US3531253A

US3531253A - Method for determining the extent of conversion of ethylene to alpha olefins

Info

Publication number: US3531253A
Application number: US593028A
Authority: US
Inventors: Herbert B Fernald; William Gall; Alfred N Kresge
Original assignee: Gulf Research and Development Co
Current assignee: Chevron USA Inc
Priority date: 1966-11-09
Filing date: 1966-11-09
Publication date: 1970-09-29
Anticipated expiration: 1987-09-29
Also published as: JPS4939641B1

Description

Sept. 29, 1970 I -H. B. FERVNALD T 3,531,253:

METHOD FOR DETERMINING THE EXTENT OF CONVERSION OF ETHYLENE TO ALPHA OLEFINS Filed Nov. 9, 1966 I 2 Sheet'sSheet 1 EFFECT OF TEMPEMTl/RE 0 PRODUCT DISTRIBUTION IN ALPH 01. FIN E TETRA ENE, c

W516 7' PERCENT OF TOTAL PRODUCT OC c C SE -I c .360 370 l REACT/0A! TEMPERATURE, F'

' INVENTORS HERBERT B. FERNALD 1 3 WILLIAM GALL ALFRED N. KRESGE' United States Patent Office Patented Sept. 29, 1970 3,531,253 METHOD FOR DETERMINING THE EXTENT OF CONVERSION OF ETHYLENE TO ALPHA OLEFINS Herbert B. Fernald and William Gall, Glenshaw, and

Alfred N. Kresge, Verona, Pa., assignors to Gulf Research & Development Company, Pittsburgh, Pa., a corporation of Delaware Filed Nov. 9, 1966, Ser. No. 593,028 Int. Cl. C07c 3/10; G01n 31/08 US. Cl. 23230 10 Claims ABSTRACT OF THE DISCLOSURE This invention relates to a method for determining the extent of catalytic conversion of ethylene to normal alpha olefins.

The conversion of gaseous ethylene to a product comprising predominantly liquid alpha olefins is accomplished in the presence of an organometallic catalyst, such as triethylaluminum. The conversion is only partial and the reactor eflluent stream contains unreacted ethylene together with product alpha olefins. The product comprises mostly normal alpha olefins in the liquid phase in which all the catalyst is dissolved or suspended, and also includes gaseous normal alpha olefins which are admixed with the gaseous ethylene reactant. Since not all of the product is a liquid, if the extent of conversion were to be determined by a material balance a separation of the liquid phase from the gaseous phase in the product stream would be required plus a separation of gaseous phase components from each other.

The present invention relates to an advantageous method for measuring the total extent of reaction in a process for the catalytic conversion of ethylene to both gaseous and liquid normal alpha olefins by sampling only a small portion of the liquid product and disregarding the gaseous phase of the product stream, even though a portion of the product is in the gaseous phase. In accordance with this invention, the extent of conversion of gaseous ethylene to only liquid product is indicated by obtaining a small sample of the liquid phase of the product stream and measuring the percentage of catalytic metal, such as aluminum, in said liquid sample. All of the catalytic metal is found in the liquid phase of the product because the catalyst tends to be completely dissolved or dispersed in said liquid phase. Since the charge rate of catalytic metal to the process is generally constant but the amount of liquid product obtained from the process increases with increasing conversion, a small percentage of aluminum in a sample of the liquid phase of the product indicates a large total conversion of ethylene to alpha-olefin product, and

vice versa. Therefore, measurement of percentage of aluminum in a sample of liquid product provides an indication of conversion of charge ethylene to liquid phase normal alpha olefins, but falls short of indicating the extent of conversion of charge ethylene to both liquid and gaseous phase alpha olefins, i.e., to total alpha-olefin product.

The method of the present invention is based upon the observation that it is possible to determine the conversion of ethylene to all the alpha-olefin components of the product stream, both liquid and gaseous, by selecting a certain specific alpha-olefin component which is entirely in the liquid phase of the product and measuring the percentage of that component in the previously indicated sample taken of the product liquid phase. By selecting the proper component in the liquid phase of the product for measurement an indication of the total conversion of ethylene to all liquid and gaseous alpha olefins can be achieved even though no measurement is made of any other liquid components of the product or of any gaseous component of the product.

