• How can I know the water mole fraction in the overhead stream in the CDU? I need it to know the optimum temperature of the top reflexes.



  • Marcio Wagner da Silva, Petrobras, marciows@petrobras.com.br

    I believe that you can carry out a mass balance using the stripping steam flow rate and eventually another water injection to the atmospheric column to determine the water concentration in the overhead system indirectly, if you have a safe and adequate sample point you can make a chemical analysis to determine it. But it's important considering that the corrosion in the overhead system of the atmospheric tower is strictly related with the performance of the desalting system.

    According to the characteristics of the processed crude, the performance of desalters needs to be optimized, especially considering the concentration of magnesium and calcium chloride salts which tends to suffer hydrolysis and generate HCl which will concentrates in the overhead system of the atmospheric column, this is a special concern to refiners processing heavy crudes, slop blendend with crude oil, and opportunity crudes. The adequate reflux temperature is fundamental to control the corrosion and fouling in the top plates of the atmospheric column which is caused by low reflux temperature and presence of NH3, HCl and amines which are absorbed by the water which vaporizes along the downward of the reflux leading to the salt fouling in the top plates which causes severe pressure drop leading to poor fractionating performance and can limit the the operating cycle of the processing unit. 

    Some refiners adopt the amine injection in the overhead systems to control the salts precipitation, especially H2S scavengers but the side effect here is the trend of deposition of corrosive salts under low temperature.

    The literature highlights that the design of overhead systems needs to consider the probability of the corrosion and fouling in the top section of the atmospheric column, if the probability is high a overhead system with two separating vessels needs to be considered once to avoid low reflux temperature which can cause water condensation inside the column. A very good reference about this topic is the article published by Mr. Tony Barletta and Mr. Steve White in the Q3 2007 issue of PTQ Magazine, which you can read HERE.