Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.


  1. Advanced Patent Search
Publication numberUS3406111 A
Publication typeGrant
Publication dateOct 15, 1968
Filing dateJul 12, 1962
Priority dateMar 18, 1960
Also published asDE1239048B
Publication numberUS 3406111 A, US 3406111A, US-A-3406111, US3406111 A, US3406111A
InventorsJr Shirley C Bartlett, Wynkoop Raymond
Original AssigneeSun Oil Co
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Transformer oil
US 3406111 A
Abstract  available in
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

Oct. 15, 1968 R. WYNKOOP ET AL 3,406,111

TRANSFORMER OIL Filed July 12, 1962 5 Sheets-Sheet l Oxidation Time, Hrs.



ATTORNEY Oct. 15, 1968 WYNKQOP ET AL 3,406,111

TRANSFORMER OIL 5 Sheets-Sheet 2 Filed July 12, 1962 QEP Q 3,9 10;:03 JaMo wag ma I I RAYMOND WY NVENTORJ NKOOP SHIRLEY C. BARTLETT, JR

ATTORNEY Oct. 15, 1968 WYNKOQP ET AL 3,406,111

TRANSFORMER OIL Filed July 12, 1962 '5 Sheets-Sheet 5 4'0 Ibo Gzsidaiion Time, Hrs.


ATTORNEY United States Patent 3,406,111 TRANSFORMER OIL Raymond Wyukoop, Gladwyne, and Shirley C. Bartlett, IL, Broomall, Pa, assignors to Sun Oil Company, Philadelphia, Pa., a corporation of New Jersey Continuation-impart of application Ser. No. 94,433, Mar. 9, 1961. This application July 12, 1962, Ser. No. 209,345

4 Claims. (Cl. 208-14) This invention relates to novel transformer oil compositions and more particularly to uninhibited transformer oils which, under Doble Oxidation Test conditions, consistently exhibit power factors less than two percent throughout surprisingly long periods of oxidation and also low sludging tendency and high oxidation stability under the test conditions.

This application is a continuation-in-part of copending application Ser. No. 94,433, filed Mar. 9, 1961, now abandoned, which in turn is a continuation-in-part of application Ser. No. 15,810, filed Mar. 18, 1960 and now abandoned.

Properties of commercial oils used as insulating media in transformers are well known in the art and a list of typical characteristics is given in the text by F. M. Clark entitled, Insulating Materials for Design and Engineering Practice (1962), page 135. Such oils typically boil in the range of 460-775 F. and preferably have viscosities in the range of 55-60 S.U.S. at 100 F., as may be seen by reference to Wasson et al. Patent No. 3,000,807.

Commercial transformer oils customarily are tested by the Doble Oxidation Test for electrical insulating oils developed by the Doble Engineering Company of Belmont, Massachusetts. This procedure has been described in ASTM Standards on Electrical Insulating Liquids and Gases, pp. 307-313, December 1959, under the title Suggested Method of Test for Oxidation Characteristics of Mineral Transformer Oil. It involves bubbling air through a known amount of the oil held at a temperature of 95 C. in the presence of copper and iron and making two types of tests daily on small samples of the oil. One type of test is an acidity measurement. The other is a precipitation test in which one volume of the oil is diluted with five volumes of pentane, the mixture is allowed to stand at least eight hours and the presence or absence of a sludge precipitate is noted. The endpoint of the Doble test is taken as the number of days of oxidation either before the acidity of the oil reaches 0.25 mg. of KOH per gram or before a positive precipitation test for sludge is obtained. The best commercial transformer oils heretofore available generally have a life of only about three days under Doble test conditions. Longer life values could be obtained by adding an oxidation inhibitor to the oil but the use of such additive traditionally has been considered unacceptable by transformer manufacturers and users.

An amplification of the Doble test recommended by the Doble Engineering Company is the so-called Power Factor Valued Oxidation (PFVO). This involves operating in the manner described above but also determining the power factor of the oil at two hour intervals throughout the oxidation period. A curve is obtained by plotting the power factor against the oxidation time. The accompanying FIGURE 1 illustrates the type of curves found to be characteristic of commercially available transformer oils. FIGURE 1 shows two curves that were obtained for commercial transformer oils A and B. It can be seen that for each oil a sharp hump in the power factor curve occurs in the earlier stages of oxidation. The power factor rapidly increases to a value considerably in excess of two percent, following which it decreases to a relatively low value and thereafter rises continuously. The time of oxiice dation at which the peak of the hump is reached may differ with different oils, but it is characteristic for all uninhibited commercial transformer oils previously available that a hump in the PFVO curve is obtained and that the power factor at the peak of the hump is considerably above two percent. The ideal transformer oil of course would be one that exhibits no increase whatever in power factor throughout the test life.

