CA1103188A - Method of dispersing iron sulfide in light hydro- carbon (c.sub.1-c.sub.6) distillation columns - Google Patents

Abstract

Abstract of the Invention The present invention is directed to a method of dispersing iron sulfide to avoid deposition thereof in light hydrocarbons (Cl-C6,i,e., methane through hexane including isomers thereof) distillation columns in gas purification. More specifically it has been discovered that iron sulfide deposits commonly encountered in light hydrocarbon distillation columns can be controlled by adding to the light hydrocarbon a dispersant agent produced by the reaction of a succinic acid or anhydride derivative with a polyamine.

Description

li~3i~B

Back~ro~nd of the_Invention Process equipment fouling caused by the formation of adherent deposits which interfere with equipment operation is a problem of the petroleum industry. In many instances fouling and resulting economic losses can be avoided or greatly reduced by use of special chemicals.
Fouling deposlts may be classified as inorganic and/or organic. Inor-ganic deposits such as oxides and sulfides and other salts of iron, copper and vanadium may be present in the original hydrocarbon feed or may be present as a result of corrosion where fouling is evidence upstream of the fouling problem area. Iron sulfide which is the problem to which the present invention is directed, is present in most light hydrocarbon distillation systems. The iron sulfide is either present in the light hydrocarbon being distilled or is formed by the reaction of residual sulfur with iron derived from the corrosion of the metallic part in contact with the product to be distilled. The deposition of iron sulfide results in the necessity of taking off-line heat exchange equipment and reboilers for cleaning thereby decreasing productivity.
Because of the iron sulfide deposition and fouling generally experienced in light hydrocarbon distillation, a research pro~ect was initiated to develop dispersants for iron sulfide in the hopes of eliminating or at least mitigating the problems caused thereby.

During the development tests it was discovered that a family of compounds when used at certain dosage rates, provided the desired disper-sion of the iron sulfide. Of interest is the fact that one species of this family has been utilized in formulations with other important ingredients for other purposes in hydrocarbon streams. However the species has always been a minor component of these formulations and the formulations themselves were never used or recommended at the dosages where the species was able to provide the dispersant effect. The formu-lations were recommended as antifoulants at upper limits of 25 parts per million parts of hydrocarbon stream with the species being approximately 11~3~88 8.9~ by weight of the formulations. Accordingly, the amount of compound of the species actually used was in the range of 2.2 parts per million.
It was discovered that in order for the family of compounds to show the efficacy with respect to iron sulfide dispersion a great deal more of the compound had to be used than used as contained in the formulations.
To obtain the dispersing effect, a minimal range of 18 to 500 parts per million parts of light hydrocarbon and preferably in the range of 25 to 250 parts per million had to be used.
The compounds used in accordance with the invention, as earlier stated, are low molecular weight, hydrocarbon soluble reaction products, a succinic acid or anhydride derivative with a polyamine. The reaction products are commercially available from Lubrizol Company and have been described in U.S. Patents, 3,271,295 and 3,271,296. The succinic-amine reaction dispersing agents have found according to the above patents use as antifoulants in other areas of crude processing, gas-oil production (heavy oil, e.g., fuel), residuum (tars), naphtha and light distillate (straight run gasoline). These areas of use are quite distinct from the light hydrocarbon (Cl-C6), e.g., methane through hexane and isomers thereof, which are being ~reated in accordance with the present invention.
The additives employed in the present invention are the reaction products of H _ _ H
CH - CO

H - C0 n or a substituted succinic acid having the following formula:

-,~

-I --H ~ -I
C~ - CO~
~1 _ ~2- COOH n in which R is alkyl or alkenyl radical having from 30 to 200 carbon atoms in the carbon chain, and n is an integer between 1 and 5 inclusive, with a polyamine of the formula:

~ 2N (CnH2nN)y H
in which n ~s an 1nteger between 2 and 10 inclusive, y is an integer between l and lO inclusive, and CnH2n is a straight chain hydrocarbon group.

In the above reaction, from l/2 to 2 chemical equivalents of poly-amine are used for 1 chemical equivalent of succinic compound.

