Thermal cracking is the first commercial conversion process developed in the early 1900s principally to produce more motor gasoline from crude oils and produce high-octane gasoline for aircraft use, initiating an attempt to change the composition of crude oil in petroleum refinery. The purpose of thermal cracking is to make light middle distillates from heavier ends by pyrolysis, or thermolysis. With the advent of catalytic cracking in the 1930s and 1940s and its capability to produce higher yields of gasoline with higher octane number, thermal cracking of gas oils has ceased to be an important process for gasoline production in modern refineries. In countries where the principal petroleum fuel with a high demand is diesel fuel, thermal cracking is still important in fuel refineries. A principal application of thermal cracking of distillate fractions in current refineries is limited to naphtha cracking for the purpose of producing ethylene (C2H4) for the petrochemical industry. However, thermal cracking of residual fractions, particularly VDR, is still practiced in association with visbreaking and coking processes in the refineries. The chemistry of thermal cracking and thermal cracking processes is discussed in this section.
By the end of this lesson, you should be able to:
This lesson will take us one week to complete. Please refer to the Course Syllabus for specific time frames and due dates. Specific directions for the assignments below can be found on the Assignments page within this lesson.
Readings | J. H. Gary, G. E. Handwerk, Mark J. Kaiser, Chapter 5 |
---|---|
Assignments | Exercise 5: A coil visbreaker operates at 500°C for 1 min. How long will it take to achieve the same thermal severity at 450°C in a soaker visbreaking process? An apparent Arrhenius activation energy for thermal cracking is given as 50 kcal/mol. |
If you have any questions, please post them to our Help Discussion (not email), located in Canvas. I will check that discussion forum daily to respond. While you are there, feel free to post your own responses if you, too, are able to help out a classmate.
Thermal cracking produces shorter straight chain alkanes from longer straight chains found in gas oils or other crude oil fractions. Free radicals (reactive species with unpaired electrons, but no electronic charge) are the active species that govern thermal cracking reactions. Because of the free radical chemistry, thermal cracking of gas oil would produce gasoline with relatively low octane numbers, as will be discussed later in this section.
Figure 6.1 lists the three steps of free radical chain reactions as initiation, propagation, and termination. In the figure, R-H represents a paraffin chain which can be expanded such as (H3-(CH2)n – H) where n represents the number of carbon atoms in the alkane. In other words, R represents a radical with an unpaired electron that becomes an alkane (R-H) when combined with a hydrogen atom. A Hydrogen atom with one proton and one electron is the simplest radical.
The free radical chain reaction starts with breaking the weakest C-C bond in the reactant alkane (R-H) to form two free radicals R1 and R2, each with one unpaired electron resulting from the homolysis of the C-C bond (initiation). Once formed by the initiation step, each free radical can go through two different propagation reactions:
In a hydrogen abstraction reaction, for example, the radical R1 removes (abstracts) a hydrogen atom from an alkane (R-H) to produce a shorter chain alkane product (R1-H) and a new radical R (hydrogen abstraction), thus propagating the free radical chain. Alternatively, the radical R1, (or R2) can go through a β scission reaction to produce an olefin (ethylene) as a product, and a radical R to propagate the chain. The β scission refers to the breaking of the covalent bond in the β position relative to the position of the unpaired electron, as shown below:
Note that the initiation step produces two free radicals, the propagation step produces a reaction product and one radical to continue the chain. The last step in the chain reaction, the termination step, removes two radicals to produce one or two stable compounds depending on the termination reaction, as seen in Figure 6.1. The principal end result of the free radical chain reactions in thermal cracking is to produce from long chain alkanes shorter-chain alkanes, light olefins, and some aromatic compounds. One important feature of free radical reactions is that isomerization reactions, e.g., shifting of the unpaired electron site from an edge atom of a molecule to the interior atoms (as shown in Figure 6.1), are not favored reactions. In other words, isomerization reactions take place at a slower rate than other propagation reactions, e.g., β scission reaction. The critical importance of this observation is that the thermal cracking reactions produce shorter straight-chain alkanes and olefins without any significant formation of branched-chain (or iso-alkanes). This is the reason why catalytic cracking processes have virtually replaced thermal cracking processes to produce high octane number gasoline, as will be discussed in the next section on catalytic cracking.
