INDUSTRIAL AND ENGINEERING CHEMISTRY
822
of these plants has been higher than elsewhere, the daily operating costs and the item of depreciation are lower. The writer has a friend who purchased stock at par in one of these projects as a matter of local duty and patriotism. During the second year he was continually solicited to sell the stock a t an advance. Its par value increased in 1928 from $100 to $230, and a t the latter price the stock showed a yield of 11.32 per cent. This reflects the summation of economies that have been effected in the operating and producing costs of Pacific Coast pulp and paper mills. They represent the fruitage of technical experience in the selection of equipment that reduces labor costs. I n them technical control, scientific research, and experimental development have a prominent place which insure quality of product that can nowhere be surpassed. I n the new cellulose industries utilizing pulp for purposes other than paper, the Pacific Coast product promises to assume a prominent place.
Vol. 22, No. 8
Future of Industry
With 40 per cent of our wood-pulp requirements still met by iinportations from Canada, Scandanavia, and other European countries, it is obvious that our nation is not a self-contained unit in the supply of this essential commodity. The future trend of the industry must also be inevitably connected with the economic and trade factors a t work in other parts of the world. The very favorable conditions in the Paciiic Northwest for the utilization of our forest products will insure a large participation whatever the competitive struggles of the immediate future may be in supplying the increasing per capita consumption of pulp and paper products. Literature Cited (1) Partridge,
IND.ENG.CHEM.,22,
422 (1930).
Carbon Black in Rubber Insulating Compounds* W. B. Wiegand and C. R. Bog@ BINKEY8; SMITHCo., 41 EAST4
2 ST., ~ KEW ~ YORK,N. Y., AND SIMPLEXWIRE& CABLECo., BOSTON, MASS,
Up to about 10 per cent by weight of properly made As regards r e s i s t i v i t y , and dried carbon black may be added to each 100 parts t h e r e i s n o g r e a t change U B B E R i s a nonof rubber hydrocarbon present in rubber insulating when added i n m o d e r a t e conductor or dieleccompounds with marked improvement i n dielectric a m o u n t s ( 4 ) . As regards tric. Carbon black strength, resistivity, and power factor, and without power f a c t o r , most fillers consists mostly of carbon, serious increase i n dielectric constant. increase it ( 5 ) . which is a conductor and so The exact amount of carbon black required depends Regarding dielectric connot a dielectric. Therefore, somewhat on which electrical property is to be brought stant, we a r e t o l d t h a t the addition of the latter t o its maximum value. Fresh, uncompounded carbon fillers i n c r e a s e this value, to the former must detract black m u s t be employed. the ‘(extent depending on from its dielectric properElectrical improvements of u p to 50 per cent are the amount used and the ties. Thus current belief, possible. characteristics of the filler” and thus the literature. The improvement is thought to be due to the removal (6). The effect of carbon The contrary is the case. by the carbon black of t h e ultimate traces of moisture black as a comDoundinn inThe admixture in suitable and electrolytic impurities. gredient is statkd to be- difproportions of carbon black ferent from that of other t o -rubber insulating compounds not only does not injure but may improve them, fillers, owing to its being a conductor. As little as 0.2 per cent is stated to have a perceptible effect, whereas “with and to a striking extent. 20 per cent, the dielectric constant is more than double Dielectric Properties of Rubber that of the base compound. The increase in dielectric constant is almost linear and is much greater than with CRUDERUBBER-crude rubber is an excellent insulator, equal amounts of other fillers.” the resistivity being of the order of 5000 X lo6megohms per Thus, for example, we read ( 7 ) : cc. (1968 X lo6 megohms per cu. in.). The power factor is The introduction of carbon greatly decreases the resistivity of the order, 0.2. The dielectric constant, at 1000 cycles, of rubber. This effect is not apparent, however, with 0.2 to is of the order, 2.5 (10). 2 per cent of carbon. With 10 per cent carbon some of the Note-Power factor, as applied to a dielectric, measures its tendency specimens show a resistivity of 2000 X 10’ M. 0. per cc. which PART I-THEORETICAL
R
t o dissipate electrical energy when subjected to alternating voltage. It is usually expressed as a percentage. Thus the “power factor” is the electrical analog of friction in a machine.
is not lower than that of some pure rubber compounds. With higher percentages of carbon, however, the resistivity falls off with greater rapidity.
