V o l . 25. No. 185. M a r c h 1939.

T A N K ST R A P PIN G .*

By P . K e r r , M.A., B.Sc., A.I.C. (Member).

S y n o p s i s .

T he p ap er suggests decision in advance on th e m axim um u n certa in ty p e r­m issible in th e ta n k tab le which is to be produced. M easurem ents on th e ta n k m ay th en be m ade w ith th e lowest perm issible precision, and therefore w ith th e g reatest ease. Only those corrections an d allowances w hich are necessary to secure th e desired accuracy in th e final tab le need be included.

A pplying th is principle, th e p ap er sets o u t th e chief p recau tions necessary w hen tables of high accuracy are to be p rep ared from ex ternal m easurem ents o f ta n k circumferences. O ther m ethods o f ta n k calib ra tion are discussed only so fa r as th ey bear on th e m ethod considered.

The chief corrections to a p p aren t ex te rn al circum ferences requ ired to ob ta in th e m ean horizontal in te rnal cross-sectional a rea for each course are th en discussed. These include th e effects of th e tem p era tu re , tension or errors o f g raduation of th e tap e , and o f d isp lacem ents o f th e tap e from the ta n k surface b y vertical seam edges or o th er obstacles, or th ro u g h local d isto rtio n o f th e tan k p lates. The effects o f slight conicity an d e llip tic ity of th e courses are considered, as also th a t of t i l t o f th e ta n k as a whole. Corrections depending on th e m ean tem p era tu re o f use of th e ta n k an d its expansion u nder th e head o f oil in i t a re discussed, as also those for p la te and p a in t thickness, and for in te rnal fittings. The ad ju s tm en t o f tab les for irregu larities in th e shape of th e ta n k b o tto m is also considered briefly.

T a n k s m ay be satisfactorily calibrated in m any different ways. The m ethod discussed below is th a t of “ strapping ” or measuring apparent circumferences outside the tank. Correction is made la ter for all factors which affect the final tables.

W ith this m ethod the principal measurements m ay be taken while the tank is in use, and less im portant measurements m ay usually be taken from the drawings. In a busy installation it is seldom easy to em pty a tank completely. I t is, however, difficult to measure the effect of corrosion in the plates of old tanks w ithout this.

P r e l i m i n a r i e s .

Before beginning work, the degree of accuracy required in the final tables should be decided. This decision avoids waste of tim e in taking unneces­sarily exact measurements, and again secures inclusion of all precautions in working and corrections really required. I f oil shortages or discrepancies require investigation later, i t is essential to know how far one can rely on the tank tables.

This degree of accuracy is m ost simply expressed as a tolerance-—say, the percentage error permissible in the final tables. The exact error present in tables is never known : if i t were, the tables would be corrected. There is therefore some advantage in expressing the degree of accuracy aim ed a t or achieved as a “ probable error ” or “ standard deviation.”

Many non-technical considerations bear on th e choice of such tolerances. Time and staff m ay be limited, and costs have usually to be kep t down. A highly valuable product would call for smaller tolerances th an would

* P ap er p resen ted for D iscussion a t th e One H u n d red an d E ig h ty -fifth G eneral M eeting of th e In s t i tu te of Petro leum held on 10th J a n u a ry , 1939.

110 K ERR : TANK STRAPPING.

water. On the other hand, if a tan k table is to rem ain in use for several years, it may pay to avoid even small errors. The to ta l effect of these might ultimately add up to a considerable cash value, even w ith a relatively cheap product.

I t is sometimes argued th a t since it is im practicable to measure the amount of oil in a tank exactly, and since the uncertain ty of a single measure­ment may represent a large num ber of gallons, i t is therefore unnecessary to aim a t high precision in tank tables. In the w riter’s view, th is is incorrect. First, the unavoidable errors in measuring oil quantities are often over­estimated. Again, if gauging is properly carried out, the measurements will sometimes slightly over-estimate, sometimes under-estim ate, the oil quantity. Taken over a sufficient series of measurements, such fortuitous errors tend to cancel out in percentage. I f the tan k tables contain any given percentage inaccuracy, this error will be present in all quantity calculations. The error is systematic, and does not tend to cancel out in percentage, no m atter how m any consecutive measurem ents are averaged. The utility of good tank tables in controlling working losses is obvious.

Technical considerations also bear on the selection of tolerances. Meas­urement difficulties increase rapidly as the error allowable is reduced. The precision with which measurements m ust be taken increases. Cor­rections negligible for rougher work m ust be included in finer calibrations.

To a great extent, therefore, details m ust depend on the circumstances of the case. To be of general use, this note necessarily describes methods of high accuracy. Those who feel th a t too m any corrections and precautions are included are reminded th a t by choosing lower standards of accuracy, many of these may be neglected.

Ca l ib r a t io n b y F i l l in g .

For a given amount of care, strapping tends to give higher percentage uncertainties on small tanks than on large. A common m ethod of cali­brating smaller tanks, often applied also to calibrate irregular parts of larger tanks, is to fill measured quantities of liquid into the tank , the resulting depth of liquid being measured after each addition. For any previously chosen degree of uncertainty in the results, the trouble and expense of this method by filling increase with the size of the tank calibrated. I t is usual, therefore, to strap larger tanks and fill smaller. As the size of the tank decreases, a point is reached a t which the cost of securing a given accuracy by strapping becomes greater than th a t by filling. I t is a t this point— admittedly somewhat indefinite—th a t one should change over from the one method to the other. This note does not fully discuss methods other than th a t of strapping, and this implies th a t tanks of suitably large diameter are alone considered. These normally approximate to vertical cylinders, and other shapes are not considered here.

C a l ib r a t io n f r o m D r a w in g s .

Tanks are now usually made in standard sizes, and the accuracy of their construction is much greater than formerly. The plates for any one course are usually interchangeable. Such tanks approxim ate very closely to the

K ERR : TANK STRAPPING. I l l

dimensions shown on their drawings. Tables of very satisfactory accuracy m ay then be calculated from the drawings in advance of erection, this allowing the tank to be brought into use w ithout delay.- Such tables do not, and cannot, allow for irregular deformation of the tank

bottom on its foundations, bu t this p a rt of a tan k is seldom used for precise measurements. Oil received into or delivered from a tan k m ay be fairly satisfactorily measured w ithout bottom calibration, provided th a t the tank is not emptied. I t is perhaps the rule ra ther th an the exception to neglect such irregularities. Oil-stock figures are affected by this neglect. W here these are im portant, the tan k bottom m ay be calibrated. In some cases w ater bottom s are pu t into tanks, which eliminates the effect on stocks, bu t storage space is wasted by this if the tan k bottom is known to be tight.

Official regulations in m any countries dem and the production of tank tables from measurements made on the tan k after erection. Except for this, in the au tho r’s view the production of the final tables for standard tanks by calculation from the drawings m ight usefully be applied more widely th an a t present. The verifications necessary are perhaps most open to those who purchase a large num ber of tanks. Comparison of recorded measurements on tanks previously erected with the corresponding figures calculated from the drawings, then shows whether the standard tanks of a particular maker m atch the drawings sufficiently closely to allow tables to be tentatively calculated as suggested.

Before tables so calculated are accepted as final for a further tank , the erection engineer should report for this th a t little or no reaming of the rivet holes has been necessary before rivets were driven, and also th a t the tan k after erection showed little or no leakage a t the seams. A single circumference m ay be measured a t some convenient height on the tan k as a check th a t all is in order. Such other check measurements as are easily made m ay also be taken. The savings are considerable, for not only are measurements largely avoided, bu t the standard table calculated for any particular tan k m ay be used again for other tanks of the same size, so th a t calculation and printing are also reduced.

Even when direct measurem ents m ust be taken, the tan k drawings, if available, give a valuable check. I t is useful to calculate out beforehand the circumferences and other measurements which should be found by the strapper. A ny notable deviation of his measurements from those expected then directs a tten tion a t once to the need for careful verification of the measurements found. Especially if the crew has to be sent some distance for the work, or if th e production of reliable tables is an urgent m atter, the saving as compared w ith la ter detection and verification of the discrepancies is often considerable.

In these prelim inary calculations i t is useful to give no t only the figures expected, bu t also the tolerance allowable on each of these while still m ain­taining the required accuracy in the final tables. Rechecking of minor discrepancies in observed measurem ents is thus obviated.

R e c a l ib r a t io n .

Decision to recalibrate tankage is usually taken only when discrepancies in measurements have become troublesome. Before th is is decided on, it

1 1 2 K ERB : TANK STRAPPING.

is well to verify local measurem ent procedure on all the simpler points in which faults are likely to he found. I f visual inspection of a tan k shows it to be well made and free from flats or deformations, m easurem ent of a single circumference a t some convenient level, usually near the bottom of the second course, taken in conjunction w ith external m easurem ent of the plate thicknesses, overlaps and heights for all courses, is usually sufficient to show whether the cost of recalibration is or is no t justified. The course heights used in calculating the tank capacities m ust naturally be the internal heights, but one can usually estim ate these with sufficient accuracy from the external course heights and the seam overlaps. Much depends on the condition of the tank, but for tanks w ithout visible irregularities an excellent check is often obtained.

Further check on the accuracy with which the final tables have been constructed is often obtainable from oil transfers. The measurements made in another tank from which the newly calibrated tan k is filled, or into which it is emptied, are compared with those taken in the tank under check. Auxiliary tanks used for this m ust themselves be known to be well calibrated. The precision with which such measurements, purely internal to the installation, are made in ordinary routine is sometimes not very high. I t may be necessary to average a sufficiently long series of transfers before a good comparison is obtained. The greater the variation in percentage difference between separate comparisons, the greater the number of individual cases which m ust be averaged before the comparison is relied on.

This comparison is affected by pumping losses in th e case of volatile products. Some compensation for this evaporation m ay be obtained by taking comparisons both into and out of the tank concerned. I f the pump­ing operations are reasonably similar, then the shortage due to loss in receipts will be balanced, a t least to some extent, by the shortage due to loss in deliveries, both losses being best expressed as percentages on the quantities handled.

S t r a p p in g L e v e l s .

The upper edges of external courses form a very convenient ledge on which to rest the tape if a circumference a t the bottom of the next higher inner course is being measured. In the au thor’s view, the tem ptation to use this as a level a t which to measure a circumference should be resisted. Rivets are driven hot, and unless the plate is very thick, it is often possible to detect a local spreading of the plate. This local distortion is usually completely negligible in its real effect on the tank capacity. I f the tank is strapped a t this level, however, the distorted circumference is then taken as applying to the whole height of the course, or even to more than one course, according to the number of levels measured. The edges of plates are the parts most hable to slight irregularities. I t is preferable to avoid these levels as much as possible in strapping.

I f the tank to be strapped has been painted, a little pain t often drains down and is caught by the upper edges of outer plates. The effect is irregular, and a satisfactory circumference can often be obtained a t this luv el only after removal of the paint. The trouble is best avoided by making the rule of not strapping within, say, four inches of the seam. This rule is also applied when strapping lower circumferences on outer courses.

K ERR : TANK STRAPPING. 113

“ Referee ” methods for strapping tanks usually call for a t least one circumference to be measured on each course. I f there is sufficiently close agreement between the observed measurements and those calculated from the drawings, then some of th e courses—say, every second—m ight reason­ably be om itted from measurement. A ppropriate rules for such omissions are, however, a little difficult to form ulate in a m anner suitable for use by the strappers. I t m ay aLso be doubted whether such rules would effect substantial economies. Once the expense of getting the strapper’s crew and gear to the tank has been faced, the additional cost of strapping a few more circumferences is not very great. The saving m ight be useful in the case of a very busy crew. Those methods in which a smaller num ber of courses to be strapped is laid down in advance of any inform ation as to the regularity of the tank, leave one somewhat uncertain as to the real accuracy of the resulting tables, if not of the results themselves.

I f th is view is accepted, then a circumference is measured near the bottom of each course of the tank , w ith another near the top of the top course. These circumferences should not be measured too close to the joints, bu t should be taken, say, four to six inches above horizontal seams. The exact levels used are chosen so th a t the tape passes w ith a minimum am ount of interference between the rivets of the vertical seams.

S t r a p p in g T a p e s .

Circumferences are m ost conveniently measured by means of a steel tape. This is stretched around the tan k a t some convenient tension—usually 10 lbs. I t is a little difficult to be quite sure th a t the tension is transm itted uniformly around the tank. Possible errors resulting from th is would be minimized if the tape were of large cross-sectional area. This would be secured by using a wide or th ick tape.

Against this, a narrow tape is to be preferred if it is to pass among the rivet-heads a t the vertical seams w ith the minimum am ount of interference. I t is not desirable th a t the tape should be so wide th a t i t m ust pass entirely over these rivet-heads. Correction of circumferences so measured for the displacement of the tape by the rivet-heads would not be difficult, bu t the great practical advantage of having the tape pass nearer the tan k surface is th a t the rivet-heads help greatly in keeping the tape to its proper pa th round the tank.

Again, tapes are m ost conveniently kep t wound up on a frame when not in use. I f the tape is thick, it tends when unwound to re ta in the set given it by this winding, which leads to great practical difficulties.

Both the w idth and the thickness of the tape are therefore a m atter of compromise. I t is desirable th a t the tape should be flat, since it then tends naturally to follow its proper p a th between the supporting rivet- heads.

The methods used by reputable makers in their m anufacture of steel tapes are of very high precision. Errors of graduation in these seldom am ount to more than a few parts per 100,000 of the length measured. Tapes used in tank strapping should of course be calibrated stretched horizontally a t the tension a t which they are to be used. They vary in length with their tem perature. The tem perature a t which the small corrections occasionally

114 K ERR : TANK STRAPPING.

found necessary are given is now always 68° F. (20° C.) • I f errors are present in the tape itself, the apparent circumference figure is corrected for their effect, the corrections being given on the calibration certificate.

T a p e T e m p e r a t u r e .

Since it is difficult to measure the tape tem perature, and since this affects the readings obtained, it is desirable, when good figures are required, to strap tanks a t such a time of day th a t the air and the tan k contents are a t about the same tem perature. Especially if the p late surfaces are rusty, pitted or painted, the tape m ay easily not be in good therm al contact with the tank. In strong sunshine the high tem perature taken up by the tape in such cases is readily detectable by running the finger along it. Another method for avoiding the uncertainties which arise from this cause is to use a nickel alloy tape of very low coefficient of therm al expansion. Such tapes are, however, rather expensive. Since they are brittle, such care must be taken in their use tha t, in the w riter’s opinion, they are hardly practicable for routine strapping.

I f their tem perature changes, steel tanks expand or contract both hori­zontally and vertically, bu t such changes in their dimensions are not measurable with steel tapes, since these rapidly take up the tem perature of the materials with which they are in contact. The effect of this on tank tables is perhaps rather unexpected. I f a tank is strapped when the air, the tape and the tank plates are all a t 20° C., then, setting aside other corrections possibly necessary, the circumference measured will be correctly th a t of the tank a t 20° C., if this is the tem perature a t which the tape is correct. I f the air, tape and tank tem peratures are all raised—say, to 30° C.—then the steel tape will expand a t closely the same ra te as the steel tank. The circumference obtained will not be the true circumference of the tank at 30° C., but will still be th a t a t 20° C., the tem perature of cali­bration of the tape. I f the precautions described in the preceding paragraph have been taken, the circumferences obtained m ay and m ust be regarded as those which the tank would have a t the tem perature of calibration of the tape.

I t is then permissible and possible to correct the tables to the normal or average working temperature of the tank. I f the tank has been strapped without these precautions, a small error, the limit of which is not exactly known, will be introduced.

T a p e D is p l a c e m e n t s .

The courses of tanks, especially those of in-and-out construction, approxi­mate closely to right cylinders on the same axis, each course differing slightly in its dimensions from those above and below it. I f the tape, in strapping, is made to take a path lying in a plane a t right angles to the common axis, the circumference so measured will be approxim ately circular. The exposed plate edges define such planes, even if the tan k is tilted. The path w ich the tape should take may either be m arked in advance by measure­ment from the nearest convenient edge, or the tape m ay be adjusted to the path after it has been placed in approxim ate position.

KEEK : TANK STRAPPING. 115

An assum ption underlying th e m ethod is th a t the circumferences so measured are sufficiently circular to be corrected to true circles, and th a t the cross-sectional area of the tan k a t th a t level m ay be calculated from these. I f the tan k were tru ly circular, as m ight be the case w ith a welded tank , then the tape would he in contact w ith th e p late surface throughout its whole length. Any departure of the tape from the tan k surface is there­fore a warning th a t steps m ay have to be taken to apply an appropriate correction.

I f the tan k plates have been badly handled before erection, or if th e tank sides have been blown in by the wind during this, perm anent flats or deformations m ay be left in the tan k sides. Slight flattening is also usually noticeable above and below vertical seams, especially with half-plate staggering of these between courses. The form ation of a flat generally forces the p late on each side of th e flat to project a little outside the circum­ference of an undam aged plate. Relatively slight flattening is therefore sufficient to cause the tape to leave the tan k surface. The strapper should be required to verify th a t the tape m aintains norm al contact w ith the tan k over each circumference measured.

W hen the flattening is severe and a considerable length of tape comes out of contact, the levels affected m ust be considered as irregular. The m ethod of filling m ight be applied to obtain the table for the levels affected, or the effect of the deformation on the tan k capacity m ay be obtained by suitable survey. Such cases are relatively rare, since serious deform ation of the plate will generally result in erection difficulties. The question is discussed more fully below.

The tape m ay leave th e tan k surface through passing over rivet-heads. The level strapped m ay usually he chosen to avoid these supporting in ternal fittings, bu t i t is often im practicable to avoid those of doubly or trebly riveted vertical seams. Even w ith single riveting, vertical seam edges norm ally displace the tape. B utt-straps, and in some cases external fittings, cannot be avoided.

S t e p - O v e r s .

All these cases are m et by using the “ step-over,” a rigid frame supporting two scribing points. I ts shape and the gap between the points are so chosen th a t when one of these is applied to the tan k surface on one side of the obstacle, the other m ay be brought in contact on the other side. The points should touch th e plates well clear of any local distortions. Con­venience m ay dem and more th an one size of step-over.

For large obstacles, such as fittings, the level strapped is first m arked on both sides of the obstacle. A fine vertical line is then draw n a t right angles to th is tape pa th on one side, and the partia l circumference up to th a t line is measured with the tape. One of the scribing points is then placed on this “ zero ” line, and the other point made to scribe a second zero line a t the proper level on the other side. Strapping is continued on from this second line, the effective length of th e step-over gap being added in la ter to the parts of the circumference measured w ith the tape.

This effective length is not, of course, the linear distance between the scribing points, bu t the arc of the undisturbed course circumference con­tained between them. This varies with the radius of the course, most

116 K ERR : TANK STRAPPING.

rapidly if the tank is of small diameter. I f the linear distance between the points is measured j it is easy to calculate the effecti\ e length for any course radius. I f preferred, the value of the gap m ay be determ ined on the course itself. For this, the tape is stretched out on the course exactly as if a cir­cumference were being measured, though it is not necessary th a t the tape should encircle the tank completely. The step-over is then applied to the tape where it lies in good contact with the tank surface—th a t is, usually about the middle of a plate. The distance read off on the tape between the scribing points is then the effective length of the gap for th a t course.

I t is best to repeat this measurement a t several points around the tank circumference, and to take the average of the results as the value for the gap. This is esssential if the course is out of round—say, elliptical—in section.

Alternatively, the circumference m ay be strapped over the obstacles, as if they were not there. This is the more convenient m ethod for a series of small similar obstructions—say, vertical seams and their rivets, or hutt- straps. W ith the tape still in position under tension after strapping, the step-over is applied across the obstacles, and the apparent distance between the points is read off on the tape. The true value of the gap for the par­ticular circumference being known, the difference between this and the length read off on the tape is the effect of the obstacle on the circumference measured.

The tape will normally pass in the same way through all vertical seam or butt-strap rivet-heads on any one course. The effect of one obstacle will then be the same as tha t of another of the same type. I t is then unneces­sary to apply the step-over across each obstacle. A sufficient number of corrections are obtained to verify th a t the obstacles are really similar in effect, and to give a fair average figure for the correction. In correcting the circumference, this average is multiplied by the known number of obstacles. Since the average is to be multiplied, often by a considerable number, great care should be taken in the individual observations. The average may usefully be calculated a decimal place further th an is obtain­able with the individual measurements, to avoid arithm etical error from the subsequent multiplication.

B utt-strap corrections naturally vary with the size and thickness of the butt-strap. Even for vertical seam corrections, the average correction for one course should not be applied to correct circumferences on another course, unless the courses are similar and the tape passes in the same way through the rivet-heads.

When the types of tank calibrated are standard, it is possible to tabulate adequate corrections, best by averaging previous measurements made on similar tanks. This obviates much of the need for the step-over, especially on the higher parts of the tank, which are usually troublesome to reach. The width of the tape used in strapping m ust also have been standardized before such tables can be applied, since in general this w idth will affect the amount by which the tape is raised from the tank surface in passing through vertical seam or butt-strap rivet heads.

On tanks which are not standard, or not to the same standard, tape paths across vertical seams may vary, altering the value of the correction applic- a j c . ie edge of a seam normally projects above the cylindrical surface of

K ERR : TANK STRAPPING. 117

the tank , bu t seams sufficiently re-entrant to allow the tape to pass over them clear of all interference are occasionally m et with.

O v a l a n d C o n ic a l Co u r s e s .

W ithout causing the tape to leave the tan k surface, tank courses m ay depart from cylinders either horizontally or vertically. They m ay be elliptical in section, or otherwise out of round. These general deformations m ust be severe before they have any substantial effect on the tables for the tank. The circle is the figure of maximum area for a given circumference. I f therefore a course, originally circular in section, is slightly deformed in such a way th a t it still gives the same measured circumference, the resulting alteration to the cross-sectional area is necessarily of a t least th e second order. If, for example, a course originally circular is slightly deformed into an ellipse of the same circumference, the percentage correction to the calculated cross-sectional area of the tan k is approxim ately 9e4, where e is the eccentricity of the ellipse formed. The tank bottom and the heavy plates of lower courses tend to keep these levels circular, as also roof-girders the top courses. Tanks are m ost often out of round about mid-height.

W hat has ju st been said applies also to local deformation of plates. Slight local flattening of a plate throws its centre towards, and its borders ou t­wards from the tank centre. The real perim eter of the tan k is no t altered, and its cross-section a t th a t level only negligibly. I f the tape still m ain­tained contact w ith the tank surface, there would seldom be any need to correct the tables. The need for correction usually arises from the ta p e ’s leaving the tank surface, which, so to speak, gives an erroneous circumference.

I f a circumference on each course is called for, the flat m ay often be avoided by choosing a different level a t which to strap, relying on the proof just given th a t slight flats do not appreciably affect capacities. Special survey of the flattened area, or the use of the m ethod of filling, is seldom really justified.

I f tanks are strapped when they are full of oil or nearly so, the oil head tends to round out irregularities. As has been shown, this does not usually alter the tan k capacity, bu t i t does reduce the risk of troubles from local flattening, so th a t i t is advantageous to strap tanks when they are full.

“ Shingled ” tanks, or those in which the lower edge of each course lies outside the upper edge of the course below, are sometimes m et with. Such tanks m ay be telescopic in construction, or the general diam eter m ay be preserved by coning the courses.

I f the course is only slightly coned, then a circumference m ight be measured a t the middle of the courses. This would then be sufficiently close to the mean circumference of its course to be used in calculating the tank table. The author prefers, however, to retain the m ethod sug­gested above—th a t of measuring a circumference near th e bottom of each course, with one more near the tan k top. The circumference measured a t the bottom of any course is then greater th an the proper average for the course. A sufficiently correct average for the course is, however, obtained by calculating a circumference for the top of the course, w ith subsequent averaging of this with the m easured bottom circumference.

118 KERB : TANK STRAPPING.

Calculation of the required top circumference is usually no t difficult. The bottom circumference of the next higher plate is available. The chief cause of difference between this and the calculated circumference required is the thickness of the overlap between the two courses. I f therefore the measured circumference on the next higher plate is corrected for the effect of its plate thickness, the required second circumference is obtained. The labour consists in taking the appropriate correction from the tabulated effects of plate thickness mentioned below, subtraction of this from the upper plate circumference, with subsequent averaging as described.

Neither of the two procedures just discussed is theoretically quite correct. In forming the table, it is the true capacity of the course th a t is to be aimed at. The volume of the frustum of a cone is not obtained exactly by multi­plying the area of its mid-section by the height of the frustum . The correction is usually entirely negligible in the cases under consideration, but the method suggested allows a small and easily calculated correction to be applied, if this is thought necessary. The chief advantage of the suggested method, however, is th a t it allows a single standard strapping procedure to be adopted for any common form of tan k construction.

Where coning is intentional and severe, radically different methods of calibration would be adopted. Such cases arise w ith coned bottoms of agitators, and the like.

E x p a n s io n u n d e r O il H e a d .

When tanks are strapped full of oil, slight conicity of their courses is always present. The metal of the plates expands under the oil head, and this effect is greater a t the bottom of a course than a t the top. The procedure just given is sufficient to allow for this conicity, except some­times for the bottom course.

The bottom plates stiffen the resistance to expansion of the lower part of the bottom course. I t is usually impossible to detect any expansion under oil head near the bottom angle iron. Higher up the course the resistance lessens, until a t the top, if not before, the expansion under oil head is very closely th a t calculated from theory. The shape of the bottom course is then roughly th a t of an inverted conical frustum , bu t the sides of this frustum are slightly convex outwards. To allow properly for this effect in the tables, it may be necessary to measure a circumference near the middle of the course, in addition to th a t near the bottom . The bottom course is then treated in calculation as if it were two courses, each of slight conicity.

T il t .

If a tank is tilted from the vertical, then the oil sui'face is not circular, but elliptical. The capacity of the tank per un it depth, gauged in the usual way, is greater than if the tank were tru ly vertical. In almost all practical cases the effect is so slight th a t it m ay be neglected. I f a tilted tank were brought back to the vertical and then inclined to the other side, the surface area of the oil in it would pass through a minimum as the tank passes through the vertical. The effects of slight t ilt are, therefore, of the second order. I t may easily be shown th a t a tank m ust be tilted by 0-8

K ERR : TANK STRAPPING. 119

of a degree from the vertical before its volume per un it depth, taken vertically, is altered by 0 01 per cent. Long before such tilts are reached— say, by settlem ent of the tan k foundations—the tan k has usually been re-levelled by the engineers in charge. Especially if the tilted tan k stands among others which are vertical, the eye appears to be very sensitive to any lack of verticality. The strapper m ay be asked to report any visible tilt, bu t th is usually leads to more cases being reported th an really require correction.

P l a t e a n d P a in t T h ic k n e s s .

The capacity of a tan k is determ ined by its internal, not its external dimensions, and the la tte r m ust therefore be suitably corrected. The thickness of the plates and of the paint, if any, over which strappings have been taken m ust be allowed for. One m ay, if one chooses, calculate the external radius of each course, and sub tract the corresponding plate and pain t thicknesses. I t is, however, m athem atically equivalent, and in practice more convenient, to reduce external circumferences directly to internal by suitable allowances for plate and pain t thicknesses.

These two thicknesses are conveniently taken together. For every l/1 6 th of an inch in th is combined thickness, th e external circumference should be reduced by 0 0327 of a foot. I t is convenient to make a short table of multiples of this figure, from which m ay be taken directly the correction for any whole num ber of sixteenths of thickness. The corrections for l/32nd and l/64 th of an inch are proportional, and m ay be given separately, to be added in when necessary.

L ater on, when the fully corrected mean internal circumference of a course has been obtained, m ultiplication of the square of this figure ex­pressed in feet by the factor 0-0413063 gives directly the num ber of Im perial gallons per inch depth in the course. The result is then used in building up the table over the course concerned.

For new tanks, the plate thicknesses m ay be taken from the m akers’ drawings if the tolerances in thickness are low enough not to affect the accuracy required in the tables. The plates m ay otherwise be measured by suitable calipers before erection, or by weighing, using the known area of the plates and weight per cubic foot of their steel.

I f the tan k is already erected, the thicknesses given on the drawings m ay be checked by measuring the exposed edges of the plates. In th is it is often necessary to make allowance for th e thickening of th e edge which results from caulking. Experience helps in th is estim ate, bu t if drawings are available, i t is better to regard edge measurements as checking these, rather th an as measurements to be included in subsequent calculations. Drilling of plates to determ ine their thickness is seldom practicable.

The plates of tanks which have been in use m ay show corrosion of varying extent and character. The choice of a good average figure for the thickness of such plates is often a m atter of considerable difficulty. I f the tan k can be emptied, then the ex ten t of any general corrosion can usually be estim ated by examining the plates near internal fittings or seam overlaps. Taking the overlaps as example, the inner p late protects the outer p late from corrosion. I f the original jo in t between the two plates is located, the radial distance between th is and the general surface of th e outer plate near

120 KERB I TANK STRAPPING.

it is a fair measure of any general corrosion. Several measurements should be made and an average calculated.

If corrosion is of the pitted variety, it is often difficult to make more than an estimate of its effect on the average plate thickness. Magnetic or electrical plate thickness testers m ight perhaps be used to determine the residual plate thickness, but one usually contents oneself with estimates made with less elaborate equipment.

Paint thicknesses can usually be satisfactorily m easured by lifting a few flakes with a penknife from levels over which the tank has been strapped. These flakes are then held against the edge of a finely divided steel rule and their thickness measured. Soft paint, especially if thick, or paint containing many lumps—say, the bodies of insects caught in it when wet—is best cleaned from the tape paths before strapping is begun.

Lagged tanks are best strapped before the lagging is p u t on. I f these come° forward for calibration later, it is convenient to calculate their tables from the drawings, or, if this is insufficient, to calibrate them from internal measurements.

C o r r e c t i o n f o r O i l T e m p e r a t u r e .

As explained above, circumferences properly measured are those which would have been obtained if the tank and its contents were a t the tem­perature of calibration of the tape. W hen the tem perature a t which the tank is to be used differs from this, correction m ay have to be applied if the tables are to maintain high accuracy.

Heavy fuels are often kept artificially a t high and roughly constant temperatures all the year round. For other products the temperature of the oils stored varies with the season of the year, the range of variation depending on the local climate. I t is, however, hardly practicable to use more than one table for the same tank. In practice the tables may be corrected to the average temperature of the oil stored. t This average may be sufficiently estimated from records a t the same installation, or from records at other installations in similar climates.

Correction of measured circumferences for the linear expansion of the tape between its temperature of calibration and the average temperature of the oil would be sufficient for this purpose. In norm al gauging, however, the temperature of the dip-tape is also affected. This in tu rn affects gauges or dips obtained in routine, so th a t these m ay reasonably be corrected to the mean tank temperature also.

If the dip-tapes used are standardized to be correct a t the same temperature as the tape used in strapping, it is possible and convenient to combine these tv o corrections and insert them in the tank table. Strappings are then left uncorrected, but the volumes calculated from them are later corrected by means of the cubical coefficient of expansion of steel, using the appropriate temperature difference. The tank table then shows the volumes contained in the tank at its mean tem perature of use, and as measured by steel dip- tapes. In the method of calculation preferred by the author, this cor­rection is applied to the volume per unit depth calculated for each course, this being an equivalent and more convenient process.

K ERR : TANK STRAPPING. 121

C o r r e c t io n f o r E x p a n s i o n u n d e r O i l H e a d .

Tanks expand under the pressure of the oil contained in them . The effect is negligible in tanks of small diam eter, bu t it increases w ith the square of the diam eter. For large tanks, the increase in capacity on f i l l i n g

with oil, as compared with the volume when em pty, m ay require considera­tion. I f the tan k is not norm ally full on strapping, its to ta l capacity cal­culated from the strappings will be less th an its actual capacity for oil. If, before strapping, the tan k has been filled to its normal maximum content, then the resulting expansions are included in the circumferences measured.

Factors affecting the am ount of th is expansion are the radius or circum­ference of the tank, the gravity and depth of the oil above the level con­sidered, and the thickness and elasticity of the plate a t th a t level. The usual calculation for boiler expansion m ay be applied to obtain an exact formula. An approxim ate formula, which assumes fixed values for some of these factors, bu t which has m et the w riter’s needs, is given in Bell’s “ American Petroleum Refining,” 2nd edition, page 467. This gives the linear expansion in feet of the circumference as being the product of the square of the circumference in feet and the head of oil in feet, divided by three million times the plate thickness, expressed in sixteenths of an inch. In symbols :

E = 27(72/3,000,000f

Since the expansion is proportional to the head H of oil, the average expansion for each course will occur a t mid-level if the course is full of oil. Since for any tank of the types considered the circumference C is approxi­m ately the same for all courses, (7/3,000,000 m ay be w ritten as K , a constant, when the formula gives :

E /C = K H ¡t

E[C is the fractional increase in circumference, and if this is measured or calculated a t mid-level in a course full of oil, it is the mean fractional increase in circumference for th a t course. The mean fractional increase in area of the course is then sufficiently closely 227/(7, or 2K H /t. For a course completely filled with oil, therefore, the increase in volume due to oil head will be :

v = 2 K V H /t

where V is the approxim ate volume of the course, and H is measured upwards from mid-level in the course to the oil surface.

If the tan k courses are of equal height, their volumes will be approxim ately equal. I t is often sufficient in this calculation to take the average volume of all courses as the volume for any course, when 2 K V is a constant for the whole tank, and m ay be w ritten as R. The formula for any course then becomes

v = R H /t

For any particular course, the p late thickness t is constant. Taking R¡t, which is now constant, as Q, the formula reduces to

122 KERB, : TANK STRAPPING.

the suffixes being inserted as a reminder th a t all the quantities will varyfrom course to course.

Summarizing, the general constant for the tank , R = C F / l ,500,000 is first calculated. The course constants Qn are then calculated by dividing R by the known plate thicknesses for each course. W hen H n, which will vary from course to course, is known or assumed, m ultiplication of Qn by Hn for all the courses which are completely full of oil, and sum m ation of the resulting values of vn, gives the to tal expansion of the tank under the head of oil considered.

The units of volume in which the tank expansion is obtained are of course those in which V has been expressed. The units in which C, H and t are to be taken are given above. The calculation does not consider the expansion of courses which are not completely full, bu t tables may be satisfactorily corrected without this, as explained below.

I t is usual to strap tanks full, and to take this as correcting the tables sufficiently for the expansion of the tank under oil head. W hen relatively small quantities of oil are delivered from a large tan k which is nearly full, however, the removal of the oil reduces the oil head on all courses below. These lower courses contract as a result of this. An error in the quantity calculated as delivered therefore arises when the tank tables have been made on the assumption th a t the plates of all courses are always fully expanded.

Calculated as a percentage of the quantity delivered, the effect may be substantial, over 0-1 per cent. In the case just mentioned rather more oil is delivered than the usual tables indicate. Since the tables give the full capacity of the tank correctly, it follows th a t if relatively small quanti­ties of oil are delivered from the lower courses, the am ounts delivered are overestimated. Corresponding effects are shown when relatively small quantities are received into large tanks.

By calculating the to tal expansion of the tank when the oil surface is at the top of the top course, the top of the next course, and so on down the tank, the variations in capacity are obtained a t a sufficient num ber of levels to allow correction of the tables. The way in which this correction is most conveniently incorporated depends on the method adopted for calculation of the tables.

If the tank has been strapped full of oil, all circumferences will have been fully expanded. Course capacities calculated from these effectively contain the corrections applicable when the tank is completely full. Correction at any oil level then involves applying the difference between the correction already contained in the tables, and th a t properly applicable when the tank is filled only to the level under consideration.

If the tank has been strapped empty, the corrections are simply additive to the capacities calculated w ithout them. I t is useful to make the rule th a t a complete gauge of the oil in the tank is taken a t the time of strapping. Tanks may have to be strapped partly full, when the d a ta so given allow adequate correction of the resulting tables. W hether the tank has

een full, partly full, or em pty a t strapping, convenient methods for incorporating the appropriate corrections are not difficult to devise.

ver aps a t horizontal seams strengthen the tank against this expansion, an s are less expanded by oil pressure near these seams than above or e ow em. In those few cases which have been carefully examined by

K ERR : TAKE STRAPPING. 123

the author, the effect does not appear to extend more th an a few inches above or below the seam. This is w hat one would expect, since the over­laps are of small depth. The real effect on the tan k capacity is very slight, bu t the fact th a t it is possible to detect the effect offers another reason for not measuring circumferences too close to horizontal seams.

As a result of the additional resistance to expansion given by the bottom plates and angle iron, Bell’s formula is inapplicable near the bottom of the bottom course. Expansion under oil-head is usually not detectable a t the bottom of the bottom course, bu t full expansion in close agreement w ith the formula is usually obtained towards the top of this course. In calculating the expansion of the bottom course, a reasonable approxim ation is usually given by assuming th a t the bottom of the course does not expand, while the top is fully expanded.

D e a d w o o d .

Tanks usually contain “ deadwood,” or fittings which affect their capacity for oil. F ittings which affect the desired accuracy of the tables m ust be allowed for, a t the levels which they occupy. Swing-pipes are seldom allowed to project from the oil while measurement takes place, so th a t i t is sufficient to allow for these and their incidental fittings a t their lowest position. Manholes add to the tank capacity, and m ust be so trea ted in the calculations. No a ttem pt need be made here to give a complete list of possibilities : the same principle is applied to all cases. Calculation of the volumes of deadwood, though simple enough, takes time. I t is therefore best begun well in advance of the actual strapping. W hen standard fittings are used, results should be filed for ready reference in subsequent cases.

In ternal seam edges project inwards from the inner surface of the plates, and so form a type of deadwood. In ternal rivet-heads sim ila r ly displace oil. Sufficient check on such small deadwood should be kept, b u t i t will often be found, especially for large tanks, th a t these small corrections do not affect the required accuracy in the tables.

H a n d l in g o f C o r r e c t io n s .

For any previously chosen standard of uncertain ty in the final tables, i t will be found th a t certain of th e deadwood and other corrections given above are so large th a t their effects m ust certainly be included. Others will be as certainly negligible, whilst a th ird class is on th e border line between inclusion and exclusion. Care has to be taken with th is th ird class, since, although individually negligible, neglect of their to ta l m ay affect the accuracy desired.

T a n k -B o t t o m Ca l ib r a t io n .

Unless on adequate concrete foundations or iron-work supports, tank bottoms are seldom really flat. Tables are, however, frequently calculated as if the bottom s were flat, even when great irregularity is p resen t; th is on the grounds th a t these lower parts are seldom used in measuring either receipts or deliveries of oil. The stock of oil calculated as being in the

124 KERR : TANK STRAPPING.

tank will however contain an error owing to this neglect. W here this is important, it may be necessary to correct the tables for the irregularity of the bottom. The method of filling is usually used in such cases.

In the simplest method measured quantities of water are filled into the tank until all bottom irregularities are fully covered, when th e resulting depth of water is measured. This gauge or dip and the corresponding measured volume form the first entries in the tank table, no good measure­ments below this being possible. I f a t the same tim e the level of the water relative to the under side of the bottom angle iron, or to the first horizontal seam, is measured, the tank table may be completed to the top of the tank by the method of strapping. The volumes deduced from the strappings are added on to the first volume determined by filling.

A more thorough method is to fill the tank with water till the point on the bottom which will later be touched by the dip-weight in routine dipping is j ust wet, the volume of water necessary for this being measured. Measured volumes of water are then added, a water gauge or dip being taken after each addition, until bottom irregularities are fully covered. From the tabulated results a suitable table for the irregular part of the tank is readily formed.

Me n is c u s C o r r e c t io n .

One further correction may be included in forming the final tables. When the oil in a tank is dipped, it touches the steel tape in a meniscus, this reaching about 1 /12th inch above the true surface of the oil. Oil depths are therefore always read about 1 /12th inch too great. I t is how­ever easy to adjust the tables so th a t the true volume is given against the gauged depth of the oil.

A c k n o w l e d g m e n t .

This paper covers the results of work done for the Measurement and Loss Department of the Asiatic Petroleum Company, whose permission to publish the writer gratefully acknowledges.

125

ttúiítitji

T H E IN S T IT U T E O F PE T R O L E U M .Slli)®íe.; T h e One H undred and E ighty-Fifth Meeting of th e In s titu te was held

a t the Royal Society of A rts on Tuesday, Jan u ary 10th, 1939, following the Special General Meeting a t 5.30 p.m. The President, Lt.-Col. S. J . M. Auld, O.B.E., M.C., D.Sc., occupied the Chair. Mr. P. K err, M.A., B.Sc., A.I.C. (Member), presented the paper on T ank Strapping (pp. 109-124).

M e . P. K e k r , in presenting his paper, said th a t i t had been w ritten in response to a request made by Sub-Committee No. 10 (Measurement of Oil in Bulk) of the In stitu te th a t he should write ou t his views on tan k calibration, so th a t they m ight have a chopping-block, so to speak, on which to work out their own standard methods. He had therefore set out the theory on which his own views were based, ra th er th an th e practical methods by which the measurements necessary m ight be made.

In his own view, the most useful suggestion made in the paper was th a t of choosing in advance the percentage error, or ra th e r uncertain ty , perm iss­ible in the final tables. I f th a t was done, i t seemed to him th a t everything else followed. He could not be sure th a t he had seen everything w ritten on tank calibration, and he made no suggestion th a t the idea of adopting such a lim it was novel, bu t i t was original in the sense th a t he had th ough t of i t himself.

Some of the uses to which such a lim it m ight be p u t were as follows. The expected external circumference of any course m ight be calculated in advance from the drawings, and, in the light of th e accuracy required in the tables, upper and lower lim its also, w ithin which th e actual s trap ­pings m ust lie if they were to be in the required agreement w ith the draw ­ings. If, on strapping, the circumference found lay w ithin those limits, there was no need to repeat the m easurem ent: i t was already checked. Alternatively, if circumference measurements were to be checked by direct repetition, then, even if the duplicate figures differed slightly, fu rther checks were unnecessary, provided th a t bo th figures lay w ithin th e allowable range. Calculation of the allowable range was simpler in th e second application; th e check obtained did no t depend on th e drawings, b u t a little more work was throw n on th e strapping crew. The use of the limit, for including significant and rejecting insignificant deadwood or other corrections was obvious. Again, th e num ber of decimal places to which calculations should be taken was th a t which left the final tables free from arithm etical errors affecting the desired accuracy. In practice, for any particular tank, the accuracy aimed a t for its tables m ight no t be fully attained, or it m ight be slightly exceeded. I t was the order of accuracy, rather th an too rigid a figure, which he suggested should be chosen in advance.

There was one way in which he thought th e lim it should not be applied. I t was possible to calculate beforehand th a t m easurem ents to th e nearest half inch, say, were all th a t were necessary to secure the accuracy aimed at, and one m ight instruct the strappers to read their tape to th e nearest

1 2 6DISCUSSION ON “ TANK STRA PPIN G .”

half inch only. To read the tape as accurately as its graduations per­mitted, however, was really no more troublesom e; he felt one should not reject any additional accuracy so easily obtainable. I he strapping tape should always be read as accurately as possible, interpolation being made between its graduations. Even if it was found th a t the circumference figures were a little more accurate and the final tables a little more reliable than was a t first intended, surely no one would be disappointed.

There was, of course, a practical lim it to the accuracies obtainable. Independent checks suggested th a t, in normal cases and where strapping was a suitable method, it was easy to make sure of the tables w ithin 1 part in 1000, or 0-1 per cent. Particularly for larger tanks, accuracies of about 1 part in 10,000, or 0-01 per cent., were obtainable w ith a little more care and attention, but still w ithout serious practical difficulties. T hat was for the body of the ta n k ; stock figures m ight be less reliable through uncertainties in bottom calibration; but, as was well known, it was still usually possible to m aintain good accuracies in receipts and deliveries. He thought th a t higher accuracies were outside practical politics. The increase in practical difficulty between accuracies of 0-1 and 0-01 per cent, was not very great, but serious trouble was encountered if one tried to go much further. The list of corrections to which th e m ajor part of the paper was devoted was therefore limited to those which, either individually or in total, might affect the resulting tables to the ex ten t of 0-01 per cent. Within th a t limit, the list was intended to be exhaustive, and he would therefore very much appreciate the advice of the members if he had omitted any correction which should appear in the list.

The importance of any particular correction depended on circumstances, and notably on the local climate. He had tried to consider w hat might be necessary in any climate, and th a t was reflected in the paper. In any single case or single climate, the smaller corrections were often inherently negligible; occasionally one or two might cancel ou t against other corrections of opposite effect.

DISCUSSION ON “ TA N K ST R A P P IN G .”

Mb . J . K e w l e y said th a t he had worked in close con junction w ith Mr. K err for m any years, so th a t Mr. K err’s views were n a tu ra lly his views.

He would like, in the first place, to act as spokesm an for th e num erous people who were going to ask Mr. K err questions in thank ing h im for th e excellent paper he had w ritten. The whole subject of the m easurem ent of oil was one of very great importance. Vast quantities were involved. Very often in th e p a s t m easurem ent h ad been done in a slipshod fashion, standard instrum ents no t having been used. W hen one looked into the subject, one was surprised a t th e large num ber of p itfa lls th a t existed. For example, a tan k of oil when m easured em pty h ad a sm aller capacity th an when measured full, because of its expansion under th e heavy liquid. There were many points like th a t which were of im portance.

Ho thought th a t the subject of th e paper, which h ad arisen in connection w ith the work of Committee 10, of which Mr. K err was an activ e m em ber, was one which s ould be followed up and one which gave th e In s t i tu te an op p o rtu n ity of taking a definite lead, because, as far as he knew, very little had been published on the question of tan k m easurement and m easurem ent of oil generally. H o hoped th a t eventually

ommittee 10 and the Am erican Petro leum In s ti tu te , for exam ple, m ight come together and try to arrive a t some general agreem ent on m ethods of measurement.

o far success had not been achieved by th e a tte m p ts to o b ta in in ternational agree­m ent even on testing m ethods,

DISCUSSION ON “ TANK STRAPPING.” 127

W ith reference to th e s ta te m en t in th e second pa rag rap h of th e paper, “ I t is difficult to m easure th e effect of corrosion in th e p la tes of old tan k s w ithou t th is ,” i.e., w ithou t em pty ing th e ta n k com pletely, he h ad seen a t th e Physical E x h ib itio n tw o years ago a sm all a p p a ra tu s w ith w hich th e th ickness of a p la te of iron could be m easured by m eans of changes in m agnetic flux, an d he th o u g h t th a t possibly one could explore th e outside surface of a ta n k in th a t w ay an d get some idea of th e e x te n t to which corrosion h a d been tak in g place.

M e . K e r b said th ere were electrical in stru m en ts w hich enabled th e thickness of a ta n k p la te to be determ ined, b u t, so fa r as he h ad looked in to them , th e y scarcely seemed to offer th e accuracy th a t w as desired. The line along w hich he th ough t in te rn al corrosion could be m ost easily tack led— a line which he h ad n o t y e t h ad tim e to explore fully—w as as follows. T he expansion of a ta n k u nder oil head depended on th e th ickness of th e p la te ; if th e lower courses of a ta n k were s trap p ed w ith th e ta n k em p ty an d also w ith th e ta n k full, inversion of th e usual boiler expansion form ula enab led one to calculate th e ac tu a l p la te thickness.

M e . H . H y a m s , referring to th e s ta tem en t m ade b y Mr. K err in his rem arks th a t evening, th a t w ith his m ethod of ta n k s trap p in g he could ob ta in an accuracy of 1 p a r t in 10,000 in th e case of large tan k s , sa id i t would be in te resting to know w hat evidence Mr. K err could bring forw ard in support of th a t claim.

M e . K e r r said i t w as difficult to ob ta in independent m ethods of check which would verify accuracies of th e order in question . The first tr ia ls th a t he h ad m ade were to determ ine how closely he could rep ea t th e m easurem ents of th e circum ference of a tan k . A t th a t tim e he h a d a ta p e w hich w as only 50 f t. in leng th , an d he s trap p ed a ta n k th e circumference of which was a lit t le less th a n 250 ft. in five sections. H e set o u t five fine vertica l lines a round th e ta n k circum ference a n d m easured th e in te rvals betw een these, adding th é m easurem ents tog e th e r to g e t th e to ta l circum ference of th e tan k . T aking a new se t of lines, he rep ea ted th e process; th e m anager of th e installation , who w as in te res ted , rep ea ted th e m easurem ent tw ice independently , again w ith new zero lines. On th e to ta l circum ference th e g rea tes t difference betw een any p a ir of th e four resu lts which th e m anager an d he ob tained w as only 1/32 in. T h a t was perhaps acciden tal, as th ey were w orking u n d er very good conditions, strap p in g a level w hich w as easily accessible an d n o t a less accessible level h igher up th e tan k , b u t i t a t least show ed th a t th e fundam en ta l p a r t of th e m easurem ent, th e circum ference, could be tak en to a high degree of accuracy.

Again, he h a d s trap p ed tan k s in F rance, where th e original tab les were, of course, m ade by th e F ren ch B ureau of W eights and M easures, an d he h ad found th a t his resu lts u sually cam e w ith in 0-01 or 0-02 pe r cent, of th e figures which th e B ureau had ob tained . T h a t m igh t be challenged, because th e B ureau also calib ra ted th e tan k s by m easurem ent ; i t m igh t be held th a t th is p roved m erely th a t m ethods of strapp ing were self-consistent.

A t one in sta lla tio n w hich he h a d v isited , fuel oil deliveries were weighed in ra il cars, an d as com pared w ith th e am oun ts recorded as being tak e n ou t of th e storage tan k , th e in sta lla tio n w as gaining on its deliveries, th e gain shown being 0-35 per cent. On re-strapp ing th e ta n k he found th a t th e tab les under-estim ated its capacity by 0-33 p e r cen t. A gain th a t agreem ent w as perhaps a little accidental, because, in order to ob ta in i t , he h a d to see th a t th e weighbridge used by th e in sta lla tion was in nearly perfect o rder and, of course, th ere were alw ays difficulties in weighing rail wagons ; th e w ind h ad a n effect on th e ir weights, an d if th e w eather was w et th a t had an effect also. H e th o u g h t, however, th a t th e agreem ents which he h ad ju s t set ou t indicated th a t th e order of accuracy ob tainab le on large tan k s by s trap p in g was abou t 0-01 per cent.

D r . F. H . G a r n e r asked Mr. K err w h a t ap p aren t change in volum e w as obtained if a m easured volum e of oil in a ta n k ca lib ra ted b y his m ethods w as tran sferred to ano ther ca lib ra ted ta n k ?

M r . K e r r said he h ad n o t carried o u t m any very close transfers of oil. The question of errors in gauges an d errors in tem p era tu re was so troublesom e th a t m easurem ents of th a t so rt w ould have to be rep ea ted several tim es w ith th e b est a tta in ab le p re ­

128 DISCUSSION ON “ TANK STR A PPIN G .”

cision before a check was obtained on a m ethod w hich seem ed to give an accuracy of about 0 01 per cent. Fuel oil would have to be used, so th a t th ere would be no loss by evaporation; otherwise there would be a difficult correction to be m ade for the am ount of vapour lost in transfer. H e had no sufficiently accu rate se t of figures for such transfers; th ey did n o t prom ise enough in th e w ay of checking.

Mb W. F. J e l f f s said he believed it w as n o t th e general custom to app ly cor­rections for expansion under oil-head, an d he w ould like to know w hether Mr. K err considered th a t th a t should be th e practice in fu tu re .

M b . K e r b said it would be seen th a t th e p ap er d id n o t suggest any particular standard of accuracy which should be ad o p ted in th e calib ra tion of tankage. The need for the correction of tan k tab les for th e expansion of th e ta n k under oil-head depended on the standard of accuracy th a t one w ished to a tta in in th e tab les them ­selves. I t was only for larger tan k s th a t th is correction w as of m uch importance, but, as he had explained in th e paper, i t was of ra th e r m ore im portance th a n it appeared to be a t a first glance, since when a sm all q u a n tity of oil w as delivered from th e upper levels of a large tank , con traction of th e p la tes took place on all courses below the level from which delivery h ad been m ade. As a resu lt of neglecting th e expansion correction, there m ight th en be an unexpected ly serious percen tage error in the quantity calculated as delivered or received. H e w ould like to see as good accuracy as was commercially possible in oil m easurem ent, an d therefore he favoured the insertion of an expansion correction in all cases where th a t seem ed to be justified.

D r. A. I. V. U n d e r w o o d asked w hether in th e form ula given in th e paper for expansion under oil-head— i.e., E = H C 2/3,000,000i— it w as th e d ep th of oil th a t was referred to or th e pressure head of th e oil. The la t te r w ould v a ry w ith th e specific gravity of the o il; th e correction required would be different, according to whether the tan k was used for petro l or for heavy fuel oil. T he effect of th e correction for expansion under oil-head was given by Mr. K err as being approx im ate ly 0-1 per cent, in certain circumstances. If one took th e difference in th e specific g rav ity of petrol and heavy fuel oil—a difference of roughly 25 per cen t.—th e erro r due to th a t difference in specific gravity would be 1 in 4000. T h a t would be appreciab ly larger th an the error of 1 in 10,000 which Mr. K err regarded as th e ideal.

M r . K e r r said th a t th e s ta tem en t in th e paper : “ C alculated a s a percentage of the quan tity delivered, th e effect m ay be su b stan tia l, over 0 1 per cen t.,” applied to relatively small deliveries from th e tan k . T he to ta l correction, considered as a percentage of the tan k contents, reached 0 05 or 0 07 pe r cen t, on large tan k s only, so th a t the variation was a little less th a n Dr. U nderw ood h ad ju s t suggested.

He was glad th a t a tten tio n had been directed to th e p o in t in question, because he had already been to ld th a t he ought to apologize for inflicting e lem entary mathematics a t such great length on so distinguished a body as th e In s t i tu te of Petroleum . The discussion on the correction for expansion under oil-head w as very s im p le indeed, and it could have been condensed, b u t he h ad a certa in reason for se tting it out at some length. I t was clear th a t i t was n o t every ta n k th a t could be trea ted in the way described. Some tan k s h ad courses which were n o t equal in height, and the user of the formula m ight dep art from th e calculations given in th e paper a t any particular stage of the discussion. H e had therefore p u t in each stage in full, so that anyone who wished to calculate th e expansion m ore accu rate ly th a n th e approximate formula allowed could easily see th e po in t a t which he m u st d ep art from th e procedure suggested in the paper.

He accepted Dr. Underwood’s com m ent in principle. I t w as tru e th a t th e formula in the paper was approxim ate o n ly ; i t was derived from th e exac t form ula by taking 0-750 as a fixed gravity for the oil, a m ean value being also tak e n for Y oung’s Modulus. The formula given did n o t balance in d im ensions; th e co n stan t 3,000,000 necessarily contained factors which had dimensions.

As suggested by Dr. Underwood th e accuracy of th e form ula given would be improved by taking in its place

E = HC'-g12,250,000twhere the new factor g is the specific g rav ity of th e oil, th e o th e r quantities being as given m the paper.

DISCUSSION ON “ TANK STR A PPIN G .” 129

T he general form ula isE = p C 2l2ntM

where E is th e expansion of th e circum ference, p th e pressure due to oil-head, t th e p la te th ickness, an d M Y oung’s M odulus for th e steel p lates, a t th e level concerned.

E ith e r of these m ay obviously be applied as suggested in th e paper.

