US3254213A - Determination of aluminum concentration in hydrocarbon streams by use of k-capture radioactivity - Google Patents

Description

I May 31, 1966 R. J. FANNING ETAL 3,254,213 DETERMINATION OF ALUMINUM CONCENTRATION IN HYDROCARBON STREAMS BY USE OF K-CAPTURE RADIOAGTIVITY Filed Jan. 12, 1962 5 Sheets-Sheet 1 maimq 932% Q INVENTORS 20,592.- J, FA/v/vf/vq f 000M): PEP Mix/are May 31, 1966 I DETERMINATION OF AL R J. FANNING ETAL UMINUM CONCENTRATION IN HYDROCARBON STREAMS BY USE OF K-CAPTURE RADIOAGTIVITY Filed Jan. 12, 1962 5 Sheets-Sheet 2 lllll lllll o 5 MM CELL A JMM can I O I INVENTORS 205527 J, FA/v/v/w 5 T .4 E- Z ONALO 5241:? BY 4 6 May 31, 1966 R J. FANNING ETAL 3,254,213

DETERMINATION OF ALUMINUM CONCENTRATION IN HYDRQCARBQN STREAMS BY USE OF K-CAPIURE RADIOACTIVITY Filed Jan. 12, 1962 :5 Sheets-Sheet 3 20 475 [6r Vou/M) 22% ATE fir 24% A72; [5r l/aLuMk) 475 [5r VOLUME) ticles emanating from a strontium 90 3,254,213 DETERMINATION OF ALUMINUM CONCENTRA- TION IN HYDROCARBON STREAMS BY USE OF K-CAPTURE RADIOACTIVITY Robert J. Fanning, Ponca City, Okla., and Ronald G. Bruce, Lake Charles, La., assignors to Continental Oil Company, Ponca City, Okla., a corporation of Oklahoma Filed Jan; 12, 1962, Ser. No. 165,853 1 Claim. (Cl. 250-435) This invention relates to a process for determining the concentration of aluminum in hydrocarbon streams.

More particularly, but not by way of limitation, the invention relates to a process for continuously determining both bound and free alumina present in flowinghydrocarbon plant streams, and for controlling the aluminum content of such streams in accordance with such continuous determination. a

In an important, more specific aspect, the invention relates to a process for continuously determining the percarbons, and the use of radiation-induced X rays to determme the concentration of tetraethyl lead in a flowing stream of gasoline.

In the latter process, X rays produced by bombarding molybdenum foil with beta parsource are passed through the sample of leaded gasoline. The intensity of the beam of X rays is measured after it has passed through the sample, and the concentration of tetraethyl lead in the hydrocarbon stream is determined by the decrease in beam intensity caused by the absorption of X rays by the lead atoms. A

In the sulphur determination, X ray absorption also forms the basis of the analysis, but the source of the X rays utilized in a radiosotope, preferably iron 55. Like I a number of other radioisotopes, iron 55 demonstrates the phenomenon of K-capture radiation. In this phenomena, one of two electrons in the innermost, or K, shell of the atom of the radioisotope is captured by the nucleus, thereby transmuting the atom into the next lower element of the Periodic Table. When the empty space left by the captured electron is filled by another electron, most frequently from the L shell, the excess energy is radiated as X rays. It is these X rays produced by an iron 55 source which are passed through a small cell containing the sulphur-in-hydrocarbon sample. The extent to which the X rays are absorbed depends upon the amount of sulphur present. The transmittance of the X rays is measured using a Geiger tube and an electronic scaler. The percent of sulphur is then calculated from the observed absorbance, the density of the sample, and the known or calculated carbon-hydrogen ratio of the sample.

The present invention comprises a novel method for continuously determining the concentration of aluminum in a flowing hydrocarbon stream using soft X rays emitted by a radioisotope as a result of the K-capture phenomena.

Although both bound and free aluminum can be de- United States, Patent such processes, accurate and continuous control is required by virtue of the inevitable production of certain side products, the quantity of which depends upon the quantity of aluminum trialkyl in the stream. The data continuously obtained by the method of the invention may be employed for continuously and automatically maintaining the alumina trialkyl content of the streams within optimum limits for carrying out the alcohol synthesis.