In the process for the conversion of ethylene to normal alpha olefins which are predominantly liquid but which also include some gaseous olefins, a product is obtained whose components comprise substantially C to C or C normal alpha olefins. The percentage of each component tends to vary as the temperature of reaction varies. Generally, at any particular reaction temperature a given alpha-olefin component in the product will comprise a different percentage of the total product than at any other reaction temperature. However, the method of the present invention makes use of the observation that within certain limited temperature ranges it is possible to select one particular alpha-olefin component of the liquid product or a group of alpha olefins which remains substantially a constant percentage of the total product, including liquid and gases, or remains a relatively steady percentage thereof, notwithstanding variations in reaction temperature within said ranges.

This observation is illustrated by reference to FIG. 1 which is a graph showing the effect of temperature on product distribution in a process for the production of normal alpha olefins utilizing an aluminum triethyl catalyst in a long tubular reactor at a pressure of 4,000 pounds per square inch. The graph shows that the percentage of certain alpha-olefins components in the product stream changes greatly 'with minor changes in reaction temperature throughout the temperature range shown in the graph. For example, the percentages in the total product of the gaseous product component, i.e., butene-l, vary to a relatively great extent upon small temperature changes throughout the indicated temperature range. Similarly, the percentage in the total product of the C and heavier components varies to a relatively great extent upon small temperature changes throughout the indicated temperature range. In contrast, the graph shows that a particular component of the liquid product can be selected which remains substantially constant over a limited temperature range. This is indicated by the relatively flat plateau in the C product curve within the 390 F. to 400 P. reaction temperature range. FIG. 1 shows that the percentage in the total product of dodecene-l remains substantially constant throughout the reaction temperature range of 390 F. to 400 F. A similar constant temperature plateau is observed in the curve representing the C component of the product within the 370 F. to 380 F.

temperature range. FIG. 1 also shows that the total of the C C and C alpha olefins produced in the 390 F. to 400 F. temperature range remains relatively constant.

The graph of FIG. 1 therefore shows that in an alphaolefin process performed in the temperature range 390 F. to 400 F. the percentage of the C components in the liquid product or the percentage of the C C and C alpha olefins in the liquid product can be utilized to indicate the total conversion in the process. Since all of the C component of the product or of the C C and C component group is in the liquid phase, the result of chromatographic analysis of a small sample of the liquid product to determine the percentage thereof which is the C component or the C C and C components can be used to indicate the amount of total liquid and gaseous product corresponding to said sample. Because it is not possible to maintain the temperature throughout a long tubular reactor at a single point but rather only within a limited range, it is important to select a particular component or group of components of the product as indicator whose percentage in the product remains substantially constant within the control temperature range. For example, if the temperature of the reactor is permitted to fluctuate only within the range 390 F. to 400 F., the C component of the product is chosen as indicator because its proportion in the product is substantially constant with temperature changes within this range and varies less within this temperature range than ony other component of the product. Similarly, if the temperature of the reactor is controlled to permit fluctuation thereof only within the range 370 F. to 380 F., the C component of the product is chosen as indicator because, as shown in FIG. 1, its proportion in the product is substantially constant with temperature changes within this range and Varies less with temperature than any other component of the product within this range.

By selecting the appropriate indicator component in the liquid product for the temperature range in which the reaction is performed, a highly accurate indication of the total weight of the alpha-olefin conversion product, including both liquid and gaseous products, corresponding to the weight of the liquid sample is obtained. For example, if the temperature range of the reaction is 390 F. to 400 F. and the percentage of the liquid product sample which is C is measured, the total amount of both liquid and gaseous reaction product is readily obtained by the following calculationz:

Total weight of liquid Percentage of X Weight of X10. 0

and gaseous alpha the liquid the olefins based on the product product liquid product sample sag: ple which sample.

