US2200703A - Process for refining hydrocarbon oils - Google Patents

Description

May 14, 1940. E. R. P. E. RETAILLI'AU 2.200.703

PROCESS FOR REFINING HYDROCARBON OILS Filed May 18, 1938 Dilure H2so4 3 Disfillare ll Vapors: 5

lnvenfor: Edmond Qeroilliau Patented May 14, 1940 PATENT OFFlCE 2,200,703 PROCESS FOR- REFINING HYDROOARBON OILS Edmond R. P. E. Retailliau, Edwardsville, 111., as-

signor to Shell Development Company, San Francisco, Calif., a corporation of Delaware Application May is, 1938, Serial No. 208,697

9 Claims.

This invention relates to a process for refining hydrocarbon oils in a heated vaporous condition by contacting same with a chemically acting acid refining agent, and more particularly is concerned with the vapor phase refining of gasoline distillates at elevated temperatures with sulfuric acid or water-soluble bisulfates.

Many attempts have been made to use sulfuric acid in the vapor phase treatment of hydrocarbon oils, but such attempts have been substantially unsuccessful because of charring and burning of a portion of the oil by the acid at the elevated temperatures. This difficulty was in part overcome by adding to the acid substantially non-volatile spacing reagents, such as neutral salts, organic bases, etc. However, the presence of these spacing agents often leads to various difficulties and the presence of volatile spacing agents was believed to give insufiicient protection.

It is a purpose of this invention to provide a method for refining unstable hydrocarbon distillates, such as cracked gasoline, tractor fuel and other distillates to produce stable oils of good color and color stability, low gum and high gum inhibitor susceptibility. It is another purpose to achieve this result by treating same with sulfuric acid or a related refining reagent in the vapor phase and in the absence of non-volatile spacing to about 575F., and distillate fuel oils such as the No. l to No. 3 fuel oils by the A. S. T. M. specifications D396-34T, which oils may boil between about 300 to 600 F. A common difficulty of burners designed to operate on distilled fuel oils is the clogging of the screen through which the fuel must pass before combustion. This difficulty is largely overcome when treating the distillate fuel oils by my process. Moreover as the result of the treatment carbon forming bodies which have a tendency to cause soot are at least partially eliminated.

Frequently a considerable desulfurization is noted in particular when operating on high sul-,

fur distillates; and nitrogen bases contained in the distillates may be and usually are completely removed in the course of my treatment.

I have discovered that chemically acting acid reacting refining reagents which cause charring or burning at normal vapor phase refining temperatures, in particular sulfuric acid and watersoluble hydrosulfates such as the alkali metal bisulfates, aluminum hydrosulfate, chromium hydrosulfate, ferrous hydrosulfate, etc., can be employed successfully in the vapor phase refining of hydrocarbon oils at elevated temperatures and in the absence of non-volatile spacing agents, if certain conditions are maintained. My method. consists essentially of introducing the exact amount of sulfuric acid or hydrosulfate, which is required to refine a given hydrocarbon oil to a desired degree and which is consumed completely, i. e., converted tosludge in a single pass, provided sufficient time of contact is allowed, in the form of adiluteaqueoussolution of less than about 20% concentration into the hydrocarbon vapors and passing vapor and reagent as a mixture through a treating zone containing solid obstructions, for instance in the form of a packing. After passing through the treating zone, refined hydrocarbon vapors and resulting sludge are separately withdrawn and vapors are condensed. The resulting distillate is neutralized to remove the small amounts of S02 normally evolved during treatment and may be sweetened by any conventional sweetening process. However, it never contains entrained sludge such as is the case when concentrated acid is used. in the vapor phase treatment of hydrocarbon distillates, because no gaseous S02 is everliberated in my process. S03, if liberated, as is the case when using concentrated sulfuric acid, continues to react with the vapors even outside the reaction zone proper until used up, making necessary an elaborate system for the separation of the resulting finely divided entrained sludge.

To provide an efiective contact between the solution of the reagent and the vapors, the former is effectively dispersed within the latter as by spraying, spreading over a large contact surface such as that of a packing, or other suitable means. The volume ratio of the reagent solution to hydrocarbon vapors being very small in my process in spite of the high degree of dilution of the former, treatment by passing the vapors through a liquid pool of the solution is impractical, if not impossible.

The treating zone may conveniently consist of a vertical tower or column containing an inert relatively coarse packing such as broken brick, tile, pumice, coke, steel wool, rings made from silicious materials, iron, copper, etc. In general it is unnecessary that the packing be corrosion resistant against sulfuric acid, because within a short timeit is completely covered by a protective coating of sludge. It is, however, necessary that the packing material be inert towards the sludge, and it should preferably have a compressive strength sufiicient to support a bed at least 20 to 40 feet deep, even after prolonged exposure to the sludge.