The X rays which are employed in the process of the present invention preferably have an effective Wave length of about 2.0 A. For X rays of this wave length, aluminum has a mass absorption coeflicient of 103.9 cm.'-/gm., while for hydrogen it is only 0.49 cm. /gm., and for carbon 9.59 cm. /gm. For oxygen, the mass absorption coeflicient is 24.2 cm. /gm. Thus, at this Wave length, for a reasonably constant C/H ratio, soft X ray absorption varies substantially according to the variation of the aluminum concentration in a hydrocarbon stream. Although several other radioisotopes may be constructed of a material which is relatively transparent to the soft X rays used as compared to aluminum; passing the X rays through the sample cell, and continuously measuring the amount by which the intensity of the beam of X rays is attenuated by its passage through the sample. The continuous measurements so obtained are converted to a recordable signal and/or a signal which may be usedfor controlling the opening and closing of a solenoid valve, or other suitable means for controlling the addition of aluminum or hydrocarbon to the stream as desired.

Ashas been indicated, the method of the invention is particularly useful in permitting continuous analysis and control of the aluminum trialkyl content of a hydrocarbon stream to be effected. .Such streams. are utilized in the production of alcohols by the process of ethylating an aluminum trialky1,-usually aluminum triethyl or aluminum tripropyl, to lengthen the alkyl chains to conform to the molecular weight of the alcohol desired; then oxidizing such long chain aluminum trialkyls to produce aluminum trialkoxide; and finally hydrolyzing the latter compound to the alcohol. All of the reactions are carried out in an inert organic solvent. In such alcohol producing processes, there is a critical need to known the percent of bound aluminum present as an aluminum alkyl or alkoxide compound in the various streams. This .is particularly true at certain key points in the process, such as before and during the 'ethylation step in order to closely control the chain growth, and also before the hydrolysis step.

Heretofore, the organo-aluminum compound content of the alcohol process streams has been determined by periodically manually taking samples of the various streams and then transporting these to the control'laboratory for wet chemical analysis. A number of disad vantages characterize this type of analysis and control. Perhaps=the most obvious of these is the inability to maintain constant and accurate control by this method. By the time the relatively lengthy wet chemical analysis of the sample has been completed, the composition of the stream from which it was taken may have changed radically. In any event, the analyses of the samples can, at best, only give, an indication of what the composition of the streamhas been previously, not its current com Patented May 31, 1966 position. There is also, of course, the need to frequently sample the streams, requiring for this both sample collectors and carriers and laboratory personnel to run the analyses.

Perhaps the most serious disadvantages of the periodic sampling-wet chemical analysis type of process control in such alcohol synthesis processes result from the properties of the aluminum trialkyl compounds utilized. These compounds are pyrophoric, and are extremely reactive with a number of materials combining violently with air and Water. Thus, not only is the sampling and subsequent handling of this material extremely dangerous and requisite of the utmost caution, but it is very difficult to obtain samples of the various streams and subject these samples to analysis without allowing the aluminum trialkyl to be chemically altered to some extent by contact with airand moisture, thus giving an inaccurate, or at least, unreliable analysis.

The present invention permits the aluminum trialkyl content of the flowing hydrocarbon streams of the described alcohol process to be continuously, accurately and safely determined. Moreover, the continuous determinations which are made give an accurate indication of the current character of the streams being analyzed.

The data obtained may then be used to continuously control the addition of the aluminum trialkyl to the stream so that that its concentration remains within optimum limits. Since the aluminum trialkyl is continuously retained in an inert hydrocarbon solvent, there is no danger of explosion.

Other advantages of the process of the invention are the economies realized by the reduction in the number of operating personnel required, the saving in space and equipment effected when the relatively compact soft X ray equipment is employed, and the speed with which analytical data may be obtained. We have found that by careful control of process conditions, the percent of aluminum present as aluminum trialkyl may be determined to within 0.1 percent of that actually present.

