INDUSTRIAL A X D EXGINEERIiYG CHEMISTRY
1276
Vol. 23, No. 11
Degrading of Rosins Due to Heating in Air' A r t h u r R. H i t c h CHEMICAL LABORATORY, GILLICAN-CHIPLEY Co.,
I
T HAS been known for a long time that rosins will, when heated in air above their melting point, gradually increase in color, but as far as the writer is able to ascertain there is no published record of the extent to which various kinds and grades of American gum rosins will darken when heated for a definite period of time in air. In this investigation an attempt was made to determine the effect of heating in air upon gum rosins of various grades and produced by several processes, and the relative degree of effect which took place. For this work the following rosins were used: commercial American gum rosins grading X, WW,TVG, N, M, K, and I ; special extra light-colored American gum rosins grading 1OA and 8A; special vacuum-distilled rosins grading 8A, 63, 4A, and 2A; and French rosins grading N and I. The commercial American gum rosins were taken from the rosin yard of the Downing Company, Brunswick, Ga., and were representative of southeastern Georgia production. The special extra light-colored American gum rosins were obtained by distilling, under vacuum with steam, a waterclear uncontaminated pine gum freshly exuded from the tree. The grading of all rosins paler than the American X was done by means of a special set of French types. The special distilled rosins were obtained by distilling a Nancy gum rosin under very high vacuum. The French rosins were taken from a shipment received directly from the Bordeaux region in France. Methods of Procedure
In this investigation three procedures were carried out as follows: PROCEDURE 1-The various samples of rosin were crushed coarsely and put into small test tubes of sufficient size to hold 26 grams of molten rosin each. The tubes were placed in a rack and the whole put in an electric oven a t 130" C. until the rosin in each tube had entirely melted. This required 20 minutes. The different rosins were then poured as rapidly as possible into 7/8-inch (2.22-cm.) cube molds and allowed to solidify. When cold the rosin cubes were graded for color against a set of United States Government standards. Duplicates were run in each case. Standard cubes, 7 / 8 of an inch (222 c m ) thick, were cut from the original samples of rosin and graded against the United States Government standards, and the drop in grade of the heated samples noted. Duplicates were run in each case. The results given in Table IA are averages of the duplicates. PROCEDURE 2-This was carried out similar to Procedure 1 except that the heating oven was set a t a temperature of 175" C. and the rosin samples allowed to remain in the oven a t this temperature for l ' / ~hours. Averages of the results obtained by Procedure 2 are recorded in Table I B . It was found difficult to obtain accurate checks on duplicate samples by the above procedure, owing probably to the reflux action in the test tubes of the oils in the rosins. Procedure 3 was therefore tried and found to be an accurate method for obtaining relative comparisons. PROCEDURE 3-Samples of the original rosins to be compared were cut into 7/s-inch (2.22-cm.) cubes and graded against the United States Government standard types. One hundred and twenty grams of each rosin were dissolved in 120 grams of turpentine a t a temperature not above 80' C. Two 4-ounce2 oilsample bottles were filled with the turpentine-rosin solution in each case. One sample bottle in each case was set aside as typical of the particular grade of rosin employed. A set of standard solutions was thus obtained, representing the grades of rosin in 50 per cent turpentine solution ranging as follows: SA, 6A, 4A, 2A, X, \VI&', WG, X, hi, K, I, and H. In the other 1 2
Received June 20, 1931. 120-cc.
SAVANNAH,
G.4.
set of duplicate 4-ounce2 bottles filled with 50 per cent rosin solutions, each bottle was emptied into a separate 250-cc. beaker and the gross weight of the beaker and contents determined. The beakers were put into an oil bath and allowed to remain a t a temperature of 130" C. for 45 minutes, open to the atmosphere. The beakers were then withdrawn and allowed to cool. The loss of turpentine was cared for by adding enough turpentine to regain the original gross weight. The contents of each beaker were then poured back into the original 4-ounce2 bottles and their color compared to the original standard solutions in order to determine the loss of grade of the rosin due to heating. This procedure gave a relative comparison only. The standard solutions were kept in the dark to prevent bleaching and afforded an accurate color comparison. Standards of greater rosin concentration were made but were too dark in the darker grades to obtain an accurate color comparison. Standards of lower rosin concentrations were also made but found to be too light in color, especially in the higher grades, to give accurate grading. The 50 per cent rosin-turpentine solutions were found to be best. Results are shown in Table IC. Table I-Results A-ROSINS
H E A T E D I N O V E N A T 130'
Distilled gum rosin Extra light-colored American rosin Fire still rosin French rosin Steam still rosin E-ROSINS
FINAL GRADE
LOSS I N
GRADE
C. FOR 20 M I N U T E S
8A
7A
1
10A
9A
1 1 1 1
X
N N
H E A T E D I S O V E N A T 115'
ww M M
C. F O R 90 M I N U T E S
N
French rosin C-50
of Heating Rosins
ORIGINAL GRADE
ORIGINALROSIN
P E R CENT ROSIN-TURPENTINE SOLUTIONS H E A T E D A T 130" C. FOR 45 M I N U T E S
Extra: light-colored A met.ican rosin Distilled gum rosin Distilled gum rosin Distilled gum rosin Fire still rosin Fire stiil rosin Fire still rosin Fire still rosin Fire still rosin Fire still rosin Fire still rosin French rosin
8A 6A 4A 2A X
ww WG N 11
K I
I
5A X+
3 5%
WG M+
'4 3l/2 3 2 1/1 2
ww *M
K+
K I
I H H
5
2
1 1 1
S u m m a r y a n d Conclusions
Various kinds and grades of rosin have been heated in air for definite periods of time a t 130" and 175" C. in order to ascertain the comparative degree to which they will darken or degrade. Fifty per cent solutions of these rosins in turpentine have also been studied for degrading when heated a t 130" C. for 45 minutes. When various kinds of light-colored rosins, grade N or above, were heated in air a t a temperature of 130" C. for 20 minutes and then immediately cooled, a loss of one grade resulted in all samples studied, regardless of the kind of rosin used. When various kinds of light-colored rosins, grade N or above, were heated in air a t a temperature of 175" C. for 11/? hours and then quickly cooled, a loss of three grades resulted in all samples except that of the extra light-colored American gum rosin, which dropped twelve grades (from 10A to WG). This rosin contained considerably more turpentine than the other rosins studied and showed a much greater reflux action in the test tubes during the heating, which probably explains its unusual increase in color. Fifty per cent rosin-turpentine solutions, when heated a t 130" C. for 45 minutes, showed a loss of from three to five
November, 1931
1277
INDUSTRIAL AND ENGINEERING CHEMISTRY
and one-half grades for all rosins above WW and from one to two and one-half grades for all rosins from I to WG. Distilled gum rosins, in general, showed the greatest loss in grade. Extra light-colored American gum rosins (grade 8A) appeared to be no more sensitive to heat when in solution in turpentine than the ordinary commercial grades, WW and X.