The factor of 10.0 shown in the above calculation is obtained from the graph of FIG. 1 which shows that within the 390 F. to 400 F. temperature range the C component of the product represents 10 percent of the total liquid and gaseous product. Therefore, if the weight of the C component based upon the sample taken is known, it is multiplied by 10.0 to obtain the weight of the total liquid and gaseous product based upon the product sample.

Since we now know (1) the total weight of liquid and gaseous alpha olefins based upon the liquid product sample and (2) the total weight of catalytic aluminum in said sample we can obtain the ratio of (l) to (2) in tlr: product. Assume the ratio of total olefin product baserl upon the sample to aluminum in the product samp'e (ratio of (l) to (2)) is 100 to 1. Now refer to the charge stream to the process wherein ethylene and triethylaluminum catalyst are charged to the reactor. Since (3) the charge rate of ethylene and (4) the charge rate of catalyst are controlled and therefore known in any commercial process, dividing (3) by (4) will provide the ratio of ethylene in the charge to catalytic aluminum in the charge.

For purpose of discussion, assume the ratio of charge ethylene (3) to charge aluminum (4) is 200 to 1. Now, since the stated charge ratio is 200 to 1 and the stated efiluent ratio is 100 to 1, this leaves 100 parts in the effiuent which is unaccounted for, i.e., is unreacted ethylene. Since of the stated 200 parts of ethylene in the charge only 100 parts remain as unreacted ethylene in the effluent stream, there has been a 50 percent conversion of ethylene to alpha-olefin product in the process.

Under some process conditions calculations to measure changes in ethylene conversion based upon changes in aluminum concentration in product liquid phase samples which have been stabilized prior to analysis in a standardized manner to remove lower boiling components will give fairly comparable results as compared to calculations based upon changes in the above-described ratio in the product stream. Under these conditions, for purposes of calculation measurement of aluminum concentration in product liquid samples which have been stabilized prior to analysis can be substituted for measurement of the aforementioned product stream ratio. Similarly, under some process conditions, a knowledge of changes in concentration of aluminum in the charge stream is fairly comparable for purposes of calculations of changes in ethylene conversion to a knowledge of changes in the aforesaid charge stream ratio. Furthermore, if the aforesaid charge stream ratio remains substantially constant, in the absence of calculation merely observing changes in aluminum concentration in uniformly stabilized samples of the liquid phase of the product will by itself indicate changes in ethylene conversion. For example, with a constant aluminum concentration in the charge a reduction in aluminum concentration in the product liquid phase can indicate an increase in liquid yield.

In calculating the ratios of the present invention it is entirely equivalent to substitute the weight of the aluminum compound from which the aluminum is derived for the weight of aluminum metal per se, as long as consistency in this regard is observed in both the charge ratio and in the product ratio. Therefore, in the claims of the present invention wherever the weight of aluminum per se is recited, the weight of the corresponding aluminum compound is considered a complete equivalent.

The present invention is applied to a process for the step-wise conversion of gaseous ethylene to higher straight chain normally liquid olefins having the double bond in the terminal or alpha position, which reaction proceeds as follows:

etc. This polymerization occurs catalytically in the presence of organometallic compounds, such as aluminum alkyls, which participate in the reaction. As the reaction proceeds in the presence of excess ethylene, an increasing quantity of gaseous ethylene is converted to liquid olefin so that the density of the reaction system progressively increases. The chemistry of the alpha olefin process can be described in terms of three major reactions. In the propagation (growth) reaction, an alkyl group on an aluminum atom containing n ethylene units can add an ethylene molecule to become an alkyl group of n+1 ethylene units, as follows:

(CHzCH2) H (CHzCH2)pH The transalkylation (displacement) reaction which occurs concurrently with the growth reaction consists of two steps. These are, first, thermal decomposition of an aluminum alkyl group to a hydride plus alpha olefin followed by a rapid reaction of the hydride with ethylene to regenerate an ethyl group which can start another growth cycle.