PUREVULCANIZED RUBBER-with pure-gum mixes of rubber and sulfur the resistivity ranges irregularly from 1000 X lo6 to 15,000 X lo6 megohms per cc. (11) (394 X 106 to 591 X 106 megohms per cu. in.). The power factor at given frequency is a function of the sulfur ratio reaching 9 per cent (12). The dielectric constant also ranges irregularly from 2.6 to nearly 5, depending on frequency (12). COMPOU~;DED RUBBER-Most types of inorganic compounding ingredients are stated to behave as follows:
As to power factor, it is stated that “with 10 per cent of carbon black the power factor is about ten times that of the base compound, and with 20 per cent it is nearly 30 times that of the base compound.” EFFECT OF MOISTURE-Moisture is injurious because of its power to dissolve and ionize any traces of electrolytes, on the one hand, and of its own effect, on the other. For details as to its effect on the dielectric constant, resistivity, and power factor, the reader may consult Boggs and Blake ($), Williams and Kemp (19), and Nuttall (14). I n brief, moisture profoundly affects all of these electrical properties, not only in the case of rubber, but also in the
1 Received April 15, 1930. Presented before the Division of Rubber Chemistry at the 79th Meeting of the American Chemical Society, Atlanta, Ga., April 7 to 11, 1930.
I N D U S T R I A L A N D ENGINEERING CHEMISTRY
August, 1930
case of insulating oils, where 1 part in 10,000 reduces the breakdown voltage by 70 per cent (14). The view has even been advanced that the “insulation resistance of an insulating material is really a measure of its degree of dryness.” Properties of Carbon Black CohfPoSITIoP;--The composition of carbon Hack varies according to the method of manufacture. For example, a “roller” black as manufactured for the ink trade will analyze about 91 per cent of carbon, the remainder consisting es-
pooo
ADSORPTIOS BEHAYIOR-Like other forms of active carbon, carbon black may show a strong avidity for moisture and for dissolved substances. Table I shows, merely by way of illustration, the “removal” effect of different types of carbon black on water, on an alcoholic solution of diphenylguanidine, and on an aqueous solution of potassium hydroxide (these solutions being approximately 0.01 normal in strength). The details of carbon black “removal” will be discussed in a later communication. It is mentioned a t this time to explain the reasoning which led up to the experiments of this paper. We pass now to a consideration of the known facts about other conducting materials when in a state of colloidal dispersion in dielectrics.
6
-TF 100 20 30 40 50 60 70 80 POUNDS CARBON BLACK PER 100 POUNDS RUBBER
IO
Figure 1-Effect
90
Electrical Properties of Conducting Substances Colloidally Dispersed in Dielectrics Smoluchowski (1.5) has calculated the conductivity to be expected in a pure sol of gold, for example, and arrives a t a figure of 1.2 x lo-’, an extremely low value. It would appear that the conductivity of dilute metallic sols is influenced very little by the conductivity of the dispersed metallic particles. Svedberg (16) states:
100
“In some cases the conductivity of the sol has been found to be even less than the conductivity of the disDersion medium. If we disperse mercury in- pure conductivity water by means of the oscillatory arc, the conductivity decreases. That probably is due to the adsorption on the mercury particles formed of the impurities present. The conductivity, therefore, decreases. The small contribution to conductivity from the particles can, of course, be disregarded.”