D r . M. B . B l a c k l e r co n gratu la ted Mr. K err on v en tila ting a sub ject which w as of un d o u b ted im portance to th e oil in d u stry in general. H e did n o t believe th a t ques­tions of head, expansions an d contractions of ta n k m ateria ls seriously affected com ­m ercial m easurem ents. T anks as m anufactu red to -day , w hen tran sp o rted and erected u nder reasonable conditions, could be calib ra ted w ith in th e lim its of com ­m ercial accuracy from th e m akers’ draw ings, b u t if erected ab ro ad a fte r th e p lates had been tran sp o rted long distances b y sea, ra il and road , th en calib ra tion b y d irect m easurem ent w as essential. I n such cases he h ad adop ted th e following p lan in order to ob ta in reliable calib ra tion figures. F irs t levels were tak e n over th e b o tto m of th e tan k , an d th e general configuration w as d raw n o u t so th a t th e calib ra tion of th e first foot of height could be determ ined w ith reasonable accuracy. The d iam eter of each s trak e of th e ta n k was m easured a t th e to p , m iddle and b o ttom , and these d iam eters were m easured a t angles of 45° round th e tan k , so th a t actu a lly tw elve d iam eters were m easured for each s trak e an d th e average d iam eter w as tak e n as th e accep ted d iam eter of th e strak e . The calculations were m ade in litres p er centim etre, and as a m a tte r of rou tine , allowances were m ade for volum es of riv e t heads, angles, ladders, e tc ., in th e calib ra tion volum es per centim etre.

The w ork w as included in th e drawing-office rou tine , an d w hen com pleted th e calibration tab les were d is trib u ted from th a t D epartm en t.

D r. B lackler d irected a tte n tio n to calib ra tion difficulties in those cases w here th e ta n k b o tto m m oved under vary ing oil-head, m entioning a case in his experience in w hich th e g round under a ta n k was found to be sa tu ra ted w ith oil which h ad leaked th ro u g h th e ta n k b o tto m as a resu lt of th is m ovem ent.

Mr . K e r r said he qu ite agreed th a t th e calib ra tion of a ta n k b o tto m w as difficult if th e foundations were n o t sound. M ovem ent d id occur in p a rticu la r cases, as in th a t m entioned b y D r. B lackler, where i t h ad led to leakage. T he suggestion th a t m ovem ent was tak in g place was, however, som etim es m ade w ithou t very clear proof, often as a ra th e r speculative exp lanation of discrepancies in m easu rem en t; in these cases one should estim ate probabilities from th e n a tu re of th e ground an d th e founda­tions. I n doub tfu l cases i t was well to m ake sure, for m ore th a n m easurem ent was invo lved ; as D r. B la d d e r’s rem arks im plied, th e ta n k should th en be w atched for possible leakage. To m eet b o th leakage an d m easurem ent, he suggested p u ttin g a tem p o rary w a ter b o tto m in th e tan k , sufficient to cover th e b o tto m com pletely, and determ in ing b y dipping w hether th e level of th e w a ter surface changed w ith th e oil- head.

This could be done w ithou t tak in g th e ta n k o u t of use, an d w ithou t m uch trouble , since rou tine w a ter dips tak e n from day to d ay w ould usually give th e inform ation req u ired . I f th e ta n k b o tto m h ad a lready begun to leak, th e expected re la tion betw een th e w a ter dips and th e oil-head m ight be com plicated b y ac tu a l loss of w ater, b u t by prolonging th e tria ls u n til th e ta n k h a d been filled and em ptied tw ice or th rice, th e po in t w ould be cleared up . I f m ovem ent were confirm ed, th e w a ter b o tto m m ight be left in th e tan k , w hen gradual d isappearance of th e w a ter would show th a t leakage h a d beg u n ; m easurem ent errors would be reduced also. I f th e w a ter p u t in was sufficient to keep its surface above th e b o tto m p la tes even w hen th e ta n k w as full of oil, th en it would be possible to calcu late from th e w a ter dips th e volum e added to th e ta n k th rough depression of th e b o ttom , an d so estim a te th e seriousness of th e m ovem ent tak in g p la c e ; th e figures ob tained should be regarded as a m inim um , since th e p o in t a t which th e w ater-finder rested on th e ta n k b o tto m m igh t itself m ove dow nw ards w ith increasing head of o il ; b u t th is m ovem ent could in tu rn be m easured an d allowed for b y recording th e vary ing heigh ts from th e edge of th e dip-hole to th e p o in t in question.

As regards D r. B lacker’s m ethod of ta n k calib ra tion , he agreed th a t tan k s might, be sa tisfac to rily ca lib ra ted in various w ays. C alib ra tion of ta n k b o tto m s b y survey m ethods led easily to a figure for th e to ta l con ten t of th e ta n k b o tto m to some chosen

130 DISCUSSION ON “ TANK STR A PPIN G .”

depth of liquid • from such d a ta i t was possible, b u t perhaps less easy, to form a table allowing measurement in the tan k b o ttom itself of sm all q u an titie s of w ater, say those accidentally delivered w ith a cargo. The m ethods of filling ou tlm ed in his paper were useful when bottom calibration had been o m itted w hen th e ta n k was first erected, but was later found desirable ; for it was generally possible to pum p in m easured quantities of w ater under th e oil, and to m easure th e resu lting d ep ths w ith water- finding paper or paste, w ithout tak ing th e ta n k o u t of use.

That the expansion of tanks under oil-head could alw ays be neglected was, of course, the current opinion. His own view was th a t th e need for th is correction varied w ith circumstances; it was only necessary w ith large tan k s . W hen even a large tan k was gauged only when it was nearly full or nearly em p ty , as m igh t be th e case with tanks used only for cargo shipm ents, th en th e effect was often sufficiently allowed for by strapping the tan k when it was full, w ithou t fu rth e r a d ju s tm e n t of th e tables. When small deliveries were m ade o u t of large tan k s , th e effect w as m ore serious, as the paper suggested, and as could readily be calcu la ted . H is own view was th a t in these cases a t least, a correction so readily calcu lated an d so easily incorporated once for all in the tan k table should scarcely be om itted .

Mb . A . P. C a t h e r a l l said he would like to ask Mr. K err w h a t his experience was with regard to the compensation of th e various errors th a t en te red in to th e measure­ment of oil. The calibration of a ta n k was only a p a r t of th e m easurem ent of the oil ; there were irregularities in tan k dippings, tem p era tu re m easurem ents, and so on. His own experience was th a t there was a num ber of such errors which generally cancelled themselves out, so th a t, even w ith qu ite casual ta n k strapp ing , and even in taking the m easurements from th e blue p rin t, prov ided th a t th ey were checked with a t least one strapping, one could ob tain qu ite accu rate resu lts w ithou t extremely accurate tan k strapping.

M r . K e r r agreed w ith Mr. C atherall in th ink ing th a t strap p in g was often unnecessary if the tan k was well made and th e p la tes were undam aged. Official regulations, however, often required the tables to be m ade from d irect m easurem ents, so th a t one was compelled to strap more frequently th a n was necessary on purely technical grounds. Calculations from drawings were usefully checked b y strapp ing a single circumference, as Mr. Catherall had suggested ; th a t th is was his view would be seen in the paper.

In strapping, the length of tape required to encircle th e ta n k was a m inim um when the tape followed its proper p a th ; it followed th a t sm all d epartu res of th e tape from its proper path could only have second-order effects on th e ta n k tables, and th is was thought to be a reason why Mr. Catherall had obtained good resu lts w ith o u t extrem e care. I t also followed th a t if care had been tak en w ith th e tap e tension, there was some­thing to be said for taking the sm allest figure obtained for any given circumference, rather than the average of repeat figures, in calculating th e ta b le ; b u t if tolerances had been calculated as suggested in th e paper, and all rep ea ts fell w ith in these, their average, or indeed any of the figures, could be tak en in to calcula tion w ithout affecting the accuracy desired.

\V hen discrepancies in m easurem ent were troublesom e, it was som etim es necessary to find out how much of these should be a ttr ib u te d to errors in th e ta n k tab les; here the abbreviated m ethod suggested in th e paper under “ R ecalib ra tion ” was often sufficient.

To discuss the question of compensation of errors in oil m easurem ent fully would lead too far from the subject of the paper, b u t Mr. C atherall’s experience on this question was also his own, and it was in line w ith theory . H e h ad found th a t errors in 01 measurement followed the usual m athem atical theo ry of errors ; given a sufficient num ber of cases, it was usually possible to separate th e effect of system atic errors iom a o ortuitous errors. The la tte r were usually d is trib u ted norm ally ; in the ormal distribution small errors were m uch m ore frequent th a n large, in line with

f™™ a ..leia s exPenence. His view was th a t system atic errors should be eliminated he tnl°' ,ncjahUremellt as carefully as possible; m uch larger fo rtu itous errors could ont » f m i t’- 8mue - I 7 oeeurred only occasionally, and since th ey tended to cancel

“ 6d V paper under “ Prelim inaries.” E rro rs in ta n k tables tended to be system atic in their effect.

DISCUSSION ON “ TANK STR A PPIN G .” 131

M r . H . H y a m s said th a t Mr. K e rr ind ica ted in his pap er th e correction to be applied for th e difference betw een th e calib ra tion tem p era tu re of th e tap e and th e average tem p era tu re of th e oil th a t was going to be s to red in th e tan k . H e q u ite agreed th a t such a correction was necessary an d should be incorpora ted in th e ta n k tab les, b u t he th o u g h t i t m ight be borne in m ind, in connection w ith m easurem ent on a weight basis, th a t hydrom eters in th e oil in d u stry were usually ca lib ra ted a t 60° F . or 15° C., and th ey were frequen tly used in trop ical clim ates or in connection w ith artificially heated oils w ith o u t any correction a t all being m ade for th e cubical expansion of th e glass. T he e rro r w hich would resu lt from neglecting th a t correction was, in th e case of an oil whose specific g ra v ity was determ ined a t, say, 90° F . ab o u t equal to th e com bined corrections th a t Mr. K e rr said were necessary u nder th e heads of expansion of th e ta n k and also expansion of th e tap e . H e saw no w ay of m eeting th a t con­tingency o th er th a n th a t th e hydrom eter should itself be ca lib ra ted a t a tem p era tu re corresponding to th e oil w ith w hich i t w as to be used, or, if th e s tan d ard hydrom eter ca lib ra ted a t 60° F . w as used, i t seem ed to h im th a t , in those cases in which th e oil was a t a tem p era tu re su b stan tia lly above 60° F ., th e ap p ro p ria te correction for cubical expansion of th e glass should be m ad e ; otherw ise th e correction which Mr. Ken- in d ica ted w as necessary would be nullified.

T he im portance of correct calib ra tion of th e ta n k b o ttom , w here irregularities are know n to exist, has received due em phasis in th e paper. The provision of a perm anent w a ter b o tto m is a rem edy, b u t th ere a re m an y countries w here very cold w in ter conditions m ake such a w a ter b o tto m undesirable or where w ater bo ttom s are, in any case, p roh ib ited by local au tho rities. I n such cases th e correct m easurem ent of oil q u an titie s in tan k s w ith deform ed or w eak ta n k b o tto m s becom es a real difficulty, and th is difficulty can be overcom e only by careful calib ration . T he shape of th e b o tto m and its cubical con ten ts will possibly v a ry w ith th e head of o i l ; i t has been suggested th a t w ith certa in ty p es of ta n k s th e b o tto m ac ts as a d iaphragm , a lthough conclusive evidence has n o t been forthcom ing to sup p o rt th is con ten tion . B u t in such instances i t w ould be undesirable to co n stru ct tab les for th e b o tto m levels from th e ta n k draw ings. Mr. K e rr’s m ethod of b o tto m calib ra tion b y w a ter could be followed w hen th e new ta n k is sub m itted to its w a ter te s t.

T here was one fu rth e r question th a t he would like to ask. Mr. K err said th a t i t w as desirable to s tra p tan k s w hen th e tem p era tu re qf th e a ir and th e tem p era tu re of th e con ten ts of th e ta n k were ap prox im ate ly th e sam e. I t could n o t be sim ple to ob ta in a predeterm ined a ir tem pera tu re , an d he would like to know how th a t difficulty was overcom e b y Mr. K err.

M r . K e r r said th a t Mr. H yam s h ad raised an im p o rtan t question in his first poin t, w hether m easurem ent should be considered as a whole o r sectioned off in to questions of ta n k calibration , dipping, tak in g specific gravities, an d so on. H e th o u g h t th a t ta n k calib ra tion could perhaps be m ore fairly re legated to its own d ep artm en t th a n th e re s t of th e rou tine process of gauging. I t was tru e th a t one m ight offset th e effect of th e neglect of th e expansion of glass correction for th e h y drom eter against tan k -tab le corrections, b u t Mr. H yam s h ad him self suggested a b e tte r m ethod . If tan k s were co rrectly calibra ted , one should n o t re ly on com pensation of errors, b u t should ra th e r correct th e hydrom eter for its errors w hen used a t a high tem p era tu re or, a lte rn a tiv e ly , for high tem pera tu res supply a h y drom eter which requ ired no cor­rection . G enerally speaking; he was opposed to allowing errors to com pensate each o ther, unless one was qu ite certain th a t com pensation would be reasonably perfect. A gain, a lth o u g h in some countries m easurem ent was b y w eight, in o thers i t w as by volum e, w hen th e com pensation proposed w ould fail. I t was possible to have one m ethod of ta n k calib ra tion for areas where m easurem ent was by w eight an d an o th er m eth o d for areas where m easurem ent w as b y volum e, b u t i t w as ra th e r doub tfu l w he ther it w as advisable to duplicate m ethods in th a t w ay for th e sake of com pensation of errors.

T he la s t p o in t raised by Mr. H yam s was a v ery difficult one, and one on which he would a d m it th a t he w as n o t him self q u ite satisfied. I t w as v ery difficult to arrange th a t th e a ir, oil, and tap e tem p era tu res should all be th e sam e. The linear expansion of steel w as a b o u t 0 000012 p er degree C., so th a t if th e tap e tem p era tu re w as 8° C. aw ay from th e tem p era tu re of calib ra tion of th e tap e—th e tem p era tu re a t which he assum ed th e tap e was correct,— an error w ould be in troduced of ab ou t 1 p a r t in 10,000

132 DISCUSSION ON “ TANK STR A PPIN G .”

in the circumference measured. An error of th a t am oun t in th e circum ference corre­sponded to an error of 2 p a rts in 10,000 on th e cross-sectional a re a ; in order to main- tain an accuracy of 1 p a rt in 10,000 in th e ta n k tab les, one w ould have to be sure th a t the tape was within 4° C. of its correct tem p era tu re , or else m easure in some way the tem perature a t which the tape was being used. N either of those w as very easy. The simplest course to adopt, of course, was to reduce th e accuracy of th e tab les, to choose such a tolerance as would cover th e errors th a t were m ade w ith th e tape. An alternative m ethod was to su b stitu te a tap e of low coefficient of expansion for the normal steel tape used. I t was tru e th a t in th e pap er he h ad expressed himself as being opposed to the use of such special tapes, b u t in th e case in question one was confronted by a special difficulty and had to choose betw een tw o undesirable courses.

The tape tem perature was im portan t m the tropics. I here, if m full sunshine a tape was pu t round a tan k which was p a in ted or h ad a p itte d surface or was ru sty , when one drew one’s finger along th e tap e th e sensation was n o t m erely one of w arm th, but of shght pain. The tapes m ight reach tem pera tu res of 140° F . or m ore under these conditions, which was quite a departu re from th e 68° F . a t which th ey were normally calibrated. He hoped th a t w ith fu rth er work some m ethod m igh t be found of cor­recting for the m ajor portion of th e error in troduced by failure to contro l th e tem pera­ture of the tape. As an illustra tion of w hat m ight be done, one m igh t be able to hang a therm om eter of some stan d ard type on th e sunny side of th e tan k and, by strapping a few tanks bo th w ith an ordinary steel tap e and w ith a n inv ar tape, obtain a satisfactory table of corrections, to be applied in fu tu re eases against th e readings of a similar therm om eter similarly placed during strapp ing .

E xcept for tanks heated artificially, in all clim ates th ere w as alw ays a tim e in the morning, w ith another in the evening, a t which th e air, oil an d tap e tem peratures would be the sam e; b u t i t was extrem ely inconvenient, an d often im practicable, to have to lim it strapping to these tim es. He agreed th a t th e p o in t was a difficult one, not yet fully solved.

He agreed th a t insertion of perm anent w ater bo ttom s would be im practicable in very cold clim ates; in such cases check on th e stab ility of th e ta n k bo ttom might still be possible in summer. The fact th a t m ovem ent of th e b o tto m was a p t to start leakage, combined w ith the impossibility of m ain tain ing all th e year round a water bottom against leakage, would no doubt lead engineers to m ake very sure of tank foundations under such circumstances. W hile he h ad n o t h ad to handle the specific case of inserting a tem porary w ater bo ttom when local regulations did no t allow these, he had seldom found any real difficulty in obtain ing tem porary re laxation of official regulations, of course w ithin reason, a fte r th e need for an d in ten tio n of the work proposed had been suitably explained.

M r . W . E . D o u g l a s said one point to which Mr. K err h ad n o t referred in th is paper was th a t there were lim itations w ith regard to tan k -calib ra tio n tab les. There were certain persons who were responsible for those lim ita tions. I n th e first place, the tank-calibration table was lim ited according to th e lim ita tions of th e person or persons making the measurements. I t was unfo rtunate th a t th e people who had to calibrate tanks had not always the fine, scientific, accurate m ind of Mr. K err, and it was very difficult to get them to realize th e im portance of some of th e finer po in ts. One might possibly produce more accurate results by n o t aim ing a t such a h igh degree of accuracy. But the lim itation which still more affected th e tab les w as th e lim ita tion of the people using them . If the tables were too com plicated or requ ired too m uch judgem ent on the p a rt of the people using them , it would invariably be found th a t th e tables were wrongly applied. One of the greatest difficulties in a tta in in g a high degree of accuracy was t e fact th a t the m ental calibre of some of th e people using th em was no t sufficiently good, and th a t, of course, was particu larly th e case in places fa r overseas. He there- . ore t oug t it was advisable no t to aim a t too high a degree of accuracy if that involved m aking the tables them selves too com plicated.

M r . K e r r said th a t if one had not a t one’s disposal s ta ff who were com petent to +a an ProPerly > then the results would be unreliable. To send ou t a special

riI, , • ° remo e districts m ight be out of th e question , on account of th e cost. In one mig i ia i e to adm it th a t in such cases one could no t reach satisfactory

DISCUSSION ON “ TANK STR A PPIN G .” 133

accuracies b y s trapp ing . I n these cases calcula tion from th e draw ings would alm ost certain ly be m ore sa tisfac to ry th a n s trapp ing .

H e was en tire ly in agreem ent w ith Mr. D ouglas as regards ta n k tab les. These s ta te against th e dips th e calcu lated volum e of oil in th e ta n k ; in his opinion, th e volum e figures given should a lready con ta in all corrections considered desirable. Fo r th e ty p e of ta n k considered in th e paper, an y desired correction could be incor­po ra ted in calcu lation in to th e volum es given, th is leaving one com pletely free to se t ou t th e tab le in any desired fo rm ; he agreed w ith Mr. D ouglas th a t th e user should be given as sim ple a tab le as possible.

M r . J . L e w i s said th a t Mr. K err s ta te d in his pap er th a t th e strap p in g tap e should n o t come w ith in four inches of a seam . H e w ould like to ask Mr. K err how he arranged for th a t , because in h is own experience of s trap p in g it h ad been difficult enough to keep th e tap e on to p of th e p la te in its correct position.

M r . K e r r said th a t a lthough he h ad n o t m et th e troub le , i t m igh t be difficult to m ain ta in a tap e in correct position on a tan k , b u t if th e circum ference was to be of value, i t h ad to be done. T he tap e w as p u t round th e ta n k in tension , an d it w as m oved in to p lace b y m eans of a soft m eta l ring which passed round th e tap e . A strin g was led from th e soft m eta l ring to th e to p of th e ta n k an d an o th er string was led to th e ground. Two m en a t th e ends of th e s tring were able to place th e tap e in a sa tisfac to ry position . T en pounds tension on th e tap e was generally enough to hold i t in position , b u t if th is difficulty w as experienced w ith any p a rticu la r tap e , he w ould increase th e tension u n til he go t th e tap e to keep in place. A correction for th e add itional tension m ight th en have to be applied, b u t since th e sam e tap e would p ro bab ly have to be used on o th er tan k s , th is would be calcu lated as a percentage correction once for all.

U sing a tap e \ in . w ide he h ad n o t m et serious difficulty in keeping i t to th e p roper p a th , even w hen th e tw o m en assisting him were na tives new to th e w ork. T he w ider th e tap e th e m ore easily i t would keep to i ts p a th , once i t h ad been placed in t h i s ; th e so lu tion m igh t be found in using a w ider tap e , b u t on th is see th e section on “ S trapp ing Tapes ” in th e paper.

M r . J . L e w i s said his experience was th a t as soon as one increased th e tension, th e m an a t th e o ther end of th e tap e invariab ly le t i t go. H e a tte m p ted to keep th e tap e some d istance above th e horizontal seam, b u t he found th a t i t cam e dow n on th e horizon tal seam , and he h ad to go round th e tan k s him self to see th a t th e tap e was actu ally in position. H e gave up a ll hope of ever f i n d i n g th e tap e 4 in. above th e seam.

M r . K e r r said th a t in th e m ethod which he preferred th e tap e was long enough to encircle th e ta n k com pletely. One end of th e tap e w as he ld th ro u g h a spring balance a t th e desired tension, and th e o th er end w as ju s t held in th e hand . The friction necessary to pull th e tap e o u t of position was ab o u t 18 lb ., on a ta n k of fair size, so th a t a t 10 lb . one still h ad a m argin betw een th e am oun t necessary to pu ll th e tap e o u t of position and th a t w hich gave th e requ ired tension. H e h a d n o t found th e difficulty in ac tu a l strapp ing , b u t i t no d o u b t depended to a considerable ex te n t on th e q u a lity of th e assistance th a t one had.

A strap p in g crew m u st, of course, include a t least one m an who understood tho rough ly w h a t w as to be done; i t was useful if th e second m an also understood, since th is saved t im e ; b u t he h ad frequen tly h ad to tra in as second h an d a reasonably b u t n o t unusua lly intelligent n a tive leading-hand or forem an new to th e work, th is add ing perhaps ha lf a n hour to th e tim e tak en . T he th ird m an need have ve ry few qualifications beyond being willing to do as he is t o ld ; b u t th is is essential.

T he use of a long tap e p u ts b o th ends of th is in th e hands of one m an, who should be th e b est qualified of th e th ree . The tap e is k e p t a t th e requ ired tension b y th e leader while i t is being finally p laced by th e tw o o thers. I f i t is desirable th a t th e leader should inspect th e placing of th e tap e , and if th e th ird h an d is n o t to be tru s te d to keep it a t p roper tension, th e following procedure h as been found useful. T he ends of th e tap e a lread y o v e rla p ; still u n d er th e p roper tension, th e overlap is ex tended , so th a t a fa ir leng th of th e one end lies closely above th e o th er on th e ta n k surface.

134 DISCUSSION ON “ TANK STR A PPIN G .”

The th ird m an is n o t allow ed to ta k e th e ta p e in h is h a n d s , b u t h o ld s i t in p o sitio n bv Dressing his h ands fla t ag a in s t th e t a n k o v er th e tw o en d s . M a rk s h a v e a lre a d y been m ade against g rad u a tio n s on b o th ends of th e ta p e , a n d i t is p o in te d o u t to him th a t th e ta p e m u st s till be a t th ese m a rk s w hen th e le a d e r com es b a c k . T h e p osition of th e tap e is th e n in spected b y th e lead er, a n y fa u lts b e in g p o in te d o u t to th e second m an The leader again ta k e s over th e ta p e fro m th e th i rd m a n ; be fo re th e th ird m an rem oves h is h ands , th e p ro p e r ten s io n is re s to re d o n th e free e n d th ro u g h th e spring balance. T he ta p e is th e n re a d ju s te d if n ecessa ry b y th e seco n d a n d th ird hands, and its p lac ing ag a in verified ; if th e p ro cess is in s is te d on ca re fu lly a t the beginning, i t is seldom n ecessa ry to c a r ry i t o u t o n a l l co u rses . I f th e to lerances w ith in w hich th e m easu red circum ferences m u s t lie in o rd e r to ag ree sa tis fac to rily w ith th e draw ings h av e been c a lcu la ted in ad v a n c e , th e n a s th e le a d e r com es to each new circum ference he know s a t once w h e th e r f u r th e r in sp e c tio n of th e ta p e p a th is desirable.

M r. A. P . C a t h e r a i a sa id he h a d s tr a p p e d a co n s id e rab le n u m b e r of ta n k s w ith 500-ft. tap e , I in . w ide, a n d using co lou red la b o u r, he h a d h a d n o d ifficu lty a t all in keeping th e tap e on th e ta n k a t th e re q u ire d te n s io n b y th e m e th o d suggested by Mr. K err.

M r . I v e r r sa id h is difficulty g enera lly h a s b een to p re v e n t th e m e n p u lling too heavily on th e ends of th e ta p e . H e h a d seen th e m p u llin g a t th e en d s of th e tape as though th e y were engaged in a tug -o f-w ar, an d t h a t h a d , of cou rse , to b e stopped. As he h ad said, a g rea t dea l dep en d ed on th e q u a li ty of th e a s s is ta n c e av a ilab le .

H e th o u g h t i t m ig h t be use fu l to e x p la in m ore fu lly th e tro u b le s w h ich arose from accep ting too low a s ta n d a rd of acc u ra c y in ta n k ta b le s . A n y fa u lts in ta b le s which tended to short-delivery to cu stom ers w ou ld n o t b e to le ra b le to a n y re p u ta b le Com pany. F au lts in strap p in g delivery ta n k s cou ld be d iv id ed o n th is b a s is ; s lack n ess in the tap e , neglect of seam overlap co rrec tions, a n d th e like , te n d e d to sho rt-delivery ; excess tension in th e ta p e , if n o t co rre c ted fo r, te n d e d in th e o th e r d ire c tio n ; while there were o th e r fau lts w hose effect v a r ie d acco rd in g to c irc u m sta n c e s . A nother consideration w as th is , th a t i t c e r ta in ly p a id to c o n tro l w o rk in g losses closely, a t any ra te w hen th e q u an titie s invo lved w ere large . I f a to le ra n c e of sa y p lu s or minus 0-1 per cen t, w ere accep ted , w hen a l i t t le m o re ca re w o u ld g iv e h ig h e r accu racy , the resu lting confusion in w ork ing losses w as o f te n se rious. T w o ta n k s a t th e same in sta lla tio n m igh t th e n differ b y 0-2 p e r c e n t, in th e e x tre m e case ; if th ese tanks received oil from ta n k e rs , th e n th e a p p a re n t t r a n s i t losses to th e in s ta lla tio n could v ary b y 0-2 p er cen t., even if th e rea l losses w ere c o n s ta n t in p e rc e n ta g e . T he p er­form ance of th e ta n k e rs w ould com e u n d e r susp ic ion , a n d on e m ig h t easily w aste tim e in looking in to th is , w hen th e fa u lt rea lly la y elsew here.

D raw ing oil from one ta n k in s tea d of th e o th e r , th e in s ta l la t io n w o u ld show 0-2 per cen t, difference in w orking loss. L ocal s ta ff w o u ld becom e a c c u s to m e d to apparen tly inexplicable v a ria tio n s; if th e y d id n o t a p p re c ia te th e p a r t th e ir ta n k tab le s were p laying, th e ir efforts to p roduce com prehensib le loss figu res w o u ld be ineffective, w hen in te re s t in th e questio n te n d e d to b e lo s t. T h e e x tre m e ran g es in working losses o ften rep o rted from tw o in s ta lla tio n s w o rk in g u n d e r clo se ly th e sam e con­d itions w ould be fam iliar to h is au d ience ; p o o r t a n k ta b le s w ere o f te n a fa c to r in this.

A lthough he him self reg ard ed th e a c tu a l p rocess of s t r a p p in g as one requ irin g care a n d little else, he p o in ted o u t th a t i t seem ed to be w o r th w h ile overcom ing w hatever difficulties m igh t be m e t w ith , considering t h a t th e re su ltin g ta b le s m ig h t be in use for m any years. T an k s se ldom lay id le ; ta k e n o v er a few y e a rs , th e re w ere few cases m w hich th e oil volum es d e a lt w ith b y a ta b le w ere n o t la rg e . H e w o u ld n o t a tte m p t to ay dow n w h a t s ta n d a rd s of acc u ra c y sh o u ld be a d o p te d ; a c c u ra c y for i ts ow n sake was a m istake , b u t in s tra p p in g ta n k s i t seem ed to h im t h a t i t p rom ised definite economies.

M r. 13. C F e r g u s o n asked w h e th e r M r. K e r r h a d a n y ex p e rien ce of th e correlationween ca i ra tio n by in te rn a l d iam e te rs a n d b y e x te rn a l c ircum ferences .

Mr . K e r r said th a t th e m e th o d of s tra p p in g m e t h is ow n need s so m u ch more osety th a n th e m ethod of m easu rin g in te rn a l d ia m e te rs t h a t h e h a d n o t h ad m uch

DISCUSSION ON “ TANK STRAPPING.” 135

experience of th e la t te r m ethod . H e saw no reason, however, w hy th e m easurem ent of in ternal d iam eters should n o t give one an excellent tan k -tab le . I t was clear th a t if a ta n k was m easured em p ty th e d iam eters ob tained d id n o t con tain th e correction for ta n k expansion which he preferred should be added in th e case of large tan k s , b u t th a t could be calcu la ted and allowed for. H e believed th a t very close agreem ent would be ob tained betw een th e tw o m ethods.

M r . M a l c o l m B r o w n e said th ere w as one question he w o u l d like to ask w ith regard to buried tan k s . T anks above ground were usually circular, vertica l or horizontal, b u t in th e case of buried storage o th e r shapes were used. H ad Mr. K err h ad any experience of calib ra ting tan k s th a t were square or rectangular, for instance, and , if so, w h a t difficulties m ight be expected in connection w ith th em ?

M r . K e r r , in a w ritten rep ly to th e above question , said th a t he h a d h ad only a little experience w ith rectangular tan k s . C alibration m ethods th a t suggested th em ­selves were calculations from th e drawings, m easurem ent of in te rn a l dimensions, or d irect filling w ith m easured q u an titie s of liquid, since rectangu lar tan k s were usually sm all. H e th o u g h t th a t d irect filling w as likely to be th e m ost accu rate m ethod ; especially if th e ir dim ensions were large, fla t ends and sides ten d to d is to rt ra th e r n o tab ly outw ards under th e oil head, unless th ey are very well supported by th e ea rth outside th e tan k , or otherw ise streng thened . H e believed he h ad seen in th e lite ra tu re a discussion of th e am oun t of th is d isto rtion , b u t a t th e m om ent of w riting he was abroad , so th a t he could n o t give a m ore definite re ference ; he had m ade no tria ls to com pare th e resu lts of th e various m ethods of calib ra tion on th is ty p e of tan k .

I f th e ta n k were laid w ith a t i l t for drainage, or if it developed a t i l t th ro u g h weakness in th e foundations, th is m igh t lead to inaccuracies unless allowed fo r; w ith sm all tan k s , errors of a few gallons were h igh in th e ir percentage effects. F o r th is reason, deadwood h ad to be tre a te d carefully if good percen tage accuracies were to be ob tained from m easurem ent m ethods in these cases. Once th e ta n k had se ttled on its founda­tions, calib ra tion by filling w ould allow au tom atica lly for t i l t effects.

In m easuring in ternally , th e end loop of th e tap e m igh t be held against one side of th e tan k , an d a g raduated steel ru le in th e requ ired position against th e o th er s id e ; th e tap e would th en be read off opposite some convenient g raduation on th e rule, th e tap e an d ru le lengths being added to ob ta in th e to ta l leng th . The tap e m igh t be used sim ilarly betw een tw o rules. I t should n o t sag betw een th e ta n k e n d s ; i t m ight be supported on a wood b a tte n . Sm all varia tions from th e correct tap e tension would n o t be im p o rtan t for short len g th s ; b u t w hen large d iam eters were m easured, special a rrangem ents would have to be used to secure th e p roper tension an d to perm it correction for sag, if good accuracies were required .

M r . A. C. H a r t l e y asked w hether Mr. K err h ad h ad any experience w ith welded tan k s , an d w hether there was any difference in th e behaviour of w elded tan k s and riv eted tan k s in regard to expansion under a head of oil.

M r . K e r r said he h ad h ad very lit t le experience w ith w elded tan k s , h u t he w as credibly inform ed th a t th e welded ta n k was o ften less regu lar in shape th a n th e r iv e ted tan k . The only cause of difference in expansion u n der oil-head, he th o u g h t, was th a t a t th e horizon tal seams of a riv e ted ta n k th e p la tes overlapped and a t th a t level th ere was a g reater resistance to expansion th a n in th e open p la te . In th e case of w elded tan k s th e re were no such overlaps, so th a t he would expect th e expansion of a w elded ta n k to follow th e boiler expansion fo rm ula m ore closely th a n w ould a r iv eted tan k , unless it h ad stiffening rings. H e h ad y e t to calcu late o u t th e effect of th e overlap of horizontal seams on th e ex p an s io n ; th e theo ry underly ing th e calcula­tio n was ra th e r com plex, and he h ad n o t y e t h a d tim e to go in to th e question properly . H e h ad checked up th e effect p ra c tic a lly ; his re su lts suggested th a t a t four to six inches above th e seam th e expansion form ula given in th e p ap er could be applied sa tisfac to rily . A t th e seam itself th e re s tra in ing effect of th e overlap could be d e tected , b u t i t w as m uch sm aller th a n w ould be expected if th e overlap ex tended over an y g rea t d ep th . I t w as for th a t reason th a t he suggested s trap p in g 4 to 6 inches above th e b o tto m of th e course.

136 DISCUSSION ON TANK STRAPPING.

Dk F H G a b n e r said th a t in connection w ith correction for oil tem pera tu re Mr. Kerr s ta te d 'th a t the volumes calculated from th e strapp ings were la te r corrected for the cubical coefficient of expansion of steel. Suitable corrections could qu ite well be incorporated in the standard volume correction ta b le s ; ad o p tio n of th is procedure would considerably simplify accurate calculation of volum es from ta n k m easurements.

M r Kerb, said th a t Dr. G arner h ad referred to an o th er side of a question already raised in the discussion— i.e., as to how far oil m easurem ent should be considered as a whole and how far i t should be separated ou t in to its co n stitu en t p a rts . I t was quite possible to allow for expansion by ad justing th e coefficient used for th e oil; it was largely a m atter of opinion, b u t, on th e whole, he w ould like h is ta n k tab les to show true volumes directly. H e was qu ite willing to agree, how ever, th a t th e final results would be the same.

Dr. G am er’s suggestion assum ed th a t oil volum es w ould be reduced in routine by suitable tables to 60° F ., or a t least to some s ta n d a rd tem p era tu re . W hile th is was standard practice in th e U nited S ta tes system , m easurem ent b y w eight, without reduction of either volumes or gravities to s tan d ard tem p era tu re , w as common in both the British and m etric system s. H ere th e observed specific g rav ity of the oil was corrected to th e tem perature of th e oil in th e tan k , an d th e oil volum e, measured a t the same tem perature, was m ultiplied by th e figure so o b ta ined to get th e weight. The process had th e advantage th a t th e effect of possible errors in th e coefficients used to reduce the gravities were g reatly m inim ised. Allowance for th erm al expansion of the tan k in the m anner suggested by Dr. G arner was inapplicable when th is method was used.

A part from this, the m ethod of correction suggested b y D r. G arner would be advantageous, since instead of ad justing th e ta n k tab les to an average temperature of use the correct allowance would be m ade a t each gauge for th e ta n k ’s actu al tempera­ture. He apologized for having suggested a t th e m eeting th a t th e two methods were equivalent; they would average ou t equally , b u t D r. G arner’s m ethod undoubt­edly had the advantage in th e individual cases.

T h e P r e s i d e n t , in proposing a h earty vote of th an k s to Mr. K err, said he was sure everyone present would agree w ith him th a t th e discussion th a t evening had been most enlightening and interesting. H e th o u g h t th e m ethod which Mr. Kerr had adopted of answering each question as i t was asked was a useful one, and might be adopted in m any other cases; it had certain ly b ro u g h t o u t a very large number of points which m ight no t otherwise have been raised. H e would rem ind the members th a t remarks in writing would also be welcomed, and he w as sure th a t Mr. Kerr would be pleased to answer in w riting any questions w hich were asked in th a t way.

He thought the In stitu te was to be congratu la ted on being able to give the lead to which Mr. Kewley had referred by having a paper on th e sub ject of tank-strapping in its Journal. The subject was n o t an aspect of pe tro leum technology to which he himself had devoted m uch a tten tio n ; he had alw ays been co n ten t to leave it in other people s hands, and he had been surprised th a t evening to lea rn th e accuracy with which measurement of oil in tan k s could be carried ou t.

He wished to express th e th anks of th e In s ti tu te to th e A siatic Petro leum Company for perm itting the publication of Mr. K e rr’s work, and th u s helping th e In stitu te to give a lead on the subject of in ternational stan d ard iza tio n of such work. He felt th a t the prestige of the In s titu te would be greatly enhanced by th e publication of Mr. Kerr s paper and the tables which should accom pany i t. I f any arrangement could be made for coining to some agreem ent w ith th e A m erican Petro leum Institute on the question of the standard ization of m easurem ent, i t w ould be one of the finest ways of creating reciprocal understanding betw een th e petro leum technologists of the world. The In s titu te was m uch indebted to Mr. K err an d his Company for the presentation of the paper th a t evening.

The vote of thanks was accorded w ith acclam ation, and th e m eeting then terminated.

E N G IN E K N O C K A N D IT S E F F E C T ON F U E L D E V E L O PM E N T .*

By Professor D. M. N e w i t t , M.C., D.Sc., D.I.C., A.R.C.S.

I t is m y purpose this evening to give some account of the combustion of hydrocarbons and, more particularly, of th a t aspect of their combustion which is peculiar to the internal combustion engine, and which, in certain circumstances, gives rise to the phenomenon of knock or pinking. The subject is one of considerable technical importance, for the incidence of knock in petrol fuels is a dom inating factor in the design of engines and has exerted a profound influence on the trend of development in the petroleum industry.

The origin and cause of knock has long puzzled both chemists and engineers, and attem pts to find an explanation which will account for all the facts have hitherto not been entirely satisfactory. All the evidence, however, points to it being a chemical phenomenon resulting in a sudden acceleration of the ra te of combustion in the engine and the initiation of intense compression waves in the charge. A knowledge of the chemical changes occurring during a working cycle is therefore an essential pre­lim inary to an understanding of the problem, and I propose to give a brief account of some recent researches which have a direct hearing upon the m atter.

I t is well known 1 th a t a quiescent combustible gas-air or oxygen m ixture of suitable composition will, under certain conditions {e.g., when ignited in a horizontal tube closed a t one end), propagate flame a t a uniform slow rate of the order of 5 to 10 m etres per second. Provided the tube is suffi­ciently long the uniform propagation m ay be disturbed by vibrations which cause compression waves to traverse the medium, and i t m ay eventually develop into detonation, the flame then travelling a t the enormously en­hanced speed of some 3000 m etres per second. These are w hat m ay be term ed the normal types of flame propagation, and they m ay conveniently be studied by photographing the flame on a film moving a t right angles to the axis of the tube.

There is no need to enlarge upon these types of combustion, which have frequently been described and are probably familiar to you all, bu t I should like to draw attention to three photographs taken by m y colleague, Mr. R. P. Fraser, which illustrate certain points of interest in connection w ith them .2

Plate I (a) is a photograph of flame traversing a carbonic oxide-oxygen medium in a horizontal tube closed a t both ends. I t shows very clearly the way in which a series of compression waves m ay be set up and traverse the gases.

P late I (b) illustrates the setting up of detonation a t two points each a little ahead of the flame front in a similar m ixture burning in a horizontal

* P ap e r p resen ted to a J o in t M eeting o f th e South W ales B ran ch o f th e In s ti tu te o f P e tro leum an d th e local section o f th e In s titu te o f Chem istry, he ld a t Swansea on D ecem ber 9 th , 1938.

138 NEW ITT : ENGINE KNOCK AND

tube I t is interesting to note th a t from the point of detonation two flames travel forwards, the one a t a com paratively slow speed of 360 metres per second and the other a t about 3000 metres per second.

Plate II (a) shows a detonation wave traversing a carbonic oxide-oxygen medium at a uniform speed of 1760 metres per second. The regularly distributed striations in the photograph indicate th a t the detonation head is moving in a spiral path, and measurement shows th a t it is rotating at a speed of 44,000 revolutions per second.

Now the type of combustion producing knock, although associated with compression waves and a high rate of reaction, cannot be identified with either uniform slow propagation or with true detonation. This is shown by Plate II (6), which is a photograph of the flame traversing an engine cylinder in which knock is taking place.3 I t will be seen th a t towards the end of the stroke, when only a relatively small proportion of the charge remains unburnt, there is a sudden increase in velocity accompanied by a violent vibratory movement of the gases. This vibratory movement, together with resonance effects in the cylinder walls, gives rise to the familiar sound of knock.

Let us consider for a moment the condition of the unburn t charge near the end of the stroke. I t is in contact with the hot walls of the combustion chamber and with the faces of the exhaust valve, it is subjected to adiabatic compression, and it is irradiated from the flame and bombarded by highly energized radicals and molecules projected ahead of the flame front. These are the environmental conditions th a t m ust be borne in mind. I t is clear tha t they are such as would be expected to cause partia l oxidation with the formation of intermediate oxygenated products, atom s and radicals, and it is the presence of one or more such bodies in critical concentrations which sensitizes the mixture and doubtless renders it prone to knock.

A certain amount of information, both direct and indirect, is available about the kinetics of these initial processes. There is no question, for example, but th a t pressure, tem perature and tim e are im portant factors; and experiment has shown th a t the ra te of sensitization m ay be accelerated or slowed down by the introduction of small quantities of promoters and inhibitors, respectively. These facts suggest th a t the high rate of com­bustion producing knock is due to a reaction of the branched chain type the chain-carriers being in all probability radicals derived from the thermal decomposition of some intermediate product formed in adequate concen­tration during the sensitizing period.4

A large number of inhibitors or anti-knocks is known, although only one, namely lead tetrethyl, has come into general use. Usually, though not invariably, they turn out to be volatile organo-metallic compounds which decompose to give metallic atoms and radicals, and, as Egerton has pointed out, the metal is always one th a t is capable of forming both a lower and a higher oxide. They have very little efiect upon the speed of flame propa­gation or upon the setting up of true detonation, bu t they appear to act either by retarding the initial slow reaction or by limiting the number of reaction centres from which chains are set up.

From the point of view of the production of high-duty fuels (i.e., fuels of high knock-rating), considerable interest attaches to the precise mechan­ism of the chain reactions which are responsible for knock, and a large

P l a t e I (a).

G R A P H IC A L P IC T U R E O F F L A M E IN ' A M IX T U R E O F 2 C 0 + 0 2 IX A CL JS E D T U B E 1 -6 m . L O N G IG N IT E D C E N T R A L L Y .

[To face p 138.

P l a t e I (6 ) .

DETONATION IN A CARBONIC O X ID E -O X Y G E N M E D IU M .

P l a t e I I (a).

THE SPIN NING DETO NATION W AVE IN A M IXTU RE OF 2CO -j- O a IN A T U B E 1*3 Cm . DIA M ETER. ROTATIONAL SPE ED 44,000/sec. D E TO N A TIO N S P E E D 1760 I t I . / s e C .

P l a t e I I 'b). FLAME TRAVERSING AN

ITS EFFECT ON FU EL DEVELOPM ENT. 139

amount of work has been done in recent years in attempts to identify them. The problem is by no means an easy one. The slow combustion of any of the higher hydrocarbons leads to the formation of a very complex mixture containing aldehydes, alcohols, peroxidic bodies and acids in varying pro­portions, and the chain-carrier may be present amongst them in quantities so small as to defy detection by ordinary analytical methods. I t is possible, however, to narrow the field. The dependence of knock on pressure, temperature and time suggests that the substance we are looking for must be relatively unstable under the working conditions and must be capable of breaking down to give active radicals which can act as chain-carriers. The fact that very small quantities of inhibitors (e.g., 1 vol. of lead tetraethyl in 1300 vols. of fuel), are effective, indicates tha t it need be present only in low concentrations, and it is also reasonable to suppose that it would belong to the class of promoters and, when added to a fuel, would act as a powerful knock-inducer.

The substances which might be formed during slow oxidation and which conform to the above specification are alkyl-, alkyl hydrogen- and alcoxy- peroxides, aldehyde peroxides and peroxides of oxygenated ring compounds (e.g., vinyl ether peroxide). Oxides of nitrogen which are also present are not pro-knocks. The presence of one or more organic peroxides in the exhaust gases of an engine has been demonstrated by Dumanois, Mondain- Monval and Quanquin,5 and they have also been detected by Egerton, Smith and Ubbelohde 4 in the gases extracted from an engine just prior to knock.

On examining the “ promoting ” characteristics of typical members of each of the above classes of peroxide it is found that whilst define peroxides have only a slight effect, acetyl-, ethyl hydrogen- and diethyl-peroxides are powerful pro-knocks. The results as a whole suggest tha t only those peroxides are effective which decompose to give -O R and -O H radicals, the former most probably acting as chain-carriers. In this connection mention may be made of a recent detailed study of the decomposition of diethyl peroxide by Harris and Egerton.6 They find that whilst the slow decomposition takes place according to a unimolecular law there is a certain critical pressure above which it becomes explosive; the stabilized products in the former case are acetaldehyde and ethyl alcohol and in the latter formaldehyde and ethane.

Having inferred something about the type of reaction which is or may be responsible for knock we may now attem pt to find a correlation between the tendency to knock and the chemical reactivity of the fuel. I t is knowm tha t all straight-chain paraffins, with the possible exception of methane, knock readily, whilst branched chain paraffins, defines, diolefines and the simple aromatic hydrocarbons have comparatively high knock ratings. Generally speaking, the more compact the structure of the molecule the less likely is it to knock. Now, a convenient measure of the chemical reactivity of a fuel is its spontaneous ignition temperature, and the latter should, therefore, be closely related to the knock rating.

M. Prettre ' has measured the ignition temperatures of a wide range of hydrocarbons a t atmospheric and reduced pressures, and D. T. Townend 8 has investigated the relation between the spontaneous ignition temperatures and pressure up to pressures of about 15 atmospheres. We may refer to

the latter investigation, which is of particular significance in connectionwith the knock problem.

The spontaneous ignition temperatures over a wide range of pressures are conveniently measured by admitting the previously mixed fuel-gas medium into a heated vessel. Provided the temperature is sufficiently high there is usually a short delay period or “ lag ” during which no appreciable change of pressure occurs, followed by an abrupt increase in pressure, and ignition. The time-lags can be measured by means of a stop-watch and the ignition indicated by a suitable pressure gauge.

If we consider first some non-knocking fuels, such as methane, ethylene and benzene, we find that the ignition temperature falls progressively and

] ^ 0 NEW ITT : ENGINE KNOCK AND

F ig . l .

(1) 1 3 % Methane.(2) 1 -3 6 % Benzene.(3) 1 0 % Ethylene.

slowly with increase in pressure, as shown by the smooth curves in Fig. 1. The small figures adjacent to the curves indicate the time-lags before ignition. This is the type of relation that might be expected on the basis of the thermal definition of ignition and of a comparatively simple oxidation process, and needs no further comment. On comparing the results with the corresponding curves for some typical knocking fuels, namely «-octane and «-heptane, which have a low knock-rating, and iso-octane which has a high one (Fig. 2), we are at once struck by a marked difference in the character of the pressure effect.

In the case of the two normal paraffins the first effect of increase of pressure is to cause a very rapid fall in the ignition temperature and a corresponding decrease^ in the time-lags preceding ignition. Below about

ITS EFFECT ON F E E L DEVELOPM ENT. 141

400° C. there is a low-temperature system in which may be noted two pressure minima of ignition, and above 3 atmospheres ignition occurs at about 250° C., with progressively increasing time-lags. The shaded area to the left of the ignition curve marks the region in which “ cool ” flames are formed. The curve for iso-octane is somewhat similar, but there is only one pressure minimum in the low-temperature system occurring a t a pressure of about 3 to 4 atmospheres higher than in the case of the normal paraffins. The minimum ignition temperature is also some 40° higher and the cool-flame region is more extensive.

F i g . 2.C O R R ESP O N D IN G M IXTU RES W ITH A IR OF ( 1 ) n -O C T A N E ; ( 2 ) « -H E P T A N E ;

(3) ISO -O CTANE.

(Figures along curves denote tim e-lags (seconds)).

The conditions of temperature and pressure in an internal combustion engine are such as to suggest that knock is related to the reactivity in the low-temperature ignition system as determined by the curves in Fig. 2; and if we compare, as in Table I, the pressure minima read off from a series of such curves with the critical compression ratio of the fuels in question, as measured by a variable compression engine, we find th a t the fuels may be arranged in the same order.

In comparing these results it must be borne in mind that the minimum pressures relate to time-lags of the order of 1 sec. or more, whereas under engine conditions the time-lags would have to be of the order of a few thou­sandths of a second.

142 NEW ITT : ENGINE KNOCK AND

T a b l e I .

A Comparison of the M inimum Ignition Pressures and the Critical Compression Ratios of a Number of Paraffin Fuels.

Fuel.Minimum pressure,

atms.Critical compres­

sion ratio.

Methane . Propane . Butane . Pentane . Heptane . Octane . isoOctane

6-83-2 2-2 1-58 1-34-85

1401206-4 3-8 2-8 2-87-6

I t now becomes a matter of interest to discover something about the chemical changes which are taking place in the low-temperature ignition

F ig . 3.SPONTANEOUS IG N ITIO N T E M PER A TU R E S OF A C3H g + 0 2 M IX TU R E .

system. Newitt and Thornes 9 have examined propane in detail and have made careful analyses of the products from combustion in the cool-flame region and in the slow-combustion zone immediately above it. The ignition temperature-pressure curve (Fig. 3) for an equimolecular propane- oxygen mixture resembles that for iso-octane, inasmuch as there is only one pressure minimum situated a t about 0-5 atm. The c o o l- f la m e region is also veil defined, and there are zones within it in which two o r more separate cool flames traverse the medium at intervals o f s o m e 7 0 seconds,

ITS EFFECT ON FU EL DEVELOPM ENT. 14 3

each one being accompanied by a small pressure pulse. By carrying out the reaction in a quartz vessel it is possible to remove the vessel from the furnace in which it is being heated, and plunge it immediately into a freez­ing mixture, so as to arrest the reaction at any desired intermediate stage. In this way Newitt and Thornes have succeeded in following the course of the cool-flame- and slow-reactions in some detail.

Time, mins.F ig . 4.

PR O D U C TS PROM TH E REAC TION OP A C3H 8’ + 0 2 M E D IU M AT 400 M M . A N D 294°.

Attention may he drawn to two series of experiments carried out by them along the isobaric line ED (Fig. 3) in the cool-flame region and in the slow- combustion region immediately above it, respectively. The results are shown graphically in Figs. 4 and 5 and are summarized in Tables II and III.

Considering first the slow-combustion experiments, it will be seen that there is an induction or lag period of 10-15 secs, followed by a reaction period of 2-5 mins. during which intense luminescence is visible in the reaction vessel. In the gaseous products upwards of 30 per cent, of the carbon of the propane burnt appears as propylene at all stages of the re­action ; both acetylene and methane are also present throughout, diminish­ing in quantity as the reaction approaches completion. Amongst the

144 NE WITT : ENGINE KNOCK AND

jjquid products peroxidic bodies and aldehydes appear in comparatively lame quantities, the peroxides reaching a maximum when the reaction is proceeding with maximum velocity and representing no less than 19-9 per cent, of the carbon of the propane burnt. The higher aldehydes reach a maximum somewhat earlier, whilst formaldehyde, present a t the com­mencement of the reaction to the extent of 4-6 per cent., gradually diminishes as reaction proceeds.

F ig . 5.PRODUCTS FROM THE COM BUSTION OF A C 3H S - f 0 2 M E D IU M AT 3 6 0 MM. AND 4 0 0 ° .

Attempts to identify and isolate the"peroxidic substances have not yet been successful. I t seems probable that they are responsible for the ethyl alcohol which constitutes the greater part of the higher alcohols found in the cool-flame experiments. They are also decomposed by concentrated potassium hydroxide solution liberating a gas consisting of 80 parts of hydrogen to 20 parts of oxygen. These and other facts point to the presence of a mixture of peroxides containing an alkyl peroxide, a per-acid and/or hydrogen peroxide.

Turning now to the series of experiments in the cool-flame region, data are presented up to and including the passage of the second cool flame. The induction period for the reaction is 8 mins., the first cool flame is o ser\ ed after 8-75 mins., the second after 9-4 mins., and reaction is complete

•if

ITS EFFEC T ON FU EL DEVELOPM ENT. 1 4 5

00g C

o sr CO

•“Q IIee C

_oo V>

•_© ~C l ■-e U-A

0

3 o

«! 3=. -3

° TI X+ 3

w 1

dr -e

ei .2

© O O O C i;O C :x»Q C !N C<l !N

> © io © so

> l-: -«t c; w c; N Tj-

0 0 0 i ' C C i (,# 0

f l h C X X o , o iiC r t o o o M « '£ g o

3■p.........................OQ

§ -f i ^ l o o ..................

: « 0 0 0 00 *-« v v * ’Sw f t j o ©* I §>£ • • • • £

.3 8 « ^ |• o c

t £ o O . oo . . Oc ^ P fl "S © ©.5 1 2 ^ n C Ti « C o o ir © *p. o cu-g 3 b - a a£ fl ® c3^ . © >i . «§ o « o ® -n-a om o 5 ® s "3 •§ ®■g 3 -g Js 5 , . 2 3 0e> 8)5 ^ ° 2 j : a S 2 £ S 'S 'S fc S H E g ‘5 .5 X S h » 3 o e f E-iQPM < K fe p H P i

146 NEW ITT EN G IN E KNOCK AND

+Wn

H W„o

3TJ.3

acOM '^Nt'OM LOHW CC .......................................................... < 6 c D H H

OOCOO<MI>»OCO?Ot (NC<JCp

(C O iO X O M H O N Q f' .................. <(NO^COrH

^■^Xt>)COOCDO(MlO

Ot^rH0cDoo^ooOOOlO(N ^

iocopyicoa>'^t>coc5^t^t'- CO CO <—i l i 5 0 o b H ^ H ' 2 ( O Q CO TH CO .—I — i S

c34 3 .................................0} ....................................fl£ « ®

I ^ W H o o V ....................: ¿ p o o o o

S f ' T i ' 1” § ..............i■5 s 1« I„•« fl o dc -p ® a -g 8 • • • • 811, § & .£ 5 U ® o | h3 a s 3 - ° - s U .'g 'g 11!do 0 3 o o -d,d o S o

-S®SO g p « . ' 2 ' r “ o§ S s | a 8 | \ 3 . s | M* s ° # § gS.SPSB O Ph WpuFm^ M P h

ITS EFFEC T ON FU EL DEVELOPM ENT. 1 47

after 11 mins. The chemical changes occurring during this reaction are rather complex, but it is possible to correlate them to some extent with the cool-flame phenomena. Thus the analytical figures show th a t higher aldehydes, acids, propylene and carbon monoxide are present in detectable quantities a t the outset of the reaction, but no peroxides, or formaldehyde, and only a trace of carbon dioxide. The higher aldehydes rise to a maxi­mum 45 secs, after reaction starts, and a t this point the first cool flame is observed. During its passage through the reacting medium the percentage of higher aldehydes diminishes whilst peroxides and formaldehyde make their appearance; the rate of oxidation of the propane also increases. Shortly after the extinction of the cool flame, the peroxides reach a maxi­mum and the higher aldehydes a minimum. A little later the aldehydes again increase rapidly to a second maximum, higher than the first, whilst the peroxides simultaneously decrease; the second cool-flame ignition then takes place and is accompanied by changes in the products similar to those noted with the first flame. The rate of formation of propylene appears to follow th a t of the higher aldehydes a t all stages, whilst carbon monoxide increases fairly uniformly throughout.