The thermal decomposition is much slower than reaction of ethylene with a hydride and, therefore, is the rate-deter mining step for the over-all reaction.

(C HzC H2) m H A zCI-I2) n".

(C HzC H2) r-H (CH2CHz)mI-I \(CH2C H2) p H C H2 0 H2 The growth and displacement reactions occur repeatedly as long as there is unreacted ethylene present. Therefore, the reaction is advantageously afforded a very high residence time. As long as there is free ethylene in the presence of catalyst in the reactor under reaction condition, each mole of catalyst present will produce additional normal alpha-olefin product. Therefore, a long residence time is conducive to a high alpha-olefin yield per mole of catalyst, Le, a high catalyst efficiency.

The third reaction is similar to the first except that the aluminum alkyl adds a product alpha olefin, rather than ethylene, to form a branched chain aluminum alkyl group. However, this structure is very unstable and rapidly decomposes to form a hydride and an olefin of vinylidene structure.

The decomposition is so rapid compared to the addition of another ethylene molecule to the branched alkyl that essentially all reactions of this type result in an olefin of vinylidene structure and regeneration of an aluminum ethyl alkyl group. As a result, there will be few, if any, alpha olefins with branching beyond the beta carbon.

Low temperature favors the growth reaction and will result in a higher average molecular weight product. At high temperatures, the average molecular weight 'will be lower because the transalkylation reaction predominates. The proportion of C alpha olefin in the product tends to remain relatively constant with temperature changes within the most preferred range of this invention, with lower temperatures favoring a relatively higher proportion of product above C and higher temperatures favoring a relatively higher proportion of product below C In view of the fact that the production of normal alpha olefins is the object of the above reactions, ethylene is the sole olefin which can be employed in the charge. The normal alpha olefins produced will have from four to about 40 carbon atoms and will be primarily liquid with practically no solid polymer produced except as an undesired hy-product. The normal alpha olefins produced, particularly the C C and C alpha olefins, have high utility for the production of detergents.

The catalyst employed in the alpha olefin process can be defined by the following structural formula:

'a it e d wherein M is a metal selected from the alkali or alkaline earth metals and a can be either 0 or one; M is a metal selected from the group consisting of aluminum, gallium, indium and beryllium and b can be either 0, one or two, except that a-I-b is at least equal to one; R is selected from the group consisting of monovalent saturated aliphatic or alicyclic radicals, monovalent aromatic radicals or any combination thereof; X is selected from the group consisting of hydrogen and halogen. The sum of c and d is equal to the total valences represented by the metals, and when X is a halogen 0 must be at least one. Examples of catalysts which can be employed include Be(C H etc. The catalyst can be used as such, but preferably is employed with about 70 to about 98 percent by weight thereof of an inert hydrocarbon solvent such as saturated aliphatics (n-pentane, isopentane, hexane, n-heptane, isooctane, n-dodecane, merusol oil, paraffinic oils, kerosene, etc.), .alicyclics such as cyclohexane, cyclopentane, etc., aromatics such as benzene, toluene, etc. Since it is desired to produce a liquid alpha-olefin product rather than a relatively high molecular Weight solid polymer, the catalyst defined above should be substantially free of catalyst components such as, for example, TiCl which tend to cause production of relatively high molecular weight solid polymers. The amount of catalyst required herein is not critical and can be from about 1X10- to about 1 x10- moles thereof per mole of ethylene.

The temperature of the reaction can range from about 285 F. to about 615 F., generally, from about 350 F. to about 430 F., preferably, and from about 380 F. to about 400 F., most preferably. The upper range of pressure employed is not critical and can be as high as about 1000 atmospheres or even higher, but the lower pressure range, however, is critical. The pressure should be sufficiently high that most of the alpha-olefin product is a liquid under reaction conditions and so that the catalyst and most of the ethylene are dissolved or dispersed in said liquid. As soon as liquid alpha-olefin product is produced, the catalyst tends to entirely dissolve therein. It is important to have as high as possbile a concentration of ethylene in the phase containing the catalyst, otherwise liquid olefin product rather than ethylene will tend to react with the catalyst to produce vinylidenes. Therefore, the pressure should be sufiiciently high to force as much ethylene as possible into the liquid phase together with the catalyst. After there has been a conversion of to percent of the ethylene, there is suflicient liquid product to dissolve substantially all the ethylene and produce a single homogeneous phase in the reactor. Thus, the pressure in the reactor must at all times be at least about 1000, and preferably at least about 2000 pounds per square inch gauge.