of Increasing Dosages of Carbon Black in Rubber
sentially of oxygen (3). Carbon black made on channels as furnished to the rubber industry will analyze about 96.7 per cent of carbon, together with oxygen, hydrogen, etc. I n addition to these non-carbonaceous elements, carbon black of course contains a relatively large amount of entrained air, which, however, is practically all eliminated during admixture into rubber. Owing to its extremely small sizeca. 80 millimicrons (mp)-the exact structure and relationship of the carbon l o the non-carbon components are not known. I n any case it cannot, a priori, be assumed that the electrical behavior of carbon black, when dispersed in rubber or in any other medium, will be that expected of a pure carbon phase. Table I-Adsorption Properties of Carbon Black GAINI N WEIGHT AFTER L O N G DPG KOH EXPOSURE OVER ADSORPTION ADSORPTION BLACK WATER INDEX INDEX “Rubber” grade Special No. 1 Special No. 2
823
%
%
%
4
10 40 88
47
30 56
20
82
STRUCTURE WHENDISPERSED IS RUBBER-Despite several serious efforts to study the morphology of carbon black dispersions in rubber by means of the microscope (Q), it i s not known how completely carbon black is dispersed nor how its particles “pile” or arrange themselves. By indirect reasoning, however, we may be certain that with increasing concentration agglomeration sets in (18). The curves in Figure 1 illustrate the decay in physical properties as the percentage of carbon black is increased beyond a certain point (1). )Vote-The onset of agglomeration is also clearly indicated by heat of wetting experiments. See the excellent studies of Hartner, Kollotdchem Bezhefte, 30, 83 (1929).
At very high concentrations there would appear to be a partial inversion of phase in which some rubber units are surrounded by carbon black instead of the rarbon-black particles being surrounded by rubber. This results in a leather-like structure which, curiously enough, is already the basis of an important division of the rubber industry. At the lower concentrations the dispersion of carbon black becomes relatively complete.
The work of Whitneg and Blake ( I ? ) and of Xordenson ( I S ) indicates that with most metallic sols the small conductivity observed arises from residual electrolytes (remaining a s impurities on the dispersed phase). I n view of the foregoing it seemed unnecessary to assume that a conductor such as carbon black should prove any more deleterious when dispersed in low percentages in rubber than does gold, a much better conductor, when dispersed in water. Furthermore, the injurious action of water, and of watersoluble impurities, taken in conjunction with the well-known adsorptive activity of carbon black toward water (and such dissolved substances) suggested that the addition of carbon black in the proper amount might not only leave the dielectric properties of rubber intact but even improve them. I n other words, it was hoped that the improvement due to the possible removal (the word “removal” is here used in the electrical sense) of these impurities, might be attained before being overtaken by the ill effects associated with the higher concentrations of carbon black. PART 11-EXPERIMENTAL
The following experiments on carbon black insulating compounds were carried out in the laboratories of the Simplex TTire and Cable Co., Boston, Mass. Experimental Conditions The crude rubber used was smoked sheets broken down 30 minutes in a full-size internal mixer, washed, and vacuum-
dried in the factory. Zinc oxide (Red Seal) and whiting were dried to constant weight a t 212” F. (100”C.). The carbon black (Micronex) was heated 2 hours a t 325” F. (166’ C.), The compounds were mixed on a small laboratory mill, great care being taken to insure uniform dispersion of the mineral ingredients. The dried powders were carried direct from the oven to the mill. After mixing, the compounds
INDUSTRIAL A N D ENGINEERING CHEMISTRY
824
were stored 24 hours in a closed vessel containing calcium chloride. The samples were cured in a laboratory hydraulic platen press, the temperature being 270" * 0.5" F. (132' C.). Preliminary cures were carried out on the base compound and on the compounds containing the minimum and maximum percentages of carbon black. From these results the best technical cure was determined as 30 minutes, which was the cure used for all samples. The samples were tested a t 73" F. (23' (2.). B R 0 P