The process of the invention and its advantages will be more fully understood when the following description of the steps and conditions of the process is considered in conjunction with the accompanying drawings which illustrate schematically the apparatus employed in the process, and the results obtained using the process.

In the drawings:

FIGURE 1 is a liquid flow and electronic circuit diagram illustrating the apparatus utilized in practicing the process of the invention.

FIGURE 2 is a graph illustrating the variation in absorption of soft X rays of 2.05 A. wave length by a number of samples comprising varying percentages of aluminum trialkyl in kerosene. Cells of 5 mm. and 3 mm. effective thickness were utilized to give the two lines plotted on the graph.

FIGURE 3 is an illustration of the strip chart of a continuous recorder showing the trace obtained when the recorder is used to continuously record the aluminum content of a flowing hydrocarbon stream.

Referring now to the drawings in detail, and particularly to FIGURE 1, reference character designates a pipeline utilized in a process, such as the alcohol synthesis process hereinbefore described, for carrying an organo-aluminum reactant entrained in a hydrocarbon solvent. In the alcohol synthesis process, the organoaluminum compound utilized is an aluminum trialkyl, and the hydrocarbon carrier is generally a kerosene fraction having an API gravity of between 39 and 43. Other hydrocarbon solvents which might be employed include, but are not limited to, a petroleum naphtha fraction containing little or no unsaturated or aromatic hydrocarbons and sold under the trade name Soltrol, and also a substantially pure iso-octane stream. The primary considerations in selection of the hydrocarbon carrier is that it not be reactive with the aluminum trialkyl, and that it contain predominately compounds composed only of carbon and hydrogen atoms. A sampling line or conduit 12 communicateswith the process line 10 for conducting the sample to be analyzed from the process line into a sample cell designated by reference character 14 and having cell faces 14a.

Enroute tothe cell 14, sampling conduit 12 passes into a hollow cooling jacket designated generally by reference character 15. The cooling jacket 15 is of double walled construction, having an outer wall 16 and an inner wall 18. The outer and inner walls, 16 and 18, respectively, define a chamber 20 through which a suitable coolant liquid is circulated by way of the coolant inlet conduit 22 and coolant discharge conduit 24. Although other materials may be utilized as a coolant, we prefer to use kerosene for this function because of its ready availability and its inertness to reaction with aluminum trialkyl. Materials which react vigorously with the aluminum trialkyl, such as water, should not be utilized as the cooling material because of the explosion hazard which is involved in such use.

After the sampling conduit '12 enters the cooling jacket 15, it describes a number of convolutions 26 around the internal wall 18 and in the chamber 20, so that an enhanced opportunity for heat exchange between the sample and the coolant material is provided. In general, the sample from the process stream will be relatively warm as it enters the cooling jacket 15, having an average temperature of from 140 to 180 F. The kerosene which is circulated through the chamber 20 in the cooling jacket 15 is maintained at a temperature of approximately to F. so that the temperature of the sample is lowered to this value before .the conduit 12 leaves the chamber 20 and enters the space 28 existing inside of the inner wall 18 of the cooling jacket 15. The step of initially cooling the sample to a temperature of between 90 and l110 F. is an important step of the process for reasons subsequently to be explained.

After the sampling conduit 12 enters the space 28 within the inside wall 18 of the cooling jacket 15, it is passed in continuous convolutions 30 around an isothermal block 32. The isothermal block 32 is internally heated by heating elements (not seen) which are connected to electrical leads 34 passing out through the bottom of the cooling jacket 15. The temperature of the isothermal block 32 is precisely controlled by means of a thermister 36 placed in the top of the block, and connected to the controller (not shown) of the heating elements located inside the block. As the sample is heated by passing through the coils 30 of the conduit 12 around the isothermal block 32, its temperature is raised .to a temperature exceeding about F.

The control of the temperature of the sample which enters the sample cell 14 is very important to the efficient performance of the process of the present invention. We have found that a temperature variation of /2 F. may cause an error 0.5 percent in the aluminum determination. In order to bring the sample stream to the desired temperature, and to maintain it within /2 F. of such temperature, we have found that it is of considerable advantage to initially cool the stream below the desired temperature, and then raise the temperature of the stream to the desired value by the use of the isothermal block. Very close and accurate control of the temperature of the block 32 can be maintained utilizing the thermister 36 and a thermocouple (not shown) located in the space 28 inside the inner wall 18 in the cooling jacket 15.

