Context. Thermal instability plays a major role in condensation phenomena in the solar corona, e.g. for coronal rain and prominence formation. In flare-relevant current sheets, tearing instability may trigger explosive reconnection and plasmoid formation. However, how both instabilities influence the disruption of current concentrations in the solar corona has received less attention to date. Aims. We explore how the thermal and tearing modes reinforce each other in the fragmentation of a current sheet in the solar corona through an explosive reconnection process, characterized by the formation of plasmoids which interact and trap condensing plasma. Methods. We use a resistive magnetohydrodynamic (MHD) simulation of a 2D current layer, incorporating the non-adiabatic effects of optically thin radiative energy loss and background heating using the open-source code MPI-AMRVAC. Multiple levels of adaptive mesh refined grids are used for achieving a high resolution to resolve the fine structures during the evolution of the system. Results. Our parametric survey explores different resistivities and plasma-β to quantify the instability growth rate in the linear and nonlinear regimes. We notice that for dimensionless resistivity values within 10 −4 − 5 × 10 −3 , we get explosive behavior where thermal instability and tearing behavior reinforce each other. This is clearly below the usual critical Lundquist number range of pure resistive explosive plasmoid formation. We calculate the mean growth rate for the linear phase and different non-linear phases of the evolution. The non-linear growth rates follow weak power-law dependency with resistivity. The fragmentation of the current sheet and the formation of the plasmoids in the nonlinear phase of the evolution due to the thermal and tearing instabilities are obtained. The formation of plasmoids is noticed for the Lundquist number (S L ) range 4.6 × 10 3 − 2.34 × 10 5 . We quantify the temporal variation of the plasmoid numbers and the density filling factor of the plasmoids for different physical conditions. We also find that the maximum plasmoid numbers scale as S 0.223 L .Within the nonlinearly coalescing plasmoid chains, localized cool condensations gather, realizing density and temperature contrasts similar to coronal rain or prominences.