Please take a few minutes to answer the question below. Click "Check My Answer" to see some feedback.
Question: Why is hydrogen abstraction reaction favored over β scission at high pressures?
ANSWER: Hydrogen abstraction is a bimolecular reaction that is favored at high pressures over a monomolecular reaction (β scission).
Management of thermal reactivity is important in thermal cracking processes for optimum conversion of the feeds with a wide boiling range. Figure 6.2 shows an example of a thermal cracking process to convert a heavy gas oil fraction primarily to light gas oil (LGO). Typically, lighter products (gasoline and gas) and the heavier product (fuel oil) are considered by-products in this process. A heavy gas oil feedstock with a wide boiling range is fed to a fractionator to separate the feed into a light oil and a heavy oil fraction, depending on the desired cut points. Light oil (which would have a relatively low thermal reactivity) is heated in a separate furnace to higher temperatures so that the cracking takes place in a vapor phase. Heavy oil fraction (with a high thermal reactivity) is heated to a lower temperature for cracking in the liquid phase. The heated feed streams are combined in a soak drum to provide sufficient time for the completion of the cracking reactions. After the soaker, the products are sent to a flash separator to separate the heavy end as a side product (fuel oil) and the lighter products are sent to the fractionator for separation into LGO, gasoline and the gaseous products. Separating the feed into two fractions will avoid heating the reactive longer alkane chains to high temperatures to keep coke formation and gas production under control to maximize the LGO yield. In naphtha cracking for ethylene production, all reactions are carried out in vapor phase at low pressures to promote β scission reactions for high ethylene yields. Coking of reactor tubes creates a major maintenance problem in naphtha cracking for ethylene production.
Visbreaking is a mild thermal cracking process applied to reduce the viscosity of VDR to produce fuel oil and some light products to increase the distillate yield in a refinery [1]. Depending on the feedstock properties and thermal severity in the reactor, the process will typically achieve 10–25% of conversion of the heavy ends to gas, gasoline, and distillates while producing fuel oil with the desired specifications. Carbon rejection in small quantities on the reactor surfaces during thermal cracking helps reduce the viscosity of the fuel oil product (Figure 6.3). The process decreases the demand for a cutter stock used as diluent (e.g., kerosene) that might otherwise be used to reduce the viscosity of the heavy ends to meet the fuel oil specifications. Adding a diluent may still be needed, depending on the sulfur content of the product and the fuel oil specifications. Although the principal objective of visbreaking is to reduce viscosity, some refineries may use this mild cracking process to convert fuel oil into lighter distillates.
As in all chemical reactions, conversion in visbreaking depends primarily on temperature and time. As a measure of “thermal severity” under reactions conditions, one can use a thermal severity index (TSI) as a function of temperature and time that is shown in Figure 6.4. The exponential dependence of TSI on temperature relates to the general exponential term that constitutes the chemical reaction rate constants. The chemical conversion in visbreaking reactions can be expressed as the reduction in concentration (cA) of long-chain alkane (or high-molecular weight compounds) in the feedstocks. One can see from Figure 6.4 that the conversion in the visbreaking reaction can be expressed by the integral in Figure 6.4, assuming an apparent first-order kinetics for the reaction. It can also be seen, in Figure 6.4, that the conversion that can be related to the extent of visbreaking depends on (kt); and the TSI to establish the interchangeability of T and t for a given conversion relates to (e(-Ea/RT)t), where Ea is the apparent activation energy of the reaction, R is the universal gas constant, T is the temperature, and t is time. In using the TSI for comparing thermal severity of different T and t combinations as major operating variables of visbreaking, care should be taken to use the right units for R and T. As a general convention, an apparent activation of energy of 50 kcal/mol is assumed for thermal cracking reactions involving the homolysis of C-C bonds to produce free radicals.
