Reactive semivolatile contaminants, such as isocyanates, can be particularly difficult to assess in occupational environments. While standard methods exist for isocyanates, there are still significant differences between the results they provide for various occupational environments or processes. This study presents the validation of a laboratory system for the generation of controlled atmospheres of isocyanates. A system consisting of different modules generated airborne methylene diphenyl diisocyanate (MDI) by nebulizing a solution into mixing and exposure chambers with control of flow rate, temperature, and relative humidity. Sampling was performed through an eight-port flow splitter that allowed only very slight within-test variability. MDI was measured using the Asset EZ4-NCO® and a modified version of the Iso-Chek® sampling system. MDI specific particle-size distribution was measured by a Marple Sierra cascade. Aerosol real-time monitoring was performed using a condensation particle counter, an electrical low-pressure impactor (ELPI+), and an aerosol optical spectrometer, providing additional information on system stability and particle-size distribution of the generated aerosol. The system was able to generate MDI concentration levels ranging from 4 to 233 μg m⁻³, with a steady-state level reached within 5 minutes, and with well-documented intra-test and inter-test variability (RSD of 4% and 15%, respectively). Accuracy and representativeness of MDI data were confirmed by the agreement between MAMA and Asset EZ4-NCO (used as reference), with a mean bias of 3%. Using the Asset EZ4-NCO capability, the vapor-particle partitioning of MDI was evaluated to be 8% and 92%, respectively, at a concentration ranging from 20 to 25 μg m⁻³. The system may therefore be used for exhaustive method intercomparison studies and could also be adapted to generate other emission types of semivolatile compounds.