During the contact of the reagent solution and hot hydrocarbon vapors, several reactions take place almost simultaneously. Water of solution begins to vaporize immediately, whereby the reagent within the individual droplets or the film is concentrated rapidly. Simultaneously the 'sulfuric acid or hydrosulfate begins to react with the most reactive components in the vapors. The result is that by the time the reagent has been concentrated to a point sufficient to cause charring or oxidation, it is already sufficiently spent and diluted with its own products of reaction, so that it no longer can develop the harmful charring or burning effect.

In contact with the vapors, the sulfuric acid or hydrosulfate reagent is converted to a fluid sludge which deposits on the packing, enveloping and protecting same, and eventually runs down to form a liquid pool at the bottom of the treating zone from which it can be withdrawn readily. In the early stages of its formation this sludge exerts a definite refining action on partly refined vapors until its free acidity is substantially consumed or the vapors become unreactive. The final sludge differs considerably from acid sludges formed in conventional sulfuric acid treatment. For instance, it contains no free acid and is substantially non-corrosive to iron, copper and many other metals. It has a pleasant aromatic odor, is'normally completely soluble in gasoline, insoluble in water and ethyl alcohol, and does not evolve S02 on standing at room temperature. Its sulfur content is high, e. g., of the order of 1.6%. Hydrolysis with a concentrated I-ICl at 300 F. under reflux has shown it to be free from alkyl sulfates.

Sulfuric acid used in my process may be of most any commercial or recovered grade and may contain various impurities. For instance acid recovered by hydrolysis of acid sludges obtained in the conventional acid treatment of light or heavy hydrocarbon oils may be used. Such acids may contain among other impurities acid, sulfuric acid esters and. sulfonic acids, Also nitrogen bases are frequently contained therein.

The reagent solution and vapors may be passed through the treating zone concurrently or countercurrently. In general I prefer concurrent flow,

as it normally results in smaller treating losses for a given degree ofstabilization of the hydrocarbon oil. This may be due to the fact that in concurrent vflow the most reactive and most unstable components of the hydrocarbon vapors are removed by the fresh reagent. If the correct amount of the reagent is used, i. e., that which is substantially spent and used up when the most unstable components have been converted to sludge and/or polymers, other components which are more stable and yet capable of reacting with the reagent are but little affected, because the reaction which they undergo in contact with sludge in the digesting period following the first rapid reaction, appears to be of the nature of reforming resulting in little additional formation of sludge and polymers, if any, but usually resulting in a considerable increase in stability, gum inhibitor susceptibility and other properties of the distillate which depends upon such reformation. In other words, concurrent flow is highly selective, eliminating in the form of sludge and polymers essentially the most reactive components only and converting reactive components of lesser instability to stable components boiling largely within the boiling range of the untreated hydrocarbons.

'In this manner minimum loss in the form of sludge and polymers is combined with satisfactory chemical stabilization.

In countercurrent flow on the other hand the fresh reagent comes in contact with vapors from which at least a portion of the most reactive components have already been eliminated in the form of sludge and high boiling polymers. Consequently, the more stable reactive components will be attacked'to form additional sludge and high boiling polymers. I s I As hereinbefore stated I use an amount of treating reagent which is completely used up. This has the advantage in addition to forming a non-corrosive sludge, of obviating recirculation of unconsumed reagent, a feature which greatly simplifies the treating equipment and reduces the cost of installation. Moreover, at no time are vapors exposed to an excess of reagent, so that only the very minimum amount of reagent is consumed and only the most unstable components are converted to yield the minimum amount of sludge and polymers.

The distillat'es resulting from my treatment are improved in regard to their color, gum stability, susceptibility toward gum inhibitors, as is usual in chemical refining processes. In one important aspect, however, the distillates from my process differ from those of other sulfuric acid treating processes. Normally sulfuric acid treatment materially reduces the octane number of gasolines. My treatment, however, normally does not affect the octane number and in some ing the gasoline, had induction periods by the oxygen bomb test, before and after removal of 1% high boilingpolymers formed in the treatment, of 3% and 2 /3 hours, respectively.

The most essential factors in my treatment are amounts and concentration of the reagents, and temperature and time of the reaction.