On the basis of the above experiments it may be concluded that a neces­sary condition for cool-flame inflammation and for ignition in the lower region is a critical high concentration of higher aldehydes. During the passage of the flame the aldehydes are further oxidized to give aldehyde peroxides and/or per-acids and ultimately formaldehyde. The reactions by which propylene and the higher aldehydes are formed also appear to be related.

Now although the conditions in these experiments differ in some material respects from those in an engine, particularly as regards the time factor, there is reason to suppose tha t the mechanism of the combustion is essen­tially the same in the two cases. I t may, indeed, be assumed that the sensitized mixture in which knock originates is one having a composition not very different from that of a mixture in which cool-flame inflammation is about to take place. And if such a mixture were ignited by an external source there is little doubt but th a t a rapid combustion closely resembling th a t associated with knock would occur.

The products of the combustion of all the higher hydrocarbon fuels including branched-chain paraffins and defines are similar to those of propane in the sense tha t aldehydes, alcohols, peroxidic bodies and acids are always formed. The proportions in which they survive a t any par­ticular stage of the combustion, however, varies from fuel to fuel and depends upon temperature and pressure; and it is this circumstance perhaps more than any other which constitutes the difference between a low and high knock rating fuel. I t will be recalled, for example, that although a striking similarity exists between the ignition tem perature- pressure curves for w-octane and iso-octane a higher pressure is always required to affect the ignition of the latter a t any given temperature in the ignition range. This may be interpreted as indicating that a higher pres­sure is required to stabilize the initial products so th a t concentrations adequate to initiate cool flames may be built up.

The same considerations apply to the knock ratings of some of the more recently tested high-duty fuels such as acetone and diisopropyl ether ;8 in

both cases the pressures required to promote ignition in the low-tempera- ture range are comparatively high. In the case of mixed fuels, as for example the petrols obtained by the hydrogenation of coal, the combustion characteristics depend largely upon their straight chain- and branched- paraffin, naphthene and volatile aromatic hydrocarbon content; and methods of controlling the relative proportions of these constituents by vapour phase hydrogenation in the presence of selective catalysts have recently been discovered and put into operation for the production of high-duty fuel.

I t would be incorrect to assume that anything like finality had been reached in the development of high-duty fuels and of anti-knocks. During the past ten years there has been a steady improvement in the knock ratings of commercial fuels, and a number of new fuels have been introduced to meet the increasingly stringent demands of engine designers. In this connection some of the researches I have described or referred to are proving of considerable utility, particularly in indicating directions in which further advances may profitably be made. Finally, I think it will be ad­mitted that, although the precise reactions responsible for knock and the inhibiting effect of anti-knocks have not yet been identified, the general character of the mechanisms involved are well established.

References.

1 Bone and Townend, “ Flame and Combustion,” Longmans, 1927.2 Fraser, “ Science of Petroleum Bone and Fraser, Phil. Trans., 1935, 235, 29.3 Withrow and Rassweiler, Auto. Eng., 1934, 24. 281, 385.4 Egerton, Smith and Ubbelohde, Phil. Trans., 1935. 234, 433.5 Dumanois, Mondain-Monval and Quanquin, Ann. Chim. Phys., 1931, 15, 309.6 Harris and Egerton, Proc. Roy. Soc., A, 1938, 168, 1.7 Prettre, Ann. Comb, liq., 1931, 6, 7, 269, 533 et seq.8 Townend, “ Science of Petroleum,” Proc. Roy. Soc.., A , 1936, 154, 95: J. Chem.

Soc., 1938, 238.9 Newitt and Thornes, J . Chem. Soc., 1937, 1656.

1 4 8 ENGINE KNOCK AND ITS EFFECT ON FU EL DEVELOPMENT.

T H E CO M PO SITIO N O F SOME RO U M A N IA N S T R A IG H T -R U N G A SO LIN ES.*

By C. D. N e n i t z e s c u a n d A. C o n s t a n t i n e s c u .

S y n o p s i s .

In the present paper the composition of some Roumanian straight-run gasolines is studied, not so much from the point of view of the identification or estimation of certain individual hydrocarbons occurring in these gasolines, but chiefly to obtain a general idea of the main classes of hydrocarbons occurring therein. The basis of modem research work dealing with the composition of gasoline is a distillation process, using high-efficieney apparatus, which has lately been greatly improved.12 This method has been employed. The gasolines were distilled in a sim ilar apparatus, and the narrow fractions obtained were then analysed by the most convenient methods, with a view to establishing the percentage of aromatic, naphthenic, normal paraffin and branched paraffin hydrocarbons occurring therein.

F r a c t i o n a t i n g C o l u m n .

For the construction of this apparatus we utilized the information given by M. It. Fenske and co-workers.3 The height of the packed section was 6 m., whilst the to tal height of the distilling apparatus was 7 m. The column consisted of a brass cylinder which was packed with single-turn rings of 4 mm. diameter made of 0-5 mm. brass wire.

The still employed in our experiments (See Plate 1) was a vertical cylinder, 220 mm. diameter, 330 mm. high, in which 8-5-9 litres of liquid could be distilled. In order to test the efficiency of the apparatus, a smaller still of 1-5 litres capacity was used.

At the top of the column was fitted a total reflux condenser which could be connected by means of a small stopcock with the product line. The distilled product was collected in a graduated burette of 100-ml. capacity provided with two glass condensers.

In order to obtain a distillation under adiabatic conditions, the column was insulated with a mixture composed of asbestos and magnesium oxide over a layer of 10 cm. About the middle of this layer—i.e., a t a distance of 35 mm. from the column—a chrome-nickel resistance belt 0-1 mm. thick and 5 mm. wide was fitted. The column was wound in three separate sections, 2 m. long, each being controlled by a variable resistance and an amperimeter. By this method the column could be maintained, as desired, a t a temperature equal to, or slightly below, th a t of the vapour phase. The same insulating and heating system was used for the still.

In each of the three heated sections, as well as on the top of the column, a Hartmann and Braun resistance thermometer was fitted, with which temperatures within 50-150° C. can be measured with a precision of 0-1° C. The boiling points of the fractions were measured with a thermometer fitted on the top of the column.

* Paper received September 27th, 1938.

15 0 NENITZESCU AND CONSTANTINESCTT : THE COMPOSITION OE

T e m p e r a t u r e s a r e r e a d a n d c o n t r o l l e d f r o m a c o n t r o l - p a n e l c o n t a i n i n g

a l s o t h e c o l l e c t i n g b u r e t t e f o r t h e p r o d u c t s . _The actual volume of the packing material is 390 ml. The free space

amounts, therefore, to 79-3 per cent. The total surface of the packing is approximately 3 sq. m., and the hold-up, which was determined as usual with kerosine, amounts to 78 ml.

During distillation a superpressure of 30-40 mm. Hg was observed, which was measured with a manometer fitted on the above-mentioned control- panel.

The column was tested under total reflux with a binary mixture consisting of «-heptane and methylcyc/ohexane,4 the former being derived from Jeffreys pine and having the following constants : D — 0-6839 andn = 1-3878, whereas the boiling point determined with our column (at a pressure of 750 mm.) was 98-0° C. The methylcyc/ohexane used was purified by us in the usual manner. With our column we found a boiling point of 103-3° C. and a refractive index n = 1-1232.

The efficiency test was carried out by distilling under total reflux a mixture of 2-73 mols. of each of the above-mentioned hydrocarbons. When equilibrium was reached, which took about 1 hour, two samples, a few millilitres each, were withdrawn from the top of the column as well as from the still. At this moment the temperature at the top of the column was 98° C.

The refractive index of the samples was determined a t the temperature of 20 4- 0-05° C. by means of a Pulfrich refractometer provided with a Hópler ultra-thermostat, which was also used for the other experiments in the present study.

For the sample taken on the top of the column we found a refractive index of n = 1-38887, which, according to the tables published by E. C. Bromiley and D. Quiggle,5 corresponds to 96-5 ml. per cent, «-heptane; for the sample taken from the still we found a refractive index of n = 1-40999, corresponding to 33-7 mols. per cent, «.-heptane.

We applied the formula of M. R. Fenske, C. O. Tonberg and D. Quiggle6

X nA , „ X1A = ( a ” - 1 ! _____X n B > X IB

, X IA . . „wnere ^ — molar ratio of «-heptane (A ) to methylcyclohexane (B) on

. XnAany plate, = molar ratio of A to B on any plate removed from the

first; and n = number of perfect plates required for separation, a = the relative volatility, was taken from Beatty and G. Calingaert’s paper7 as equal to 1-07. o n e

By means of the above formula we found th a t n = 60, corresponding to the theoretical number of plates of our column. Next, according to the following equation, we have :

H .E .P .T . = = 10-2 cm.60-1

After taking the two samples under total reflux, the product line was opened and the liquid distilled a t the rate of approximately 0-8 ml. per

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 151

minute. So as not to complicate the apparatus, we discontinued determin- ing the reflux ratio which, however, was later estimated at approximately 1 : 20-30. Twenty-six fractions, 20 ml. each, were collected, and the

6 *

C 6 0

■ f t 5 6

10 5 0 90 150 170 110 250 2 9 0 330 570 910 950 990 530570 610 Total Vol. clist. in ml.

F ig . 1.

distillation process was then interrupted when the boiling point of the fractions attained 100-3° C. The residue amounted to 102 ml. The com­position of each fraction was then determined by the above-mentioned method.

152 NENITZESCIJ AND CONSTANTINESCTJ : THE COMPOSITION OF

A n a l y s e d G a s o l i n e s .

In the present study three gasolines were examined—namely, one gasoline from a Bucsani paraffinic crude (Concordia Company), a gasoline from Merisor asphaltic crude (I.R.D.P. Company) and a high-octane-number gasoline selected from a Dacian crude of Gura Ocnitei oilfield (Concordia Company). The first two gasolines were prepared from crude oil in our laboratory, whilst the last one was placed a t our disposal by the Concordia Company.' All the gasolines were shaken with a sodium hydroxide solution to remove any acid which they might contain, and were then dried over metallic sodium.

The Merisor and Bucsani gasolines were first distilled in the column as such. As we found, however, that the aromatic hydrocarbons distilled over a too wide range, and thus the other hydrocarbons could be drawn together, we first removed the aromatic hydrocarbons from the Bucsani gasoline and then distilled it in our column, obtaining in this way sharper maxima. The Gura Ocnitei gasoline was distilled only after the aromatic hydrocarbons were removed.

The removal of aromatic hydrocarbons from gasoline was carried out by repeated shaking with concentrated and fuming sulphuric acid, until the presence of aromatic hydrocarbons could no longer be detected by the usual analytical methods.

All gasolines, as well as the fractions and residues, were completely saturated towards bromine water.

The characteristics of the gasolines thus obtained are shown in Table I below.

T a b l e I.

Gasoline of : ^15-Ar,

% vol. O.N. Engler Distillation.

Meritor . . . . 0-7362 6-5 63-0 I.B .P . 60° C .; 40-5% at 100° C .; F.B.P. 173° C.

Bucsani . . . . 0-7365 10-8 51-5 I.B .P . 68° C .; 29% at 100° C .; F.B.P. 182° C.

Gura-Ocnifei 0-7435 7-5 76-0 I.B .P . 53° C .; 53% at 100° C .; F .B .P. 136° C.

Bucsani (aromatic-free) 0-7262 0 48-0 I.B .P . 68° C .; 26-5% at 100° C .; F.B.P. 185° C.

Gura-Ocnitei (aromatic-free) 0-7350 0 75-0 I.B .P . 53° C .; 50% at 100° C. ; F.B.P. 140° C.

The gasolines were then distilled in the column a t the rate of 0-7-0-8 ml. per minute. Eight to nine litres were distilled, and a complete distillation lasted 8-10 days. The distillation was interrupted for only a very short time, sometimes being kept working continuously for 50 hours. During the interruptions the heat of the column was kept constant, whilst the heating of the still was interrupted. At each new start the column was worked under total reflux until a perfect equilibrium was established.

n order to avoid auto-oxidation, a small quantity of diphenylamine was added to the gasoline.r.„PpI' nS ^ le distillation process we collected first a top fraction up to

G, then 100 fractions, each distilling a t intervals of 1° C., with the only

Perce

ntage

Dg

rretgn

r tgg

rererr

ea wo

we

total

wsn/

'/are

teensn

y ow

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 1 5 3

2 .-MERISOR GASOLINE

i

7uxaput dA

tpeJ/a//

exception of the “ Gura Ocnitei ” gasoline, where the distillation was interrupted at 123-5° C. ; we thus obtained only 74 fractions.

In order to obtain comparative figures, the measured volumes of the respective fractions were converted into weights, and these were then referred to the bulk of the distilled fractions. The gross results of the distillation process are shown in Table II. The great losses observed are due to the uncondensable gases of the gasolines.

1 54 NENITZESCÜ AND CONSTANTINESCTJ : THE COMPOSITION OP

T a b l e I I .

Gasoline of : Ml.Fractions

below 50° C.

Sum of D istil­

late, % .Residue,

%•Losses,

%•

Merisor . 9200 3-5 63-8 24-5 8-2Bucsani . . . . 8000 2-6 56-5 300 10-9Bucsani (aromatic-free) . 8000 2-6 55-5 30-7 11-2Gura-Ocmtei (aromatic-free) . 8000 5-1 69-9 12-3 12-7

As will be seen from Figs. 2, 3, 4 and 5, the various fractions have very unequal weights, whilst the distillation curve has very sharp maxima and minima.

I n v e s t ig a t io n o f t h e F r a c t io n s O b t a i n e d .

For each fraction obtained we first determined the density and the refractive index. The former was measured with a picnometer at 20 + 0-05° C., and was then reduced to water a t 4° C. The refractive index was measured with a Pulfrich refractometer as explained above.

In the case of gasolines still containing aromatics, we then proceeded to estimate and remove these hydrocarbons. In the aromatic-free fractions obtained we then estimated the naphthenes and paraffins.

Before showing the results of these estimations, we must state briefly the reasons which guided us to select the various methods adopted.

T h e E s t im a t io n o f A r o m a t ic H y d r o c a r b o n s .

One of the three classes of hydrocarbons occurring in gasoline—i.e. the aromatic hydrocarbons—are characterized by a far greater reactivity and by physical constants much more specific than the other two. How­ever, as we shall see below, no method exists up to the present for the estimation of these hydrocarbons with the precision required by the usual analytical methods.

The methods described in literature for the estimation of aromatic hydro­carbons occurring in gasoline may be divided into three groups : chemical methods, physical methods and mixed methods. The chemical methods use as reagents the sulphuric-nitric acid mixture, concentrated or fuming s u i;h"ric ac^ or formaldehyde in the presence of sulphuric acid.

I his last method is not precise enough to be considered.i umerous methods utilize the sulphuric-nitric acid mixture.8 The

methods of Hess and of Egloff and Morrell are sufficiently known, but from the studies of Riesenfeld and Bandte9 and those of Faragher, Morrell and

evine it follows that these methods, in spite of the fact that they give

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 1 5 5

satisfactory results, present no advantage whatever as compared with those which use 98 per cent. acid. The use of fuming sulphuric acid in various concentrations is to be found in the older literature; 11 this reagent, however, readily attacks the hydrocarbons of the other classes occurring in gasoline. Zelinsky,12 for the first time, uses acids of lower concentration than 100 per cent, for the estimation of aromatic and ethylenic hydrocarbons. An acid of approximately 98 per cent, was then suggested by Thole,13 by Tizzard and Marshall,14 by Dänäilä and his co-workers,15 by Morgan and Soule,16 by Sachanen and W irabianz17 and others. Finally, we must mention the suggestions of Manning and Shepherd,20 and also those of Katwinkel,21 who add silver sulphate and phosphorous pentoxide, respectively, to the sulphuric acid.

Dänäilä and his co-workers use a sulphurimeter similar to that used by Kramer and Boettcher, which enables a direct reading of the volume of gasoline not absorbed by the sulphuric acid. In this manner the above method becomes very rapid and convenient. The estimation is carried out with 10-ml. gasoline, and the tube of the sulphurimeter is graduated in 0-1 ml., so th a t one division corresponds to 1 per cent, aromatics. Later on the authors have suggested sulphurimeters graduated in 0-05 ml., which are filled and read in special thermostats. We shall see below that these improvements were not of much use, because of the lack of precision of the method.

The main criticism brought against this method of directly estimating the volume of aromatic hydrocarbons by absorption in sulphuric acid is that the saturated hydrocarbons are also absorbed by the sulphuric acid : thus we have known for a long time th a t octane is absorbed by 97 per cent, sulphuric acid, under certain mild conditions, in a proportion of 2 per cent. The same conclusion was reached by Cazimir,19 working on aromatic-free gasoline with an acid of 98-3 per cent. The values found for the aromatic hydrocarbons are, therefore, greater than the real ones. Similar remarks are made by Ormandy and Craven.23

Taking these conclusions into consideration, it would seem tha t the best way to estimate aromatic hydrocarbons in gasoline would be a combined method : the determination of the density and refractive index of the respective fraction, the removal of aromatics from the fraction with sul­phuric acid under mild conditions and a final determination of the density and refractive index. The percentage of aromatics is then calculated by means of Thole’s 13 formula, using the density, or with Hoyte’s 22 formula, which uses the refractive index. This method is connected with another source of error which escapes control—namely, when aromatics are mixed with saturated hydrocarbons a contraction of volume takes place, depending on the nature of the saturated hydrocarbons present, so tha t the additivity rule on which the above calculation is based is no longer valid.24

With regard to the physical methods applied or suggested for the estima­tion of aromatics in gasoline, the only one worth mentioning is the well- known “ aniline-point ” method. This method, which may be of service when it is only a question of estimating aromatics in large fractions of gaso­line, was not sufficiently precise for our purpose, because, as Tilitschejew and Dumskaja 24 have shown, the coefficients vary with the concentration of the aromatic hydrocarbons as well as with the nature and chemical

156 NENITZESCU AND CONSTANTINESCU : THE COMPOSITION OP

composition of the noil-aromatic portion of the sample. The coefficients mentioned in literature which serve to determine the percentage of aromatics (1-15 for a fraction boiling at 60-95° C .; 1'20 for a fraction of boiling point 25—122° C ) were determined for wide gasoline fractions containing a large number of compounds. The application of these coefficients to narrow fractions such as ours would have introduced an element of inaccuracy which we do not believe to be less than tha t encountered in the 98 per cent, sulphuric absorption method. I t is true that this difficulty might be over­come in a large measure by determining some of the aniline points for each of our aromatic-free fractions, by adding the pure aromatic hydrocarbon and by determining the aniline point afterwards. This procedure would, however, have greatly complicated the method, and would have lessened much of its practical value. Therefore, we chose for the estimation of aromatic hydrocarbons in our narrow distilling fractions the method of Dänäilä and his co-workers, which is, likewise, very expedient.

With regard to the errors which may be made by the application of Dänäilä’s method, we have observed the following : if aromatic-free gasoline fractions are treated exactly according to this method with a 98 per cent, acid, we observed an absorption of approximately 1-0-1-2 per cent. In the case of the Bucsani gasoline, the 92-93° C. fraction had, before and after treatment with sulphuric acid according to Dänäilä’s method, the same density and refractive index as before. Consequently this fraction does not contain any aromatic hydrocarbons. I t still shows, however, an absorption of 1-2 per cent, in sulphuric acid when treated by Dänäilä’s method, even when avoiding with care any heating during the first minutes of the shaking. This applies also to fractions 93-91, 94-95, 118-119 and 119-120. I t follows, therefore, that all the values for the aromatic hydro­carbons obtained by us are about 1 per cent, higher than the real figures. As these remarks were made only at the conclusion of this study, we did not attempt to make any revision in this sense.

E s t i m a t i o n o f N a p h t h e n i c H y d r o c a r b o n s w h e n M i x e d i v i t h

P a r a f f i n s .

After estimating the aromatic hydrocarbons by Dänäilä’s method, the sulphurimeter is emptied in a separating funnel, the acid removed and the fractions are washed and dried over metallic sodium. A series of fractions were obtained in this way containing only paraffins and naphthenes. The fractions obtained by previously distilling the aromatic-free benzines showed the same composition. The specific gravity and refractive index were then determined for these fractions by the methods outlined above.

For the estimation of the amount of naphthenes contained in these fractions, we had two methods at our disposal: th a t of the aniline point and that of specific refraction.

The first-mentioned method is too inaccurate to be considered for narrow fractions. Data published in various papers on the subject 14,17,28,25 show that some paraffins have very similar boiling points, yet widely differing aniline points.

On the other hand, the classic method of specific refraction is very useful m this case, as the substances concerned may be mixed together without

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 157

contraction of volume, so that the additivity rule becomes applicable in its entirety. The specific refraction of the fraction may be calculated on the basis of the specific gravity and the refractive index by the well- known Lorentz-Lorenz formula, and is a constant quantity independent of temperature. The specific gravity and the refractive index may be determined with great accuracy, and specific refraction is, therefore, in a large measure free from experimental error.

The specific refraction of the mixture and tha t of the paraffins and naphthenes contained in this mixture being known, the percentage naph- thenes by weight may be obtained from the formula :

33 .

Naph. * = lO o g g

in which Rf, Rp and Rn are the specific refractions of the analysed fraction, and of the paraffins and naphthenes contained in the respective fraction.

The specific refractions of the pure substances may be calculated from the molecular refraction by dividing same by the molecular weight of the respective substance. Likewise, molecular refraction may be calculated from the atomic increments,27 which, being determined for a very large number of substances, offer an incomparable degree of accuracy compared with other constants, such as, for example, critical solution temperature in aniline.

Gasolines with a boiling point up to 200° C. contain only monocyclic naphthenes with a general formula CnH 2„. These have a constant specific refraction equal to 0-3296, independent of molecular weight.

Thus, two out of the three factors of the above formula may be deter­mined with great accuracy. Unfortunately, the specific refraction of paraffins contained in a wide fraction of gasoline cannot be given with the same precision, as in the case of paraffins specific refraction varies with molecular weight. This is, doubtless, the reason why this method is not more widespread to-day.*

How-ever, in the case of narrow and well-separated fractions, such as those with which we have worked, the specific refraction of the paraffins may be indicated with sufficient accuracy. Thus, if we examine the distillation curve of the Merisor gasoline, we see th a t it is composed of a series of maxima separated by a series of well-defined minima. Minima are thus encountered a t 75° C., a t 103° C., and again at 128° C. Taking the boiling points of various paraffins into consideration, we may infer that in the interval between 50° and 75° C. only isomeric hexanes are encountered, between 75° and 103° C. only isomeric heptanes and from the latter temperature to 150° C. only isomeric nonanes.

In the case of the other distilled gasolines, we encounter minima a t the same temperatures, so th a t the intervals admitted above for the distillation

* Marder,26 in a comprehensive study concerning the analysis of gasoline by physical methods, attempts to avoid this disadvantage by measuring the average molecular weight of gasolines and determining by a graphic method the mean specific refraction corresponding to the mean molecular weight. Th is method naturally loses all accuracy with fractions boiling over somewhat wider limits. In calculating the percentage of naphthenes, Marder uses a formula similar to that shown above, the percentage of naphthenes being, however, expressed in volumes.

M

NENITZESCTT AND CONSTANTINESCTT : THE COMPOSITION OF

3 -BUCSANI GASOLINE

li

ii

ii

fi

li

ii

ii

ii

ii

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 159

of hexanes, heptanes, octanes and nonanes are the same for all gasolines, if the distillation column has sufficient separating power.

Consequently the above-mentioned formula may be applied by making use of the specific refractions of paraffins calculated from Eisenlohr’s increments, as shown in Table III .

T a b l e I I I .

Specific Refractions of Paraffins (R p ) Contained, in Various Gasoline Fractions.

Fractions. + 2* Rp.

50-75 CrH u 0-346375-103 c 7h 16 0-3450

10 3-128 Cs-ELs 0-3431128-150 O9H 20 0-3416

On examining the distillation curves of the gasolines, and by taking into consideration the boiling points of pure paraffins,28 it is apparent that it is not possible to find appreciable quantities of hexanes in the maxima above 75° C. and of heptanes below tha t temperature. Similarly, heptanes cannot be found in the maxima above 103° C. nor octanes within the maxima below that temperature. Uncertainty exists only with regard to the nature of the paraffins in the fractions corresponding to the minima of the curve—for example, fractions 74-76, 102-104, etc. The paraffins in these fractions may be hexanes mixed with heptanes, heptanes with octanes, etc. As these fractions, however, represent only a small quanti­tative percentage of the total gasoline, the error resulting from the uncer­tainty regarding the composition of these fractions, and from arbitrary choice of the limits shown in Table III , becomes negligible.

E x a m p l e s o f C a l c u l a t i o n .

As shown above, the percentage of hydrocarbons may be determined by the method of specific refractions by weight, and we were thus forced to compute all our results obtained in terms of weight. For this reason we consider as specific gravity of aroma tics distilling up to 86° C. (benzene) = 0-8736, of those distilling up to 120° (toluene) = 0-8660, and of those dis­tilling up to 143° (ethyl benzene, o-, m-, and p-xylenes) = 0-8680 whilst those distilling between 133° and 150° C. (in which o-xylene predominates)=0-8790.

Likewise, the to tal weight and mean specific gravity of the sum total of distilled fractions were calculated on the basis of the volume and specific gravity of each fraction.

Gasoline of : Total Volume, ml.

Total Weight, gms.

d T,average.

Meritor 5871 4357 0-7421Bucsani 4519 3318 0-7343Buesani (aromatic-free) 4438 3206 0-7224Jura-Ocnfiei (aromatic-free) . 5589 4135 0-7398

16 0 NENITZESCU AND CONSTANTINESCU : THE COMPOSITION OF

We give below as an example the calculation for the fraction 97-98' C. of Bucsani gasoline. I t had the following constants: == 0*/033;h20 = 1-39673 and Ar per cent. vol. = 1-3 per cent. After removal of aromatics D f = 0-7027 and «5 = 1-39624. The specific refraction ofthe aromatic-free fraction is therefore 0-3421.

The percentage volume of aromatics were transformed into weight as follows :

aromatics % vol. X density of aromatics _^ — specific gravity of original fraction

1-3 X 0-8660- W i i m — = 1-6 per cent.

The percentage, by weight, of naphthenes in the aromatic-free fraction is, according to the above reasoning :

The percentage, by weight, of paraffins in the same fraction amounts to : = 100 - 18-8 = 81-2.

Referring these figures to the original fraction we have :(100 — A) X IVj (100 - 1-6) x 18-8

” 100 100 p _ (100 - A) X P x _ (100 - 1-6) X 81-2 = ?g g

100 100

This type of calculation was used throughout for each separate fraction. The calculation is, naturally, simplified in the case of fractions obtained from aromatic-free gasoline.

In order to have comparative figures, the results were referred to the total fractions distilling between 50° and 150° C. as shown above, in the follow­ing manner :

Fraction 97-98 of Bucsani gasoline had a vol. = 227 ml. and D = 0-7033. The weight of the fraction is 227 X 0-7033 = 159-6 gm. Referred to the total distillate, this fraction represents :

Q = 100 = 4-812 per cent.

If we represent the contents in aromatic, paraffinic and naphthenichydrocarbons referred to the total distillate as QA, QP and Q^, we obtain:

n 159-6x1-6_ ----3318----- = '' Per cent-

n _ 159-6 x 79-9 0v p ------= 3-845 per cent.

n 159-6 x 18-5 „ „w — 33l8-------~ Per cenf-

Q = 0-077 + 3-875 -f 0-890 = 4-812,

n

iM

’’i ■i i JU

* *!S

4. - BUCSANI GASOLINE, DE-AROMATIZED

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 161

Paraf finie Hydroa/im ü t Piphtmicdi/droaríxxn ■z£l .Density

Pefredive index

' t :

' I

ïTfï

ïïT

Tïn

Refra

ctive

index

n„’°

From the above it follows that for each fraction a table has been drawn up containing the following data, for the fraction 97-98 of Bucsani gasoline :

162 NENITZESCU AND CONSTANTINESCU : THE COMPOSITION OF

T a b l e IV .

Fraction.

Original.

ml. Gm. Q- D f .4 " d - A r, % vol.

97-98 227 159-6 4-812 0-7033 1-39673 1-3

Aromatic-free :Composition of the

Fraction, % by weight.

Percentage of the Total Distillate.

D f. rfi°.D Rj. N x. P i- A . N . P . Qa- Qn- Qp-

0-7027 1-39624 0-3421 18-8 81-2 1-6 18-5 79-9 0-077 0-890 3-845

In order to avoid making our present paper too long, and in view of the fact that figures for all fractions have been published in extenso in Mr. Constantinescu’s thesis, we shall not reproduce here the tables.

D i s c u s s i o n o f t h e R e s u e t s .

An examination of the distillation curves will show tha t each fraction is composed of hydrocarbons belonging to two or three classes, respectively.

Aromatic hydrocarbons are distributed over wide ranges of temperature, with the result that large concentrations of aromatics are not obtained in any fraction. On the other hand, paraffinic and naphthenic hydrocarbons distil at much closer intervals.

A maximum concentration of aromatic hydrocarbons is usually found in fractions boiling a few degrees below the pure aromatic hydrocarbons.

The fact that aromatic hydrocarbons are distributed over a wide range of temperature results, likewise, in a disturbing effect on distillation of the saturated hydrocarbons which are partly carried away by fractions having nearly the same boiling point. This fact clearly follows from the distillation curves, which in the case of gasolines containing aromatics show less pronounced minima than in the case of aromatic-free gasolines.

For this reason we abandoned the distillation of gasolines containing aromatics, with a view to distilling in future only aromatic-free gasolines, as it is much more precise.

The proportion of heptane and methylcyc/ohexane in the Bucsani gaso­line is very typical, as illustrating the reciprocal influence exerted by paraffins and naphthenes with close boiling points. Both hydrocarbons which occur in almost equal quantities distil mostly a t an interval of 3° C., from 97 to 100 C. The distillation curve shows two maxima : one at 97-98 corresponding to a fraction rich in n-heptane, and another at 99-100 C. rich in methylcycZohexane, and containing, likewise, n-heptane.

This example will suffice to show, therefore, the separation limit of our

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 16 3

column. Effective separation can be achieved only by redist illing the fractions, which would necessitate, however, much larger quantities of gasoline. Since, however, our object is to obtain as uniform fractions as possible, and not the separation of individual compounds, in order to enable the application of the analytical methods used, the efficiency of our column is quite sufficient.

In order to estimate the percentage of paraffinic and naphthenic hydro­carbons, we have adopted as basis the intervals shown in Table V, established on the basis of the maxima and minima of the distillation curve and the boiling points of the pure hydrocarbons. I t is to be noted that in the case of all distilled gasolines the maxima and minima of the curves meet a t the same points : i 1° C. I t is thus possible to determine the intervals for the various hydrocarbons as in Table V, since each interval comprises a maximum accompanied by clearly discernible minima.

T a b l e V .

BoilingIntervals. Nature of Hydrocarbon.

Paraffins50-65 isoHexanes(2- and 3-methylpentane).65—75 n-Hexane.75-95 isoHeptanes(2- and 3-methylhexanes).9.5-103 n-Heptane.

103-120 ¿soOetanes (a dimethylhexane in fract. 10 3 -110 ° C. and 2- or 3-m ethyl-

120-128heptane in fract. 1 17 -12 0 ° C . ).

n-Octane.128-145 isoNonanes.145-150 n-Nonanes.

Naphthene50-57 cycZoPentane.65-75 Methy IcycZopentane.75-85 cycZoHexane.85-95 DimethylcycZopentanes.95-104 Methylcyctohexane.

10 4 -113 Naphthene non-identified (ethylcycZopentane, trimethylcycZopentanes).113 -12 6 DimethylcycZohexanes.126-133 EthylcycZohexane and a nononaphthene with a 5-carbon cycle.133-145 Nononaphthene in which trimethylcycZohexanes predominate.

Table V was established on the basis of the above considerations, and gives the composition of the respective gasolines. The figures in brackets corresponding to the original benzines represent the percentage of hydro­carbons of the respective class calculated in relation to the bulk of saturated hydrocarbons contained in the gasoline. These figures are, therefore, comparable with those found in fractions derived from aromatic-free gasolines.*

* It will be seen that in the case of the Bucsani gasoline, which was distilled, both in the aromatic-free state as well as with the aromatics, it contained the figures obtained for paraffins, isoparaffins and naphthenes show differences ranging from 1 to 2 per cent. Most of these differences are doubtless due to systematic errors during the estimation of aromatics.

/.jęto ; 11900 "g

<s/oo Q:Arrío/.i/OO

P a r a f f i n i c H y d r o c a r b o n s

N a p h th e n ic H y d r o c a r b o n s

D e n s i t y ( 0/ aromatk-fn* frwctknt)

R e f r a c t i v o In d e x

NENITZESCU AND CONSTANTINESCU : THE COMPOSITION OF

5 - C U R A O C N I T E I G A S O L I N E , D E - A R O A 1 .A T IZ E D

T a b l e V I .

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 1 65

Composition of Fractions 50-150° C. of Gasoline Derived from :

Merisor, O.N. = 63.

Bucsani, O.N. = 51-5.

Aromatic- free

Bucsani, O.N. ’= 48.

Aromatic- free *

Gura- Ocnitei,

O.N. = 75.

n-Paraffins :n-Hexane 5-4 5-3 6-5 30n-Heptane 4-8 8-3 8-8 3 1n-Octane 3-2 3-2 8-1 _n-Nonane ( — 150°) . 1-2 1-2 6-0 —

Total 14-6 (15-8) 23-9 (31-2) 29-4 6 1

¿soParaffins .-isoHexanes 5-4 5-1 5-7 9-4¿soHeptanes 6-9 7-7 8-6 10-9isoOctanes 6-4 9-1 10-3 9-6iaoNonanes 5-2 8-4 9-2 —

Total 23-9 (*25-9) 30-3 (34-4) 33-8 29-9

Naphthenes :cycZoPentane 0-4 0-1 0-1 0-2MethylcycZopentane . 3-7 2-0 2-4 7-1cycZoHexane 3 5 2-7 3-6 9-1DimethylcycZopentane 60 2-7 3-4 10-8Methylcycfohexane 9-3 7-3 9-2 20-4Naphthene tract. 10 4 -113 ° C. 3-6 1-2 1-5 4-9DimethylcycZohexane 1 1 0 5-7 7-0 11-6Naphthene tract. 126 -150 ° C. 1 6 1 8-4 9-6 —-

Total 53-6 (58-2) 30 1 (34-3) 36-8 64-1

Aromatics :Benzene . . . . 1- 7 2- 0 — _Toluene . . . . 2- 1 3- 6 _ _Ethylbenzene. Xylene 4-1 6-5 — —

Total 7-9 12-1 — —

* 7 -5 % aromatic hydrocarbons—Gura-Ocnitei gasoline.

We see, therefore, tha t the percentage of isoparafnns shows a variation of ± 5 per cent, and amounts to approximately 30 per cent. The per­centage of naphthenes shows large variations. Whilst the proportion of naphthenes is about 60 per cent, for the two asphaltic gasolines, it is only 37 per cent, in the case of the gasoline of paraffinic origin.

The disturbing influence of normal paraffins on the octane number is clearly evident: whereas gasolines with a high O.X. contain only 6 per cent, of «-paraffins, paraffinous gasolines (with O.X. = 48) contain 30 per cent, of «-paraffins. The wide difference of the O.X. for the Merisor and Gura Ocnitei gasolines is solely due to a comparatively small difference (8 per cent.) in the contents of «-paraffins. The proportion of iso- and normal paraffins in Bucsani gasoline is roughly equal to 1 (33-8 : 29-4); in the Meritor gasoline it is 1-6 (25-9 : 15-8) and in the high-octane Gura Ocnitei gasoline it is 5 (29-9 : 6-1).

NENITZESCU AND CONSTANTINESCIT : THE COMPOSITION OF

With regard to the nature of ¿soparaffins, examination of the distillation curves will show that in gasoline the predominating derivatives are those with a single side chain (methyl) and, namely, 2- or 3-methylhexane, or a mixture of both 2 : 3 - or 4-methylheptanes, or a mixture of them, etc. The percentage of isoheptane together w'ith dimethylcycZopentane in all fractions from 89 to 91 is considerable in all gasolines. The proportion of isoparaffins with several side-chains is smaller. Such compounds are found in fractions which boil a t approximately 80° C. (an isoheptane), 107° C. (iso-octane) and 135° C. (isononane). No hypothesis is put forward here as to their structure.

Naphthenes are found throughout the 50-150° C. interval, w i t h t h e exception of the range from 55° to 66° C., which is free f r o m n a p h t h e n e s in all the gasolines investigated. A small proportion o f c y c Z o p e n ta n e is found in all gasolines. Methylcyc/opentane, cycZohexane, methylcyclo- hexane and dimethylcycZohexane are also surely found, as s h o w n i n t h e diagrams. The last two naphthenes are, quantitatively s p e a k i n g , t h e most important hydrocarbons in gasoline. MethylcycZohexane c o n s t i t u t e s 20 per cent, of Gura Ocnitei gasoline (fraction 50-123-5) a n d a b o u t 10 p e r cent, of the other gasolines.

In conclusion, it may be said that the three gasolines i n v e s t i g a t e d i n o u r laboratories possess very similar distillation curves. The m a x i m a a n d minima of the curves are found at the same points. The g a s o l i n e s d if fe r only in the percentage of the various components contained in the m ix t u r e . In the paraffin series normal hydrocarbons predominate, o r t h o s e w i t h a single side-chain; in most cases the side-chain i s p r o b a b l y c o m p o s e d o f methyl groups. Paraffins with several side-chains o r w d th c o m p l i c a t e d ramifications exist only in unimportant concentrations. In t h e n a p h t h e n i c series up to the mononaphthenes, all possible homologues a n d i s o m e r s a re represented. Among these, derivatives of cycZohexane a r e c o n t a i n e d in larger proportions than those of cycZopentane.

Laboratory of Organic Chemistry,Polytechnic “ King Carol I I ,”

Bucharest.

Literature Cited.

1 F. D. Rossini, Refiner, 1935, 14, 255.2 M. R. Fenske, D. Quiggle and C. O. Tonberg, Ind. Enq. Chem., 1932, 24, 542, 814;

1936 ,28,201,644; 1937,29,70.3 M. R. Fenske and co-workers, ibid., 1932, 24, 4 12 ; 1934, 26. 1169; 1936, 28, 645.4 S. T. Sehicktanz and J. H . Bruun, E at. Bur. Stand. J . Res., 1931, 7, 851 ; See also

Ref. 6.« ™ Bromiley and D. Quiggle, Ind. Eng. Chem., 1933, 25. 1136.

M. R. Fenske, C. O. Tonberg and D. Quiggle, ibid., 1934, 26, 1169.H . A. Beatty and G. Calingaert, ibid., 1934, 26, 504

9 Mareusson, Chem. Ztg., 1909, 33, 987; Halphen, M atières Grasses, 19 11, 3, 1987;R. Zaloziecki and Haussman, Z. angew. Chem., 1906, 20, 1761 ; D. Florentin andVanderberge, Bull. Soc. chim., 1920, 27, 204; M. Heyn and Z. Dunkel, Brennstofi-Chem., 1926, 7, 245; G. Egloff and J. C. Morrell, Ind. Eng. Chem., 1926, 18. 354;

inoon,m J ' lnsL PetroL Techn., 1928, 14, 695; A . B . Manning, J . Chem. ooc., ly#29. 1014.10 w Blesenfeld ancl G - Bandte, Erdôl und Teer, 1926, 2. 491.

2 18 ara er’ G' Morrell and J. M. Levine, Ind. Eng. Chem. Anal. Ed., 1930,

SOME ROUMANIAN STRAIGHT-RUN GASOLINES. 1 67

11 ß . Krämer and W . Böttcher, Ber., 1887, 20, 595; W . Markownikoff, Annalen, 1886, 234, 86; L . Edeleanu and G. Gane, Öster. Chem.-Techn. Ztg., 1889, 830, 850; Engler-Höfer, Das Erdöl, 1913, 1 , 353.

22 N. D. Zelinsky, Ber., 1912, 45, 3679.13 F . B. Thole, J . Soc. Chem. Ind., 1919, 38, 3 9 t .11 H . T . Tizzard and A. G. Marshall, ibid., 1921, 40, 2 0 t .15 N. D än äili, A. V . Andrei and E . Melinescu, Bul. chim. Soc. romdnä de Stiinte,

1923, 26, No. 4—6; 1924, 24, No. 4 -6 ; N. Dänäilä and V . Stoenescu, ibid., 1926,29, No. 1 -3 ; N. D ä n iilä , Th. Ionescu and R . Verona, ibid., 1932, 35, No. 1 -6 ; R . Verona, ibid., 1934, 35, No. 1-6 .

16 G. Morgan and R . P. Soule, Ind. Eng. Chem., 1923, 15, 587.17 A. N. Sachanen and R . Wirabianz, Petroleum, 1929, 25, 867.18 S. Senkte, Engler-Höfer, Das Erdöl, 1913, 1 , 231.19 E . Cazimir, C. Creanga and M. Dum itriu, Petroleum, 1930, 26, 617.20 A. B. Manning, J . Chem. Soc., 1929, 1014; A. B. Manning and M. Shepherd,

Brit. Chem. Abstr., B, 1931, 284.21 R. Katwinkel, Chem. Ztg., 1935, 49, 57; Brennstoff-Chem., 1927, 8, 353.22 W. N. Hoyte, J . Inst. Petr. Techn., 1925, 1 1 , 76.23 W. Ormandy and E . C. Craven, ibid., 1929, 13, 76, 3 11, 846.24 M. D. Tilitschejew and A. T . Dum skaja, ibid., 1929, 15, 465.25 G. Chavanne and L . Simon, Compt. rend., 1919, 168, 1 1 1 1 ; Bull. Soc. chim. Belg.,

1922, 31, 331.26 M. Marder, Kohle, Erdöl und Teer, 1935, 18, No. 1, 3, 5, 9, 11 , 1927 F . Eisenlohr, Z. physikal. Chem., 1910, 75, 585.

T H E R A PID D E T E R M IN A T IO N O F SO LU BLE B ITU M EN IN RO A D C A R PETS.

By L. J. C h a l k , M.Sc., A.I.C. (Member).

S u m m a r y .

A detailed account is given of two rapid but reliable methods employed by the author for the determination of soluble bitumen in road carpets.

The first consists of a hot extraction process in which a tall metal beaker replaces the usual Soxhlet form of apparatus. The sample is placed in a wire gauze basket suspended inside the beaker and loss of solvent is pre­vented by closing the top of the beaker with a condenser made from lead compo’ tubing. The bitumen is determined by difference.

In the other method, a known volume of solvent is added to the sample and the mixture agitated until the bitumen has completely dissolved. An aliquot portion of the solution is then evaporated to dryness and the bitumen determined by direct weighing.

The necessary precautions and manipulative technique are described in detail in each case. Both methods are extremely simple in character but give results in good agreement with the British Standard method.

A method of determining by wet sieve analysis the proportion of material passing 200 British Standard mesh in the recovered mineral aggregate is also described.

F e w chemists engaged in the asphalt industry have not experienced t h e need for a simple rapid and reliable method for determining soluble bitumen in road carpets, and particularly is this the case with those employed in t h e laboratory supervision of road constructional works. The procedure recommended by D. M. Wilson,1 and adopted by the British Standards Institution,2 is often unsuitable where results are urgently required, owing to the length of time necessary for the filtration. Asphalt cements con­taining large amounts of Trinidad Lake Asphalt and carpets incorporating these cements are notable in this respect. The difficulty is further accen­tuated where the size of the coarse aggregate necessitates the use of a large sample. For certain routine purposes, it may be possible to dispense with the determination of the percentage and sieve sizes of the coarse aggregate, and save time by taking comparatively small samples, but the procedure is obviously inapplicable where the sieve analysis of the recovered mineral aggregate is required.

I t is not surprising, therefore, that many rapid methods of determining the soluble bitumen have been developed. In some, no attem pt has been made to secure a very high degree of accuracy and the object has been simply to detect the major discrepancies from specification and enable appropriate remedial methods to be applied without loss of time. Never­theless, a number of rapid methods is available, which can be relied upon to give results in close agreement with the standard m ethod; these fall into two main classes :

(i) Hot extraction methods utilizing solvents such as trichlor- ethylene, perchlorethylene and benzene, in which the bitumen is usually found by difference.

SOLUBLE BITUM EN LN ROAD CARPETS. 1 69

(ii) Cold extraction methods with similar solvents, but employing a direct bitumen determination.

(i) H o t E x t r a c t io n M e t h o d s .

Apparatus and Method for Hot Extaction of Bituminous Aggregates is described by Hubbard in Laboratory Manual for Bituminous Aggregates, 1916. This method, using the New York Testing Laboratory Extractor, has been widely used and consists of a cylindrical brass vessel for bolding the solvent, inside which is suspended a cylindrical wire basket of 80 mesh wire cloth; an inverted conical condenser is fitted in the top so as to ensure that the vapours are condensed and percolate through the bituminous aggregate contained in the wire basket.

A well-known method belonging to this group is that described by D. M. Wilson,3 which embodies a filtration under reduced pressure through a Kieselguhr filter cylinder of the Berkefeld type. The analysis may be completed comfortably in a working day and is suitable for the simultaneous extraction of a number of samples of road carpet. I t does not appear to have been extensively adopted, probably because many works laboratories do not possess the requisite compressed air and vacuum lines.

A number of other forms of hot extraction apparatus are based on the principle of the Soxblet extractor. The filter thimble is frequently re­placed by a wire-mesh basket to avoid difficulties in adjusting the rate of refluxing to suit filtration. With the type of road carpets laid in this country, it has been found that if the filter paper is close enough to retain the fine mineral m atter, the rate of filtration is usually slower than the rate of refluxing, with the consequence tha t fine mineral matter is liable to be washed over the top of the paper. Whatever type of apparatus is used, it is desirable th a t the dim ensions of the extractor should be large enough to take samples weighing 1 to 2 kgms. A Soxhlet extractor of this size is necessarily an expensive item of equipment and is unlikely to find favour in laboratories where numerous soluble bitumen determinations are carried out.

The extractor developed a t these laboratories (Fig. 1) described below does not suffer from this objection. I t may be easily and cheaply con­structed and is suitable for all grades of stone-filled asphalt. For mastic asphalt, a slightly modified form, described in a later section, is recommended.

Extractor for Stone-filled Asphalt.The extractor for stone-filled asphalt consists of an aluminium or stainless

steel beaker, 6 ins. diameter and 12 ins. high, inside which is suspended a cylindrical gauze basket, 5 | ins. diameter and 6 ins. high. Loss of solvent during extraction is prevented by placing on the top of the beaker a con­denser made from a spiral of lead compo’ tubing.

The basket may be easily constructed from a cylindrical tin 5 | ms. diameter with a press-on f id as follows :—three circular holes 11 ins. diameter are cut in both fid and base and 85 B.S. wire gauze soldered over the holes. The major part of the cylindrical portion of the tin is then cut away leaving only a ¿-in. rim attached to top and bottom. These two pieces are joined together with four brass strips \ in. w id e and 6 in s . long and the basket

1 7 0 CHALK : THE RAPID DETERM INATION OF

completed by soldering 85 B.S. wire gauze around the outside of the frame- work If desired, the top portions of two tins may be employed and a reversible basket made with a lid at the top and bottom. The basket is suspended by a wire passing around the rim of the beaker and attached to two hooks on the top of the basket.

The condenser is made from lead compo’ tubing J in. external diameter wound in the form of a spiral and wired to a disc of perforated zinc. The spiral is made to fit inside the beaker and is provided with a few extra coils on the outside rim to give it the shape of a shallow trough.

Method for Stone-filled Asphalt.A representative sample weighing a t least 1 kgm. is obtained by the usual

quartering procedure and dried in an oven a t 105° C. The sample is then

SOLUBLE BITUM EN IN BOAD CABPETS. 171

introduced into the weighed basket. Approximately 500 ml. of trichlor- ethylene are transferred to the beaker and the basket suspended by wire about 1 in. above the surface of the liquid. The condenser is placed in position and the solvent maintained in a state of steady ebullition for 4 to 5 hours, after which time the apparatus is allowed to cool and the contents of the basket tipped out on to a shallow tray and placed in a warm position until the odour of solvent has disappeared. The basket and mineral aggregate are then dried in an oven at 105° C. for 1 hour and weighed.

The solvent contained in the beaker is carefully decanted into a 1 litre graduated flask and the mineral residue washed several times with tri- chlorethylene allowing 5 to 10 minutes before decanting each portion of wash liquid into the litre flask. The beaker containing the residual mineral matter is then dried and weighed; alternatively, the mineral m atter may be transferred with solvent to a smaller receptacle.

The bitumen solution is diluted with solvent to exactly 1 litre and well shaken. An aliquot portion of 100 ml. is immediately measured into a graduated flask and transferred together with rinsings to a weighed silica dish. The solvent is evaporated as far as possible on a water bath, after which the dish is placed over a low bunsen flame and the temperature gradu­ally raised until all carbonaceous m atter has been removed. Precautions are taken during this process to prevent the bitumen from catching fire.

Finally, the residue is recarbonated and the weight of ash multiplied by the appropriate factor. The quantity of ash recovered by this process does not exceed 1 per cent, of the sample and for the purpose of computing the sieve analysis of the mineral aggregate it may be assumed to consist entirely of material passing the 200 mesh B.S. sieve.

The weight of bitumen in the sample is obtained by subtracting the total recovered aggregate from the weight originally taken for analysis.

A centrifuge may be employed if desired to separate the mineral m atter from an aliquot portion of the bitumen solution. Such separations are, however, rarely complete, and the precaution of determining the ash con­tent of the centrifuged aliquot should not be neglected. The centrifuge method is recommended when the sample contains organic m atter insoluble in the usual solvents for bitumen, as is the case when certain native asphalts are present or mineral fillers which undergo considerable change in weight on ignition and recarbonation. The error arising from these causes is, how7ever, comparatively small, and the soluble bitumen figure is unlikely to be more than 0-3 per cent, in error, even when the centrifuge process is not employed.

Extractor for Mastic Asphalt.A basket made from 200 B.S. wire gauze is employed for mastic asphalt.

I t is constructed in a similar manner to the basket used for rolled asphalt, except that the interior of the basket is divided into four compartments by means of gauze discs supported on a central brass rod. The discs may be constructed from tin lids by cutting out 3 or 4 circles or sectors and covering the holes with wire gauze. The central brass rod is made in 4 tapped and screwed sections, each 1 | ins. long. A somewhat similar form of basket has been employed a t the Road Research Station, Harmondsworth, Middle­sex with very successful results.

172 CHALK : THE RAPID DETERM INATION OF

The sample is warmed and pulled apart into pieces as small as possible and introduced into the basket as follows :—the first section is screwed to the bottom of the basket and approximately quarter of the sample added. The first disc and second section of rod are fitted into position and another portion of sample transferred to the basket. The process is repeated with the second and third discs and subsequent procedure is the same as that described for rolled asphalt, except th a t a longer period of extraction is usually necessary. If circumstances permit, the extraction may be allowed to proceed overnight.

The foregoing methods give results in good agreement with those for the standard procedure described in B.S.S. 598.4

Some typical figures obtained with samples of steam-rolled and mastic asphalts are shown in Table I.

T a b l e I.

Soluble bitumen.

Sample. A .C.Hot extrac­ B.S.S. 598,

tion, per cent. per cent.

Steam-rolled asphalt Fluxed epure 8-8 8-5Asphaltic bitumen 7-4 7-2

,, 7-7 7-5Mastic asphalt i i i f 12-7 12-7

Fluxed epure 12 0 11-8a a Asphaltic bitumen 9-4 9-4

12-7 12-7a a i i i i 1 1 0 11-2

(ii) C o l d E x t r a c t i o n M e t h o d s .

A feature of the majority of these methods is that a known volume or weight of solvent is added to the weighed sample and after taking suitable precautions to secure complete solution of the bitumen and the removal of the bulk of the mineral matter, an aliquot portion of the solution is evapo­rated to dryness. The automatic burette method described by D. C. Broome,5 and designed for use with mastic and rock asphalts follows these lines, as does that put forward by I. Hvidberg.6 The method described below is based on similar principles, but incorporates certain modifications which give improved accuracy and renders the method of more general utility. The method is suitable for all types of bituminous road carpets and can be recommended for routine control purposes. I t is extremely simple, requires no elaborate apparatus and can be completed in approxi­mately 3 hours.

Method. (Applicable to B.S.S. Nos. 347, 348, 594, 595, 596 & 597.)The sample is warmed in an oven until soft and, if present, precoated

chippings are removed and discarded. The sample is then pulled apart into pieces as small as possible and quartered to the approximate weight git en in Table II. A sampling procedure is employed and no attempt is made to adjust the weight of the sample to the exact figure shown in the table.

SOLUBLE BITUM EN IN ROAD CARPETS. 173

T a b l e I I .M inim um Weight of Sample.

Mastic asphalt, ungritted . . . . . . . 100 gms.Mastic asphalt containing aggregate passing 1 in. . . . 100 ,,Mastic asphalt containing aggregate J—f in. . . . 250Sand carpet . . . . . . . . . . 100Rolled asphalt containing less than 25 per cent. } in. stone . 500Rolled asphalt containing more than 25 per cent. \ in. stone . 1000Rolled asphalt, binder course . . . . . . . 1000Compressed rock asphalt . . . . . . . 100 ,,

The sample is dried in an oven a t 105° C. and weighed. I t is then placed in a glass bottle or metal drum and a measured quantity of solvent (Table III) added from a graduated flask, allowing 30 seconds for the flask to drain. The solvent may be carbon disulphide, chloroform, perchlorethylene or trichlorethylene.

T a b l e I I I .

Sample weight. Volume of solvent.

100 gms. 1000 ml.250 „ 2000 „500 „ 3000 „

1000 „ 5000 „

The temperature of the solution is read, after which the container is vigorously shaken for 30 seconds and then set aside and re-shaken four more times at quarter-hourly intervals; finally the solution is allowed to remain undisturbed for a further quarter of an hour.

Approximately 50 ml. of bitumen solution are removed and centrifuged for 2 minutes a t 2000 to 4000 r.p.m., the average radius of the path of the liquid in the centrifuge tube being 4 |" . The centrifuge tubes are provided with tin-foil caps to minimize loss of solvent by evaporation. A portion of the centrifuged liquid is decanted into a beaker and the temperature quickly adjusted to its former value. 10 ml. of solvent are then transferred with a pipette (calibrated for use with the particular solvent) into a weighed dish. The solvent is evaporated on a water bath and the residue dried a t 160— 170° C. for half an hour, or a t 105° C. for one hour and weighed. The residue is ignited and the ash recarbonated. In calculating the soluble bitumen content, an allowance is made for the fact that the volume of bitumen solution is always slightly greater than the volume of solvent.

If A = weight of bitumen in aliquot,B = weight of sample,C = volume of aliquot,D = total volume of solvent,E = weight of ash in aliquot,

then the volume of bitumen solution = D ^l +

a j. 100 A D ( . .ana percentage soluble bitumen = — — (1 + q )

100 EDThe percentage ash given by — — should not exceed 1 per cent.EL/

N

1 7 4 c h a l k : t h e r a p i d d e t e r m i n a t i o n o f

If a centrifuge is not available, the container after shaking in the manner described above may be allowed to stand overnight and 10 ml. of the solution withdrawn without disturbing the mineral m atter a t the bottom of the container. Should the ash exceed 1 per cent., another 10 ml. aliquot is diluted to 150-200 ml. and filtered through a No. 5 Whatman, 11-cm. filter paper. The filtrate and washings are then transferred to a weighed dish and evaporated to dryness and ashed. The extra time involved in this procedure amounts to approximately 2 hours. The same method may be used in conjunction with a centrifuge process if so desired.

T a b l e IV .

Soluble bitumen.

Sample. A.C. Cold extrac­tion, per cent.