The synthesis of compounds used in the present application is described in British Patent 922,831, published April 3, 1963, for Met~l-Free Lubr1cant Additives. The alkyl succinic anhydride discussed in this reference was prepared from a chlorinated polyisobutylene and maleic an-hydride. The compounds of similar character discussed in the remainder of this disclosure were prepared by thermal condensation of non-chlorinated polyisobutylene and maleic anhydride. Either a chlorinated or a non-chlorinated polyisobutylene may be used to prepare additives for the purposes of the present inventlon. The alkyl or alkenyl radicals ~mono or poly unsaturated) substituted in the succinic acid or succinic an-hydride are con~only obtained from polyolefins such as pol~ethylene, polypropylene, polybutylene or copolymers of styrene or any other alkenyl ~3~8~

group capable of forming copolymers with maleic anhydride or maleic acid. These substituent groups are large, having between 30 and 200 carbon atoms in the molecule and preferably between 30 and 100 carbon atoms in the molecule, both inclusive. The reaction involves thermal condensation of these reactants at temperatures between200 and 300 C.
The copolymer resulting is predominantly aliphatic in nature.

The most commonly used sources of these substantially aliphatic hydrocarbon substituents are the polyolefins. Examples are polyethylene, polypropylene, polylsobutylene, and so on up the series. A particularly desirable polyolefin for this use is polyisobutylene present in polymers having a molecular weight between 600 and 1,000 inclusive.

It has been noted that at least 1/2 of a chemical equivalent of the amine per equivalent of substituted succinic acid or succinic anhydride should be used in the synthesis to produce a satisfactory product with respect to iron sulfide dispersing properties, and it is preferable to use one equlvalent of polyamine per equivalent of succinic compound. It is not desirable to use more than two equivalents of polyamine per equivalent of succinic compound.

~le reaction is believed to involve spontaneous salt formation and half amide formation upon heating to temperatures between 80 and 120 C.
The third step is better carried out by reacting in xylene and removing the water by azeotropic distillation from 150 to 180C.

Structural formula for the family of compounds which are preferred is determined to be R~ ~
O N ~0 (Cll~)n 0-~=0 R~

11~31~8 where n is an integer from 2 to 6 and R is an alkyl or substituted alkyl of from 30 to 200, and preferably 30 to 100, carbon atoms. Molecular wei~ht does not appear to be critical except that it should not be so large as to render the additive insoluble in the distillation medium.

Example 1 through Example 11 of the U.S. Patent 3,271,295 illustrate the preparation of the type products which are used in accordance with the present invention. The products were obtained utilizing the reaction of isobutylene with maleic anhydride to form the succinic derivative which was further condensed with the polyalkylene polyamine, such as ethylene diamine and triamine.

Specific Embodiments Example 1 The product which is particularly significant was produced by obtaining the commercial product referred to as Lubrizol 890 which contains sixty percent (60%) of the isobutylene succinic anhydride -ethylene diamine reaction product in a hydrocarbon solvent (Example 4 of U.S. 3,271,295). Forty-two percent (42%) by weight of the commercial product was blended with a fifty-eight percent (58%) heavy aromatic naptha producing a product containing twenty-five percent (25%) of the reaction product. The product obtained will be referred to as the Product of Example 1 in the subsequent Tables and is referred to as such in the drawings.

At this time it should be pointed out that all of the other Lubrizol products tested were produced in the same manner in the same concentrations.
Accordingly, each product produced from the Lubrizol commercial products contained twenty-five percent (25%) by weight of the actual isobutylene -maleic anhydride - amine product.

Test Procedure Room Temperature 1. Products to be tested were added to 5 mls laurelsene (kerosene) -`

!~' ~ i, 1~3~38 in 4 oz. bottles. (Laurelsene was chosen as the test environment since it is essentially a non-fouling medium. It contributes no polymerization products or impurities, therefore allowing a true picture of the dispersive effect of the additives. This effect was subsequently confirmed with actual field trials ~hich are described later herein).

2. 100 ppm iron (Uversol iron liquid 6% - Harshaw Chemical Co.) dilution in laurelsene was prepared.

3. The iron-laurelsene solution was purged for ten minutes with hydrogen sulfide (500 mls solution in 1 liter flask).

4. 60 mls samples of resulting iron sulfide suspension were immediately pipetted into the 4 oz. bottles containing the products to be tested, placed on a shaker for one minute, then allowed to sit for six hours.

5. 5 mls samples of the iron sulfide "suspension" were then pipetted and placed in 50 mls laurelsene (sampling pipette extends approx-imately 2/5 of the distance below the surface of the suspension).

6. Samples were shaken, unstoppered and allowed to sit exposed to the atmosphere for 18 hours.

7. Samples then shaken for 30 seconds. 25 mls then pipetted lnto a sample. cell and turbidity measurements made using a Hach Analytical Nephelometer (Model 2424).

8. The unit of measure used in this instrument is a Nephelometric Turbidity Unit (NTU). High readings indicate good dispersion, low readings lndicate poor dispersion.

