Field of the Invention

The invention relates to chemical compositions and methods for scavenging

sulfhydryl compounds, particularly hydrogen sulfide (H ₂ S), from "sour" aqueous

and hydrocarbon substrates. More particularly, the invention relates to hydrocarbon

soluble sulfhydryl scavengers comprising preferably substantially water free

bisoxazolidines.

Background of the Invention

The removal of H ₂ S from a liquid or gaseous hydrocarbon stream is a

problem that has challenged many workers in many industries. One such industry

is the petroleum industry, where the H ₂ S content of certain crudes from reservoirs

in many areas of the world is too high for commercial acceptance. The same is true

of many natural gas streams. Even where a crude or gas stream contains only a

minor amount of sulfur, the processes to which the crude oil or fractions thereof are

subjected often produce one or more hydrocarbon streams that contain H ₂ S.

The presence of H ₂ S in hydrocarbon streams presents many environmental

and safety hazards. Hydrogen sulfide is highly flammable, toxic when inhaled, and

strongly irritates the eyes and other mucous membranes. In addition, sulfur-

containing salts can deposit in and plug or corrode transmission pipes, valves,

regulators, and the like. Flaring of natural gas that contains H ₂ S does not solve the

problem for gas streams because, unless the H ₂ S is removed prior to flaring, the

combustion products will contain unacceptable amounts of pollutants, such as sulfur

dioxide (SO ₂ )— a component of "acid rain. "

Hydrogen sulfide has an offensive odor, and natural gas containing H ₂ S often

is called "sour" gas. Treatments to reduce or remove H ₂ S from hydrocarbon or

other substrates often are called "sweetening" treatments. The agent that is used to

remove or reduce H ₂ S levels sometimes is called a "scavenging agent."

The problem of removing or reducing H ₂ S from hydrocarbon substrates has

been solved in many different ways in the past. Most of the known techniques

involve either (a) absorption, or selective absoφtion by a suitable absorbent, after

which the absorbent is separated and the sulfur removed to regenerate and recycle

the absorbent, or (b) selective reaction with a reagent that produces a readily soluble

product. A number of known systems treat a hydrocarbon stream with an amine,

an aldehyde, an alcohol, and/or a reaction product thereof.

Previously known sulfhydryl scavengers theoretically may require about

2-3 ppm of scavenger per ppm of hydrogen sulfide; however, the amount actually

required is much higher— in the range of about 5-10 or more ppm per ppm of

hydrogen sulfide. A high amount of scavenger is required because of the difficulty

of distributing the scavenger evenly throughout the fluid. Much of this difficulty

is the result of inadequate solubility of the scavenger in the hydrocarbon substrate.

A continuing need exists for effective and efficient processes and composi¬

tions to reduce and/or remove sulfhydryl compounds from hydrocarbon substrates.

Summary of the Invention

The present invention provides a method for scavenging sulfhydryl

compounds from hydrocarbon substrates using bisoxazolidines.

Brief Description of the Drawings

Fig. 1 is a Table giving the results of Example 2.

Fig. 2 is a chart of the results in Fig. 1.

Fig. 3 is a Table giving the results of Example 3.

Detailed Description of the Invention

The scavenging agents of the present invention may be used to treat

hydrocarbon substrates that are rendered "sour" by the presence of "sulfhydryl

compounds," such as hydrogen sulfide (H ₂ S), organosulfur compounds having a

sulfhydryl (-SH) group, known as mercaptans, also known as thiols (R-SH, where

R is a hydrocarbon group), thiol carboxylic acids (RCO-SH), dithio acids (RCS-

SH), and related compounds.

A wide variety of hydrocarbon substrates can be treated using the scavenging

agents of the present invention. The term "hydrocarbon substrate" is meant to

include unrefined and refined hydrocarbon products, including natural gas, derived

from petroleum or from the liquefaction of coal, both of which contain hydrogen

sulfide or other sulfur-containing compounds. Thus, particularly for petroleum-

based substrates, the term "hydrocarbon substrate" includes wellhead condensate as

well as crude oil which may be contained in storage facilities at the producing field.

"Hydrocarbon substrate" also includes the same materials transported from those

facilities by barges, pipelines, tankers, or trucks to refinery storage tanks, or,

alternately, transported directly from the producing facilities through pipelines to

the refinery storage tanks. The term "hydrocarbon substrate" also includes product

streams found in a refinery, including distillates such as gasolines, distillate fuels,

oils, and residual fuels. As used in the claims, the term "hydrocarbon substrate"

also refers to vapors produced by the foregoing materials.