B .S.S. 598, per cent.

Mastic asphalt Asphaltic bitumen 10-09-4

9-99-2

9 9 9 9 9-3 9 19 9 9 9

Steam-rolled asphalt9-4 9-4

9 9 9 9 9-4 9-2Mastic asphalt 9 9 9 9 15-7

14-015-613-9

15-4 15-5,, Fluxed epure

9 9 9 9

15-2 15 09 9 9 9 16-0 15-7

The results given in Table IV were obtained using trichlorethylene as a solvent. The figures quoted are in close agreement with those obtained by the standard method.

When a sieve analysis of the mineral aggregate is required, the bitumen solution remaining after the determination of soluble bitumen is agitated and passed through an 8- and a 200-mesh B.S. sieve, fitted together over an 8-in. funnel. The residual mineral m atter is washed several times with trichlorethylene and finally transferred to the 8-mesh B.S. sieve. Fine particles of filler adhering to the coarse aggregate are removed with a jet of trichlorethylene from a wash-bottle fitted with a blow-ball. The aggre­gate retained on the 8-mesh B.S. sieve is then dried, weighed and sieved in the usual manner. I t is not assumed to consist entirely of plus 8 material, but is re-sieved on the 8-mesh B.S. sieve.

The fine material on the 200-mesh B.S. sieve is washed three or four times with trichlorethylene, then once with alcohol (95 per cent. I.M.S.) and finally with a stream of tap water until the —200 material has been removed. This point is ascertained by examining the water passing through the sieve. I t is an advantage to employ a metal spiral in which a number of small holes have been bored to distribute the water over the entire surface of the sieve, as by this means sieving may be carried out with practically no attention or effort on the part of the operator. When the washing process has been completed, the residue is rinsed with alcohol, dried and weighed. The proportion of material passing 200 mesh is obtained by subtracting the weight of aggregate retained on the 8- and 200-mesh sieves

SOLUBLE BITUM EN IN KO AD CAKPETS. 1 7 5

from the total weight of mineral aggregate. The aggregate retained on the 200-mesh sieve is dry-sieved in the usual manner.

Wet sieving possesses pronounced advantages over the more usual dry sieving and gives more accurate and consistent results. I t is to be hoped that in course of time it will entirely replace the older method. The pro­portion of —200 material found by dry sieving (B.S.S. 598) is invariably low, as filler adheres to the coarse aggregate and is lost when the latter is sepa­rated on the 8-mesh sieve. The error so introduced depends both on the quantity and nature of the coarse aggregate and is frequently very marked. For many years now, it has been the custom to dry sieve mineral aggre­gates and fillers, but the process has little to recommend it. The separation of the —200 fraction by wet sieving not only ensures correct and consistent results, but eliminates an extremely tedious and dusty operation and re­duces the wear and tear on the 200-mesh sieve. I t can be employed in the presence of a Portland cement filler and is particularly useful for soft lime­stone aggregates which are difficult to deal with by the normal method.

In the course of investigating the cold extraction method, attention was directed to certain possible sources of error and a number of experiments were carried out with a view to checking the accuracy of the method and improving the general manipulative technique. These experiments may be grouped under the following headings :—

(а) Correction for solution volume of bitumen.(б) Correction for volatilisation of solvent during the centrifuge process.(c) Calibration of measuring vessels.(d) Procedure for drying residue.(e) Ashing procedure.(/) Check determinations.

(a) Correction for Solution Volume of Bitumen.Weighed amounts of bitumen were added to measured volumes of solvent

and the increase in volume noted. I t was apparent from the results that for the purpose of calculating the correction it would be sufficiently accurate to regard the increase in volume as equal to the volume of bitumen and to take the density of the bitumen as unity. Thus, employing the nomen­clature given above, it could be assumed that the volume of bitumen

ADsolution would be greater than the volume of solvent by —7 r .c

(b) Correction for Volatilisation of Solvent during the Centrifuge Process.A centrifuge tube approximately 2 cms. diameter and 35 ml. capacity

with a hemispherical end was employed in these experiments and was weighed before and after centrifuging. The percentage loss of solvent when the tubes were filled to capacity was found to be somewhat greater than when the tubes were partly filled, but by placing tin-foil caps over the open ends of the tubes, the loss in all cases could be reduced to insignificant proportions. The caps were made by placing a circle of tin-foil about 2 cms. diameter over the top of the tube and smoothing the projecting portion into contact with the outside of the tube. The efficacy of this device is clearly indicated by the figures given in Table V.

1 7 6 CHALK : THE RAPID DETERM INATION OF

T a b l e V .

Solvent.

Loss after centrifuging 2 mins.Room

temp., ° C.W ith. cap. per cent.

W ithout cap, per cent.

Carbon disulphide 0-06 0-60 28-8Trichlorethylene . 002 0-27 » »Chloroform . . . . 003 0-33 , ,

Carbon tetrachloride 0-02 0-28 , ,

Perchlorethylene . 0003 0 1 1 J t

(c) Calibration of Measuring Vessels.Pipettes. I t is usual to calibrate pipettes with distilled water. For the

present work, however, it was considered preferable to calibrate with the particular liquid to be employed and ascertain by direct weighing the ratio of the weight of solvent delivered by the pipette to the weight delivered by the measuring flask.

Measuring Flask. A 1-litre measuring flask filled to the mark with trichlorethylene was found to deliver 999-0 ml. when a drainage time of 30 seconds was allowed. The error in measuring the requisite volume of solvent could, therefore, be assumed to be negligible.

(d) Procedure for Drying Residue.The following experiments were carried out with a view to ascertaining

whether loss of volatile hydrocarbons occasioned by drying at 160-170° C. introduced an error into the analysis. 0-1 gm. of several grades of bitumen and flux oil were transferred to porcelain dishes, dissolved in 10 ml. trichlor­ethylene and the solutions evaporated to dryness on a water bath. The residues were dried either a t 105° C. or a t 160-170° C. and reweighed. The results shown in Table VI suggest that in the absence of volatile flux oils or soft bitumens a drying time of half-an-hour a t 160-170° C. will give satis­factory results. In case of doubt a drying time of one hour at 105° C. is recommended.

T a b l e V I.

Bitumen.

D rying period. Change in weight, mgms.105° C.,

hrs.16 0 -170 ° C.,

hrs.

Mexphalte 65 2 __ + 1-0,, . . . — 1 Nil

Mexphalte 80/90 .— 4 - 20— h - 0-4

Texaco E flux oil . _ I - 1-6Shell flux oil . . . 2 — 2-9

,, »» »» — h - 15-81 6 % Shell flux oil, 8 4 % epure 1 Nil

,, ,, ,, ,, ,, 2 — - 0-6,, ,, ,, ,, ,, — - 1-3’* »» >> M »» 4 — - 1-3” — è - 5-2

SOLUBLE BITUMEN IN ROAD CARPETS. 1 77

(e) Ashing Procedure.A porcelain dish may lose in weight on strong ignition and for this reason

it is advisable not to heat the residue more than is necessary to effect removal of carbonaceous matter. The recarbonation process is carried out as follows :—

The ash is allowed to cool and moistened with a few drops of ammonium carbonate solution. The excess is evaporated on a steam bath and the dish then gently warmed until the smell of ammonia is no longer perceptible. The dish is then re-weighed.

(f) Check Determinations.10 gms. of Mexphalte 30/40 were placed in a flask and 1000 ml. solvent

and 100 gms. limestone powder added. After shaking and adjusting the temperature of the solution as prescribed in the method, a 10 ml. aliquot was evaporated to dryness, dried half-an-hour a t 160-170° C., weighed, ashed and the residue recarhonated. Similar experiments were carried out adding pulverized silica and Portland cement. In all cases the weight of bitumen obtained was slightly less than the theoretical amount, presumably on account of absorption of bitumen by the mineral aggregate. The actual figures are shown in Table VII.

T a b l e V I I .

Solvent. Aggregate. W t. of bitumen in aliquot.

Carbon disulphide Limestone 00996Chloroform 00994Carbon tetrachloride 0-0981Perchlorethylene 9 9 0 0985Triehlorethylene 9 9 00986

*» Pulverized silica 0-0988> f 9 9 9 9 0-09859 9 Portland cement 0-0978

In conclusion, the author wishes to express his thanks to the Director of these laboratories, Mr. H. B. Milner, M.A., for permission to publish and to record his appreciation of the efficient manner in which Mr. A. H. Clarke performed most of the analytical work.

Geochemical Laboratories,London, $ . JT.l.

References.

1 D . M. Wilson, J .S .C .I ., 1931, 50, (28), 599-600.2 British Standards Institution, B .S.S. 598, 1936, “ Methods for the Sampling and

Exam ination of Bituminous Road Mixtures.”3 D . M. W ilson, J .S .C .I ., 1933, 52, (28), 597-81.4 B ritish Standards Institution, B .S.S. 598, 1936, “ Methods for the Sampling and

Exam ination of Bituminous Road Mixtures.”5 D . C. Broome, “ The Testing of Bituminous Road Mixtures,” Arnold, London

(1934).6 I . Hvidberg, B it., 1936, 6, (8), 169.

1 7 8

T H E IN S T IT U T E O F P E T R O L E U M .

SPECIAL GENERAL MEETING.

A Special General Meeting of the Institute was held a t the Royal Society of Arts, John Street, London, W.C.2, on Tuesday, 10th January, 1939, at 5.30 p.m., for the purpose of considering and, if thought fit, passing with or without amendment certain Temporary Regulations relating to the Transfer of Members, Associate Members and Associates.

T h e P r e s i d e n t , L i e u t .-C o l S. J . M. A u l d , O.B.E., M.C., D.Sc., occupied the Chair.

T h e S e c r e t a r y (M r . S . J . A s t b u r y ) read the proposed Temporary Regulations, as under :

T E M P O R A R Y R E G U L A T IO N S R E L A T IN G TO T H E T R A N S F E R OF M E M B E R S , A S S O C IA T E M E M B E R S , A N D A S S O C IA T E S .

(In force up to 30th June, 1939.)

(а) A ll Members, Associate Members, and Associates of the Institute of Petro­leum, whose names were on the register as such on the 31st December, 1938, shall be entitled to apply for transfer to one of the classes of Fellows, Members, or Associate Members in accordance with the Regulations given hereunder.

(б) A s soon as possible after the Special General Meeting at which these Tem­porary Regulations are approved by the members, and in any case before 30th June, 1939, any Member, Associate Member, or Associate desirous of being transferred to another class of membership should indicate to the Council on a Form prescribed for the purpose the class of membership to which he applies to be transferred.

(c) No transfer shall become effective before the 1st January, 1939, and no transfer shall be deemed valid until the appropriate Transfer Fee, if any, shall have been paid.

M e m b e r s .

(d) A Member of the Institute of Petroleum whose name was on the register as such on the 31st December, 1938, is entitled to apply for transfer to the class of Fellow, and the decision of the Council on his application shall be final. No Transfer Fee shall be required.

A s s o c i a t e M e m b e r s .

(e) An Associate Member of the Institute of Petroleum is entitled to apply for Transfer to the class of Fellow or Member, and his application for transfer shall be submitted to and decided upon by the Council, whose decision shall be final; or he may continue an Associate Member of the Institute.

A Transfer Fee of £1 Is. shall become payable on transference of an Associate Member to the class of Fellow or Member.

A s s o c i a t e s .

(/) An Associate of the Institute of Petroleum is entitled to apply for transfer to one of the classes of Fellow, Member, or Associate Member, and his application for transfer shall be submitted to and decided upon by the Council, whose decision shall be final.

No “ Associate ” may continue as such after the 30th June, 1939.

SPECIAL GENERAL MEETING. 179

On transference to the class of Fellow or Member an Associate shall pay a Transfer Fee of £1 Is. On transference to the class of Associate Member no Transfer Fee w ill be required.

P u b l i c a t i o n o r T r a n s f e r s .

(g) Notification of applications for transfer w ill not be pubbshed in the Journal, but a ll transfers w ill be announced in the Journal as and when they become effective.

The P r e s i d e n t , in proposing the adoption of the Regulations, said they had been prepared by the Council and were the result of a considerable amount of deliberation. He did not think they required any explanation from him, as they were quite clear.

M r . J a c k s o n s e c o n d e d t h e m o t i o n .

M r . B u t t e r f i e l d said that, as a member of long standing, he would like to ask for a little explanation of the grounds of transfer of Members to the grade of Fellows. Up to the end of last year Members were in the highest category of membership of the Institute, but he gathered tha t in future the highest category would be that of Fellow. Would those who were Members at the end of last year be automatically transferred, on application, to the grade of Fellow, or was it intended that some discretion should be exercised in the m atter ? The question was raised, he thought, by the words in Regulation {d) : “ the decision of the Council on his application shall be final.” The Council knew all those who were at present Members, and must be in a position to make up their minds whether, on application, those Members should be transferred en bloc to the grade of Fellow. If the Council did not intend to do that, but to exercise some discrimination, the grounds on which the applications would be considered should, he thought, be indicated to the Members. I f it was not intended to exercise such discrimination, the words “ the decision of the Council on his application shall be final ” might well be omitted.

T h e P r e s i d e n t said he would answer Mr. Butterfield’s question in the same way as tha t in which he had answered it previously, by saying that the m atter would be dealt with most generously by the Council, but the Council wished to have the power to discriminate if and when necessary. He did not like using the word “ discriminate,” because it implied some­thing rather different from what the Council had in mind. He thought the fact tha t in the definition of Fellows it was stated tha t a Fellow must have “ advanced the science and technology of petroleum ” implied that there might be cases in which, if the standard of a Fellow was to be maintained, it would not be suitable to transfer a Member to the grade of Fellow without consideration. He thought also that the fact th a t the m atter was given consideration should be welcomed by the Fellows. They would feel tha t they had not been elected by a stroke of the pen, but tha t it had been done carefully; there being first a definite, deliberate instruction from the Election Committee to the Council and then a reconsideration of the m atter by the Council.

D r . J . A. L. H e n d e r s o n asked whether the twenty-four Founders would continue to be life members of the Institute when transferred to the grade of Fellow.

T h e P r e s i d e n t said th a t tha t would undoubtedly be the case.

If there were no other questions, he would put the resolution th a t the Temporary Regulations read by the Secretary be adopted.

The resolution was carried.

T h e P r e s i d e n t said there was no other business and declared the meeting closed.

1 8 0 SPECIAL GENERAL M EETING.

E R R A T U M .

Physical and Chemical Constants of Normal Paraffins b y D . J. W . K r e u l e n .

Journal, Vol. 24, No. 180, October 1938, p. 557. Table I I last column. Values for W p should be 0-95, 0-85, 0-77, 0-61, 0-30, 0-18, 0-13, 0 04 for molecular weights from 481 to 206.

7 3 a

ABSTRACTS.

Geology ............................P A G E

74 a

Geophysics 8 0 a

Drilling 8 5 a

Production 8 8 a

Transport and Storage ... 91 A

Crude Petroleum ... 9 2 a

Gas 93 a

Cracking ... 93 a

Hydrogenation ... 94 a

Polymerization ... 9 5 a

Refining and Refinery Plant 9 5 a

P A G E

Chemistry and Physics of Petrol­9 8 aeum

Analysis and Testing 1 0 4 a

Motor Fuels 1 0 4 a

Gas, Diesel and Fuel Oils 1 0 5 a

Lubricants and Lubrication 1 0 6 a

Special Products ... 1 1 0 A

Detonation and Engines... 1 1 1 A

Coal and Shale 1 1 1 A

Economics and Statistics 1 1 1 A

Book Reviews 1 1 2 A

Book Received 1 1 6 a

A U T H O R IN D E X .

The numbers refer to the Abstract Number.

Adams, E., 314 Agde, G., 339 Albright, J. C., 258, 263 Altman, B., 256 Anderson, A., 267 Anderson, C. A., 289 Anderson, C. T., 267 André, E., 328 Archer, S., 312, 313, 314 Auld, S. J. M., 330

Baker, C. L., 248 Baker, J. G., 292 Ballard, N., 214 Barnes, C., 276 Barton, C. H., 325 Bass, E. L., 325 Bataafsche Petroleum

Mij., N.Y. de, 321, 335 Ba Thi, M., 303 Beckman, E. J., 250 Beeny, H. H., 332 Bignell, L. G. E., 268 Bishop, J. J., 254 Boelhouwer, J. W. M.,

320Boulter, G. M., 254 Bouman, C. A., 322 Boyd, W. B., 215 Bray, U. B., 327 Breth, F. W., 335 Bridges, C. H., 293 Brimsdown Chemical

Works, 336 Brooks, D. B., 301 Brown, A. B., 335 Brown, E. K., 335 Brownsdon, H. W., 335 Buchler, C. C., 335 Burch, E. A., 283, 284 Burn, W. S., 341 Byers, A. M., Jr., 296

Cerf, C. S., 273 Chave, C. T., 335 Clark, C. C., 213 Clark, C. L., 287 Clark, L. V. W., 262 Conn, J. B., 310 Cook, L. W., 335 Cowles, L. G., 238

F

Cox, W. F., 267 Cozzens, F. R., 245 Crickmer, C. C., 267 Crider, A. F., 216

Danforth, R. S., 290 Damielle, R. B., 269 Davidson, J. R., 267 Deambrosis, R., 288 Decker, E. L., 253 Dehn, F. B., 335 Denison, I. A., 269 Deussen, A., 229 Diggs, S. H., 335 Diwoky, F. F., 335 Donnell, J. W., 284 Dorn, N. L., 253 Downey, V. I., 335 Dreyfus, H., 321 Du Pont de Nemours,

E. I. & Co., 336

Edeleanu Gesellschaft m.b.H., 335

Egloff, G., 271 Elkins, T. A., 237 Evans, E. B., 316

Fawcett, E. W., 282 Fenske, M. R., 335 Fischer, F., 280, 338 Flagg, R., 254 Foster, A. L., 277 Frenzel, W. G., 254 Friedman, B. S., 298, 309 Fussteig, R., 319

Garrison, A. D., 248 Germe, A. E. J. L., 321 Gillingham, W. J., 261 Ginsberg, B., 335 Glavis, F. J., 307 Godlewicz, M., 335 Green, C., 341 Green, C. H., 235 Gruse, W. A., 335 Gunness, R. C., 292

Hammer, S., 237 Hanes, D., 267 Hanson, V. F., 336

Harrington, P. J., 335 Hartigan, A. W., 335 Hass, H. B., 317 Hawley, P. F., 233 Henderson, L. M., 285 Hendry, W. B., 335 Henne, A. L., 297 Hewitt, J. C., Jr., 267 Hiestand, T. C., 219 Hildorf, W. G., 287 Hjerpe, E. B., 335 Hodge, E. B., 291 Hoover, C. O., 321 Hotine, L., 341 Houdry, E. J., 278 Humphrey, G. T., 267 Hunter, T. G., 303 Hutchins, G. H., 335

I.G. Farbenindustrie, A.-G., 335

Imperial ChemicalIndustries, Ltd., 282,335

Ipatieff, V. N., 298, 309

Jenkins, R. D., 276 Johnson, 0. H., 234 Johnson, G. W., 281, 321,

336Johnstone, R. G., 272 Jones, S. O., 299

Katz, D. L., 318 Kelly, S. F., 241 Kendall, J. M., 254 King, H. H., 257 Kinney, C. R., 315 Kinsel, A., 335 Kistiakowsky, G. B., 310 Klaus, H., 232 Knox, G. S., 267 Kreulen, D. J. W., 302 Küster, V. E., 267

Ladd, E. C., 297 Launspach, E. H., 300 Lawlor, R., 240 Lee, J. W., 259 Liberthson, L., 335 Liddle, J. C., 336

Lippincott, S. B., 317 Livingston, H. K., 260 Lloyd, A. H., 332 Loomis, A. G., 254 Love, F. H., 259 Lyon, N. X., 251

McCluer, W. B., 335 McDonald, A. T., 326 McFarland, E. G., 335 Mair, B. J., 294, 295 Marker, R. E., 306 Martin, E. J., 335 Martin, M., 261 Marvel, C. S., 307 Mason, H. J., 333 Merrill, D. R., 327 Meyer, H., 336 Migaux, L., 243 Miller, J. F., 331 Mitera, Z. A., 231 Monroe, W. H., 220 Montgomery, P., 252 Moore, 0. C., Jr., 327 Moos, J., 335 Morrell, J. C., 271 Murphy, W., 270 Murray, G. H., 261 Muskat, M., 230

Nash, A. W., 262, 303 Nay, B., 267 Nissan, A. H., 262

Oakwood, T. S., 306 Orem, H. P., 305 Otto, C.& Co., 274

Palmer, R., 272 Park, T. S., 267 Paton, J. G., 282 Pease, R. N., 296 Penick, A. J., 267 Penick, K. T., 267 Pennington, H., 254 Perrin, M. W., 282 Pichler, H., 280 Pilat, S., 335 Pines, H., 298, 309 Pippin, J. W., 254 Plummer, W. B., 282

ABSTRACTS.

Pohl, H ., 338 Po tt, A., 340 Powell, R. 0 ., 290 Précoul, M., 337 Prévost, E., 323

Ramser, H ., 335 Raymond, É. P ., 254 Reed, P., 266 Reeside, J . B., J r ., 217 Reich, H ., 244 Reid, E. Em mett, 299 Rhodes, F. H ., 293 Ridgway, C. M., 285 Riffkin, J ., 324 Roche, J ., 328 Rock, S. M., 239 Roper, E. E., 308 Rosaire, E. E ., 242 Rosenblum, 0., 311 Ross, W. B., 285

Roulston, W. R ., 247 Rude, R. L., 276 Ruhrchemie, A.-G., 321 Russell, W. L., 221

Saegebarth, E ., 335 Saltman, W., 318 Sawdon, W. A., 249 Schuerenberg, H ., 339 Sclater, K. C., 246 Setzler, H . B., 335 Shaw, S. F ., 255 Sherrill, M. L., 300 Simons, J . H ., 312, 313,

314Skinner, W. H ., 341 Smith, E. A., 310 Southgate, H . A., 304 Sparrow, S. W., 329 Staatsm ijnen, De Direc­

tion Van de, 274

S tandard Oil Develop­m ent Co., 321, 336

Standlee, H . R ., 254 S tchepinsky, V., 227 Stead, H . V., 341 Stephenson, L. W., 217,

220Stone, B. H ., 267 Stone, F ., 267 Story, Le R. G., 335 Swift, H . W., 334

Taylor, O. 0 ., 254 Terres, E ., 335 Thaheld, F . A., 267 Thiele, E. W., 335 Thom as, G. D., 212 Thorne, H . M., 270 Tillotson, A. W., 218 Turner, R . H ., 286

V an W ijk, W. R., 320 Vogel, H. E. R., 336

W allis, J . S., 335 W eber, G., 265 W est, S. S., 236 W hite, A. E., 287 W hite, F . L ., 335 W hitm ore, F . C., 304,

305W illiam, E . G., 282 W illingham , C. B ., 294,

295Willson, C. O., 222, 223,

224, 225, 226 Wilson, R. R., 335 Wolf, O., 290 W ood, O. E ., 262

Ziegenhain, W. T ., 275 Zim m erm an, G. B ., 271

Geology.212. Carterville-Sarepta and Shongaloo Fields, Bossier and Webster Parishes, Louisiana. G. D . Thomas. Bull. Amer. Ass. Petrol. Oeol., 1938, 22, 1473—1503.—Thesub-surface rocks of these fields consist of Tertiary sediments (Claiborne & Wilcox, 1150 ft., and Midway 500 ft.), and the Cretaceous and Comanche formations.

The Annona chalk is the most easily recognized chalk member of the Cretaceous, having a thickness of 120 f t . ; its base has been used as a datum plane on the structure maps included in this paper. In the Comanche, no well at Shongaloo or Carterville has yet penetrated below the Glen Rose anhydrite, which is approximately 500 ft. thick at Shongaloo.

The Shongaloo field, discovered in March 1921, is an east-west elongate dome, separated from the Carterville-Sarepta field on the west by a saddle and from the Cotton Valley structure on the south also by a saddle. The only producing sand at Shongaloo is the Buckrange of basal Ozan age. I t is found at an average depth of 2600-2650 ft. and is very variable in character and thickness. I t is thought that the amount and extent of production is governed as much by the variable sand conditions (porosity and thickness) as by structure.

The Carterville-Sarepta field appears to be a north-west to south-east extending anticline on which six small local closures exist. The dip on the south side of the anticline is apparently steeper than that on the north. A description of the six local closures is given. Three producing sands occur, the Buckrange (oil), the topmost sand member of the Tokio (gas) and the second sand member of the Tokio— 60 ft. below the gas sand-—produces oil in the Carterville area.

The origin of the structures in both fields is uncertain, although it is generally agreed that they are the result of a small amount of flowage of salt beds of Comanche or pre-Comanche age which are thought to underlie most of North Louisiana and part of Arkansas. G. S. S.

213. Sugar Creek Field, Claiborne Parish, Louisiana. C. C. Clark. B ull. Amer. Ass. Petrol. Oeol., 1938, 22, 1504-1518.—This field was discovered in March 1930 and has a producing area of about 4000 acres. A generalized section is given of the formations occurring in the field; these are : the Eocene (Claiborne), Upper Cretaceous (Gulf Series) and the Lower Cretaceous (Trinity).

Structurally, the field is an anticline about five miles long by three miles wide. At the surface it occurs as an imperfect inlier of the Cook Mountain formation sur­rounded by the overlying Cockfield. Two structure contour maps are furnished.

Gas is obtained from two reservoirs in the T r in ity series-—the K ilp a trick and Darrett zones. O il is found only in economic quantities in the Darrett, where both gas and oil occur in porous beds distributed throughout a thickness of 175—275 ft. The total accumulated oil production on 1st January, 1938, was 86,000 brls., a ll from the Darrett zone. g_

ABSTKACTS. 7 5 a

214. Stratigraphy and Structural History o! East-Central United States. N . Ballard. Bull. Amer. Ass. Petrol. Geol., 1938, 22, 1519 -1559.—This paper summarizes some facts concerning the stratigraphy of Michigan, Ohio, Indiana, Illin o is, Kentucky and Tennessee.

A series of cross-sections are given in order to show as many stratigraphical details as possible, and samples from approximately 525 wells have been examined to assist in this work. A tentative correlation table and seven logs are also included in the paper.

Two major periods of folding occurred in the area, the first at the beginning of Onondago time. This folding probably reached its maximum towards the end of Hamilton deposition, for on the flanks of the Ozarks there was much post-Ham ilton faulting.

The second period of major faulting occurred at the end of Mississippian time. When the Pennsylvanian sea encroached over the old land surface, hundreds of feet of beds had been removed as the Pennsylvanian overlaps St. Peter sand in northern Illinois.

The last uplift to affect the area arose after the deposition of the Pennsylvanian and before the Cretaceous was laid down, as faults in Kentucky displace the Pennsylvanian but do not cut the Cretaceous. G . S. S.

215. Jesse Pool, Pontotoc and Coal Counties, Oklahoma. W . B . Boyd. B ull. Amer. Ass. Petrol. Geol., 1938, 22, 1560-1578.—The Jesse Pool and the F itts Pool (one mile distant) are closely related both stratigraphically and structurally. The Jesse Pool itself is about four miles long and one mile wide.

The oldest formation drilled in the Jesse is the Arbuckle limestone, penetrating it 87 ft. This is followed by rocks of the Simpson Group, Siluro-Devonian (Hunton Limestone), Mississippian and Pennsylvanian (Morrow Group and Des Moines Group).

The Jesse Pool lies on a large anticline which is faulted on the south side. The faulting has a maximum throw of 1300 ft. and occurs in a series of step faults, a ll of which are normal.

Production may be generally divided into four zones and areas : “ W ilco x,”Bromide, Hunton and Pennsylvanian. The best production is found in the “ W ilcox,” which is confined to a small area on the crest of the anticline. The average in it ia l potential gauge was approximately 4000 brls. a day, the estimated ultimate production being 6,650,000 brls.

From the Bromide sand, the average in itia l production was 1800 b r ls .; the estimated ultimate yield is 1,680,000 brls. Production from the Hunton is erratic, the in it ia l yield ranging from 63 to 3480 brls., and an estimated ultimate production of 4,000,000 brls. from the present developed area.

From the Pennsylvanian, production is confined to the W apanucka and Atoka formations. Three gas wells have been completed in the W apanucka sand, but no showings of oil have been observed, and five wells have been completed in the “ Gilcrease ” sand (Atoka) with an average production of 72 brls. a day.

On 1st January, 1938, the Jesse Pool had produced 1,704,164 brls. G. S. S.

216. Geology of Bellevue Oilfield, Bossier Parish, Louisiana. A . F . Crider. Bull. Amer. Ass. Petrol. Geol., 1938, 22, 1658-1681.—The Bellevue Oilfield is located in T.19 N., R . l l W ., Bossier Parish, and is surrounded by the producing oilfields of north-western Louisiana.

The formations occurring in the area are of Upper Cretaceous and Tertiary age. The oldest sediments reached to date are black marine shales and limestones of the Cotton Valley formation which have been penetrated at 1429 ft. The remaining formations of the Upper Cretaceous follow, the close of the System being marked by a chalky shale of Arkadelphia age, which on the crest of the structure is about 40 ft thick. Overlying this bed is the Midway clay (Tertiary). Th is consists of shale which immediately underlies Pleistocene and Recent deposits. Its uniform thickness is about 400-600 ft., though on the crest of the dome the thickness has been partly reduced by truncation, leaving only 100 ft.

Structurally, Bellevue is a deep-seated dome, probably of salt origin, w ith overlying sands across the crest in which oil has accumulated. From the top of the dome to the base of the syncline separating Bellevue and Cotton V alley, the amount of u p lift

ABSTRACTS.

as shown on Upper Cretaceous data, is about 1800 ft. and 2500 ft. on Lower Cretaceous

daO il is obtained from the Nacatoch sand at a depth of 300-400 ft., and gas and oil from the junction of the Upper and Lower Cretaceous at 1800 ft.

Since its discovery in 1921 to 1st January, 1938, the field has yielded 9,860,430 brls., the peak production being reached in 1923 (2,250,057 brls.). The field was closed during 1932-1933 owing to depression conditions. The producing area of the field is 900 acres and the yield per acre to 1st January, 1938, is 10,996 brls.

217 Comparison of Upper Cretaceous Deposits of Gulf Region and Western Interior Region. L. W. Stephenson and J . B. Reeside, Jr. B ull. Amer. A ss. Petrol. Geol., 1938, 22, 1629-1638.—In the Gulf region the Upper Cretaceous sediments are chiefly of marine origin. The most complete section is the classic example in east central Texas, and this may be taken as a standard. The series includes the Woodbine Sand, Eagle Ford shale, Austin Chalk, Taylor marl and the Navarro group, having an aggregate thickness of about 3000 ft. Traced both south-west and north-east from east central Texas, however, the sequence changes markedly.

In the Western Interior region the Upper Cretaceous deposits m ay be classified into three types : (i) an eastern belt of marine fine sediments (Great Plains sequence);(ii) a middle belt of mixed continental and marine sediments (Rocky Mountain sequence); (iii) a western belt of sandstones and conglomerates m ainly non-marine, an incomplete sequence.

In both regions fossils are abundant. Although numerous species are restricted to either one region or the other, many of them are indentical or analogous to allow of correlation. The ranges of some of these fossils are shown in a table.

G. S. S.

218. Olympic Pool, Hughes and Okfuskee Counties, Oklahoma. A . W . Tillotson.Bull. Amer. Ass. Petrol. Geol., 1938, 22, 1579-1587.—The Olym pic pool is in the east central part of Oklahoma, just off the north-east edge of the Seminole uplift.

The main producing horizon is from the Senora formation, a lenticular sand member—“ Olympic Sand.” Minor production is also obtained from the Pennsyl­vanian formation (Calvin Sand and Cromwell Sand) and the Hunton limestone of Siluro-Devonian age.

To date 326 oil wells have been completed and all but four of these derive their oil from the Olympic Sand. I t is estimated that the pool w ill eventually recover 16-24 m illion barrels from this horizon. G . S. S.

219. Studies of Insoluble Residues from “ Mississippi Lime ” of Central Kansas. T. C.Hiestand. Bull. Amer. Ass. Petrol. Geol., 1938, 22, 1588-1599.—From a study of the stratigraphy aided by the use of insoluble residues, it is shown that the Mississippi lime in Central Kansas can be subdivided into zones or formations. These zones are proved to be very sim ilar to the members of the Boone limestone of Missouri.

Oil is obtained from one or both of two of the zones in certain pools. Accumulation in many places is apparently due to stratigraphic traps rather than to merely structural closures. Accompanying the paper are two cross-sections arranged w ith the top of the Missouri series as the datum plane to show the subdivisions of the Mississippi lime.

The author concludes that the subdivision of this formation in a ll parts of Kansas and adjacent areas should furnish data to make more complete interpretations of the history of the Central Kansas buried uplift and other important structural features.

G. S. S.

220. Stratigraphy of Upper Cretaceous Series in Mississippi and Alabama. L . W.Stephenson and W. H . Monroe. Bull. Amer. A ss. Petrol. Geol., 1938, 22, 1639- 1657. The Lpper Cretaceous sediments of the Eastern Gulf region outcrop in a great crescentic band which wraps around the south-west end of the plunging Appalachian Highlands. This band is 500 miles long and centrally is 75 miles wide, the sediments having a maximum thickness of about 2300 ft.

The Upper Cretaceous in west central Alabama and in east central Mississippi is

ABSTRACTS. 7 7 a

readily divisible from below upward into four formational u n its ; the Tuscaloosa, Eutaw, Selma chalk and Prairie Bluff chalk.

Deposition of the Upper Cretaceous was not continuous, sedimentation having been interrupted from time to time. These breaks are recorded in at least four uncon­formities. The first of these is between the Tuscaloosa and the overlying Eutaw , and has been traced throughout the eastern G ulf region.

The next unconformity is that separating the Tombigbee sand member of the Eutaw from the overlying Selma ch a lk ; the break is believed to be widespread in the area.

The third break is about 300 ft. above the base of the Selma chalk and a few feet above the Areola limestone. The position of this break is marked by a thin band of phosphatic fossil moulds.

The fourth break, separating the Selma chalk from the overlying Prairie Bluff chalk, is an important one. I t has been traced for more than 300 miles in Mississippi, Alabama and Georgia; in places it marks a northern transgression of the Prairie Bluff chalk over a great thickness of older Cretaceous sediments.

The following fossil zones have been recorded through M ississippi and Alabama : the Exogyra ponderosa zone which includes the Tombigbee sand and the lower two- thirds of the Selma chalk. A narrow and persistent zone in the Selma chalk about 180 ft. above the Areola limestone is characterized by Diploschiza cretacea and by Terebratulina filosa. Th is zone has been traced from Montgomery County to Tupelo (Mississippi), a distance of 220 miles. Including the upper third of the Selma chalk and the Prairie Bluff chalk, the Exogyra costata zone can be traced, and the E. cancellata zone can be followed throughout the area in the lower 200 ft. or less of theE. costata zone. G. S. S.

221. Relation o! Rough Creek Fault of Kentucky to Ouachita Deformation. W . L.Russell. Bull. Amer. Ass. Petrol. Oeol., 1938, 22, 1682-1686.—A belt of thrust faults and folds extends from the Mexican border to Arkansas, forming one of the major structural features of the continent.

The Ouachita structures exhibit no sign of diminishing where they disappear below the Cretaceous and Tertiary sediments of the Mississippi Embayment. The Rough Creek fault system of Kentucky and its associated faults are sim ilar to the Ouachita structures on the west.

The author therefore discusses the evidence as to whether the Rough Creek fault is an eastern extension of the Ouachita deformation. Of various suggestions made, one is quoted from evidence obtained from wells drilled in the Mississippi Embayment in north-west Tennessee between the Ouachita and Rough Creek structures. Here wells have penetrated Palaeozoic rocks which are definitely disturbed and altered.

G. S. S.

222. Oil Bringing a New Epoch in Eastern Venezuela. C. O. Willson. Oil Gas J .,3.11.38, 37 (25), 14.—The Eastern Venezuelan oil production constitutes about 1 6 % of the 530,000 brls. daily output of Venezuela. In the Maracaibo region the older fields are now declining in output, but exploratory work is still going on.

The area of search in Eastern Venezuela is about 450 miles long and 350 miles broad. In it Quirequire with a daily production of 70,000 brls., provides most of the oil. This field has 247 wells, of which 209 are flowing. 12,700 acres have been proved, and oil is obtained from Pliocene sand lenses under shore-line conditions at depths of 1800-3250 ft. The very thick oil zone is subdivided by a water sand. The Peder- nales field gives oil from the Middle Miocene at depths of 5000-6200 ft. on a north­east to south-west trending anticline. H igh pressures and steep dips render the work rather difficult.

Production from the 40 wells at Temblador is derived from the Middle Miocene on the up-thrown side of a normal fault at depths of about 3900 ft.

Prolonged seismic work preceded drilling at Oficina, where five producing wells, which have proved an area of 3 sq. ml., are shut in. The average depth of the pro­ducing sand is 5700 ft., but deeper and shallower oil-sands have been reported. A t Santa Ana there is gas-distillate production from a depth of 7500 ft., and low -gravity oil production has been found at E l Tigre.

The geological knowledge of this area is largely the result of geophysical work, for the area is cloaked by a thick cover of Pliocene deposits. G. D . H

ABSTRACTS.

223 Exploratory Work in Venezuela. C. O. W illson. Oil G a sJ ., 10.11.38, 37 (26), U 17 —Most of the exploratory work is taking place in eastern and central Venezuela, but some is going on in the west, extending to the Colombian border. A t present twelve geological crews, four magnetometer crews, nine gravity crews and fourteen seismic crews are at work. There is also one E ltran party.

A geosyncline appeared in Cretaceous times, and in the Oligocene the well-defined Maracaibo basin was separated off. This covers about 63,000 sq. km ., whereas the less well-defined Maturin basin in the east covers 68,000 sq. km. About two-fifths of the total area is under concession.

There are about 2700 producing wells, of which 300 are in eastern Venezuela. About half of them are being pumped. Western Venezuela and the Maracaibo area give about 440,000 brls. of oil per day. The bulk of the production comes from depths of 2000-3750 ft. The deepest well is the E l Roble wildcat near Oficina, which isdrilling at over 9000 ft.

A geological map shows the positions of the fields and of the wildcats.

224. Western Venezuela Drilling Operations Show Efficiency Against Hazards. C. O.Willson. Oil Gas J ., 1.12.38, 37 (29), 26-28.—Western Venezuela is essentially a supplier of heavy oils. Oilfields now run almost continuously for nearly fifty miles along the northern half of the eastern side of Lake Maracaibo. Many wells have been drilled in the lake. I t is possible that the decline in production of the old fields may be offset by newer ones. Bachaquero is the most southerly and Tarra the most westerly field.

The Lake fields have about 2000 producing wells, of which 33 % flow, 62 % are pumped and the rest are on air or gas lift. Ambrosia, L a Rosa and Benitez have about 775 wells, T ia Juana 300, and the rest are at Lagunillas. The Am brosia-La Rosa wells are about 2300 ft. deep ; those of Lagunillas about 5000 ft.

The Maracaibo pools yield oil from Cretaceous-Mid-Miocene beds ; (a) on anticlines and domes of considerable variety ; (6) on gentle monoclines w ith sands or sandstones which lense out up dip ; (c) on combinations of (a) and (6). Shore line conditions are responsible for most of the oil accumulations now exploited in the Maracaibo basin. These are not near the margin of the basin, but close to the central part of the present topographic depression. G. D . H .

225. Colombia to Witness Active Drilling Campaign in 1939. C. O. W illson. OilG asJ., 22.12.38, 37 (32), 14 -16 .—Most of the concessions granted lie in the Magdalena valley, in the main east of the river. The area extends along the river for about 200 miles from Bodega Central in the north to Laclorada in the south. A 335-mile pipe-line connects the fields of the De Mares concession (Infantes and L a Cira) with Cartegena on the Caribbean. The daily oil output is 60,000 brls.

The wildcat Socony-Narina No. 1 is drilling at 5000 ft. on the Narina uplift. Three wells have been drilled on the Las Monas anticline, one of which is a small producer. A well is being drilled at San Fernando, and drilling is expected to begin shortly on the Cantemplora and Carare concessions. Elsewhere geological surveys are in progress. G. D . H .

226. Remarkable Developments in Barco Concession. C. O. W illson. Oil Gas J .,15.12.38, 37 (31), 18.—When the 260-mile pipe-line from Petrolea to Covenas on the Caribbean is completed, rapid developments are expected to take place in the Barco concession. This concession, near the Colombian—Venezuelan border, is about 75 miles long and 25 miles broad, and is the site of numerous oil and gas seeps. These were exploited primitively, and in 1919 a shallow test in the R io de Oro district in the north showed heavy oil. In later work in 1933 Petrolea No. 1 blew in at about 500 ft. and caught fire. Thereafter several wells were drilled at Petrolea and a further two at Rio de Oro. Geological work and wildcatting have been going on in other parts of the concession.

At Petrolea, production is from a much-faulted anticline. Most of the oil comes from siltstones and limestones in the Cogollo formation at 450-650 ft. The initial

ABSTRACTS. 7 9 A

output per well ranges from a few hundred to 3000 brls. per day of 45° A .P .I. oil, giving 5 0 % gasoline on atmospheric distillation.

The four R io de Oro wells yield 37-40° A .P .I. oil from a sandstone in the Catatumbo formation at depths of 1200-1500 ft. (j, D . j j .

227. Petroliferous Beds of Egypt. V . Stchepinsky. Ann. Off. Corrib, liq., Sept.- Oct., 1938, 13 (5), 823—874 (1 map).—The dj. Zeit seep has long been known, and the search for oil began in 1868. Evidences of oil have been observed in the Gulf of Suez zone, on the Red Sea coast and in north Sinai. The first producing field was opened up at Djemsah in 1909, but from 1914 onwards Hurghada has been the main producer. In late years the production has fallen off. New legislation was recently introduced which has led to the taking up of m any permits, in regions from Sinai to Ras Benas on the Red Sea, and west of the N ile between Alexandria and Cairo.

There are two main features in the stratigraphy of the Red Sea basin : (i) A big gap between the Crystalline and the Miocene in the prim itive anticlinal zones, pointing to intense erosion in Cretaceous, Eocene and Oligocene times, (ii) Lateral variation of facies. There is also a lack of correlation between some areas, and altogether the establishment of a general stratigraphic column is rather difficult.

The various stratigraphical horizons are briefly described. The term “ Nubian sandstones ” has only a lithological and not a stratigraphical significance, for it has been used for beds ranging from Carboniferous to Senonian in age. The Oligocene is missing and the Jurassic is best known at d j. Moghara.

The principal earth movements took place at the end of the Eocene, and there were accessory movements in Pliocene times. The oil-bearing beds are grouped as occurring on (a) fundamental anticlines and (6) on superficial anticlines. The former have crystalline cores and the latter lie in the synclinal areas between the former. In the Cretaceous there was a transgression from north-west to south-east. A new trans­gression came from the same direction in the Miocene, which period ended in a long lagoonal phase. The fundamental anticlines are suitable for testing Eocene and Cretaceous beds, where present, but not so the superficial anticlines where the Miocene is thick.

The beds and structure of the various areas with evidences of oil are described : North Sinai, Habachi, B ir Abou K it ifa , dj. Kochera, Oued Gharandel, dj. Tanka, Oued Matulla, dj. Mezzazat, Abou Durba, Ras Mohammed, Ras Gharib, Ras D ib, Zeitia, Ranim Island, Ras el Bahar, Ras Djemsah, islands east of Djemsah, Abou- Chaar, Hurghada, Abou-Mingarh, Jiftoun Kebir Island, and Dichet el Daarba. Many beds show traces of oil, but few are important. Many seeps are associated with faults and some with dykes. The Nubian sandstone is generally water-bearing. The productive horizons are : (i) granitic sands of uncertain age between the granite and the Campanian (Hurghada); (ii) Campanian (?) (Hurghada); (iii) base of Miocene (Hurghada) and below the Lagoonal formation (Djem sah); (iv) Lagoonal formation (Hurghada and Djemsah). A ll the present producing structures have a granite core.

The history of the oil exploitation is summarized. In 1931 a maximum output of 285,000 metric tons per annum was attained.

Several mother rocks m ay exist, amongst which the Cretaceous is prominent. As yet the exploration has only been shallow. G. D . H .

228. Recent Exploration in the Carpathian Foreland. Anon. Bohrtech. Z ., 1938, 56 (11), 164—166.—In the last two years some 20,000 sq. km. of the foreland have been mapped and seismic surveys made over 9000 sq. km. The sub-Carpathian Salifere has the most favourable oil prospects, although the locating of test wells is extremely difficult. Surface shows and shows in shallow borings are abundant, but, from the point of view of deep drilling, the region is almost unknown. The presence of the Boryslaw tectonic unit hew been proved in the north-west, near Sambor. In the south­east the geology of the area around Niebylow has been elucidated by means of hand- borings and pits, and it is shown that it compares with Bitkow rather than Boryslaw.

Little progress has been made in the anticlinal zone of the Stebnika Beds. The area is free of a ll surface shows and none were obtained in the test boring— 1504 m.—- at Gaje-Nizne.

In the Tortonian area it has been shown that the natural gas zones at Konigsau, Daszawa, Balicze, Opary and Kosow are associated w ith the overthrust contact

ABSTRACTS.

between Stebnika Beds and the Tortonian. The gas-hearmg beds at Kosow are in the form of gentle domes of a size and type not hitherto known in Poland. The investigation of the Tortonian is in its infancy, but it may be said that the Carpathian foreland is very similar to the oil series of the outer Alpine arc, and the Tortonian has all the characteristics of an oil facies. Although the gas at Daszawa is pure methane, in other places it contains as much as 2 0 -3 0 % of ethane and higher homo- logues Two problems now to be considered are the search for oil in the Tortonian and the exploration of the older, underlying beds. A further problem is the explanation of the oil shows in the Miocene in the Western Foreland. S. E . C.

229 Discoveries. A. Deussen. Oeophys., 1938, 3 (3), 17 7-19 7 .—Estimates of the oil reserves of the U .S.A. have always proved to be too low' and the discovery rate higher than anticipated. Since 1925 no estimates of undiscovered oil have been made because previous figures were so inaccurate, and yet this is the most important aspect of reserves. The Texas-Louisiana G ulf Coast may be taken as a type region for a con­sideration of this problem. The total number of known structures and fields plotted against time shows that there is as yet no tendency for the curve to flatten and, hence, on statistical grounds, the end of discovery in the G ulf Coast is far off. A s another method of approach, a typical oil-producing county in the G ulf Coast, H arris County, is chosen and the production and discovery rates examined. The discovery rate curve for this county is similar to that for the G ulf Coast as a whole and thus, whilst the structures appear to be crowded, it seems that the lim it has not yet been reached. In Harris County the area of oil land, assuming a ll the structures to carry oil, is 4 076%, and so a final figure of 5 -6 % might be attained. Assuming an average of 1 % for the whole of the Gulf Coast and south-west Texas, 461 fields (average 500 acres per field) remain to be discovered. In terms of undiscovered oil this is ap­proximately 9725 million brls. S. E . C.

See also Abstract No. 262.

Geophysics.230. Reflection of Longitudinal Wave Pulses from Plane Parallel Plates. M. Muskat.Oeophys., 1938, 3 (3), 198-218.—A mathematical analysis is made of the reflection of elastic wave pulses from plane parallel plates, the problem being restricted to the case where the incident waves are longitudinal, and the discussion to the reflected wave pulses. The treatment given resolves the reflected wave system into the various individual reflection and refraction processes, and gives the resultant reflection co­efficients for the different types in a single step. Further, it shows that the reflected pulses are of the same form as the incident pulse, but are characterized by varying amplitudes and phase shifts. Numerical values are given for the expressions deter­mining the reflection coefficients obtained in the above analysis. These calculations are restricted to the three strongest longitudinal reflected pulses, and the values are represented in graphical form as well as being tabulated. The resultant wave amplitudes, given by the combination of the individual reflected waves, are illustrated graphically for cases where the ratio of the thickness of the reflecting plate to the length of the incident pulse is 2, 1 and S. E . C.

231. Present Status and Future Aspects of Geophysical Exploration in Poland. Z. A.Mitera. Oeophys., 1938, 3 (3), 225-233.—The early geophysical work, commenced in 1923, is outlined. Large scale activ ity commenced in 1934 when the Pioneer Co. formed its own geophysical department. Seismic reflection methods proved to be of very limited application within the Carpathian mountain system, due m ainly to the very complex structures, their relatively small size and the definite lack of uniform beds. Conditions are much more favourable in the foreland. Persistent reflections have been followed throughout several thousand square miles and contour maps were made of the reflecting surface. Extensive gravimeter surveys have also been carried out and Schlumberger logging is now standard practice.

It is planned to extend surveys in West Poland for the discovery of salt-domes

ABSTRACTS. 81 A

sim ilar to the North German type. In the province of W ilno, data is now believed to be sufficient to indicate new oil reserves in Palaeozoic and younger sediments associated with the structural uplifts. The structural features may be better outlined by geophysical methods. S. E . C.

232. Second Derivative Contour Method of Interpreting Torsion Balance Data. H .Klaus. Geophys., 1938, 3 (3), 234-246.—Recently the tendency has been to use the torsion balance to measure total gravity, whereas actually it is an instrument for measuring gravity gradients and curvature. I t is advocated that these second derivatives should be used in conjunction w ith a contouring system. In the conven­tional method they are computed w ith reference to astronomic north, but in the contour method one of the axes, preferably the y-axis, is taken parallel to the geological strike. Stations should be well spread out over an area, instead of being concentrated along profile lines, and separate contour maps produced of each of the four derivatives. Where very accurate work is being done the contour intervals m ay be 2 -3 E . for the gradient components and 4—5 E . for the curvature components. In other cases, intervals may be 5 E . and 10 E . respectively. The method of interpreting these maps is outlined. I t is not advocated that this method be used for quantitative work. Its application lies m ainly in the recognition of the types of structure present and their relations to one another. Quantitative work m ay then be carried out by seismic methods. S. E . C.

233. Transients in Electrical Prospecting. P. F . Hawley. Geophys., 1938, 3 (3), 247-257.—A constant P .D . is applied between two electrodes fixed in the ground some distance apart and the resultant surge of current is recorded. The potential surge between two other electrodes is also recorded. The occurrence of “ nicks in the curves so obtained is said to indicate the presence of different strata below the electrodes. The depth to an horizon is proportional to the time taken for a “ n ick ” to develop and to the resistivity above that horizon. A high-speed, revolving drum camera was used in conjunction with a cathode-ray oscillograph and a special switch gear incorporating a thyratvon. The apparatus was mounted in a truck and sufficient cable supplied to enable each current electrode to be spaced 6000 ft. from the truck. The Wenner electrode pattern was used in a ll voltage transient tests. F ie ld tests were made in the San Joaquin Valley, and details of the results are given. I t was found that none of the records shows a “ nick ” in the wave-front, and that the b u ild ­up of the current and potential transients was far too rapid to permit of correlation between the results and the known subsurface conditions. Further tests w ill be made with modifications of technique. S. E . C.

234. Locating and Detailing Fault Formations by means of the Geo-Sonograph. C. H .Johnson. Geophys., 1938, 3 (3), 273-291.—The normal type of seismograph is handi­capped in some regions on account of the confused seismograms which are obtained. This is frequently the case in the v ic in ity of faults. The cause of the confused records is a m ultiplicity of waves arriving at the detectors from the neighbourhood of the fault. The method of multiple recording assists to some extent in clarifying the records when confusion is caused by criss-cross waves from depth. I f combined detector outputs obtained in this method are plotted as rad ii vectors, then the envelope of these vectors w ill be a polar curve which represents the effective sensitivity of the detector group to waves of the reflected wave frequency. The Rieber Geo-Sonograph enables one to rotate this polar sensitivity curve to any desired direction so that one feature may be examined whilst the others are eliminated. Th is is done by obtaining records of the output of each detector and then combining them, not only in their original phase relationship, but also in any other as desired.

The application of this method of detailing faults is illustrated by three examples.S. E . C.

235. Velocity Determinations by means of Reflection Profiles. C. H . Green. Geophys., 1938, 3 (4), 295-305.—I t can be shown sim ply that to measure the velocity down to any particular horizon, it is only necessary to measure the reflection time for any two different spreads. However, since the relation between V 2 (X = spread) and

82 a ABSTRACTS.

5p2 i p = time) is linear, it is recommended that several corresponding values be obtained and the best straight line drawn through the plotted data to average out errors Although the method is indirect in dealing with second order quantities, it is believed that good reflection data w ill give results within 3% accuracy.

To obtain such accuracy several conditions must be fulfilled : (a) at least one, and preferably several reflections must be known to be continuous over the coverage of the proposed profile ; (6) weathering must be closely uniform for a ll the recorder set-ups; (c) topographic changes must be a minimum over the several recorder positions; (d ) all depth shots should be in the same material and at the same ap­proximate depth, so as to minimize the effect of variable shot hole conditions upon reflection reception times, as well as on character and frequency ; (e) reflecting horizons should be “ flat,” or at least it is desirable to be able to shoot along the strike. In order to gain the necessary information which w ill permit the choice of an area com­plying with the above conditions, preliminary work is necessary which involves the assumption of a depth-velocity relation. B y symmetrical disposition of shot and recorder positions, the importance of a “ flat ” subsurface can be minimized.

Two examples are given with details of the manipulation of the d a ta ; and the results are compared with the velocities obtained from shots at the nearest wells. The agreement is good, but it is admitted that the values are inferior to data obtained from wells. G. D . H .

236. Electrical Prospecting with Non-Sinusoidal Alternating Currents. S. S. West. Geophys., 1938, 3 (4), 306-314.—Any non-sinusoidal alternating current, of which the waveform has a sufficiently simple representation in terms of Fourier series, can be used for electrical-resistivity measurements, and should have a ll the advantages of the transient while being free from many of the practical disadvantages. Such a method bears some resemblance to the transient method, and can make use of its experimental arrangement. The best type of electrode system has four electrodes in a straight line with the detecting electrodes outside the current electrodes. How­ever, this arrangement is not so sensitive as the Wenner system to resistivity con­trasts between horizontal strata. A scheme was adopted w ith the electrodes in a line at 1000 ft. intervals. The detecting and current circuits were therefore 1000 ft. apart. This separation can be altered to meet special circumstances. A 50-cycle alternating current of square form was used. B y using a circuit of suitable design one or more cycles of the modified detected wave could be kept stationary on the screen of an oscilloscope connected to the output of the amplifier. The current source was a thyratron relaxation oscillator, controlled by a 50-cycle tuning fork. The average current was 1 -2 amp. I f the linear electrode system w ith equal spacings is moved by only a fraction of 1000 ft., the change in the transient can often be easily seen on the oscilloscope screen or its photograph. The same is true for the modified rectangular waveform. A special circuit was designed to measure the changes in the waveform, but it is not entirely suitable for absolute determinations of the waveform.

The theory of the method is briefly outlined. Although the rectangular A .C. does not usually permit the E.M .F. in the detecting circuit to reach the transient steady state, the effect upon each half-cycle is much the same as the effect on the transient, with regard to change of shape. Inasmuch as there exists no mathematical analysis yielding a solution which can predict the shape of the transient or of the modified rectangular wave for any useful cases, it is not very important what quantity is chosen as characteristic of the waveform. I t is only essential that this quantity be deter­mined uniquely by the structure of the subsurface, and be sensitive to changes in it. Hence for each prospect fixed values were chosen for a ll but one of the parameters of the balancing network, and by means of the last the E .M .F . was balanced out from approximately 60° to nearly 180° after the beginning of the half-cycle. The values of this parameter required for balancing the longest possible part of the cycle can then be plotted and contoured. An electrically anomalous closed structure has thus been determined on prospects, some of which have oil. The anomalies may indicate structures in which oil can collect. G. D . H.