Transformer oils of the present invention are unlike any previously available in that the power factor curve obtained is relatively fiat. This is illustrated by FIG- URES 2-A and 2-B representing three uninhibited transformer oils (C, D and E) according to the present invention. These oils were prepared by the process described and claimed in the aforesaid copending applica tion Ser. No. 94,433, hereinafter described. FIGURE 2-A shows the power factor curves for the three oils during the first 200 hours of oxidation by the Doble procedure, while FIGURE 2-B presents an extension of the curves for the oxidation period of 200-400 hours. It can be seen that the curves are remarkably fiat as compared to those obtained for other transformer oils. While there may be a slight hump in the power factor curve for some of the oils of the present invention, the hump is insignificant and the power factor consistently remains less than two percent for at least 96 hours of oxidation (4 days). Preferred compositions, such as the oils illustrated in FIGURES 2-A and 2-B, have power factors that remain consistently below two percent for oxidation times exceeding 192 hours (8 days) and often considerably in excess of such time.

Transformer oils of the present invention can be characterized as follows:

(1) They are napthenic petroleum distillates having a viscosity in the range of -65 S.U.S. at 100 F. and more preferably -60 S.U.S. at F.

(2) They have viscosity-gravity constants in the range of 0.84-0.92. This corresponds to aromatic contents generally in the range of 15-65% by weight.

(3) They have nitrogen contents less than 4 ppm. and more preferably not exceeding 2 ppm.

(4) When tested by the Doble Oxidation Test in the absence of an inhibitor, they exhibit the following characteristics: (a) a neutralization number less than 0.25 mg. KOH/ g. at 96 hours oxidation time; (b) absence of sludge at 96 hours oxidation time; (c) power factors during the oxidation period of 0-96 hours consistently less than 2 percent. Preferred compositions have power factors that are consistently below 2 percent for at least 192 hours of oxidation time and a Doble test life of at least 192 hours (8 days).

The improved transformer oils of the invention can be prepared by the process described in the aforesaid application Ser. No. 94,433. The preferred procedure involves first treating a naphthenic distillate of the abovespecified viscosity range with strong sulfuric acid, removing the acid sludge and then contacting the oil with adsorptive clay. Typically the oil is contacted with 10 lbs./bbl. of concentrated sulfuric acid and 20 lbs./bbl. of clay. At this stage the oil generally will have a Doble test life of only about 2 days and a nitrogen content of the order of 10 ppm. The oil is next treated at 250- 280 F. with an alkali metal alkoxide in which the alkoxide portion corresponds to a secondary or tertiary alcohol of the C -C range. The preferred alkoxide is sodium isopropoxide. Other alkali metals, such as potassium or lithium, can be used and the alkoxide portion of the treating agent can correspond to other secondary or tertiary alcohols, such as t-butyl alcohol, but not to primary alcohols such as ethanol or n-propanol. The alkoxide can be pre-formed by reacting sodium with an appropriate alcohol or it can be formed in situ in the oil by dispersing metallic sodium therein and then adding the alcohol. Generally it is preferred to use a slight molar excess of alcohol over the stoichiometric proportion required to form the alkoxide. The amount of sodium used typically is 05-10% by weight on the oil. After treatment with the alkoxide, the oil is contacted at 250-280 F. with carbon dioxide in molar excess of the amount of alkali metal used. Alternatively the oil can be contacted with carbon dioxide while being treated with the alkali metal alkoxide although this is not preferred. The oil is then washed with water to remove the organo-metallic materials and thereafter is heated under vacuum while being sparged with an inert gas (e.g. nitrogen) to remove residual water. Finally the oil is contacted again at say 225 F. with adsorptive clay in amount, for example, of 20 lbs./bb1. The resulting refined product will have a nitrogen content less than 4 p.p.m. and generally not exceeding 2. p.p.m. and will exhibit the unique Doble test characteristics specified above. The product will give outstanding performance in transformer service.

