Atmospheric Residue Desulfurization Research Articles

Neural network models for atmospheric residue desulfurization using real plant data with novel training strategies

Oil refining industry has been using processes to convert heavy oil into high-value materials due to environmental regulations and decreasing demand for heavy oil. Atmospheric Residue Desulfurization (ARDS) process is important as a pre-treatment process for removing impurities from heavy oil. In this paper, neural network (NN) models are developed to predict the impurity removal amounts in the ARDS process using real plant data. To improve the reliability and reflect the chemical reaction characteristics, three training strategies are proposed: (1) restricting NN weights to positive values, (2) penalizing imbalances for two identical reactor trains, and (3) updating NNs with the latest 4-week data in a receding manner. The trained models satisfy the chemical consistency and provide well-balanced predictions. The model with a pre-defined aging factor structure can learn the catalyst aging status. The maximum percent error in the predicted values is smaller than or equal to 9.20%.

Development of a novel chelation-based recycling strategy for the efficient decontamination of hazardous petroleum refinery spent catalysts

The conventional hydrometallurgical methods for recycling refinery spent hydroprocessing catalysts are ineffective in simultaneously removing all metals (Ni, V, and Mo) in a single-stage operation. In this study, a novel octadentate chelating agent, diethylenetriaminepentaacetic acid (DTPA-C14H23N3O10), has been proposed for the first time to remove toxic metals (Ni, V, and Mo) in a single stage of operation from an industrial spent atmospheric residue desulfurization (ARDS) catalysts. It was discovered that the efficient formation of metal-DTPA complexes was attained under the optimum experimental conditions (60 °C, stirring - 150 rpm, S/L ration (w/v) of 2.5%, 7.5% DTPA, and medium pH-9) that resulted in the high removal of Mo (83.6%), V (81.3%) and Ni (64.1%) from the spent ARDS catalyst. Kinetic studies suggest that the leaching process followed a semi-empirical Avrami equation (R2 > 0.92), which predicted that the diffusion control reaction controlled the leaching. Species distribution and ecological risk analysis of the remaining metals in the insoluble residue (mostly Al2O3) indicated that the potential bioavailability of the remaining metals (except Ni) was significantly decreased, and residue poses a low ecological and contamination risk (individual contamination factor <1). Furthermore, the textural properties of the residue (BET surface area-103 m2/g and pore volume- 0.49 ml/g) were dramatically improved, suggesting that fresh hydroprocessing catalyst support can be synthesized using the leached residue. Compared to the conventional processes, the proposed chelating process is highly selective, closed-loop, and achieved high metal recovery in a single-stage operation while decreasing the environmental risks of the hazardous spent catalysts.

Hydrogen recovery from ARDS unit by membranes: A simulation and economic study

Atmospheric residue desulfurization (ARDS) is widely used in the refinery to produce low-sulfur fuel oil (LSFO). The need for this fuel is increasing to meet the high demand in transportation and power generation. In ARDS, excessive amounts of hydrogen is required for the hydrotreatment process. Therefore, the purge gas of the ARDS unit usually contains significant amounts of unreacted hydrogen. In this study, a commercial membrane was integrated into one of the ARDS units at the Clean Fuels Project (CFP) of Kuwait refinery. The simulation case was built in UniSim with the development of a membrane module. Results show that the membrane can recover 92% of hydrogen with a product purity of 98 mol%. The membrane system has an attractive payback duration of four to six months.

Neural network model for atmospheric residue desulfurization conversion prediction using real plant data

Atmospheric residue desulfurization (RDS) is used to treat the invaluable atmospheric residue from crude distillation column before it is sent to the fluid catalytic cracking (FCC) process. RDS is operated by increasing the operation temperature of catalysts to keep the desirable conversion as the catalyst aging progresses. To use the catalyst as long as possible while satisfying the criterion on the impurities in the RDS product, we need to predict the conversion based on the operation histories. In this study, we propose neural network models for predicting removal amount of the impurities in RDS. As the temperature, hydrogen flow rate, and impurities feed flow rates are selected as the features for neural network models. In addition, an approximated aging factor for the catalyst is considered with the cumulative amount of treated impurities and trainable parameters. The neural networks are trained in a moving horizon manner and tested using the real plant data during 123 days. If we train the neural networks in a general way, the trained model shows unrealistic results that are physically and chemically unexpected. When we increase the temperature with the other factors fixed, the catalyst conversion predicted by the trained neural network decreases. This absurd prediction is because the model is trained using the data where the temperature increases when the conversion is low with aging catalyst. To address this issue, we use a clipping method that the weights of neural networks are restricted to positive values. This strategy always ensures the increasing (decreasing) tendencies of the removal amount with respective to the increase (decrease) of the temperature and hydrogen flow rate.

