August , 1928
INDUSTRIAL AND ENGINEERING CHEMISTRY
mention was, I believe, given in a paper presented before the Society of Automotive Engineers. D. P. BARNARD, 4TH-YeS. R. R. MATHEWS-AS I remember it, they had worked out a method of feeding fresh oil to the cylinders of the engine, after which the oil dropped into the crankcase and was circulated through the rest of the system. I was very much interested in it, because it has always seemed that we would have better piston lubrication if we could always have fresh oil on the cylinder walls. Has any effort been made to adopt this system generally? D. P. BARNARD, ~TH-No, I don't believe so. The system to which you refer is the Madison Kipp system and, while it has been successfully applied t o the larger types of industrial engines, it has up to the present time received no attention from the manufacturers of automobiles. As Doctor Mathews states, this system feeds fresh oil to the cylinders-that is, oil from a small storage tank which has not previously been in circulation in the engine. I think that this work was a revelation in that it showed what an extremely small amount of oil the cylinders of an automobile engine really require. I n connection with a 4 by 5 inch truck engine I believe it was found that 15 drops a minute were sufficient to lubricate all of the cylinders a t 1000 r. p. m. This is certainly a long cry from the present system in general use, which involves the throwing of oil in great quantities on to the cylinder walls. R. R. MATHEWS-I believe that this work indicated that it was possible to keep oil consumption under control and a t the same time prevent undue cylinder wear. D. P. BARNARD, ~TH-Ithink it is the general experience of all those engaged in dynamometer work that an engine on the block
849
will show less wear for a period of operation equivalent to 100,000 miles or so than it will actually show in less than one-tenth of that mileage on the road. It is almost impossible t o wear out an engine on the block where it does not have to digest the vast quantities of dirt which it must breathe on the road. The fresh oil system supplies to the cylinder oil that contains no grit, and it has been definitely proved that the life of the engine is greatly prolonged by the use of such a system. I am not sure that we oil men should agitate too strongly for the general adoption of such a system, as it is very effective in reducing the oil consumption of a n engine and would therefore force us to depend almost entirely on crankcase leakage to keep up the year's volume of business. We have probably nothing t o worry about for a year or two a t least, as the cost of developing and putting into practice such a system for the various types of production cars would be enormous. It is doubtful if the oil saved by the prevention of excessive consumption would pay for its installation cost. From the point of view of cold-weather lubrication, however, a positive-feed oiling system to the cylinder walls has much to recommend it. I understand that quite beneficial results have been obtained by the use of a modification of this system on some automobile engines in which oil is forced to the cylinder walls when the engine is being cranked. Of course, this supply of oil takes place only when the starting pedal is depressed and as soon as the engine catches it must depend for lubrication upon its regular lubricating system. However, the small shot of oil given the cylinders during cranking period in most cases is probably just enough t o prevent damage to the pistons and will lubricate them adequately until the main oiling system can come into play.
Miscible Carbon Disulfide' Walter E. Fleming a n d Reinhold Wagner JAPANESE BEETLELABORATORY, U. s. BUREAUOF
H.E treatment of soil about the roots of nursery stock with dilute emulsions of carbon disulfide is one of the most effective methods for destroying the Japanese beetle (PopiUia japonica N e w . ) in, the soil without causing serious damage to plants. A miscible carbon disulfide which has been perfected recently is equally as effective as emulsions of this compound for destroying the soil-infesting stages of the beetle. It is no more injurious to plants, and furthermore, it has certain physical and chemical advantages.
T
Properties
Miscible carbon disulfide is a mobile, transparent liquid with a specific gravity of 1.1156 at 15" C. It does not form a heavy foam when shaken, and it can therefore be poured easily and measured accurately in small quantities. It mixes readily with water in all proportions, forming a white emulsion. Unlike emulsions containing soap and water, it does not stratify, but remains homogeneous. It does not disintegrate under normal conditions, and it can be held at a temperature of 32" F. (0" C.) for one week without breaking down. Ingredients
Miscible carbon disulfide is a mixture of carbon disulfide with castor oil, potassium hydroxide, denatured alcohol, and water. As it can be made satisfactorily only by using high-grade materials, it is necessary to specify carefully the quality of each ingredient. Carbon disulfide. Use a good grade of technical carbon disulfide which contains only small proportions of free sulfur, hydrogen s a d e , sulfuric acid, or sulfurous acid. Castor oil. Use a blown castor oil with a specific gravity be-
ENTOMOLOGY, MOORESTOWN, h'.J.