Concentration of the aqueous reagent solution affects the process in at least two ways: it influencesthe effectiveness of distribution of the reagent throughout the vapors, and, if too high,

may be the cause of charring, burning and/or oxidation of the vapors. The greater the dilution, the greater is the volume of treating solution to be used and the more efliciently can it be injected into and distributed throughout the vapors. From this point of view it is in general desirable to use concentrations not in excess of about 10% by weight, although when operating at relatively low temperatures in particular, concentrations up to about 20% by'weight maybe employed without substantial danger of charring and burning Experiments made with aqueous solutions containing 25% sulfuric acid at temperatures of about 300 led to a rapid coking andclogging of the packing even in the presence of an equal percentage. of a. so-calledspacing agent. Continuous operation over extended periods of time is obviously impossible under these conditions. on the other hand at relatively high temperatures I have successfullyused solutions having concentrations as low as .2% and even lower concentrations may be employed, if desired.

I ,Amountsof reagents suitable for my treatment are between about02 to 4 lbs. and preferably less than aboutl lb. of reagent per barrel of distillate. To minimize consumption of reagent, the

minimum amount which gives a satisfactory treating result is obviously the most desirable. This minimum varies considerably with the type of distillate under treatment and also with the reagent, in general it being lower for sulfuric acid than for hydrosulfates.

I have found that best treating results are obtained if the amount and concentration of the reagent solution, particularly in the case of sulfuric acid, are correlated with the treating temperature in such a way that thehypothetical potential concentration of the reagent in the treating zone, ascalculated from the temperature and the partial pressure of the water vapor in the resulting vapor mixture, neglecting consumption of the reagent and the dilution caused by products of the reaction, is between about 50 and 75%. It is understood that these figures do not represent actual potential concentrations of reagent solution, because as pointed out the reagent actually is consumed and dilutedby its own products of reaction. What I call hypothetical potential concentration is the concentration which would result from the vaporization of water of solution if no chemical reaction would take place between the reagent and the hydrocarbon vapors. At hypothetical concentrations higher than the above there is a tendency to form coke due to the charring effect of the reagent,

coking becoming rapid at hypothetical potential concentrations of 80% and higher; and at hypothetical potential concentrations below 50% there is danger of poor refining action.

Suitable contact temperatures are in general of the order of about 200 to 400 F., temperatures between about 230 to 300 F. being preferred. At relatively high temperatures i. e.', 300 F. and higher, the sludgehas a considerable tendency to decompose, thereby liberating S02 and forming coke, and at temperatures below about 250 F, average gasoline distillates are incompletely vaporized. If relatively low temperatures are desirable, resort may be had to subatmospheric pressures in order to effect complete vaporization. It is desirable that the hydrocarbon distillate be completely vaporized in the treating zone, as I have found that the presence of hydrocarbons in the liquid state adversely affects the treating results and, in particular raises th polymerization losses.

Superatmospheric pressures may be employed if desired but are limited by considerations of temperatures and volatility of the hydrocarbons as explained above. Theeffect of superatmospheric pressures is mainly that of lengthening the time of contact in the treating zone for a given throughput and also of reducing the hypothetical potential concentration of the reagent at a given temperature, due to an increase of the steam pressure in the vapors. On the other hand when treating relatively high boiling distillates a considerable vacuum may have to be maintained inthe treating zone to enable substantially complete vaporization and to prevent condensation of the distillate at the most desirable treating temperatures.

The time of contact may be varied between the limits of about 3 to 90 seconds and may if desired be extended beyond these limits. In general at relatively low temperatures longer times of con tact are permissible and required than at higher temperatures under otherwise equal conditions. Reaction times of less than about 3 seconds are normally insuflicient. With increasing time of contact above about 3 seconds the stability of the hydrocarbon oil increases while treating losses may rise but little and often remain substantially unchanged. On the other hand if contact is extended beyond a certain time which depends largely upon the temperature, stability is not further improved materially, and certain disadvantages make themselves felt such as the extremely large size of the reaction vessel which is required for a given throughput, a tendency of the sludge to decompose thereby liberating more S02, unfavorable reaction between the S02 and the hydrocarbon vapors and coke which is liable to plug the packing.

Coking of the sludge can be minimized Within limits by employin multi-stage treating systems in which the treating reagent is divided into as many portions as there are stages, one portion being injected into each stage, sludge, if formed, being withdrawn separately from each stage, and the vapors are passed serially through the sev eral stages. The concentration of the solutions injected into the several stages need not be the same, neither is it necessary to maintain the same treating conditions in the various stages. On the contrary, I have found that it is advantageous to maintain progressively lower temperatures between 200 and 400 F. in successive stages, and in order to maintain the optimum hypothetical potential concentrations of the reagent in the several treating zones as hereinbefore explained, I may inject aqueous reagent solutions of progressively higher concentrations. As to the amount of sludge formed in the several stages I have found that most of it is produced in the first stage, very little if any being formed in subsequent stages.