Hach Analytical Nephelome~e.r - Principle of Operation The Analytical Nephelometer operates on the principle that light passing through a substance, is scattered by particulate matter suspended in the substance. A strong light beam is sent upward through a cell ~ 7 --lill3~

containing a sample. As the beam passes through the turbidity particles, an amount of llght (proportional to turbidity present) is scattered at a 90 angle to the beam and is received by a photomultiplier tube. This light energy is, in turn, converted to an electrical signal which is measured by the instrument panel meter.

Discussion The test as developed is a comparative technique for evaluating products relative to one another. Very low turbidimeter readings were obtained with the black sulfide particles and solution so the "Test Procedure" technique was employed. The 18 hour air exposure modified the black color of solution and particles to an orange-brown coloration by which significant turbidimeter numerical values were obtained. The actual particles being measured after air oxidation may be some combination of sulfides, hydroxides or oxides of iron composed however primarily of iron sulfide. Since the time duration of each step in the test was kept consistent, the test results should be valid for comparative purposes.
Also, visual observations support the data in that low turbidimeter readings were equated with samples,that are seen to have precipitated and high turbidimeter readings are equated with samples that are very black in appearance.

Table 1 which follows shows that it requires significantly more of the commercial product containing Lubrizol 890 to produce the results achieved by the product of Example 1 at lower temperatures. As earlier stated, the commercial products have not and would not likely be fed to hydrocarbon at the rates where the efficacy was noted.

11~318~3 ~ . .
g o o o C~l .~ ~ ~

~ o o o o u~ r~

o o o U~ ~

_ o o o g ,~ ~ o ~ ~ C~
o o C~
._ O d- r~
~ UO~
L
Y ~
O O ' L g C~
C O ~ ~
~ D
L

tn ~ r~ cn ~_ E ., L ~ ~0 U~

C~
_ o~
C~l C U~
O

~J ~0 ~0 O
o~o a~ ~c~
o, o, CO L O L O
O ~ ~ r~l ~ ,~
L -- L
~_ ~ J ~ J
~IJ N J ~J L L
C~ ~1 11 0 ~ ~ F. ~
~E ~ t~l c la o co o co C~
o ,~

11~31~38 Table 2 establishes the ef~icacy of the family of succinic acid or anhydride derivatives (isobutylene/maleic anhydride reaction product) -a~ine condensatlon products as regards to dispersing of iron sulfide.

~0 X

l o l~ lB8 -. ., E
O O O
O r~ O~
U~ . <~ ~ ~

O 1~ 1~ E
C~
~?

. ~

I ~ol o _ I o ~ ~
J L~) o ~ ~
Cl~ E C C
~e,~ O
L L ~ tO CO

C N . t 1~
~ ~ N
1~1 . ~a~
=) . SC o ~ ~ 1.~) CO C C
X O ~ ~ ~ ~
V~

C C ~
5 o C _ '~ o U~
~ ,.
_ _ o ~
E ~ 0 u~ . . . .. . . .....
-Addi tional C.t,~ s Effect of Tron Concentration -Figure l provides the graphical test data for the effect of iron concentration on NTIJ value in hydrocarbons. Preliminary test data had ~ndicated that the product of Example l was an effective iron sulfide dispersant at concentrations of lO0 ppm and greater. A steady state value, i.e.,an inhibitor concentration where consistent reproducibility was obtained, 250 ppm Product of Example l, was chosen for the effect of lron concentration turbidity measurements.

As expected, these data show an increase in NTU value with increas-~ng iron concentration. The decrease in NTU value after one hour is greater at 75 and lO0 ppm iron concentration than at 25 or 50 ppm. All samples show a leveling effect after 3 hours. Since this was a compara-tive rather than an absolute test, the iron concentration of lO0 ppm with the highest NTU measurements was chosen as the standard iron test concentration.

Time Dependency Study Figure 2 gives the graphical test data for the effect of time on the settling rate of an inhibited sample vs the control over the time ~nterval 0-6 hours. The control sample data indicate that the majority of the iron sulfide precipitates within the first hour while a signficant port~on of the iron sulfide in the inhibited sample remains dispersed for s~x hours. These data 1ndicate that six hours was sufficient set-tling time for comparative test purposes.

The difference between the NTU value of the control and that of the - ~nhibited sample at 0 time may be due to particle size and number.
Visual observations indicated that the uninhlbited or control iron sulfide precipitate appeared mor~ granular or fla~y than the inhibited .

.
~3~8B

sample. The inhibited sample appeared to be a very fine, mud-like precipitate. There is evidence in the literature that detergent-dispersants have the ability to peptize '(break up into small particles) sludge deposits, causing them to be dispersed and kept in the colloidal suspension. It is probable that the peptizing action of the inhibitor accounts for'the difference in NTU value at 0 time.