tween 0.991 and 1.004 a t 15.5' C. This oil should have an iodine number2 between 60 and 53, a saponification number3 between 205 and 216, and a titer4 of 3. The fatty acids of this oil should have an iodine number between 63 and 53 and an acid value between 210 and 225. An unblown castor oil cannot be used with satisfactory results. Alcohol. 'Cse 190-proof ethyl alcohol. Ethyl alcohol which has been completely denatured according to formula No. 1 of the United States. Bureau of Internal Revenue has been found satisfactory and :s recommended. This alcohol contains 10 parts by volume of methanol and 0.5 part of benzene to 100 parts of ethyl alcohol. No permit from the United States Bureau of Internal Revenue is required to handle this grade of denatured alcohol. Potassium hydroxide. Use a high-grade potassium hydroxide. It should be a t least 80 per cent pure and contain not more than 4 per cent potassium carbonate. It should be soluble in alcohol, and should have only traces of sulfates, chlorides, nitrates, or silicates. Water. Use distilled water, rain water, or water containing not more than traces of dissolved salts. Calcium and magnesium salts are particularly objectionable. Apparatus
The castor oil must be saponified by alcoholic potassium hydroxide in a closed kettle which is equipped with a heating device and a mechanical agitator. I n the laboratory small quantities of oil have been saponified in a three-necked flask heated over a water bath. The flask was equipped with a mechanical agitator, a thermometer, and a reflux condenser to prevent loss of alcohol. Preparation
Dissoh-e sufficient potassium hydroxide in a solution composed of 7 parts by volume of alcohol and 3 parts of water 2
Received April 18, 1928. Laboratory. 1
Contribution No. 40 of the Japanese Beetle
3 4
Assocn. Official Agr. Chem. Methods, p. 244 (1920). Ibzd., p. 246. I b z d . , p. 242.
Vol. 20, No. 8
I N D U S T R I A L A N D ENGINEERlNG CHEMISTRY
850
to give a concentration of 24.5 to 25.0 per cent potassium hydroxide. (In the laboratory an excess of alkali was dissolved in the alcohol and water, aliquot samples were taken and were standardized against a normal hydrochloric acid solution, and sufficient alcohol and water were added to obtain the proper concentration of potassium hydroxide.) Then mix 55 parts by volume of castor oil with each 10 parts of alcoholic potassium hydroxide. Close the container, start the agitator, and gradually increase the temperature t o 200" F. (93.3" C.). Hold a t this temperature and agitate until the alkali has reacted completely with the oil. In small batches, 2 hours a t 200" F. were sufficient. Test saponification by withdrawing small samples a t intervals, mixing them with half their volume of carbon disulfide, and observing their action in water. When the drop breaks into its component parts in water so that carbon disulfide settles to the bottom in droplets, saponification is not complete; when it breaks into a milky-white liquid which gradually diffuses throughout the water, the proper stage has lbeen reached. The alcoholic soap, when completed, should have the following composition hy weight: P e r cent Total solids: Castor oil Potassium hydroxide Total volatile constituents: Alcohol Water
83.2 3.7
P e r cent 86.9 ... 13.1
8.6 4.5
When the alcoholic castor-oil soap has been prepared, cool it to room temperature and mix with carbon disulfide, using 35 parts by volume of carbon disulfide for each 65 parts of the soap, and stirring the mixture until it is homogeneous. MisEible carbon disulfide can be used immediately, or placed in storage in tight containers. When completed, i t should have the following composition by weight: P e r cent Total solids: Castor oil Potassium hydroxide Total volatile constituents: Carbon disulfide Alcohol Water
49.5 2.2
Per cent 51.7
40.5 5 1 2.7
48.3
The proportion of no ingredient should vary more than 1 per cent from the figure given in the formula. Analysis
I n order to assist the manufacturer in the preparation of a standard product, the following procedure is suggested as a means of analyzing the finished product. Analytical Tesults as obtained in this laboratory are given in Table I. Table I-Quantitative ANALYSIS No. of detns.
TOTAL SOLIDS KOH
12 10 P e r cent P e r cent Maximum 52.33 2.29 Minimum 50.65 2.00 Average 51.86 2.10 Theoretical 51.7 2.20 0 Estimated.
