604 MANLEY: THE ANALYSIS OF PEI‘ROL - KEROSINE MIXTURES, [Vol. 75 The Analysis of Petrol - Kerosine Mixtures, with Specific Reference to the Boiling-point BY C. H. MANLEY (Read at the meeting of the Society on Wednesday, Nay 3rd, 1950) SyNoPsIs-The value of the density determination, as well as that of the distillation range, as a means of detecting and determining the proportion of kerosine present in a petrol - kerosine mixture is discussed. A third method, giving results in close agreement with the first two methods, is described. This depends upon the determination of the boiling-point of the motor spirit under examination in an all-glass reflux apparatus, in which the boiling-points of mixtures of known composition have been previously ascertained. FOLLOWING the coming into force on June lst, 1948, of the Motor Spirit (Regulation) Act and the Regulations made thereunder, as well as of the Motor Fuel (Control) Order, the Public Analyst found himself called upon not only to identify “commercial” petrol (with or without the red colour removed), but also to examine “private” petrol for the presence of kerosine.In this connection it is to be noted that the Motor Fuel (Control) Order, unlike the Motor Spirit (Regulation) Act, was made under the Defence (General) Regulations, 1939, with the result that the penalties involved for offences committed against it differ materially from those liable to be incurred under the Act. Further, Section 5 (5) of theNov., 19501 WITH SPECIFIC REFERENCE TO THE BOILING-POINT 605 Order makes it an offence (except with the authority of a licence) to use any kerosine, whether or not in admixture, for supplying motive power to any motor vehicle chargeable with duty under Section 13 of the Finance Act, 1920, and Section 4 (2) of the Order states when motor spirit shall be deemed to contain kerosine in admixture. No methods, however, are given for calculating the proportion in which any kerosine so found is present.Section 4 (2) simply states that motor spirit or any other hydrocarbon oil shall be deemed to contain kerosine in admixture if, with a closed flash-point less than 73" F., the final boiling-point thereof exceeds 215" C.; or if, with a closed flash-point greater than 150" F., not less than 20 per centum thereof distils at a temperature not greater than 210" C.The method 1.P.- 28/42 in the Institute of Petroleum publication, "Standard Methods for Testing Petroleum and its Products," only details apparatus and defines distillation procedure, and those standard textbooks that have been consulted only describe tests that are applicable to individual petroleum fractions. It is important therefore that, if marked discrepancy is to be avoided in the conclusions reached by two analysts examining portions of the same motor spirit, there should be some agreed methods of analysis. It became evident in due course that valuable use could be made of the fact that at least 40 per cent. of ordinary petrol (or gasoline) distils below 100" C. and that where the final boiling-point of a sample exceeded 200" C., and less than 40 per cent. distilled below 100" C., the amount of petrol present could be calculated by simple proportion, the difference between 100 per cent.and the figure for petrol being kerosine. A sample of which only 20 per cent. distilled below 100" C. therefore represented a 50 : 50 mixture of petrol and kerosine, or expressed in another way, contained not less than 50 per cent. of kerosine by volume. To confirm this method, a mixture of petrol and kerosine in the proportions found in an adulterated sample of petrol was prepared and distilled, as in the example shown in Table I. TABLE I Proportion Calculated distilling below proportion of Sample Final b , p. 100" c., kerosine, % % P.260 .. .. .. . . 272" C. (96%) 10 75 30% petrol : 70% kerosine.. .. 272" C. 11 724 In examining a number of suspected samples received on one particular day, it was appreciated that valuable time could be saved in testing for samples of genuine petrol if two quick reliable tests could be applied.DENSITY AND REFRACTIVE INDEX- In the majority of the petrols examined, whether for the presence of kerosine or diphenyl- amine, either the specific gravity or the density (grams per millilitre) has been determined, measurements being originally made at 15.5" C., but later at the rather more convenient temperature of 20" C., at which all later density determinations have been carried out, 0.740 and 0.800 being taken as the densities of petrol and kerosine respectively, and the proportion of kerosine in a sample having a density in marked excess of 0.740 being calculated from the rise in density above 0.740.This was done on the assumption that the densities of mixtures of the two petroleum fractions concerned lie on a straight line, which was, in fact, borne out by direct measurement. As, moreover, it had been privately suggested that helpful indications might also result from a determination of the refractive index, this was ascertained for each of the mixtures examined, the values obtained also lying on a straight line and ranging from approximately 1-420 to 1.445. The mixtures used were made by volume, the requisite proportions being measured a t 20" C. in pipettes graduated at this temperature. The density was then determined in a stoppered 10-ml. graduated bottle correct at 20" C., and the refractive index in an Abb6 instrument adjusted to 20" C.BOILING-POINT- Whilst still relying upon the distillation and density methods for ascertaining the pro- portion of kerosine in petrol - kerosine mixtures, consideration was given to other possible means of achieving this. Eventually, determination of the boiling-point of the sample under examination seemed a possibility, and this, together with the density, has proved606 MANLEY: THE ANALYSIS OF PErROL - KEROSINE MIXTURES, particularly useful, for when both density and boiling-point have been found to be normal, it has been considered unnecessary to proceed with distillation. As petrol and kerosine consist of mixtures of substances of similar chemical composition, it seemed reasonable to conclude that a steady rise in the boiling-point would be observed with increasing percentages of kerosine.Accordingly, boiling-point determinations were carried out on approximately 50-ml. quantities of (a) petrol, (b) kerosine and (c) various mixtures of these in a Quickfit and Quartz apparatus consisting of a 150-ml. fla.sk connected to a reflux condenser (effective length, 8 inches) in which was held, by copper wire passed through the loop, a thermometer (range 0" to 250" C.), the bulb of which dipped into the liquid under examination. In order to maintain the composition of the boiling liquid as uniform as possible it was kept just gently boiling, thereby reducing to a minimum the proportion of the more volatile constituents vaporised into the reflux condenser. On plotting the results, a slightly concave curve was obtained, the temperature range covered lying between 80" C.and 200" C. The figures shown in Table I1 were obtained for the mixtures studied, and the results were plotted. [Vol. 75 TABLE I1 DENSITY, REFRACTIVE INDEX AND BOILING-POINT OF PETROL .. KEROSINE MIXTURES Kerosine, Y O 0 10 20 30 40 50 60 70 80 90 95 100 Density at 20" C. 0.741 0.747 0.753 0.757 0.763 0.769 0.773 0.779 0.784 0.790 0.791 0.795 Refractive index at 20' C. 1.421 1-423 1.425 1.429 1.430 1-434 1.436 1.437 1.440 1.442 1.443 1.444 ?P., C. 80 83 85 93 97 102 115 131 147 166 179 195 The three methods involving density, distillation and boiling-point were applied with success to four samples in which appreciable proportions of kerosine had been mixed with petrol, P.259 to 261 concerning one defendant and P.274 another, as shown in Table 111.TABLE 111 APPLICATION OF THE THREE METHODS FOR DETERMINATION OF KEROSINE Proportion distilling Density below Sample at 20" C. Final b.p. 100" c., % P.259 0.772 246°C. (92%) 20 P.260 0.779 272OC. (96%) 10 P.274 0.773 242OC. (95%) 18 P. 26 1 0.764 264OC. (94%) 24 Proportion of Kerosine by b e n s i t y t i l l a - 7 Bdp., a t 20" C., tion, B.P., C. % % % 105 63 50 52 133 65 75 72 102 40 40 48 118 55 55 64 The minimum proportions of kerosine certified for the purposes of the Court proceedings which followed were returned as 50, 65, 40 and 55 per cent. respectively, the two defendants concerned pleading guilty. Whilst the proportions deduced from the boiling-point in two of the examples quoted are higher than those (obtained by the other two methods, it should be pointed out that appreciably higher figures would result from the distillation test if the probable average figure of 45 per cent.was usetl for the proportion of petrol distilling below 100°C., when the proportions of kerosine would become 56, 78, 47 and 60 per cent. in the preceding table, and higher still if the figure of 50 per cent. (sometimes encountered) was used as the basis of calculation. As an example of a case in which suspicions proved unfounded we have that of the car owner who alleged that a garage proprietor had supplied him with petrol containing whiteNov., 19501 WITH SPECIFIC REFERENCE TO THE BOILING-POINT 607 spirit. The police sample submitted on his behalf had a density a t 20" C.of 0-734 and boiled at 78" to 79" C., both being figures normal to genuine petrol. It is appreciated that whilst most samples of petrol have densities at 20" C. of approxi- mately 0-740, a few reach 0-750 or even 0.760, and determinations of the boiling-point and distillation range may well afford useful information as to whether or not kerosine is present. Finally, it should be possible t o apply the boiling-point method to the elucidation of the approximate composition of other binary mixtures of petroleum products and to various binary mixtures of organic liquid compounds of similar composition. CITY ANALYST'S LABORATORY LEEDS DISCUSSION NR. P. MCGREGOR said that he had had considerable experience in examining samples of suspected petrol - kerosine mixtures taken from all parts of the country.When there had been an admixture the proportion of kerosine found was frequently less than 20 per cent. He would like to be able to estimate it by specific gravity or refractive index, but owing to the wide variation in the physical constants of petrol and kerosine, this would not be satisfactory. Samples taken under the Motor Fuel (Control) Order, 1948, have to be examined by the prescribed methods. When using the standard method for the distillation (1.P.-28/42), it did not follow that, if a final boiling-point above 215" C. was obtained, kerosine was necessarily present. Similar effects could be obtained if gas oil or lubricating oil were added, but then the residue after distillation was greater than that normally obtained from petrol or petrol - kerosine mixtures.It was therefore necessary to take the nature and volume of the residue into account. An approximate idea of the amount of kerosine present could be obtained by assuming that a t 200" C. all the petrol and half the kerosine had been distilled. DR. E. J. MILLER said that he had listened with great interest to Mr. Manley's talk and to the subsequent discussion, as he and his colleagues, at the North-Eastern Forensic Science Laboratory in Wakefield, dealt with a steady stream of samples of the nature described. Although they were primarily interested in the qualitative aspect of petrol - kerosine mixtures, nevertheless they felt that it was of great value to have a reasonably accurate knowledge of the quantities of each of the two "constituents" of the mixture.Their method of determining this, which they had found to be sufficiently near the truth, was to assume that in a genuine petrol all of i t (or a t least 95 per cent.) would have distilled over a t a temperature of 190° c. They distilled the mixture as laid down in 1.P.