I-
= BREAKDOWN VOLTS
= RESISTIVITY
= DIELECTRIC CONSTANT = POWER FACTOR
Vol.'22, No. 8
Reduction of Data I n view of the experimental difficulties in obtaining close duplicates in the electrical measurements, arithmetic means were used, first rejecting wild values. The "four-times" rule was used to determine "wildness." That is, the mean of the two closest values was computed. Their average deviation and the deviation of the remaining value from the above mean compared with this average deviation. Any value showing a deviation four or more times the average deviation was rejected. 40 PER CENTRUBBERCOMPOUND Smoked sheets.. . . . . . . . . . . . . 40.46 Whiting.. . . . . . . . . . . . . . . . . . . 42.10 Zinc oxide.. . . . . . . . . . . . . . . . . 13.30 Paraffin.. . . . . . . . . . . . . . . . . . . 2.21 Agerite powder.. . . . . . . . . . . . . 0.22 Monex.. . . . . . . . . . . . . . . . . . . . 0.12 Sulfur.. .................... 1.60
100.00
B P- IOOC D- IO00
P-440.0004
0
2 4 6 8 1 0 VOLUME PERCENT CARBON B L A C K F i g u r e 2-Effect of C a r b o n Black in 40 P e r C e n t R u b b e r Insulating C o m p o u n d
ELECTRICAL TEsTs-Embedded aluminum electrodes were used in all tests except break-down, and carefully centered on the uncured sample prior to vulcanization. An electrode, of 6-inch (15.24-cm.) diameter, was placed over, and a 61/2inch (16.5-cm.) electrode on the bottom of, each sample. The samples were approximately 9l/r inches (24.4 cm.) square, the thickness averaging 0.060 inch (0.15 cm.) between electrodes. Triplicate samples were tested for each cure. The voltage (break-down) results were obtained on the same samples, employing %inch (5.08-cm.) disks. The 1000-cycle dielectric constant was obtained by means of a General Radio capacitance bridge, using a tuning fork oscillator as the source of power. The 1000-cycle power factor was obtained simultaneously with the capacity (to which the dielectric constant is directly proportional), and was calculated from the circuit values, when balance was obtained. The d. c. dielectric constant was calculated from the dimensions of the sample and the capacity obtained by comparing the deflection of a ballistic galvanometer when the sample was discharged through it immediately after a 10second charge a t 300 volts, with the standard deflection obtained from a standard condenser under similar conditions. The resistivity was measured by the loss-of-charge method. The sample was charged a t 300 volts for 10 seconds, disconnected from the charging battery and, after standing 1 minute, discharged through the ballistic galvanometer. The resistivity was calculated from the capacity and the ratio of the deflection due to the delayed discharge to that obtained by discharge immediately after the charging battery was disconnected. Dielectric-strength tests were made by using 2-inch (5.08-cm.) round disks, the voltage being raised at the rate of 1000 volts per second.
Carbon black (Micronex) was used in the above compound to replace whiting by progressive volume substitution. The results are shown graphically in Figure 2. These data may be summarized as follows: (1) At approximately 3 l / 4 per cent of carbon black (Micronex) by volume or 4 per cent by weight, the breakdown voltage was raised from 408 to 571, an increase of 40 per cent. (2) ' The resistivity was raised from 76 X 106 megohms per cubic inch to 128 X lo8 megohms per cubic inch (193 X 106 M. 0. per cc. to 325 X lo6 M. 0. per cc.), an increase of 68 per cent. (3) The dielectric constant was changed for d. c. from 4.94 to 5.34; a t 1000 cycles from 4.39 to 4.99; a t 440,000 cycles from 4.26 to 4.67. It is seen that the specific inductive capacity has not been materially raised by this addition of carbon black. (4) At 1000 cycles the power factor has decreased from 1.05 to 0.793, an improvement of 24 per cent. (With 2 per cent by volume of carbon black (Micronex) the power factor was only 0.541, a decrease a f 48 per cent.) At 440,000 cycles the power factor dropped from 1.81 to 1.46, a decrease of 19 per cent.
