BOREAM, a detailed model for the gas-phase oxidation of α-pinene and its subsequent formation of Secondary Organic Aerosol (SOA), is tested against a large set of SOA yield measurements obtained in dark ozonolysis experiments. For the majority of experiments, modelled SOA yields are found to agree with measured yields to within a factor 2. However, the comparisons point to a general underestimation of modelled SOA yields at high temperatures (above 30 °C), reaching an order of magnitude or more in the worst cases, whereas modelled SOA yields are often overestimated at lower temperature (by a factor of about 2). Comparisons of results obtained using four different vapour pressure prediction methods indicate a strong sensitivity to the choice of the method, although the overestimated temperature dependence of the yields is found in all cases. Accounting for non-ideality of the aerosol mixture (based on an adapted UNIFAC method) has significant effects, especially at low yields. Our simulations show that the formation of oligomers through the gas-phase reactions of Stabilised Criegee Intermediates (SCI) with other molecular organic products could increase the SOA yield significantly only at very low relative humidity (below 1%). Further tests show that the agreement between model and measurements is improved when the ozonolysis mechanism includes additional production of non-volatile compounds.