237. Resolution of Combined Effects, with Applications to Gravitational and Magnetic JJata. T. A. Elkins and S. Hammer. Geophys., 1938, 3 (4), 3 15 -3 3 1.—A simple

ut rigorous and quite general mathematical method is given for finding the minimum

ABSTRACTS. 8 3 A

separation of two nearby bodies, at which their observed combined effect indicates the presence of two separate bodies. Geophysical applications of the method are illustrated by investigating the resolution of gravity and torsion balance data for the two lim iting cases of spheres and infinite horizontal cylinders, the resolution of the vertical magnetic intensity for infinite rectangular plugs and the direct inter­pretation of the infinite horizontal rectangular block. The horizontal gradient profile does not satisfy the restrictions imposed by the derivations, and consequently cannot be discussed by this general method. Indeed, the gradient is of little use as a resolution criterion.

If the probable depth of the anomalies in an area to be surveyed is known, the existence of a resolution lim it w ill yield a value of station spacing below which there is no point in going, at least from the viewpoint of detecting individual anomalies. The actual field data w ill never attain the theoretical resolving power, and allowance must be made for the precision of practical data in applying the above analysis.

G. D . H .

238. The Adjustment of Misclosures. L . G . Cowles. Qeophys., 1938, 3 (4), 332- 339.—The adjustment of misclosures by the method of least squares is accomplished by solving a system of simultaneous equations which m ay be written down by in ­spection of the traverse diagram. The solution of these equations can be effected by measuring currents in an analogous electrical resistance network. The resistances of this network are determined by the geometry of the survey, and the voltages introduced in the loops are proportional to the misclosures. The substitution of the electrical analogy and some simple measurements eliminates the necessity of the laborious calculations required by the method of least squares. The latter otherwise becomes prohibitive in extensive networks. G. D . H .

239. Three-Dimensional Reflection Control. S. M. Rock. Geophys., 1938, 3 (4), 340—348.—A pattern is presented in which A T ’s are obtained from intersecting lines of detectors. Assuming (a) plane wave fronts at the detectors, and (6) rectilinear wave propagation, formula; are presented for : (i) i/r, the angle of arrival of the reflected wave in the wave travel plane : i.e., the plane through the line of exploration and perpendicular to the reflecting plane ; (ii) 9, the angle between the wave travel plane and a vertical plane through the line of exploration ; (iii) a, the dip component in the wave travel plane; (iv) 8, the total dip ; and (v) y, the angle between direction of total dip and the line of exploration.

Application of the method to field work is described and illustrative examples are depicted. G. D . H .

240. NomogTam for Dip Computations. R . Lawlor. Geophys., 1938, 3 (4), 349- 357.—I t can be shown that the position, strike and dip of a reflecting bed are com­pletely determined by a, 8, T and V where : a = component of dip in wave travel plane; 9 = angle between wave travel plane and vertical plane containing line of exploration ; T = travel tim e ; V = average velocity of the seismic waves. The chart is applicable to all problems in which geophones are arranged in two m utually perpendicular bisecting lines, an in-line arm with geophones in the line of exploration and a cross-line arm with geophones in a line perpendicular to the line of exploration. I f no cross-line arm is used, a is the only dip component determinable. I f both lines are used, the chart may be employed to compute a and 9 from the A T'b for any spread, any geophone spread, any travel time and any velocity function provided only that the following assumptions are sufficiently accurate for interpretation of the data : (a) The average velocity is a known function of wave-travel time only, and (6) the seismic waves travel in straight lines.

The principles underlying the construction of the chart are given. B y a simple change of scales it can be adapted to all shot distances, geophone spreads and all wave velocities which are functions of wave travel time only. G. D . H .

241. Geophysical Exploration. S. F . K e lly . M in . and M etall., January, 1939, 20, 6 1 - 65.—A review of geophysical progress and activities in various parts of the world during 1938. In the United States the present trend of gravimetric work seems to be toward

ABSTRACTS.

the gravimeter, the number of torsion balance parties showing a considerable decline. The pronounced decrease in seismic crews suggests that the peak in seismic prospect­ing has been passed in the U .S.A. On the other hand the number of magnetic crews for reconnaissance on oilfield structures shows a slight increase. Brief reference is made to improvements in seismic instruments and to the new mobile electrode for continuous electrical profiling of deep structures. Photographs of some of the latest instruments are shown, including the electrical Echo-Meter which utilizes the principle of sound-wave reflection for ascertaining fluid levels in oil wells. D . W.

242. Stratigraphie vs. Structural Prospecting. E . E . Rosaire. Oil Gas J ., 22.12.38,37 (32), 43. Each exploratory method was considered the most effective of all,whilst it was operating effectively ; but each in turn has experienced the inevitable increase in costs and rapidly diminishing returns enforcing its relegation to oblivion or to a secondary position. A change in exploration technique is only justified inso­far as reward is not proportional to effort.

The Gulf Coast salt dome province is examined to illustrate the ideas. The three periods 1901-1924, 1924-1932 and 1932 to date are associated w ith three major exploration campaigns, each based on the use of one or more distinct exploration techniques : (a ) Wells were drilled on surface anomalies such as topographic highs and lows, gas seeps and parafifin dirt beds. At first the costs per discovery were low, but they rose. (6) Torsion balance and refraction methods followed, and were abandoned when the costs rose excessively, (c) The last phase corresponds with the use of the reflection seismograph. Here discovery apparently lags two years behind application. The “ difficulty of discovery ” and the reasons for rises in costs are analyzed. During the period 1930-1936 the cost per discovery has increased almost fourfold.

Surface geological exploration went through sim ilar phases and was replaced by subsurface geological methods which will, however, persist u n til a ll the geological provinces have been exhaustively wildcatted by drilling. I t is not so spectacular as the other methods.

Each period of discoveries had more successes than that preceding, but the dis­coveries were of progressively lower relief.

Structural prospecting methods depend on vertical changes in the sediments ; stratigraphical methods depend on lateral changes. The former have tended to be stressed.

The reflection seismograph approaches the d rill in penetrating power, but it does not provide such complete information. I t is, however, cheaper. The general assumption at the present time is that each new deeper discovery justifies re-shooting the whole Gulf Coast at or a little below that depth. The reflection seismograph is not used to the best advantage in reconnaissance. The same is true of the drill. The reflection seismograph should be developed to yield structural detail in advance of the drill.

There are lateral or stratigraphie changes which result from the underlying structure —induration, mineralization of shallow ground waters, locally increased seismic velocities in shallow sediments, local variations in electrical properties and haloes of leaking hydrocarbons. These evidences are independent of the depth and relief of the underlying structure. Hence the easiest way of discovering structure is to locate such a feature and later to examine it with the reflection seismograph and the drill. Numerous structures of low or zero (true stratigraphie traps) relief have been overlooked by the reflection seismograph. The proper use of stratigraphical methods w ill serve to locate practically a ll favourable structures of high and low relief.

The reflection seismograph w ill continue as a useful tool only to the extent that the observational errors now present remain, or can be made, appreciably less than the relief characteristic of the remaining undiscovered structures. G. D . H .

243. Electrical and Thermal Exploration of Boreholes. L . Migaux. B ull. Ass.franç. Tech. Pétrol., 1938, (45), 4-28.—Electrical coring consists essentially in making continuous records of resistivity and spontaneous polarization. The resistivity recorded as the electrodes pass down the hole is not the true resistivity of the forma­tions traversed, but an apparent resistivity which includes factors due to the presence

ABSTRACTS. 8 5 A

of the drillin g mud, etc. For ordinary correlation purposes this apparent resistivity is sufficient. Spontaneous polarization is due principally to electro-filtration and electro-osmose phenomena. In order to separate resistivity measurements from spontaneous polarization measurements, an A .C. supply is connected to the current electrode. Two P .D .’s are thus superimposed and picked up by the measuring electrodes, one, alternating, which measures the resistivity, and one, continuous, which measures the spontaneous polarization. These are separated and recorded independently. The magnitude of the spontaneous polarization gives a measure of the permeability of the strata. A n oil-sand gives high spontaneous polarization and resistivity, whilst a water-sand gives high spontaneous polarization and low resistivity. Strata such as dense limestones give low spontaneous polarizations and high resistivities.

I t has been found that the conductivity of strata is always greatest in the direction parallel to the bedding and the equipotential surfaces are ellipsoids. B y using two sets of measuring electrodes at right angles to each other, and symmetrically placed with respect to the centre of the borehole, the shape of this ellipsoid may be deter­mined, and hence the direction of dip of the strata. The electrodes are orientated by means of the magnetic compass, and a teleclinometer is also run with the instrument, so that corrections may be made for deviation of the borehole from the vertical. A set of measurements is made at intervals of 10-20 cms. in order to reduce the effect of small local anomalies.

Continuous temperature measurements in a borehole are made with an electrical resistance thermometer. The temperatures obtained when the borehole fluid is in thermal equilibrium w ith the strata, give useful data to the engineer and also show the junction between strata of widely differing thermal conductivities. Measurements made before thermal equilibrium is complete indicate the points of entry into the well of water and of oil and gas. Electrical coring is impossible in cased wells and tem­perature surveys are the only means of getting information. The temperature diagram is comparable with the resistivity-porosity diagram, although the details are far less well defined. The thickness and position of cement are clearly shown if a temperature survey is made several hours after completion of the cementing operation. S. E . C.

244. Work of the State Geophysical Survey in German Petroleum Areas. H . Reich. Oel u. Kohle, 1939, 15, 23-26.—An outline is given of the progress of the State Geo­physical Survey (which is stated to have already yielded very valuable results) under four headings : gravimetric measurements, torsion balance measurements, seismic measurements and magnetic measurements. For technical reasons, the work done in Austria is not included, though it is stated that a considerable amount has been carried out.

Maps are given showing the areas already covered under the four headings in the old Reich, the gravimetric and magnetic surveys being most advanced. Examples are quoted showing that each method is liable to give apparently conflicting data, and insistence is placed on the difficulty in correctly interpreting the results. For this purpose a knowledge of geology as well as geophysics is necessary, and the author stresses how essential it is that a ll geophysicists should also be geologists, and vice versa. T . T D

Drilling.245. Control of Heaving Shale by Blasting. F . R . Cozzens. Petrol. Engr, December, 1938, 10 (3), 46.—W axy shales that behave much like quicksands are encountered at about 600 ft. in the Ohio, West V irg in ia and Kentucky fields. The moment an opening is made the walls slip and crumble. The thickness of the shale ranees from 30 to 70 ft.

Explosives are being used in some oil-producing districts to control the grainless shales. I t is found that instead of crumbling the walls the force of the blast stiffens them, permitting drilling to proceed without interruption.

A torpedo was made from an 8-ft. section of metal rain pipe, closed at the bottom and provided with a bail in the top. Twelve cartridges of ammonia dynamite, 4 0 %

86 a ABSTRACTS.

in strength, were loaded into the shell. Into the top cartridge was inserted a No. 6 electric blasting cap, which was connected by an insulated copper wire sufficiently long to reach to the firing battery.

The charge was fired and the débris was removed by a suction bailer. D rilling was continued for a further 15 ft. and a second torpedo containing ten cartridges was then exploded. The walls stood up until the casing was set.

The principle is to stem the charge in such a manner that sufficient force is exerted against the side walls to make them firm and sear them for a short time.

L . V . W . C.

246. World’s Longest String of 9 |- in . Casing Set. K . C. Sclater. Petrol. Engr, January, 19 39 ,10 (4), 21.-—The longest string of 9f-in . casing ever to be run in a well has recently been set in the Gulf Coast. Th is string weighed approximately 253 tons at a depth of 11,631 ft.

The well is situated in the shallow water on the Louisiana Coast and is being drilled from the Texas Company’s specially constructed submersible barge, on which is mounted the derrick and the drilling rig, steam being used as a source of power.

Extreme care was exercised in conditioning the mud and before pulling the drill pipe the weight of the mud was 1 1 T lb. per gal. viscosity 5 1° A .P .I., sand content 2 % , salt 3150 parts per m il., temperature at flow line 123° F ., and solids 3 7 % by weight.

The casing was set in less than 24 hrs., and with no serious trouble.L . V . W . C.

247. Under-reaming Reduces Cost of Drilling and Deepening Wells in California. W. R.Roulston. Petrol. Engr, January, 1939, 10 (4), 27.—The Montebello field in California is underlain by an unknown number of possibly productive deep zones. Deepening is an advantage in some wells producing from upper formations. When production from upper formations is disappointing under-reaming has been used successfully and has saved a considerable amount of money in the cost of drilling and completing the well.

A hole that is under-reamed to provide clearance for the largest liner that w ill pass through the casing w ill permit the use of smaller water strings and surface pipe and, therefore, smaller bits from the surface to the total depth of well. L . V . W. C.

248. Chemical Control of Heaving Shale. C. L . Baker and A. D . Garrison. Petrol. Engr, January, 1939, 10 (4), 50.—Heaving or caving of the hole has prevented deep drilling in certain areas. Th is phenomenon may result from the establishment of a serious pressure differential in the hole or from hydrous disintegration of “ heaving shale.” Th is heaving can be avoided by using a drilling mud rich in silicate of soda and sodium chloride. The requirements in such a mud, its formulation and use in the field are discussed with a brief description of the equipment required and special problems that must be met. Heaving due to hydrous alteration of shale has been prevented when drilling has been carried out using this new method. L . V . W. C.

249. Improvements in Methods of Gravel Packing Oil W ells. W . A . Sawdon. Petrol. Engr, January, 1939, 10 (4), 95.—The general method of gravel packing in use in California is the reversed circulation method. Most of the gravel packing has been done in new wells, although in the Bakersfield area gravel has been used to repair a few wells.

In the two general practices now being followed, using the reverse circulation method, the gravel is carried in a mud fluid down the annular space between the casing and the tubing on which the liner is run and supported during the operation. When the fluid and gravel reach the liner they continue downward outside the liner. The gravel builds up around the liner and the carrying fluid enters the liner at the bottom returning to the surface through the tubing.

Centralizing guides aid m aterially in distributing the pack uniformly around the liner.

An adjustable nipple and gravel-mixing trucks are recent developments.L . Y . W . C.

ABSTRACTS. 8 7 a

250. Drilling with Reverse Oil Circulation in Wider Use. E . J . Beckman. Oil Wkly,5.12.38, 91 (13), 19.—Considerable work has been done during the last two years in developing the reverse circulation method of drilling. W ith this method it is possible to (i) d rill new wells, (ii) drill, plug and wash well, (iii) d rill out cement plugs, (iv) underream while drilling, (v) underream and d rill new holes in salt water disposal wells, (vi) clean out wells quickly and (vii) perform special jobs such as rolling out collapsed casing, side tracking, etc. Many advantages are gained by its use.

The method is sim ply rotary drilling using a pack-off head at the top of the casing in order to circulate the drilling fluid in the reverse direction of conventional methods.

Accurate logging can be accomplished, and the velocity of the fluid up the drill pipe is greater than is possible with the normal method, and a sample of the cuttings for observation is available in much less time. L . V . W . C.

251. Rotary Conquers Michigan’s Traverse Lime. N. X . Lyon. Oil W kly, 12.12.38, 92 (1), 17.—The first deep well in Michigan to be drilled by rotary has recently been completed and has shown a marked saving in drilling time over cable tools. No special drilling muds were used, and a density of 9-7 of 0-3 viscosity was carried on this test. L . V . W . C.

252. Survey of Cementing Jobs to Shut off Salt Water in East Texas. P. Montgomery. Oil Wkly, 12.12.38, 92 (1), 34.—This article is an account of the various types of cementing jobs used to shut off salt water in the east Texas oilfield.

These jobs are classified into four groups with a total of twelve different types or methods and each is dealt with in some detail.

As a result of this investigation it may be stated that :

1. The method best adapted to the conditions should be used. There is no one kind of job that can be recommended for any and all wells.

2. In wells making a high percentage of salt water, retainer squeeze and short- string squeeze jobs are recommended.

3. The retainer squeeze job w ill be most used owing to the light-weight casing used.

4. The braden head squeeze job has been successful where the cement job is done shortly after the well has started to make salt water.

5. High-pressure squeeze jobs w ill be much more permanent.6. Slow setting gel-forming cements have assisted considerably in improving

the method of shutting off salt water. L . V . W . C.

253. Care and Maintenance of Instruments Used on Drilling Rigs. E . L . Decker and N . L . Dorn. Oil Wkly, 19.12.38, 92 (2), 42.— D rilling crews are unprepared to handle the instruments necessary to progress w ith the advances made in rotary drilling technique.

This article brings out the need for care and maintenance routines, and reports are quoted to show the benefits to be gained from such programmes. Fixed re­sponsibility, instruction of crews and steps to m inim ize the occurrence and effects of abuse are recommended. L . V . W . C.

254. Drilling Patents. A . G. Loomis. U .S.P. 2,143,990, 17.1.39. Appl. 27.11.34.A method of differentially shutting off water in oil wells penetrating both oil and water without permanently hindering the flow of oil into the well, consisting of the use of two aqueous solutions, the first containing an acid and the second a reagent, which on admixture with the acid, reacts to form a water-insoluble, oil-soluble plastic. The plastic is caused to form in the interior of the water formation.

A. G. Loomis. U .S.P. 2,143,991, 17.1.39. Appl. 27.11.34. A method of shutting off water in brine-bearing formations which comprises injecting into the formations in liquid form a water-soluble soap of the polycyclic naphthenate type, the calcium and magnesium derivatives of which are soluble in oil. The soap is caused to mingle with the brine in the brine-bearing formations to form insoluble sealing deposits.

J. J. Bishop. U .S.P. 2,144,100, 17.1.39. Appl. 11.2.36. Rotary jar.

ABSTRACTS.

C C Taylor. U.S.P. 2,144,669, 24.1.39. Appl. 16.3.35. Bailer having meansfor opening the valve controlling the discharge port.

J W Pippin U .S.P. 2,144,687, 24.1.39. Appl. 4.1.37. D rilling apparatus comnrising a tubular drill stem, an assembly adapted to be lowered through the stem a n d including a drilling cutter pivoting about a transverse axis. Means provided for actuating the cutter into transverse position to engage the lower end of the stem upon upward movement of the assembly.

J.M . Kendall. U .S .P . 2 ,1 4 4 ,7 6 2 , 2 4 .1 .3 9 . Appl. 2 9 .9 .3 7 . Recording thermometer for use in boreholes comprising a casing, a plug closing one end of the casing, a heat- insulating tube, a bulb attached to the free end of the tube, a coil of Bourdon tubing, having one end closed and the other fixed to the plug, and means for recording the movement of its free end and a supply of volatile liquid in the bulb.

E . F . Raymond. U.S.P. 2,144,810, 24.1.39. Appl. 27.4.35. Adjusting device for drilling jars.

G. M. Boulter. U.S.P. 2,144,869, 24.1.39. Appl. 20.3.36. Rotary jar.

H. R. Standlee. U .S.P. 2,144,944, 24.1.39. Appl. 18.1.37. W ell swab.

R . Flagg. U .S.P. 2,145,168, 24.1.39. Appl. 21.10.35. Method of treating threaded pipe connections to obtain a leak-proof lock connection.

W. G. Frenzel. U.S.P. 2,145,170, 24.1.39, 30.1.36. Core barrel.

H . Pennington. U.S.P. 2,145,191, 24.1.39. Appl. 16.7.29. Core barrel whichcan be retrieved by means of circulating fluid.

H. Pennington. U .S.P. 2,145,192, 24.1.39. Appl. 12.1.37. Tubing catcher sub for use with retrievable core barrel. L . V . W . C.

Production.255. Use of Gas Lift in Handling Small Allowables. S. F. Shaw. Oil Wkly, 19.12.38, 92 (2), 38.—Gas lift has now become a popular method of handling small allowables in many fields and entails a low capital expenditure per well for installation of equip­ment, and reduces the lifting cost to a low point.

In new fields the method is favourable for the following reasons: (1) where high flowing pressures are available, high fluid levels are found; (2) gas oil ratios obtain that are at least as high as those originally associated with the oil before the reservoir was tapped; (3) small allowables require low cost equipment; (4) centralizedmachinery may be installed; and (5) infrequency of pulling and other well jobs.

L . V . W . C.

256. Measurements of Rates of Gas Delivery Under Conditions of Non-Critical Flow and Critical Flow. B. Altman. Oil Wkly, 26.12.38, 92 (3), 24.—The instruments most commonly used in calculating open flow of gas-wells are : pitot tube, choke nipple, critical flow provers, the back pressure method, funnel meter, orifice well tester and orifice meters. The first quoted four methods are suitable for critical flow measurements, whilst the remainder are suitable for non-critical flow. The funnel meter is better adapted to use on small wells, since the maximum rate of flow that can be measured with it is 8,000,000 cu. ft. per day.

The uses of each of these instruments is explained in detail, and diagrams are given of a number of them. L . V . W . C.

257. East Texas Experiment Shows Feasibility of Returning Salt Water to Producing Formation. H . H . King. Oil Wkly, 28.11.38, 91 (12), 23.—Input wells situated on the water drive flank of the east Texas field may be used to take the huge volume of salt water that must be handled in draining this field.

Steps are being taken to instal input wells in order to be prepared when the produc­

ABSTRACTS. 8 9 A

tion of brackish water can no longer be accommodated in the surface pits. Facilities for treating the water to cleanse it of foreign compounds that would clog the exposed sand have enabled 3000 brls. of water d aily to be returned to the input well already in existence. The brine flows by gravity from a gun barrel through a baffle tower and thence to a receiving tank. The water is delivered to a filter box above the tubing head on the well.

Experiments are now under way to coagulate the fine particles of foreign matter carried by the brine to accomplish greater precipitation and at a faster rate. Tests are to be made w ith slaked lime and, also, chlorine. L . V . W. C.

258. Salt Water Disposal System in the Fitts Pool. J . C. Albright. Petrol. Engr, December, 1938, 10 (3), 66.—A mutual co-operative association has been formed in the Fitts Pool to deal with the quantity of salt water being produced in the field.

The first salt water produced was collected in large reservoirs and pools and per­mitted to soak into the soil and to evaporate into the atmosphere. Fear of polluting the streams impelled the operators to develop the disposal system.

Core analysis of a ll the formations indicated that water could be pumped into the Wilcox and O il Creek sands in large quantity, if filtered.

A gravity drainage system to transport the brine to a central reservoir was next designed.

About 11,000 brls. of brine per day are pumped at a pressure of about 250 lb. No pressure build-up has been noticed showing that filtration has clarified the water sufficiently that the sand face is not plugged to any noticeable extent.

l . y. w. c.

259. Gas Lift Project in East Texas Utilizes Gas from Distillate Well. J . W . Lee andF . H . Love. Petrol. Engr, December, 1938, 10 (3), 74.—The gas from a wildcat high- pressure well has been piped to a nearby field for gas-lift purposes. This gas, which would otherwise be v irtually useless, is made to produce oil economically.

The gas is separated from the distillate and metered before entering the line. The well pressure, which is 1000 lb., is reduced to 350 lb. at the injection well by means of high-pressure regulators. The gas is injected into the annular space between the tubing and the casing.

Flow valves are an important feature, and from five to eight valves are installed in the tubing string at intervals ranging from 200 to 350 ft., the top valve at about thestatic fluid level. L . V . W. C.

260. Effects of Surface Phenomena on the Production of Oil. H . K . Livingston.Petrol. Engr, January, 1939, 10 (4), 84.—T h is article describes an investigation of surface phenomena of liquids and solids, and shows that interfacial tensions appreciably affect the rate of movement of fluids in the reservoir. L . V . W . C.

261. Determination of the Potential Productivity of Oil-bearing Formations by Re­sistivity Measurements. M. M artin, G. H . Murray and W. J . Gillingham. Geophys., 1938, 3 (3), 258-272.—The resistivity of an oil-sand depends on the amount of saline water in the sand. I t has been observed that this may be as high as 3 0 % , even in the case of a sand producing pure oil. Resistivity measurements were made in the laboratory on sands in which there were varying proportions of salt water and oil. I t was found that the curve relating electrical resistance to oil content did not rise steeply until the percentage of oil was approximately 8 0 % . Thus, rich oil-sands may have a comparatively low resistance and, furthermore, large variations in re­sistance are associated with only small changes in oil saturation. This relationship enables one to obtain a measure of the oil saturation of a sand, provided its true resistivity can be measured. In electrical logging apparent resistivity is obtained and this is not sufficiently accurate for the present purpose. B y making measure­ments opposite the same bed with successively increased electrode spacing, apparent resistivity m ay be plotted against electrode spacing. From the curve so obtained the true resistivity of the bed m ay be obtained. Experience has shown that only a few of these curves need to be drawn for a given district in order to determine the

G

9 0 aABSTRACTS.

best electrode spacing. A ll other determinations of true resistivity in the district are then single m e a s u r e m e n t s at that spacing. The method can be used only for beds of greater thickness than 10 -15 ft. c -

262 Application of Physico-chemical Principles to the Investigation of the Properties nf ¿neks Part III Comparison of Methods and Conclusions. A. H . Nissan, C. E . Wood L V W Clark and A. W. Nash. J . Instn Petrol. Tech., 1938, 24, 585-597— The two methods for determining porosity described in Parts I and I I are compared and the overall accuracy of each method is calculated and demonstrated experimentally.

D . L . S.

263 Gas Lock Protective Devices and Mobile Rotary Pumps Used on K-M-A Gathering Systems J. C. Albright. Oil Wkly, 2.1.39, 92 (4), 21.—The design of the gathering system of the Texas Pipe Line Co. in the K -M -A field of north Texas and the use of mobile rotary type pipe line pumps has enabled the Company to move a much greater amount of oil than is ordinarily possible.

A ll pipe line companies at K -M -A have built gravity systems as far as practicable, but because of the gassy characteristic of the oil, oil firms are faced with gas locking in their systems. In addition air in the feeder lines from the tank batteries is as troublesome as the gas.

The conventional method of handling oil is to allow gravity to move the oil as far as practicable. At the station side of the lines pumps are frequently used that pull a high vacuum on the oil in an attempt to accelerate the movement of oil. K -M -A crude makes this almost impossible, for when this is attempted the oil invariably breaks into gas pockets. Automatic air release and vacuum valves have been installed. By slowing down the permanently installed pump and by having a slight head on the intake the oil would move faster from the batteries than if a good vacuum was pulled.

Mobile rotary pump units have been developed. The pump is mounted on a light maintenance truck, power being taken from the transmission take-off of the truck.

L . V . W . C.

264. High Pressure Gas From Another Field Flows East Texas Wells. Anon. OilWkly, 2.1.39, 92 (4), 26.—High pressure gas is being transported a distance of about ten miles in the east Texas field for the purpose of gas-lifting production. A six- mile branch is run to other properties where four different makes of gas lift valves or intermitters were tried in a number of wells.

The gas from the gas well had an in itial pressure of 2770 lb. at the well and is carried to the other properties at 750 lb. after it has been passed through a separator.

A recent survey shows that over 100 wells are now operating on gas lift from this source. L . V . W . C.

265. Fluid Level Indicator is Useful in East Texas Field Operations. G. Weber. OilOas J ., 15.12.38, 37 (31), 44.—A number of new methods for testing and equipping wells in which natural flow has ceased have been introduced in the east Texas field due to the increased use of artificial lift. Knowledge of fluid levels, bottom hole pressure, build-up and other characteristics are needed in fitting the well for best results in secondary production.

The new procedure employs the use of sound wave reflections to determine the depth of the fluid level in the well tested.

Equipment for conducting tests of this type comprises a heavy duty well-head connection incorporating a gun and a sensitive microphone, an amplifier equipped with tuning and filtering provisions and a pen recorder.

Since speed of sound varies in different wells, depending upon the density and characteristics of the gas medium in the column, depth determinations are greatly simplified by rising the tubing collar reflections for depth calibrations. The average tubing length is derived from well records, and by direct measurement on the ribbon, o a length including a certain number of tubing collar reflections, the basis for the calculations is established. y \y (j

266. Magnolia Water Project Represents Advance in Disposal Technique. P. Reed.i (is , 22.12.38, 37 (32), 30.—Several unique features are included in Magnolia

ABSTRACTS. 91 A

P e tro le u m C o.’s s a l t w a te r d isp o sa l p ro je c t in th e F i t t s P o r t , O k lohom a. A m ong th e m ore im p o r ta n t a re : (i) am p le p ro v is io n s fo r s e p a ra tin g o il fro m b rin e before i t reaches th e t r e a t in g p l a n t ; (ii) p ro v is io n s fo r c lean in g lines b y p u m p in g sc rape rs th ro u g h th e lin es w h ile o p e ra t in g u n d e r p re ssu re o r b y g ra v ity f lo w ; (iii) g a th e rin g lines w ith p ro v is io n s fo r in c re a s in g d e liv e ry c a p a c i ty ; (iv) th e p ip e in th e g a th e rin g lines is o f m a te r ia l t h a t w ill n o t be co rro d e d b y e i th e r s a l t w a te r o r th e s o i l ; (v) th e line is m a in ta in e d a t a d e fin ite g ra d ie n t th ro u g h o u t ; (vi) a d d it io n a l conn ec tio n s m ay be m ad e a t a n y t im e ; (v ii) re se rv o ir s to ra g e is m u c h sm a lle r th a n u s u a l ; (v iii) t r e a t ­ing p la n t is f le x ib le ; (ix) th e d isp o sa l w ell w ill ta k e la rg e q u a n ti t ie s of w a te r und er low p ressu res a n d sh o u ld c o n tin u e to do so fo r a long p e r io d ; a n d (x) th e eq u ip m en t is su ch t h a t a h ig h p e rc e n ta g e c a n be sa lv ag ed w hen n ecessa ry . L . V . W . C.

267. Production P atents. C. T . A n d e rso n . U .S .P . 2 ,143,836, 17.1.39. A p p l.30.8.37. G as a n c h o r .

G . T . H u m p h re y . U .S .P . 2 ,143,945, 17.1.39. A p p l. 2.11.36. C ran k fo r w ell p u m p .

B . H . S to n e . U .S .P . 2 ,143,962, 17.1.39. A p p l. 14.12.36. S u b te rra n e a n flu id flow m ea su rin g dev ice , in w h ic h p a c k in g m e a n s a re a v a ila b le to p ro d u ce non-flow a n d open-to-flow a re a s . P lu r a l i ty of m e te rs w ith in th e d ev ice to m easu re th e flow of flu id in to th e w ell fro m e a c h o p en -to -flow a re a .

T . S. P a rk . U .S .P . 2 ,144 ,026 , 17.1.39. A p p l. 6.2 .36. W ell p ack e r.

W . F . Cox. U .S .P . 2 ,144,045, 17.1.39. A p p l. 16.10.36. W ell p u m p .

J . C. H e w it t , J r . , a n d V . E . K u s te r . U .S .P . 2 ,144,061, 17.1.39. A p p l. 15.2.35. A w ell-su rvey ing a p p a r a tu s co m p ris in g a c a tc h e r p la c e d in a fixed p o s itio n w ith a w h ipstock fixed in k n o w n o r ie n ta t io n w ith th e p ip e . A w ell-su rvey ing in s tru m e n t is low ered in to th e d i i l l p ip e to a fixed p o s it io n o f re s t w ith th e c a tc h e r . M eans p ro v id ed for in d ic a tin g th e o r ie n te d p o s it io n o f th e in s tru m e n t w ith in th e casing .

C. C. C rickm er. U .S .P . 2 ,144,144, 17.1.39. A p p l. 5.10.35. M eans fo r ra is in g liq u id fro m w ells.

A. J . P en ick a n d K . T . P e n ic k . U .S .P . 2 ,144,227, 17.1.39. A p p l. 14.2.36. C om ­b in a tio n w ell-h ead a n d h a n g e r .

A. J . P en ick a n d K . T . P e n ic k . U .S .P . 2 ,144,228, 17.1.39. A p p l. 16.7.36. C on tro l valve m ech an ism fo r w ell-h ead s.

J . R . D av id so n . U .S .P . 2,144,403, 17.1.39. A p p l. 28.10.38. O il saver.

F . A . T h ah e ld . U .S .P . 2 ,144,420, 17.1.39. A p p l. 21.3.36. W ell p ack er.

A. A nderson . U .S .P . 2 ,144,422, 17.1.39. A p p l. 7.8.35. S u rv ey a p p a ra tu s fo r ta k in g p h o to g ra p h s o f th e h o le a t a n y d e s ire d p o in t .

F . S tone a n d G. S. K n o x . U .S .P . 2 ,144,818, 24.1.39. A p p l. 23.5.36. W eU -pipe p lug .

C .C C rickm er. U .S .P . 2 ,144 ,833 ,24 .1 .39 . A p p l. 2.11.36. W ell flow ing a p p a ra tu s .

D. H an es . U .S .P . 2 ,144,842, 24 .1 .29 . A p p l. 27.4.37. A p a c k e r a d a p te d to be used in a b o re ho le a n d m e a n s fo r s e t t in g a n d re leas in g i t in a well.

B . N ay . U .S .P . 2 ,145,190, 24 .1 .39 . A p p l. 24.5.38. Sw ivel jo in t b a rd o il rem o v er fo r tu b in g s tr in g s . 1-. V . W . C.

Transport and Storage.268. Oil Meters. L . G . E . B ig n e ll. O il Gas J ., 5.1 .39, 37 (34), 39.— O il m e te rs fo r m e te rin g c ru d e o il a n d i t s p ro d u c ts h a v e been d ev e lo p ed w ith in th e p a s t few y ears . T hese can be g ro u p e d in to tw o g en e ra l ty p e s a n d classified as q u a n t i ty m e te rs o r ra te - of-flow m e te rs a n d r a te m e te rs .

I n th e q u a n t i t y -m e te r c lass th e re a re tw o ty p e s— p o s itiv e m e te rs a n d in fe re n tia l

a b s t r a c t s .

m eters I n th e firs t th e flow is d iv id e d a u to m a tic a l ly in to s e p a ra te iso la te d p a r ts eq u a l b y w eigh t or b y vo lum e. I n th e in fe re n tia l m e te r th e flow is “ in fe rre d ” from th e ac tio n of th e s tre a m on th e p r im a ry e lem en t o f th e m e te r .

C urren t m ete rs m ay also be d iv id ed in to tw o k in d s , th e free im p e lle r ty p e a n d th e confined im peller ty p e . I n th e firs t c lass th e im p e lle r is p la c e d in th e c e n tre of th e cross-sectional a re a of th e c o n d u it, w h ils t in th e seco n d th e c as in g is so a r ra n g e d th a t a ll th e liqu id , a f te r p assing th ro u g h s tra ig h te n in g v an es , is fo rce d th ro u g h th e im peller.

T here is also a w eigh ing ty p e of m e te r . V o lu m e -ty p e m e te rs m a y b e classified in to m u ta tin g -d isc , o sc illa tin g -p is to n ro ta ry a n d p is to n m e te rs . L . V . W . C.

269. Correlation of an Electrolytic Corrosion Test w ith the A ctual Corrosiveness of Soils. I- A. D enison a n d R . B . D am ie lle . Bur. Stand. J . Res., Wash., 1938, 21 (6 ),819-830. T he e lec tro ly tic t e s t fo r m ea su rin g th e co rro s iv en ess o f so ils co n s is ts ind e te rm in in g th e p o la r iz a tio n v o lta g e a t v a r io u s c u r re n t d e n s itie s of a specia lly d e ­signed cell in w hich th e e lec tro d es a re s te e l a n d th e e le c tro ly te is th e so il u n d e r te s t.

T he corrosiveness of so ils a lo n g a 128-m ile se c tio n of a p ip e - lin e sy s te m w as e s tim a te d from d a ta on th e occurrence of leak s a n d le n g th o f lin e r e c o n d itio n e d a n d from th e re su lts of th e e lec tro ly tic co rros ion te s t .

I t a p p ea rs t h a t th is co rrosiveness decreases a s d ra in a g e im p ro v e s a n d increases as th e soils becom e h eav ie r in te x tu re .

I t w as found t h a t a ro u g h l in e a r c o r re la tio n e x is te d b e tw e e n th e a c tu a l corrosive­ness of th e so ils a n d th e re su lts of th e la b o ra to ry e le c tro ly tic te s t . D . L . S.

Crude Petroleum.270. Chemical and Refining Study of som e W yom ing B lack Oils. H . M. T h o m e a n d W . M urphy . V .S. Bur. M ines. Rep. Invest. No. 3423. N o v e m b e r , 1938.— I n general these crude oils a re ch a ra c te riz e d b y th e ir h ig h su lp h u r , h ig h a s p h a l t a n d low gasoline co n ten t, a n d a re ca lled b la c k o ils becau se of th ese p ro p e r t ie s a n d th e i r co lou r. Crudes from O regon B asin , D a lla s -D e rb y a n d G a rla n d F ie ld s h a v e th e fo llow ing p ro p e rtie s : sp . gr. 0-924—0-940, su lp h u r c o n te n t 2 -89-3-25% , v isc o s ity S .U . a t 100° F . 260-530 secs., p o u r p t . below — 5° F ., c a rb o n re s id u e 8 -0 -10 -2% . T h e d e m a n d fo r th e crudes is a t p re se n t la rg e ly co n tro lle d b y th e r e q u ire m e n ts o f a s p h a l t a n d ro a d o il p ro d u c ts in th e m a rk e t te r r i to ry . W ith a v iew to in c reas in g p o ss ib le o u t le ts fo r th e crudes, c rack in g s tu d ie s h av e b een co n d u c te d a t th e L a ra m ie s ta t io n o n to p p e d c ru d es under d ifferen t co n d itio n s of p ressu re a n d te m p e ra tu re , a n d th e re fin in g o f l ig h t d is tilla te s to m a rk e t s ta n d a rd s ex am in ed . T h e s tr a ig h t- ru n n a p h th a s p ro d u c e d b y to p p in g (10 -18% y ield ) a re defic ien t in lig h t f ra c tio n s , h ig h in s u lp h u r c o n te n t a n d low in o c tan e ra t in g (31-34). T h ey th e re fo re re q u ire b le n d in g w i th s u ita b le n a p h th a s or to be “ refo rm ed ” o r c rack ed a n d su b se q u e n tly t r e a te d . C rack in g of th e to p p ed c rude in a special la b o ra to ry b a tc h e q u ip m e n t y ie ld e d n a p h th a s h a v in g o c ta n e ra tin g s of 48-61, acco rd in g to th e te m p e ra tu re a n d p re ssu re u til iz e d , of goo d v o la t i l i ty b u t c o n ta in in g a b o u t 1% of su lp h u r in a ll cases. O p tim u m y ie ld o f c ra c k e d n a p h th a b ased on cru d e is a b o u t 4 5 % . R efin in g b y ac id to su lp h u r c o n te n t o f 0 -1% requ ires excessive q u a n titie s of ac id a n d low ers th e o c ta n e n u m b e r o f th e c ra c k e d gasolines app rec iab ly . A m ore co m p le te in v e s tig a tio n o f th e re fin in g p ro b le m is p lanned . A p p ro x im ate ly one th i rd of th e su lp h u r in th e c ru d e w as c o n v e r te d to H 2S, w hich m ay be rem oved a n d co n v e r te d to ac id . C onsid erab le q u a n t i t ie s o f non -condensib le gases w ere p ro d u ced , a n d th e y ie ld of coke w as u su a lly a b o u t 25—3 0 % . T h e su lp h u r c o n te n t of th e coke w as u su a lly a b o u t 5 % o r m o re , w h ich w ill cau se l i t t le d ifficulty for fuel pu rposes , b u t m a y be d e tr im e n ta l fo r c e r ta in m e ta llu rg ic a l p u rp o ses . As a re su lt of th e w ork so fa r c o n d u c te d i t is p ro p o se d to in v e s tig a te th e a l te rn a t iv e m ethod of p roducing specification a s p h a lts fro m th e c ru d e s a n d c ra c k in g th e in te rm e d ia te d istilla te s . T he cru d es a re p a r t ic u la r ly w ell su ite d to th e m a n u fa c tu re of road- b in d in g m a te ria ls a n d o th e r a s p h a ltic p ro d u c ts o n a c c o u n t o f th e i r h ig h asp h a lt co n ten t. p

271. Refining Value of Foreign Crudes. G. E g lo ff, J . C. M orell a n d G . B . Z im m erm an . J 1 29.12.38, 37 (33), 74.— A naly ses a n d y ie ld s fro m la b o ra to ry d is t i l la t io n ofth e follow ing crudes a re g iven :

ABSTRACTS. 93 A

(1) P oza R ica , M exico.(2) C o lo m b ia (h igh co ld t e s t a n d low co ld te s t) .(3) T u rn e r V alley (44-0° a n d 49-2° A .P .I .) a n d W a in w rig h t, A lb e r ta .(4) O il f ro m A lb e r ta T a r S a n d s .(5) C om odoro , A rg e n tin a .(6 ) G bely (E gbell) , C zecho slo v ak ia .(7) L isp e i a n d a 50 /5 0 b le n d of L isp e i a n d B u k k szek i, H u n g a ry .(8) D h u lia n , B r i t . I n d ia . C. L G

272. Laboratory Method for E valuating Crude Oil, w ith Special Reference to Trinidad Crude. R . G . J o h n s to n e a n d R . P a lm e r . J . Instn Petrol. Tech., 1938, 24, 6 0 5 - 620.— A la b o ra to ry p ro c e d u re is d esc rib ed fo r e v a lu a tin g c ru d e o il in te rm s of th re e fra c tio n s— gaso lin e , g a s o il a n d fu e l o il. D , l . S.

273. Patent on Crude Oil. C. S. C erf. U .S .P . 2 ,140,574, 20 .12 .38. A p p l. 16.8.37. P rocess of d e sa ltin g c ru d e o il. \ y g p; q .

Gas.274. Patent on Gas. D e D ire c tie n V a n de S ta a tsm ijn e n a n d C. O tto & Co. E .P . 497,330, 14.12.38. A p p l. 14.6.37. R e m o v a l o f H 2S fro m gases b y m ean s of a n am m o n iaca l so lu tio n c o n ta in in g F e co m p o u n d s , w h ich is re g e n e ra te d b y m ean s of 0 2 a f te r w ash in g . W . S. E . C.

Cracking.275. 85% of G asoline of 81 Octane (number) from Gas Oil. W . T . Z iegenhain . Oil Gas ./ . , 8.12.38, 37 (30), 23.— A sem i-co m m erc ia l scale c a ta ly t ic c rack in g u n i t in o p e ra tio n a t th e re se a rc h p la n t of th e U n iv e rsa l O il P ro d u c ts Co. c an p ro d u ce 85% of gasoline of 81 o c ta n e n u m b e r fro m M id -C o n tin en t gas o il of A .P .I . g ra v ity 36-7°, I .B .P . 407° F ., F .B .P . 750° F ., V ise . S .U . a t 100° F . 40 secs. T h e gaso line p ro d u ced is s ta b le in re sp e c t o f co lo u r a n d o c ta n e r a t in g a n d h a s a v a p o u r p re ssu re of 10 lb. T he y ie ld of gaso lin e is b a se d u p o n a recycle o p e ra tio n , a n d inc ludes t h a t o b ta in e d from c a ta ly tic p o ly m e riz a tio n o f th e c ra c k e d gases. T h e p la n t co n sists e ssen tia lly of a h e a te r , tw o b a n k s o f c a ta ly s t re a c to rs , a n d a u to m a tic c o n tro ls fo r a l te rn a tin g th e flow o f h e a te d o il f irs t th ro u g h one h a lf of th e re a c to rs an d th e n th ro u g h th e o th e r h a lf w h ils t th e c a ta ly s t in th e f irs t is b e in g re a c tiv a te d . R e a c tiv a tio n is accom plished b y p u rg in g th e re a c to rs o f o il v a p o u rs , a n d b u rn in g off th e c a rb o n w h ich co llec ts on th e c a ta ly s t b y p a ss in g a p re d e te rm in e d a m o u n t of a ir th ro u g h th e tu b e m a in ta in in g a m o d e ra te te m p e ra tu re of co m b u s tio n . T w e n ty ty p e s of ch a rg in g s to c k s , ra n g in g from resid u es to g a s o ils, h a v e b een in v e s tig a te d in se lec tin g th e m o s t p ra c tic a b le c a ta ly s t w h ich show ed p ra c tic a lly c o m p le te reco v e ry o n r e a c tiv a tio n o v er long p erio d s of tim e in la b o ra to ry te s ts . H ig h ly para ffin ic o v e rh e a d d is t i l la te s a re th e m o st su ita b le feed s to c k s . R . A . E .

276. D eterm ination of Yield per Pass, Time of Treatm ent and “ In Situ ” D ensity and Instrum entality for their Control in Com mercial Cracking U nits. R . L . R u d e , R . D . Je n k in s a n d C. B a rn e s . Refiner, N o v em b er, 1 9 3 8 ,17 (11), 583.— I t is n ow recogn ized th a t th e t im e -te m p e ra tu re re la tio n sh ip is th e m o s t im p o r ta n t o f th e c o n d itio n s g o v e rn in g th e y ie ld of a c ra c k in g u n i t a n d t h a t th e p ro p e r t ie s of th e m a te r ia l in a n y se c tio n of th e u n it m u s t be k n o w n w ith a m in im u m of d e lay . A ch an g e in p ro p e rtie s n ec e ss ita te s a change in p la n t c o n d itio n s if th e p ro p e r tie s of th e final p ro d u c t a re to re m a in u n ­changed . I t h a s b een e s ta b lis h e d t h a t th e d e n s ity of th e m a te r ia l c an be u se d a s a n ac c u ra te g u id e of i t s re le v a n t p ro p e r t ie s , a n d in s tru m e n ts h a v e b een d ev ised fo r th e d e te rm in a tio n o f d e n s itie s “ in s i t u ,” t h a t is co n tin u o u s ly a n d w ith o u t reco u rse to th e w ith d ra w a l of sa m p le s. T h e re is th u s n o t im e la g in th e p ro d u c tio n of in fo rm a tio n . T he d e n s ity m e a s u rin g d ev ice is d esc rib ed in g re a t d e ta il . I t co n s is ts of a n orifice

g 4 A ABSTRACTS.

connected to a d iffe ren tia l h e a d m e te r . T h e w e ig h t r a te of flow of a f lu id p ass in g th ro u g h a p ip e a n d orifice of k n o w n d es ig n c a n b e e x p re sse d th u s :

Qjyi __IF = 0 0997 v f — = Vdh„

w here I F = r a te of flow in lb . p e r sec., C = coeffic ien t o f d isc h a rg e , D = d iam . of

orifice in in ., 1 - = ap p ro a c h fa c to r , hu = d if fe re n tia l h e a d in in . w a te r , d =’ V i — jS1

d en s ity of th e flow ing flu id , in lb . / f t .3. T h e t im e of d e te n t io n of a f lu id in a zone of know n d im ensions can be c a lc u la te d th u s

md . Vt = cIF

w here t = tim e of d e te n tio n , md = m e a n d e n s ity of th e f lu id , V = v o l. of th e zone, IF = w eigh t r a te of flow a n d c = c o n s ta n t . H . G.

277. Cracking Pennsylvanian Gas Oil for 90 O.N. B lending V alue G asoline. A. L.F o ste r. Nat. Petrol. News, 14.12.38, 30 (50), R . 594.— I n th e c ra c k in g of P e n n sy l­v an ian gas oil b y th e T ru e V ap o u r P h a se P ro c ess co n s id e ra b le im p ro v e m e n ts in yield and o c tan e n u m b e r h a v e b een ach iev ed b y recy c lin g th e g as d isc h a rg e d b y th e com ­presso r-coo ler sy s tem a n d m a d e u p of g as fro m th e gaso lin e a c c u m u la to r a n d stab ilizer to p . P a r t of th is gas is p assed in to th e cycle g as h e a te r a n d p a r t is s e n t th ro u g h a “ flash gas ” h e a te r co il in to th e flash d ru m to a id in th e v a p o r iz a tio n of th e charge. T here is in d ire c t ev idence t h a t a n ap p re c ia b le p ro p o r t io n o f th is g a s is po lym erized . T he to ta l y ie ld is 67% on th e g as o il, w h ich b r in g s th e t o t a l y ie ld f ro m th e crude to 82% . T he final re fo rm ed p ro d u c t h a s a n o c ta n e n u m b e r o f 75 a n d a b le n d in g octane value of 90. T he ra w gaso line is sw ee ten ed in a c o n tin u o u s d o c to r t r e a te r , w ashed w ith w ate r, sod ium b isu lp h ite so lu tio n , w a te r a g a in , f i lte re d th ro u g h a sa n d b ed and finally tr e a te d w ith in h ib ito r . T h e p la n t p ro d u ces so m e 820,000 cu . f t . d a ily of po lym erizab le h y d ro ca rb o n s , w hich , so fa r , a re n o t u t i l iz e d e x c e p t, in p a r t , a s fuel.

H . G.

278. Patent on Cracking. E . J . H o u d ry . U .S .P . 2 ,141 ,185 , 27.12.38. Appl.27.4.35. P rocess of re fo rm in g n a p h th a in o rd e r to in c rease th e a n ti-k n o c k q u a lity .

W . S. E . C.

Hydrogenation.279. H ydrogenation as a Step Towards Italian Oil A utarchy. A n o n . World Petrol., 1939, 10 (1), 56.— A h y d ro g e n a tio n p la n t h a s b een p u t in to o p e ra t io n a t L ivorno, I ta ly , a second is u n d e r c o n s tru c tio n a t B a ri a n d a th i rd m a y be a d d e d la te r to handle A lb an ian cru d e oil. I t is p ro p o sed e v e n tu a lly to s u b s t i tu te to a la rg e e x te n t im ported I ra q oil b y th is cru d e a n d o ils fro m I ta l ia n a s p h a lts , sh a le s a n d lig n ite ta r s . The L ivorno refinery also co n ta in s d is t i l la t io n u n i ts , a c ra c k in g u n i t a n d a so lv e n t tre a tin g p lan t. T he h y d ro g e n a tio n u n i t o p e ra te s u n d e r th e B e rg iu s p a te n t s g iv in g a y ield of 80% of gasoline fro m th e A lb an ia cru d e , w h ich is h ig h ly su lp h u ro u s a n d y ie ld s only 12% of s tra ig h t- ru n gaso line . I t w ill a lso b e u se d to p ro d u c e 60,000 to n s of high q u a lity lu b r ic a n ts a n d in a d d it io n iso -o c tan e . T h e tw o p la n ts w ill p ro d u c e 240,000 to n s of au to m o b ile a n d a v ia t io n gaso line— h a lf th e n a t io n a l c o n su m p tio n — b u t this m ay ev e n tu a lly be in crease d to 400,000 to n s .

Storage fac ilitie s a t L iv o rn o a re eq u ip p e d w ith a n itro g e n c irc u la t io n sy s tem for fire p ro tec tio n , w hile th e e lec tric po w er p la n t u til iz e s th e en e rg y p ro d u c e d th ro u g h the boracic soffioni of L ard ere llo . T h e co m p le tio n of th e V enice a n d T rie s te refineries a n d th e cu rta ilin g of d o m estic c o n su m p tio n of gaso lin e to 340,000 to n s p e r annum enables considerab le q u a n ti t ie s to be e x p o r te d . T h e p o te n t ia l p ro d u c tio n of the A lban ia wells is g iven as 120,000 to n s a n d t h a t f ro m d o m e s tic l ig n ite a s 170,000 tons, an d th a t from calcareous a s p h a lts , b itu m e n s a n d sc h is ts a s 100,000 to n s . I ta lia n crudes yield 5000 to n s of gaso line , w hile a q u a n t i ty e q u iv a le n t to 12,000 to n s is o b ta in ab le by th e use of n a tu ra l gas, a n d a q u a n t i ty e q u iv a le n t to 85,000 to n s by

ABSTBACTS. 9 5 A

the use o f so lid fuels. P ro d u c tio n of a lcoho l fo r co m p u lso ry a d m ix tu re w ith gaso line has been suspended ow ing to a n in a d e q u a te su g a r-b e e t c ro p a n d its" low su g a r content. B enzo l p ro d u c tio n h a s b een in c rease d fro m 17,000 to n s in 1937 to 22,000 to n s in 1938. I t a l ia n im p o r ts o f refin ed p ro d u c ts w ere red u ced b y 40-6% d u rin g 1938, the I ta l ia n re fineries b e in g n o w c a p a b le o f m e e tin g th e e n tire gaso line d e m a n d , 65% of the a g r ic u ltu ra l a n d illu m in a tio n o il, 7 5 % o f th e lu b r ic a tin g o il a n d 100% of th e w hite s p ir i t , t r a n s fo rm e r o il a n d a s p h a l t d e m a n d . C. L . G.

280. Mean Pressure Synthesis of Paraffins from CO and H with a Co Catalyst. F .F isch e r a n d H . P ic h le r . Brennst.-Chemie, 1.2.39, 20 (3), 41 -48 .— H y d ro g e n a tio n te s ts h a v e b e e n c a r r ie d o u t a t te m p e ra tu re s a n d p ressu re s in te rm e d ia te b e tw een th o se of th e n o rm a l p re ssu re gaso lin e sy n th e s is a n d th o se of th e h ig h -p ressu re m e th a n o l sy n th e sis . T h is “ m e a n p re ssu re sy n th e s is ” d iffers b as ica lly fro m h i th e r to kn o w n processes. S im ila r t o th e gaso lin e sy n th e s is CO a n d H a re co n su m ed a t a r a t io o f a p p ro x im a te ly 1 : 2. U s in g a Co c a ta ly s t o p tim u m c o n d itio n s a re 160—200° C., a n d 4 -20 a tm . M a x im u m y ie ld is o b ta in e d n e a r th e low er te m p e ra tu re lim it W ith th e sam e Co c a ta ly s t a s u se d in th e g aso lin e sy n th e s is o p tim u m te m p e ra tu re s of th e m ean p ressu re sy n th e s is a re low er. A t th e p re ssu re s p re v a ilin g in th e m e a n p re ssu re sy n th e s is a n d u s in g Co c a ta ly s ts , m a x im u m y ie ld o f so lid para ffin , w h ich is th e m a in p ro d u c t of th e r e a c tio n , a n d a lso m a x im u m to ta l y ie ld , is o b ta in e d . A t th e sam e tim e c a ta ly s t life w as lo n g e s t. E x c e e d in g th e p re ssu re reg io n o f th e m e a n p re ssu re sy n th e s is re su lte d in no a d v a n ta g e . O p e ra tin g in one s ta g e w ith th e sam e c a ta ly s t a n d w ith o u t c a ta ly s t r e a c t iv a t io n a so lid p a ra ffin y ie ld 6 -7 tim e s t h a t of th e g aso line sy n th e s is w as o b ta in e d , p a ra ffin s b o ilin g o v e r 450° C. w ere u p to tw en ty -fiv e tim e s m ore , w h ereas to ta l y ie ld o f l iq u id a n d so lid h y d ro c a rb o n s w as 2 5 -3 0 % h ig h er. T h e life of th e c a ta ly s t w as s ix tim e s lo n g e r th a n in th e n o rm a l p re ssu re sy n th e s is .

I t is concluded t h a t , c o n tr a ry to p re v io u s o p in io n , h ig h m o lecu la r h y d ro c a rb o n s do n o t possess a g re a te r te n d e n c y to red u ce th e a c t iv i ty of Co c a ta ly s ts . T h e m ore ra p id decline of c a ta ly s t a c t iv i ty in th e n o rm a l p ressu re sy n th e s is is ex p la in ed b y th e fo rm a tio n of m in u te q u a n t i t ie s o f c o m p o u n d s o f a d v e rse effect on th e c a ta ly s t , w hereas such co m p o u n d s a re m a d e h a rm le ss in th e m e a n p ressu re sy n th e s is b y th e m ore effective h y d ro g e n a tio n . L . R .

