Background of the Invention The stability of heavy fuel oils obtained from thermally cracked residual oils is a well known problem with significant economic ramifications. Residual fuel oil consists predominantly of an oil phase, the composition of which is almost entirely related to the crude oil from which it originates. In this oil phase are dispersed relatively large hydrocarbon molecules called asphaltenes. It is the nature of asphaltenes to be attracted to one another, and it is this tendency, along with size and concentration of the asphaltene molecules, that are consequences of both the crude oil type and the thermal cracking manufacturing process. The compositions of the various thermally cracked residual fuel oils can thus vary widely.

The stability of a residual fuel oil can be defined as its ability to resist the formation of carbonaceous sludge during storage and handling. The effects of sludge formation in a residual fuel oil in systems where that fuel oil is used to power an engine can result in choked centrifuges, filter blocking, heater fouling, and ultimately, engine shut down and damage. However, the simple formation of sediment over time in the bottom of storage tanks causes problems because these sludge layers are difficult to remove. These sediments are due to the aggregation of the unstable, high molecular weight polynuclear aromatic asphaltenes.

The traditional approaches of trying to stabilize thermally cracked fuel oils is to blend them with valuable refinery stocks or add any one of a variety of different chemicals to the fuel oils stored in tanks. However, these techniques have the disadvantage of having to be customized for each particular fuel oil. Moreover blending of fuel oil with other refinery cutter stocks requires the availability of aromatic heavy boiling cuts from Fluid Catalytic Cracking plants. If such streams are not available, any attempt to blend unstable cracked fuel with atmospheric or vacuum gas-oil will result in a de-stabilization of asphaltenes. Addition of chemicals in storage tanks also requires good mixing, which is seldom available.

Similar stability problems affect visbreaking and delayed coking processes, and potentially any bottoms upgrading process where the feed is stored at elevated temperatures prior to processing. Although delayed coking is used herein as a specific embodiment, as delayed cokers are units where the problem is often seen, it will be appreciated that the problem is present in any operation where feed is preheated and heat exchangers experience fouling.

Delayed coking is a bottoms upgrading process. It involves raising a feedstock to approximately 950°F (510°C) using a process furnace, and then transferring the hot stream to a coke drum. The coke drum functions as a residence chamber for the oil to allow time for cracking to occur. The products of the cracking are coke (a highly enriched carbon polymer) which forms in the coke drum, and some quantity of cracked distillates (gasoline and gas oil boiling range) which are removed by fractionating a stream that leaves the coke drum. A common problem with this process is the formation of fouling (coke formation) in the process furnace.

Delayed coker furnace fouling is believed to result from at least two mechanisms. The first involves the pyrolysis of hydrocarbons, followed by polymerization and dehydrogenation, leaving behind a nearly pure carbon structure commonly called coke. The second mechanism involves the destabilization of already existing asphaltene polymers in the feedstock.

These asphaltenes exist as a colloidal dispersion in the feedstock, with

another class of high boiling hydrocarbons, resins, acting as the dispersing agent. Any of several changes to which the feedstock is exposed can disturb this colloidal state, with precipitation of the asphaltenes on the furnace tubes.

These asphaltenes further dehydrogenate and result in a coke-like residue, very similar to that derived from the former mechanism.

The feed to the coker unit is typically composed of crude unit vacuum tower residue or bottoms (VTB). In most coker units, the VTB is partly routed directly as coker feed, while a portion is routed to intermediate storage. The storage exists as a buffer, to allow the upstream crude unit to continue producing VTB even while the coker unit is down for furnace tube decoking.

This decoking is periodically necessary to remove the coke formed from the two mechanisms described above. The primary economic impact of this furnace coking or fouling includes lost production penalties and potentially shorter furnace tube life.

There is thus a need to find a method and/or composition that will help stabilize thermally cracked fuel oils and coker feedstocks that is more effective than current techniques.

Summary of the Invention Accordingly, it is an object of the present invention to provide a chemical composition for improving the stability of thermally cracked residual fuel oils.

It is another object of the present invention to provide a method for treating thermally cracked residual fuel oils that improves their stability.

An additional object of the invention is to provide a fuel oil that has improved stability.

In carrying out these and other objects of the invention, there is pro- vided, in one form, a method for stabilizing a hydrocarbon stream containing asphaltenes that involves heating the hydrocarbon stream containing asphaltenes ; and adding to the hydrocarbon stream a magnesium overbased compound. The magnesium overbased compound may be a magnesium overbased carboxylate, a magnesium overbased sulfonate, a magnesium

overbased phenates, or mixtures thereof. The magnesium overbased compound is added in an amount effective to improve the stability of the hydrocarbon stream. The hydrocarbon stream is subjected to heat after the addition.