It will be noted that, with no significant raising of the dielectric constant, the addition of 4 per cent (10 per cent on the crude rubber) of carbon black-a conductor-has improved the dielectric strength, resistivity, and power factor of a typical high-quality insulating compound by about onehalf. 30 PER CENTRUBBERCOMPOUND Smoked sheets. . . . . . . . . . . . . . 30.17 Whiting.. . . . . . . . . . . . . . . . . . . 53.29 Zinc oxide.. . . . . . . . . . . . . . . . . 11.16 Litharge. . . . . . . . . . . . . . . . . . . 2.48 Ozokerite., . . . . . . . . . . . . . . . . . Agerite powder., . . . . . . . . . . . . Monex.. . . . . . . . . . . . . . . . . . . . Sulfur.. ....................
1.24 0.15 0.12 1.39
100.00
Carbon black again replaced whiting in equivalent volume increments. The tests on this compound are shown graphically in Figure 3. The results may be summarized as follows: (1) At approximately 2 per cent by volume (slightly more by weight) breakdown increased from 505 volts per mil to 528 volts per mil, an increase of 5 per cent. (2) Resistivity increased from 131 X 108 M. 0. per cubic inch to 186 X 108 M. 0. per cubic inch (333 X 10' M. 0. per cc. to 473 X 108 M. 0. per cc.), an increase of 42 per cent. (3) Dielectric constant (d. c.) increased from 4.83 to 5.20; at 1000 cycles from 4.66 to 5.02; at 440,000 cycles from 4.37 to 4.70. Here again it will be noted that there is no significant increase in the dielectric constant. (4) The power factor at 1000 cycles increased from 0.57 to 0.59; substantially no change.
Here again we note a definite electrical improvement in the case of the lower quality (30 per cent) compound, the effect being less than with the 40 per cent compound. This
INDUSTRIAL AiVD ENGINEERING CHEMISTRY
August, 1930
suggests that the beneficial effect of the added carbon black may be due to the removal of certain impurities which are associated with the crude rubber present in the compound. 35 PER CENTCOMPOUND CONTAINING RECLAIMED Smoked sheets.. . . . . . . . . . . . . . Alkali reclaimed tiresa.. . . . . . . Zinc oxide.. ................. Mineral rubber.. . . . . . . . . . . . . . Agerite powder., . . . . . . . . . . . . Accelerator.. . . . . . . . . . . . . . . . . Sulfur.. ....................
35.5 21.7 23.0 17.3 0.5 0.1 1.9
825
m-ell-known ability of carbon black to adsorb water and dissolved electrolytes endows carbon black-rubber insulating compounds of various types with improved dielectric strength, resistivity, and power factor, the specific inductive capacity remaining substantially unchanged. I n some cases this improvement may exceed 50 per cent.
-
100.0 The reclaimed rubber contained the carbon black originally present
in the tire treads, This carbon black is not shown in the data, since it is not “fresh,” having passed through the reclaiming process.
I n this case both the reclaimed rubber and the crude rubber were vacuum-dried to constant weight. Zinc *de was replaced by carbon black (Micronex) in progressive volume for volume substitutions. The results are shown in Figure 4,and may be summarized as follows: (1) At approximately 3 per cent by volume (5 per cent by weight) of carbon black breakdown in volts per mil increased from 370 to 420, an increase of 14 per cent. (2) Resistivity at approximately 1 per cent by volume (1.5. per cent by weight) increased from 627 X lo8 M. 0 . per cubic inch to 643 X 106 (1590 X 108 M. 0. per cc. to 1635 X 106 M. 0. per cc.), an increase of 2.5 per cent. (3) Dielectric constant increased from 4.27 to 4.35, an increase of 1.9 per cent. At 1000 cycles it increased from 3.94 to 4.02, an increase of 2.1 per cent. (4) At 2 per cent carbon black by volume (3 per cent by weight) the power factor was improved from 1.25 to 1.12, a decrease of 10 per cent.