Higher visbreaking severity would produce a higher reduction in viscosity. Thermal severity is limited by the reactivity of the feedstock and the storage stability of the residual fuel in accordance with the desired conversion level and desired reduction in viscosity. Asphaltene content and concarbon of the feedstocks are important factors to consider when selecting an appropriate thermal severity for the process to prevent excessive coking in the visbreaking reactor.
[1] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 5, pp.111-116.
There are two types of visbreaking processes: coil or soaker visbreaking. Figure 6.5a shows a schematic diagram of the coil visbreaking process. For visbreaking, the feedstock is introduced into the coil heated in the furnace, where the thermal cracking reactions take place. At the furnace outlet, the reaction products are immediately quenched using a portion of the gas oil product from the fractionator to stop the thermal cracking reactions. The quenched products are sent to the fractionator for separation into gas, gasoline, light gas oil, and visbroken residue streams. A steam stripper can be used with the fractionator for better separation of the visbreaking products. In the soaker visbreaking process, a soak drum is placed after the furnace, Figure 6.5b. Most of the thermal cracking reactions, in this case, take place in the soaker drum.
Depending on the process objectives and feedstock characteristics, reaction temperatures range from 450°C to 485°C and pressures ranging from 3 to 10 bar. Higher temperatures and lower residence times are used in the coil visbreaking process.
Residence times can vary from 1 min (associated with high temperatures in coil visbreaking) to 10 min (for lower temperatures used in soaker visbreaking).
Similar to deasphalting and distillation, the environmental impact of visbreaking is associated with burning fuel in the furnace to provide energy for thermal cracking, and, to a lesser extent, burning off the coke deposited in the coil or soaker drum leading to emissions of CO2, oxides of nitrogen (NOx), and oxides of sulfur (SOx) in the flue gases.
A coil visbreaker operates at 500°C for 1 min. How long will it take to achieve the same thermal severity at 450°C in a soaker visbreaking process? An apparent Arrhenius activation energy for thermal cracking is given as 50 kcal/mol.
Despite the development of catalytic cracking processes, coking processes have survived as a popular refining process all over the world to refine the heavy end of crudes or heavy oils through carbon rejection as coke. Coking is the most severe thermal process used in the refinery to treat the very bottom-of-the-barrel of crude oil, i.e., vacuum residue.
Because of the high severity of thermal cracking during coking, the residue feed is completely converted to gas, light and medium distillates, and coke with no production of residual oil. Three different coking processes are used in the refineries: delayed coking, fluid coking, and flexi-coking (a variation of fluid coking). The common objective of the three coking processes is to maximize the yield of distillate products in a refinery by rejecting large quantities of carbon in the residue as solid coke, known as petroleum coke. Complete rejection of metals with the coke product provides an attractive alternative for upgrading the extra-heavy crude and bitumen, and that is particularly useful for initial processing of tar (or oil) sands for liberating the hydrocarbons from the sand that is left behind with the coke. Finding markets for the coke product as fuel or as filler for manufacturing anodes for the electrolysis of alumina (possible only with petroleum coke from delayed coking) makes the economics of coking more attractive by creating value for the rejected carbon. Sulfur and metal contents of the petroleum coke, as determined by the sulfur and metal contents of the residue feed, are two important factors that affect the commercial value of petroleum coke. Of the two coking processes, delayed coking is the preferred approach in many refineries that process heavy crudes.
Figure 6.6 shows a flow scheme in a delayed coking progress and a photograph of a delayed coking unit. The derricks above the drums that contain the drill stems are used to drill out the coke from the coke drums at the end of the coking cycle.