Temperature Dependency Study Figure 3 provides the graphical test data for the effect of temper-ature on the settling rate of an inhibited sample of iron suIfide in hydrocarbon at 23C and 84C. Test conditions were as indicated in "Experimental" except the 84C samples were placed immediately in the oven for 8iX hours and sampled immediately upon being removed from the oven. These data compare the efficacy of the product of Example 1 in preventing iron sulfide precipitation over the concentration range of 25-1000 ppm at the two test temperatures.

These data show a decrease in inhib~tor efficiency at the elevated temperature of 84C. This decrease i6 reflected in a shift to higher inhïbitor concentrations required at 84C and not to a complete failure of the inhibitor to disperse the iron sulfide particles. This is a static system and a change in test temperature would be expected to change certain critical parameters such as solution viscosity, particle interactions, etc. No study of the effect of temperature other than of dispersant inhibitors was made.

In summary, the effect of increasing the test temperature appeared to be the shift to higher inhibitor concentrations required for effective dispersant inhibitors. Room temperature test conditions were considered adequate for evaluating materials as potential iron sulfide hydrocarbon dispersants.

~Li ~ 39L8~3 Field Evalu~tion A refinery straight run depropanizer reboiler was fouled with iron sulfide and pol~ner deposits to the point where it was impossible to obtain the degree of separation required to make saleable propane from straight-run gasoline. Because the overhead product contained 30 to 40 percent butanes, ~t had to be burned as fuel rather than sold for LPG. It was suggested that an onstream cleanup program be tried using the Product of Example 1.

The cleanup program lasted about one month. After the month cleanup the plant was able to make a saleable 95 percent pure propane p~oduct. By being able to sell the overhead product as LPG rather than burning it as fuel, the refinery now shows an increased profit of ~5,000/day.

This is now a continuous application of the product. The specifics of the cleanup program were as follows:

To remove the foulants gradually, the cleanup operation was carried out in three separate stages:

1. feed to the reboiler, 2. then to the column charge, 3. and, finally to the overhead reflux line.

Column feedrate is 37,000 barrels/day with about 5000 barrels/day of overhead product.

1. 30 gallons/day of the Product of Example 1 (23 ppm) were injected to the reboiler inlet line for 10 days. After this time, the ternp-erature drop of the reboiler heat medium fluid increased from 85F
to 140F and the temperature drop on the tower side of the reboiler ~ncreased from 19F to 27F. Both of these temperature drop increases are indicative of reboiler cleanup.

2. 40 sallons/day of thc Product (26 ppm) wcre then injected into the tower feed for two weeks. After one week, the butane and heavier content of ~he overhead product had decreased to nearly 20 percent.

3. On a pemlanent basis, 40 gallonsJday of the Product (30 ppm) are ~njected into the overhead ref7ux line. The overhead product is now 95 percent pure in propane.

This cleanup case history indicates the effectiveness of the Product of Example 1. For best results the treatment should be started with clean equip~ent. Feedrates for removing iron sulfide deposits are somewhat lower than those used for dispersion, i.e., 10 to 50 parts of the Product per million parts of light hydrocarbon.

Having thus described the invention, what is claimed is:

. ~15-

Claims (8)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY OR
PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method of dispersing iron sulfide in a light hydrocarbon dis-tillation column which comprises adding to the hydrocarbon a compound or compounds having the formula wherein n is an integer of 2 to 6, and R is alkyl or substituted alkyl of 30 to 200 carbon atoms and the compound is added in an amount of 18 to 500 parts per million parts of light hydrocarbon.

2. A method according to claim 1, wherein the light hydrocarbon stream is composed of C1 - C6 hydrocarbons.

3. A method according to claim 2, wherein n is an integer of 2 and R is an alkyl of 30 to 100 carbon atoms.

4. A method according to claim 3, wherein the compound is added in an amount of 25 to 250 parts per million parts of light hydrocarbon.

5. A method of removing iron sulfide deposits in a light hydrocarbon distillation column which comprises adding to the light hydrocarbon a compound or mixture of compounds having the formula where n is an integer from 2 to 6, and R is an alkyl or substituted alkyl of from 30 to 200 carbon atoms and the compound is added in an amount of 10 to 50 parts per million parts of light hydrocarbon.

6. A method according to claim 5, wherein the light hydrocarbon is composed of C1 to C6 hydrocarbons.

7. A method according to claim 6, wherein n is an integer of 2 and R is an alkyl of 30 to 100 carbon atoms.

8. A method according to claim 7, wherein the compound is added in an amount of 10 to 25 parts per million parts of light hydrocarbon.