Carbon is one of the important basic elements in nature, and carbon materials with carbon as the main constituent element play an irreplaceable role in human society. The synthesis of new carbon materials and the study of their properties have greatly promoted the development of science. Especially in the past four decades, the discovery and application of new carbon materials such as fullerene (1985), carbon nanotube (1991), graphene (2004) and graphyne (2010) have made great progress in the fields of optoelectronics, information and energy (Figure 1)^{[1,2][3,4][5]}. In view of the significant contribution of this new class of nanocarbon materials to the development of science, the discovery of fullerene and graphene won the Nobel Prize in 1996 and 2010, respectively. However, most carbon materials obtained by physical synthesis, such as graphene and carbon nanotubes, still have differences in length, width, and even chemical structure between their molecules, and there is no reliable and convenient separation method to purify them^[6]. Therefore, scientists have set their sights on the means of synthetic chemistry. Based on the "bottom-up" organic synthesis pathway, nano-carbon molecules with precise structure are gradually constructed and used as template molecules. The precise construction of carbon materials is achieved by developing the expansion reaction of the conjugated system of template molecules (Fig. 7). Among them, nanocarbon molecules are conjugated structures with nanometer size and atomic-level accuracy, mainly composed of sp² carbon atoms. The synthesis of nanocarbon molecules with novel topologies is very challenging research. By designing and adjusting the molecular structure of nanocarbon, a clear structure-activity relationship can be obtained, and then its physical and chemical properties can be regulated, which will also strongly promote the development of new carbon materials^[7,8]. As one of the most promising new carbon materials in the future, carbon nanomaterials have made rapid progress in this field in recent years^[9]. Especially in the last fifteen years, organic synthesis techniques have achieved important applications in this field. In this paper, from the perspective of the structure and synthesis of carbon nanomolecules, the latest research progress of carbon nanomolecules will be introduced from four parts: carbon nanoribbons, negative curvature carbon nanorods, cyclic carbon nanorods and other structures.

Graphene is a two-dimensional material with the thickness of a single layer of carbon atoms, which is based on a hexagonal honeycomb lattice formed by carbon atoms with sp² hybrid orbitals, while carbon Nanoribbons (GNRs) are Graphene carbon nanoribbons with a width of less than 10 nm, which is a new type of carbon nanomaterials with quasi-one-dimensional structure^[10,11]. GNRs synthesized precisely at the atomic level have special electronic structure, magnetic edge state and carrier transport properties. The research results on their synthesis methods and physical properties have been published in Science and other journals for many times, and have attracted wide attention from scientists in various fields around the world^[12⇓~14].

GNRs can be divided into zigzag-edge, armchair-edge, cove-edge and gulf-edge according to their length, direction and edge structure (Fig. 2), and these edge types are also one of the main factors determining the properties of GNRs. Calculations made by early scientists based on the tight-binding approximation model predicted that the zigzag type had metallic bond properties. The armchair type has metallic bonding properties or semiconducting properties, and its specific properties depend on the bandwidth of the nanoribbon^[13]. The edge configuration of GNRs not only affects their electrical, magnetic and chemical properties, but also their width and number of layers^[15,16].

At present, there are two main strategies for the preparation of GNRs: "top-down" (etching graphene or shearing and exfoliating carbon nanotubes) and "bottom-up" (building from molecular modules by organic synthesis and chemical vapor deposition)^[17]. The "top-down" approach does not allow precise control of the molecular structure, which in turn leads to uncertainty in the width and edge structure of GNRs. In order to precisely control the physical and chemical properties of GNRs, the "bottom-up" organic synthesis strategy can be used to control their structures to atomic level accuracy, and then to realize the synthesis of GNRs with specific widths and edge structures. Many simple and efficient strategies for organic synthesis have been developed in the past two decades. For example, monomers with substituents or defined edge structures are generally designed first, and then linear polymer precursors are prepared through coupling reactions such as Yamamoto or Suzuki-Miyaura. Finally, GNRs with various sizes, symmetries and edge structures were synthesized by intramolecular-based oxidative dehydrocyclization (Scholl reaction). In 2008, M Müllen et al. Used 1,4-diiodo-2,3,5,6-tetraphenylbenzene and hexaphenylbenzene diborate as monomers to obtain polymer precursors through Suzuki-Miyaura coupling reaction, and then used FeCl₃ as an oxidant to realize the intramolecular oxidative dehydrocyclization of the polymer precursors, pioneering the synthesis of armchair GNRs (Fig. 3)^[18]. In 2012, they reported the synthesis of GNRs based on Yamamoto coupling reaction, and obtained laterally extended GNRs with absorption range extending to the near-infrared (NIR) region (Fig. 4)^[19]. There are still some problems to be solved in the organic synthesis of GNRs, such as the complexity of the synthesis of GNRs precursors, the solubility of GNRs and the structural defects caused by the side reactions in the Scholl reaction.

Allotropes made entirely of carbon atoms hybridized with sp² can exhibit different surface structures, either lying flat like a carpet or bending like a bowl or saddle. The curvature of this surface reflects the overall geometric properties of the nanocarbon structure. Among them, "graphene composed entirely of hexagonal carbon skeleton is a flat surface with zero curvature;" Fullerenes built from hexagonal and pentagonal carbon skeletons are closed spherical structures with positive curvature. The introduction of heptagonal or octagonal carbon skeletons into the hexagonal network results in a saddle-shaped surface with "negative" curvature^[20]. Mackay crystal is the first three-dimensional periodic carbon structure with negative curvature (Fig. 5), which has been challenged by synthetic scientists as a new carbon allotrope^[21]. At present, the exploration of Makai crystal synthesis is mainly based on the "bottom-up" strategy, which introduces seven-membered or eight-membered rings into polycyclic aromatic hydrocarbons to realize the synthesis of nano-graphene fragments with negative curvature, and then uses them as templates to construct Makai crystals.

