(560hd) Development of an Accelerated Deactivation Protocol for Vacuum Gasoil Hydrocracking Catalysts

Authors: 
Dubreuil, A. C., Institut Français du Pétrole (IFP)
Dubreuil, A. C., Institut Français du Pétrole (IFP)

color:#1F497D">INTRODUCTION

justify;line-height:150%">Hydrocracking is one of the most flexible refining processes
for converting heavy,
high-boiling feedstock molecules to smaller, lower boiling products. This
process is carriedout using hydrogen at high pressure conditions in
presence of heterogeneous bifunctional catalysts, combining an acid and a
metallic function.  This catalyst  is continuously deactivated in the presence
of hydrocarbon feedstocks at hydrocracking conditions. Deactivation at normal
operational conditions occurs slowly, depending on the feed and the operation
severity; cycle lengths can vary between a few months and a couple of years.
The heavier the feedstock the more severe the deactivation. The loss of activity
determines the time of an operation cycle and consequently the profitability of
the process. This is due to the fact that the i " arial>ncrease of cycle length reduces the number of catalyst
replacements and consequently the downtime of the unit. Deactivation is
inevitable, but it can be slowed down or prevented and some of its consequences
can be avoided or delayed. In that sense an obvious action is the purification
of the feed to prevent poisoning. One common industrial practice is to gradually
increase the operating temperature to compensate the activity loss. A different
option could be to reduce the deactivation rate through catalyst stability
improvement. The knowledge of the chemical and physical aspects of catalyst
deactivation is fundamental for the design of more stable catalysts. For this reason, one of the main
concerns in the hydrocarbons processing is the study of catalyst  deactivation,
for which different approaches exist. Some authors have evaluated the effects
of coke or metal deposition through the characterization of industrial spent
catalysts. However, this
approach was not proved to give enough insight on deactivation phenomena due to
the changing-character of the industrial operation. As mentioned above, deactivation
in an industrial plant takes place over a period of time that makes impractical
to do absolutely equivalent experiments in the laboratory.   Additionally  they
are expensive and high time-consuming and, depending on the sampling program
and the analyzes carried out,  would not provide much more information than a
full scale operation. The accelerated deactivation appears as an instrument
capable of providing relevant information on the deactivation phenomenon in
reduced time. Despite of
the importance of this subject,  it is hardly discussed in the literature for
hydrocracking catalysts. Most of deactivation studies are focused on catalysts
used for residues hydrotreatment, where the main sources of poisoning are the
deposition of asphaltenes and metals contained in these heavy fractions. In the
case of  vacuum distillates, these two poisons are much less abundant than in
residues; therefore, they may not be the main sources of deactivation. Some
other studies have been carried out with model feedstocks, whose behavior is
not representative of real feedstocks.

normal"> 

normal"> 

color:#1F497D">OBJECTIVES AND METHODOLOGY        

color:#1F497D">                                                                   
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   

The purpose of this work is to develop
an accelerated deactivation experimental procedure for a vacuum gasoil
hydrocracking catalyst, that allows to assess this phenomenon in a short period
of time, compared to the one in an industrial unit. The experiments were
carried out in a fixed bed pilot unit. The feed was a vacuum gas oil for
studying the effect of real compounds that are typically in contact with the
industrial catalyst. The catalyst is constituted of a nickel-molybdenum
metallic phase dispersed on Y zeolite. The methodology consists of two main
steps: the first one is a preliminary screening of the variables that impact
the deactivation rate. Based on the literature these variables are:
temperature, space velocity, hydrogen/hydrocarbon ratio, and organic and total
nitrogen content. This step consist on submitting the catalyst to the following
conditions: 403°C, 140 bar, H2/HC: 1500 NL/L, LHSV: 2h-1,
total nitrogen content 2200 ppm and an organic nitrogen content of 150 ppm.
This set of conditions  constitutes the reference case. Subsequently each one
of this variables is changed once at a time and catalyst is under each new
condition for 10 days. After each variable change, conditions goes back to the
reference case. Measurement of deactivation is carried out by tracking the
conversion comparing to the conversion of the last point at reference
conditions. According to these results we can determine what variables have the
bigger impact on the deactivation rate.