Preferred substrates for the bisoxazolidines of the present inventions are

those in which the presence of water can be detrimental. Such substrates include,

but are not necessarily limited to dry crude oils and fuels, such as natural gas,

particularly dry natural gas condensates.

The scavenging agents of the present invention preferably have the following

general formula:

wherein n is between about 1-2 and R ¹ and R ² independently are selected from the

group consisting of hydrogen, phenyl groups, and linear, branched, and cyclic

alkyl, alkenyl, and alkynyl groups having between about 1- 6 carbon atoms. In a

preferred embodiment, n is 1 and R ¹ and R ² independently are selected from the

group consisting of phenyl groups, and linear, branched, and cyclic alkyl, alkenyl,

and alkynyl groups having between about 1- 3 carbon atoms. A most preferred

embodiment is 3,3' methylenebis-[5-methyl oxazolidine], in which n is 1 and R ¹

and R ² are methyl groups.

While specific examples of R ¹ and R ² have been described, R ¹ and R ² may be

any substituent that does not substantially interfere with the solubility of the

bisoxazolidine in the hydrocarbon substrate. Materials with equivalent properties

should include products of the reaction of 1, 2 or 1, 3 amino alcohols containing 3-7

carbon atoms with aldehydes containing 4 or fewer carbon atoms. A substituent

"substantially interferes" with the solubility of the bisoxazolidine if the

bisoxazolidine cannot be rendered readily soluble in the substrate with the use of an

acceptable cosolvent. In this regard, when R ¹ and R ² are hydrogen, a cosolvent may

be required to maintain the solubility of the bisoxazolidine. A preferred cosolvent

in such instance comprises between about 10-50% BUTYLCELLOSOLVE™, a

monobutylether of ethylene glycol available from Union Carbide, and between about

50-90% FINASOL™, available from Fina Oil & Chemical Co. , Dallas, Texas.

The bisoxazolidines of the present invention exhibit a high uptake capacity

for hydrogen sulfide, and the raw materials required to manufacture the

bisoxazolidines are low cost materials. Bisoxazolidines may be made by reacting

an alkanolamine, with between about 1.1 to 2.1 equivalents, preferably 1.5

equivalents, of paraformaldehyde to yield an aqueous solution of reaction products.

In a preferred embodiment, monoisopropanolamine (MIPA) is reacted with

paraformaldehyde to form an aqueous mixture which, after distillation, yields

substantially water free 3,3'-methylenebis[5-meethyloxazolidine]. The water

formed by the reaction preferably should be removed by distillation, preferably

after the reaction is complete, to give a substantially water free bisoxazolidine. In

this preferred embodiment, the reaction takes place at ambient pressure and at a

temperature of between about 100-200°C (212-392°F). Preferably, the resulting

bisoxazolidine should contain less than about 20% water, most preferably less than

about 5% water.

Bisoxazolidines are commercially available in Europe as preservatives for

oil base paints and fuel oils. An example of such a product is GROAN-OX™,

which is commercially available from Sterling Industrial, UK. The bisoxazolidine

preferably should be added to the hydrocarbon substrate at a high enough

temperature that the substrate is flowable for ease in mixing. The treatment may

take place at temperatures up to the temperature at which the material being treated

begins to decompose. Preferred treatment temperatures are between ambient to

about 200°C (392°F).

The hydrocarbon or aqueous substrate should be treated with the

bisoxazolidine until reaction with hydrogen sulfide, or with other sulfhydryl

compounds, has produced a product in which the sulfhydryls in the vapor (or liquid)

phase have been removed to an acceptable or specification grade product.

Typically, a sufficient amount of bisoxazolidine should be added to reduce the

sulfhydryls in the vapor phase to at least about 200 ppm or less.

In order to determine how much bisoxazolidine to add to a given substrate,

the amount of H ₂ S in the vapor phase above the hydrocarbon may be measured.

The bisoxazolidine may be added to the hydrocarbon in an amount equal to about

2/3-1 ppm by weight of scavenger per 10 ppm by volume of H ₂ S concentration in

the vapor phase. Alternately, the total concentration of hydrogen sulfide in the

system can be measured, and a molar ratio of between about 1/3-2/3 mole of

bisoxazolidine to 1 mole of hydrogen sulfide in the system may be added. The

molar amount of bisoxazolidine added as a scavenger should be proportional to the

molar amount of sulfhydryl compound(s) present in the substrate and will depend

on the level of sulfhydryl reduction required. Hydrogen sulfide contents of up to

about 100,000 ppm in the vapor phase may be treated satisfactorily with the

bisoxazolidines of the present invention. The bisoxazolidines will be most

effective if the substrate is treated at temperatures between ambient to about 200°C

(392°F).

