecular cross-linking, enabling efficient linear polymerization, and serves as a core technology for preparing one-dimensional carbon materials with controllable structures and high performance. This paper systematically reviews the research progress of confined synthesis technology in the preparation of one-dimensional carbon materials, focusing on the structural advantages, application scenarios, and preparation technical pathways of four types of hard templates: carbon nanotubes, boron nitride nanotubes, metal-organic frameworks, and molecular sieves. The main conclusions are as follows:
(1) Carbon nanotubes are currently the most mature confined synthesis templates, capable of precisely synthesizing various one-dimensional carbon materials such as graphene nanoribbons, carbon chains, and small-diameter carbon nanotubes. The products feature long lengths and high purity, with flexible precursor design. They also enable doping and isotopic labeling of the products, serving as the core template for the confined synthesis of one-dimensional carbon materials.
(2) Boron nitride nanotubes possess advantages such as high thermal stability, optical transparency, absence of electrical shielding effects, and weak interfacial interactions. These features provide a new pathway for confined synthesis under high-temperature conditions, facilitate the tracking of reaction processes, and enable the extraction of pure-phase products. However, the difficulty in obtaining small-diameter samples limits their widespread application.
(3) Metal-organic frameworks offer high flexibility in structural design, enabling precise control over pore size, shape, and chemical environment. They facilitate the confined synthesis of one-dimensional carbon materials and polymers under mild conditions, with templates that can be gently removed; however, insufficient thermal stability remains a major bottleneck for their application.
(4) Molecular sieves possess ultra-high thermal stability, excellent catalytic activity, and the advantage of easy template removal. They enable the combination of confinement effects and catalytic effects, making them suitable for confined synthesis under high-temperature conditions while enhancing the degree of polymerization of polymers. As a novel potential template for the confined synthesis of one-dimensional carbon materials, their acidic sites within the pores can sometimes affect the confined synthesis.
Based on the current state of research, future studies in the field of confined synthesis of one-dimensional carbon materials should focus on the following directions:
(1) Template structure optimization and performance enhancement: Conduct targeted research on existing issues with various templates, such as developing preparation technologies for single-diameter, single-chirality carbon nanotubes, overcoming bottlenecks in preparing small-diameter boron nitride nanotubes, researching metal-organic frameworks with high thermal stability and tunable pore sizes, and improving the pore channel control precision and surface functionalization levels of molecular sieves to further leverage the structural advantages of various templates.
(2) Seeking novel template systems suitable for confined synthesis: Covalent Organic Frameworks (COFs) possess uniform one-dimensional channels and tunable pore sizes (0.5~5.0 nm). Their frameworks are metal-free, exhibit good chemical inertness, and hold potential for achieving the confined synthesis of graphene nanoribbons or conjugated polymers at lower temperatures. Their channels can be precisely regulated through monomer structural design, and the templates can be dissolved and removed under mild conditions. Furthermore, new types of porous materials with uniform sub-nanometer channels, such as Zeolitic Imidazolate Frameworks (ZIF series) and Porous Aromatic Frameworks (PAFs), are also expected to be applied in confined synthesis.
(3) Deepen research on the mechanism of confined synthesis: Utilize in-situ characterization techniques (such as in-situ Raman, in-situ electron microscopy, and in-situ optical characterization) to track in real-time the assembly, polymerization, and transformation processes of precursor molecules within template pores. Clarify the regulatory patterns of confinement effects and interfacial interactions on product structure and performance, establish the structure-activity relationship of "template structure-reaction conditions-product performance," and provide theoretical support for the precise control of confined synthesis.
(4) Expanding precursor types and reaction systems: Break through the structural limitations of traditional precursors, develop more non-structural precursor molecules containing heteroatoms, and achieve diversified doping and modification of one-dimensional carbon materials; explore new reaction systems such as low-temperature polymerization, photocatalytic polymerization, and electrocatalytic polymerization to reduce reaction temperatures and meet the application requirements of low-stability templates such as metal-organic frameworks.
(5) Promote scaled-up preparation and product separation: Address the current issues of low product yield and difficult separation/purification in confined synthesis; develop continuous, large-scale confined synthesis processes; optimize template removal and product separation methods; improve product purity and yield; and lay the foundation for the practical application of confined synthesis technology.
MakarovaT, PalacioF. Carbon Based Magnetism:An Overview of the Magnetism of Metal Free Carbon-Based Compounds and Materials. Amsterdam, The Netherlands; San Diego, CA: Elsevier, 2006.
[23]
KanE J, LiZ Y, YangJ L, HouJ G. J. Am. Chem. Soc., 2008, 130(13): 4224.
PDF
