US2533878A - Method of pumping viscous petroleum - Google Patents

Description

Patented Dec. 12, 195A? METHOD OF PUMPING VISCOUS PETROLEUM Albert F. Clark, Los Angeles, and Abraham Shapiro, Monrovia, Califi, assignors to Socony- Vacuum Oil Company, Incorporated, New York, N. Y., a corporation of New York No Drawing. Application May 31, 1949, Serial No. 96,392

4 Claims. (Cl. 137-78) 1 The transportation of extremely viscous petroleums by pipe line has presented great difficulty, particularly during cold weather. Even when pressures near the maximum permissible in standard pipe and pumping equipment are employed, the flow of such oil is so sluggish that special procedures are required to make pipe-line transportation practical.

It has been customary to facilitate the fiow of very heavy petroleum by reducing its viscosity, either through the addition of a hydrocarbon diluent or through the installation of heating equipment at frequent intervals along a pipe line. The former expedient is practicable only in the somewhat unusual case where a supply of light petroleum is available in the same region from which the heavy petroleum is taken, and the latter expedient is inconvenient and costly,

We have forgid that the flow of viscous petroleum in pipe lines can be greatly facilitated by the use of water containing minute proportions of a waterr ace-active agent and of an alkali-metal phosphate and having its pH adjusted to within the range pH 5.7 to pH 7.0, preferably pH 6.5.

The water is placed in the pipe line together with the viscous petroleum, in proportions of about 8% to of the total liquid carried by the line. A portion of the water wets the inner surface of the pipe to form a film or layer between the oil and the metallic surface. This layer, by preventing contact between the metal and the viscous, adhesive oil, acts as a lubricant. The remainder of the water is dispersed through the oil body in the form of large drops and lenses. The dispersed water functions to reduce the overall internal cohesion of the flowing heterogeneous fluid and thus effects much the same result as would be effected by a reduction of the viscosity of the oil.

Water containing only th surface-active agent is capable of preferentially wetting the pipe surface and it is capable of facilitating flow through a short length of pipe line. However, such water eventually forms an emulsion with the oil which has a greater apparent viscosity than the oil itself and which has so much stability that little or no water is available for wetting the pipe surface. Thus, in a long pipe line, such water is of no benefit and may actually decrease the volume of flow obtainable with a given pressure.

The addition of the phosphate and the adjustment of the pH inhibit the formation of waterin-oil emulsions and thereby permit the beneficial effect of the water to extend throughout a long pipe line.

The anionic surface-active agents comprise a well recognized group of partially ionic compounds in which the cation is definitely hydrophilic (an alkali metal, ammonium, or substituted ammonium including no long organic chains) and in which the anion is in on portion of a mostly nonpolar, lipophilic molecular structure. The best known of these agents answer to the formulae RCOOM (soaps), RSO3M (sulphonates), and ROSOaM (alkali-organic sulphates), in which M denotes an alkali metal or the equivalent and R denotes a lipophilic organic radical, frequently a hydrocarbon radical which includes an alkyl chain of ten or more carbon atoms in length. I

These agents generally have som solubility or pseudo-solubility in both oil and water. However, they are classified into three groups on the basis of preferential solubility. The water-soluble agents are those which have greatly preferential solubility in water, the oil-soluble agents are those which have greatly preferential solubility in oil, and the oiland water-soluble agents are those which, when introduced to a two-phase system of oil and water, enter into and remain with both phases to substantial degrees. The agents of the water-soluble group are those which we employ in the present invention, and of these the sulphates, sulphonates, and other agents known to be fully effective in slightly acid environments are greatly preferable.

A particular agent which we have found to be very efficient and which We prefer becaus of its availability and low cost is a water-soluble petroleum sulphonate prepared by treating transformer-oil stock with concentrated sulphuric acid, separating the acid sludge from the treated oil, neutralizing the sludge with a strong aqueous solution of sodium hydroxide or sodium carbonate, and separating the neutralized product into an aqueous layer and an oil layer. We employ the material of the aqueous layer, or the product obtained by evaporating the water from the aqueous solution as the surface-active agent in our method. Corresponding treatment of kerosene stock, absorber-oil stock, and other petroleum distillates of similar boiling range yields agents which are equally useful.

The term petroleum sulphonates customarily used for materials derived in this manner does not accurately indicate the nature of these surface-active agents. They actually are complex mixtures including sodium sulphonates, sodiumorganic sulphates, and sodium-organic sulphites, in which the organic radicals are extremely numerous and varied. Many of the commercially available petroleum sulphonates have been treated with selective solvents for the removal of sodium sulphate and other impurities; such treatment does not decrease the usefulness of the agents for our purpose but it does unnecessarily increase the cost.