The foregoing treatment removes little of the aromatic hydrocarbons present in the charge stock. The presence of these naturally occurring aromatics in the refined product is highly important in providing the performance characteristics desired. It appears that the aromatics function as natural oxidation inhibitors and are essential for securing the unique Doble test characteristics as described above and illustrated in FIGURES 2-A and 2-B. The extremely low nitrogen content obtained by the above-specitied treatment also is important in contributing to the outstanding stability of the present oils under the oxidizing conditions of the Doble test, as it has been found that if the nitrogen content exceeds 4 p.p.m. the oil generally will have a Doble test life of only about 2 days and a pronounced hump in the initial portion of the power factor curve will appear.

Oils which have viscosities substantially dilferent from the range of 50-65 S.U.S. at 100 F. curiously do not exhibit the outstanding Doble test characteristics found in the present oils, even when such other oils have been treated by the same refining procedure as described above. This is illustrated by the PFVO curves shown in FIG- URE 3 for two such other oils designated as F and G. Both of these oils were naphthenic distillates which had been treated by the procedure described above using 0.75% sodium by weight on the oil and a molar ratio of isopropanol to sodium of 1.10. Oil F had a viscosity of 165 S.U.S. at 100 F. and a viscosity-gravity constant of 0.882, while the corresponding properties for oil G were 735 S.U.S. at 100 F. and 0.870. It can be seen from FIGURE 3 that while neither curve exhibits a distinct hump as obtained for the conventional transformer oils of FIGURE 1, the power factors increased relatively rapidly and exceeded two percent after only about 40 hours oxidation time. The reason for the poorer oxidation stabilities of these oils, as compared to oils such as C, D and E of FIGURES 2-A and 2-B, is not known with certainty but it is thought to be likely that the difference in the molecular weight ranges of the aromatics therein is probably responsible. In other words the aromatics in naphthenic oils having viscosities in the range of 50-65 S.U.S. at 100 F. appear to have molecular weights and structures such that they function more effectively as inhibitors under Doble test conditions than do the aromatics in oils of a substantially different viscosity range. In any event the fact is that the oil should have a viscosity of 50-65 S.U.S. at 100 F. to exhibit the outstanding Doble test characteristics illustrated in FIGURES 2-A and 2-B.

The following examples illustrate the preparation of the novel transformer oils of the invention and properties of the oils so produced.

EXAMPLE I The charge oil was a naphthenic distillate stock of boiling range suitable for electrical transformer use. The oil initially had the following properties: A.P.I. gravity: 24.4; flash point:280 F.; fire point=3l0 F.; S.U.S. viscosity at F.=55.3; S.U.S. viscosity at 210 F: 33.6; viscosity-gravity constant=0. 855; nitrogen content=50 p.p.m.; sulfur content=0.18%; refractive index=1.5009; Doble life of a sample which was furfural extracted and then treated with 20 lbs./bbl. of 99% acid and 35 lbs./bb1. of clay=2 days. The foregoing Doble test life indicates that conventional treatment of this oil does not produce a satisfactory transformer oil.

The charge oil was first contacted at room temperature with 99% sulfuric acid in amount of 10 lbs./bbl. and was water washed and dried by air blowing. The oil was then heated to a temperature of about 270 F., and sodium in amount of 1% by weight was dispersed therein by means of a high speed stirrer. Then isopropanol was added in amount of 0.5% by volume on the oil, whereupon reaction occurred and the temperature rose about 10 F. over a'time of about 6 minutes and thereafter dropped. After about 30 minutes of mixing, carbon dioxide was bubbled into the mixture at a temperature of about 245 F. This caused the temperature to rise during a period of about 10 minutes, the amount of rise being somewhat greater than during the treatment with sodium and alcohol. Contact with carbon dioxide was continued for a time of about 10 minutes. Stirring of the mixture was then discontinued and it was allowed to stratify while cooling. Unreacted sodium and a reaction product layer settled to the bottom and the clear oil was decanted. The oil finally was contacted with 20 lbs./bbl. of clay. Testing of the finished oil showed that it had a Doble life of 8 days, thus indicating that it would be excellent for transformer use. The oil product had a nitrogen content of only about 2 p.p.m., which shows that the present treatment is highly effective in removing nitrogen compounds. The nitrogenous bodies are concentrated in the organo-sodium reaction product layer which separates from the oil.