Effect of catalyst and feed properties on the deactivation parameters of an ARDS process model for HDS and HDM reactions

A mathematical model for the atmospheric residue desulfurization (ARDS) process was used to estimate the deactivation parameters for the hydrodesulfurization (HDS) and hydrodemetalation (HDM) reactions. These parameters, namely Thiele moduli and contact efficiency, were assessed for three commercial ARDS catalyst systems and three atmospheric residues. The model’s deactivation parameters were validated with the literature, even though the model’s structure did not explicitly model these free parameters mathematically. It was demonstrated that the Thiele modulus had a positive correlation with the pore volume of the fresh catalyst and a negative correlation with its bulk density for the HDS reaction. The lighter feed, in comparison to the other feeds, contacted the catalysts more efficiently under the current operating conditions. Metal removal depended on the catalyst pore structure, and vanadium removal seemed to be controlled by diffusion.

Development of Gui for Ards Model Tool Using Matlab® Software

An interface for the atmospheric residue desulfurization (ARD) modeling tool developed at petroleum research center (PRC). The modeling tool simulates and predicts the concentrations of sulfur, nickel, vanadium, asphaltene, and conradum carbon residue (CCR) versus time on stream. This tool for built in MATLAB®, requires experimental data for developed for its execution and produces a large number of figures as outputs. Using the developed interface data templates are populated in a friendly manner. In addition, the produced figures are presented in an organized format. The interface is built in the graphical user interface (GUI) of MATLAB.

Predicted Effects of Heavy Feeds on the Deactivation of a Commercial Atmospheric Residue Desulfurization Catalyst System

The effect of three atmospheric residual oils on the deactivation of a commercial atmospheric residue desulfurization (ARDS) catalyst system was assessed using a commercial catalyst system consisting of five catalysts loaded inside a two-reactor pilot plant. The hydrodemetalation (HDM) and hydrodesulfurization (HDS) reactions of the three residues over the catalyst system were studied. A deactivation model considering metals and carbon deposition was used to fit the life test data. Experimental and simulated data were compared. The effect of the residual oils on each of the catalysts was evaluated, and the catalyst contribution was calculated highlighting the importance of predicting different catalyst activities during the design of a composite catalyst bed.

Robust Predictions of Catalyst Deactivation of Atmospheric Residual Desulfurization

A robust parameter estimation procedure was proposed to obtain accurate, confident parameters for the atmospheric residue desulfurization (ARDS) model in this work. A comprehensive ARDS model includes too many parameters such as deactivation and loading parameters. However, past studies have used data fitting in the early phases of reactor system life testing to obtain the parameters. This procedure was found to be too sensitive to the initial estimates used in fitting. In order to use the ARDS model for designing reactor systems and evaluating operating strategies, a robust parameter estimation procedure must be available. In this work, we demonstrate that the kinetic and deactivation parameters should be obtained from kinetic tests and accelerated test data, respectively for individual catalysts. Only loading parameters such as contact efficiencies should be obtained using early life test data. The procedure was validated with long-term experimental data. The results showed that the procedure could be applied on different catalyst systems as well, which allows the model to predict mid-term and long-term performance well during a life test. The procedure described here will substantially improve the robustness of model predictions in the refining industry.

