US3173851A

US3173851A - Electrode pitch binders

Info

Publication number: US3173851A
Application number: US45295A
Authority: US
Inventors: Laurence F King; Clellie T Steele
Original assignee: Exxon Research and Engineering Co
Current assignee: ExxonMobil Technology and Engineering Co
Priority date: 1960-07-26
Filing date: 1960-07-26
Publication date: 1965-03-16
Anticipated expiration: 1982-03-16
Also published as: GB938301A; SE185308C1; DK108462C; NL267528A; DE1258532B; GB913730A

Description

Patented Mar. 16, 1965 3,173,851 ELEQTRGDE FETCH BENBERS Laurence F. King, Mooretown, Gutario, and Cleliie T.

Steeie, Sarnia, Gntario, Canada, assignors to Esso Research and Engineering Company, a corporation of Delaware No Drawing. Fiied duly 26, 1969, Ser. No. 45,25 2 Claims. ($1. 208-23) This invention relates to a process of producing oarbonaceous materials useful as binders for carbon electrodes and to a composition useful as a binder for carbon electrodes.

Most carbon electrodes are usually made from calcined coke, which may be petroleum coke, coke from different coals, etc. As the coke has no natural adhesiveness, it must be bound together by a binder. In making electrodes the coke is ground, mixed with a suitable binder, compressed and heated to carbonize the binder.

The nature and quality of the binder is extremely important. Petroleum tars and pitches have not been heretofore proved useful because of many reasons, such as that the electrodes so made are of uneven mechanical strength and are variable in electrical conductivity. Even highly aromatic tars resulting from cracking processes have failed to produce a satisfactory pitch by conventional means.

Coal tar pitch is almost universally used as a binder in the manufacture of amorphous and graphitized electrodes, blocks, cylinders, powders and other carbon prodn ucts. The pitch is mixed with powdered or granular petroleum coke or similar material and the mixture is pressed, extruded or molded to obtain the desired shapes before baking. Coal tar pitch is especially suitable because it contains a high content of benzene insoluble resinous materials which yield coke when carbonized at relatively low temperatures. The formation of this coke between the grains of solid carbonaceous materials comprising the bulk of the structure cements the latter together into a dense and highly conductive product.

According to the present invention in its broader concept a carbon electrode pitch binder is preferably made in two stages from a particular petroleum residue. The petroleum residue is an aromatic tar obtained by steam cracking gas oil. The residue is the bottoms fraction taken from the fractionator used for fractionating steam cracked products. The first stage of the invention includes preparing a high softening point, relatively insoluhie base material of high coking value by a suitable heat treating process, preferably destructive distillation or heat soaking, and the second stage includes fluxing this base material with a minor proportion of a lighter or lower boiling hydrocarbon fraction to the softening point desired when using destructive distillation, and flashing when using heat soaking under superatmospheric pressure. In some cases the two stages can be carried out in one apparatus. The first stage may be an air oxidation process instead of the preferred processes named but is not considered a full equivalent of the preferred processes.

No single stage process appears capable of producing a suitable electrode binder when starting with aromatic tar petroleum residues. For example, if the coking value and benzene insoluble content of the binder are high enough, the softening point is too high and if the softening point is in the correct range, the other values are too low. Further, according to the present invention a carbon electrode pitch binder is made from a petroleum residue to simulate coal tar pitch binders by a two stage process in which the first stage requires the use of a high temperature to produce the benzene insoluble resins or materials which are important in the binder and considered to be the best electrode binding agents. About 30 Wt. percent benzene insoluble content is desired and the minimum should be at least about 15 wt. percent benzene insoluble content.

Aromatic tar yields of between about 15 to 25 vol. percent on feed, depending upon the oil feed character, are obtained when steam cracking gas oils to about 40- 45% conversion to C and lighter hydrocarbons. The aromatic tar has a high asphaltene content between about 12 and 25 wt. percent. These alphaltenes or resins in the pitch or binder, which are solublein benzene but insoluble in petroleum ether, contribute to the adhesive properties.