Further provided in another non-limiting embodiment is a method for inhibiting coke furnace fouling that includes heating a coke drum feedstock containing asphaltenes and then adding to the coke drum feedstock a magnesium overbased compound. The magnesium overbased compound can be a magnesium overbased carboxylate, a magnesium overbased sulfonate, a magnesium overbased phenate, or mixtures thereof. The coke drum feedstock is then stored at an elevated temperature.

There is additionally provided a method for stabilizing heavy fuel oils that involves thermally cracking a residual oil to provide heavy fuel oil ; and adding to the thermally cracked heavy fuel oil a magnesium overbased compound that is a magnesium overbased carboxylate, a magnesium overbased sulfonate, and/or a magnesium overbased phenate, in an amount effective to improve the stability of the fuel oil, where the adding is conducted sufficiently soon after thermal cracking to improve stability. In one non-limiting embodiment of the invention, the magnesium overbased carboxylate is added within about 2 hours or less of thermally cracking the fuel oil. In another non- limiting embodiment of the invention, the magnesium overbased carboxylate is added within about 40 hours or less of thermally cracking the fuel oil.

There is additionally provided in another non-restrictive form of the invention a stabilized heavy fuel oil that includes a thermally cracked residual oil. The stabilized heavy fuel oil also includes a magnesium overbased compound in an amount effective to improve the stability of the fuel oil. The magnesium overbased compound is added to the thermally cracked residual oil sufficiently soon after thermal cracking to produce the residual oil to improve stability. The magnesium overbased compound may be a magnesium overbased carboxylate, a magnesium overbased sulfonate, and/or a magnesium overbased phenate.

Detailed Description of the Invention The present invention relates to the addition of overbased magnesium compounds relatively soon after the heavy fuel oil is produced by thermal cracking before it is stored and/or blended, if blending is necessary. In particular, the invention is concerned with the treatment of visbreaker tars.

According to one embodiment of the invention, the application of these compounds is much more effective if they are applied to the visbreaker tars before storage and blending of the tar. In particular, early application of these compounds helps prevent or reduce aging difficulties, resulting in far better results than attempts to cure aging and stability problems after their occurrence.

In one non-limiting embodiment of the invention, the method operates by treating heavy fuel oils from fractionation of thermally cracked atmospheric or vacuum residuals. By"heavy"is meant with a boiling range above 350°C, a density ranging from 0.9 to 1 kg/m3 and a viscosity range from 200 to 500 centistokes (2 to 5 cm2/s) at 50°C. These properties are averages reported as non-limiting examples only; it should be clear that these parameters are not comprehensive of all thermally cracked residuals to which the present invention applies. The exact method of production of the fuel oils and their characteristics are not critical to the method of this invention. Producing heavy fuel oils by thermal cracking of residual oil is a well-known process in the industry. Within the context of this invention, it will be understood that the term"heavy fuel oil"includes, but is not necessarily limited to, visbreaker tars (or vistars), thermally cracked resids (residual fuel oils), turbine fuels, and the like.

Suitable overbased magnesium compounds include, but are not necessarily limited to, magnesium overbased carboxylates, magnesium over- based sulfonates and/or magnesium overbased phenates and the like. In one non-limiting embodiment, magnesium overbased carboxylates may be used.

A particularly suitable magnesium overbased carboxylate is Kl-85 available from Baker Petrolite. In another non-limiting embodiment of the invention, the overbased magnesium compound contains from about 21 to about 26 wt%

magnesium. In another non-limiting embodiment of the invention, the magnesium content is from about 24 to about 25 wt%. In a particular non- limiting embodiment of the invention, the compound has at least about 21 wt% magnesium, and in another non-limiting embodiment, has at least about 25 wt% Mg. Typically, these magnesium proportions are average amounts.

These magnesium overbased compounds may be readily produced by methods well known in the art.

In another non-limiting embodiment of the invention, the residual fuel oil is treated by the addition of about 25 to about 2000 ppm of a suitable magnesium overbased compound, based on the heavy fuel oil. In another non-limiting embodiment, the proportion of the magnesium overbased compound ranges from about 150 to about 2000 ppm of based on the heavy fuel oil.

An important part of the method of the invention is to add the mag- nesium overbased carboxylate (or other compound) to the thermally cracked heavy fuel oil sufficiently soon after thermal cracking to improve its stability.