-28/42, and from the results they determined graphically the amount that had distilled over a t this temperature. The difference represented. the kerosine fraction. Any small amount of petrol that had been held back by the kerosine was compensated for by the small amount of kerosine which had distilled over a t this temperature. MR. J. HASLAM said that for the method to be valid it appeared that petrol and kerosine must be substapces of reasonably constant composition. Mr.Manley had presumably examined several samples of petrol and kerosine in the first place, and Mr. Haslam asked what variations Mr. Manley had found on application of the tests he had described to these samples. DR. J. H. HAMENCE said that he had listened to the paper with considerable interest, as he had been called upon to examine a number of samples of private petrol for the presence of kerosine. He would like to ask the author if he had considered the influence of benzene on the physical constants which formed the basis of his assessment of the proportion of kerosine. He understood on good authority that the proportion of benzene in private petrol varied considerably, and this was borne out by analysis of many hundreds of samples which he had made during the past two years.The presence of benzene had a profound effect on the specific gravity of the petrol, and also affected considerably the initial boiling-point. With these facts in mind, he was of the opinion that any calculations based on specific gravity or boiling-point might well be rendered completely erroneous if benzene in fair quantities were present. He pointed out that in some of the examples which had been quoted by the author the high specific gravity might equally well be due to the presence of benzene in the petrol, and not to kerosine. Dr. Hamence said that in his view, one of the most reliable methods of arriving a t an assessment of the proportion of kerosine in a petrol - kerosine mixture was from a study of the boiling-point curve, particularly the fractions distilling above 190" C., as compared with the boiling-point curves given by mixtures containing known amounts of kerosine.He found that the presence of benzene did not affect the boiling-point range above 190" C., whereas i t had a marked effect on many of the other physical constants. MR. MANLEY, replying to Mr. McGregor, pointed out that, whilst it was true that the Motor Fuel (Control) Order, 1948, prescribed special methods and apparatus for determining whether or not motor spirit contained kerosine in admixture, the analyst was left free to use such methods as he deemed fit for determining the actual proportion of kerosine once its presence had been indicated. The chances of gas oil or lubricating oil being used were decidedly small, kerosine being the most likely alternative fuel available to the motorist; in fact, in none of the cases with which he had been concerned had the allegation that kerosine was the adulterant been contested.It was true that fuels other than kerosine had (in his own608 JONES AND MORRIS: THE MICI3OBIOLOGICAL ASSAY OF [Vol. 75 experience) been used to supplement the petrol ration. Special Boiling-Point Spirits No. 2 and Aviation Spirit were examples, but their use was understandable by reason of their much lower distillation range. He had attempted to calculate the percentage of kerosine present in petrol - kerosine mixtures by having regard to the proportion distilling above 190" C., but had found that this had led to a gross under-estimate.Later, as pointed out, results not far removed from the truth were obtained by having regard to the proportion distilling below 100" C. Moreover, i t was necessary to emphasise the fact that qualitative findings alone singularly failed to impress a Court of Law, which always wanted to know how much of the offending ingredient was present, and took a lean view of an expert witness who, for instance, could not say whether the proportion was of the order of 5 or 50 per cent. In reply to Mr. Haslam's observations, Mr. Manley said that he had examined a number of samples of both petrol and kerosine, and had found them to be of reasonably constant composition. Certain variations in distillation range and density were inevitable, just as there were variations within limits in the various chemical and physical values of the numerous foods examined by the public analyst, who both in his own interest and in fairness to others concerned was careful to consider these variations before issuing his report. A similar approach had been made to the petrol - kerosine problem. Replying to Dr. Hamence, Mr. Manley stated that he had considered the possible influence of benzene on the physical constants studied inasmuch as he was aware that most of the various pool grades contained, in addition to paraffins and naphthenes, small percentages of aromatics, the proportion being characteristic of the particular grade. It was perhaps well to point out that there was no intention of condemning a product on density or boiling-point alone. In fact, when suspicion was thereby aroused, it was necessary to proceed to determine the distillation range in order to comply with Section 4 (2) of the Motor Fuel (Control) Order, 1948. If confirmation followed, then the method giving the lowest figure for the kerosine present would be used for' purposes of certification. Density and boiling-point determinations were extremely useful for genuine petrols, for where normal figures were obtained, it was rarely necessary to proceed with distillation. Mr. Manley said that he agreed with Dr. Miller regarding the final boiling-point of petrol.