As shown in Figure 6.6, the residue feed is introduced to the fractionator after being heated in the heat exchangers with the coker gas oil products. The bottoms from the fractionator, including the heavy ends of the vacuum residue feed with heavy coker gas oil recycle, are mixed with steam and sent to the tubular heater in the furnace to be heated to approximately 475°C at a pressure of 10-30 psi. Steam is added to prevent coking in the heater, and the heated feed is introduced from the bottom of one of the coke drums. The coking takes place in the insulated coke drum as the drum fills up for a period of 16–18 h. While drum A is being filled up, drum B is decoked by using hydraulic cutters and the drilling stem, and the coke is removed from the bottom of the drum. As the coking in drum A is completed, drum B should be decoked, sealed, heated, and prepared for switching the feed. The coking cycle is controlled such that the vacuum residue is continuously fed to the unit (because the vacuum column works around the clock) and the fluid products are recovered continuously, while coke is removed intermittently in a semi-continuous process scheme. Therefore, there are at least two coke drums in every delayed coking unit, and some units have more than two drums. All of the heat necessary for coking is provided in the heater, whereas coking takes place in the coke drum; hence, the process is called “delayed coking.”
The hot product vapors and steam from the top of the drum are quenched by the incoming feed in the fractionator to prevent coking in the fractionator and to strip the lighter components of the vacuum residue feed. The fractionator separates the coking products into gasses, coker naphtha, coker light gas oil, and coker heavy gas oil. A side-steam stripper is used with the fractionator to ensure a good separation between the coker naphtha and light gas oil streams [2].
The delayed coking operating variables include heater outlet temperature, pressure, recycle ratio, and cycle time. These variables are selected based on feed properties such as the characterization factor, asphaltene content, and Conradson Carbon Residue (CCR) to ensure that coking in tubular heaters is minimized, and liquid product yield is maximized. The recycle ratio, which is typically 3–5%, is used to control the endpoint of the coker heavy gas oil. The coke yield can vary from 20% to 30% depending on the feed properties and coking conditions. In the textbook, you may find some proposed equations to predict coke and other product yields on the basis of the CCR of the vacuum residue and estimates of the distribution of sulfur in the feed among the coking products, suggesting that up to 30 wt% of the sulfur in the feed ends up in the coke, 30 wt% in the gas product, and 20 wt% in the coker heavy gas oil.
[2] Petroleum Refining, by J. H. Gary, G. E. Handwerk, M. J. Kaiser, 5th Edition, CRC Press NY, 2007, Chapter 5, pp.97-111.
There are two kinds of coke produced by delayed coking of VDR: high-density shot coke, and porous sponge coke. Figure 6.7 shows the appearances of shot coke, consisting of aggregates of ~5 mm diameter spherical particles (resembling buckshots) and sponge coke (with a porous structure resembling a sponge). The formation of shot coke is usually troublesome because of difficulties in removing the coke from the drums and problems with grinding, although shot coke has some niche applications, such as in titanium dioxide (TiO2) production. Sponge coke is used as solid fuel, and manufacturing anodes for aluminum production, if its sulfur and metal concentrations are sufficiently low.
Among the delayed coking products, needle coke is a specialty coke produced mostly from coking of a highly aromatic FCC decant oil. The major properties of the needle coke include a low coefficient of thermal expansion, a low puffing (sudden volume expansion) tendency during graphitization because of lower nitrogen and sulfur contents, and high mechanical strength. The anode coke has limits on metal contaminants, requiring less than 500 ppm of Ni and V in the coke. The price of fuel coke depends on its carbon purity (S, N, and metal contaminants); however, the fuel coke is traded at a price comparable to that of coal.
Fluid coking and flexi-coking are fluid-bed processes developed from the basic principles of FCC, with close integration of endothermic (cracking, coking, or gasification) and exothermic (coke burning) reactions. In fluid coking and flexi-coking processes, part of the coke product is burned to provide the heat necessary for coking reactions to convert vacuum residua into gasses, distillate liquids, and coke. Flexi-coking, as a variation of fluid coking, provides the options of partial or complete gasification of the coke product to produce a fuel gas with some or no coke in the product slate. Different from the bulk liquid-phase coking in delayed coking, coking takes place on the surface of circulating coke particles of coke heated by burning the surface layers of accumulated coke in a separate burner. Figure 6.8 shows a schematic flow diagram of the fluid coking process. The preheated vacuum residue is sprayed onto the hot coke particles heated in the burner by partial combustion of coke produced in the previous cycle. Using fluid beds in the reactor and burner provides efficient heat transfer and fast coking on a collectively large surface area of the small coke particles circulating between the reactor and burner. The products of coking are sent to a fractionator (similar to that used in delayed coking after recovery of fine coke particles). Steam is also added at the bottom of the reactor (not shown in the figure) in a scrubber to strip heavy liquids sticking to the surface of coke particles before they are sent to the burner. This steam also provides fluidization of coke particles in the reactor. The reactor and the burner operate at temperatures of 510–570°C and 595–675°C, respectively.