The optimal time or time range of addition will vary depending on the nature of the overbased magnesium compound, how much is added, how much magnesium is present in the compound, the temperature of addition and the nature of the thermally cracked fuel oil. In one non-limiting embodiment, the overbased magnesium compound is added at least within about 40 hours or less. In another non-limiting embodiment the overbased magnesium compound is added at least within about 2 hours or less after separation of heavy fuel oil from other thermally cracked streams. In another non-limiting embodiment of the invention, the magnesium overbased compound is added within 20 hours or less after thermal cracking, and in an alternate embodiment, within 10 hours or less.

In one other non-limiting embodiment of the invention, the adding of the overbased magnesium compound is performed within a temperature range of about 250 to about 490°C. Typically, this temperature will be at or near the temperature of the thermally cracked residual fuel oil shortly after it is

produced. In yet another non-limiting embodiment, the overbased magnesium compound is added within a temperature range of about 250 to about 380°C.

Although it is acceptable to use the overbased magnesium compounds made according to the methods of U. S. Pat. No. 6,197, 075, it will be appreciated that in another non-limiting embodiment of the present invention, the method is practiced in the absence of adding a co-promoter reaction product of a succinic anhydride and a lower carboxylic acid, in all of the forms described in the'075 patent. It is also noted that the method of the'075 patent does not appreciate the need for adding the overbased magnesium compounds very shortly after the residual fuel oils are produced by thermal cracking.

Further, it will be appreciated that it is not necessary for the heavy fuel oils to completely prevent asphaltene sediments or aggregation or to produce a heavy fuel oil that is stable forever for the invention to be considered successful. Rather, the methods and compositions of this invention are successful if the stability of the heavy fuel oils is simply improved.

The present invention also relates to inhibiting or preventing furnace fouling caused by asphaltenes in other hydrocarbon streams including coker feedstocks. Again, it will be appreciated that fouling need not be entirely prevented for the invention to be considered successful in this context.

It has been observed that furnace fouling rates are correlated to the amount of coker unit feed that comes from storage. Higher amounts of feed from storage result in generally higher rates of furnace fouling. The basis for the invention is the possibility that a significant portion of the coker furnace fouling results due to the 10-20 % of feed that comes from the storage tank.

This stored-feedstock-induced fouling is believed to result from degradation of the feedstock during this storage time. These materials are very viscous, and, as such, must be kept at elevated temperatures (250°F and higher; 121°C and higher) while stored to allow pumping to the coker unit. Storage times can range from a few days to several weeks.

The degradation of the stored materials is believed to involve several possible mechanisms :

1. Thermal destruction of the resins, followed by aggregation of asphaltenes.

2. Oxidative destruction of the resins, followed by aggregation of asphaltenes.

3. Oxidation and polymerization of hydrocarbons to form additional, poorly soluble polymers.

In all three cases, metal catalysis may promote the chemical reactions.

The metal catalysis may be caused by impurities in the hydrocarbon stream or possibly the metal conduits and vessels. Also, in cases 1 and 2, the stability of asphaltenes is negatively affected, which causes this portion of the coker feed to exhibit a higher degree of asphaltene destabilization and precipitation than the portion of feed that does not experience extended storage time. In case 3, a polymer is formed that, like asphaltenes, has poor solubility in the oil, and with the extreme temperatures seen in the furnace, can precipitate as foulant.

Finally, it has been discovered that these forms of degradation can be successfully inhibited using the magnesium overbased compounds previously described. A variety of thermally stable dispersants such as the magnesium overbased compounds have shown promise in controlling this degradation, thus greatly reducing this component of the furnace fouling. Laboratory testing thus far has shown that two magnesium overbased compounds within the definition of this invention, both thermally stable dispersants, give significant inhibition to the degradation, as measured by solids by hot filtration.

Previous attempts to control or reduce furnace fouling in delayed coking units and visbreakers have generally met with little success. Only sporadic success has been reported. The prior attempts have involved adding either a dispersant, anti-coking or antioxidant additive directly to the feed that goes to the furnace in the delayed coker unit. Additionally, previous attempts to improve on the furnace treatment by moving the injection point back up stream, as far as the vacuum tower bottom, to enhance mixing of additive and oil and to provide additional residence time appear to have shown some

benefit. The benefit may well have occurred as a result of some of the additive going with the feed that was stored. This invention involves introducing a more concentrated chemical treatment into the stored feed or just prior to the hydrocarbon stream being stored. By"more concentrated"is meant the high-magnesium content compounds of this invention.