Other specific agents which are known to be effective in our method include the sodium and potassium salts of dodecylbenzyl sulphonic acid and decylbenzenesulphonic acid, and numerous water-soluble anionic wetting agents which are known by trade names and trade designations and which, like the petroleum sulphonates, are not susceptible of chemical identification.

The phosphate component may be any alkalimetal phosphate, including the orthophosphates, metaphosphates, and the various polyphosphates. As regards their primary function of inhibiting emulsification, there is little if any difference among the various alkali-metal phosphates, but secondary considerations may indicate the choice of particular phosphates. For example, if the available water is very hard, it is preferable to use polyphosphates having a sequestering elfect, such as sodium hexametaphosphate, to avoid the accumulation of deposits of calcium and magnesium phosphates in the pipe line. A convenient method of bringing the pH of the water to within the required limits is to choose a phosphate or mixture of phosphates having the necessary degree of acidity or alkalinity.

After the addition of the surface-active agent and the phosphate, if the pH of the water is not within the limits 5.7 to 7.0, the solution is adjusted to within that range by the addition of an acidic or basic electrolyte. Sodium hydroxide, sodium carbonate, or trisodium phosphate is convenient if it is required to increase the pH to bring it into the correct range; sulphuric acid, monosodium orthophosphate, or a sodium metaphosphate is convenient if the pH is to be reduced. The use of additional phosphate for this purpose is not harmful, as there is no critical upper limit for the proportion of phosphate in the aqueous solution.

It is not possible to set forth numeric limits for the permissible proportions of the surface-active agent and the phosphate. Petroleum is not a standard material having fixed properties; in particular, it contains varying amounts of natural ingredients (e. g., naphthenic acids) which act as oil-soluble emulsifying agents and wetting agents, and the water therefore requires varying proportions of phosphate and of surface-active agent to overcome the emulsifying and surface-wetting tendencies of the oil. Furthermore, the watersoluble surface-active agents include a vast number of individual compounds which difier from one another as to effectiveness at a particular concentration. Even the water is a nonstandard material, since it contains varying amounts of calcium and magnesium which precipitate or inactivate varying amounts of the phosphate ions. The usefulness of a particular aqueous solution with a particular petroleum may be deter mined in the laboratory by the use of low-speed stirring apparatus having metallic surfaces. Tendency of the oil to adhere to the metal may be readily estimated by inspection, resistance to stirring is a good measure of the overall internal c0- hesion of the heterogeneous liquid, and emulsification may be detected by heating the liquid and observing the degree of separation into layers of oil and water or by microscopic examination of thin layers of the liquid.

Failure of the water to markedly reduce adhesion of the oil indicates that not enough of the surface-active agent is employed. Failure of the water to reduce resistance to stirring indicates that insuificient water is used or that emulsification has occurred. Formation of a water-in-oil emulsion indicates (assuming the pH to be correct) that not enough phosphate is used. A slight tendency to form oil-in-water emulsions is permissible, and while a strong tendency to such emulsification would be disadvantageous, it is unlikely to be encountered unless the proportion of surface-active agent is far in excess of that needed and economically desirable.

We have also employed repeated circulation of petroleum and water through coils of small pipe for testing the utility of aqueous solutions.

Although the entire range of useful proportions cannot be described numerically, it can be stated that any water-soluble anionic surfaceactive agent which is sufificiently effective to be known commercially as an acid-compatible wetting agent can be depended on to yield good results when employed in proportions of 0.02% to 0.04% by weight with respect to the water. The phosphate yields good results when added in proportions of 0.005% or more by weight with respect to the water.

We prefer to use the treated water in proportions of 8% to 15% of the total liquid carried by the pipe line. However, the water is capable of facilitating the flow of viscous petroleum when employed in greater proportions. The upper limit is critical only in the sense that the use of additional water does not further facilitate flow sufficiently to compensate for the additional load which it imposes upon the capacity of pipe and pumps.

While our method permits the elimination of line heaters, it does not in every case permit the pumping of cold oil into a pipe line. This is because the pressure available for forcing petroleum from a tank to the intake of a pump does not exceed the total of atmospheric pressure plus the hydrostatic pressure of the oil in the tank, and such low pressure is often insufficient for causing an adequate flow of extremely viscous petroleum through even a very short passage. Some cases in which the rate of fiow of cold oil from a tank is inadequate may be corrected by the substitution of larger pipe between the tank and the pum intake, but in other cases the application of heat to the tank is the preferred procedure.