When it is desired to terminate the reaction, the product is withdrawn from the tubular reactor and is reduced in temperature and pressure, whereupon most of the gaseous olefins are flashed off. The liquid product is then treated in any suitable manner to deactivate the catalyst and the desired product fractions are recovered. The catalyst may be deactivated, for example, by contact with sufficient acid, base, Water or alcohol to react stoichiometrically with the catalyst. When an acid or base is employed an aqueous layer is formed, which is then separated from the organic layer, and the remainder, including the solvent for the catlayst, can be separated into its component parts by distillation. If desired, the catalyst can be deactivated by contact with oxygen or halogens or any other material which reacts with and suitably destroys the catalytic activity of organometallic compounds. In a preferred method the aluminum catalyst is removed from the alpha-olefin product by reaction with caustic solution to form Na OAl O plus paraffin as follows:

It is shown in Ser. No. 153,815, filed Nov. 21, 1961, now abandoned, that the amount of the desired normal alpha olefin in the product is always greater when the polymerization reaction is carried out in a tubular or coil reactor rather than in a single continuous stirred autoclave or series of stirred autoclaves for a given total conversion of ethylene to some kind of polymer. That application explains that in order to achieve high. selectivity toward normal alpha olefins the reactants and product should flow substantially as a column through the tube whereby there is a minimum of backmixing so that the percentage of normal alpha-olefin product increases throughout the length of the reactor. Since a given molecule of aluminum alkyl catalyst can undergo growth and transalkylatio reactions repeatedly, it is important that ethylene charge and catalyst be permitted a high residence time in order to achieve a high catalyst efiiciency, i.e., the production of a large amount of normal alpha olefins per mole of aluminum alkyl catalyst charged. A high residence time and avoidance of backmixing is most conveniently achieved by utilizing a very long tubular reactor.

However, there is a severe practical limitation on tube length. This limitation is tied to the problem of maintaining as constant a temperature as possible at substantially every point along the length of a very long tubular reactor. The molecular weight distribution of the normal alpha-olefin product, in addition to rate of conversion, is determined by the temperature of the reaction and it is therefore important to maintain as constant a temperature as possible along the length of the reactor tube. If the tube length is very great, it is apparent that concurrent or countercurrent flow of a coolant along the length of the tube on the outside thereof will produce a temperature gradient in the cooling fluid and, therefore, also in the reactor tube. To overcome this disadvantage, the reactor tube is adavntageously submerged in a bath of a pressurized boiling liquid, such as water, whereby a constant temperature is maintained throughout the entire body of cooling fluid. A change in the pressure exerted on the boiling liquid produces a rapid change in temperature at every point throughout the liquid bath, especially because of the agitation provided in the bath because of the boiling of the coolant.