The invention will be better understood with reference to the following

examples:

Example 1

In a liter flask was placed 600 gm of monoisopropanolamine (MIPA).

The MIPA was stirred and cooled in a water bath. Paraformaldehyde was added

in three equal portions. During the first two additions, the pot temperature reached

a maximum of about 95 °C (203 °F). The second and third portions of

paraformaldehyde were added after the mixture had cooled to about 65°C (149°F).

After the third portion of paraformaldehyde was added, the mixture was warmed

and kept at 95 °C (203 °F) until all of the paraformaldehyde had dissolved. The

mixture was gradually warmed to 140°C (284°F) and about 242 gm of distillate

were collected. The material remaining in the flask was determined to be essentially

pure 3,3'-methy-enebis-[5-methyloxazolidine].

Example 2

The following basic protocol was used for each of Examples 2-3:

Septum bottles were half filled with hydrogen sulfide laden marine or No.

6 fuel oil from a Louisiana refinery. The head spaces were blanketed with

nitrogen. The bottles were septum sealed and placed in an oven at 65°C (149°F).

After 18 hours, samples were shaken and the head spaces were analyzed for

hydrogen sulfide by withdrawing a known volume from the head space with a gas-

tight syringe. The sample (or a dilution of the sample in air) was injected into a

gas chromatograph (GC) and the area counts of hydrogen sulfide measured. The

results were noted as the initial vapor phase hydrogen sulfide concentration for

comparison to final readings.

A known amount of the candidate and comparative materials were injected

into all of the sample bottles except controls. The control bottles were designated

blanks (i.e., untreated). The bottles were shaken vigorously for 30 seconds to mix

the additives into the oil, and placed in an oven at 65.5°C (150°F). The bottles

were shaken periodically, and samples of the head space vapor were withdrawn

using a gas tight μL syringe at various intervals. The samples were analyzed by gas

chromatography. If the measured amount of vapor phase hydrogen sulfide was not

significantly abated, the process was repeated after additional incremental injections

of candidate.

The hydrogen sulfide content of the head space in the samples and the control

were calculated by comparing the area counts with a standard curve for hydrogen

sulfide. The results are shown in the respective Figures.

The efficacy of the candidate may be expressed as the treatment effectiveness

ratio ("TER"). The TER is defined as

PPM _V of vapor H ₂ $ abated

PPM.„ of candidate added

The higher the value of "T.E.R. , " the greater the efficacy.

For purposes of this experiment, several products commercially available for

the same purpose (designated "A" and "B") were compared with samples internally

designated "RE-3019" and "RE-3175", which contain 3,3'-methylene bis-[5-methyl

oxazolidinc] and a mixture of reaction products, a major proportion of which

comprises 3,3 '-methylene bisoxazolidine, respectively. The objective was to

produce a series of dosage response curves for the additives.

The oil was dosed to a level of 18,000 ppm H ₂ S and dispensed into the serum

bottles. The bottles were allowed to equilibrate for approximately 2 days. Initial

vapor space hydrogen sulfide concentrations in the serum bottles averaged between

92,000-100,000 ppm-v. The results are given in FIG. 1, and charted in FIG. 2.

Fig.l shows the results for the additives two hours after the first injection of

1500 ppm-w of candidate. The samples were allowed additional reaction time

overnight. The vertical drop line in Fig. 1 shows the additional amount of hydrogen

sulfide abated after 16.5 hours at 1500 ppm-w of each additive. Finally, Fig. 1

displays the results 3.5 hours following the second dosage injection totaling 3500

ppm-w of each additive. The two experimental additives, RE-3019 and RE-3175,

reduced hydrogen sulfide to nearly zero. For chart clarity, the test results for the

replicate run of RE-3175 were not included. The replicate results mirrored the

results for the original RE-3175 sample.

Example 3

The commercial candidates again were compared with RE-3019 and RE-

3175. The commercial candidates were tested in their "as sold" concentrations; RE-

3019 was tested as a 100% concentrate; and, RE-3179 was tested as 80% active

gel dispersed in xylene. The reaction times for all of the samples was slower than

expected, but uniformly so for an undetermined reason.

The results are given in Fig. 3. Both RE-3019 and RE-3179 had a very high

TER- from about 8 to 5 times higher than commercial candidates.

Persons of ordinary skill in the art will appreciate that many modifications

may be made to the embodiments described herein without departing from the spirit

of the present invention. Accordingly, the embodiments described herein are

illustrative only and are not intended to limit the scope of the present invention.