Due to their physical properties, neutrons are an excellent probe for the investigation of matter in different scientific fields, such as physics, chemistry and biology as well as for specific medical and industrial applications. Neutron detection is usually achieved via nuclear capture reactions, where the neutron is absorbed by the nucleus of an atom, which decays into two heavy charged particles. These reactions only occur with significant cross-section for a few isotopes and the ones of practical interest for detection applications are, by decreasing cross-section, 3He, 10B and 6Li. Until recent years, proportional counters filled with 3He gas were considered the golden standard for neutron detection. However, when a severe shortage of this gas was acknowledged, prices skyrocketed and heavy acquisition restrictions were implemented, which urged to pursue alternative technologies. Consequently, over the last decade, a lot of effort and investment was put into the development of 3He-free neutron detectors for a wide range of applications. Gaseous detectors equipped with boron layers, deposited on the inner walls of the detector or on substrates that are then inserted into it emerged as the most obvious alternative. Due to momentum and energy conservation, the reaction products of the 10B neutron capture are emitted along the same line, in opposite directions. Consequently, in conventional boron coated detectors, for each neutron capture, only one of the reaction products can travel towards the gas to generate a signal in the detector, while the other is absorbed by the boron layer or the substrate. Furthermore, depending on the depth in which the nuclear capture, the range of the α particles in typical gases used in gaseous detectors at atmospheric pressure can extend up to about 10 mm, which intrinsically limits their spatial resolution. In this work, we propose an alternative approach that aims at simultaneously detecting both secondary products of neutron capture reactions which can be achieved if thin enough converter and substrate layers are deployed. Monte Carlo simulations were developed to validate the detection concept and to optimize its geometry by computing the detection efficiency as a function of substrate and converter thicknesses. A 0.9 µm Mylar foil was stretched over an aluminium frame obtaining a smooth 100×100 mm2 effective surface, suitable for boron carbide (B4C) deposition. Considering this substrate, simulation results indicate that a 1.5% detection efficiency for thermal neutrons can be reached with a total 1 µm thick B4C (99% enriched) layer, equally distributed over both sides of the substrate surface. Although this value is inferior to that of conventional boron-based detectors employing a thick conversion layer (~4.5%), it offers the advantage of allowing a more precise estimation of the interaction site for each detected neutron, which leads to an improvement of spatial resolution.