It is usually unnecessary though permissible, to precede my treatment with a caustic alkali or an acid wash for the removal of impurities such as mercaptans, alkyl phenols, nitrogen bases, etc. A preliminary liquid phase wash has the disadvantage of necessitating the steps of condensing and redistilling the treating stock, whereas in the absence of such awash these steps are obviated.

My process will be further understood from the drawing which represents a simplified flow diagram of one form of my process.

A dilute aqueous solution of a suitable reagent, such as sulfuric acid is admitted to the top of treating tower I through line 3, and a similar solution may be injected into the top of treating tower 2 through line 4 from sources not shown. Distillate vaporsof the proper temperature enter tower l nearits top through line 5, or near its bottom through line 6. In the tower the acid and the vapors are closely contacted and react as hereinbefore described, forming. a fluid sludge which is withdrawn through bottom line I. If the vapors are admitted at the top they are withdrawn near the bottom through line 8 at a point sufiiciently above the sludge line! to permit a clean separation of sludge and vapors. The treated vapors may be passed through line 9 directly to condenser I0, in case tower I only is in operation. If both towers I and 2 are in use, the vapors may pass through reheater II in line I2 which serves to readjust the temperature and the reheated vapors enter tower 2 near its top. If desired the reheater Il may be bypassed through line 2|. The vapors in tower 2 in contact with the sulfuric acid solution from line i produce a small amount of sludge which is removed through bottom line I3 and is collected in tank I4 together with sludge from tower I and line I. Vapors treated in tower 2 leave through line I5 near the bottom and pass through line 9 to condenser I0 where they are condense-d, condensate being accumulated in tank Ill.

A similar procedure applies, if the raw distillate vapors enter tower I through line 6 near its bottom. Treated vapors are withdrawn through top line H and may pass to condenser I0; or they may go to the bottom of tower 2 through reheater I9 situated in line l8 orby-pass 22, to be further treated. Retreated vapors leave tower 2 through top line 20 passing to condenser I0 and tank I6.

Steam may be injected into towers I and/or 2 if desired at some convenient points through lines 23 and/or 24 respectively, to facilitate the vaporization of the distillate and to help regulate the hypothetical potential concentration of the reagent solution. In some cases it may be advantageous to use concurrent flow in treater I and countercurrent in treater 2 or vice versa, depending on the refractoriness of the charging stock. I

The following examples further serve to illustrate my process:

Example I A cracked, sour, highly unstable gasoline having an induction period of 45 minutes when sweetened with silver nitrate was washed with aqueous caustic soda and dilute sulfuric acid of about 50% concentration to remove alkyl phenols and nitrogen bases therefrom. The washed gasoline was vaporized and a .5% aqueous solution of sulfuric Polymerization loss per cent by weight 2.5

Sludge loss do- .3 Recovered gasoline do 97.2

Induction period hours 4 Upon addition of 001% of a commercial inhibitor (U. 0. P. No. 2) the induction period was raised to 6% hours.

Example II The washed gasoline of Example I was treated in concurrent flow in the vapor phase at 280 F. with an aqueous .5% solution of sodium bisulfate in an amount of 12% by volume of the liquid gasoline (equal to .156 lb. of sodium bisulfate per barrel of gasoline), for 5.5 seconds. The poly merization loss was 1.8% and the properties of the treated gasolinewere similar though slightly inferior to those of Example. I.

Example III A cracked gasoline distillate, having a specific gravity of .745, was washed with dilute solutions of caustic soda and sulfuric acid, and was treated countercurrently in the vapor phase at 280 F. and atmospheric pressure with a 2% aqueous solution of sulfuric acid with. the results shown below:

Contact time, in seconds 8 9 11 12 19 Total loss, percent by weight 1.6 9 1. 0 1. 8 1.4 Recovered gasoline, percent by 7 weight 8. 4 99. l 99. 0 98. 2 98. 6 Acid consumption, lbs./bbl .5 7 .35 17 Properties of treated gasoline after sweetening: Saybolt color 30+ 30+ 30+ 30+ Induction period, hours. 213 1% 1% 1 1% Induction period,'hours, after addition of .001% by weight gum inhibitor (U. 0. P. No. 2) 2% 2% 2% 2 2% It will be noted above that amount of acid consumed and time of contact are interchangeable within limits, to secure substantially identical results.