Quinoline insoluble material is low for the binders or pitches made from aromatic tars according to the present invention, usually about 2 wt. percent and this is a good feature or characteristic. Quinoline soluble material in the binder is fusible and capable of great binding strength on solidification. The quinoline soluble material is considered to be much superior to the quinoline insoluble material.

The second stage in destructive distillation includes fiuxing or blending back with a lighter material which may be a portion of the original aromatic tar, a distillate thereof or fraction-ator bottoms from catalytic cracking. In testing pitches it is usual to distill them to about 6-80 F. and then determine the softening point of the bottoms. The fraction distilled a his temperature should not eX- ceed 6% by weight. This limits the proportion of light material in the flux to about 20 wt. percent on the basis of 30 wt. percent flux oil in final blend. This is the only apparent limitation to the flux material that can be used.

The pitch binder made according to the present invention is useful in the production of prebaked or Soderberg electrodes and is especially useful in the production of electrodes for use in aluminum manufacture but not restricted thereto.

In one specific example the aromatic tar used as the starting product is formed as a high boiling residual product or bottoms when steam cracking gas oil to produce olefins, diolefins, aromatic liquid hydrocarbons, etc.

The inspections of two residual aromatic tars which may be used as feeds in the present invention are as follows:

Gravity, AP! -1.6 -4. Specific Gravity at 60 F 1. 089 1. 110 Flash COG, F. 300 320 V1210 SSU 95 213 Carbon. wt. percent 89. 90. 4 Hydrogen, Wt. percent 7. 2 6. 8 H1O atomic Ratio D. 96 g 0. 90 Sulfur, wt. percent- 1.9 2. 0 Ash. wt. percent 0.003 E 0.01 Aspl1altenes.wt. percent- 17 i 22 ASTM distillation:

The invention includes residual aromatic tars having values between the two columns given and some variations therefrom.

The steam cracking process is a well known process and comprises cracking a gas oil in the presence of about mol percent of steam at a coil outlet temperature between about 1200" F. and 1450 F. and immediately quenching with an oil having a boiling range between about 500 F. and 675 F. to a temperature of about 3 .540-F. to 560 F. The gas oil has an initial boiling point between about 400 and 525 F. and a final boiling point between about 750 and 825 R, an API gravity of 30.5 mol weight of about 290, weight percent aromatic rings of about 11.3 and weight percent naphthenic rings of about 32.2.

The gaseous products containing hydrogens, saturated hydrocarbons, olefins, diolefins, etc. are taken overhead and the'liquid hydrocarbons are fractionated to separate an aromatic gasoline boiling between about 100 F. and 460 F., a higher boiling aromatic fraction boiling between about 370 F. and 570 F. which may be recycled to the quench zone of the steam cracking unit. The amount of residual aromatic tar obtained is about vol. percent to vol. percent of the feed.

The aromatic tar is used as the feed and is then heat "treated at 'a temperature between about 750 F. and

900 F. as'by destructive distillationor mild coking to yield a volatile material which is taken overhead and condensed and a high boiling polymeric component which remains behind and is substantially benzene insoluble. Then the heat treated aromatic tar is fluxed or blended back with a lower boiling liquid to the desired softening point. This fluxing material may be a portion of the aromatic tar starting product. I

Wlrere the heat treating step is maintained under superatmospheric pressure, such as visbreaking or heat soaking, condensation and polymerization of some of the light material occurs because of the longer contact time. From a practical operational standpoint, heat soak ing plus flashing is the preferred method of carrying out the process. Moderate pressure between about 50 and 1500 p.s.i.g. may be used. A thermal cracking coil having a soaking drum can be used for heat soaking the heated aromatic tar feed. The products from heat soaking under. pressure are flashed to recover bottoms from light overhead products. Sufiicient of the light material is removed to give a bottoms product of the desired softening point for a petroleum pitch binder.