FIG. 2 shows a tubular reactor system for the practice of this invention wherein ethylene is charged to a very long tubular reactor through a flow control valve 12. Tubular reactor 10 is disposed substantially entirely within outer shell 14. Cooling water is charged to shell 14 through line 16. Level control valve 18 maintains a constant water level 20 Within the shell which completely submerges reactor 10. A relatively small stream of alkylaluminum catalyst, such as triethylaluminum, dissolved in a suitable solvent is pumped by positive displacement action to an intermediate point 22 in coil 10 through line 32 and valve 34 so that the region 24 of said coil upstream from point 22 serves as an ethylene preheat zone and the region 26 of said coil downstream from point 22 serves as a reaction zone. Point 22 is essentially the point in said reactor coil closest to the inlet end wherein the ethylene is substantially effectively preheated to the reaction temperature. Thereafter, regulation of steam pressure within shell 14 by means of steam pressure control valve 28 in line 30 establishes the temperature of the boiling Water throughout shell 14 and maintains a uniform reactor temperature substantially throughout the length of reaction zone 26 of the coil 10. Reaction zone eflluent comprising predominantly normal alpha olefins, unreacted ethylene, and catalyst is discharged through reactor pressure control valve 36, whereat the pressure is reduced to between about 50 and 1,000 pounds per square inch gauge, and is then discharged through cooling chamber 38 whereat product temperature is reduced to the lowest practical temperature While still maintaining the product in a liquid state ,i.e., to about 150 F., by means of water charged through line 40 and removed through line 42. Finally, product which is cooled and at a reduced pressure is passed through line 44 and a product measuring device 46, such as a flow recorder or chromatograph, and is then discharged through line 48 to a caustic treatment chamber, not shown, for removing the catalyst from the desired normal alpha olefin product by reacting the aluminum with caustic to produce sodium aluminate and paralfins.

In order to achieve the highest conversion of ethylene to normal alpha olefin per mole of catalyst used the length of the reactor is made as long as possible. For example, tube 10 can comprise between about 500 and 20,000 feet of about oneto four-inch pipe. There are a number of reasons for utilizing a very long tubular reactor. First, a 'very long tubular reactor permits excellent heat transfer for removal of heat of reaction. Secondly, it advantageously reduces backmixing for the reason explained above. Thirdly, a long reactor length permits achievement of a high catalyst efliciency because of additional conversion per mole of catalyst. Finally, a long reactor length tends to minimize the percentage of paraflin in the alpha olefin product. The final reason is based upon the fact that upon separation of the alkyl aluminum catalyst by treatment with caustic the alkyl components of the catalyst are converted to paraffins which have boiling points close to those of the most desired alpha-olefin components of the product and are therefore difficult to remove from the desired normal alpha olefins. Since the absolute amount of parafiins produced is fixed by the quantity of catalyst used, the greater the quantity of alpha olefins produced with said catalyst the smaller will be the percentage of paraflins in the product.

The steam pressure in shell 14 is maintained at about 50 and 500 pounds per square inch, generally, and at about between 140 and 340 pounds per square inch, preferably. The reactants in reaction zone 26 are generally at a temperature only about 10 F. to 12 F. above the bath temperature. As noted above, the reaction temperature not only affects the degree of conversion of ethylene but, more importantly, it also establishes the molecular weight distribution of the alpha-olefin product. Since relatively low reaction temperatures favor conversion to relatively high molecular weight product it is important to preheat the ethylene to within about 1 F. to 10 F., generally, and 3 F. to 6 F., preferably, of the coolant bath temperature prior to catalyst addition. It is believed that the relatively high molecular weight alpha olefins produced at low reaction temperatures grow into polymers which can foul the downstream region of the reactor tube and thereby increase the frequency of periodic reactor down times due to fouling because of polymer formation. For example, it was found that recycle of a portion of the high molecular weight alpha olefins increases the amount of solid polymer produced. For this reason, it is important not to add catalyst to the reactor tube until the ethylene has been preheated to as near as possible to reaction temperature, and at least to within about 10 F. of reaction temperature.

Finally, the reactor tube should not be so long that more than about 75 Weight percent, generally, or more than about weight percent, preferably, of the ethylene is converted to product. The reason is that at high conversion levels, there arises excessive competition between olefin product and ethylene in the growth reaction, whereby conversion to vinylidene compounds becomes excessive.