5.5c
3.5;
Here again there has been definite improvement in electrical properties without important change in dielectric constant.
P- 1 0 0 0 ~
-
P- 440.000
0
2 4 6 VOLUME PERCENT
Figure 3-Effect
8
Per C e n t C o m p o u n d Containing Reclaimed
3-The prevailing opinion that carbon black is injurious to rubber insulating compounds which are to be used next
B = B R E A K D O W N VOLTS
0
Figure 4-35
I 2 3 4 5 6 VOLUME PERCENT CARBON B L A C K
to the wire, or which in general are expected to serve as electrical insulation, has been shown to be erroneous, provided the proper proportions are employed. &These results would seem to render advisable the rewriting of many specifications dealing with rubber insulating compounds, and thus make it possible to apply the wellknown beneficial effects of carbon black compoundingimproved toughness, density, wearing resistance, imperviousness to light, tear resistance, etc.-to the electrical insulation field, from which it hap hitherto been barred. %Although it is strongly recommended that the proper dosage of carbon black (which must be of suitable quality and thoroughly dry) be redetermined in each case, the writers’ results would indicate that up to 10 per cent of carbon black on the crude rubber (plus the rubber content of any reclaimed rubber present) will effect the desired improvement in electrical properties.
IO
Acknowledgment
CARBON BLACK
of Carbon Black in 30 Per C e n t Rubber Insulating Compound
Conclusions
1-It has been shown that, in conformity with published behavior of other conducting substances (metallic sols, etc.), carbon black may be incorporated in a dielectric such as rubber without detracting from its insulating or dielectric properties. Published results to the contrary were in error, probably because the material was added in excessive amounts. 2-In addition to tjhis effect, i t has been shown that the
The writers desire to express their thanks for aid in preparation of the data to E. M. Follansbee and W. N. Eddy, of the Simplex Wire & Cable Company, and to J. W. Snyder, of Binney & Smith Company. Literature Cited Binney & Smith Laboratories, Unpublished results. Boggs and Blake, IND. ENG.CHBM.,18, 224 (1926). Bur. Mines, Bull. 192, p. 73. Curtis and McPherson, Bur. Standards, Tech. Paper 2Q9,p. 722. I b i d . , p. 721. ( 6 ) I b i d . , p. 720. (7) I b i d . , p. 713.
(1) (2) (3) (4) (5)
INDUSTRIAL AND ENGINEERING CHEMISTRY
826 (8) (9) (10) (11) (12) (13)
Green, “Microscopy of Paint and Rubber Pigments” (1922). Grenquist, IND. ENG. CHEM.,20, 1073 (1928). International Critical Tables, Vol. 11, p. 272 (1927). Ibid., p. 273. Kitchin, Trans. Am. Insl. Elec. Eng., 48, 495 (1929). Nordenson, Koltoid-Z.. 16. 65 (1915).
(14) (15) (16) (17) (18) (19)
Vol. 22, No. 8
Xuttall, Chemistry 5 Industry, 41, 1359 (1928). Smoluchowski, Physik. Z.,6, 529 (1905). Svedberg, “Colloid Chemistry,” p. 238 (1928). Wbitney and Blake, J . Am. Chem. Soc., 26, 1385 (1904). Wiegand. Canadian Chem. J . , 4 (1920). Williams and Kemp, J . Franklin I n s f . , 203, 35 (1927).