More specifically in one method of destructive distillation, a 700 ml. sample of the aromatic tar of the type "above described from steam cracking was charged to a two liter'stainless steel round bottomed still positioned in a heating pot. proof Pyrex glass top containing four standard taper openings. These openingswere used for two thermocouples (vapor and pot temperatures), a mercury sealed stirrer, a distillate outlet and a flux arm which replaced one of the thermocouples during the fluxing operation which should be done isothermally.

The vapor temperature in the still was maintained between about 525 F. and 710 F. and the pot temperature was maintained between about 650 F. and 850 F. Heat was supplied by gas heating.

The destructive distillation proceeded at a rate of about 5-30 ml. minute and for about 25 to 60 minutes until the desired bottoms product was obtained. This was controlled by the percentage of overhead product taken. The required amount of flux oil which in this case was part of the feed aromatic tar was then added through the flux arm and the heating and stirring continued for about fifteen minutes. The heating mantle around the top of, the still was removed just prior to fluxing to prevent the removal of additional overhead.

The flux liquid may be;

(a) Feed aromatic tar of the type specifically described above; or

(b) Fractionator bottoms from catalytic cracking having an initial boiling point of about 450 F. and a boiling point of about 815 F.; or

(c) -Any other high boiling point fluid residuum or distillate having not more than 20% distillable at 680 F. In one example of destructive distillation in the stainless steel equipment described above the following data were obtained:

The still was provided with a leak In one example about 72.8 parts by weight of the residue in Table 1 were mixed with 27.2 parts by weight of flux (aromatic tar feed) and heat soaked at a temperature of about 800 F. for about 15 minutes. The resulting pitch had a softening point of about 218 F., benzene insoluble of 26.4 wt. percent and a coking value of about 52.7 wt. percent.

In another example the weight percent residue was about 37.6. 78.4 parts by weight of this residue were heat soaked 15 minutes at a temperature of about 800 F. After adding about 21.6 parts by weight of flux oil, the resulting pitch having a softening point of about 235 F., benzene insoluble of about 32.3 wt. percent and a coking value ofabout 54.5 weight percent.

When using the fluxing method the actual distillation rate does not seem critical. But the end point is critical. In one example where the aromatic feed was heated to a pot temperature of about 895 F. and the time of distilling was about 45 minutes, the weight percent residue was 27.9 and using 37.3 Weight percent of quench oil the residue could not be blended to give a homogeneous product. In another example where the pot temperature was about 844 F. and the time of distilling was about 30 minutes, the wt. percent residue was 31.2 and was blended into a pitch binder when using 36 wt. percent quench oil. The pitch product had a softening point of about 252 F., benzene insoluble about 32.5 wt. percent and a coking value of about 55 WtLpercent.

Coking value of the product is the residue in Wt. percent not volatilized obtained after putting a sample in a crucible and heating to about 1022 F. for about 2% hours.

The flux used was a portion of the aromatic tar feed. Stirring was used throughout the entire operation. Less than about 70 wt. percent of the aromatic tar should be removed overhead during destructive distillation as it has been found that when more than about 70 wt. percent of the aromatic tar is taken overhead during the destructive distillation step it is not possible to successfully flux or blend back the residue with a liquid to give a homogeneous product and satisfactory binder.

The following data in Table 2 were obtained by destructive distillation of the aromatic (tar feed at temperatures between about 525 F. and 710 F. vapor temperature or 650850 F. pot temperature to the wt. percent residues based on the aromatic tar starting product. The times of heating were from about 30 minutes to 60 minutes. The data also include the fiuxed compositions comprising the residues obtained on distillation fluxed with the aromatic tar feed or starting product. The inspeotions show that the composition blends closely simulate accepted specifications for coal tar pitch used as an electrode binder. The overall yield of electrode binder pitch based on the aromatic tar feed was about 43.3 wt. percent.

Blends-Composition, Wt.

percent:

Residue 67.

Aromatic Tar Flux Inspections:

Softening Point, F Coking Value, Wt.

percent Benzene Insoluble, Wt.

percent Pitch binders made from heat treated aromatic tars according to the present invention have the characteristics given in Table 3.