Effect of the Operating Pressure on Residual Oil Hydroprocessing

The atmospheric residue desulfurization (ARDS) process is one of the main processing units in Kuwaiti refineries. In the ARDS process, a graded catalyst system consists of different types of catalysts. The performance of the catalyst system is mainly influenced by feedstock quality and reaction conditions, such as temperature and pressure. In this study, the effect of operating hydrogen pressure on the life cycle of the catalyst is investigated using two feedstocks, Kuwait heavy crude atmospheric residue (KHC-AR) and Eocene-AR, using an analogous process condition that is generally used in refinery. The pilot-plant hydrotreatment results demonstrated an effect of hydrogen pressure on the catalyst performance for the hydrodesulfurization (HDS) activity and sustaining a target sulfur level of 0.6 wt % in the product. The required sulfur level in the product was adjusted by raising the temperature gradually to compensate for the catalyst deactivation. The activity test for two different feeds were studied up to 4000 h time-on-stream (TOS) at 120 and 150 bar hydrogen pressure. The results showed that the performance of the catalyst system under an operating pressure of 150 bar was better than that at 120 bar. The reaction rate is accelerated by high pressure because it reduces the effect of the coke deposition on the catalyst.

Development of zeolite hydrocracking catalyst and system for resid hydrodesulfurization unit

Idemitsu Kosan Co., Ltd. had developed the iron (Fe) containing zeolite catalysts for the fixed bed residual oil (resid) hydrodesulfurization (HDS) unit. The catalysts, named R-HYC, showed high resid hydrocracking (HYC) activity by increasing the middle distillates and concurrently decreasing the bottom fuel oil. Commercial application with resid HDS units had been realized in 1980s–1990s. Because of relatively less HDS activity of R-HYC, the catalyst has been replaced with commercial HDS catalysts to correspond the lowering sulphur regulation in recent years. In order to meet the present demands on refineries, such as coping with the soaring oil price and shrinking market for fuel oil as the primary energy sources, we resumed developing new R-HYC catalysts, which should have good abilities in both HYC and HDS. For more practical purpose to be applicable to the existing atmospheric residue desulfurization (ARDS) unit, a suitable combination for the R-HYC catalyst with the conventional hydrotreating catalysts, such as hydrodemetallization (HDM) and HDS, has been under investigation.

Activity of hydroprocessing catalysts prepared by reprocessing spent catalysts

In this study, two spent catalysts from the 3rd and 4th fixed beds of the atmospheric residue desulfurization (ARDS) process and one spent catalyst from the H-Oil process employing the expanded bed were used as the starting material for preparation of hydroprocessing catalysts. The spent catalysts were used either in a coked or decoked form. The procedures involved pulverizing the spent catalysts before mixing, kneading and extruding with boehmite in the presence of peptizing agent such as HNO 3. The hydrodesufurization (HDS) and hydrodemetallization (HDM) activities of the new catalysts were evaluated in the microreactor using the atmospheric residue derived from Kuwait crude. The new catalysts prepared from the spent catalysts used in the fixed bed reactors exhibited higher activity than that of commercial catalysts. The HDS activity of the new catalysts prepared from spent catalysts used in the ebullated bed reactor was much lower than that of the other catalysts. This was attributed to a high content of vanadium in the former spent catalyst. At the same time, the adverse effect of vanadium on HDM was much less evident.

Optimization Study of Residuum Hydrotreating Processes

This paper presents an optimization study of catalytic hydrotreating reactors processing heavy residuum feedstock. The focus is on conversion, throughput and catalyst life. The core of the proposed optimization model is a cost function representing the essential economical parameters of hydrotreating processes and accounts for additional costs imposed by deeper desulfurization in addition to the monetary benefit of lower sulfur products. Operational variables are estimated using a mathematical model, which accounts for catalyst deactivation. Simulation results are presented to illustrate the effect of various operating variables on the process performance. An industrial scale atmospheric residue desulfurization process has been selected as a typical hydrotreating unit to demonstrate the capabilities of the optimization model. Optimization results were found quite reliable and consistent with actual industrial practices.

Multicatalyst System Testing Methodology for Upgrading Residual Oils

Atmospheric residue desulfurization (ARDS) is one of the major processes installed in refineries worldwide for upgrading petroleum residues to more-valuable clean products. Because of the importance of the process, research and development related to the process and its associated catalysts has gained increasing attention internationally. Hence, it is necessary to develop a reliable testing methodology for evaluating ARDS catalysts in a relatively short time. From our experience, it has been determined that, within 2−3 months, an ARDS catalyst system consisting of 3−5 catalysts can be evaluated and their performance in an industrial unit can be predicted. This requires generating quality data related to the initial activities, kinetic parameters of individual catalysts, the deactivation behavior, and maximum metal capacity in the catalyst system. Using the proposed catalyst evaluation methodology, values of the apparent reaction kinetics and catalyst life can be determined. These data are fed into an in-house developed mathematical model, which enables the prediction of the performance of the same catalyst system in an industrial scale. Very satisfactory results, from both pilot-plant test and commercial plant prediction, are realized from the developed methodology. When conducting long-life tests using the same feedstock and catalyst system, but under different operation modes, such as the constant-sulfur mode or the constant-temperature mode, the catalyst deactivation behaviors between the two life tests are significantly different, However, their respective values of the maximum metal-on-catalyst (MMOC) of the total catalyst system are very similar.