Typically in a delayed coking operation, the stored feed comes from a pair of storage tanks. One embodiment of this invention would involve concentrating the treatment chemicals in the rundown to storage, rather than treating the furnace directly.

The invention will now be further described with respect to particular Examples that are not meant to limit the invention, but rather are intended to illustrate it further with respect to certain, more specific non-limiting embodiments.

Hot Filtration Test The Hot Filtration Test (HFT) is a relatively standard test to determine the stability of a particular fuel oil.

Materials Four Whatman fiber glass GF/A type filters, 1.6 micrometers porosity or equivalent Hot filtration test equipment Heating plate Analytical balance, 0.0001 grams 100°C thermometer with 1 °C precision n-Heptane, analytical grade Mixture of 85% n-heptane, 15% xylene by volume, analytical grade

Procedure: Install on a hot filtration test apparatus four pre-weighed (0.0001 grams precision) filters, two for each filtration heated flask. Heat the filtration flasks at 100°C with vacuum applied to the hot filtration test filter holders.

Heat the fuel to about 70-80°C to have a fluid fuel. Weigh 10 grams of heavy fuel oil sample, with 0.0001 grams precision. Heat fuel oil in a range of 99-101 °C and pour about half of the fuel oil in the first HFT filter holder, with the two Whatman filters. Register, by weight, the exact quantity poured.

Repeat the operation with the other filter holder. Wait for complete filtration under vacuum. Cool filter holders to ambient temperature. Wash each filter couple, with filters on the filter holder, with two washings of 25 mis each of n- heptane/xylene mixture and two washing of 10 mls each of n-heptane. Dry each filter couple, and re-weigh, with 0.0001 gram precision.

Sediments are determined by average of the weight difference after and before filtration. Hot filtration results can be calculated as follows.

HF1 = First couple of filters weight after HFT test procedure-First couple of filters, 1 weighed after HFT, 1 before HFT HFT1 = 100* (HF1)/Fuel oil filtered HF2 = Second couple of filters 1 weight after HFT test procedure-Second couple of filters, 1 weighed after HFT, 1 before HFT HFT2 = 100* (HF2)/Fuel oil filtered Hot Filtration Test Result, HFT = (HFT1+HFT2)/2 Results As widely recognized, the acceptable sediment content for fuel oil is less than 0.5% by hot filtration test (HFT). These sediments are due to aggregation of unstable high molecular weight polynuclear type aromatics known as asphaltenes. Higher contents than 0.5% have a negative impact on fuel filters (plugging) and on the burning quality of fuel. High contents of

sediments also have a negative impact on storage tanks as they tend to settle out on the bottom of the tank with a layer of sludge that is difficult to remove.

The invention consists in limiting the content of sediment formation with time in storage tanks for fuel oil. Particularly, the invention is concerned with treatment of residues from thermal cracking, used as heavy fuels or blended with gas-oils for fuel oil no. 6 production, in one non-limiting embodiment. More particularly, this invention is related to the treatment of resids from visbreaking, commonly known as vistar or tar.

These feedstocks are very problematic with respect to sediment for- mation since thermal cracking in the furnace gives rise to instability. This can be partially solved by decreasing thermal cracking temperatures or reaction time at cracking temperatures (about 430-490°C), although this leads to a strong decrease in the yield of valuable 360°C+ distillates from thermal cracking.

In attempts to limit the problem of sludge formation, several products were tested: oil soluble magnesium carboxylate overbased products (having 14-18% magnesium), oil soluble magnesium carboxylate overbased formulations with a higher magnesium content (23-26%; average 25%), an asphaltene dispersant (Baker Petrolite BPR34260) and a sterically hindered phenol, which acts as a radical stopper-scavenger, commonly marketed as antioxidant.

To test the products'effectiveness, samples of vistar and vistar blended with gas-oil streams were submitted to meet no. 6 fuel oil viscosity specifications, the blank (untreated) samples and treated samples were subjected to controlled"aging", that is, keeping them at 80°C for a period of time of more than 100 hours. This is well representative of typical storage tank temperatures and after more than 60-80 hours aging (sludge/aggregates) is complete.

Products are qualitatively considered to be effective whenever they are able to keep a sediment content of less than 0.5% by Hot Filtration Test.

Reported results are from duplicated measurements with differences in measurements of less than 10%.

The results on samples from storage tanks, immediately after storage, show an unexpected ineffectiveness of these products. The charge from which the tar sample came had been processed in a visbreaker plant about 10 hours before the sampling of finished product, the tar. In fact, the initial Hot Filtration Test value of the tar is low (0.07%), and products are ineffective even at higher dosages as reported in Table I below.