Table 3 Softening Point, F 226 215 0, wt. percent i 91. 9 92. 8 H, wt. percent 5. 9 6.1 (3/51, Atomic 1. 31 1. 38 Coking Value, percent 53. 9 52. 6 Benzenelnsoluble, \vt. percent... 16. 5 25. 6 Sulfur, percent 1. 7 1. 5 Ash, percent 0. 08 0. 03

The binder in Column B was used to make an electrode. In general, the calcined petroleum coke is screened and various fractions of particle sizes are used. A general method of making and testing compressive strength of test electrodes, etc. is given in A Laboratory Evaluation of Pitch Binders Using Compressive Strength of Test Electrodes, by H. L. Jones et al., Journal of Chemical and Engineering Data, vol. 5, No. 1, January 1960, pages 8487. This publication on page 84 gives a typical particle size distribution. In general the course fraction of the calcined coke which may be a petroleum coke is mixed with the heated binder and then the other finer fractions of coke are added. The mixture is molded under pressure and heated to about 1832 F. for an extended period of time up to about 24 hours, the heating at 1832 F. being only for about one hour to carbonizc or coke the binder. One electrode was made using the binder of Column B and it met the minimum compressive strength and electrical resistivity requirements. Other methods of making electrodes may be used.

The pitch binder can also be made according to the present invention by heat treating at a temperature in the range of about 800 F. to 900 F. for about 5 minutes to 150 minutes or longer and such treating is preferably heat soaking under a pressure between about 30 p.s.i.g. and about 2000 p.s.i.g. The heat soaked products are then flashed to a lower pressure zone to recover a bottoms fraction suitable for use as a pitch binder. In an atmospheric pressure process, while it is not necessary to blend back or flux, as the distillation or heat treatment may be stopped when the desired amount of lower boiling material is removed overhead from the reaction zone as it is formed, it is desirable and advantageous to blend back as it produces a higher content of benzene insolubles. In one set of experiments where no flux oil was used, the benzene insoluble was about 7% by weight, whereas in those experiments where 22 to 31 wt. percent flux or quench was used the benzene insolubles were about 28 to 35 wt. percent.

In a superatmospheric pressure such as vis-breaking or heat soaking, similar chemical reactions occur but the lighter or lower boiling material contributes more to the condensation and polymerization reactions because of longer contact time. Heat soaking plus flashing of light products is the preferred way of carrying out the process.

Heat soaking runs were carried out under pressure at 780 F. to 860 F. The times of heating were between about minutes and about minutes when the temperature was above 700 F. At 780 F. there was only mild cracking at 100 p.s.i.g. maximum pressure. At 85 0 F. maximum temperature, cracking was more severe, resulting in 1100 psig maximum pressure and about 14 wt. percent benzene insolubles. Asphaltene content and coking value also increased with. increasing severity.

The products from the operation at the higher severity (850 F. and 1100 p.s.i.g.) were vacuum distilled according to ASTM Method D1l60 and gave a 62 wt. percent yield of bottoms having a 187 F. softening point, a coking value of 50 wt. percent and about 25 Wt. percent benzene insolubles; and a 55 Wt. percent yield of bottoms of 220 F. softening point, 54 wt. percent coking value and 31 wt. percent benzene insolubles. The hydrogen content (5.55.9%) of these products was lower than that of the pitch made by fluxing the bottoms from destructive distillation hereinbefore described. In this respect the heat soaked products approached coal tar pitch more closely than other binders previously prepared.

Yields of high coking value material (bottoms) are much greater from a heat soaking operation than by other methods investigated to date, such as destructive distillation.

Heat soaking plus flashing gives a higher yield of binder pitch, namely about 55-60% by Weight, than the other methods investigated. Long residence times of about 150480 minutes at above about 860 F. resulted in agglomeration and separation and precipitation of largecoke particles. A thermal cracking coil can be used to heat the aromatic tar feed to about 800 F. to 850 F. and the soaking drum used to provide time for the reaction.

