This review discusses the possibility of using electrical conductivity to characterize the structure of low-temperature carbon, which is obtained by carbonizing organic materials at temperatures below 1000 °C. Such carbons are attractive materials due to their potential to develop various functions by controlling the carbonization process. A knowledge of the structures of the low-temperature carbons is essential in making them functional carbon materials; however, there are serious difficulties in analyzing their structures by conventional analytical methods. Instead, the use of electrical conductivity is proposed as a method to obtain the structural information about the carbons. This idea was validated by reviewing the general concept of electrical conductivity, the structural models deduced from the changes in electrical conductivity during carbonization, and the factors influencing the intrinsic electrical conductivity of powder carbons. Electrical conductivity is concluded to be a good characterization tool for low-temperature carbons.

Graphene, single-wall carbon nanotubes, and porous carbons have zero-dimensional nanowindows, one-dimensional (1D) pores, and three-dimensional pores consisting of two-dimensional (2D) unit pores, respectively. A molecular dynamics simulation study on nanowindows shows that they enable the efficient separation of O₂ from N₂ and Ar at an extremely high permeance. Intense confinement of sulfur atoms in 1D-carbon nanotube pore spaces in vacuo leads to the formation of metallic 1D-sulfur chains. Evidence of the partial breaking of the Coulombic law for ionic liquids confined as monolayers in 2D pores, which is caused by an evident image charge effect of conductive carbon-pore walls, was provided by hybrid-reverse Monte Carlo simulation-aided X-ray scattering. The solid-like structural formation of oxygen molecules in 2D carbon pores induces highly efficient and rapid adsorption separation of ¹⁸O₂ from ¹⁶O₂ near 112 K, which is the boiling point of methane. Abundant water vapor can be adsorbed on microporous carbon through cluster-associated hydrophobic-hydrophilic transformations. Water molecules form aggregated clusters on the adsorbed branch, which transforms into a continuous structure at a relative pressure of 1, giving rise to obvious adsorption hysteresis. Logical and challenging studies to improve our fundamental understanding, along with new approaches, produce completely new findings, even for carbon materials that have been thoroughly studied.

Carbon materials are mainly composed of giant polycyclic aromatic molecules. The differences in the size and shape of such macromolecules and the way they assemble lead to an immense variety of structural possibilities, which allows carbon materials to be used for a surprisingly wide range of applications. It is therefore generally recognized that control of the carbon structure not only at the molecular level but also at the nanometer and micrometer levels is indispensable for both the understanding of carbon materials and the development of high-performance carbons. Nevertheless, our approach to carbon research is rather different from the general methodology. Almost all of our efforts have been directed toward the control of carbon structure only at the molecular level. This review introduces such a molecular level approach to our carbon research such as carbon gasification, template carbonization techniques and the understanding of carbon structure and physicochemical properties in terms of carbon edges sites.

Attractive functions of nanocarbon materials are derived from the local structures such as edges or functional groups. To maximize the function of nanocarbon materials, it is important to introduce and maintain such local structures abundantly in the synthesis of nanocarbon materials. In this study, the manufacturing process of the bottom-up nanocarbon material that can control the local structures was developed, and the applications were further proposed.