Determination of boiling point distribution by TG using a novel crucible and comparison with true boiling point distillation and simulated distillation by GC–MS and Py-GC/MS

To effectively guide the processing and utilization of oils, their true boiling point distillation (TBPD) data must be determined accurately and simply. In this investigation, the experimental distillation of crude oil (RP), low-temperature coal tar (MT), and high-temperature coal tar (HT) were investigated by using the TBPD apparatus, simulated distillation based on gas chromatography-mass spectrometry (GC–MS) and Py (pyrolyzer)-GC/MS, and thermogravimetry with a self-modified crucible (TG-Distillation). GC–MS calibration curves were calibrated using alkane and aromatic external standards. Several parameters for simulated distillation were calibrated, and the dependence of alkane and aromatic components on column temperature was examined. An in-depth analysis and comparison were conducted on TBPD, GC–MS, Py-GC/MS, and TG-Distillation methods. The results show that the difference in TBPD between GC–MS and Py-GC/MS is less than 10 %. Aromatics have a stronger temperature-dependence than alkane components, and the column heating rate of 5 °C/min is better than 10 °C/min. The resolution of oils (especially aromatic-rich coal tar) by GC–MS and Py-GC/MS can be effectively improved by calibrating the components with multiple parameters. TG-Distillation can achieve good distillation and separation effects comparable to that of GC–MS after adjusting the reflux ratio. At the heating rate of 5 °C/min, the TG-Distillation effect is close to the results of TBPD, especially the RP, with a difference of less than 10 %. Adjusting the height of the crucible lid achieves a wider boiling point distribution examination. Additionally, a conversion method was employed to accurately convert TG linearizable data into data from other simulated distillation methods, and predict boiling point distribution up to 400 °C with a prediction deviation of less than 5 % (R2 > 0.98).