Another run was made in the plant using a thermal cracking coil. The aromatic tar feed was passed through the furnace (residence time about 13 minutes) and flashed in the separator to produce binder pitch as bottoms. The coil outlet temperature was about 865 F. and the coil outlet pressure was about 50 p.sli.g. T he inspection of the bottoms are as follows in Table 4.

T able 4 Softening point, F. 226 C, wt. percent 92.0 H, wt. percent 6.0 C/H, atomic 1.30 Coking value 5 4.0 Quinoline insoluble 0.49 Benzene insoluble 15 S, Wt. percent 1.7 Ash, wt. percent 0. 054

In order to increase the benzene insolubles it is necessary to include soaking in the heat soaking drum forming part of the thermal cracking coil to increase the time of reaction. The binder described in Table 4 is useful in making carbon electrodes. Three electrodes made from this pitch met the minimum compressive strength and electrical resistivity requirements.

In destructive distillation the work done indicated that the benzene insoluble resins or materials were formed at 800 F850 F. when the light components were continuously removed overhead.

Heat soaking aromatic tar feed under pressure and flashing are believed to be the best means of preparing electrode pitch binders as the yields of high coking value material are much greater than the other methods invc-stigated.

What is claimed is:

l. A composition for use as a binder for carbon electrodes including a high boiling high molecular Weight hydrocarbon residue produced by heating a residual aromatic tar having an API gravity of about l.6 to 4.0 and an asphaltene content between about 12 and 25 Wt. percent,

obtained as bottoms from a high temperature gas oil steam cracking process to produce 4045% C and lighter hydrocarbons, to temperature at least about 800 F. for a period of time sufiicient to remove overhead between about 60 and 70 wt. percent of said aromatic tar the resultant heat treated residue being then blended into a homogeneous product With an aromatic flux liquid consisting of said residual aromatic tar starting product in amounts of about 3035% by weight of the latter to about 70 to 65% by weight of said heat treated residue.

2. A pitch binder for carbon electrodes made solely from residual aromatic tar from high temperature cracking of gas oil as defined in claim 1 which has a softening point of between about 175 and 230 F., a coking value between about 50 and 57 wt. percent and benzene insoll5 uble material between about 15 and 40 Wt. perecnt.

References Citedin thefile of this patent UNITED STATES PATENTS Brautigam et a1 Sept. 29, 1942 Beattie June 26, 1956 Nash Oct. 23, 1956 Dunkel et al Nov. 27, 1956 Waddill Mar. 11, 1958 Hackley J an. 26, 1960 Goldthwait et al July 1 2, 1960 Renner July 4, 1961 Renner July 11, 1961 FOREIGN PATENTS Great Britain June 6, 1955

Claims

1. A COMPOSITION FOR USE AS A BINDER FOR CARBON ELECTRODES INCLUDING A HIGH BOILING HIGH MOLECULAR WEIGHT HYDROCARBON RESIDUE PRODUCED BY HEATING A RESIDUAL AROMATIC TAR HAVING AN API GRAVITY OF ABOUT -1.6 TO -4.0 AND AN ASPHALTENE CONTENT BETWEEN ABOUT 12 AND 25 WT. PERCENT, OBTAINED AS BOTTOMS FROM HIGH TEMPERATURE GAS OIL STEAM CRACKING PROCESS TO PRODUCE 40-45% C3 AND LIGHTER HYDROCARBONS, TO TEMPERATURE AT LEAST ABOUT 800*F. FOR A PERIOD OF TIME SUFFICIENT TO REMOVE OVERHEAD BETWEEN ABOUT 60 AND 70 WT. PERCENT OF SAID AROMATIC TAR THE RESULTANT HEAT TREATED RESIDUE BEING THEN BLENDED INTO A HOMOGENEOUS PRODUCT WITH AN AROMATIC FLUX LIQUID CONSISTING OF SAID RESIDUAL AROMATIC TAR STARTING PRODUCT IN AMOUNTS OF ABOUT 30-35% BY WEIGHT OF THE LATTER TO ABOUT 70 TO 65% BY WEIGHT OF SAID HEATED RESIDUE.