Secondary Esters and Their Use in Lacquers’ J. G. Park and M. B. Hopkins STANCO DISTRIBUTORS, 2 PARK AvE., N E W YORK, N. Y.
regarding them have been published. It is the purpose of this paper to present a systematic study of their properties based on the use of experimental data scientifically obtained with special reference to use in lacquers. The investigation has been confined to the study of the secondary butyl, amyl, and hexyl acetates, as it is believed that these esters hold the greatest interest for lacquer manufacturers from the standpoint of both their inherent properties and their economic position. The secondary esters used in this work were practically chemically pure mixtures of 85 per cent ester and 15 per cent corresponding alcohol. Although the process of manufacturing secondary esters permits their production as single individuals or as any blend the lacquer manufacturer may desire, it was deemed advisable to determine the properties of the pure ester mixtures. For this reason in those cases where comparisons are made between secondary and other esters, the comparisons do not necessarily offer a basis for evaluation. The boiling ranges of the esters used, which are recorded graphically in Figure 1, will bear out the statement that the three secondary esters used are evidently purer chemical compounds than the three esters used as a basis of comparison. Of the latter, butyl propionate has the widest boiling range, no doubt owing to the presence of homologous compounds. The secondary esters are colorless mobile liquids having the properties listed in Table I. of Secondary Esters SP.GR. REFRACTIVE ESTER BOILING 15.5O/ INDEX CONCN. POINT 15.5’ C. (20.0’ C.) DEXSITY Lbs. per % O c. Rd. .” 0.861 1.3915 7.17 85-88 107-114 0.863 1.4021 7.18 85-88 128-134 1.4081 7.18 85-85 146-156 0.863
T a b l e I-Properties SOLVECNT sec-Butyl acetate sec-Amyl acetate sec-Hexyl acetate
These compounds, as now commercially available, have characteristic clean ester odors quite different from the odors of normal esters. It is the general experience of those who work with secondary acetate that it does not seem to affect the lower part of the throat to the same extent as normal acetate, and consequently there is less “gagging” effect. The earlier production of secondary esters was not particularly satisfactory from the odor standpoint, but great improvement has been made in this respect and the present 1 Received April 27, 1930. Presented before the Division of Paint and Varnish Chemistry at the 79th Meeting of the -4merican Chemical Society, Atlanta, Ga., April 7 t o 11, 1930.
precautions to prevent d e hydration of secondary alcohols during esterification has led some to the erroneous idea that secondary esters are unstable. This fact should not be overlooked when running saponifications to determine the ester content. The writers’ laboratories use regular saponification methods, but care is taken that the samples are refluxed with the alcholic potash at least 4 hours before titrating. Solvent Power
It is generally conceded that the dilution ratio is a reasonably accurate measure of solvent power of an ester; therefore this test was applied to the solvents under examination. Six solutions containing 20 per cent of dissolved nitrocellulose were prepared, and the non-solvent was slowly added, with thorough mixing, as long as the precipitated nitrocellulose would redissolve. The dilution ratios were then calculated as the ratio of non-solvent by weight at the end of the experiment. The values obtained are presented in Table 11. T a b l e 11-Dilution SOLVENT
R a t i o s of Secondary Esters DILUTIOX RATIOBY WEIGHT With toluene With naphtha
sec-Butyl acetate sec-Amyl acetate sec-Hexyl acetate n-Butyl acetate %-Butylpropionate Pentacetate
The naDhtha used in these tests was the usual Detroleum distillate having a boiling range of approximateiy 80” to 130” C. These dilution ratios are plotted graphically in Figure 2, in descending order. It will be noted that practically the same order exists for both non-solvents. It will also be noted that the dilution ratio decreases with the increase in molecular weight of the solvent. As a further test of the solvent power of secondary esters, viscosity determinations were made of nitrocellulose solutions of the same concentration, in various solvents. It is generally recognized that viscosities of solutions of nitrocellulose are a measure of the solvent value of the solvents used. The viscosities of 20 per cent solutions of dry ’/zsecond R. S. nitrocellulose in the various solvents under examination were determined a t 25” C. by the means of a MacMichael viscometer. These values were determined on solutions which had been allowed to stand for approximately six weeks after being prepared. (Table 111)