Once the preliminary screening was finished,
the operating conditions for a standard protocol were selected. In order to
validate this protocol, the catalyst was submitted to these conditions until
the conversion decreases between 30 to 50%, in a maximum period of one month. This
loss of conversion is representative of the typical increase in temperature in
a commercial unit between the start and the end of the run, which is around 20
°C. At the end of the test  temperature was increased until the initial
conversion is achieved. In that way, it was possible to determine the
deactivation in terms of a delta temperature, the common practice in industrial
scenarios. 

During the test, catalyst samples were
taken at different bed locations and in the middle and at the end of the test
run, in order to assess the evolution of the catalytic properties with time on
stream and reactor location. An
experimental procedure for catalyst sampling without catalyst modification and catalyst
reloading by keeping the same conversion as the one immediately obtained before
catalysts unloading for sampling purposes was developed.  In the best of our knowledge there are
no publications on accelerated deactivation experimental procedures for
hydrocracking catalyst with real feedstocks nor do they consider the evolution
of  the catalyst properties with time and location inside the reactor.

Respect to the laboratory analysis, concerning
the catalyst these are mercury porosimetry (to evaluate changes in
porosity, establishing which pores are more sensitive to the variables that are
going to be modified), nitrogen isotherms (to establish the loss of active area
after catalyst deactivation), elemental analysis (to quantify the H/C ratio
which gives an idea about the aromaticity of the sample, type of coke deposited), 
TGA (to measure number and strength of acid sites), HTA to evaluate the changes
in the metallic phase and eventually mass spectroscopy to quantify known
materials, to identify compounds and its structure, infrared spectroscopy and
eventually STEM (to characterize the morphology and surface of solid materials,
occurrence and location of zeolites embedded in the matrix, size of supported
metal particles and changes in their size, shape, and location during catalyst
use). Respect to the
reactor effluents gases are online analyzed by gas chromatography meanwhile the
feed stock and liquid products are characterized by density, refraction index,
nitrogen content, simulated distillation, type of carbon, type of aromatics and
viscosity.

color:#1F497D">RESULTS

The methodology described above allows
first of all to determine the effect of the variables on deactivation rate
which should  follow the subsequent behavior:

margin-left:36.0pt;margin-bottom:.0001pt;text-align:justify;text-indent:-18.0pt;
line-height:150%;border:none">●      
Increase
temperature to 410°C: conversion increases, coke precursors formation
increases.

margin-left:36.0pt;margin-bottom:.0001pt;text-align:justify;text-indent:-18.0pt;
line-height:150%;border:none">●      
Decrease
H2/HC ratio to 1000 LN/L: conversion decreases, increase of
dehydrogenation reactions involved in coke formation.

margin-left:36.0pt;margin-bottom:.0001pt;text-align:justify;text-indent:-18.0pt;
line-height:150%;border:none">●      
Increase
LHSV to 3 h-1: conversion decreases, quantity of contaminant
compounds (nitrogen compounds and coke precursors) flow from the feedstock
increases.

-18.0pt;line-height:150%;border:none">●      
Increase
organic nitrogen content to 300 ppm: conversion decreases, increase of
neutralized acid sites

Based on the results of the conversion
obtained when operating conditions return to the base case conditions, the
impact of each variable change on deactivation was determined and the specific value
for each variable were set for the establishment of the final accelerated
deactivation experimental procedure.

Taking into account the information
from the samples of the accelerated deactivation test, the set of operating
conditions was validated in terms of the loss of activity reached during one
month. As mentioned in the methodology description, this experiment was carried
out until obtaining a loss in conversion representative of 20°C of delta
temperature. Besides, information about the differences on catalyst properties
such as porosity, surface area, pore volume, coke deposition (quantity,
location and type), and the effect on the metallic function for three different
locations of the catalyst inside of the reactor and at two different times were
obtained.

normal"> 



" advot596495f2> 

" advot596495f2>