EXAMPLE II Another run was carried out as in the preceding example except that the amount of sodium used was increased to 2.5% by weight and the isopropanol to 1.0% by volume on the oil. The finished oil in this case had a Doble life of 11 days,

EXAMPLE III Sodium recovered from the run of Example II was used for treating another batch of oil in the same manner except that in this case contact of the oil with sodium and with carbon dioxide was done concurrently. Again a Doble life of 11 days was obtained. This shows that the excess sodium can be reused and that successive and concurrent treaments with sodium and carbon dioxide each are efiective in preparing transformer oils according to the invention.

EXAMPLE IV Another run was carried out in the same manner as in Example II except that after the sodium treatment the oil was cooled to room temperature before being contacted with carbon dioxide. The resulting oil had a Doble life of 9 days. This shows that an elevated temperature is not required for the carbon dioxide treatment to be effective.

EXAMPLE V upper layer of clear oil. Dissipation factor tests on the oil layers gave the following results:

Amount of acid used, Dissipation factor lbs./bbl.: at 100 C.

The low dissipation values obtained indicate that most of the sodium compounds can be removed from the reaction mixture by centrifuging and that a final water-washing or clay-treating step is not necessarily essential for producing a good transformer oil.

EXAMPLE VI The present example illustrates preparation of the present transformer oil composition in continuous manner. The reactor used was provided with a high speed stirrer and with means for separately introducing streams of the charge oil, the alkoxide reagent which had been pre-formed and carbon dioxide and for continuously withdrawing the reaction mixture. The alkoxide reagent was prepared by forming a concentrated dispersion of sodium in a small amount of the charge oil heated to a temperature above the melting point of sodium, cooling the dispersion to room temperature and then slowly adding isopropanol while stirring the mixture. The charge oil was the same stock as described in Example I but had been treated with 10 lbs./bbl. of 99% sulfuric acid, neutralized by aqueous caustic soda and then dried by air blowing. It had a Doble test life of only 2 days.

The treatment was carried out by continuously passing the oil at a temperature of about 280 F. through the reactor at a rate providing a residence time of minutes, continuously feeding in a stream of the isopropoxide slurry in amount such that the sodium content of the reaction mixture was 0.56% by weight and continuously passing CO into the agitated mixture in molar excess of the isopropoxide. Sodium-containing material was separated from the reactor efiiuent by sedimentation and the treated oil was finished by contact with 20 lbs./bbl. of adsorptive clay at 220 F. The treated cil product was found to have a Doble test life of about 5 days, showing that it had good stability for transformer service.

EXAMPLE VII Another continuous run was made in the same manner as described above except that the sodium content of the reaction mixture was increased to about 1.1% by weight. The Doble test life of the oil was found to be about 7 days.

EXAMPLE VIII A sample of the acid-treated and neutralized charge oil used in Example VI was treated batchwise with tertiary butoxide as the treating agent. The alkoxide was prepared by dispersing 0.5% by weight of sodium in the oil while hot, cooling to room temperature and slowly adding 1.1 moles of tertiary butanol per mole of sodium. The mixture was then heated to about 280 F. and agitated for 30 minutes. Carbon dioxide was then passed into the mixture at about 280 F. for 20 minutes. Sodium-containing material was removed from the mixture by settling. The treated oil had a Doble Test life of 5 days.

By way of comparison, when this run was repeated, except that n-propyl alcohol in one case and ethyl alcohol in another case were substituted for tertiary butyl alcohol, the Doble test life was only 2 days which was the same value as obtained for the charge oil before treatment. This indicates that primary alcohols are not operative for the present purpose.

EXAMPLE IX The charge oil was another batch of naphthenic distillate oil having the following properties: A.P.*I.

6 gravity=24.8; flash point=290 F.; viscosity at 100 F.=58 S.U.S.; viscosity-gravity constant=0.88; R.I., d/ 20-=1.5000; total aromatics=37.5% by weight; monocyclic aromatics=25.6% by weight; dicyclic aromatics =10.2% by weight; tricyclic aromatics=1.7% by weight; sulfur content=0.20% by weight; nitrogen content= p.p.m. The oil was treated with 20 lbs./bbl. of 99% H SO then with 2 lbs./-bbl. of coagulation clay, neutralized with aqueous caustic soda, water washed and finally brightened by air blowing. At this stage the oil had a Doble test life of only 2 days.