Having adjusted the temperature of the sample stream to the desired value within the range of 115 to about F, the sample is immediately passed through the sample cell 14 located adjacent the isothermal block 32. The construction of the sample cell 14 is important to the proper accomplishment of the invention. As a first consideration, the effective thickness of the cell 14 is of importance, and we have found that an effective cell thickness of 3 to 5 mm. gives the best results. Thinner cells cause a marked loss in the sensitivity of the X ray detecting instrument, while thicker cells require a stronger source of radiation. The material of construction of the cell faces 14a is also of extreme importance, particularly when the method of the invention is to be utilized in determining the concentration of aluminum present in a hydrocarbon stream as the highly reactive aluminum trialkyl. Such material of construction must be relatively transparent to soft X rays of the wave length employed, must withstand bursting pressures up to 100 lbs. per sq.

inch, must have an operating temperature of at least 180 F., must be nonreactive with-aluminum.trialkyl and must have good dimensional stability. The requirement. that the cell faces be substantially transparent to X rays of 2.05 A. wavelength eliminates the usual metals, as well as those plastics which contain fluorine or heavier elements. Even plastics containing oxygen tend to impair the free passage of the X rays therethrough. A material which most nearly meets these cell face requirements is high density polymerized polyethylene sold under the tradename of Type #2 Marlex. Othen materials which may be utilized include industrial diamond, certain types of graphite, polypropylene and high molecular weight poly-isobutylene. When Marlex is employed as the material of construction of the cell faces 14a, a thickness of 0.015 inch may be employed for the sheet of Marlex at each face. The preferred maximum tempera- .ture within the previously cited temperature range of 1l5-l80 F. is based upon the operating temperature of the Marlex cell faces. However, higher temperatures might be used in the case of other cell face materials.

As the sample continuously flows through the sample cell 14, it is continuously exposed to radiation with a beam of soft X rays emanating from a suitable radioisotope source 40. The soft X rays should have a wave length in the vicinity of 2.0 A. and when iron-55 is the radioisotope used, the weighted wave length is 2.07 A. and the efl ective monochromatic wavelength is 2.05 A. In performing the method, we have employed l0 millicuries of iron-55 as the X ray source material.

The low level gamma radiation, or soft X rays, from the source 40 pass through the Marlex cell faces 14a and through the sample. A portion of the rays are absorbed by the aluminum atoms present in the sample as aluminum trialkyl so that the intensity of the X ray beam is diminished as a result of its passage through the sample. The rays exiting'from the cell impinge upon a suitable sensing or detecting instrument, such as a Geiger-Muller tube 42, disposed on the opposite side of the sample cell 14 from the iron-55 source. The Geiger-Muller tube 42 is sensitive to the soft X rays and provides a small positive pulse for each X ray received. Using. the millicurie source, the count level at 25% aluminum trialkyl is approximately 20,000 counts/min. when an effective tube 42. If the count level exceeds 50,000 counts/mm,

as in the case of low concentrations of aluminum, the

, efiiciency of the Geiger-Muller tube 42 falls off and it is desirable to employ a more efficient detector such as a solid-state detector or a scintillation crystal.

The pulses from the Geiger-Muller tube 42 are carried to a pre-amplifier 44 where they are amplified and inverted. The pulse signals are then passed on to a rate count meter 46 which provides an output proportional to the time average of the signal pulses. The zero reading and span of the rate meter 46, may be made adjustable by the provision of suitable span-zero control circuitry if desired. The output from the rate meter 46 is visually displayed on the recorder 48, and the output signal may also be directed to a suitable control device (not' shown), such as a solenoid valve, for controlling the addition of the aluminum trialkyl or kerosene to the process of the space 28 in the sampling conduit 12 and is re turned to the process line 10. The fluid time constant for the system is approximately 4 minutes; that is, this is the time required for sample to flow into the sample cell 14, from the process line 10. The flow rate of the sample through the system is about ml./min. It will thus be seen that a continuous determination of the concentration of aluminum in process streams may be carried.

out with very little time lag occurring between initial diversion of the sample from the process stream 10, andfinal readout of the concentration value upon the recorder 48. Those skilled in the art, will, of course, appreciate that an initial calibration of the apparatus using samples containing known concentrations of aluminum is required before commencing the determinations.