OIL
Analytical Data
TOTAL
VOLATILE CSz 12 10 P e r cent P e r cent P e r cent 49.35 41.52 47.67 40.20 49.76O 4 8 . 1 4 40.59 49.5 48.3 40.5
ALCOHOL WATER 10 P e r cent
AND
7.554 7.8
TOTAL SOLIDS-weigh a 20-gram sample of miscible carbon disulfide on an analytical balance in a glass-stoppered, flabbottom weighing bottle, and place on a steam bath to remove the readily volatile constituents. Then place in an oven a t 105" C. and dry to a constant weight. POTASSIUM HYDRoxmE-Take a lo-gram sample of the miscible carbon disulfide, drive off the volatile constituents on a steam bath, and dissolve the residue in 500 cc. of boiling distilled water. Add 10 cc. of normal sulfuric acid and boil with vigorous agitation until the fatty acids separate
and the aqueous solution becomes clear. Allow the solution to stand overnight. Filter through a wet filter, wash the fatty acids thoroughly with distilled water, and titrate the filtrate with normal sodium hydroxide solution, using phenolphthalein as an indicator. After deducting from the total volume (cc.) of the normal sulfuric acid used the volume of normal acid neutralized by the sodium hydroxide, there is left the volume of sulfuric acid which reacted with the potassium in the castor-oil soap. The percentage of potassium hydroxide in the sample is determined by using the following formula:
R = M X 0.0561 X
100
W
where R = per cent of KOH in samples M = cc. of N H2S0, reacting with KOH W = weight of sample, grams 0.0561 = grams KOH per 1 cc. normal solution
OIL-An approximate quantitative estimation of the oil can be obtained by deducting the percentage of potassium hydroxide from the percentage of total solids in the sample, since these two ingredients, together with traces of impurities in the volatile constituents, form the non-volatile part of the miscible carbon disulfide. The quality of the oil is determined by the characteristics of its fatty acids. Take a 0.2- to 0.3-gram sample of the washed fatty acids left on the filter paper in the potassium hydroxide determination and determine the iodine number of the fatty acids by H a n d s method.* Determine the acid value by dissolving a 2-gram sample of washed fatty acids left on the filter paper in a known excess volume of standard alcoholic potassium hydroxide, and determining the excess alkali that was added by titrating against standard hydrochloric acid, using phenolphthalein as an indicator. The milligrams of potassium hydroxide reacting with 1 gram of fatty acid is the acid value. Four samples were taken from each of two different batches of miscible carbon disulfide. The following results were obtained: IODINE ACID SAMPLEh T VALUE ~ Batch 1 1 61.0 210 2 60.7 210 3 60.0 210 4 60.0 210
~
IODIXE ACTD SAMPLE ~ NUMBER ~ ~ VALUE Batch 2 1 53.9 222 2 54.1 217 3 54.1 220 4 56.0 223
Since the fatty acids of the oil should have an iodine number between 63 and 53, and an acid value between 210 and 223, it is obvious that the proper oils were used in the preparation of the material. TOTALVOLATILE COXSTITUENTS-Weigh a %)-gram sample of miscible carbon disulfide on an analytical balance in a glass-stoppered, flat-bottom weighing bottle, and place first on a steam bath to remove the readily volatile constituents, and then in an oven a t 105' C. and dry to a constant weight. CARBONDISULFIDE-Dilute a 10-gram sample of miscible carbon disulfide to 1000 cc. with distilled water in a volumetric flask. Take 10 cc. of this dilute solution for analysis and mix with 10 cc. of 20 per cent alcoholic potassium hydroxide. Keutralize with 50 per cent glacial acetic acid, using phenolphthalein as an indicator. Add a slight excess of solid sodium bicarbonate. Dilute to 50 cc. with distilled water, and titrate with a standard iodine solution, using starch as an indicator, until the blue color persists for 30 seconds. One cubic centimeter of the iodine solution required is equivalent to 0.0129 gram of carbon disulfide present The total quantity of carbon disulfide cannot be detected by the iodometric method, probably because some is absorbed by the castor oil. It has been found that the total quantity of carbon disulfide can be estimated by multiply-
INDUSTRIAL A N D ENGINEERING CHEMISTRY
August, 1928
ing the number of grams of carbon disulfide determined iodometrically by the factor 1.14, which was derived by dividing the weight taken by the weight recovered in a large nnmbcr of laboratory tests. ALCOHOLAND U'ATm-Small quantities of alcohol and water caniiot be readily determined quantitatively in miscible carbon disulfide. The appearance and action of the finished product is a fair index of the proper proportions of these
851
ingredients. If there is too much water, the miscible carbon disulfide will be translucent; if there is too much alcohol, crystals will be formed in the transparent. solution; if the proportions are correct it will remain transparent without the formation of crystals. The approximate total quantity of these ingredients can be estimated by deducting the percentage of carbon disulfide from the percentage of the total volatile constituents.