EXAMPLE Into a tubular reactor with a length to inside diameter ratio of 34,547 was continuously fed a charge material of the following weight percent composition: ethylene 91.82 percent, butene and higher 3.825 percent, triethylaluminum 0.228 percent and lubricating oil (catalyst solvent) 4.l24 percent. The ethylene was introduced into the preheat zone of the reactor and at the exit of the preheat zone met with the incoming catalyst solution feed. The reactor pressure was maintained at 3350 pounds per square inch and the reaction temperature as measured by thermocouples in the reaction stream was maintained at 393394 F., throughout the length of the reactor. Temperature control of the reaction mixture was maintained by keeping the reactor submerged in a pressurized boiling water bath in which the pressure was 202 pounds per square inch. The 202 pounds per square inch steam pressure corresponds to a temperature of 389 F. At the outlet of the reactor the pressure of the reaction mixture was lowered to 600 pounds per square inch and the total reaction mixture was passed through a water cooled heat exchanger in which the temperature of the product was lowered to l45-150 F. A sample of the total product was withdrawn after leaving the product cooler. Unreacted ethylene and low boiling product olefins were allowed to vaporize away by heating the reaction mixture to 180 F. for one-half hour. The product left behind contained liquid product olefins, the aluminum alkyl catalyst and catalyst solvent. Analyses were made on this stabilized liquid product for its dodecene and aluminum content. From these data it was calculated that 47.6 percent of the ethylene fed to the reactor had been converted to alpha olefins. The extent ofconversion as determined by an entirely independent method was 46.5 percent.

Various changes and modifications can be made Without departing from the spirit of this invention or the scope thereof as defined in the following claims.

We claim:

1. In a method for measuring the extent of conversion of ethylene to a product comprising predominantly liquid alpha olefins together with gaseous alpha olefin in the presence of an alkyl aluminum catalyst in a controlled temperature range wherein the ratio of liquid and gaseous alpha olefins to aluminum in the product is compared with the ratio of ethylene to aluminum in the charge, the improvement comprising determining substantially the total liquid and gaseous phase alpha olefins for purposes of said measurement by obtaining a sample of the liquid phase of said product and measuring the amount in said sample of the alpha olefin component contained therein whose proportion in the total liquid and gaseous phase alpha olefin product remains relatively steady with temperature fluctuations within said temperature range.

2. The method of claim 1 wherein said catalyst is triethylaluminum.

3. The method of claim 1 wherein said product comprises predominantly C to C normal alpha olefins.

4. The method of claim 1 wherein said controlled temperature range is between about 390 F. and 400 F. and said alpha olefin component is dodecene-1.

5. The method of claim 1 wherein said controlled temperature range is between about 370 F. and 380 F. and said alpha olefin component is tetradecene-l.

6. The method of claim 1 wherein said conversion is carried out in an elongated tubular reactor.

7. In a method for measuring the extent of conversion of ethylene to a product comprising predominantly liquid alpha olefins together with gaseous alpha olefin in the presence of an alkyl aluminum catalyst in a controlled temperature range wherein the ratio of liquid and gaseous alpha olefins to aluminum in the product is compared with the ratio of ethylene to aluminum in the charge, the improvement comprising determining substantially the total liquid and gaseous phase alpha olefins for purposes for said measurement by obtaining a sample of the liquid phase of said product and measuring the amount in said sample of the group of alpha olefin components contained therein whose proportion in the total liquid and gaseous phase olefin product remains relatively steady with temperature fluctuations within said temperature range.

8. The method of claim 7 wherein said catalyst is triethylaluminum.

9. The method of claim 7 wherein said product comprises predominantly C to C normal alpha olefins.

10. The method of claim 7 wherein said conversion is carried out in an elongated tubular reactor.

References Cited UNITED STATES PATENTS 3,310,600 3/1967 Ziegler et a1.

PAUL M. COUGHLAN, JR., Primary Examiner U.S. Cl. X.R.

mg UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No. 3,531,253 Dated September 2L 1970 Inventor) Herbert B. Fernald, William Gall and Alfred N. Kresge It is certified that error appears in the above-identified patent and that said Letters Patent are hereby corrected as shown below:

Column 3, line 49, "Weight of the product sample." should read Weight of the liquid product sample.--.

m -ml 1. mm, m. Au j ()ffi mnion of Patents