281. Patent on H ydrogenation. G . W . Jo h n so n . E .P . 497,607, 22.12.38. A pp l.21.7.37. M echan ica l s e p a ra tio n o f so lid su b s ta n c e s fro m o il resid u es fo rm ed b y th e d es tru c tiv e h y d ro g e n a tio n o r c ra c k in g of ca rb o n aceo u s su b s ta n c e s in th e p resence of d ilu en ts b o ilin g b e tw een 160° a n d 250° C., a n d e ffec ting m ech an ica l se p a ra tio n b y filtering o r cen tr ifu g in g . W . S. E . C.

Polymerization.282. Patents on Polym erization. M. W . P e rrin , E . W . F a w c e tt, J . G. P a to n , E . G. W illiam , a n d I .C .I . L td . E .P . 497,643, 22.12 .38. A p p l. 22.4.37. In te rp o ly m e r iz a tio n of e th y len e w ith o rg an ic co m p o u n d s , e.g., p ro p y len e , iso b u ty len e , p e n te n e o r iso- pen ten e , a t 100—400° C. u n d e r 1500 a tm . p re ss u re in th e p resen ce of free o x y g en o r benzoyl p e ro x id e , to p ro d u c e h ig h m ol. w t. co m p o u n d s.

W . B . P lu m m e r. U .S .P . R e is su e 20,950, 13.12.38. A p p l. 7.2.36 (o rig inal U .S .P . 1,991,353, 15.2.35). P o ly m e riz a tio n o f u n s a tu ra te d gases to in crease th e d e f in e co n ten t a n d s u b je c tin g to recy c lin g a n d p o ly m eriz in g tr e a tm e n t . V i. S. E . C.

Refining and Refinery Plant.283. Control of W ax D istillate Q uality. E . A . B u rch . N at. Petrol. News, 30.11.38, 30 (48), R . 582.— T h e a n a ly tic a l m e th o d s a p p lie d to th e q u a n t i ta t iv e se p a ra tio n of w ax from oil—w a x m ix tu re s in th e c o n tro l o f so lv e n t e x tra c tio n p rocesses fo r lu b r ic a tin g oil refining a re p re se n te d . W a x is d efin ed as t h a t m a te r ia l w h ich , w h en se p a ra te d from the oil—w a x m ix tu re , leav es an o il o f th e d es ired p o u r p o in t (u su a lly 0° F .) . T he classical m ethods fo r th e s e p a ra tio n of w a x a re n o t s a tis fa c to ry a n d th e choice of th e

a b s t r a c t s .

solvent m u st a lw ays depen d on th e n a tu r e of th e o il-w a x m ix tu re T h e d e s id e ra ta of th is choice are : (i) H ig h so lv en t a c tio n fo r b o th o il a n d w ax a t re la tiv e ly h ig h tern-

a tu res (ii) H ig h so lv en t a c tio n fo r oil a n d low so lv e n t a c t io n fo r w ax a t low tem n era tu res (iii) Low bo iling p o in t to f a c ili ta te reco v e ry , (iv) C hem ica l s ta b il i ty , low to x ic ity an d low price. M ethy lene ch lo ride , w ith o r w ith o u t a d d it io n s of e th y len e dichloride benzol o r ace tone , h a s been fo u n d to be m o s t su ita b le . T h e a c tu a l con ­ditions of n a tu re of so lv e n t, so lv e n t-o il-w ax m ix tu re r a t io a n d d e w a x in g te m p e ra tu re s are de te rm in ed ex p erim en ta lly , th e p rec ise ro u tin e of th e a n a ly tic a l p ro ced u re being a m a tte r for th e choice of th e o p e ra to r. M inus 14° F . is reco m m en d ed a s a s ta n d a rd dew axing te m p e ra tu re w hen 0° F . p o u r t e s t o il is d es ired . S uffic ien t o il a n d w ax are separa ted fo r th e d e te rm in a tio n of th e v isc o s ity in d e x o f th e fo rm er a n d th e m eltin g p o in t of th e la t te r . T h e scale-w ax c o n te n t of th e s e p a ra te d w a x is th e n d e te rm in ed by filtering a 2-5% w t. so lu tio n of th e w a x in a m ix tu re o f 7 5 % a c e to n e a n d 25% m ethy lene chloride a t 0° C. T h u s o p tim u m co n d itio n s m a y b e d e te rm in e d fo r m a x i­m um yields of o il of a desired p o u r p o in t a n d v isc o s ity in d e x to g e th e r w ith d a ta concerning th e y ie ld a n d m e ltin g p o in t o f th e w a x . H . G.

284. Control of W ax D istillate Quality. J . W . D o n n e l a n d E . A . B u rc h . N at. Petrol. News 14.12.38, 30 (50), R . 602.— T h e d e s id e ra ta of th e e x p e r im e n ta l c o n tro l of slack- w ax sw eating is d iscussed a n d a p i lo t p la n t fo r th is p u rp o se d esc rib ed . T w o types of slack-w ax are considered , (a) h o t slack , s e p a ra te d a t 35 -40° F . , a n d (6) co ld slack, se p ara te d a t m in u s 5-0° F . T he p o in ts co n s id e red a re : t im e cycle o f sw ea tin g in re la tion to y ield , th e recycling of in te rm e d ia te w ax es , o r ig in of c ru d e a n d n a tu re of w ax d is tilla te c u t, b len d in g of d iffe ren t s lack s, sw e a te r c ak e th ic k n e s s , effect of a tm ospheric cond itions, e tc . C o n d itio n s m u s t b e f u r th e r co n s id e red in re la tio n to (i) th e co n cen tra tio n of w ax in th e slack , (ii) th e d is t i l la t io n ra n g e o f th e w ax o r of th e o rig inal w ax y d is til la te , (iii) th e m e ltin g -p o in t ra n g e of th e w a x in th e slack. I n general slacks co n ta in in g m o re th a n 5 0 % o il c a n n o t b e sw e a te d . S lacks should be cooled a t a m in im u m r a te to in d u c e th e co arse c ry s ta l l in e s t r u c tu r e w h ich is necessary fo r efficient sw eating . O th e r fa c to rs b e in g e q u a l th e y ie ld of scale varies as th e sw eating tim e , econom ic co n s id e ra tio n s b e in g th e lim itin g fa c to r . H . G.

285. Tetraethyl-lead Susceptibilities of Gasoline. L . M . H e n d e rso n , W . B . R o ss andC. M. R idgw ay . Indm tr. Engng Chem., 1939, 31 (1), 2 7 -3 0 .— T h is in v es tig a tio n w as u n d e r ta k e n w ith th e o b jec t of co m p arin g th e te t r a e th y l- le a d su scep tib ilitie s of caustic-w ashed gasolines w ith th o se t r e a te d w ith so d iu m p lu m b ite a n d sulphur. T he ex p erim en ta l ev idence in d ic a te d th a t , w hen gaso lin es a re t r e a te d w ith sodium p lu m b ite a n d su lp h u r, th e y req u ire m ore te t r a e th y l- le a d to p ro d u c e a g iv en octane nu m b er th a n do th o se tr e a te d w ith so d iu m h y d ro x id e a lo n e , a n d s im ila r ly , m ore th a n do th o se w h ich a re th o ro u g h ly sc ru b b ed w ith c a u s tic a n d th e n su b s e q u e n tly tre a te d w ith sod ium p lu m b ite a n d su lp h u r. T h e d ifferences in a m o u n t o f te tra e th y l- le a d are q u a n tita tiv e ly re la te d to th e a m o u n t of m e rc a p ta n s u lp h u r rem o v ed b y the caustic w ashing.

The decreased req u irem en ts of te t r a e th y l- le a d acco m p lish ed b y efficient caustic scrubb ing w ould re su lt in la rge econom ic sa v in g s in th e p e tro le u m in d u s try .

H . E . T.

286. Sample Computations of Natural Gasoline Absorber D ata . R . H . T u rn e r . Petrol. Engr, D ecem ber, 1 9 3 8 ,10 (3), 54.— T h e p rin c ip le s in v o lv e d in th e c a lc u la tio n of abso rber tow er capac ity , analyses of p ro d u c ts a n d o th e r fa c to rs a re d isc u sse d m a th e m a tic a lly an d ty p ica l ca lcu la tions m ade. B u b b le to w e r c a p a c ity d e p e n d s on se v era l facto rs including (i) d is tan c e b etw een t r a y s a n d p la te s , (ii) m o l. w t. o f th e gas or vapo u rs,(iii) th e abso lu te o p e ra tin g p ressu re , (iv) th e q u a n t i ty of liq u id p a ss in g dow nw ards an d (v) th e d e p th of liq u id on th e p la te s . E x p e rien ce h a s sh o w n t h a t 18 in . is th e m ost econom ical d istan ce betw een p la te s fo r gaso line p la n ts . I n b u b b le to w ers the allow- 3 fl6 ■ ! ," V e lo c ity ' s ^ Per se c- a t a tm o sp h e ric p re ssu re a n d 18 in . t r a y spac in g for a u id of m ol. w t. 125, th is v e lo c ity v a ry in g as th e r a t io o f th e sq u a re ro o ts of the mo ecu a r w eight a n d also a s th e r a t io of th e sq u a re ro o t o f th e a b s o lu te o p e ra tin g pressure.

T ypical ca lcu la tions a re g iven of th e a llo w ab le c ap v e lo c ity a n d th e c a p a c ity of an

ABSTRACTS. 97 A

ab so rb e r u n d e r specified c o n d itio n s . A ta b le is g iv en show ing th e c a p a c ity of b u b b le tow ers p e r 1000 cu . f t . o f g as p e r c a p p e r 24 h r . f ro m w h ich th e o p tim u m size of to w er can be c a lc u la te d . O th e r c a lc u la tio n s a re w o rk ed o u t fo r th e o il r a te fo r a n y g iv en se t of co n d itio n s , th e f a t o il s a tu ra t io n , th e q u a n t i ty of th e o il p re se n t, f rac tio n s condensing a n d w e a th e r in g . L . G.

287. H eat Characteristics of 5 Silicon-C hrom ium -M olybdenum Steels. A . E . W h ite ,C. L . C la rk a n d W . G . H ild o rf . Oil Oas J . P a r t I , 1.12.38, 37 (29), 43. P a r t 2,8.12.38, 37 (30), 43.— I n th e d es ig n of h ig h te m p e ra tu re e q u ip m e n t, p e rm iss ib le stresses are a t p re se n t u su a lly b a se d u p o n th e creep c h a ra c te r is tic s of th e p ro p o sed ste e ls a t th e g iv en o p e ra tin g te m p e ra tu re s . T h e u su a l creep te s ts , how ever, g ive l i t t le in ­fo rm a tio n w ith re sp e c t t o in flu en ce of la c k o f su rface o r s t r u c tu r a l s ta b il i ty on th e lo ad -ca rry in g a b i l i ty o r o n th e d u c t i l i ty c h a ra c te r is tic s . S tre ss -ru p tu re te s ts a re be lieved to g ive th e a d d it io n a l in fo rm a tio n a n d to p ro v id e a m o re sim p le m ean s for d e te rm in in g d es ig n s tre sse s fo r c e r ta in ty p e s of h ig h te m p e ra tu re e q u ip m e n t. I n ­creased re s is ta n c e to o x id a tio n a n d su lp h id e co rro s io n ca n be o b ta in e d th ro u g h p ro p e r co m b in a tio n of s ilica a n d ch ro m iu m .

F iv e s te e ls , k n o w n a s S icrom o 1, 2, 2 \, 3 a n d 5, w ere e x am in ed . T h ey c o n ta in e ssen tia lly th e sam e a m o u n ts o f M n, S, P a n d M o, a n d w ith one e x c e p tio n th e Si c o n te n t is 1-32 to 1-57% . T h e C r c o n te n ts , h ow ever, ran g e fro m 1-30 to 4-83% a n d th e C c o n te n t fro m 0 09 to 0 -15% . E a c h stee l, p r io r to m ach in in g , w as an n e a le d a t 1550° F . T h e fo llow ing p ro p e r t ie s w ere d e te rm in e d on th ese a n d c e r ta in o th e r steels, a n d re su lts a re re c o rd e d : C hem ica l co m p o s itio n , g ra in size, B rinell h a rd n ess , ten s ile p ro p e rtie s a n d C h a rp y im p a c t re s is ta n c e s a t te m p e ra tu re s ra n g in g fro m 80° F . to 1500° F ., c reep c h a ra c te r is t ic s a t te m p e ra tu re s of 800° F . to 1300° F . a n d stress- ru p tu re c h a ra c te r is tic s a t te m p e ra tu re s of 900° F . to 1500° F . O x id a tio n resis tan ce c h a ra c te ris tic s , th e effec t of te m p e ra tu re a n d t im e o n im p a c t re sis tan ce , ten sile p ro ­perties , im p a c t re s is ta n c e s a n d m e ta llo g ra p h ic s tru c tu re s of com pleted creep specim ens, a n d m e ta llo g ra p h ic s tru c tu re s o f s tre s s - ru p tu re specim ens w ere a lso d e te rm in ed , a n d v arious ta b le s a n d i l lu s t r a t io n s su m m arize th e re su lts o b ta in ed .

T he m a in conc lu s io n s re a c h e d a re t h a t a c c e p ta b le stee ls fo r h ig h te m p e ra tu re service u p to 1300° F . c an b e o b ta in e d th ro u g h th e a d d it io n o f Si in a m o u n ts u p to 1-50% to stee ls c o n ta in in g v a ry in g a m o u n ts o f C r a n d 0-50% of Mo. T h e re su ltin g steels possess a v e ry su ita b le c o m b in a tio n of s tr e n g th , d u c ti l i ty a n d im p a c t re sis tan ce a t room te m p e ra tu re . O n th e b as is of 1000-hour c reep te s ts u n d e r s lig h tly ox id iz ing cond itions, th e a d d i t io n o f S i in th is a m o u n t low ers creep s t r e n g th a t c e r ta in of th e e lev a ted te m p e ra tu re s , b u t o n a c c o u n t of th e in c rease d su rface a n d s tru c tu ra l s tr e n g th im p a r te d i t d oes in c rease lo a d -c a rry in g a b i l i ty a t th e se te m p e ra tu re s as in d ic a te d b y th e s tre s s - ru p tu re r e su lts . In c re a se d Si c o n te n t th u s increases a c tu a l serv ice life, th o u g h i t d ecreases c reep s t r e n g th . T h e S icrom o stee ls possess a la rg e degree of h ig h te m p e ra tu re s ta b il i ty , r e ta in in g th e ir o rig in a l p ro p e rtie s u n d e r th e com bined influences of tim e a n d s tre s s a t e a c h te m p e ra tu re consid ered . H o t d u c t i l i ty c h a ra c te ris tic s a re also good u n d e r th e in flu en ce of t im e , te m p e ra tu re a n d stre ss , th u s a ssu rin g t h a t b r i tt le ty p e f ra c tu re s w ill n o t be o b ta in e d in se rv ice . B - A . E .

288. U ruguayan G overnm ent Refinery U ses Crudes from South A m erica. R . D eam - brosis. Oil Gas J ., 29 .12 .38 , 37 (33), 134.— T h e A N C A P refinery a t L a T e ja n e a r M o n te ­v ideo, U ru g u a y , b e g a n o p e ra tio n s in J a n u a ry 1937 a n d now su p p lies p ra c tic a lly a ll th e c o u n try ’s fuel re q u ire m e n ts . T h e re fin ery co n s is ts of a to p p in g p la n t of c a p a c ity 800 cu. m e tre s p e r d a y , a c ra c k in g u n i t of c a p a c ity 300 cu. m e tre s p e r d a y , refinery e q u ip m en t fo r c a u s tic so d a , d o c to r a n d a c id tr e a tm e n t a n d th e u su a l a u x ilia ry p la n t . E cu ad o rian a n d P e ru v ia n c ru d e is d is ti l le d , th e to ta l y e a r ly th ro u g h p u t of 265,000 cu. m etres y ie ld in g 80,000 cu . m e tre s gaso line , 58,000 cu . m e tre s kerosine , 23,500 cu . m etres gaso il a n d 100,000 cu . m e tre s o f to p p e d cru d e . T h e c rack in g u n i t p ro d u ces from th is s to c k 50,000 cu . m e tre s gaso lin e , 39,000 cu . m e tre s of fuel oil a n d 10,000,000 cu. m etre s of g as (b u rn t in th e re fin ery ). P ro p e r t ie s o f th e tw o c ru d es, a n d of th e p ro d u c ts fro m th e to p p in g o f th e m ix e d c ru d es a n d f ro m th e c rack in g of th e m ix e d redu ced c ru d es a re g iv e n . L . L . G .

289. Methods of Testing the A dequacy of E lectrical Grounds. C. A. A n d erso n . Petrol. Engr, D ece m b e r, 1 9 3 8 ,10 (3), 51.— A d e sc rip tio n is g iv en o f a n A .C. te s t in g a p p a ra tu s

ABSTRACTS.

for th e adequacy of serv ice g ro u n d s u sed fo r l ig h tn in g a n d h ig h p o te n t ia l h a z a rd s or for th e re tu rn p a th for e lec trica l e q u ip m e n t c irc u its , e tc . T h e a p p a ra tu s co n sists of an induction bridge c ircu it in w h ich re s is ta n c e m one a rm c a n b e v a r ie d to e q u a l th e non- inductive resis tance of th e o th e r a rm . T h e n e u tr a l p o in t is o b ta in e d a u d ib ly usrng an electrical w ave from a m a g n e to te s t se t. E x a m p le s a re g iv en of a p p lic a tio n s of th is ap p a ra tu s , show ing i ts v a lu e in lo c a tin g ineffic ien t g ro u n d in g , th is m e th o d being preferred to D .C. te s tin g .

290. Patent on Refining. R- S. D a n fo rth . U .S .P . 2 ,140,450, 13.12.38. A ppl.26.4.34. Im p ro v ed m e th o d of d is t i l la t io n of h y d ro c a rb o n o ils.

O. W olf a n d R . C. Pow ell. U .S .P . 2 ,141,633, 27.12.38. A p p l. 6 .7.35. N ew typeof furnace a n d h e a te r for h e a tin g h y d ro c a rb o n o ils to te m p e ra tu re s ab o v e 800° F .

W f i P . f !

Chemistry and Physics of Petroleum.291. Limits of Flammability of Mixtures of Propane, Air, and Nitrogen Oxide. E. B.H odge. Industr. Engng Chem., 1938, 30 (12), 1390.— T h e u p p e r a n d low er lim its for th is m ix tu re w ere d e te rm in e d fo r u p w a rd p ro p a g a tio n o v e r th e com p le te range. The resu lts are g iven in a ta b le a n d p lo t te d o n tr ia n g u la r c o -o rd in a te s . P . D .

292. Performance of Commercial Perforated-Plate D istillation Column. R . C. G unness an d J . G. B aker. Industr. Engng Chem., 1938, 30 (12), 1394.— P e rfo ra te d p la te colum ns are com m only u sed in in d u s try w hen d is til l in g s to c k w h ich d ep o s its solid m a te ria l t h a t w ou ld clog a n d re n d e r b u b b le -cap co lu m n s in effic ien t. T w o te s ts were m ade o n a s ix te e n -p la te co lu m n u se d fo r th e d is t i l la t io n o f a lco h o l from m ash. T he co lum n w as 5 J f t. in d ia m e te r , w ith a p la te sp a c in g of 18 in . T h e size of the perfo ra tions w as J in . a n d th e re w ere a b o u t 2500 p e r p la te . T h e tw o te s ts gave reasonab ly co n co rd an t re su lts , a n d th e in d iv id u a l p la te efficiencies (based on the M urphree defin itio n of efficiency) ra n g e d fro m 2 0 % a t th e b o t to m to 5 0 % a t th e top, w ith a n average of 40% . T h e f irs t t e s t w as m a d e 34 d a y s a f te r c lean in g a n d th e second a f te r 72 days. T here w as no in d ic a tio n of a d ro p in efficiency in th e second test, show ing th a t fou ling w as n o t su fficient to im p a ir o p e ra tio n . P . D.

293. Heat Transfer to Boiling Liquids. F . H . R h o d e s a n d C. H . B rid g es . Industr. Engng Chem., 1938, 30 (12), 1401.— A s th e h e a t t r a n s f e r r a te f ro m a tu b e w all to a bo iling liqu id is p rogressively in creased a ch an g e in th e c h a ra c te r of b o ilin g is ob tained. A t low ra te s (sm all te m p e ra tu re d ifference b e tw e e n tu b e a n d liq u id ) v ap o riza tio n is nuclear, a n d th e h e a t tra n s fe r coeffic ient is h ig h . A t h ig h r a te s (large tem p era tu re difference) th e v a p o u r is p ro d u ced in a film , sp re a d o v e r th e w a ll su rface , w h ich gives a considerab le th e rm a l re sis tan ce a n d co n se q u e n tly low coeffic ien t. T h e change tak es p lace a t a c r itic a l te m p e ra tu re d ifference b e tw e e n w a ll a n d liq u id . T h is te m ­p e ra tu re difference w as d e te rm in e d fo r w a te r b o ilin g in a s te e l tu b e 0 1 9 2 in . in ternal d iam ete r a t p ressu res fro m a tm o sp h e ric d o w n to 2 in . o f m e rc u ry . W ith a c lean tube a critica l te m p e ra tu re d ifference of 50 to 60° C. w as o b ta in e d a t p re ssu res of 5 an d 10 in . of m ercu ry , b u t th e e v a p o ra tio n re m a in e d n u c le a r a t p re ssu re s of 1 a tm ., 20 in. a n d 2 in . u p to th is te m p e ra tu re d ifference. T h e p resen c e o f oleic a c id o r m ineral oil as a film on th e s tee l co n s id e rab ly red u ces th e c r i t ic a l te m p e ra tu re difference, w hile th e p resence of so d iu m c a rb o n a te o r ch lo rid e in a n o ily tu b e rem o v es th e oil lay er an d favours n u c lear boiling .

E x p erim en ts w ere also m ad e in a ch ro m iu m -p la te d tu b e , a n d th e c r it ic a l tem p era tu re difference w as found to be low er th a n w ith c lean ste e l. P . D.

294. Exhaustive Fractionation of the “ Extract ” Portion of the Lubricant Fraction from a Mid-Continent Petroleum . B . J . M a ir a n d C. B . W illin g h a m . Bur. Stand. J . Res., Wash,, 1938, 21 (5), 535-563 .— L u b ric a tin g s to c k f ro m a m id -con tinen t crude, d is tilled in a com m ercial v a c u u m s t i l l , w as g iv en th e fo llow ing t r e a tm e n t :

ABSTRACTS. 99 A

(i) s e p a ra tio n of a n e x t r a c t p o r t io n b y e x tr a c t io n w ith l iq u id S 0 2 a t ro o m te m ­p e ra tu re , (ii) s e p a ra tio n o f a w a x p o r t io n b y c ry s ta lliz a tio n fro m e th y len e ch loride a t — 18° C., (iii) s e p a ra tio n o f re m a in d e r b y f i l t r a t io n th ro u g h s ilica gel in to (a) a w a te r w h ite f i lt ra te , (6 ) p o r t io n reco v e red fro m s ilica gel a n d ca lled s ilica gel ho ld -u p .

I n th is p a p e r th e in v e s tig a tio n of f ra c tio n s 1 a n d 3(6) is d escrib ed . F ra c tio n 1 w as e x tr a c te d w ith l iq u id S 0 2 a n d p e tro le u m e th e r a t — 55° C., th e p e tro le u m e th e r so luble p o r t io n c o m b in ed w ith f ra c tio n 3(6) a n d th e w hole s u b m itte d to a sy s te m a tic f rac tio n a l d is t i l la t io n in h ig h v a c u o u n t i l s u b s ta n tia l ly c o n s ta n t b o ilin g f ra c tio n s were o b ta in e d .

F iv e ch arg es o f th e se f ra c tio n s of fro m 500-700 gm . e ach w ere s e p a ra te d in to 30 or 40 f ra c tio n s b y e x tr a c t io n w i th m e th y l cy a n id e o r m e th y l cy an id e a n d ace to n e . K in e m a tic v isc o sitie s a t 100° F . a n d 210° F ., re fra c tiv e in d ices a t 25° C., d en s itie s , re frac tiv e d isp e rs io n s , specific o p tic a l r o ta t io n s a n d an ilin e p o in ts w ere d e te rm in ed for m o st o f th e f ra c tio n s f ro m th e e x tr a c t io n p rocess . I n a d d i t io n C, H ra tio s , m o lecu lar w e ig h ts a n d b o ilin g p o in ts w ere d e te rm in e d on 41 k ey fra c tio n s a n d fo r som e of th e m also % S a n d % N 2.

A n u m b e r of in te re s t in g fa c ts in c o n n e c tio n w ith th e p ro p e rtie s of th ese p e tro le u m frac tio n s em erge f ro m th is w ork . D . L . S.

295. H ydrogenation of the “ Extract ” Portion of the Lubricant Fraction from a M id- Continent Petroleum . B . J . M air, C. B . W illin g h a m a n d A . J . S tre iff. Bur. Stand. J . Res., Wash., 1938, 21 (5), 565—580.— T o in v e s tig a te th e c o m p o s itio n of th e e x tra c t p o r tio n of th e lu b r ic a n t f r a c tio n fro m a m id -c o n tin e n t p e tro leu m , 15 p o r tio n s p r e ­p ared b y e x te n s iv e d is t i l la t io n fo llow ed b y e x h a u s tiv e e x tra c tio n , w ere com plete ly h y d ro g en a ted .

H y d ro g e n a tio n w as c a r r ie d o u t in a b o m b u sin g a n ick e l c a ta ly s t , a n d w as c o n ­tin u e d u n ti l th e specific d isp e rs io n d ecreased to a b o u t 100, a v a lu e c h a ra c te r is tic of n ap h th en es . T ab le s a re in c lu d e d c o m p a rin g th e p h y sic a l c o n s ta n ts of th e f ra c tio n s before a n d a f te r h y d ro g e n a tio n . E v id e n c e is g iv e n w h ich in d ic a te s t h a t u n d e r th e c o n d itio n s of th e se e x p e r im e n ts— te m p . 230-250° C. a n d p ressu res of H 2 170-210 a tm o sp h e res , th e f ra c tio n s w ere c o m p le te ly h y d ro g e n a te d a n d th a t no b reak d o w n of th e m olecules o ccu rred . D . L . S.

296. The Ethane Equilibrium . R . N . P ease a n d A . M. B y e rs , J r . J . Amer. chem. Soc., 1938, 60, 2489-2491 .— I t h a s b een su g g e sted t h a t th e ex p e rim e n ta l d a ta on th e p o sitio n of e q u ilib riu m in th e re a c tio n C2H 4 + H 2 ^ C2H 6 a re in e rro r because of th e occurrence of s id e -re a c tio n s in g en e ra l, a n d p a r t ic u la r ly , because th e alleged presence of p ropy len e h a s in te r fe re d w ith th e a n a ly tic a l d e te rm in a tio n of e th y len e . A d d itio n a l ex p erim en ts con firm th e e a r lie r conc lu s io n s of T ra v e rs a n d H ock in .

A co m p ariso n of a ll e x p e r im e n ta l v a lu e s w ith v a lu es b ased o n th e N e m s t H e a t T heorem (and th e T h ird L aw ) em p h as izes th e in te rn a l co n s is ten cy of th e d a ta .

I t is co n c lu d ed t h a t th e re is n o goo d g ro u n d fo r ig n o rin g th e e x p e r im e n ta l d a ta on th e e th a n e e q u ilib riu m in d e a lin g w ith th e q u e s tio n of free ro ta t io n in th e e th a n e m olecule. T . C. G. T .

297. Fluorinated D erivatives of Propane. H . A. L . H en n e a n d E . C. L ad d . J. Amer. chem. Soc., 1938, 60, 2 4 9 1-2495 .— T h ree c h lo r in a te d p ro p an es C3C18, CHC12CC12CC13, CC13CHC1CC13 h a v e b een sy n th e s ized , a n d th e n f lu o rin a te d to g ive e igh t fluorine d e r iv a tiv e s . T h e s tru c tu r e s a n d p ro p e rtie s of th o se d e r iv a tiv e s , to g e th e r w ith th ree o th e rs , a re p re se n te d . T . C. G. T .

298. Reaction of ¿soButene and d i-isoB utene w ith Phenol, w ith and w ithout Scission of C C Linkages. V . N . Ip a tie f f , H . P in e s a n d B . S. F rie d m a n . J . Amer. chem. Soc., 1938, 60, 2 495-2497 .— P h e n o l is a lk y la te d w ith d i- iso b u ten e , a t te m p e ra tu re s up to 150°, in th e p re sen c e of p h o sp h o ric a c id o r w ith a la rg e excess of su lp h u ric ac id , to y ield 4 -f-b u ty lp h en o l a n d 2 : 4 -d i-f-b u ty lp h en o l.

P henol is a lk y la te d b y iso b u te n e in th e p resen ce of p h o sp h o ric ac id a t 100° to give good y ie ld s o f p - f -b u ty lp h e n o l a n d 2 : 4 -d i-f-b u ty lp b en o l.

W hen p - (a a ,y y - te tr a m e th o b u ty l)p h e n o l w as h e a te d u n d e r p re ssu re w ith pho sp h o ric acid , th e s id e -ch a in w as c leav ed to p ro d u c e p h en o l, o c tenes , p - i-b u ty lp h e n o l, an d 2 : 4 -d i-i-b u ty Ip h en o l. T . C. G . T .

100 A ABSTRACTS.

299. Addition of Sulphur, Hydrogen Sulphide and M ercaptans to Unsaturated Hydro­carbons. S. O. Jo n e s an d E . E m m e tt R e id . J . Amer. chem. Soc., 1938, 60, 2452—2 4 5 5 S u lphur s tr ip s H 2 from u n s a tu ra te d s to fo rm H 2S a p a r t of w h ich a d d s to thedouble bond to p roduce a m e rc a p ta n , w h ich a d d s to m o re o f th e u n s a tu ra te to g ive a sulphide. T he a d d itio n of H 2S ta k e s p lace re a d ily a n d is c a ta ly z e d b y S. The ad d itio n of m ercap tan s , as also t h a t of H 2S, to d o u b le b o n d s , follow s M arkow nikoff’s rule in th e absence of perox ides. A s th e re a re u su a lly su ffic ien t p ero x id es , fo r c a ta ly tic purposes, in th e h y d ro ca rb o n , th e a b n o rm a l a d d i t io n is d ifficu lt to supp ress.

T . C. G. T.

300. Physical Constants of c is Pentene-2. M. L . S h e rrill a n d E . H . L aunspach .J. Amer. chem. Soc., 1938, 60, 2562-2563 .— T h e c fs-pen tene-2 w as p re p a re d b y the sem i-reduction of p e n te n e -2 a n d w as fo u n d to possess th e fo llow ing c o n s ta n ts , b .-p t., 3 7 .O» -j- 0-05°, n™D 1-3822, d |° 0-6562. T . C. G. T .

301. Properties of Purified Normal H eptane and isoOctane ( 2 : 2 : 4 -Trimethyl Pentane).D. B. B rooks. Bur. Stand. J . Res., Wash., 1938, 21 (6 ), 847 -852 .— W o rk is in hand to develop specifications for n -h ep tan e a n d iso o c ta n e w h ic h a re u se d as p rim ary s ta n d a rd reference fuels for th e k n o ck r a t in g o f a u to m o tiv e en g in e fuels. I n th is connection i t w as necessary to o b ta in d a ta o n th e p h y s ic a l p ro p e r tie s of sam ples of th e h ighest possib le p u r ity .

To th is en d th e p u re s t a v a ilab le m a te r ia ls w ere fu r th e r p u rified a n d th e boiling- p o in ts , freezing-po in ts, re frac tiv e ind ices, d e n s itie s a n d c e r ta in o f th e ir differential coefficients accu ra te ly m easu red o n th e r e su ltin g m a te r ia ls . I n e ach case th e freezing p o in t of th e b es t m a te r ia l o b ta in e d w as h ig h e r th a n c u r re n tly a c c e p te d values.

D . L . S.

302. Physical and Chemical Constants of Norm al Paraffins. D . J . W . K reu len . J.Instn Petrol. Tech., 1938, 24, 554-561 .— T h e d e n s ity , re fra c tiv e in d e x , an ilin e poin t, v iscosity a n d su rface ten s io n of a series of f ra c tio n s c o n s is tin g e n tire ly of norm al paraffins h av e been d e te rm in e d a n d th e ir r e la tio n sh ip s d e m o n s tra te d b y a num ber of g raphs. D . L . S.

303. Principles of Solvent D ew axing. Part IV . The Precipitation of W ax from Solution in Oil by Oil-miscible Liquids. M. B a T h i, T . G. H u n te r a n d A . W . Nash. J . Instn Petrol. Tech., 1938, 24, 562-576 .— T h e fa c to rs in v o lv e d in p re c ip ita t in g wax from so lu tion in oil a re d iscussed . I t is p o in te d o u t t h a t th e s u i ta b i l i ty of a liquid for dew axing o p era tio n s is d e p e n d e n t p r im a rily o n th e so lu b il i ty of w a x in th e oil, in con junction w ith th e so lu b ility of w ax in su ch a d e w a x in g liq u id . D . L . S.

304. The Nonanes. F . C. W h itm o re a n d H . A . S o u th g a te . J . Amer. chem. Soc., 1938, 60, 2571-2572.— T h e p a p e r describ es th e sy n th e s is o f th re e n ew Nonanes, nam ely , 3 -e th y lh ep tan e , 2 : 3 -d im e th y lh e p ta n e a n d 2 : 2 : 4 : 4 -te tra m e th y lp e n ta n e . As th e d a ta on 2 -m eth y lo c tan e w ere m eagre , th is c o m p o u n d w as a lso p re p a re d .

T he 2 : 2 : 4 : 4 -te tra m e th y lp e n ta n e w as o b ta in e d fro m 2 : 2 : 4 -trim ethy l-4 -b rom o- pen tan e , u sing d im e th y l zinc. T h e th re e o th e r n o n a n e s w ere o b ta in e d b y dehyd ra tio n of th e te r t ia ry alcohols. T h e p h y sic a l p ro p e rtie s of th e h y d ro c a rb o n s o b ta in ed are show n below :

N onane. F .-p „°C.

B .-p . (760 m .)

°C.n™°D. 2)20 1.

V iscosity . 37-8 100 °C. °C.

P o ises X 10.

2-M ethy loc tane2 : 3 -D im ethy lhep tane

3 -E th y lh ep tan e

2 : 2 : 4 : 4 -T e tra m e th y lp e n ta n e .

- 80-1 ( -1 1 6 -7 ) glass

( -1 1 4 -9 ) glass - 66-9

to - 6 7 - 1

142-80 140-65 in itia l143-10

122-30

1-402851-40850

1-40900

1-40695

0-71070-7235

0-7260

0-7185

5-25 2-795-21 3-30

4-90 2-61

6-80 3-39

W . E . J . B.

ABSTRACTS. 101 A

305. New Synthesis of Tertiary H ydrocarbons. F . C. W h itm o re a n d H . P . O rem . J . A m et. chem. Soc., 1938, 60, 2573~2o74.— F iv e t e r t ia r y a l ip h a t ic h y d ro c a rb o n s h a v e been p re p a re d f ro m th e co rre sp o n d in g t e r t ia r y a lcoho ls b y a m od ified L evene m e th o d . T he a lcoho ls a re t r e a te d w ith d ry h y d ro g e n io d id e u n t i l \ \ t im e s th e th e o re tic a l q u a n ti ty h a s b e e n a d d e d . Z inc d u s t is th e n a d d e d in sm a ll q u a n ti t ie s u n t i l re a c tio n ceases. D ry h y d ro g e n ch lo rid e is th e n p a sse d in w ith th e a d d it io n of m ore zinc u n ti l 4 tim e s th e th e o re tic a l q u a n t i ty is a d d e d . T h e re a c tio n te m p e ra tu re is ra ise d to 70-80° C. w ith s t ir r in g , a n d th e a d d i t io n of h y d ro g e n ch lo rid e c o n tin u ed .

T he h y d ro c a rb o n is o b ta in e d b y w a te r d i lu t io n fo llow ed b y s te a m d is ti l la t io n . A fter p u r ific a tio n , th e p h y s ic a l c o n s ta n ts d e te rm in e d a re show n below :

H y d ro c a rb o n s . B .-p .,°C. M n. n 20. < ■ F .-p .,

°C.Y ie ld .

0//o-

2 -M ethy lhexane 90-3 760 0-6794 1-3851 - 1 2 0 -3 23-72-M ethy loctane 142-8 760 0-7132 1-4030 - 1 2 0 -3 48-83 -E th y lh e p ta n e 1 4 3 1 760 0-7272 1-4090 - 1 2 0 -3 38-33-M ethy lnonane 167-6 760 0-7319 1-4123 - 9 0 - 0 30-14-M ethy ldecane 188-1 760 0-7422 1-4177 - 9 2 - 9 27-8

W . E . J . B .

306. H exam ethylethane and T etra-alkylm ethanes. R . E . M a rk e r a n d T . S. O akw ood . J. Amer. chem. Soc., 1938, 60 , 2598.— A p ra c tic a l m e th o d is g iv en fo r th e p re p a ra tio n of h e x a m e th y le th a n e a n d o th e r h y d ro c a rb o n s co n ta in in g a t e r t ia ry g ro u p b y co n ­d e n sa tio n of a n a lk y l G rig n a rd re a g e n t w i th a t e r t ia ry h a lid e b y m ean s of cu p ro u s iodide.

T he h y d ro c a rb o n s n o te d below , w ere o b ta in e d :

B .-p .,(760 m m .)

°C. v,20 d20.4

2 : 2 -D im e th y lb u ta n e 49-6 1-3709 0-64912 : 2 -D im e th y lp e n ta n e . 79-0 1-3825 0-67392 : 2 -D im e th y lh e x an e 105-2 1-3942 0-69342 : 2 -D im e th y lh e p ta n e 130-4 1-4035 0-71053 : 3 -D im e th y lp e n ta n e 86-0 1-3911 0-69373 : 3 -D im e th y lh e x an e 112-0 1-4008 0-71073 : 3 -D im e th y lh e p ta n e 137-2 1-4087 0-72543 : 3 -D im e th y lo e ta n e 161-2 1-4165 0-7390

W . E . J . B .

307. Vinyl H alide Polysulphones. Peracetic Acid as a Catalyst for the R eaction between Sulphur D ioxide and Olefins. C. S. M a rv e l a n d F . J . G lav is . J . Amer. chem. Soc., 1938, 60, 2622-2626 .— T h e c a ta ly t ic a c t iv i ty w h ich c e r ta in sam ples of a g e d p a ra ld eh y d e h a v e show n in p ro m o tin g th e r e a c tio n b e tw e e n olefins a n d s u lp h u r d iox ide c an b e o b ta in e d b y u se of p e ra c e tic ac id so lu tio n s .

V iny l ch lo ride a n d v in y l b ro m id e co m b in e w ith su lp h u r d io x id e in th e p resence of ac tiv e p a ra ld e h y d e o r p e ra c e tic ac id to g ive a p o ly m e r of th e c o m p o s itio n ((C H -p C H X ^S O j),,, r a th e r th a n th e u su a l one to on e ty p e o f p o ly m er. A p re lim in a ry s tu d y of th e re a c tio n o f th e v in y l h a lid e p o ly m e r h a s n o t in d ic a te d i ts s t r u c tu r a l u n i t ,

No ev idence fo r co m p o u n d fo rm a tio n w as o b ta in e d fro m th e e x a m in a tio n of free z­ing -po in t c o m p o s itio n cu rv es fo r so lu tio n s of s ty re n e , 1 -p en ten e a n d 10 -hendecenoic acid in liq u id su lp h u r d io x id e . W . E . J . B .

308. Instability of Liquid isoB utene. E . E . R o p e r. J . Amer. chem. Soc., 1938, 60, 2699-2701.— I t h a s b e e n show n t h a t iso b u ten e (2-m e th y lp ro p en e ) in th e liq u id p h ase fo rm s a h ig h -b o ilin g liq u id u p o n s ta n d in g , w h ich p ro b a b ly is a p o ly m er, w ith c e r ta in evidence p o in tin g to d im e riz a tio n a s th e f irs t s te p .

102 aa b s t r a c t s .

At 0° C th is reac tio n causes a low ering of th e p re ssu re o f th e sa tu ra ted vapour of pure w obutene to th e e x te n t of a b o u t 0 -6% . T h e o th e r three butenes do n o t exhib it such behav iour. W - E ’ J ‘ B '

qnq Reaction of AUphatic Olefins w ith Thiophenol. V . N . Ip atie ff, H . P ines andB S F riedm an . J. Amer. chem. Soc., 1938, 60, 2 7 3 1 -2 /3 4 . T h e ad d ition of th io ­phenol to a lip h a tic olefins p ro ceeds c o n tra ry to M arkow nikofif’s rule. Propene, b u ten e -1 u o b u ten e , p e n te n e -1 , fsop ropy l, e th y le n e , a n d tn m eth y le th v len e react w ith th iopheno l in th e absence of ac id c a ta ly s ts to y ie ld n -p ro p y l, n -b u ty l, t-butyl, n-am vl i-am vl a n d s-isoam yl p h e n y l su lp h id es , re sp e c tiv e ly . T h e presence of 90% phosphoric ac id scarcely affec ted th e cou rse of th e reaction , since th e olefins react m ore read ily w ith th io p h en o l th a n w ith th e ac id .

In th e presence of su lp h u ric a c id (d ilu te d w ith w a te r o r acetic acid) th e reaction proceeds in accordance w ith M arkow nikoff’s ru le : ¡so b u ten e and trim ethylethylene y ield i-b u ty l a n d i-am yl p h en y l su lp h id es , resp ec tiv e ly .

Iso m eriza tion occurs d u rin g th e re a c tio n of ia o p ro p y le th y le n e w ith thiophenol in th e presence of su lphuric ac id . T h e p ro d u c t is i-a m y l p h e n y l su lph ide and not the expected s-isoam yl p h en y l su lph ide .

The p hysical p ro p e rtie s of a n u m b e r o f a lk y l p h e n y l su lp h id e s a re d esc rib ed . Several of these were charac te rized b y p re p a ra tio n of so lid d e r iv a tiv e s . W . E . J . B.

310. Heats of Organic Reactions. VII. Addition of H alogens to Olefins. J . B . Conn,G. B. K istiak o w sk y a n d E . A . S m ith . J . Amer. chem. Soc., 1938, 60, 2764-2771.— The au th o rs h a v e d e te rm in ed th e h e a ts of re a c tio n in v o lv e d in certa in brom inations a n d ch lo rin a tio n s of v ario u s u n s a tu ra te d h y d ro c a rb o n s . T h e paper should be con­su lted fo r th e a c tu a l d a ta o b ta in e d , b u t one co n c lu sio n to b e draw n from the work is th a t s u b s titu tio n o n th e e th y len ic g ro u p re su lts in a n in c re a se d b eat evo lu tion on ad d itio n of b rom ine , w hereas p rev io u s w o rk h ad in d ic a te d t h a t as f a r as hydrogenation w as concerned, th e h e a t of re a c tio n w as d im in ish e d b y su c h su b s t i tu t io n .

The observed tr e n d s afford a th e rm o -d y n a m ic b as is fo r th e so -ca lled M arkownikoff ru le, an d fo r th e te n d e n c y of b ro m in e to re p lace h y d ro g e n o n th e m o st substituted carbon a to m of a s a tu ra te d h y d ro ca rb o n .

A m e th o d of ca lcu la tio n is g iven w h ich en ab les th e p re d ic t io n of th e h ea ts o f addition of hyd rogen h alides a n d of b ro m in e s u b s t i tu t io n , to be m ade. W . E . J . B.

311. Interaction between Methylene Radicals and H ydrogen. C. R osenblum . J.Amer. chem. Soc., 1938, 60, 2819-2820.— T h e p h o to ly s is o f k e te n e in th e presence of hydrogen in d ica tes a reac tio n be tw een m e th y le n e ra d ic a ls a n d h y d ro g e n . A t 200° C.— e.g., reac tio n p ro d u c ts consisted of m e th a n e , a gaseous f ra c tio n w ith a m ean com position C2.68H 7.36 a n d a residue w ith a n av erag e co m p o s itio n C4H 10.

A suggested reac tio n is C H 2 + H 2 = C H 3 + H .A secondary source of m e th y l ra d ic a ls m ig h t be a reaction b etw een m ethylene

rad ica ls a n d m e th an e a lread y fo rm ed .T he source of m e th an e is p ro b a b ly a n in te ra c tio n of m e th y l radicals w ith hydrogen

o r h yd rocarbons, w h ils t th e re c o m b in a tio n o f m e th y l ra d ic a ls to form ethane would account for th e m ean co m p o sitio n of th e gaseous p h ase . W . E . J . B.

312. Hydrogen Fluoride as a Condensing A gent. H . The A lkylation of B enzene by Olefins. J . H . S im ons a n d S. A rch er. J . Amer. chem. Soc., 1938, 60, 2952—2953.— H ydro g en fluoride h a s b een fo u n d to b e a v e ry u se fu l a g e n t for th e alkylat ion of benzene b y olefins. P ro p y len e , ¡'.sobutene, p en ten e -2 , tr im e th y le th y le n e , and cyclo- hexene were used . V ery good y ie ld s w ere o b ta in e d .

U n d er th e co n d itio n s em p lo y ed , n o ev idence w as fo u n d fo r a n y reaction except a lk y la tio n . T he a lk y l benzenes fo rm ed w ere id en tified b y m e a n s of p h ysica l constants an d by conversion to th e m o n o ace tam in o d e r iv a tiv e s . ' W . E . J . B.

313 Hydrogen Fluoride as a Condensing A gent. H I. A lkylation of Arom atics with Aliphatic Halides. J . H . S im ons a n d S. A rch er. J . Amer. chem. Soc., 1938, 60,- o - .9 5 4 . H ydro g en fluoride h a s been fo u n d to be a n e ffe c tiv e a g e n t to promote

ABSTRACTS. 103 A

th e re a c tio n b e tw e e n a lk y l h a lid e s a n d a ro m a tic co m p o u n d s. T h e halid es u sed w ere n -p ro p y l b ro m id e , iso p ro p y l ch lo rid e , f -b u ty l ch lo rid e a n d f-am yl ch lo ride , a n d th e a ro m a tic c o m p o u n d s w ere ben zen e , to lu e n e a n d n a p h th a le n e . W . E . J . B .

314. H ydrogen Fluoride as a Condensing A gent. IV. R eaction of a/cioPropane with Benzene. J . H . S im ons, S. A rc h e r a n d E . A dam s. J . Amer. chem. Soc., 1938, 60, 2955-2956.— c y d o P ro p a n e h a s b een fo u n d to r e a c t w ith b enzene c a ta ly z e d b y h y d ro g en fluoride to g ive n o rm a l p ro p y lb e n z e n e s in good y ie ld . A possib le m ech an ism for th e reac tio n is p o s tu la te d w h ich in v o lv es a n ion ic m ech an ism . T h e cyclopropane m olecule is a ssu m ed to a d d a p ro to n to fo rm a h y p o th e tic a l a n d tr a n s i to ry p ro p y l ion.

W . E . J . B .

it)315. System Correlating M olecular Structure of Organic Compounds with their B oiling- Points. I. A liphatic B oiling-Poin t Num bers. C. R . K in n e y . J . Amer. chem. Soc., 1938, 60, 3032-3039 .— T h e a u th o r h a s developed a sy s te m fo r c o rre la tin g th e s tru c tu re of organ ic co m p o u n d s w ith th e ir b o ilin g -p o in ts . T h e co n cep t o f b o ilin g -p o in t n u m b e r (B .P .N .) h a s b een in tro d u c e d a n d v a lu e s of 0-8 a n d 1 0 u n i ts h a v e resp ec tiv e ly been asc rib ed to c a rb o n a n d h y d ro g e n as th e a to m ic B .P .N ’s.

I n co n sid erin g a n y m o lecu le th e B .P .N . is co n sid ered to be com posed of tw o p a r ts , t h a t du e to th e h y d ro g e n a to m o r th e h y d ro c a rb o n resid u e , a n d t h a t du e to th e c h a ra c ­te r is tic a to m o r g ro u p . I n g en era l, th e c a lc u la te d v a lu es a re in good a g reem en t w ith th o se o b se rv ed , a n d i t is a n t ic ip a te d th a t th e n ew m e th o d w ill becom e a usefu l help in e i th e r c a lc u la tin g th e b o ilin g -p o in t of a co m p o u n d fro m i t s s tru c tu re , o r i t s s tru c tu re fro m th e b o ilin g -p o in t. W . E . J . B.

or)

316. Viscosities of H ydrocarbons. Parts V n and VIH. E . B . E v a n s . J. Instn Petrol. Tech., 1938, 24, 5 37-553 .— T h is p a p e r c o n ta in s d a ta on th e v isco sities of n in e teen m onocyclic a ro m a tic h y d ro c a rb o n s w ith fro m s ix to s ix ty ca rb o n a to m s a n d of s ix ty - tw o po lycyclic h y d ro c a rb o n s co m p ris in g co m p o u n d s w ith tw o o r m ore r in g s tru c tu re s . T he la t t e r c o n ta in f ro m n in e to th i r ty - e ig h t c a rb o n a to m s a n d m a n y of th e m m a y be of in te re s t in c o n n ec tio n w ith th e co m p o s itio n a n d p ro p e rtie s of lu b r ic a tin g oil frac tio n s .

D . L . S.dm

317. Action of M ineral Acids on Primary Nitroparaffins. S. B . L ip p in c o tt a n d H . B . H ass. Industr. Engng Chem., 1939, 31 (1), 118-120.— A cetic , p ro p io n ic , b u ty r ic a n d iso b u ty ric a c id s h a v e b e e n p re p a re d b y re flu x in g th e co rre sp o n d in g p r im a ry n itrop a ra ffin s w ith 8 5 % su lp h u ric ac id . H y d ro x y la m in e ac id su lp h a te is a b y -p ro d u c t

i t : in th e re a c tio n . P ro p io n o h y d ro x a m ic ac id w as p re p a re d fro m 1 -n itro p ro p an e a n dco n cen tra ted su lp h u r ic ac id . H . E . T .

318. Surface Tension of H ydrocarbons. D . L . K a tz a n d W . S a ltm a n . Industr. Engng Chem., 1939, 31 (1), 91.— T h e su rface te n s io n s of e th a n e , p ro p a n e a n d n - b u tan e w ere d e te rm in e d u n d e r e q u ilib riu m v a p o u r p ressu res o v er th e te m p e ra tu re range 0° to 45° C. T h is d a t a a n d t h a t fo r o th e r p a ra ffin h y d ro c a rb o n s a n d h y d ro - carbon m ix tu re s g iv en in th e l i te ra tu r e a re c o rre la te d b y p lo tt in g su rface ten s io n ag a in s t red u ced p ressu re . F ro m th i s g ra p h a c h a r t c o n n ec tin g su rface ten s io n , te m p e ra tu re a n d m o lecu la r w e ig h t is d ra w n . T h e su rface te n s io n s of n o n -p ara ffin

. g hy d ro carb o n s a re p lo t te d a g a in s t te m p e ra tu re o n a th i rd g ra p h . T h e red u ced te m ­p e ra tu re p lo t ca n b e u se d to e s t im a te c r it ic a l te m p e ra tu re s of m ix tu re s w hose su rface tens io n is k now n . P- D .

ttP'319. Catalytic A gent as Im portant Factor in the Pyrolysis Process. R . F u ss te ig . Petrol. Engr, D ecem b er, 1 9 3 8 ,10 (3), 33.— T h e b e h a v io u r of c a ta ly s ts is ex p la in ed o n th e basis of m o d e m th e o r ie s of th e s t r u c tu r e of a to m s , a n d th e re su lts of e x p e r im e n ta l w ork on th e p r e p a ra t io n a n d se lec tio n of c a ta ly s ts fo r use in p y ro ly s is re a c tio n s d is ­cussed. T he su rface o f th e c a ta ly s t is co n s id ered to co n s ist of u n s a tu ra te d m a t te r w hich ad so rb s a m o lecu le of, e.g., h y d ro g e n s p li t t in g a to m s in to e lec tro n s a n d nuclei, th e form er p e n e tra t in g th e lig h t m o v ab le e lec tro n s of th e c a ta ly s t a n d th e la t t e r ad h e r- ing to th e su rface of th e c a ta ly s t w h ere th e y a re ab le to re a c t w ith s p li t h y d ro c a rb o n

BtttP

1 04 aABSTBACTS.

nuclei T he adso rbed m olecules a re s u b m itte d to a d e fo rm a tio n w h ich fo rm s a new field of force opposite th e n e ig h b o u rin g gas sp ace , i.e., b eco m in g m o re ac tiv e . The q u a n tity an d q u a lity of th ese lig h t m o v ab le e lec tro n s in th e c a ta ly s t c o n tro l i t s ac tiv ity an d depend on th e n a tu re an d m e th o d of a c tiv a tio n o f th e c a ta ly s t . C a ta ly s ts form ed b y p rec ip ita tio n on e a r th a re v e ry v o lu m in o u s, a n d re q u ire a h ig h g as v e lo c ity w hich reduces yields, a n d h av e poor h e a t c o n d u c tiv itie s . T h e sk e le to n ty p e in w hich Ni is m elted w ith A1 o r Si a n d th e la t t e r rem o v ed w ith N a 2C 0 3 o r N a O H is b e t te r inth is respec t. . .

T he p a rtic le size of th e c a ta ly s t is of e x tre m e im p o rta n c e as a m ed iu m -sized gram con ta in s a g rea te r n u m b er of e lec tro n s on i t s su rface th a n th e fine-sized g ra in . The traces of N i(N 0 3)2 in p re c ip ita te d c a ta ly s ts red u ces th e i r life ow ing to th e ir d is tu rb an ce of th e electric field a t m a n y p o in ts o n th e su rface . T h e e le c tro n s a n d n u c le i of nickel skeleton c a ta ly s ts a re considered m ore a c t iv e in in d u c in g th e fo rm a tio n o f olefine hydrocarbons th a n those of p re c ip ita te d n ickel.

I n a s tu d y of c a ta ly s ts p re p a re d fro m d iffe ren t p ro p o r tio n s of A1 a n d N i i t was found th a t a 60-40 A1 N i a llo y g av e th e b e s t r e s u lts a n d t h a t th e a d d it io n of F e and Co im proved th e a c tiv ity , w hile M n a n d Cu re d u c e d i t , in s p i te of th e fa c t t h a t th e iron skeleton is q u ite in ac tiv e , a n d th e M n sk e le to n v e ry a c tiv e . I t is co n sid ered th a t the co n tac t of c e rta in m e ta l sk e le to n s causes a n in c rease of th e s p l i t t in g in to electrons an d nuclei, an d th a t , a s th e e lec tr ica l p o w er o f a to m s o f p re c ip ita te d m e ta l differ from those of th e sam e m e ta l as a sk e le to n , o p tim u m re su lts a re o b ta in e d b y com bining th e tw o ty p es in th e p ro p e r ra tio .

G asoline-form ing c a ta ly s ts m a y be of th re e ty p e s (i) th o se u n a b le to keep labile th e ac tiv e n :C H 2 rad ica ls long en ough , r e s u ltin g in h y d ro g e n a tio n to paraffinic h y d rocarbons or p o ly m eriza tio n , (ii) th o se ab le to k eep th e ~ C H 2 ra d ic a ls lab ile long enough to allow olefine h y d ro c a rb o n s to be fo rm ed , (iii) th o se h a v in g in add ition sufficiently ac tiv e nuclei a n d e lec tro n s to t r a n s fo rm th e d e f in e s in to cyclic h ydro ­carbons.