The usage of high metal feedstock for the determination of metal capacity of ARDS catalyst system by accelerated aging tests

For residue hydrotreating catalysts that are deactivated mainly by pore blockage due to deposition of metals, the capacity for accommodation of the deposited metals is an important parameter in determining its stability and life. The primary objective of the present work is to determine the maximum metal capacity of different catalysts that are used in different reactors of atmospheric residue desulfurization (ARDS) units by accelerated aging tests in a short duration. A feedstock with high metals content, namely, Boscan crude, was used for the tests in a multireactor pilot plant to cause complete deactivation of the catalysts by metal accumulation in a shorter period. The influence of operating temperature and LHSV on the rate of catalyst deactivation was also examined for the high metal feed. Catalyst deactivation rate was found to be more than five times faster for the high metal Boscan feed compared to the conventional atmospheric residue from Kuwait export crude. It was possible to determine MMOC of an ARDS catalyst system within 45–50 days by accelerated deactivation test using a high metal feed (e.g., Boscan crude).

Deactivation patterns of Mo/Al 2O 3, Ni–Mo/Al 2O 3 and Ni–MoP/Al 2O 3 catalysts in atmospheric residue hydrodesulphurization

In the present work, a comparative study on the deactivation behavior of three types of industrial hydrotreating catalysts, namely, Mo/Al 2O 3, Ni–Mo/Al 2O 3 and Ni–MoP/Al 2O 3, that are used to promote primarily hydrodemetallization (HDM), hydrodesulphurization (HDS) and hydrodesulphurization + hydrodenitrogenation (HDS/HDN) reactions, respectively, in the first, second and third reactor of commercial atmospheric residue desulfurization (ARDS) units was carried out. The main objective of the study was to contribute to a better understanding of the relationship between catalyst type and catalyst deactivation patterns. The used catalysts from these experiments were fully characterized to determine the extent and the cause of deactivation. Special emphasis was paid to understanding the nature of the coke and metal deposition on the used catalysts by applying chemical analysis and various advanced analytical techniques, such as solid-state carbon-13 nuclear magnetic resonance spectroscopy ( 13C NMR), temperature-programmed oxidation (TPO), electron probe micro-analysis (EPMA), and diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy. The results are discussed scientifically based on the physico–chemical properties of the three catalysts.

A comparative study on the performance of a catalyst system for the desulfurization of two kinds of atmospheric residues, Kuwait Export and Eocene residual oils

Crude oil feedstock is becoming increasingly heavier with higher sulfur, conradson carbon residue (CCR) and asphaltenes contents. This trend will specifically impact residual oil processes aimed at converting the residual oil to distillate products and hydrotreating the residual oil to low sulfur fuel oils (LSFO). In this paper, the performance of a catalyst system typically used in industrial atmospheric residue desulfurization (ARDS) units was investigated using residual oils of Kuwait Export Crude (KEC) and Eocene Crude as feedstock. The catalyst system consists of a combination of five catalysts having different physico-chemical properties. The performance of the catalyst system was studied using a pilot plant equipped with two reactors in series for a complete life cycle. Sulfur content in the product for both tests was maintained at around 0.6%, by raising the temperature gradually to compensate for the catalyst deactivation. The tests were continued until the temperature of the back-end reactor (R2) reached 412 °C. The results showed that the life of the catalyst system using the atmospheric residues of KEC and Eocene was around 7500 and 3800 h, respectively. The results are discussed in terms of possible differences in the deactivation pattern caused by differences in the composition of the two types of crude oils used in the study.