A series of runs was made in a batch-type pilot plant in which batches of the acid-treated oil were treated with sodium isopropoxide formed in situ in the oil by the addition of metallic sodium and isopropanol thereto. The conditions in all the runs were substantially the same. The amount of sodium used was 0.5% by weight on the oil, the molar ratio of isopropanol to sodium was 1.1 and the temperature was 250-280 F. Following this treatment the oil was blown with CO at about the same temperature level, the molar ratio of CO used to sodium being 1.2. The oil was then agitated at about 175 F. with 0.4 volume water per volume of oil, the mixture was settled for minutes and the water layer was removed. This washing procedure was repeated except that the temperature was about 145 F. The oil was then heated to about 225 C. while under vacuum and was sparged with nitrogen to remove residual water. Finally the oil was contacted with 5 lbs./bbl. of clay for 30 minutes at 225 F.

In the foregoing manner the oils designated as C, D and E in FIGURES 2-A and 2-B were obtained and their PFVO curves were found to be as shown therein. The Doble Test life values were found to be as follows:

Days Oil C 14 Oil D 10 Oil E 14 For these oils differences between the life values found and between the curves shown in FIGURES 2-A and 2-B merely represent normal variations that can be expected for repetitive processing operations and testing of products and are not considered to have substantial significance with respect to quality of the oils.

The following are typical properties of oils C, D and E:

Comparison of these values with the properties given for the original charge stock shows that the most drastic change elfected by the treating procedure employed was in the nitrogen contents of the oils. A major part of the original nitrogen content (50 ppm.) was removed by the acid treatment while a minor but highly significant part was removed by the Na isopropoxide-CO treatment. The total aromatic content was reduced only slightly while the sulfur content was reduced to approximately one-half the original value. The latter changes occurred mainly during the acid treatment step.

We claim:

1. A transformer oil comprising a naphthenic petroleum distillate boiling in the range of 460-775 having a viscosity in the range of 50-65 S.U.S. at F., a viscosity-gravity constant in the range of 0.84-0.92 and a nitrogen content less than 4 p.p.-m. and, when tested under -D0ble Oxidation Test conditions in the absence of an added inhibitor, exhibiting the following characteristics: (1) a neutralization number less than 0.25 mg. KOH/g. at 96 hours oxidation time; (2) absence of sludge at 96 hours oxidation time; and (3) power factors during the oxidation period of 0-96 hours consistently less than 2 percent.

2. A transformer oil according to claim 1 having a nitrogen content not exceeding 2 ppm.

3. A transformed oil according to claim 1 having a power factor at 192 hours of oxidation less than 2 percent.

4. A transformer oil according to claim 1 having a viscosity in the range of 55-60 S.U.S. at 100 F.

References Cited 5 UNITED STATES PATENTS 3,095,366 6/1963 Schiem'an 208-14 1,856,700 5/1932 Ford 20814 3,000,807 9/1961 Wasson et a1. 208-44 10 DELBERT E. GANTZ, Primary Examiner.

H. LEVINE, Assistant Examiner.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US1856700 *Apr 12, 1928May 3, 1932Westinghouse Electric & Mfg CoInsulating oil
US3000807 *Dec 4, 1958Sep 19, 1961Exxon Research Engineering CoBlended transformer oil
US3095366 *Mar 3, 1960Jun 25, 1963Standard Oil CoInsulating oil
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US3619414 *Feb 19, 1969Nov 9, 1971Sun Oil CoCatalytic hydrofinishing of petroleum distillates in the lubricating oil boiling range
US3714021 *Oct 22, 1970Jan 30, 1973Kureha Chemical Ind Co LtdThermally stable insulating oil
US3753188 *Dec 28, 1970Aug 14, 1973Hitachi LtdInductive electric apparatus
US3839188 *May 22, 1968Oct 1, 1974Sun Oil CoHydrorefined transformer oil and process of manufacture
US3839189 *Feb 24, 1972Oct 1, 1974Sun Oil CoHydrorefined lube oil and process of manufacture
US3932267 *Sep 11, 1974Jan 13, 1976Shell Oil CompanyProcess for producing uninhibited transformer oil
EP0497467A1 *Jan 16, 1992Aug 5, 1992Cooper Power Systems, Inc.Very low pour point dielectric
U.S. Classification208/14, 208/275, 208/254.00R, 208/291, 208/284, 208/294, 208/286, 208/285
International ClassificationC10G29/20, C09K3/14, C10G29/00, H01B3/22, H01B3/18
Cooperative ClassificationH01B3/22, C10G29/20, C09K3/1454
European ClassificationH01B3/22, C10G29/20, C09K3/14D