O th er c a ta ly s ts te s te d inc lude C o-A l sk e le to n (low a c t iv i ty — v e ry sh o r t life), and a Co ske leton im p re g n a ted w ith p re c ip ita te d Co (h igher a c t iv i ty th a n e ith e r elem ent). T he ad d itio n of 10% of p re c ip ita te d T h o riu m o r M n p lu s p re c ip ita te d Co to th e Co skeleton increased th e y ie ld v e ry co n s id e rab ly . A N i-S i sk e le to n a lso gave v ery low yields th o u g h a l i t t le im p ro v e d b y th e a d d it io n of p re c ip i ta te d N i a n d T h o riu m . On th e o th e r h a n d , th e C o-S i a lloy g av e b e t te r re su lts th a n th e A l-C o a llo y . T he highest y ields w ere o b ta in ed w ith a co b a lt sk e le to n fro m a C o -S i a llo y m ix e d w ith p rec ip ita ted Co an d im p re g n a ted w ith a Mo s a lt a n d re d u ced .

I t is em phasized th a t th e p ro p e r u se of c a ta ly s ts in v o lv es th e se lec tio n of m ix tures of c a ta ly s ts of th e co rre c t ty p e fo r th e r e q u ire d re a c tio n s . C. L . G.

Analysis and Testing.320. Determination of the Aniline Point of Dark Petroleum Products. W . R . VanW ijk an d J . W . M. B oelhouw er. J . Instn Petrol. Tech., 1938, 24 , 598-604 .— The aniline p o in t of d a rk oils is d e te rm in e d b y m e a su rin g th e i r tr a n s p a re n c y to infra-red rays. A n a p p a ra tu s fo r te s tin g s ix sam p les s im u lta n e o u s ly is describ ed .

D. L. S.

Motor Fuels.321. Patents on Motor Spirit. H . D rey fu s . E . P . 496.292, 29.11 .38. A pp l. 29.5.37.Synthesis of h y d ro carb o n s fro m CO a n d H 2 a t 150-350° C. a n d u n d e r a n abso lu te p ressure of 0 T —0-25 a tm . a t th e m o s t, in th e p re sen c e o f a h y d ro g e n a tin g ca ta ly s t, e.g., a m eta l of G roup 8 a n d a m e ta l o x ide . S te a m , CO», o r h y d ro c a rb o n gases o r vapours a re added as d ilu en t. 6

A. E . J . L . G erm e. E .P . 496,607, 2.12.38. A p p l. 9 .6 .37 . T re a tm e n t of n a tu ra l an com m ercial m ix tu re s of h y d ro ca rb o n s c o n ta in in g a ro m a tic s to e n r ic h th e arom atic con en , j a ow ing n a sc e n t 0 2 (from th e a c tio n o f so lid p o ta s s iu m b ich ro m ate and

ABSTRACTS. 105 A

c o n c e n tra te d H .,S(): ), to a c t o n th e p o ly e th y le n e d e r iv a tiv e s . T h e t r e a te d m ix tu re is w ashed f irs t w ith w a te r a n d th e n w ith 1 % so lu tio n of a n h y d ro u s N a .C O . a t 5 0 - 60° C.

N . V . d e B a ta a fs c h e P e tro le u m M ij. E .P . 496,676, 5 .12 .38. A p p l. 7.6 .37. M an u fac tu re o f ¿sobu tene d e r iv a tiv e s b y iso m e riz a tio n a t 325° C. of bu ten e-1 o r b u tene-2 o r m ix tu re s o f th e se c o m p o u n d s . T h e re a c tio n is ca rr ie d o u t in th e p resence of w a te r v a p o u r a n d a c a ta ly s t o b ta in e d b y m ix in g p h o sp h o ric ac id a n d k iese lguhr an d h e a tin g th e m ix tu re to 300° C. T h e m ix tu re o b ta in e d m a y , if d es ired , be po lym erized a t 150-250° C. a t ra is e d p re ssu re u s in g p h o sp h o ric a c id a s c a ta ly s t a n d th e p o ly m er gaso lin e h y d ro g e n a te d , if d es ired .

R u h rch em ie A .-G . E .P . 496,718, 1.12.38. A p p l. 1.6.37. S y n th esis of benzene from CO a n d H 2 in th e p re sen c e o f a so lid c a ta ly s t in a su ita b le h ea t-e x c h a n g in g a p p a ra tu s co m p ris in g a g a s - t ig h t re a c tio n c h a m b e r t r a v e rs e d b y a sy s te m of tu b e s d isposed in a n in c lin ed p o s itio n .

G. W . Jo h n s o n . E .P . 496,880, 5.12.38. A p p l. 3 .6.37. C onversion of CO a n d H 2 in to h y d ro c a rb o n s c o n ta in in g m o re th a n 1 C a to m a n d n o n -gaseous oxygen- co n ta in in g d e r iv a tiv e s o f h y d ro c a rb o n s u s in g a s c a ta ly s t red u c ib le iro n com pounds.

S ta n d a rd O il D ev e lo p m en t Co. E .P . 497,255, 15.12.38. A p p l. 24.12.37. S w ee ten ­ing p e tro le u m d is t i l la te s c o n ta in in g m e rc a p ta n s b y c o n v e rtin g th e m e rc a p ta n s in to copper m e rc a p tid e s w h ich re m a in d isso lv ed in th e d is t i l la te , a n d th e n e x tra c tin g th e Cu sa lt w i th e th a n o la m in e . W . S. E . C.

C. O. H o o v er. R e is su e 20,938, 6.12.38. A p p l. 25.5.38 (o rig inal U .S .P . 2,042,050, 26.5.38). S w eeten in g m e rc a p ta n -c o n ta in in g d is t i l la te s b y a d d in g a n o x ygen-con­ta in in g gas a n d a co n tro lle d a m o u n t of m o is tu re a n d c o n ta c tin g th e r e su lta n t m ix tu re w ith a n a d so rb e n t m a te r ia l a n d a c o p p e r c o m p o u n d w h ich fo rm s a m e rc a p tid e — e.g., cupric ch loride.

See also A b s tra c t N o . 285.

Gas, Diesel and Fuel Oils.322. Ageing of I.C. E ngine Oils. C. A . B o u m an . Ann. Off. Combust, liq., 1938, 13(2), 353-364.— A s a r e su lt o f p re v io u s o b se rv a tio n s , su p p le m e n te d b y p re se n t e x ­p e rim en ta l re su lts , i t is co n s id ered t h a t “ so o t ” is th e ch ief source of c o n ta m in a tio n of th e su m p o il, th e a m o u n t b e in g g re a te r in th e case of diesel eng ines th a n in p e tro l engines. T h e sam e o il w as te s te d in e ach of 6 eng ines, 3 d iese l (sing le-cy linder) a n d 3 petro l. T ech n iq u e w as a s fo llow s : S p eed 1000 r .p .m ., fo u r h o rn s a t h a lf-lo ad , th e n a s to p fo llow ed b y fo u r h o u rs a t fu ll-lo ad . “ S o o t ” (w hich in c lu d ed fine carb o n ) w as alw ays g re a te r t h a n “ la c q u e r ” (in so lub le in 6 0 -80 s p ir i t , so lub le in alcohol) o r a sphaltenes (inso lub le in 6 0 -80 s p i r i t a n d a lcoho l, so lu b le in benzene) o r ash . S im ila r resu lts w ere o b ta in e d in a 6 -cy lin d e r d iese l eng ine in se rv ice . T h e v a r io u s c o n ­ta m in a n ts d id n o t in c rease in d e f in ite ly w ith tim e , b u t te n d e d to re a c h lim itin g v a lu es w hen co rre c tio n is m a d e fo r new o il a d d e d . I t is co n c lu d ed t h a t since th e m a jo r p a r t of c o n ta m in a tio n is e x te rn a l a n d in d e p e n d e n t of o il q u a li ty , n o a r tific ia l ageing te s t can g ive a p ra c tic a l an sw er. J - L . T .

323. Gas Oils and Fuel Oils. E . P ré v o s t . Rev. Comb. Liq., O c to b er, 1938, 16 (158), 271—282.— T he a u th o r tra c e s th e h is to ry o f g as o ils a n d fuel o ils, a n d i t is in te re s tin g to n o te t h a t th e sp e c ifica tio n of th e U .G .I . (U n ite d G as Im p ro v e m e n t Co.) w as th e only one in e x is te n ce in 1920 a n d i t s e n tir e p u rp o se w as to specify o ils su ita b le for enrich ing coal gas. T h e f irs t official sp e c ifica tio n s w ere fo rm u la te d o n th e 3 rd F eb ru ary , 1922, u n d e r th e h e a d in g of “ U n ite d S ta te s G o v ern m en t S ta n d a rd specifica­tio n No. 2c,” b u t th e s e d id n o t in c lu d e g as o ils. F re n c h g o v e rn m e n t specifica tio n s as defined b y a g o v e rn m e n t d ecree in 1919 ca ll fo r a m in im u m v isc o s ity , a m a x im u m (5% ) ° f m a te r ia l re m o v a b le b y su lp h u r ic ac id , a n d th e m a in s t ip u la t io n t h a t i t m u s t yield a d is t i l la te o f less t h a n 10% a t 275° C. b y th e L u y n e s B o rd a 3

H

106 A ABSTRACTS.

m ethod th e th e rm o m e te r be in g in th e liq u id . I t is s t a te d t h a t fo r a n y g iv en h y d ro ­carbon th a t frac tio n w hich d is tils below 275° C. b y th e a b o v e m e th o d w ill d is t i l below 255° C b y th e s ta n d a rd A .S .T.M . m e th o d .

T he decrees w ere m odified in J u n e 1921 a n d A p ril 1923 to p e rm it th e im p o rt of „„„ oiis d is tillin g 20% a n d 30% re sp e c tiv e ly below 275° C. b y th e L u y n es B ordas m ethod , th is m eth o d being officially rep laced b y th e A .S .T .M . d is t i l la t io n in 1934.

I n 1937 a decree w as issued defin ing a s gas o ils a ll p ro d u c ts d is ti l l in g below 30% a t 255° C. b y th e A .S.T.M . m e th o d a n d as fuel o ils a ll p ro d u c ts d is til l in g below 30% a t 270° C. b y th e sam e m e th o d a n d sa tis fy in g th e co lo rim e tr ic te s t . T h e la t te r te s t was in tro d u c ed in 1936 to rep lace th e su lp h u ric t a r t e s t a n d is s t a te d to g ive an in d ica tio n of th e a sp h a lten e c o n te n t.

T he artic le gives a v e ry d e ta ile d s tu d y of th e p o s it io n b o th in TJ.S.A. a n d F rance b etw een 1919 a n d th e en d of 1937. M . M. L .

324. The Ignition Quality of D iesel Fuels. J • R iffk in . Engineering, 1939, C X L V II(3808), 1. T he m easu rem en t of th e d e lay an g le in a n en g in e is d iscussed in relationto engine knock , a n d i t is sugg ested t h a t a b e t t e r c o r re la tio n w o u ld b e o b ta in ed if com m encem ent of co m b u s tio n w ere ta k e n as co m m en cem en t of r a p id com b u stio n as m easu red b y th e com m encem en t of a ra p id ch an g e in p re ssu re . S im ila rly , i t is argued th a t com m encem ent of fuel in je c tio n sh o u ld be co n s id e red a s co m m en cem en t of rap id fuel in jec tion .

T he effect of th e in je c tio n t im in g on th e d e la y an g le is d isc u sse d w ith reference to the p rocedure to be a d o p te d in d e te rm in in g c e ta n e n u m b e rs , a n d som e d a ta o b ta in e d from e x p erim en ts m ad e o n a T an g y e o p e n -c o m b u s tio n -c h a m b e r en g in e a re used in illu s tra tio n .

G raph ical re la tio n s a re g iven b e tw e e n th e d e la y a n g le o f b le n d s of reference fuels as m easu red in th e T an g y e eng ine, a n d th e co rre sp o n d in g v a lu e s of th e A n iline P o in t, D iesel In d e x , V isc o s ity -G ra v ity In d e x , a n d S p o n ta n e o u s I g n i t io n T e m p e ra tu re .

J . G. W.

Lubricants and Lubrication.325. Aircraft-Engine Lubrication. E . L . B ass a n d C. H . B a r to n . J . Soc. aut. Engrs, 1939, 44 (1), 8 -14 .— V alu ab le d a ta c an be o b ta in e d , i t is co n s id e re d , fro m prelim inary te s ts in th e la b o ra to ry , u s in g sin g le -cy lin d er en g in es , o n su c h fa c to rs as ring-stick ing u n d e r d e to n a tin g (take-off), a n d w e a k -m ix tu re (cru ising ) c o n d itio n s , c a rb o n form ation , b earin g corrosion a n d slud g in g . S ince co rre la tio n is n o t p e r fe c t, fu ll-scale engine te s ts on th e b ench a n d in flig h t a re fin a lly n ecessa ry . E n g in e s u se d w ere J .A .P . and N o rto n (m otor-cycle) a n d B ris to l (aero.). O p e ra tin g d e ta i ls u s in g th e J .A .P . engine w ere : fo r rin g -s tic k in g te s ts , 5 h o u rs u n d e r (a) m a x im u m p o w er w ith detonation , (b) 10% w eak m ix tu re w ith no d e to n a tio n . U n d e r (a) c y lin d e r te m p e ra tu re h a d to be low ered 10° C. to o b ta in th e sam e degree of r in g g u m m in g as w ith no d e to n a tio n .

T he te s t on th e B ris to l eng ine w as lo n g er a n d v a r ie d f ro m 654—75 h r . fo r different oils. T he N o rto n eng ine is be ing u se d fo r te s ts of lo n g d u ra t io n a n d a lso to determ ine th e effect o n rin g -s tic k in g of c a rb o n fo rm a tio n in th e g ro o v es (“ rin g -p ack in g ” ) b eh in d th e rin g . So fa r th e re a p p e a rs to b e l i t t le co n n e c tio n . F o r b e a r in g corrosion te s ts , th e b ig en d w as ru n -in fo r 4 h r . a t n o rm a l o p e ra t in g te m p e ra tu re (50—60° C.) follow ed b y 20 h r . a t te m p e ra tu re s u p to 170° C. A n o il is co n s id e red sa tis fac to ry for m o st pu rposes p ro v id ed co rrosion of lead -b ro n z e b e a r in g s d oes n o t develop below 145° C. Som e d ifficu lty h a s b een ex p e rien ced in c o r re la tin g sm a ll-en g in e te s ts w ith full-scale eng ine te s ts in th e case of cad m iu m -b ase b e a r in g s . W ith su ch bearings th e re ap p ea rs to be in creased d a n g e r fro m co rro s io n b y o il. T h e re a re n o te s on the o th e r fac to rs m en tio n ed . As re g a rd s ca rb o n fo rm a tio n a n d th e effec t on i t of tem ­p e ra tu re an d oil co n su m p tio n th e in te re s tin g r e su lt is g iv en t h a t in th e case of the J .A .P . engine, a lth o u g h c a rb o n on th e p is to n cro w n d id n o t in c rease w ith cylinder h ead te m p e ra tu re , th e increase in th e c a rb o n in th e to p rin g -g ro o v e w as v e ry m arked as te m p e ra tu re increased . j p . t .

326. Some Developm ents R elative to Crankcase-Oil F iltration. A . T . M cD onald,J . ooc. aut. Engrs, 1939, 44 (1), 2 3 -2 8 .— A n a t t e m p t is m a d e in th is a r tic le to clarify

ABSTRACTS. 107 A

gome m isco n c ep tio n s a r is in g o u t o f th e c la im s m a d e fo r th e su p e r io r ity o f th e a d so rb e n t (F u llers E a r th , C ha rco a l) ty p e o v e r th e a b s o rb e n t (C o tto n W a s te ) ty p e of filte r.

I n a d d it io n to th e re m o v a l of n a tu r a l o ilin ess a g e n ts a n d a d d it iv e s des ig n ed to g ive g re a te r film s t r e n g th a n d freed o m fro m r in g -s tic k in g in diesel eng ines, th e fo rm er ty p e te n d to d is in te g ra te a n d a d d a b ra s iv e m a te r ia l to th e o il. F u r th e r , i t is consid ered th a t fresh o il te n d s t o w ash o u t im p u rit ie s fro m such f ilte rs w h ich , an y w ay , c a n n o t p oss ib ly cope w ith a ll th e a c id i ty in a n o il a n d m o reo v e r h a v e l im ite d life .

A s fa r a s th e a b s o rb e n t ty p e is co n ce rn ed , th e se su ffer f ro m th e d e fec t t h a t th e y on ly t r e a t 10% a s th e o il a t a n y t im e , th e re m a in in g 90% b e in g b y -p assed . T h ey shou ld b e su p p le m e n te d b y sc ree n f ilte rs of th e fu ll-flow ty p e .

G rap h s a re g iv en sh o w in g th e p ro g ress iv e decrease in th e efficiency of th e a d so rb e n t ty p e f ilte r in re m o v in g a d d it iv e , sp ecifica lly m e ta l so a p , p lo t t in g re su lts b o th fro m a la b o ra to ry t e s t u s in g p u m p c irc u la t io n o f o il, a n d a lso a c tu a l o p e ra tio n .

T h e c o n te n tio n s re g a rd in g th e d a n g e rs fro m th e u se o f th e a d s o rb e n t ty p e f ilte r a re su p p o rte d b y figures o n u se d o ils a n d p h o to g ra p h s of p is to n s fro m eng ines b o th w ith a n d w ith o u t th e a b s o rb e n t ty p e of filte r .

Som e u se fu l figures o n th e a c c e le ra tin g effect o f in c rease d te m p e ra tu re on c rack in g of b a b b it m e ta l h e a r in g s a n d o n co rro s io n o f c a d m iu m -b ase b ea rin g s a re also q u o ted . I n th is co n n e c tio n th e p h o to g ra p h s o f b a b b i t b e a r in g she lls fro m th e sam e eng ine o p e ra tin g u n d e r th e sam e c o n d itio n s sho w in g th e im p ro v e d c o n d itio n re su ltin g fro m fre q u e n t ch an g in g w ith o u t f i lte r a s co m p a re d w ith lo n g -tim e o p e ra tio n w ith a b so rb e n t- ty p e f ilte r a re in te re s tin g .

I n th e sing le c o n tr ib u tio n t o th e d isc u ss io n o f th is p a p e r th e re is a t im e ly w a rn in g on th e d a n g e rs o f d ra w in g co n c lu s io n s f ro m re s u lts o n one p a r t ic u la r d iesel eng ine .

J . L . T .

327. Im provem ents in D iesel-E ngine Lubricating Oils. U . B . B ra y , C. C. M oore, J r . , a n d D . R . M e rrill. J . Soc. aut. Engrs, 1939, 44 (1), 35-42.— R e q u ire m e n ts o f a diesel- eng ine lu b r ic a n t a re s ix -fo ld :

(i) D e te rg e n c y t o k eep r in g s free a n d so lid s in su spension .(ii) H ig h o ilin ess t o ta k e ca re of n o rm a l o p e ra tin g co n d itio n s .

(iii) H ig h film s t r e n g th to p re v e n t scuffing a n d sco ring u n d e r a b n o rm a l c o n d itio n so f s tre s s .

(iv) N o n -s lu d g in g p ro p e r t ie s , i.e., h ig h o x id a tio n s ta b il i ty .(v) N o n -co rro s iv e n ess to n e w - ty p e a llo y b ea rin g s .

(vi) L ow c a rb o n -fo rm in g te n d e n c ie s .

A s re g a rd s (i), m e ta l so a p s a re co n s id e red effec tiv e a d d it iv e s to m a k e good th e deficiency o f s t r a ig h t m in e ra l o ils . A s re g a rd s (ii) th e c a rb o n y l g ro u p in a so ap is th o u g h t to b e a n a c tiv e f a c to r , w h ils t fo r (iii) th e ch lo rin e in ca lc iu m d ic h lo ro -s te a ra te is co n sid ered re sp o n s ib le fo r such im p ro v e m e n t sh o w n b y o ils co n ta in in g i t . I t is b e s t a d d e d to n a p h th e n ie ty p e o ils a l th o u g h para ffin ic ty p e o ils a re also im p ro v e d . A s re g a rd s (v) n o m e ta l so a p a d d i t iv e is n o n -co rro s iv e to lead -b ase a n d cad m iu m - b ase b e a r in g a llo y s b u t th e a u th o r s th in k i t is o n ly a m a t te r of t im e befo re th is p ro b le m is reso lved .

P h o to g ra p h s of p is to n s a n d figu res fo r w e a r a n d a n a ly ses o f u se d o ils a re p u t fo rw ard to su b s ta n t ia te th e c la im s m a d e fo r ca lc iu m d ic h lo ro s te a ra te as a n im p ro v e r of diesel- eng ine o ils. J . L . T .

328. Rapid A nalysis of som e M ineral Lubricating Oils. E . A n d ré a n d J . R o c h e . Ann. Off. Combust, liq., 1938, 13 (2), 339-351.— T h e co m p o u n d s in a m in e ra l oil (R o u m an ian “ R R O ” ) co n s id e red re sp o n s ib le fo r r o ta t in g th e p la n e of p o la rized lig h t to th e r ig h t a re th o u g h t to b e c o n c e n tra te d in t h a t f ra c tio n o f th e o il w h ich is m o st so lub le in a c e to n e a n d m e th y l a n d e th y l a lcohols. T hese co m p o u n d s a re of re la tiv e ly low m o le c u la r w e ig h t a n d c o n ta in c a rb o n , h y d ro g en , o x y g en a n d su lp h u r . T h ey a re a s so c ia te d w i th h y d ro c a rb o n s (C 20- C 22) p o o r in h y d ro g e n a n d of g re a te r d e n s ity th a n w a te r . I t is n o t k n o w n w h e th e r th e y a re p re s e n t a s a q u a te rn a ry com pound o r as a m ix tu re o f te r n a ry c o m p o u n d s . J . L . T .

329. Safe V iscosity for a Motor-Car E ngine Lubricant. S. W . Sparrow. J . Soc. aut. Engrs, 1938, 43 (4), 393 -40? .— On th e cred it side, lo w -v isco sity o ils g ive h igh

1 08 aABSTRACTS.

crank ing speeds, ra p id flow a t low te m p e ra tu re s , in c rease in b ra k e to rq u e a n d b e t te r fuel econom y. On th e d e b it side a re te n d e n c y to in c rease d o il co n su m p tio n an d b low by decreased p ro te c tio n to m a in a n d b ig -en d b ea rin g s , p is to n s a n d cy lin d er walls As regards bearin g p ro te c tio n th e th e o ry is a d v a n c e d th a t th in o ils fa il because th e oil p u m p is u n ab le to b u ild u p suffic ien t p re ssu re to overco m e c e n tr ifu g a l force th e reb y m ilita tin g ag a in s t a d e q u a te flow, th e c r i te r io n o f sa fe v isc o s ity fo r con-rod bearings. L arger p u m p s a n d im p ro v e d o il sy s te m s m a y e v e n tu a lly overcom e th is . Safe v iscosity d ep en d s o n o p e ra tin g te m p e ra tu re . F ig u re s q u o te d fo r crankcase oil te m p e ra tu re range from 89-178° F . ab o v e a tm o sp h e ric te m p e ra tu re fo r a dozen 1938 cars a t 70 m .p .h ., a n d 105-195° F . ab o v e a tm o sp h e ric te m p e ra tu re a t to p speed. Surface finish p lay s an im p o rta n t p a r t . D iffe ren t c y lin d e rs in th e sa m e eng ine differed w idely in th e ir v isco sity re q u irem en ts .

The a u th o r believes t h a t th e v isc o s ity of th e o il u se d in a n ew en g in e sh o u ld be a t least as high as th a t w h ich is specified fo r a w ell ru n - in en g in e , a n d t h a t th e g rea te r p ro tec tio n ag a in s t scoring w hich is p ro v id e d b y a m o re v isc o u s o il m o re th a n com ­pensates for th e e x tra tim e re q u ire d to w ea r a w a y th e h ig h sp o ts of th e cy lin d e r w all surfaces. I n w in te r, o n cars f i tte d w ith th e rm o s ta t ic c o n tro l o f w a te r - ja c k e t te m ­p e ra tu re g iv ing figures as h ig h as in su m m er o p e ra tio n , lo w -v isco s ity lu b r ic a n ts m ay be unsafe for p is to n s a n d cy lin d ers . T h is d a n g e r is so m e w h a t offset, i t is a d m itte d , by th e low er d riv in g speeds in w in te r . N ev e rth e le ss , fo r sa fe o p e ra tio n of p is to n s an d cy linders oil v isc o sity sh o u ld b e as h ig h in w in te r a s in su m m er. T h e a u th o r p u ts h is resu lts m ore in th e fo rm of a sk in g q u e s tio n s th a n in p ro v id in g th e answ ers. T he graphs an d p h o to g rap h s a re ex ce llen t. T h e ra n g e in v e s tig a te d w as from S.A .E. 10W to S .A .E . 50, b u t v isc o sity ind ices of th e o ils co n s id e red a re n o t q u o te d .

J . L . T .

330. Specific Refractivity and Carcinogenicity of M ineral Lubricating Oils. S. J . M.A uld. J . Instn Petrol. Tech., 1938, 24, 577-583 .— T h e a u th o r h a s c a rr ie d o u t some w ork on th e specific g ra v ity a n d specific r e f r a c tiv i ty of o ils a n d o f b le n d s of d ifferent ty p es of oil from w hich h e concludes t h a t th e re a re se rio u s in co n s is ten c ie s in the M anchester C om m ittee’s specifica tion fo r n o n -ca rc in o g en ic o ils. D . L . S.

331. Manufacture of Cutting Oils. J . F . M iller. J . Instn Petrol. Tech., 1938, 24, 645-649.— V arious ty p e s of c u tt in g o ils a re d isc u sse d a n d th e p ro p e rtie s dem anded of these p ro d u c ts a re d escribed . D . L . S.

332. Cutting Fluids and the M achine Tool. A . H . L lo y d a n d H . H . B e en y . J. Instn Petrol. Tech., 1938, 24, 650-654.— T h e em u ls io n ty p e o f c u t t in g f lu id is d iscussed in con ju n c tio n w ith i t s possib le co rrosive effect o n th e c u t t in g to o l. D . L . S.

333. Selection of Cutting Fluids. H . J . M ason . J . Instn Petrol. Tech., 1938, 24, 655-661.— T he a u th o r o u tlin es som e p ra c tic a l te s ts w h ic h h a v e b een c a rr ie d o u t overa period of severa l y ears to d e te rm in e th e efficiency of c u t t in g flu id s . D . L . S.

334. Functions of Cutting Fluids. H . W . S w ift. J . Instn Petrol. Tech., 1938, 24, 662-671.— T he m ech an ism of m e ta l c u t t in g is d isc u sse d a n d th e ro le p la y e d b y the cu ttin g flu id in d ica ted . D . L . S.

335. Patents on Lubricating Oil. H . W . B ro w n sd o n a n d I .C .I . L td . E .P . 496,717,30.10.38. A ppl. 31.5.37. P re p a ra tio n of E .P . lu b r ic a n ts b y a d d in g to a m ineral lub rica tin g oil a n a d d itio n p ro d u c t— e.g., 0 1 % of p y ro g a llo l a n d 0 1 % w a te r , o r 0 0 1% of e th y l fo rm ate a n d 0 01% of w a te r , o r 0 -0005% o f s ilv e r ch lo rid e so lu tio n d issolved in aqueous am m onia , e tc .

N. V. de B a ta afsch e P e tro le u m M ij. E .P . 496,779, 6 .12.38. A p p l. 23.12.37. rocess of rem oving n a p h th e n ic ac id s fro m m in e ra l lu b r ic a t in g o ils o r lu b r ic a tin g oil

rac tio n s y co n ta c tin g th e o ils in th e v a p o u r p h a se a t te m p e ra tu re s ab o v e 300° C. in le presence of a c a ta ly s t of h ig h p o ro s ity a n d w ith a la rg e c a ta ly t ic su rface selected rom e g ro u p ; li th iu m h y d ro x id e , c a rb o n a te o r p h o sp h a te , so d iu m hydrox ide

ABSTKACTS. 109 A

or c a rb o n a te , p o ta s s iu m c a rb o n a te , b e r y l l i u m ox id e o r c a rb o n a te , m ag n es iu m h y d ro x id e o r c a rb o n a te , o r o x id es , h y d ro x id e s o r c a rb o n a te s o f ca lc iu m , a lu m in iu m , o r m an g an ese , e tc .

E d e le a n u G ese llsch aft m .b .H . E .P . 496,955, 8.12.38. A p p l. 8.6 .37. D ew ax in g of h y d ro c a rb o n o ils u s in g a n h y d ro u s p y r id in e a n d /o r i t s hom ologues u sin g n i t r o ­benzene. a n ilin e o r c h lo ran ilin e .

E d e le a n u G esellschaft m .b .H . E .P . 496,956, 8.12.38. A p p l. 8.6 .37. S o lv en t refin ing o f h y d ro c a rb o n o ils u s in g a s so lv e n t p y r id in e o r i t s hom ologues to g e th e r w ith n itro b e n zen e , a n ilin e o r c h lo ro n itr ile .

E d e le a n u G ese llsch aft m .b .H . a n d F . B . D eh n . E .P . 496.991, 8.12.38. A pp l.8.6.37. P ro c ess a s d esc rib ed in E .P . 496,956, u s in g one o f th e fo llow ing : sod . c a rb o n a te o r h y d ro su lp h id e , o r p o t . n i t r i te .

I . G. F a rb e n in d u s tr ie A .-G . E .P . 497,541, 21.12 .38. A p p l. 25.4.38. P ro d u c tio n o f lu b r ic a tin g o il b y p o ly m e riz in g p ro p y le n e a n d a -b u ty le n e o r m ix tu re s o f th ese .

H . E . R . \ ogel. E .P . 497,306, 16.12.38. A p p l. 4.12.37. P rocess a n d a p p a ra tu s for te s tin g th e lu b r ic a tin g v a lu e of lu b r ic a tin g o ils.

A . B . B ro w n a n d F . F . D iw o k y . U .S .P . 2 ,138,832, 6.12.38. A p p l. 15.10.32. S o lv en t e x tr a c t io n of lu b r ic a t in g o il u s in g m ix e d so lv e n ts— e.g., a n a lip h a tic e th e r c o n ta in in g m o re t h a n 8 C a to m s a n d ^ -c h lo r in a te d a l ip h a tic e th e r c o n ta in in g n o t m ore t h a n 8 C a to m s .

A. B . B ro w n a n d F . F . D iw o k y . U .S .P . 2 ,138,833, 6.12.38. A p p l. 2.4 .34. S o lv en t e x tra c tio n o f lu b r ic a tin g o ils u s in g a m ix tu re of d i(2 -eh lo re th y l) e th e r a n d p heno l.

A . B . B ro w n a n d F . F . D iw o k y . U .S .P . 2 ,138,834, 6.12.38. A p p l. 18.4.32. S o lven t e x tr a c t io n o f lu b r ic a tin g o ils u s in g a m ix tu re o f 2 5 -5 0 % of ort/w chloro- p h en o l, 70—40% of p a ra c h lo ro p h e n o l a n d 1 5 -1 0 % of pheno l.

L . L ib e rth so n . U .S .P . 2 ,138,868, 6 .12 .38. A p p l. 2.10.34. P ro c ess of low ering th e p o u r p o in t o f lu b r ic a t in g o ils c o n ta in in g w a x b y su b je c tin g th e w a x y o il in th e fo rm o f a th in film to th e a c t io n o f u ltra -v io le t r a d ia t io n o f 2800 A n g s tro m u n its a t te m p e ra tu re s below th e b .-p . o f th e o il a n d b le n d in g w ith a n o n - ir ra d ia te d w a x y oil.

G. H . H u tc h in s a n d A . W . H a r t ig a n . U .S .P . 2,139,161, 6.12.38. A p p l. 6 .4.36. R efin ing lu b r ic a tin g o ils b y m e a n s o f a d so rb e n t e a r th s in o rd e r to p ro d u ce co m m ercia lly u sefu l a s p h a lt a n d sp e n t a d s o rb e n t.

E . G. M c F a rla n d . U .S .P . 2 ,139,240, 6.12.38. A p p l. 24.11.36. S o lven t refin ing of h y d ro c a rb o n o ils u s in g fu rfu ra l.

F . W . B re th a n d A . K in se l. U .S .P . 2 ,139,668, 13.12.38. A p p l. 3.6.36. E x tra c tio n of P en n sy lv a n ia n o ils w ith ac e to n e .

R . R . W ilso n , E . K . B ro w n a n d F . L . W h ite . U .S .P . 2,139,871, 13.12.38. A p p l.6.8.37, a n d M . R . F e n sk e a n d W . B . M cC luer. U .S .P . 2 ,139,943, 13.12.38. A p p l.13.7.34. S o lv en t e x tr a c t io n a p p a ra tu s fo r lu b r ic a tin g oils.

E . J . M a rtin . U .S .P . 2 ,140,161, 13.12.38. A p p l. 14.11.34. M a n u fa c tu re of lu b ric a tin g o ils fro m w a x y p e tro le u m d is t i l la te s b y d ew ax in g a n d t r e a t in g th e dew ax ed oil w ith c o n c e n tra te d H 2S 0 4 a t — 10° F . a n d s e p a ra tin g th e s lu d g e a lso a t th is te m ­p e ra tu re .

J . S. W allis a n d C. T . C have. U .S .P . 2 ,140,342, 13.12.38. A p p l. 23.2.34. V acu u m d is tilla tio n of lu b r ic a t in g oils.

E . T erres , E . S a e g e b a r th , J . M oos a n d H . R a m se r . U .S .P . 2 ,140,485, 13.12.38. A ppl. 22.5.35. R e fin in g a s p h a lt ic m in e ra l o il b y d isso lv in g i t in d ich lo rod ifluo ro - m e th an e , rem o v in g th e u n d isso lv e d a s p h a lt , ch illin g th e m ix tu re to d ew ax , rem o v in g th e w ax , a n d e x tr a c t in g th e so lu tio n w ith a se lec tiv e so lv e n t in w h ich d ich lo ro d i- flu o ro m eth an e is n o t w h o lly m isc ib le .

110 aABSTRACTS.

E B H ierp e a n d W . A. G ruse. U .S .P . 2 ,141,085, 20.12 .38 . A p p l. 2.11.34. r w r o a s i n e th e v isc o sity /g r. c o n s ta n t of p e tro le u m lu b r ic a t in g o ils b y e x tra c tio n a t 0-100° F . w ith a m ix tu re of e th y len e d icb lo rid e a n d l iq u id S 0 2.

E . Torres, J . Moos a n d E . S aeg eb a rth . U .S .P . 2,141,143, 20 .12 .38. A p p l. 25.10.37. D ew axing lu b ric a tin g oil u s in g te tr a b ro m o e th a n e a n d a n a u x il ia ry so lv e n t such as benzol.

E W . Thiele a n d B . G insberg . U .S .P . 2 ,141,257, 27.12 .38. A p p l. 28.5.36. Im ­p rovem en t of th e p ro p e rtie s of s te am -re fin e d lu b r ic a t in g o ils w h ic h h a v e b een d e ­asp h a lted b y m eans of p ro p a n e a t 100-115° F . b y a d d in g i~ 2 % o f re d u c e d pressu re ta r .

P . J . H a rrin g to n . U .S .P . 2 ,141,297, 27.2.38. A p p l. 11.12.34. M e th o d of r e ­covering oil from p e tro leu m ac id sludge.

S. P i la t a n d M. G odlew icz. U .S .P . 2,141,361, 27.12 .38. A p p l. 13.4.37. D e­w axing h y d ro ca rb o n o il b y m ean s of cresol.

C. C. B uch ler a n d S. H . D iggs. U .S .P . 2 ,141,511, 27.12 .38. A p p l. 17.2.36. Solvent e x tra c tio n of o il u s in g fjp -d ich lo rd ie th y l e th e r a n d 2 -2 0 % of d ie thy lene glycol.

W . B . H e n d ry a n d L . W . C ook. U .S .P . 2 ,141,605, 27.12 .38. A p p l. 2.4.37. Solvent refin ing of lu b r ic a tin g oils u s in g as so lv e n t te t r a h y d ro fu r fu ry l a c e ta te a n d a ha logenated ace tic acid .

H . B . Setzler. U .S .P . 2,141,622, 27.12.38. A p p l. 8.10.34. S o lv e n t dew ax in g and sim u ltaneous ac id refin ing of lu b r ic a tin g o il c o n ta in in g a n in su ffic ien t q u a n t i ty of sludge-form ing su b stan ce . S til l re s id u e is a d d e d to th e o il, th e m ix tu re is ch illed to dew ax, a n d th e n t r e a te d w hile s t i l l co ld w ith 1 12SO^ in o rd e r to p ro d u c e a sludge a d a p te d to a id th e se p a ra tio n of w a x fro m th e oil.

V. I . D ow ney. U .S .P . 2,141,623, 27.12.38. A p p l. 22 .4 .37 . P ro d u c tio n of lu b r ic a t­ing oil from P en n sy lv a n ia n c ru d e o il b y f irs t re d u c in g th e v isc o s ity a n d crea ting u n sa tu ra te s o r re a c tiv e su b s ta n c e s b y s lig h tly c ra c k in g th e o il. T h ese u n sa tu ra te s a n d reac tiv e su b s tan ces a re rem o v ed to g e th e r w ith th e w a x b y ch illin g a n d th e oil is th e n a c id - tre a te d . T h e o il is s e p a ra te d f ro m th e s lu d g e , a n d b le n d e d w ith o ther oils to p roduce v ario u s com m ercia l g rad es of lu b r ic a tin g o ils.

Le R . G. S to ry . U .S .P . 2 ,141,626, 27.12.38. A p p l. 5 .2 .37 . S o lv e n t refin ing of h y d ro carb o n oil b y m ean s of a n o n-a lcoho lic so lv e n t in w h ic h th e em ulsifica tion occurring d u rin g th e e x tra c tio n a n d s e p a ra tio n in to p h a se s , is in h ib ite d b y add ing a m e ta l soap of a h ig h er f a t ty ac id . W . S. E . C.

Special Products.336. Patents on Special Products. B rim sd o w n C hem ica l W o rk s , J . C. L id d le a n d H. M eyer. E .P . 496,942, 5.12.38. A p p l. 3.6 .37. P ro d u c tio n o f a c tiv e ca rb o n by h ea tin g v egetab le— e.g., w ood ch a rco a l a n d n u t-sh e ll c h a rc o a l t o a t le a s t 700° C. an d su b je c tin g to th e a c tio n of a m ix tu re of s te a m a n d a ir .

S ta n d a rd Oil D eve lo p m en t Co. E .P . 496,966, 9 .12 .38 . A p p l. 10.6.37. M anu­fac tu re of s ta b ilized h y d ro c a rb o n p o ly m ers d e r iv e d f ro m iso b u ty le n e , of m ol. w t. above 14,000.

E . I . D u P o n t de N em o u rs & Co. a n d V . F . H a n so n . E .P . 497,234, 14.12.38. A ppl. 14.6.37. M a n u fac tu re of c a rb o n b la c k b y e le c tro th e rm a l d eco m p o s itio n of liq u id hydrocarb o n s.

G. W . Jo h n so n . E .P . 497,427, 16.12.38. A p p l. 16.6.37. C o n v ers io n of com pounds c o n ta in in g acety len e in to d e f in e lin k ag es b y t r e a t in g th e m w ith a n a q u e o u s suspension of zinc w hich is a c tiv a te d b y C u a n d Cd. W . S. E . C.

ABSTRACTS. I l l A

Detonation and Engines.337. Variation of Certain Com bustion Factors in Internal Combustion E ngines. M.Precoul. Rev. Comb. L iq., A u g .-S e p t . 1938, 16 (157), 237 -2 4 0 .— V ario u s cu rv es a re given in c o n n e c tio n w i th flam e p ro p a g a tio n , r e la t io n b e tw e e n p re ssu re a n d vo lum e of gases a n d v o lu m e d is t r ib u t io n . T h e c u rv e p — f(t) is s ta te d to b e of p a r t ic u la r in te re s t as i t g ives in fo rm a tio n o n th e su rface f ro n t of th e flam e w h ich in t u r n is p a r t ly re la ted to th e sp e ed o f co m b u s tio n . T h e m o s t im p o r ta n t f a c to r b ro u g h t o u t b y th ese curves is p ro p a g a tio n sp e ed o f th e flam e w hose v a r ia t io n ag rees in a ll re sp ec ts w ith th e a re a of th e f ro n t o f th e flam e fro m th e p o in t of v iew of en g in e p erfo rm an ce . A ll e x p e rim en ta l w o rk b e a r in g o n th e sh a p e of th e c h a m b e r is d ire c te d to w a rd s o b ta in in g a m ax im u m flam e p ro p a g a t io n sp e ed a f te r th e flam e h a s co v ered o n e - th ird o f i t s course fro m th e co m m en cem en t o f c o m b u s tio n . A lth o u g h th e cu rv es V = f(d) do n o t a p p e a r to h a v e a n y d ire c t b e a r in g o n th e c o n s tru c tio n of eng ines th e y a llow th e ca lcu la tio n of th e b e s t v o lu m e tric y ie ld o f th e ex p lo s io n c h am b er. T h u s , fro m th e above tw o cu rv es i t is p o ss ib le to o b ta in co m p le te in fo rm a tio n a b o u t th e p e rfo rm an ce a n d o u tp u t o f a n en g in e . M. M. L .

338. Effect of Sparking-plug M aterial on Octane R ating. F . F isc h e r a n d H . P o h l. Brennst.-Chemie, 15.12.38, 19 (24), 4 5 8 -4 6 0 .— R e su lts a re g iv e n of te s ts ca rr ie d o u t in a n I .G . k n o c k - te s tin g en g in e o n th e effect of sp a rk in g -p lu g d es ig n a n d m a te r ia l o n k n o ck in g . T e s te d p lu g s h a d th r e e e lec tro d es o f am p le se c tio n m a c h in e d fro m th e b o d y p ro p e r th e re b y e lim in a tin g w e ld ed o r r iv e te d jo in ts to ta k e fu ll a d v a n ta g e of th e h ig h h e a t c o n d u c t iv i ty of th e m e ta ls u se d . W ith c o p p e r a s a b o d y a n d e lec ­tro d e m a te r ia l, th e o c ta n e n u m b e r of th e fu e l in u se w as im p ro v e d fro m 54-7 o b se rv ed w ith a s ta n d a rd p lu g , to 58-5. T h is p lu g g av e sa t is fa c to ry se rv ice d u r in g th e tw o ho u rs o f te s t . S im ila r re s u l ts w ere o b ta in e d w ith a m ic a p lu g eq u ip p e d w ith Cu e lec tro d es, th e o c ta n e n u m b e r im p ro v e m e n t a m o u n tin g to 2-6 em its. W ith b ra ss as e lec tro d e m a te r ia l , th e effec t w as less, w h ile a lu m in iu m a n d le a d h a d n o effect a t a ll. L . R .

See also A b s tr a c t N o . 301.

Coal and Shale.339. Possibility of Solvent Extraction of Phenols from Coal Tar. G. A gde a n d H . S chuerenberg . Brennst.-Chemie, 15.12.38, 19 (24), 457 -458 .— T h e p o ss ib ility of e x tra c tin g low b o ilin g p h e n o ls f ro m coal t a r b y w ash in g w ith o rg an ic se lec tiv e so lv e n ts h a s b een in v e s tig a te d . T e s ts w ere c a rr ie d o u t w ith ac e to n e -w a te r , fo rm ic aeid - w a te r , a n d p e tro le u m e th e r -m e th a n o l m ix tu re s , a n d i t w as fo u n d th a t , c o n tra ry to p rev io u s a s se rtio n s , i t is im p o ssib le to o b ta in o il-free p h en o ls in one sing le o p e ra tio n .

L. R .

340. Patent on Coal. A . P o t t . U .S .P . 2 ,141,615, 27.12.38. A p p l. 30.11.35. E x ­tra c t io n of ca rb o n aceo u s m a te r ia l w ith liq u id so lv e n ts—e.g., te t r a l in e a n d p h en o l.

W . S. E . C.

Economics and Statistics.341. Report on H eavy-O il E ngine W orking Costs (1937-1938). H . V . S te a d , W . S. B u m , L. H o tin e , C. G reen a n d W . H . S k in n e r. Diesel Eng. Vs. A ss., P u b n N o. S .149.— A su m m a ry is g iv e n o f th e w o rk in g co s ts of th e d iese l p la n t r u n b y 75 d iffe ren t u n d e r ta k in g s , re p re se n tin g a to ta l of 337 en g in es . S uch fa c to rs as fue l a n d lu b r ic a tin g - o il c o n su m p tio n , a n d en g in e ren ew a ls a re d isc u sse d , w h ile m u c h a d d it io n a l in fo rm a tio n is g iv e n in ta b u la r fo rm .

C o m m en ts s e n t in b y th e c o n tr ib u tin g m em b ers a re also in c lu d ed . J . G. W .

112 A

BOOK REVIEWS.ADDlied Geophysics in the Search for Minerals. B y A . S. E v e a n d D . A . K ey s . T h ird

E d itio n . 1938. P p . 316. C a m b r id g e U n iv e rs i ty P re ss . P ric e 16s.

'_3

T he d evelopm en t of geophysica l m e th o d s o f p ro sp e c tin g is of th e g re a te s t im ­p o rtance to th e m in in g a n d oil in d u s tr ie s , a n d th e g e o p h y s ic is t to -d a y is a very essen tial lin k b etw een th e geo log ist a n d th e en g in eer.

T he d ish a rm o n y be tw een th e su rface a n d u n d e rg ro u n d fo rm a tio n s h a s been the cause of m an y expensive fa ilu res in th e p a s t , b u t to -d a y , th a n k s to th e effo rts of th e geophysicist, th e te s t bo reho le c an b e lo c a te d in th e m o s t fa v o u ra b le position .

T he v a s t a llu v ia l p la in s a re b e in g th o ro u g h ly co m b ed b y th e geop h y sic is t, and m an y g rea t discoveries h a v e a lre a d y b een m a d e .

Those in te re s te d in th e su b je c t o f a p p lie d g eo p h y sics w ill f in d lu c id accounts of th e v arious m e th o d s in th e T h ird E d it io n o f th i s b o o k b y E v e a n d K eys. I t is one of th e few books in E n g lish o n th is fa sc in a tin g su b je c t. T h e a u th o rs clearly h av e ex p e rt know ledge of som e of th e m e th o d s , p a r t ic u la r ly of th e v a r io u s electrical m ethods of p ro sp ec tin g fo r ore. In d e e d , th e p a r t o f th e b o o k d ea lin g w ith these m eth o d s co n ta in s v e ry v a lu ab le c o n tr ib u tio n s to g eo p h y s ica l l i te ra tu re .

I t is five y ears since th e a p p e a ra n c e of th e p re v io u s e d it io n , a n d in th is new volum e som e m in o r m o d ifica tio n s h a v e b e e n m a d e a n d a c h a p te r a d d e d dealing w ith th e chief adv an ces d u rin g th e p a s t y e a rs . A t te n t io n m a y b e d irec ted to Sch lum berger’s “ e lec trica l co ring ” fo r e x p lo rin g u n c a se d b o rin g s fo r oil ; the ra tio m e te r m e th o d s of e lec trica l p ro sp e c tin g ; a n d th e im p ro v e m e n ts in th e m agnetic, re s is tiv ity a n d seism ic m e th o d s of e x p lo ra tio n . A lth o u g h th e c h a p te rs on the seism ic a n d g ra v ity m e th o d s a re p e rh a p s n o t as a u th o r i ta t iv e a s th e o th ers , the s tu d e n t of geophysics w ill f ind m u ch u se fu l m a te r ia l o n th e se m e th o d s .

T he book is v ery w ell w r it te n , a n d i t w ill a p p e a l to th e e x p e r t a n d th e s tu d en t alike. J - H . J o n es .

The Exam ination of Fragmental R ocks. B y F re d e r ic k G . T ic k e ll. P p . x + 154.L o n d o n : O xford U n iv e rs ity P re ss . C a li fo rn ia : S ta n fo rd U n iv e rs ity Press.1939. §4.00.W hen P ro fesso r T ic k e ll’s b o o k w as f irs t p u b lish e d e ig h t y e a rs ago i t s p refa to ry

n o tice ad d u ced as a reaso n fo r i t s a p p e a ra n c e , th e w id e sp re a d in te re s t ta k e n in i ts su b je c t-m a tte r b y w orkers in m a n y d iffe ren t fie lds of a p p lie d p h y sic a l science.

T his in te re s t h as n o t d im in ish ed w ith th e lap se of t im e a n d a n ew e d it io n h as been long overdue. T he new vo lum e now a p p e a rs a s a re v ise d v e rs io n , w i th th e chap ter on P o ro s ity an d P e rm e a b ility e n tire ly re w r it te n , th e b ib lio g ra p h y b ro u g h t up to d a te , a n d w ith c e r ta in a lte ra t io n s in th e o rig in a l t e x t .

T hese changes h av e in creased th e size of th e b o o k b y 27 p a g e s w h ile i t s form er price has, mirabile dictu /, b een red u ced b y 20% .

A s exp la ined in th e p reface to th e f irs t e d it io n , th e p u rp o se of th e boo k is to b ring to g e th e r in one vo lum e th e b e s t of th o se in v e s tig a tiv e m e th o d s w hich have been developed in a v a r ie ty of specia lized fie lds, so t h a t th e y co u ld be of common in te re s t a n d u t i l i ty to th e w o rk ers in a ll of th e m .

I n th e in tro d u c to ry c h a p te r th e re is g iv e n a l i s t o f se v e n te e n ty p ic a l substances w hich th e a u th o r considers cou ld be p ro fita b ly e x a m in e d o r te s te d b y th e physical m eth o d s w hich he describes.

T his lis t includes su ch d iv e rg en t m a te r ia ls a s sa n d s fo r m o u ld in g a n d glass- m aking , ceram ic raw m a te r ia ls a n d p ro d u c ts , P o r t la n d c e m e n t, b u ild in g stones, oil an d w a te r san d s, c ry s ta llin e chem ical c o m p o u n d s a n d ro ta ry d rillin g m uds.

T he phy sica l a t t r ib u te s of th o se m a te r ia ls w ith w h ich th e a u th o r p roposes to deal are only seven in n u m b e r a n d six of th e se a re d e a l t w i th in th e f irs t tw o chap ters an d include th e size a n d sh a p e of g ra in s , co n s id e red u n d e r th e g en e ra l h ead in g of Size A nalysis, p o ro sity a n d p e rm e a b ility (d e a lt w ith in a s e p a ra te c h a p te r) , density a n d s ta te of agg regation .

The la s t su b je c t is how ev er o n ly co n s id ered v e ry b rie fly a t th e e n d of th e chap ter on size analysis, th e tw o pag es in v o lv ed b e in g co nfined to p r a c tic a l m e th o d s for p reparin g specim ens for su b se q u e n t e x a m in a tio n b y t r a n s m i t te d o r reflec ted light. A ctually th is a p p ea rs to be m isp lace d in th e b o o k , since th e r e is a la te r chap ter en tire ly d ev o ted to th e p re p a ra tio n of specim ens.

iter

I

BOOK REV IEW S. 113 A

T he s e v e n th a t t r ib u te , w h ich th e a u th o r re fe rs to a s “ m in e ra l c o n te n t , c o n ­s t i tu te s th e la rg e r p o r t io n o f th e w hole v o lu m e, a n d d ea ls w ith th e id e n tif ic a tio n of th e m in e ra ls lik e ly to be fo u n d in th e co m m o n er ro ck f ra g m e n ts .

T he c h a p te rs o n size a n a ly s is , p o ro s i ty a n d p e rm e a b il i ty a re a d m ira b ly p u t to g e th e r a n d p ro v id e a ll th e e s se n tia l in fo rm a tio n a n d d e s c rip tiv e d e ta i l r e la t iv e to c u rre n t m e th o d s w ith o u t u n n ecessa ry v e rb iag e . T h e se c tio n o n w a te r c lassifica­t io n m ig h t w ith a d v a n ta g e h a v e b een e x te n d e d so a s to in c lu d e e lu tr ia t io n m e th o d s w hich th e a u th o r d ism isses in fa v o u r o f th e m e th o d of u n d is tu rb e d s e ttl in g . T h is om ission is re m a rk a b le in v iew o f th e f a c t t h a t e lu t r ia t io n h a s com e to th e fo re w ith in re c e n t y e a rs fo r large-scale c la ssifica tio n in m a n y o f th e in d u s tr ie s sp ecifica lly m en tio n ed b y th e a u th o r .

T he A .P .I . t e n ta t iv e m e th o d fo r th e d e te rm in a t io n o f p e rm e a b il i ty is q u o te d a n d d iscussed a t som e le n g th (five pag es), th e a u th o r p o in tin g o u t a t th e conclusion th a t th e n ecessa ry a p p a ra tu s is ex p en s iv e to c o n s tru c t b ecau se o f th e n e c e s s ity for a n a ir com presso r a n d su b s id ia ry a p p lian ces . H e th e n m e n tio n s t h a t m e th o d s hav e been d ev e lo p ed in m o d e m fo u n d ry p ra c tic e fo r th e d e te rm in a tio n o f p e r ­m e a b ility to a ir o f m o u ld in g sa n d s , a n d re fe rs to a p o r ta b le a n d in e x p e n s iv e in s t r u ­m en t of th is ty p e d es ig n ed b y h im se lf. U n fo r tu n a te ly h e d is a p p o in ts th e r e a d e r b y o m itt in g a d e sc rip tio n of th is in s tru m e n t.

T he th re e c h a p te rs d e a lin g w ith th e id e n tif ic a tio n of m in e ra ls p ro v id e a n ex ce llen t exam ple of in te llig e n t om iss io n o f n o n -e sse n tia l d e ta ils . W ith in th e ra n g e of n in e ty -o n e p ag es th e re is com p ressed as m u c h p ra c tic a l a n d e s se n tia l in fo rm a tio n as w ou ld com m o n ly be fo u n d in a b o o k of tw ic e i t s size.

T he se p a ra tio n o f m in e ra ls b y p a n n in g , h e a v y l iq u id c o n c e n tra t io n a n d b y m agnetic , e le c tro s ta tic a n d d ie le c tr ic m e th o d s , o ccup ies th i r te e n p a g e s a n d is accom pan ied b y u se fu l ta b le s a n d i l lu s tr a t io n s , som e o f w h ich p ro v id e in fo rm a tio n w hich is n o t a v a ila b le in e v e n re c e n t e d it io n s o f c o n te m p o ra ry te x t-b o o k s o n p e tro lo g ica l m e th o d s.

T he sec tio n s d e v o te d to o p tic a l p ro p e r t ie s a n d th e p ra c t ic a l u se o f th e p e tro - g raph ie m icroscope a re e s se n tia lly p ra c tic a l a n d su ffic ien tly c o m p reh en siv e fo r th e pu rp o ses se t f o r th a s w ith in th e ran g e of th e b o o k . A se c tio n o n th e c o n s tru c ­t io n a n d in te rp re ta t io n of o r ie n ta tio n -c le a v a g e d ia g ra m s is a u se fu l f e a tu re w h ic h is i l lu s tr a te d b y tw o pag es of c lear i l lu s tr a t io n s .

R e cen t im p ro v e m e n ts in sp e c tro g ra p h ic in s tru m e n ts h a v e e n a b le d a n in c re a s in g n u m b e r of ch em is ts , m in e ra lo g is ts a n d m e ta llu rg is ts to u t i l iz e th e sp e c tro sco p e as a n a c c u ra te a n d co n v e n ie n t in s tru m e n t fo r a n a ly tic a l p u rp o ses . T h e in c lu s io n of a sh o r t acc o u n t of th e m o d e m m e th o d s of sp e c tro g ra p h ic e x a m in a tio n in th e boo k is th e re fo re a p p o s ite a n d th e ir a p p lic a tio n fo r th e id e n tif ic a tio n of s u c h m in e ra ls as b ery l, c a s s ite r ite o r t i ta n ife ro u s m a g n e ti te is d esc rib ed b y th e a u th o r .