Table I HFT Data for Samples Partially Aged, From Storage Sample, after Aged, Dosage, storage (10 108 Ex. Product ppm hours after hours production) 1 None 0 0.07 1. 25 % 2 Overbased 24-26% Mg 2000 0. 93 % 3 Overbased 23-26% Mg 200 0. 95 % 4 Overbased 14-18% Mg 2000 1. 01 % 5 Dispersant BPR34260 5000 0.01 1. 25 % 6 Sterically hindered phenol 5000 0.01 1. 23 % antioxidant BPR34017 BPR34260 and BPR34017 are available from Baker Petrolite.

It was discovered with several trials that for thermally cracked resides (tar), magnesium carboxylates products are surprisingly effective when added immediately (within about 2 hours or less) after separation of tar from other thermal cracking (visbreaking) products, before sending tar to storage in tanks, as shown below in Table II.

Table II HFT Data for Fresh Samples from Plant, Treated Immediately After Tar Fractionation Fresh Aged, Ex. Product Dosage, ppm Sample 108 hrs 7 None 0 0.01 1. 21 % 8 Overbased 25% Mg 2000 0.01 0.31 % 9 Overbased 25% Mg 200 0.01 0. 38 % 10 Overbased 25% Mg 150 0.01 0. 38 % 11 Overbased 25% Mg 100 0.01 0. 71 % 12 Overbased 25% Mg 50 0.01 0. 79 % 13 Overbased 14-18% Mg 2000 0.01 0. 7 % 14 Overbased 14-18% Mg 200 0.01 0. 95 % 15 Dispersant BPR34260 2000 0. 01 1. 0 % 16 Sterically hindered phenol antioxidant 2000 0.01 1. 05 % BPR34017 It should be noted that magnesium carboxylate with a 25% average magnesium content was effective to keep asphaltenes aggregation resulting in Hot Filtration Test sediments below 0.5% with dosages of 2000 ppm, 200 ppm, and 150 ppm (Examples 8,9 and 10, respectively). For the magnesium overbased products of these Examples, 25% is an average value, as this product has some variability in the exact magnesium content, typically between 23 and 26%.

It was also discovered that treatment on fuel oil no. 6 from blending of gas oil and tar and tar itself is ineffective to reduce Hot Filtration Test content with samples from storage tanks that have been aged for more than 60 hours and are beyond the 0.5% HFT content limit, as shown in Table III.

Table III HFT Data on Samples from Storage Tank after 60 hrs for Tar and for Blended Tar to Meet No. 6 Fuel Oil Viscosity Specifications Sample, from Ex. Product Dosage, ppm storage tank 17 None 0 > 0. 5% 18 Overbased, 24-26% Mg 2000 > 0. 5% 19 Overbased, 24-26% Mg 200 > 0. 5% 20 Overbased, 14-18% Mg 2000 > 0. 5% 21 Overbased, 14-18% Mg 200 > 0. 5% Generally, blended tars (fuel oil no. 6) showed a greater HFT than blended tars.

Example 22 Prevention of Furnace Fouling A commercial coker currently takes approximately 15% of its feed from storage. In a 5 year period, numerous additive treatments have been tried, at the furnace, to control furnace fouling. The severity of the problem is such that a spalling (a cleanup process) is necessary, on average, every 12 days.

Furnace fouling is measured by the rate of skin temperature change for several thermocouples attached to furnace tubes. A typical starting temperature is 1000°F (538°F). The limitation is an upper limit on the skin temperature, typically around 1200°F (649°C). When the limit is reached, a spall or a decoke operation to remove coke is required. When either cleanup process is carried out, production is lost, and furnace tube life is shortened.

These costs are what drive the refiner to seek solutions. It is expected that injection of an effective amount of a magnesium overbased carboxylate, such as the proportions previously mentioned, would inhibit fouling sufficiently the time between cleanings is increased from 12 days to 3 months.

In the foregoing specification, the invention has been described with reference to specific embodiments thereof, and has been demonstrated as effective in providing a method of improving the stability of heavy fuel oils and

other hydrocarbon streams containing asphaltenes. However, it will be evident that various modifications and changes can be made to the inventive compositions and methods without departing from the broader spirit or scope of the invention as set forth in the appended claims. Accordingly, the specification is to be regarded in an illustrative rather than a restrictive sense.

For example, particular magnesium-containing overbased compounds falling within the claimed parameters and added at different times and dosages, or with particular co-components, but not specifically identified or tried in a particular composition or under specific conditions, are anticipated to be within the scope of this invention.