Carbon/silicon (C/Si) heterojunction solar cells have recently achieved power conversion efficiencies exceeding 23% for small-area (6 cm²) devices. The development pathway toward these results highlights the critical requirements for engineering efficient contacts and heterojunction structures in next-generation photovoltaic technologies. If current performance trends continue, C/Si heterojunctions could provide a viable route to affordable, high-efficiency solar cells. This review examines the main carbon allotropes explored for C/Si junctions, that is, amorphous carbon (a-C), C₆₀ fullerene, graphene, and carbon nanotubes (CNTs) and evaluates their optoelectronic properties, deposition strategies, and device performance. The a-C approach enables low-temperature, scalable processing but suffers from high defect densities and limited carrier mobility. Fullerene derivatives offer favorable energy-level alignment for electron extraction, yet their low conductivity and photochemical instability remain obstacles. Graphene provides reasonable optical transparency, mechanical flexibility, and a tunable work function; however, its relatively high sheet resistance, parasitic optical absorption, and interface recombination cast doubt over its use in high-efficiency solar cells. By contrast, CNTs combine outstanding electrical conductivity, one-dimensional carrier transport, and simple solution-based fabrication, supporting efficient charge separation and extraction when integrated with optimized passivation layers. As a result, CNT/Si heterojunctions consistently outperform other carbon-based architectures, positioning CNTs as the leading candidate for future low-cost, high-efficiency C/Si photovoltaic devices. The review concludes by outlining research priorities in interface optimization and scalable large-area processing.