T o su m u p , th is is a b o o k w h ich ca n b e h ig h ly reco m m en d ed as a co n c ise a n d e m in e n tly p ra c tic a l h a n d b o o k w h ich sh o u ld p ro v e of g re a t in te re s t a n d u t i l i ty to p e tro le u m te c h n o lo g is ts in m a n y b ra n c h e s of th e ir p ro fessio n .

T he p r in tin g , i l lu s tr a t in g a n d b in d in g of th e v o lu m e a re b e y o n d re p ro a c h w ith p e rh a p s th e so li ta ry e x c e p tio n of th e la s t ta b le w h ich is in su c h sm a ll p r in t t h a t i t is p ra c tic a lly use less a s a m ed iu m fo r re a d y re fe rence . T h is d ra w b a c k is f u r th e r in tensified b y th e u se o f n u m ero u s a b b re v ia tio n s , th e k e y to w h ic h is in s t i l l sm a lle r P r ’n t - J . M c C o n n e l l S a n d e r s .

Alcool Motor e Motores a Explosao. B y E . S ab ino d e O liv e ira . P p . 356. I n s t i tu teN ac io n a l de T echno log ia , R io d e Ja n e iro . 1937.

P u b lish ed u n d e r th e ausp ices of th e B ra z ilia n M in is try of L a b o u r, I n d u s t r y and C om m erce, th is boo k su m m arizes th e re su lts of som e s ix y e a rs of s tu d y , carried o u t in th e lab o ra to rie s of th e S an P a u lo P o ly te c h n ic School, a n d th e N a tio n a l ie c h n o lo g ic a l I n s t i t u t e a t R io d e Ja n e iro .

T he o b je c t o f th ese s tu d ie s and th e e v e n ts w h ich p reced ed th e m a re ex p la in ed m a p reface , o ccu p y in g eleven pages, b y D r. F o n sec a C osta , M in is te r o f A g ricu ltu re to th e S an P a u lo g o v e rn m en t, to w hose in i t ia t iv e is la rg e ly d u e th e in te re s t ta k e n b y th e B ra z ilia n G o v ern m en t in th e a t te m p ts to fin d a s u b s t i tu te fo r gasoline a s a fue l fo r in te rn a l c o m b u s tio n eng ines.

I t is p o in te d o u t t h a t th e co n c lu s io n of th e G re a t W a r le f t in m a n y n a t io n s a deep sense o f th e i r depen d en ce u p o n e x te r io r so u rces fo r th e i r v i t a l necessities

1 14 a BOOK REVIEW S.

in m o to r fuel. F ra n ce is c re d ite d w ith th e in i t ia t io n of a se a rc h fo r a pe tro leum su b s titu te w ith in h e r ow n fro n tie rs , a n d h e r e x a m p le w as so o n fo llow ed b y G erm any, I ta ly , Sw eden a n d o th e r co u n tries .

I n B razil, w here p e tro leu m h a d n o t y e t b een d isc o v e red m com m ercia l q u an tities , g rea t in te re s t w as d isp lay e d in th e F re n c h p ro p o sa l to u t i l iz e a lcoho l, a lth o u g h a pow erful opposition cam p aig n w as b e in g w ag ed , a lleg in g t h a t th e ac id s resu lting from incom plete co m b u s tio n a t ta c k e d a n d c o rro d e d th e m o to rs , w hile i t s low calorific v a lu e m ad e i t a p o o r s u b s t i tu te fo r g aso line .

As fa r b ack as 1923 e x p e r im e n ts to p ro v e o r d isp ro v e th e se a lleg a tio n s were m ade, a n d in A u g u s t of t h a t y e a r a B ra z ilia n en g in ee r , M r. H e ra ld o de Souza M attos, carried o u t a p ra c tic a l d e m o n s tra t io n b e fo re th e B ra z ilia n A utom obile Club, using a F o rd ca r fuelled ex c lu s iv e ly w ith a lco h o l, w h ich successfu lly com­p le ted a n a llo tte d course of 230 k ilo m e tre s .

I t w as n o t u n ti l 1931 th a t a g o v e rn m e n ta l d ecree e n a c te d t h a t im p o rta tio n s of gasoline in to th e co u n try sh o u ld h a v e a t le a s t 5 % of n a t io n a lly p ro d u c e d alcohol ad ded p rio r to c o n su m p tio n , th is d ecree p e r m i tte d th e u se o f 96% s p ir it , since a n h y d ro u s alcohol w as n o t th e n a v a ila b le .

T he m an u fac tu re of d e h y d ra te d s p ir i t , a s w ell a s a ca re fu l s tu d y of m a n y technical p rob lem s invo lved in i ts a p p lic a tio n , c o n s t i tu te d a su b je c t fo r in v e s tig a tio n b y the In s t i tu te of S ugar a n d A lcohol, c re a te d in 1933 to c o -o rd in a te th e w ork of the C om m ission fo r th e D efence of S u g ar P ro d u c tio n w i th t h a t of a C om m ission for th e S tu d y of A lcohol M otors, c re a te d in 1931.

T he ex p erim en ts m ad e in 1923 h a d b een c a rr ie d o u t in th e E x p e r im e n ta l S ta tion of F u e l a n d M ining, a n d th is e s ta b lis h m e n t w as in 1933 re -n a m e d th e N ational T echnological I n s t i tu te a n d , a c tin g u n d e r th e a u sp ices of th e I n s t i tu te of Sugar an d A lcohol, p ro ceeded to s tu d y th e a p p lic a t io n of a n h y d ro u s a lcoho l as a m otor fuel.

I t w as in th is con n ec tio n t h a t th e a u th o r o f th e p re s e n t w o rk ca rr ie d o u t the in v es tig a tio n s w hich fo rm i t s su b je c t m a t te r .

I n th e in tro d u c tio n to h is b o o k M r. S ab in o d e O liv e ira p a y s h an d so m e tr ib u te to th e w ork of th e B r it is h E m p ire M o to r F u e ls C o m m itte e , w hose rep o rts he characterizes as “ jew els of c la r i ty a n d p re c is io n ,” e v e n a ffirm in g th a t previous to th e ir p u b lic a tio n ab so lu te con fusion m a rk e d th e s t a te of kn o w led g e re la tiv e to th e alcohol m o to r. H e also refe rs w ith h ig h a p p re c ia t io n to th e m a s te rly w ork of R ica rd o in E n g la n d , of D u m an o is in F ra n c e a n d C alcav ecch ia in C uba.

T he w ork is d iv id e d in to tw o p a r t s ; th e f irs t, o c c u p y in g 242 p ag es is purely tech n ica l w hile th e second is sa id to be a f ifty -s ix -p ag e risum i o f th e la s t chapters of th e firs t ren d e red in lan g u ag e in te llig ib le to th e n o n -te c h n ic a l read e r.

T he firs t tw o c h a p te rs dea l so m e w h a t sk e tc h ily w ith th e rm o d y n a m ic principles, th e gas law s a n d m o lecu lar d isso c ia tio n . I n th e th i r d c h a p te r , w h ich ex ten d s to n in e ty pages, C a rb u ra tio n a n d C a rb u re tto rs a re co n s id e red in g re a t d e ta il an d in a som ew hat o rig ina l m an n er. T h e h u m o ro u s i l lu s tr a t io n s o n p ag es 58 an d 59 seem a l i t t le o u t of p lace in th is p a r t of th e b o o k in v iew of th e e x p la n a to ry note on page 25.

T he succeeding c h a p te r o n D e to n a tio n g ives a n e x ce llen t su m m a ry of th e various fac to rs w hich affect th is p h en o m en o n , a l th o u g h th e g r e a te r p a r t o f th e d a ta presented is of a generalized n a tu re , cu lled fro m th e p u b lic a tio n s o f w ell-k n o w n au tho rities , w ith l i t t le reference to alcohol.

T here is a n in te re s tin g re m a rk on p ag e 168 to th e effec t t h a t p r io r to 1930 a com pression ra tio of 5-1 : 1 w as co n sid ered a m a x im u m fo r gaso lin e engines, and t h a t th e B raz ilian G o v ern m en t h a d in c lu d e d in o ne o f th e i r d ecrees a c lause which w as favourab le to th e use of a u to m o b ile s e q u ip p e d w ith en g in es h a v in g a com­pression ra tio g re a te r th a n 6 : 1 . T h is decree w as b a se d o n th e id e a t h a t th e owners of such engines w ould be u n ab le to u tiliz e g aso line a lone , a n d w o u ld th u s b e influenced to ad o p t b lends of th e sam e w ith alcohol.

I n C h ap te r A th e p o ss ib ility of co rros ion in a lco h o l m o to rs is d e a lt w ith , as well as th e alleged te n d e n c y of su ch fuel to d e n u d e th e c y lin d e r w alls o f lu b r ic a n t.

T he a u th o r concludes, b o th o n th e rm o c h e m ic a l g ro u n d s a n d as a re s u lt of actual experim en ts, t h a t th e fo rm a tio n of a c e tic a c id is im p o ss ib le , a l th o u g h h e adm its * nn,e° rr0S*Ve 6®®c s may b e tra c e a b le to o th e r cau ses th a n th e u se of alcohol.

I h e s ix th a n d se v e n th c h a p te rs d iscu ss re sp e c tiv e ly th e s o lu b ility of gasoline

in a lco h o l and. th e b e h a v io u r of in fla m m ab le liq u id s in th e ex p lo s io n ty p e of m o to r. T h e f in a l c h a p te r o f th is p a r t of th e b o o k d ea ls w ith th e n ecessa ry a d ju s tm e n ts w h ich sh o u ld be m a d e to a n en g in e in o rd e r t h a t i t m a y fu n c tio n e ffic ien tly w ith gaso line—alcoho l b len d s . T h is c h a p te r is o f p a r t ic u la r in te re s t , since i t g ives th e d e ta ils o f a c tu a l e x p e r im e n ts c a r r ie d o u t w ith v a r io u s ty p e s of c a rb u re t to r a n d of au to m o b ile eng ines, as w ell as c o m p a ra tiv e re su lts w i th s tr a ig h t gaso line a n d b len d s o f th e la t t e r w ith alcohol.

T h e second p a r t of th e b o o k scarce ly fulfils th e s p ir i t o f th e n o te w h ich h e a d s i t s f irs t c h a p te r , in t h a t o n ly a sm a ll p a r t of i t is d e v o te d to a rtsum i o f th e f irs t p a r t , a n d m o reo v e r i t s p ag es a re re p le te w ith cu rv es w h ich m ig h t q u ite eas ily p ro v e fo rm id ab le to th e la y re a d e r .

T h e la t t e r w ill h o w ev er a p p re c ia te th e p ra c tic a l issu e in v o lv ed in th e fo rm u la tio n of th re e ty p e s o f gaso lin e—alcoho l b le n d s , a n d th e in fo rm a tio n g iv en in e ach case re la tiv e to i t s c h a ra c te r is tic effect u p o n th e m o to r a s re g a rd s , pow er, co n su m p tio n , acce le ra tio n a n d k n o ck te n d e n c y .

I n c o n c lu s io n i t m a y b e s t a t e d t h a t t h e b o o k c o n s t i t u t e s a w e l l r e a s o n e d a n d a d e q u a t e s e r ie s o f a r g u m e n t s i n f a v o u r o f t h e a d o p t i o n o f a l c o h o l a s a n i n g r e d i e n t f o r m o t o r f u e l s , a n d a s s u c h s h o u l d b e a c c e p t a b l e a n d i n s t r u c t i v e t o r e a d e r s i n i t s c o u n t r y o f o r i g i n . T h e w o r k i s w e l l p r i n t e d a n d t h e i l l u s t r a t i o n s , w h ic h a r e n u m e r o u s , a r e e x c e l l e n t . J . M c C o n t t e l l S a h d e k s .

Statistical Y ear-Book of the W orld Power Conference. Number 3. D ata on Resources and A nnual Statistics for 1935 and 1936. E d ite d , w i th a n In tro d u c t io n a n dE x p la n a to ry T e x t, b y F re d e ric k B ro w n , B .S c . (E con .). P p . 138. T h e C e n tra lOffice, W o rld P o w er C onference, 36, K in g sw ay , L o n d o n , W .C . 2. 1938.P ric e 20s. n e t .

T h e p r im a ry o b je c t o f th e b o o k u n d e r rev iew is to p u b lish o n ly c o m p arab le in te rn a tio n a l s ta t is t ic s of p o w er resou rces , d ev e lo p m en t a n d u t i l iz a t io n in co n fo rm ity w ith th e a im a n d d e fin itio n s of th e W o rld P o w er C onference.

T h e b o o k is d iv id e d in to five se c tio n s . T h e f irs t se c tio n g ives a c lea r s ta te m e n t of th e scope a n d m e a n in g of th e s ta tis t ic s a n d in c lu d es th e d efin itio n s a d o p te d b y th e W o rld P o w er C onference. S ec tio n 3c in c lu d es th e p e tro le u m s ta tis t ic s , a n d se c tio n 4 /, n a tu r a l gas.

T h e re sp o n s ib ility fo r e n su rin g t h a t th e s ta t is t ic s p re se n te d a re a c c u ra te , a n d co n fo rm to th e d e fin itio n s a d o p te d b y th e W o rld P o w e r C onference, re s ts u p o n n a t io n a l co m m itte e s , g o v e rn m e n t d e p a r tm e n ts a n d o th e r o rg an iza tio n s w hich su p p lie d th e m . T h e p a r t ic u la r so u rces of in fo rm a tio n a re m e n tio n e d a t th e e n d of e a c h ta b le , in cop ious n o te s .

T h e b o o k c o n ta in s s ta tis t ic s on so lid , liq u id a n d gaseous fuels, w a te r p o w er a n d e le c tr ic ity . T h e l iq u id fuels in c lu d e petroleum, benzo les a n d a lc o h o ls ; gaseous fuels in c lu d e natural gas a n d m a n u fa c tu re d gas.

T h e re a re m o re th a n f ifty ta b le s in th is vo lu m e, a ll of w h ich a re se lf-ex p lan a to ry .T h e s ta t is t ic s of p e tro le u m reso u rces a n d a lso th e a n n u a l re fin ery s ta tis t ic s

p re se n te d in T ab les 9 a n d 10, a re , u n fo r tu n a te ly , in co m p le te . A c e r ta in a m o u n t of a d d it io n a l in fo rm a tio n co u ld h a v e b e e n in se r te d in th e ta b le s , h a d i t n o t b een ex p ressed in u n i ts d iffe ren t f ro m th o se a d o p te d b y th e W o rld P o w er C onference, fo r ex am p le , th e s ta t is t ic s fo r refined p e tro leu m .

T ab le 14 g ives th e w o rld ’s n a tu r a l gas resou rces a n d p ro d u c tio n in m illio n s of cu b ic m e tre s . I t in c lu d es th e n u m b e r o f d r y a n d w e t w ells d r ille d a n d th o se p ro d u c in g n a tu r a l gas.

T ab le 15 g ives th e to ta l p ro d u c tio n o f gas a c c o u n te d fo r, th e q u a n t i ty t r e a te d fo r th e re c o v e ry of n a tu r a l g aso line , th e m a n u fa c tu re of c a rb o n b la c k a n d fo r in d u s tr ia l a n d h o u se h o ld uses.

I t is in te re s t in g to n o te t h a t R o u m a n ia b e g a n p ro d u c in g ca rb o n b la c k in 1935 on a sm a ll scale , u s in g o ne m illio n cu b ic m e tre s of n a tu r a l gas fo r th is p u rp o se , w h ich in c re a se d to 18-2 m illio n cu . m . in 1936. I n n o o th e r c o u n try in th e w o rld is c a rb o n b la c k re c o rd e d as b e in g m a n u fa c tu re d f ro m n a tu r a l g as o th e r th a n U .S .A . w h ere m o re th a n 8000 m illio n cu . m . w ere u sed fo r th is p u rp o se in 1936.

M r. F re d e r ic k B ro w n o f th e L o n d o n S chool of E co n o m ics is to b e c o n g ra tu la te d fo r th e c le a r a n d th o ro u g h ly p ra c tic a l w a y in w h ich h e h a s p re se n te d th e re su lts of th is s u rv e y of th e W o rld P o w e r R eso u rces. I t is th ro u g h n o f a u l t of h is t h a t

116 ABOOK RECEIV ED.

th e p etro leum s ta tis tic s a re so sc ra p p y a n d in co m p le te . T h e N a tio n a l C om m ittees of th e W orld Pow er C onference a re w h o lly re sp o n s ib le fo r th is . T h e y sh o u ld see th a t th e a d d itio n a l in fo rm a tio n re la t in g to p e tro le u m is fo rth c o m in g in con fo rm ity w ith th e defin itions la id dow n for th e se in te rn a t io n a l s ta t i s t ic s fo r th e n ex t S ta tis tica l Y ear-B ook. w - C a d m a n .

Proceedings of the R ubber Technology C onference, 1938. P p . 1137. W . Heffer,C am bridge. P rice £2 2s.In view of th e m ag n itu d e a n d w o rld -w ide im p o rta n c e o f th e ru b b e r in d u s try it

is som ew hat su rp ris in g th a t no in te rn a tio n a l co n ference o n th is su b je c t h a d been staged u n ti l la s t year. I n 1938, how ev er, th e C ounc il of th e I n s t i tu t io n of the R u b b e r In d u s try w ere ab le , a s a re su lt of o v e r a y e a r of a c tiv e p re p a ra tio n , to organize in L ondon a conference of th e firs t m a g n itu d e w h ich m e t w ith a m agnificent response a n d w as an u n qualified success.

T he Proceed ings of th is C onference h a v e n ow b een p u b lish e d in a volum e of 1137 pages co n ta in in g 103 se p a ra te p a p e rs a n d th e c o n se q u e n t d iscussions, as well as a b rief re p o rt on th e social fu n c tio n s a s so c ia te d w ith th e C onference.

T he scope of th e Conference in c lu d ed th e w hole ra n g e o f ru b b e r techno logy , so th a t th e p apers cover an e x tra o rd in a r i ly w id e v a r ie ty of su b je c ts . Som e of these, of course, a re of specialized in te re s t, a s th e y d e a l w ith su c h to p ic s as th e p roduction an d te s tin g of la te x an d raw ru b b e r a n d of v a r io u s m a n u fa c tu re d a rtic le s , such as ty res, e lastic te x tile s , ru b b erize d fab rics a n d cab les. M an y , how ev er, w hile referring in th e m ain to ru b b e r a n d i t s p ro d u c ts , a re of m o re g en e ra l scope a s w ell, an d m ay be read w ith in te re s t a n d a d v a n ta g e b y c h e m is ts a n d p h y s ic is ts engaged in o ther lines of w ork. P ap e rs on sy n th e tic ru b b e r-lik e m a te r ia ls su c h a s N eoprene an d the B unas, on carbon b lacks, on th e a b s o rp tio n of o x y g en b y ru b b e r , th e u se of inh ib ito rs and a n ti-o x id a n ts a n d th e d ev e lo p m en t a n d u t i l i ty o f a c c e le ra te d ageing te s ts are only a few exam ples of c o n tr ib u tio n s to k n o w ledge w h ich h a v e specia l app licab ility to th e p e tro leu m in d u s try . F u rth e rm o re , th o se c o n tr ib u tio n s in w h ich physical m ethods such a s X -ray s , fluorescence, co llo id e x a m in a tio n a n d v a rio u s form s of m echanical te s tin g are d escribed , c o n ta in m a n y f ru i t fu l su g g e stio n s for physical chem ists an d for engineers, a n d a re w ell w o r th p e ru sa l.

A n excellen t in d ex , p le n tifu l d ia g ra m s a n d a n e x c e p tio n a lly goo d q u a li ty of paper an d b in d in g m ake th e vo lum e a v a lu a b le a c q u is i tio n b o th fo r re a d in g and for reference. F . B . T h o l e .

BOOK RECEIVED.Standard Specifications for Benzole and Allied Products. S econd E d itio n . 1938.

P p . 197. N atio n a l B enzole A sso c ia tio n , W e llin g to n H o u se , B u c k in g h a m Gate,S.W . 1. P rice Is. 6d.

W ith th e increasing in te re s t in th e p ro d u c tio n of b en zo le fro m coal th e new ed itio n of “ S ta n d a rd S pecifications fo r B enzo le a n d A llied P ro d u c ts ,” ju s t pub­lished b y th e N a tio n a l B enzole A sso c ia tio n , p ro v id e s co m p reh en siv e d a ta m ost necessary to th o se w hose busin ess is co n cern ed in a n y w ay w ith th e p ro d u c tio n and use of benzole an d i t s a llied p ro d u c ts .

The seven teen s ta n d a rd sp e cifica tions n ow issu e d , wTh ich co v e r v a rio u s grades of benzole, to luo le, xylo le, coal t a r so lv e n t n a p h th a a n d co a l t a r h e a v y naph th a , a re th e resu lt of a th o ro u g h rev isio n of a ll sp e c ifica tio n s b y a C o m m ittee of the A ssociation rep re sen ta tiv e of th e B enzole In d u s try . T h is c o m m itte e , in co llabora­t io n w ith th e B ritish S ta n d a rd s I n s t i tu t io n , h a s ta k e n in to co n s id e ra tio n the requ irem en ts of b o th th e m an u fa c tu re rs a n d u se rs of th e v a r io u s p ro d u c ts .

In ad d itio n to th e d e ta ile d sp ecifica tions , fu ll in fo rm a tio n as to s tandard ized m ethods of te s t a n d schedules of a p p a ra tu s a re g iv en . T h e m e th o d s of te s tin g and descrip tions of a p p a ra tu s a re th o se reco m m en d ed b y th e S ta n d a rd iz a tio n of Tar P ro d u c ts le s ts C om m ittee issued in i t s re c e n t p u b lic a t io n “ S ta n d a rd M ethods for te s tin g T a r an d I t s P ro d u c ts .”

IN ST IT U T E NOTES.M a r c h 1 9 3 9 .

FORTHCOMING MEETINGS.T h u rs d a y , 2 0 th A p ril, 1939, a t 5 .30 p .m . a t th e R o y a l S o c ie ty of A r ts , J o h n

S tre e t , L o n d o n , W .C . 2. A nnual Generäl Meeting.

SUMMER MEETING.The Summer Meeting of the Institute will be held at Birmingham

from May 22nd-24th, 1939, under the Presidency of Professor A. W. Nash, M.Sc., M.I.Mech.E. The objects of the meeting are to review and interpret recent work on fuels and lubricants for use in internal combustion engines.

The programme of the meeting, together with details regarding Ladies’ Visits, Registration, etc., are given in the circular sent out separately to members.

A Summary of the programme is given below :Monday, May 22nd.

Evening. Informal Reception by the President a t the Grand Hotel, Birmingham.

Tuesday, May 23rd.Morning. Technical Session.—Knock-Rating.Afternoon. Technical Session.—Lubrication.Evening. Reception in the Grand Hall, The University,

Edgbaston.Wednesday, May 24th.

Morning. Technical Session.—Fuels for Compression— Ignition Engines; Lubrication.

Afternoon. Visits to Austin Motor Company and Morris Commercial Cars Ltd.

Evening. Dinner and Dance a t the Grand Hotel.

S t u d e n t s ’ S e c t i o n ( L o n d o n B r a n c h ).

T u e sd a y , 4 th A p ril, 1939, a t 6.15 p .m . a t th e S ir J o h n C ass T ech n ica l I n s t i tu te , J e w ry S tre e t, A ld g a te , L o n d o n , E .C . 3. A n n u a l O p en M eeting . “ H igh Speed E n gin es,” b y H . R . R ic a rd o , F .R .S .

W e d n e sd a y , 1 9 th A p ril, 1939, a t 5.45 p .m . a t th e Offices of th e In s t i tu te , T h e A d e lp h i, L o n d o n , W .C . 2. “ The Oilfields of Iraq,” b y X . P a c h a c h i.

TRANSFERS TO NEW CLASSES OF MEMBERSHIP.The Temporary Regulations relating to the transfer of the existing

members to the new classes of membership, as set out in the leaflet sent to all members of the Institute, were approved at the Special General Meeting held on 10th January, 1939.

Members who wish to transfer to one of the new classes of member­ship are requested to submit their applications as early as possible on the forms provided for the purpose at the back of the leaflet con­vening the Special General Meeting.

STUDENTS’ MEDAL AND PRIZE.The Council has decided that the Students’ Medal and Prize in

1939 will be awarded for a thesis on a set subject, and not for theses on subjects chosen by the candidates themselves.

A short list of alternative subjects on which theses are invited will be issued by the Council to all Students of the Institute after 30th June, 1939.

^ i n s t i t u t e n o t e s .

CANDIDATES FOR ADMISSION.

The following have applied for admission to the Institute or transfer to another grade of membership, and in accordance with the By-laws the proposals will not be considered until the lapse of at least one month subsequent to the issue of this Journal, during which any Member or Associate Member may communicate by letter to the Secretary, for the confidential information of the Council, any particulars he may possess respecting the qualifications or suitability of any candidate.

The object of this information is to assist the Council in grading candidates according to the class of membership.

The names of the candidate’s proposer and seconder are given in parentheses.A r m s t r o n g , V alen tine , S tu d e n t, 27, K e n s in g to n G a rd e n s S q u a re , L ondon ,

W . 2. (A. C. Egerton ; V. C. Illing.)B r a d d i c k , H e rb e r t J a m e s W i l l i a m , E n g in e e r (Anglo-lranian Oil Co., Ltd.),

25, P ic k h u rs t M ead, H ay es , K e n t . {A. C. H a rtley ; A . E . Dunstan.) (Transfer to A . M .)

C l i f f o r d , Jo sep h , C hem ist, 5, A rlin g to n A v e n u e , S o u th S hore , B lackpool.(W. H. Cadman ; C. W. Wood.)

B r i g h t o n , R o b e rt M ill M cCom be, E n g in e e r (W illiam Briggs & Sons, Ltd.), 30, T hom son S tre e t, D un d ee , A ngus. (A. D . McLuckie ; W. F . Murray.)

C r o s s l e y , S tan ley , D irec to r (British Transformer Oil db Lubricants, Ltd.), “ S ou thv iew ,” O ld Forge Close, S ta n m o re , M id d x . (E. J . Dunstan ; B. J . Vavasour.)

F r a m e , A lexander, W orks M anager (Roman Bank Crude Oil Works), W oodside, D rum shoreland , B ro x b u rn , W . L o th ia n . (R. Crichton ; O. H . Smith.)

H a l l , Jo h n D esm ond , C hem ical E n g in e e r (Foster Wheeler, L td.), “ E arls- c ro ft,” 82, Cecil P a rk , P in n e r , M iddx . (A. W. Nash ; R . K . Fischer.)(Transfer to A . M.)

H a t t , E rn e s t R ich a rd , W o rk s M an ag er (Dussek Bitumen db Taroleum Ltd.), “ K in g a r th ,” L ongdon W ood, K e s to n , K e n t . (F . H. Oamer ; A . Osborn.)

H e l t o n , Jo h n S ydney , C hem ist (Belgrave Oil db Grease Co. (Leeds), Ltd.), “ B elgrave,” 19, S an d h ill O val, A lw oodley , L eeds. (F. Dakin.)

L e w i s , P h ilip C o ttre ll, M echanical E n g in e e r, T h e W h esso e F o u n d ry & E n g in ­eering Co., L td ., D a rlin g to n , Co. D u rh a m . (E. R . Cartwright; H. D. Demoulins.)

L u m b , E rling , C hem ist (Munster, Sim m s db Co., L td .), “ L a rv ik ,” 71, R u g b y A venue, B angor, Co. D ow n. (L. R. Phillips ; A . W. Nash.)

M a c A r t h u r , H ec to r, C hem ist ( It'. B. Dick i Co.), 38, R iv e rs id e R o ad , X ew lands, G lasgow , S. 3. (IF . M . Gum m ing; G. H . Smith.)

M a r t i n , B e rn a rd D av is , D irec to r (Germ Lubricants, L td.), C i t y G a t e H o u s e , F in sb u ry S quare, L on d o n , E .C . 2. (J. E. Southcombe.)

M a r t i n , Ja c k W illiam , E n g in e e r (Whessoe Foundry db Engineering Co., Ltd.), G reengates, C aledon R o a d , B eaconsfield , B u c k s. (E. R . Cartwright; H. D. Demoulins.)

M e t c a l f e , T hom as J o h n , C hem ist (E. Joy db Sons, L td .), 3 , G r a f t o n V illas, S tanks, Leeds. (G. V. Davies ; J . R. Smellie.)

M o r t i m e r , G eorge A b b o tt, C hem ist (Pumpherston Shale Oil Co.), 47, Q ueens A venue, B lackhall, E d in b u rg h . (R. Crichton ; G. H . Smith.)

N a k i b , M o h a m m a d A b d u l G hazi, E n g in e e r , G o v e rn m e n t Office O il M e asu re ­m e n t S ec tio n , K irk u k , I r a q . (A. IV. Nash ; L. V. W. Clark.) (Transfer to A . M . )

N e w e y , C lifford S am u el, E n g in e e r (Caribbean Petroleum Co., L td.), 49, D a r t ­m o u th S tre e t , W e s t B ro m w ich . (A. W. N a sh ; L . V. W . Clark.) (Transfer to A . M . )

N e w t o n , E rn e s t J o h n , C h em ist (Ernest Newton A Co.), 353, B irm ingham R o a d , W y ld e G reen , S u t to n Coldfield. (H. C. Tett ; A . Hamilton.)

O ’B r i e n , K e v in , D ire c to r (H. F. O'Brien A Co., L td.), “ T h o m o n d ,” B ro o k ­field A ven u e , T im p e rle y . (D . W . O’Brien ; J . Barrett.)

O ’M e a r a , T eren ce B a rry , C h e m ist, c /o C olas P ro d u c ts , L td . , N o rm a n H o u se , S tra n d , W .C . 2. (L. 6 . O abriel; J . 8 . Jackson.)

P r o c t e r , R e g in a ld H e n ry , E n g in e e r (English Drilling Equipment Co.), 65, E lg a r A venue, T o lw o rth , S u rrey . (F. E . Cherry ; A . J . Yokes.) (Transfer to A . M .)

R i g d e n , P e te r M o n tg o m ery , C h em ist (W ailes Dove Bitumastic, L td.), 12, V ic to ria S q u are , Je sm o n d , N ew ca stle -o n -T y n e . (R .Shaw ; H. C. Rampton.) (Transfer to A . M .)

S a i b , M o h am m ed A li, E n g in e e r, c /o M in is try of E co n o m ic s & C o m m u n ica tio n s , B a g h d ad , I r a q . (A. W. N a sh ; L. V. W. Clark.) (Transfer to A . M .)

T a y l o r , T h eo M allin son , C h em ist (Shell Marketing Co.), 25, M a th eso n R o a d , W est K en s in g to n , W . 14. (J. S. Jackson ; S . R . H ills.)

W h i t e , Colin M cL uckie , C h em ist (Pumpherston Oil Co.), 60, M a in S tre e t, W in ch b u rg h , W . L o th ia n (R . Crichton ; O. H. Smith.)

W i l d , E ric H e rb e r t , C h em ist (Anglo-American Oil Co., L td.), 9, R u s la n d R o a d , W ea ld s to n e , H a rro w . (H. C. Tett ; E . B . Evans.)

ARTHUR W. EASTLAKE, Honorary Secretary.

JOURNALS WANTED TO PURCHASE. The Institute is prepared to purchase copies of the following issues

of the Journal a t the price of 4s. 0d. each.No. 152, June 1936 „ 171, January 1938„ 174, April 1938

Journals should he forwarded to the Secretary, The Institute of Petroleum, The Adelphi, London, W.C.2. Only copies in good condition will be considered for purchase.

PERSONAL NOTES. Mr. M. A l l a m has returned to Egypt after a long v i s i t to the

Hedjaz.M r . B. D. C a u t h e r y is h o m e f r o m I r a n .Mr. E. K. D y k e s has left for Trinidad. Mr. J . C. J e w e l l has returned from Iran.M r . H. H . M a r t i n i s h o m e f r o m E c u a d o r .

Mr. C. A. S a n s o m is home from Burma.Correspondence or Journals forwarded to the following members

have been returned, and the Secretary would be pleased to receive any information regarding their present address : E. C. B r o w n , K. B u r t o n , M. C a p p e r , 0. C. E l v i n s , V. C. S . G e o r g e s c u , J. J . L. H a m i l t o n , J. R. H o r t h , A. D. J o n e s , J . L a n d e r , H . R. L o v e l y ,

I. L u s t y , F. M a c k l e y , A. M a c L e a n , G. P. M e l v i l l e , C. A. M o o n ,S . N i c o l , S . P a p p , R . G . R e i d , N . D. R o t h o n , H . G . S p e a r p o i n t ,and A. H. W i l l i a m s .

t a n k c a l i b r a t i o n ----------------------------------------------" B L U C H A L K " W A T E R - S E A R C H IN G C O M P O S IT I O N

and

"ULLAGE PASTE" FOR A C C U R A T EL Y C ALIBR ATIN G AND MEASURING PETROLEUM IN ST O R A G E TANKS

Used by Government Authorities and Principal Oil Companies throughout the world

Prices and Term s sent on application

Sole Patentees and Manufacturers :

THE BLUCHALK CO M PAN Y, W EST H EN D FO RD , Y E O V IL , ENG.

ANNUAL REVIEWS OF PETROLEUM TECHNOLOGY

Vol. 3 (covering 1937)

PRICE - 11s. post free

Members of the Institute and Annual Subscribers to the

Journal may purchase one copy only at 5s. 6d. post free.

Obtainable from :

THE IN STITUTE O F P ETR O LEU M THE A D ELPH I, LO N D O N , W .C .2

KNOW W H ERE THE BIT IS AT A LL TIMES BY USING

SINGLE SHOTSPreven t off le a s e d r illin g —or c h a n g e the cou rse of y o u r w e ll a n d d ire c t it to a n o th e r po in t u n d e rg ro u n d ! I t's poss ib le if y ou k eep a re c o rd of the b it a s the w ell is go in g dow n.The L ane-W ells s in g le sh o t su rv ey in stru m en t m e a su re s b o th in c lin a tio n an d d irec tion in o p en h o le d rillin g . Its u n e rrin g d e p e n d a b ility is fam o u s in oil fields th ro u g h o u t th e w o rld an d everyw here ex p e rien ced o p e ra to rs recom m end it. An i l lu s tra te d b u lle tin co n ta in s d e ta ile d in fo rm ation . W rite the L ane-W ells C o m p an y to d a y , 5610 South Soto S treet, Los A n g eles , C a lifo rn ia , or a n y b ra n c h office.

(Left)

LANE-WELLSS I N Q L E

S H O T

I N S T R U M E N T G EN ERA L O fT IC E S AND PLA N T

5610 S. Solo SL, Lo s A ngeles, CaliL

EX PO R T O r r iC E S420 Lexington Ave.. New York City. N. Y.

Kindly mention this Journal when communicating with Advertisers.

iv

HADFIELD'S1v\\Av\AAiUii/////////y^y vW W A A At 11 // // // / / / / / / / ,

t r a d e ^ ERA 131 ( m a r k & t r a d e HECLA 153 m a r k

^ / ' / y y r i T v v v v v v v ^ ^ w m v y n m n ^

S T E A M P I P E FLANGE BOLTS

HAVE A HIGH CREEP STRENGTHFor use at the highest temper­ature employed in modern steam practice. Do not become brittle as a result of operating conditions.

Steel Castings a n d Forgings of all Kinds.

iriSo*™ HADFIELDS LTD. “Kir"1------- East Hecla and Hecla Works, SHEFFIELD, Eng. -----

No. 1666.

SCHLUMBERGER ELECTRICAL CORINGF r a n c e .— Société de Prospection

E lectrique, 30 , rue Fabert, P a r is .

U .S .A .— Schlum berger W ell Surveying C orporation, 2720 Leeland, H o u s to n , Texas.

Local Offices : Long Beach, Oklahoma C ity, N ew York, C orpus Christi, B radford.

V en ezu e la .— P. Bayle, Villa Proselec, M a r a c a ib o .

T r in id a d , B .W .I.— Schlum berger E lec­trical C oring M ethods, S a n F e r n a n d o .

C o lo m b ia .— H . Rappart, P u e r to - B e r r io .

A rg e n tin e .— G . G uichardot, C o m o d o ro R iv a d a v ia , K m . 27 .

M o ro cco .— M . Texier, Société deProspection Electrique, PetITJEAN.

R u m a n ia .— A. Poirault, 18 Strada I. C . Bratianu, C a m p in a , (P rahova).

G e rm a n y .— Firm a von Flotow , Schil- lerstrasse 361, H a n o v e r .

D r. B. Paul, Kobenzlgasse 30, VlENNA.

H u n g a ry .—M . Scheibli, V adaszkürt T ü r Istvan U.5, B u d a p e s t .

I r a q .— L . Beaufort, Q a iy a r a h .

B r it is h In d ia .— A. Couret, D ig b o i, Assam .

B u r m a .— L . Bordât, K h o d a u n c .

N e th e r la n d E a s t In d ie s .— R . Sauvage, P la d jo e , Sum atra.

Schlum berger M ethods also applied in : U .S .S .R ., Japan, Italy, Poland, Yugoslavia,

Egypt and British N orth Borneo.

Kindly mention this Journal when communicating with Advertisers.

SVENSKA D IA M A N T B E R G B O R R N IN G S A K T IEB O L A G ET(The Swedish Diamond Rock Drilling C o .)

KUNGSGATAN 4 4 S T O C K H O LM SWEDENManufacturers of complete Core Drilling Outfits

system ( C r a e l i u S

for Structure Investigations in Oil Fields Drilling Contracts undertaken in all parts of the World

London R ep resen tative : T e lep h o n e : V ic to r ia 8988

REX LAMBERT, A.R.S.M., 25, Victoria Street, London, S.W.1

REPRESENTATION and P ET R O LE U M O PERA TIO N S

CHARLES DABELL AND COMPANY

" D A B ; ROS”

E N G IN E E R S A N D C O N S U L T A N T S

3 MIDAN SU A RES

C A IR O • EG YP T

LIST OF ADVERTISERS.AKTIEBOLAGET E L E K TR ISK M ALM LETNINGA l l S t e e l P r o d u c t s M f g . C o A s k a n ia - W e r k e , A .G .B a b c o c k & W i l c o x , L t d .B a k e r O i l T o o l s , I n c .B l u c h a l k C o ...........................W . C h r i s t i e & G r e y , L t d .A . F . C r a i g & C o . , L t d .C h a r l e s D a b e l l & C o D u k e & O c k e n d e n , L t d .E d e l e a n u G e s e l l s c h a f t m . b F o s t e r W h e e l e r , L t d .W . J . F r a s e r & Co., L t d .G e o p h y s i c a l P r o s p e c t i n g C o . , L t d .H a d f i e l d s , L t d .II AY WARD -T YLER & CO., LTD.I n s t i t u t i o n o f P e t r o l e u m T e c h n o l o g i s t s I n t e r n a t io n a l P a i n t & C o m p o s i t i o n s C o . , L t d . L a n e -W e l l s C o .L u f k i n R u l e C o .L u m m u s C o m p a n y

M e t r o p o l it a n V i c k e r s E l e c t r i c C o . , L t d .N a t i o n a l S u p p l y C o r p o r a t io nO i l C e n t e r T o o l C o ...........................................................O i l W e l l S u p p l y C o .S e c u r i t y E n g i n e e r i n g C o . , I n c ...............................S o c i é t é d e P r o s p e c t i o n É l e c t r i q u e S p e r r y -S u n W e l l S u r v e y i n g C o . . . .J o h n G . S t e i n & C o . , L t d .......................................... !S t e w a r t s a n d L l o y d s , L t d .SVENSKA DIAMANTBERGBORRNINGS A K TIEBO LAG ET J o h n T h o m p s o n ( W o l v e r h a m p t o n ), L t d . . . .T in t o m e t e r , L t d .

X U l

ivInside back cover

xviii vi

Inside back cover

X X

xii

xiv

xvuIX

Back cover vi xvi

Kindly mention this Journal when communicating with Advertisers.

v i

REOP

T R RKPAINTS = PETROLEUM

INDUSTRY“ DANBOLINE S IL V E R E T T E ” The super aluminium pa in t for

a ll refinery purposes. Only actual experience can prove its amazing durability.

“ TANCTECTOL” The only protective pa in t for theIN T E R IO R o f petroleum storage tanks. W ith stands perm an en t immersion in a ll petroleum frac­tions, benzole and salt or fresh w ater. U sed by the B ritish Admiralty, Royal A ir Force, and leading O il Companies.

W rite for free booklet “ PAINT IN THE OIL INDUSTRY.”

INTERNATIONAL PAINT & COM POSITIONS Co., Ltd.U .S . Enq u iries : 31-32 GrOSVenor Place. Main Factory21 W ES T S T ., ’ FELL IN G -O N -T Y N ENEW Y O R K . L O N D O N , S.W. I EN G LAN D

GEOPHYSICAL SURVEYSThe G.P.C. employs the most up-to-date

instruments and methods.

G R A V IM ETR IC M A G N ETIC E L E C T R I C S E I S M I C

Consult the G.P.C. regarding the method best suited to

your particular problems.

W O R LD W ID E EX PER IEN C EExtending over a period of 15 years

A T Y O U R SERVICE

THE GEOPHYSICAL PROSPECTING CO. LTD.Managing D irecto r: Ja s. C . T e m p le to n , B .S c ., F .G .S . , M .In s t .M .M ., M .In s t .P .T .

9-11 Copthall Avenue, London, E.C.2Telephone :

M E T R O P O L I T A N , 6363Telegrams and Cables :

G E O P R O S C O , L O N D O N

Kindly mention this Journal when communicating with Advertisers.

Just to be DOUBLY SURE)

TOP OFF YOUR WELL CONTROL

HOOK-UP WITH AN

O-C-T PREVENTER

T y p e *‘ D ” w i t h r e g u la r p a c k o f f , a v a ila b le f la n g e d o r s c r e w e d , w it h o r w it h o u t o u t le ts

Most operators now include an O-C-T Blowout Preventer a s " s ta n d a rd e q u ip m e n t" for well control hook-ups. While the O-C-T P reventer is often used independently of other b low out p rev en t­ers. it is m ore generally u sed in con junction with ram type preventers, a s show n ab ove, to p ro v id e a positive se condary seal a n d to ab so rb w ea r occasioned by ro tating an d ra ising a n d low ering pipe to p revent sticking while killing the well.An O-C-T Blowout Preventer, eq u ip p ed with regular packoff, perm its m ovem ent of drill p ipe between tool jo in ts : eq u ip p ed with s tripperrubber, it perm its p ipe to be m oved u p or dow n T y p e " D " w i t h s t r ip p e r ru b b e r , w h ic h isany desired d istance, the stripper ru b b e r read ily a v a i la b le for a l l t y p e s o-C-T P re v e n te rs ,

passing tool joints or p ipe collars.

F u ll p a rticu la rs ca rr ie d in o u r com plete ca ta lo g u e , a copy o f w h ic h w i l l be g la d ly sent on request.

c e n t e r t o q #

° 'v P S F r l1 ■ u

H O U STO N . T E X A S . U .S .A .CABkL AOORtSS 'O C £ A tT O l '

Kindly mention this Journal when communicating with Advertisers.v i i 1

SPECIALLY SUITABLE FOR OIL REFINERIEi

H E re lia b ility o f N E T T L E 4 2 /4 4 %

alum in a f ire b r ick s in high te m p e ra tu re

installation s is u n d isp u ted . U se rs o f th is

brand a re con vinced o f th is : th o se w h o

have th e ir re fracto ry p roblem s t ill w ith

th em m ay w ell find that a tr ia l o f N E T T L E

w ill p ro v id e a so lu tio n .

JOHN G. STEIN & CO. LTD., BONNYBRIDGE, SCOTLAND

Kindly mention this Journal when communicating with Advertisers.

# Pressure Saponification.

# Close scrap in g to give efficient heat transfer.

# Special type ag ita tio n to avoid stratification.

# Sound construction to perm it alternate h eatin g an d cooling.

# Large outlet valves flush with bottom to avoid pockets.

# A rrangem ent for com plete dis­charge.

# Simple and efficient m otor drive or belt drive.

% Speed control as g reases thicken.

W. J. FRASER & CO . , LTD.D A G E N H A M . . ESS EX

t a s / f z .

Kindly mention this Journal when communicating with Advertisers.

BABCOCKClüJs C W W ELD EDPRESSURE VESSELSCONSTRUCTED AT RENFREW

BOILERDRUMS

OTHERVESSELS

Total N um ber 261 273

G reatest Length 38' 4" 74' 0"

G reatest D iam eter 5' 6" 20' 0"

T h ickest Shell 3 I T •35"J8

Heaviest 36 Tons 40 Tons

Highest Pressure 1 1 24 lbs./sq. In. 1 495 lbs./sq .in.

The 400,000 Volt X-Ray Unit installed will penetrate 4£" thick steel plate.

EVERYTHING FOR THE BOILER H O U S E I N C L U D I N G VA LV E S

BABCOCK & WILCOX LTD.34 FARRINGDON STREET, LONDON. E.C.4

Kindly mention this Journal when communicating with Advertisers.

COMPLETED IN 20 WEEKS FROM

DATE OF CONTRACT

a 4 4 0 0 BARREL per day

BENZOL-KETONE D EW A X IN G PLANT

Benzol-Ketone Dewaxing plant w as signed with Sinclair Refining Com­

pany. On January 16th, 1939 —twenty w eeks to the d ay — despite snow

and the difficulties of winter w eather —the plant w as completed. » » » This

plant for Sinclair w as the eleventh Benzol-Ketone plant designed and built

by Lummus within three years. Four more Benzol-Ketone plants are now

on the drawing boards —three are additions to Lummus-built installations;

the fourth is a new unit for another major refiner.

A. C. GRONBECK

R eprin ting : THE LUMMUS COMPANYBush House, A ldw ych, London, W. C. 2

Kindly mention this Journal when communicating with Advertisers.

C A B L E T O O L O P E R A T O R S

MAKE A BULL’S EYE

O N EV ERY R U N W IT H A

B A K E R Cable ToolC O R E B A R R E L

B A K E R A ffo r d s T h e s e I m p o r ta n t A d v a n t a g e s :

H ig h e r p e r c e n ta g e o f r e c o v e r ie s in a w id er r a n g e o f fo r m a t io n s

F a s te r r u n n in g t im e

L o w e r m a in te n a n c e c o s t

S im p lic ity o f o p e r a t io n

M a x im u m sa fe ty in se r v ic e

L o n g e r l i fe

L o w in it ia l c o s t

C o m p l e t e d e t a i l s c o n c e r n in g t h is e c o n o m ic a l a n d e f f i c ie n t t o o l w i l l b e g l a d l y f u r n is h e d u p o n r e q u e s t — o r s e e y o u r 1 9 3 9

C o m p o s i t e C a t a l o g .

BAKER OIL TOOLS, INC.Telephone JEfferson 8211 - HUNTINGTON PARK, C AL IFO RN IA - 2959 E. Slauson Ave.Telephone WAyside 2 I0 3 -H O U S T O N PLANT A N D O F F IC E -6023 Navigation Blvd.

MID-CONTINENT OFFICE A N D W AREHOUSE:Telephone 2-8083—Tulsa. Oklahoma - 3 12 East Fourth Street

W EST TEXAS BRAN CH O FF IC E EXPORT SALES O FF IC E R O C K Y M O U N TAIN HEAD Q UARTERSO deiia , Te .a i-Telephone 2 17 Rm. 1914- 19 Rector St.. New York C ity Tel. 2230-Casper, Wyoming - Bo. 1464

Tel. Digby 4-SSIS

BAKER CABLE TOOL CORE BARREL'eating with Advertisers.

The en tire p u rp o se b e h in d th e d e v e lo p m e n t of "S e c u rc d o y " w a s to p erfec t a d r illa b le m e ta l for p e rm a n e n t s u b s u r fa c e in s ta l la t io n s . . . a m e ta l w ith h ig h e ro sio n a n d co rro s io n re s is tin g p ro p e r t ie s . . . a m e ta l w ith a m p le s tre n g th to b e u se d sa fe ly u n d e r a l l c o n d itio n s . . . a m e ta l th a t co u ld b e ru n in a w ell a n d left th e re th e s a m e a s s te e l, y e t c o u ld b e quick ly d rilled u p a n d c irc u la te d o u t of th e h o le w h e n e v e r n e c e s sa ry .

A nd y o u c a n do th e se th in g s w ith "S e c u ra lo y " !

It is e rosion a n d corrosion re s is ta n t, b e in g in fa c t m o re re s is ta n t th a n s tee l to m a n y com m on o ilfie ld co rro s iv es su c h a s su lp h u r a n d su lp h u r com pounds. It is s trong , b o th in c o lla p s e a n d te n s io n , a s th e s e s tre n g th fac to rs for th e p o p u la r 65/e" size "S e c u ra lo y " p ip e w ill sh o w :

“ S E C U R A L O Y " PIPE 6 5/8"

WEIGHTPER

FOOT

LENGTH OF STRING EQUIVALENT TO:COLLAPSE IN SALT WATER

(Saiety Factor 2)TENSILE STRENGTH

(Safety Factor 2«^)

10ir 6284' 7211'

A nd a b o v e a ll, "S e c u ra lo y " is d r illa b le . A n o rd in a ry d r ill in g b it w ill rem ove it a t th e r a te of 20 to 30 fe e t p e r h o u r a n d th e c u ttin g s c irc u la te com plete ly ou t of th e h o le a s tin y c h ip s s u s p e n d e d in th e f lu id s tream . T hat m e a n s re a l p ro tec tio n for y o u r o ilfie ld in v es tm en t!

★ BY EVERY E N G IN E E R IN G S T A N D A R D " S E C U R A L O Y " IS THE MATERIAL FOR P E RM ANE NT BOTTOM H OLE SERVICE!

SECURITY ENGINEERING CO., INC.WHITTIER, CALIFORNIA. PHONE 42004 7

M I D - C O N T I N E N T * 5 5 2 5 C L I N T O N D R I V E , H O U S T O N . T E X A S . P H O N E C A P I T O L 9 5 3 8 E X P O R T * S E C U R I T Y E N G I N E E R I N G C O . . I NC. , 4 2 0 L E X I N G T O N A V E N U E . N E W Y O R K C I TY

S ievesu Re&me/U * S ecw va h u f ★ S e c u su ty îb 'U lla h le ß n o d u c ti

Kindly mention this Journal when communicating with Advertisers.

Y1\7

Kindly mention this Journal when communicating with Advertisers.

x v

A LLSfm PRODUCTS MFG.CO.W i c h i t a , K a n s a s , U. S. A .

8 0 1 So. W i c h i t a S t ., P. O. D r a w e r 2 0 0 1C A B LE A D D RESS "A L L S T E E L " PH O N ES L D. 289, L O C A L 4-4381-4-4382

(Base M ap C opyrigh ted by R an d M cN ally)

A Model R Double Drum Draw Works with cable tool drilling attachment was received in Ecuador exactly 14 DAYS from the date order was placed.

" I am a d v is e d t h a t th e o th e r m achine o rd e re d by c a b le from •London ab o u t November 2 0 th , i s i n G uay aq u il. I canno t b e l ie v e i t , but- i f s o , i t i s s u r e ly a f i e l d r e c o rd f o r d e l iv e r y n e v e r even ap ­proached b e fo re " , q u o te s H -S tubbs, S u p e r in te n d e n t o f A ng lo -E cuadorian O i l f i e ld s , L td . When quick delivery and dependable performance is

: wanted; order "CARDWELL" draw works, servicing hoists or pipe line equipment. Guaranteed shipment on any standard model in five to eight days.

Kindly msntion this Journal when communicating with Advertisers.

C L A S S I

f u s i o n w e l d e d

B Y

THOMPSONtHa m p t o n ) ltd

WORKING PRESSURE OF

1 0 0 0 L B S . P E R SQ .IN -

Down toVthe bottom o f one of the deepèst w ells in Texas!

D a ily u se o f th e Syfo C l in o g r a p h p e r m its accu rate c o n tro l o f in ­c l in a tio n in d r i l l in g o p era tio n s. T h e tests a r e a c c u r a t e l y a n d e a s ily m a d e w ith o u t the u se o f d a n g ero u s ac id s. T h e S yfo C l in o ­g ra p h is sp e ed y , se lf­c h e c k in g , s im p le to o p era te , a n d in e x p e n ­sive. C a n b e u sed on a w ir e l in e o r as a “ G o - D e v i l ” r u n n in g in s id e th e d r i l l stem o r o n san d o r b a ilin g lin e in o p e n h o le .

Another drilling record with the aid of the Self- Checking SYFO Clinograph

Congratulations to the Union Producing Company for drilling the Minnie Brown No. 1 in the Agua Dulce field—considered one of the deepest tests ever drilled in Texas !The Sperry-Sun SYFO Clinograph was used in checking inclinations while drilling. Over 114 records were made. The instrument was used at first on the Halliburton Wire line. At a greater depth it was run as a “ Go-Devil ” and fished out with a core barrel overshot, or alternatively recovered when the drill stem was pulled out of the hole. The mud, at the greater depths, weighed up to 13 pounds per gallon, with a viscosity of 28 to 30.Drillers everywhere may expect equally good performance wherever the SYFO “ Go-Devil ” is used.

SPERRY-SUN WELL SURVEYING CO.1608 Walnut Street, Philadelphia, Pa., U.S.A.

Kindly mention this Journal when communicating with Advertisers.

x v i i

La c h m a n V a p o u r P h ase T r e a t in g P r o c e s s

■ ■

LACHMAN T R EA T ED spirit does notrequire any inhibitor.

LACHMAN TREA TIN G » a singleoperation conserves anti-knock q u a lity ; reduces

gum content to the vanishing point; reduces sludge

and polymerization losses to the minimum and

reduces sulphur.

T h e practical advantages also of a m ethod w hich

is fool-proof in the sense that it cannot be over­

done must appeal to all refiners.

A. F. CRAIG & CO., LTD.P A I S L E Y

R epresen tin g:

VA PO U R T R E A T IN G THE WINKLER-KOCHP R O C E SSE S IN C ., ENGINEERING CO.,

555, South Flower Street, 335, West Lewis Street,Los Angeles CALIFORNIA Wichita KANSAS

Kindly mention this Journal when communicating with Advertisers.

x v i i i

o f r o b b e r

That's what an engineer said

looking at a lively Duplex steam

pump that was struggling with a

highly volatile product. This

odd behaviour is more usually

described as 'short-stroking/

and it occurs whenever the

ordinary Duplex steam pump

draws gas instead of a full

charge of liquid; piston accelera­

tions become erratic, and the

steam ports are closed before

the completion of the stroke.

Some users have discarded

Duplex pumps on this account;

o th e rs h a v e p u r c h a s e d

Hayward-Tyler Duplex pumps

with Twells' valve gear, and

are now getting the economic

advantages of this type com­

bined with positive action.

With Twells' valve gear each

piston rod closes its own steam

port at the end of the stroke, the

opening of the port being

effected by the opposite piston

as with the standard Duplex.

The result is ability to work with

gas-laden liquids or even against

a vacuum, and incidentally a

saving in steam.

S, C O . L T D .

PUMP M A K E R S LUTON B E D SKindly mention this Journal when communicating with Advertisers.

TAPES-RULES-PRECISION TOOLStA jL C f ii J r u lu / it r iy '

STAN D ARD O F A C C U R A C Y F O R T A N K S T R A P P IN G , T A N K G A G I N G A N D G E N E R A L M E A S U R I N G

T H R O U G H O U T T H E W O R L D . . .Among others we mention:

"Atlas” — The W orld’s Best Gaging Tape "Derrick” — The Tape with Sturdy Hook

S E N D F O R F R E E C A T A L O G No. 12

t h e / u f k /n P u l e (7 o .

SAGINAW , M ICHIGAN, U. S. A.

t h e [ u f k / n R u l e f? o . o f Q a n a d a J t d .

W IN D SO R, O N T A R IO

Kindly mention this Journal when communicating with Advertisers.x x