The following are reports of three runs of a very heavy petroleum through a particular pipe line, including a run of initially heated oil without the benefit of the aqueous solution and two specific examples of the practice of our method, one with and one without initial heating of the oil.

The line employed is a three-mile length of six-inch pipe which rises approximately 300 feet in its first 3000 feet, thereby imposing a static head of 125 pounds per square inch upon the pump. The delivery end of the pipe line has an overshot which prevents drainage of the last 2000 feet of line between runs. The pump, which operates at constant speed, is provided with a bypass for preventing excessive pressure in the pipe line.

In the first run, which employed no water, the

initial temperature of the oil was 161 F., the

gravity was 13.6 A. P. I., and the M. and B. S. content was 0.1%. Results were:

oiiriiiirii iililiit"? In this run, which was of 46 hours and 40 minutes in duration, the total volume of oil pumped was 925 barrels, yielding an average pumping rate of 19.8 barrels er hour. The temperature of the oil run thepressure increased as trapped air was compressed between the column of fresh oil being pumped into the line and the plug of adhesive, dry oil which remained in the last 2000 feet of at the Point Of v y W e Pi l the line. After reaching a maximum of 330 fluctuated continually between the limits mpounds per square inch, the pressure began to (heated due o the Operatlon 0f the bypass for decrease due-to the displacement of the dry oil. Iehef 0f eXeeSSlVe D Previous to the third run, the customarily The second run wasobserved about a month heated tank from which the petroleum was 0 g z g g gfi g ffi i a pumped was allowed to cool until the temperasllml 0 e g g d b s g gg g g ture at the outlet was 73 F. The water employed p m S r was y in the third run was treated by the addition of of 0.038% by weight of the dried but unrefined 0 067 by weight of the aqueous solution of petroleum sulphonate whose derivation from etroleum Sul honates derived f m tr nsf rm transformer-oil stock is described above, 0.19% m a by weight of commercial trisodium phosphate, on stock f Was.est1mated to coma}? and enough sulphuric acid to bring the pH of surface-act1ve material, and by the addition of the water to approximately 6.0. The petroleum, 013% commercflal PrISOdmIP phosphate and drawn from the same group of wells as that of enough su1phun a1d to brmg the PH of the the first run, had an initial temperature of 146 watfar to approxlmately The Petroleum F" a gravity of 13 70 P. I. and and B. s. again drawn from the same group of wells, had content of 0.1%. Initial temperature of the a gravlty and and content of Water was 471: R As a preliminary to the pump- 0.1%. linitial temperature of the water was 60 mg of on enough of the treated Water was F. As 1n the second run, enough treated water pumped into the line to fill the first 2000 feet; was pumped into the 9 fill h first 2000 this was for the purpose of reducing the abnormal feet, then to determme 1f the 011 would fiOW resistance to flow in the early part of the run 111 suflieleht Volume the temperature to which would have been caused by the dry oil the pump mtakethe 011 w s ted a d pumped (left from previous use of the line) adhering to alone a period O an 110111 and orty m nutes. the inner surface of the pipe. Then pumping of The tank gauge Showed t a t e il Wa Cap ble oil was begun, with continuous addition of of flowing to the pump at a rate of at least 48 treated water on the suction side of the pump, barrels per hour. Finally the oil was pumped while the following results were observed. with continuous addition of treated water for a Oil pumped dur- Water pumped during Gross pumping Gauge lug period period during period Time Pressure,

lbs/sq. in.

Bbls. BblsJhr. Bbls. Bbls/hr Per cent Bbls Bbls hr.

11:00 a.m 200 11:30 a.m 150 51 102 4.9 9.5 5.5 55.9 111.5 Noon 150 22 44 5.5 7.0 13.7 25.5 51.0 1:00 p. 150 55.5 55.5 5.9 5.9 11.1 52.4 52.4 2:00 p. 200 45 14.4 14.4 24.2 59.4 59.4 5:00 p. 220 57 57 4.5 4.5 7.5 51.5 51.5 4:00 p. 210 59 59 5.9 5.9 10.5 55.9 55.9 5:00p. 200 55 4.5 4.5 5.0 57.5 57.5 5:00 p. 210 57 57 9.2 9.2 12.1 75.2 75.2 6:20 p. 250 6:36 p. 330 5:40 p. 305 55.5 55.5 5.9 5.9 11.1 52.4 52.4 6:46p. 280 7:00 p. 255 9:00p. 255 114 57 10.0 5.0 5.1 124.0 52.0

Totals and Averages--- 579 57.9 71.9 7.2 11.05 650.9 65.1

At the end of the run the temperature of the period of about ten hours while the data in the oil at the delivery point was 56 F. The higher following table were recorded.