Chipping and Abrasion Tests for Paint Coatings on Metal' A. D. Camp RBSBAACX BURBAV. ALVMINUX COIIP&NY OP A I B I I C ~ BOPPILO, , N Y.
IIE selection of paints and varnishes for the finishing of furniture made of strong aluminum alloy has p r c sented problems of a somewhat different nature fnim those connected with their use on articles composed of other metals and wood. Aluminum, being a silvery white metd, shows B very great color contrast between finish and metal, if the finish is scratched or worn off. Therefore, in addition to considering the color and beauty of finish, the manufacturer of aluminum furniture must select extremely tough and durable paints, which are as resistant as possible to abrasion and chipping. A diligent search did not disclose any tests for determining those qualities which were exactly suited to our needs, so the methods in this article were developed for the purpose. These tests hare been found very serviceable in studying the comparative qualities of paints obtained irom different sources and a.pplied to aluminum by different processes.
T
Nale-since this paper v a s prepared an article by Vogt [IND. ENO. Cxnar.. PO, 303 (1828)l has appeared, which describer a machine for testing the abrasion resistance of rubber. The operation of the paint and rubber teesting mechiner is quite similar except thst the abrading media and the test specinlens eichange positions. Some of the mathematical relationships walked out for the rubber machine m k h t be applied t o the abrasion testing of paint fiims. The chipping test is hardly cract enough to become a standard test for u s c b y operators working under dieerenf conditions, but it i s so simple iii its a.. ~ ~ l i c a t i othat n it rhotiid become B vrluabie nieans of aiiictlv . . f e r t i n ~the comparative tovzhneri and adherion of paint films. The abrarion test. on the other hand. i s subject t o quite exact ~onirol,and with further refinement might well become a standard test fordetermining the wearing guaiities of c p d i e d w i n t r . Sumeriions from paint users and manufacturers will be welromrd by t h e author.
varnish, it is necessary in most instances to test, the complete finish, in order to determine the durability of the finished article, as each coat added has an effect upon those immediately above and below it. In teatin:: finishing varnishes all samples of varnisli are sprayed upon unifomily applied backgrounds of primer and gruund color. The comparative tests of ground colors are ucually applied to primed sampleprepared under standardized conditions. It is also desirable to have all paints applied, dricd. and tested at t l e same time, since it is not always possihle to duplicate condition9 in different series of tests. The thickness of all panels is measured to 0.0001 inch with Yernier mirrometer caliprs in a t least ten spots before and after painting, and the readings are recorded on the hacks of the specimens, so thst the thickness of the paint filii1 at the point of chipping can be obtained. Tlie total thickness of the paint film for tlie chipping test has a very great influence o n the character of the chip, and for these tests it is maintained between 0.0025 and 0.0030 inch. Other thiiigs being equal, tlie heaiier the paint film the greater is the tendency to chip. Another factor which affects the test is the liardness or temper of the metal composing the sample panel; increasing softness of the metal greatly derrenees the tendency to chip.
Chipping Test
The chipping test is made by dropping on the test panel under standardized conditions R series of standard tools, and observing the extent and character of the impression made on t,he coating. PREPARATION OF TEST SPEcrmNsSince the adhesion of paint to metal surfaces is greatly influenced by their condition before the application of paint, it is of prime importance that a standardized surface condition be adopted for all chipping tests. For aluminum a sand-blasted surface has been found to offer the best foundat.ion for paint, so the test samples are uniformly sand-blasted on both sides. Specimens 4 inches wide by 8 inches long are amply large for both chipping and abrasion tests. I n testing paint films it is desirable to test, each element of the finish separately, but since most finishes for aluniiniim furniture arc compounds of primaries, ground coats, and I
Received April 30, 1928.
Cowlesy of Roszvrli Allen
Abrasion Teeflng Machine
For testing tlie finishes of aluminum furniture, 16 gege 51 SW alloy is selected for t,lie sample panels. This alloy has a composition of 0.1 per cent silicon, 0.6 per cent magncsium, balance aluminum, and in tlie heat-treated and quenched condition, denoted by the initial (W), it has a t,rnsilc strength
