The molecular composition of petroleum acids in the ultra-high acid crude oil and its fractions were characterized by Fourier transform ion cyclotron resonance mass spectrometry(FT-ICR MS).This aimed to investigate the static corrosion behaviors of four typical steels used in petroleum-refining equipment,such as Q345R,lCr5Mo,304L and 316L,in the oil products with different total acid numbers(TAN),and then establish a prediction model for high temperature acid corrosion.The results show that the TAN of crude oil is 10.403 mg KOH/g,and the O2 class species(petroleum carboxylic acids)with two oxygen hetero-atoms per molecule are the main components of petroleum acids,dominated by bicyclic and tricyclic naphthenic acids and monocyclic aromatic carboxylic acid.The static coupon corrosion test for 168 h indicates that the acid corrosion rates of four steels increase with the increasing of temperature,and the up trends gradually slow down when the temperature is higher than 280 ℃.There is a significant difference in the corrosion micro-morphology of different materials.Q345R presents an overall corrosion morphology and dense pits are formed on the surface of 1Cr5Mo.Only a small amount of superficial corrosion pits are distributed on the surface of 304L,and metallic luster can still be seen on 316L.Based on corrosion rates and considering the combination between aliphatic acid,naphthenic acid,aromatic acid and metals,the molecular composition of petroleum acids is defined as a modification item in corrosion prediction model.The model parameters are fitted by Levenberg-Marquardt(LM)algorithm.For this model,the correlation coefficient R2 is higher than 0.950,and the root mean square error(RMSE)is lower than 0.089.The model can realize the early scientific prediction of corrosion for ultra-high acid crude oil,and thus provide a support for guiding the reasonable material selection and making a check and maintenance plan during the operation of oil processing equipment.
corrosionnaphthenic acidFourier transform ion oyclotron resonance mass spectrometryultra-high acid crude oilprediction
万方数据
维普