Example IV The gasoline of Example III was treated countel-currently at 230 and 240 F., respectively, under subatmospheric pressure of 600 mm. mercury to secure complete vaporization, with .35 lb./barrel sulfuric acid in a 2% aqueous solution. The treated gasoline so obtained had the following properties:

Treating temperature, F. 240 Saybolt color 24 Copper dish gum mg./l00 m 5 9 A. S. T. M. gum .8 2 Induction period, hrs 1% 1% Induction period, hrs, after addition of .00l% by weight gum inhibitor (U. 0. P. No. 2) 3% 3? l 0 Inhibitor susceptibility 200 Inhibitor susceptibility is defined as minutes increase in induction period per mg. inhibitor in m1. of the gasoline.

When the gasoline of this example was treated by conventional liquid phase sulfuric acid treating method at roomtemperature with .35 and.

1.0 lb. sulfuric acid per barrel, respectively, the

resulting treated gasoline had the following Eat-ample V A cracked gasoline washed with dilute solutions of caustic soda and sulfuric acid was treated in the vapor phase in two stages with dilute aqueous solutions of sulfuric acid using concurrent flow of the reagent and vapors through the individual stages. Conditions were as follows;

1st stage 2nd stage Temperature F 230 250 Concentration of solution percent.. 5 2

The total amount of acid consumed was 97%,

and the total time of contact was 27 seconds. The resulting gasoline hada Saybolt color of 30+ and a gum inhibitor susceptibility of 330.

oil, which amount of reagent is completely conthe vapors at the temperature of the treatment,

maintaining said contact for a time sufiicient to vaporize water of solution, thereby concentrat ing the reagent solution and simultaneously to react at least a portion of said unstable components with the reagent, whereby a sludge is produced, and separating the resulting vapors from the sludge. I

2. In the process for removing gum forming components from hydrocarbon oil vapors in a heated condition with a refining reagent, the improvement comprising injecting into said heated hydrocarbon vapors an aqueous solution having a concentration from .2 to 20% of a reagent consisting of sulfuric acid, in an amount of from .02 to 1 pound of said reagent per barrel of said oil, which amount of reagent is completely consumed in a once-through operation under the.

conditions of the treatment, conducting such vapors and reagent solution in intimate contact through a reaction zone at a temperature of 200 to 400 F., said amount and concentration being so chosen to prevent substantial condensation of the vapors at the temperature of thetreatment, maintaining said contact for a time sufficient to vaporize water of solution, thereby concentrating the reagent solution while reacting a portion of said unstable components with the reagent, whereby a sludge is produced, extending the contact of the reacted vapors and the sludge suf-' ficiently to convert at least a portion of the remaining unstable components to more stable products, and separating the resulting vapors from. the sludge.

3. In the process of removing gum forming components from-hydrocarbon oil vapors in a heated condition with a refining reagent, the

improvement comprising passing said heated hydrocarbon vapors serially through several reacting zones, injecting into each zone aqueous solutions having concentrations between .2 and 20% of a reagent consisting of sulfuric acid in amounts between .02 and 1 pound of said reagent per barrel of said oil, which amounts of reagents are completely consumed in a once-through operation through said reaction zones under the conditions of the treatment, providing for intimate contact of said vapors and reagent solutions' in the several zones at temperatures betweenZOO to 400 F., chosen to prevent substantial condensation of said vapors, maintaining said contacts for periods of time sufficient to vaporize water of solution and to react at least a portion of said unstable components with the, reagents, thereby said amounts being so producing sludge, and separately removing sludge from each zone.

4=. In the process of removing gum forming components from hydrocarbon oil vapors in a heated condition with a refining reagent, the

improvement comprising passing said heated hydrocarbon vapors serially through several reaction zones having progressively lower temperatures between 200 and 400 F., injecting into the several zones aqueous solutions having concentrations between .2 and of a reagent consisting of sulfuric acid in amounts between .02 and 1 pound of said reagent per barrel of said oil, which amounts are completely consumed in a oncethrough operation through said reaction zones under the conditions of the treatment, providing for intimate contact of said vapors and reagent solutions in several zones, said amounts and concentrations being so chosen to prevent substantial condensation of said vapors, maintaining said contacts for periods of time sufficient to vaporize water of solution and to react at least a portion of said unstable components with the reagents, thereby producing sludge, and separately removing sludge from each zone.

5. The process of claim 1 in which the hydrocarbon vapors and the reagent are conducted through the reaction zone in concurrent flow.

6. The process of claim 1 in which the tern

Images