Utilization of Metal-Fouled Spent Residue Hydroprocessing Catalysts in the Preparation of an Active Hydrodemetallization Catalyst

In the present work, studies were carried out with the objective of utilizing spent residue hydroprocessing catalysts, which contained coke and metal (V and Ni) deposits, in the preparation of active new catalysts that can be used for hydrodemetallization of residual oils. Three types of spent catalysts that contained different levels of vanadium, molybdenum, and nickel were used in the experiments. These catalysts were collected from different reactors (front, middle, and back-end) of an industrial atmospheric residue desulfurization unit. A series of catalysts were prepared by mixing and kneading of spent catalysts with boehmite in different proportions followed by extrusion of the resulting paste. All prepared catalysts were characterized by chemical analysis and by surface area, pore volume, and pore size measurements, and their hydrodesulfurization (HDS) and hydrodemetallization (HDM) activities were evaluated by hydrotreating tests in a fixed-bed reactor using Kuwait atmospheric residue as feed. A c...

A Refined Hydroprocessing Multicatalyst Deactivation and Reactor Performance ModelPilot-Plant Accelerated Test Applications

A hydroprocessing multicatalyst deactivation and reactor performance model was originally developed to predict the life and performance of atmospheric residue hydroprocessing catalysts in a pilot-plant long-life test. The model has been subsequently applied to predict catalyst and reactor behavior in an accelerated life test using Boscan crude oil as the feedstock. Simulations from the original model did not compare well with accelerated test run data. Because Boscan crude oil has very high metal and asphaltene contents, mass balance terms that emphasized noncatalytic hydrothermal reactions in the model were needed, especially when catalysts were severely deactivated. Excellent comparisons were obtained after including hydrothermal reaction terms to refine the model. The model was then used to simulate the Boscan crude oil hydroprocessing in constant-sulfur-mode operations, and it was found that it would be more beneficial to operate the atmospheric residue desulfurization unit at a lower initial temperature.

Comparison between deactivation pattern of catalysts in fixed-bed and ebullating-bed residue hydroprocessing units

Abstract In residual oil upgrading by hydroprocessing, two types of reactor technologies namely, fixed-bed and ebullated-bed are commonly used. In this paper the deactivation behavior of catalysts in both types of units is compared based on the information obtained from the analysis of spent catalysts. In the fixed-bed unit, the concentration of vanadium deposits decreased from the entrance (first reactor) to the exit (last reactor), while, carbon content of the catalyst increased from the entrance to the exit. The loss of surface area for the catalysts in various reactors of the fixed-bed atmospheric residue desulfurization (ARDS) unit increased from the entrance to the exit. Distribution of the deposited vanadium with the catalyst pellet was also different in different reactors. Catalysts in the front-end reactors were deactivated mainly by metal deposition, while the back-end catalysts were primarily deactivated by coke deposition. The nature of the coke deposits on different catalysts was found to be significantly different by 13 C NMR, temperature programmed oxidation (TPO) and differential thermal analysis (TGA). Spent catalysts from the ebullated-bed reactor contained a mixture of highly fouled and partly fouled catalysts. The coke on the highly fouled spent catalyst particles was more aromatic and more condensed compared with that on the lightly fouled catalyst particles. Considerable changes in the physical dimensions of the catalyst pellets were noticed for the spent catalysts from the ebullating-bed reactor. In the ebullated-bed reactor, catalyst deactivation was caused not only by deposition of metals and carbon, but also by changes in the physical and mechanical properties of the catalyst under the dynamic conditions of the process.

Atmospheric Residue Desulfurization Process for Residual Oil Upgrading: An Investigation of the Effect of Catalyst Type and Operating Severity on Product Oil Quality

Atmospheric residue desulfurization (ARDS) process is extensively used in upgrading of heavy petroleum oils and residues to more valuable clean environmentally friendly transportation fuels and to partially convert the residues to produce low-sulfur fuel oil and hydrotreated feedstocks. Graded catalyst systems in multiple reactors are used in the process in order to achieve hydrodesulfurization (HDS), hydrodemetallization (HDM), hydrodenitrogenation (HDN), and conversion of residues to distillates at desired levels. The characteristics of the feedstocks processed in different reactors are significantly different. The quality of the feed entering the second reactor is strongly dependent on the operating severity in the first reactor and can have an important impact on the performance of the catalysts in the following reactor with regard to various conversions and deactivation rate. In the present work, a systematic study was conducted on the effect of two industrial catalyst types, namely, MoO3/Al2O3 (HDM)...