Higher temperatures and short residence times in the reactor lead to higher liquid and lower coke yields compared with those of delayed coking. Coke is deposited layer by layer on the fluidized coke particles in the reactor. Air is injected into the burner to burn 15–30 % of the coke produced in the reactor, part of the particles are returned to the reactor, and the remainder is drawn out as the fluid coke product. Fluid coking can process heavier VDR and gives a higher distillate yield (and lower coke yield) than delayed coking.
Figure 6.9 shows a schematic diagram of flexi-coking. A gasifier is added for conversion of some or all coke produced in the coker in reaction with air and steam to produce a synthesis gas. The hot coke particles from the combustor are circulated back to the coking reactor to provide the heat necessary for coking. The distillate products from the coker are sent to the fractionator, as is done in the fluid coking process. On the gasifier outlet, after removing the fine particles from the gas by cyclones, the gas is cooled in a direct-contact cooler to condense the sour water and recover the flexi-gas. The product gas can be used as fuel gas in the refinery. Depending on the demand, the flexi-coking process can produce both fluid coke and fuel gas, or gasify all the coke to produce only fuel gas.
Please take a few minutes to answer the questions below before attempting to complete the assignments for this lesson. Click Check when you want to check your answers.
Each week, you will be required to do a number of assignments. This week, in addition to the reading assignments listed on the overview page, you are required to complete Exercise 5.
A coil visbreaker operates at 475°C for 1 min. How long will it take to achieve the same thermal severity at 440°C in a soaker visbreaking process? An apparent Arrhenius activation energy for thermal cracking is given as 45 kcal/mol. Show each step of calculations in your answer.
Once you have a solution to the exercises, you will submit your answers as a PDF by uploading your file to be graded. The MS Word, or Excel files should be saved as a PDF before submitting the exercise. Please Note: Scans of handwritten pages are not acceptable.
Physical separation of crude oil was not sufficient to meet the demand for motor fuels that had increased significantly with the increasing number of automobiles. Thermal cracking was therefore introduced as the first conversion process to produce more distillate fuels from petroleum refining. Thermal cracking, starting typically with breaking the C-C bonds in alkanes, proceeds with free radical chain reactions. One of the common outcomes of thermal cracking is to make light or medium distillates from the heavier fractions of the crude oil. Visbreaking, a mild (low-severity) thermal cracking process, reduces the viscosity of the VDR by rejecting a small quantity of coke as deposits on reactor surfaces. The principal product of visbreaking from VDR is a heavy fuel oil. On the other end of the thermal conversion spectrum, coking, the most severe thermal cracking process, converts VDR to light distillates and gaseous products by rejecting carbon in large quantities in the form of petroleum coke. If the sulfur and metal contents of VDR are sufficiently low, the petroleum coke can be a valuable by-product that is used for producing anode coke for electrolysis of alumina to produce metallic aluminum.
You should now be able to:
You have reached the end of Lesson 6! Double-check the to-do list on the Lesson 6 Overview page to make sure you have completed all of the activities listed there before you begin Lesson 7.
Readings | J. H. Gary, G. E. Handwerk, Mark J. Kaiser, Chapter 5 |
---|---|
Assignments | Exercise 5: A coil visbreaker operates at 500°C for 1 min. How long will it take to achieve the same thermal severity at 450°C in a soaker visbreaking process? An apparent Arrhenius activation energy for thermal cracking is given as 50 kcal/mol. |
If you have any questions, please post them to our Help Discussion (not email), located in Canvas. I will check that discussion forum daily to respond. While you are there, feel free to post your own responses if you, too, are able to help out a classmate.