Oil pumped dur- Water pumped during Gross pumping Gauge 111g period period dining period '11me Pressure,

lbs./sq. in.

Bbls Bbls/hr. Bbls Bbls/hr Per cent Bbls Bbls/hr.

N00 190 1:00 p. 220 52.5 52.5 14.0 14.0 21.1 55.5 55.5 2:00p. 310 55.5 55.5 5.2 5.2 5.2 55.7 55.7 3:00p. 550 45.5 45.5 5.2 5.2 10.1 51.7 517 4:00 p. 340 49.5 49.5 5.0 5.0 10.5 55.5 55.5 5:00 p. 155 43.5 43.5 5.0 5.0 15.5 51.5 51.5 5:30p. 150 22.0 44.0 5.3 10.5 19.4 27.3 54.5 7:05p. 240 55.0 52.4 15.5 10.5 15.5 99.5 53.0 5:00 p. 200 45.5 50.7 3.5 3.9 7.1 50.1 54.5 9:50p. 510 95.0 53.5 5.9 4.9 5.4 105.9 55.4 10:05 p.m 285 Totals and Averages..- 500.0 50.85 73.0 7.42 12.7 573.0 58.3

temperature of delivery does not reflect more At the end of this run the temperature of the lenient atmospheric conditions in the second run oil at the delivery point was 60 F. The delivery than in the first but is due solely to the greater temperature was higher than those of the first velocity of flow and the consequently shorter time and second runs because the third run was conduring which loss of heat could occur. In this ducted during somewhat warmer weather.

Rates of flow were somewhat less in the third run than in the second, but this difference was due, at least in part, to the failure of the initially cold oil to flow to the pump intake rapidly enough to utilize the full capacity of the pump.

The difierences in pressure readings in the second and third runs are of little significance. The high peak pressure of the third run, which occurred at an even earlier stage than can be accounted for by the preliminary pumping of oil into the pipe, was presumably caused by a greater amount of held-over, dry oil in the line at the time of the third run.

Comparison of these three runs, with particular reference to the data recorded near the end of each run when conditions representative of continued pumping had become established, leaves no doubt as to the efiicacy of our method and when allowance is made for the static head of 125 pounds per square inch on the pump, it may be seen that possible reduction in pressure is even greater than indication in these data.

We claim as our invention:

1. In the transportation of viscous petroleum by pipe line, the improvement comprising: introducing into said pipe line with said petroleum at least 8% by volume with respect to the total liquid of water containing in solution a wetting agent and an alkali-metal phosphate and having 1 a hydrogen-ion concentration between pH 5.7

and pH 7.0.

2. In the transportation of viscous petroleum by pipe line, the improvement comprising: introducing into said pipe line with said petroleum at least 8% by volume with respect to the total liquid of water containing in solution a watersoluble anionic surface-active agent and an alkali-metal phosphate and having a hydrogenion concentration between pH 5.7 and pH 7.0, said surface-active agent being present in amount sufficient to cause said water to wet the inner surface of said pipe line preferentially with respect to said petroleum and said phosphate being present in amount sufiicient to inhibit the formation of a water-in-oil emulsion.

3. A method as defined in claim 2, in which said water is introduced in proportions of 8% to 15% by volume with respect to the total liquid.

4. In the transportation of viscous petroleum by pipe line, the improvement comprising: introducing into said pipe line with said petroleum at least 8% by volume with respect to the total liquid or" water containing in solution from 0.02% to 0.04% by weight of a water-soluble, acid-compatible, anionic Wetting agent and at least 0.005% by weight of an alkali-metal phosphate and having a hydrogen-ion concentration between pH 5.7 and pH 7.0.

ALBERT F. CLARK. ABRAHAM SHAPIRO.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 808,782 Speed Jan. 2, 1906 1,008,980 Mills Nov. 14, 1911 2,148,131 Parker Feb. 21, 1939 2,315,044 Bransky Mar. 30, 19 3

Claims (1)

1. IN THE TRANSPORTATION OF VISCOUS PETROLEUM BY PIPE LINE, THE IMPROVEMENT COMPRISING: INTRODUCING INTO SAID PIPE LINE WITH SAID PETROLEUM AT LEAST 8% BY VOLUME WITH RESPECT TO THE TOTAL LIQUID OF WATER CONTAINING IN SOLUTION A WETTING AGENT AND AN ALKALI-METAL PHOSPHATE AND HAVING A HYDROGEN-ION CONCENTRATION BETWEEN PH 5.7 AND PH 7.0.