You are currently viewing: Articles



Apr-2019

Predicting middle distillate properties

Development of on-stream correlation models to predict cloud point, flash point and freezing point and increase production of on-spec jet fuel using a single analyser

Madi ASIRI
Saudi Aramco Total Refining and Petrochemical Company (SATORP)
Viewed : 1169
Article Summary
We cannot optimise what we cannot measure continuously and we cannot increase refinery margin without continuous process optimisation. This study aims to find a simple and cost-effective way, with no capital investment, to measure and predict continuously the physical properties of middle distillate streams without the need for additional on-stream analysers. The first part of the study focuses only on hydrocracking middle distillate products (diesel and kerosene) while the second part will deal with the other middle distillate streams (CDU, DCU, and so on). The model developed in this study can be integrated with a DCS (or APC) system to minimise product giveaway and maximise the production rate of middle distillate within final specification limits.

Based on a deep understanding of hydrocracking process operation, catalyst function and reaction chemistry, we started our work with two differing hydrocracking units: a two-stage Isocracking unit (TSREC) with a name plate capacity of 59 000 b/d and 97% true net conversion; and a single-stage once-through Isocracking unit (SSOT) with a name plate capacity of 59000 b/d and 50% true net conversion (see Figure 1).

Cold flow properties, cloud point and freezing point are controlled by the reaction section. In particular, the amount of n-paraffins in the reactor effluent, which has the biggest impact on middle distillate cold flow properties, is controlled by severity of operation in the reaction section (WABT, H2PP, LHSV and catalyst type). However, in a hydrocracking process unit, as part of the catalyst selection and catalyst performance guarantee, yields and product quality must be guaranteed over the life cycle of the selected catalyst.

In other words, the chemical structure of the reactor effluent (paraffins, iso-paraffins, naphthenes and aromatics) over the cycle life of the catalyst must show absolutely minor changes to ensure product quality from start of run conditions to end of run conditions for the selected catalyst.

Our first assumption is that, since the internal chemical structure of middle distillate shows minor structural changes over the catalyst cycle life, we can assume that both cloud point and freezing point are functions of the amount of heavy hydrocarbons in the middle distillate; the heavy part of the middle distillate will impact both cloud point and freezing point. However, since D86 data can give us the required information about the middle distillate streams, how heavy or light they are and how tight or wild is the middle distillate cut, there should be a relationship with the distillation data (ASTM D86) and with both cloud point and freezing point in the hydrocracking unit. Hence, both cloud point and freezing point can be controlled in day-to-day operations by adjusting the fractionation section’s operating conditions, keeping in mind that operations in the reaction section are almost steady and not frequently changed. Usually, the reaction section’s severity is set to meet the nitrogen slip of the pretreat reactor and the global conversion of the cracking reactors during the catalyst life cycle with normal acceptable deactivation rate.

On the other hand, flash point is a function of the amount of light hydrocarbons (light ends) in the middle distillate, which can be controlled in the hydrocracking fractionation section by adjusting the operating conditions.

Based on all of the above assumptions and information, linear models were developed to predict the physical properties of hydrocracking middle distillate. The target properties are cloud point, flash point and freezing point. A total of eight correlations were developed for four product streams in two different hydrocrackers within the refinery. These correlations are:
•    Flash point and freezing point for kerosene in both TSREC and SSOT
•    Flash point and cloud point for diesel in both TSREC and SSOT.

These models are tools to allow for process optimisation to maximise the middle distillate production rate and minimise product giveaway within product specifications by continuous measurement at the operators’ panels. These correlation models will add more value to process optimisations and product quality control. Instead of waiting for lab results once a day (or in some cases twice weekly), the operations team will have continuous data and information about cloud point, flash point and freezing point to enable them to take immediate action and not wait for the next day’s results to see the impact of today’s actions on product quality.

Cloud point is the lowest temperature at which wax crystals begin to form by gradual cooling under standard conditions. However, this parameter is an important property of the fuel since the presence of solidified waxes can block filters and negatively impact engine performance. As the molecular weight rises, the cloud point is also incremented at low temperature conditions. Furthermore, cloud point is a measure of paraffin content in wax form, which is why no cloud point data are reported for light cuts like naphtha or gasoline.  

Freezing point is the temperature at which a liquid changes to a solid by cooling. Jet aircraft are frequently exposed to low operating temperatures and it is essential that their fuels do not freeze in these environments. Plugging of filters and related operational problems in the fuel system are dependent on freezing point; for this reason, jet fuel specifications include requirements for maximum freezing point.

Flash point is the lowest temperature at which vapours above a volatile, combustible substance ignite in air when exposed to flame. The lower the flash point, the easier it is to ignite the vapour if an ignition source is present. The higher the flash point, the safer the material is to handle. This means that the temperature limit to achieve safe storage is below flash point. A higher vapour pressure corresponds to a lower flash point.

Objective
The aim of the study is to develop on-stream optimisation tools (models) to enable refiners to measure physical properties in hydrocracking units continuously, and to have the opportunity for continuous process optimisation by utilising a distillation on-stream analyser (ASTM-D86). Most refiners depend on lab results to measure these additional properties, a very slow process which reduces the opportunities for process optimisation, or they need to have an individual analyser for each property they want to measure (see Table 2), which involves a high capital investment cost in addition to high operating costs. However, it is not common to have so many analysers installed in a single hydrocracking unit. The optimisation models should aim to:
• Continuously optimise middle distillate production
• Minimise middle distillate product giveaway.
• Maximise the middle distillate production rate.

Data analysis and results
Building any model for middle distillates to measure additional physical proprieties demands a deep understanding of process operations and what impact dependent and independent variables have on the target properties. For instance, in studying flash point the most important variable is the amount of light hydrocarbons (light ends) in the product cut. To study cloud point or freezing point, the important variable is the amount of heavy hydrocarbons (heavy ends) in the product cut. However, for every model certain stages must be passed to be able to achieve a proper conclusion and results:
1. Collecting lab historical data
2. Finding the relationships of dependent variables
3. Distillation (D86) analyser calibration and validation
4. Defining the on-stream model’s dependent variables
5. On-stream model validation.
Current Rating :  4

Add your rating:



Your rate: 1 2 3 4 5