In the past ten years, remarkable progress has also been made in the synthesis of nanographene with negative curvature, in which Miao Qian et al., Itami et al., Wu Yaoting et al^{[22⇓~24][25,26][27][28][29]}. Miao Qian et al. Have made detailed reviews focusing on negative curvature nanographene containing seven-membered rings and eight-membered rings respectively^[30][20]. Here, we briefly introduce the relevant synthesis strategies with emphasis on synthesis.

The synthetic idea of negative curvature nanographene is mainly to fuse more benzene rings or larger aromatic structures on the periphery of the basic structural unit, which is mainly composed of condensed aromatic hydrocarbons with seven-membered rings or eight-membered rings. In 2018, Miao Qian et al. Reported a saddle-shaped nanocarbon composed of 86 sp² carbon atoms (Figure 6A)^[23]. They first used a condensed ring containing a seven-membered ring as the basic unit, and then extended the carbon skeleton through Diels-Alder reaction and Scholl reaction. Crystal diffraction analysis shows that the naphthalene moiety of the molecular core has a bend of up to 77.2 °, which is significantly higher than the 42.5 ° bend observed in the naphthalene moiety of the C₆₀. In 2019, Miao Qian et al. Carried out Diels-Alder cycloaddition reaction with 1,3-diarylbenzofuran by using fused ring containing eight-membered ring (containing naphthalenediyne) as basic unit, and then successfully obtained saddle-shaped polycyclic skeleton by aromatization (Fig. 6B)^[22]. The above two typical synthetic strategies are to start from the basic structural unit containing a seven-membered ring or an eight-membered ring, and then use the in-out strategy to "stitch" the benzene ring outside the basic structural unit, so as to obtain the negative curvature graphene.

in addition, there is another ingenious synthesis strategy of negative curvature nanographene, namely the out-in strategy, which is to synthesize macrocycles first, and then realize the construction of seven-membered or eight-membered ring cores in the final synthesis step. In 2017, Miao Qian et al. Cleverly used the strategy of inward cyclization to synthesize [8] cycloalkene derivatives composed of 96 sp² carbon atoms^[31]. The precursor macrocycle with a hexaarylbenzene moiety is formed by two Diels-Alder reactions of cyclopentadienone with macrocyclic diyne and decarbonylation. Finally, a polycyclic twisted skeleton with an eight-membered ring as the inner core and fused with hexabenzene is formed by cyclization and dehydrogenation via Scholl reaction (fig. 6C).

Negative curvature nanographene usually exhibits higher solubility, larger HOMO-LUMO band gap, higher fluorescence quantum yield and red shift of absorption spectrum than planar nanocarbon with similar size due to its more flexible polycyclic skeleton^[32]. In view of this, the study of negative curvature nanographene will certainly become an important part of the development of new nanocarbon materials. Of course, the development of more efficient and universal synthesis strategies is also a major challenge for the synthesis of nanographene with negative curvature.

Carbon Nanorings (CNRs) and Carbon Nanobelts (CNBs), as the ring structural units of Carbon Nanotubes (CNTs), can be used as template molecules to control the precise synthesis of CNTs^[33]. In the past 15 years, carbon nanorings and carbon nanoribbons have also been precisely synthesized by means of organic synthesis, laying the foundation for the precise synthesis of carbon nanotubes^[34].

In recent years, carbon nanomolecules with other novel topologies have emerged in an endless stream, such as super helicene, carbon nanocone/bowl, thiacarbon nanoribbon and other structures (Fig. 12)^{[58][59,60][61]}. In 2018, Wang Jiaobing et al. Reported the synthesis and characterization of sextet [7] helicene^[58]. In addition to the interesting topology of superhelicene, the synthesis of nanocone topology has also made a major breakthrough, and Xie Suyuan/Zhang Qianyan et al. First synthesized carbon nanocone molecules with well-defined configuration under mild reaction conditions^[60]. Tan yuanzhi et al. Constructed carbon nanocone molecules with holes through cyclic aromatic hydrocarbons^[59]. In addition, nanoribbon molecules doped with heteroatoms have also been successfully synthesized, and Zhu Kelong et al. Reported the synthesis of a series of thiacarbon nanoribbons with thianthrene building blocks^[61]. These new nanocarbon molecules have been applied in the fields of optoelectronic materials, supramolecular recognition and living cell imaging technology^[62].

This review summarizes the recent developments in the field of nanocarbon molecules with novel topologies. These new nanocarbon molecules exhibit unique physicochemical properties, which are essential for the future development of functional materials in nanotechnology, electronics, optics, and biomedical applications. The emergence of various new nanocarbon molecules has fundamentally changed the scientific landscape and has been regarded as a revolutionary material in the future. Although the field of nanocarbon has made great progress in the past two decades, challenges still exist, and there are still two directions to be solved urgently: obtaining nanocarbon molecules with uniform structure is still a huge challenge in the field of nanocarbon science.The tedious synthesis methods still hinder their large-scale application, and how to accurately and efficiently synthesize the next new nano-carbon molecule with wide influence has always been one of the goals pursued by chemists. In a word, the discovery of various new carbon nanostructures and the study of their unique properties will have very attractive application prospects in materials science, information science, biomedicine and other fields.