As previously mentioned, control of the temperature of the sample is of extreme importance on obtaining ac curate detection and control. The apparatus is designed to permit the sample to be maintained within /2 F. of a desired temperature within the range l15180 R, which range has been found most suitable for use in analyzing the aluminum trialkyl-bearing process streams when the specified Marlex cell construction is used. This control is attainable largely because of the greater ease with which the sample is brought smoothly and quickly to the desired temperature by initially cooling it below such temperature. An accuracy of 10.5% is attainable when the temperature is maintained constant within the specified tolerance, and we have maintained sutficiently close temperature control to repeatedly obtain an accuracy of i0.1%.

In FIGURE 2 of the accompanying drawings, a semilogarithmic graph is illustrated in which the weight percent of aluminum in a kerosene stream is plotted against the counts per minute indicated by the ratemeter. Two cell thicknesses of 3'mm. and 5 mm. were employed to obtain the two lines which are plotted. The thicker cell (5 mm.) appears to produce better sensitivity to changes in the aluminum trialkyl concentration.

FIGURE 3, illustrates a trace recorded by a strip chart recorder as the concentration of aluminum triethyl in a kerosene stream is varied between 30 percent and 20 percent. Prior to the commencement of the passage of the sample through the system, a zero and span control attached to the ratemeter 46 was set to present the most sensitive, as well as acceptable, output presentation on the recorder. The concentration of aluminum triethyl in the kerosene stream was varied in 2 percent increments. Although the amplitude or height of the trace as measured along the coordinate axis of the chart increases a greater amount with each succeeding decrease of 2 percent in the aluminumtrialkyl concentration, this is to be expected in view of Beers law according to which the absorption of the soft X rays by the aluminum trialkyl varies exponentially as the concentration thereof instead of linearly.

From the foregoing discussion, it will be apparent that the present invention provides a rapid and accurate method of determining the concentration of bound or free aluminum in a flowing hydrocarbon process stream. The method permits, for the first time, close, continuous control over the composition of such streams to be maintained.

Although the process is particularly useful in permitting streams carrying highly reactive aluminum trialkyl to be analyzed and controlled, and has been described,

by way of example, as it is employed in such analysis and principles herein described and constituting the basis of the invention. Variations of this type are therefore con sidered to be encompassed by the spirit and scope of the invention except as the same may necessarily be limited by the appended claim.

We claim:

The process of continuously determining the concentration of aluminum trialkyl in a liquid hydrocarbon stream which comprises:

adjusting the temperature of the stream to a temperature between about 90 F. and 110 F. and thereafter; Warming the stream to a temperature between about 115 F. and 180 F.;

maintaining the temperature of the stream within about /2 F. of the temperature to which said stream was warmed while passing the stream. at a flow rate of 50 to 100 ml. per minute through a sample cell which confines the liquid hydrocarbon to a thickness of 3 mm. to 5 mm.;

continuously transmitting a beam of soft X rays having a wave length of about 2.05 A. from an iron-55 radioisotope source through the prescribed thickness of liquid hydrocarbon in said cell;

continuously measuring the intensity of said beam after it has passed through the hydrocarbon liquid in said cell; and from the measurement so obtained;

determining the concentration of aluminum trialkyl in said stream on a continuous basis.

References Cited by the Examiner UNITED STATES PATENTS OTHER REFERENCES Gamma Density Controls, Extraction, column by Kyle,

Chemical Engineering Progress, vol. 53, No. 11, Novem 20 her 1957, pages 551 to 555.

RALPH G. NILSON, Primary Examiner.

I. W. LAWRENCE, Assistant Examiner.

Images