The darker the original rosins were below WG, the less was the loss in grade, which was to be expected and was no doubt due to the greater amount of color required to change the grades. The commercial samples of French rosin of grades N and I when heated showed the same drop in grade as American commercial gum rosins of the same grades.
Identification of Products of Oxidation of Gas Oil by Penniman Process' Sherlock Swann, Jr., W. H. B. Howard, and E. E. Reid CHEMISTRY LABORATORY, THEJOHNSHOPKINS UNIVERSITY, BALTIMORE, MD.
from the bottom of the still ESEARCH in the oxiIn the aqueous part of a distillate resulting from an air from time to time. dation of petroleum oxidation of a Pennsylvania gas oil at 750" F. (398.9' C.) has interested many and under a pressure of 300 pounds per square inch The gas oil used to provide i n v e s t i g a t o r s Over a long (21.09 kg. per sq. cm.), the following compounds were m a t e r i a l for the work deidentified: acetaldehyde, acetone, methanol, methyl period of time on account of s c r i b e d in this paper was a the possibilities that petroacetate, dimethyl acetal, ethyl alcohol, ethyl acetate, Midcontinent distillate of the leum offers as raw material for allyl alcohol, and acetic acid. following specifications: the synthesis of industrially Specific gravity a t 60' F. (15.6' C . ) = 38.3' Be. useful products, particularly organic oxygen compounds of the aliphatic series. The studies in this field have been carried out Distillation Oil Vapor P. C.) P. C.) with a large variety of pure aliphatic hydrocarbons, or mixtures, 432 (222.2) Initial boiling point 552 (288.9) as starting materials. Both liquid- and vapor-phase oxidations Temp. 10% distilling 5S2 (305.6) 525 (273 9) Temp. 20% distilling 600 (315.6) 5% (292.2) 575 (301.7) have been tried with and without catalysts. It is evident that Temp. 30% distilling 615 (323.9) Temp. 40% distilling 631 (332.8) 592 (311.1) an entirely successful method for manufacturing organic chemiTemp, 50% djstjlljng 645 (340.6) 608 (320.0) 661 (349.4) 624 (328.9) cals from petroleum by oxidation has not been found, since Temp. 60'7 distilling 671 (355.0) 642 (338.9) Temp. 7 0 8 distilling 701 (371.7) 666 (352.2) aliphatic organic chemicals of industrial importance are, in Temp. 80% distilling general, manufactured from other raw materials. I n the cases where petroluem fractions are used, the products are The aqueous layer was the only part of the distillate investigated. obtained by means other than oxidation. The work described in this paper is the identification of Concentration of Compounds in Aqueous Layer products resulting from a liquid-phase oxidation of a gas oil I n this run thirty-seven barrels of the gas oil were used. which was carried out by Penniman (4). The process has operated successfully on a semiplant scale, using paraffin-base Sixteen barrels of nil and eight of aqueous layer were collected as distillate, but these proportions do not represent the results oils. of continuous operation. Description of Process The oil layer was washed with 135 gallons (511.7 liters) of water which was added to the aqueous layer, making a total The Penniman process consists in passing air into hydro- of 535 gallons (2027.7 liters) of an aqueous solution of organic carbons held in a still a t about 750" F. (398.9' C.) under a compounds. pressure of 300 pounds per square inch (21.09 kg. per sq. cm.) This solution was distilled by the U. S. Industrial Chemical and condensing the products which are evolved from the Company. A precise fractional distillation was not atstillhead. tempted, the distillate being cut in 7-gallon (26.5-liter) fracThe particular apparatus used for the work described in this tions for convenience in storing. At the beginning of the distillation the condenser was cooled with brine to 17' C. paper consisted of an upright cylindrical still 26 feet (7.9 meters) high and 4 feet (1.2 meters) in diameter, whose working The first two 7-gallon (26.5-liter) fractions were collected capacity was about 20 barrels [lo00 gallons (3790 liters)] of oil. under these conditions. The thermometer a t the top of the The still was equipped with heat exchangers for preheating the oil as it entered the main body of the still, and two water coils column registered room temperature. At this point the brine for controlling the reflux. The still was connected by a vapor was replaced by water, and ten more fractions were collected line through a condenser to a receiver. until the thermometer a t the top of the column registered It was found that when air was blown into the hot oil a t the 100" C., a t which point the distillation was discontinued. bottom of the still a t a rate of approximately 400 cubic feet I n all, 84 gallons (318.4 liters) of concentrated organic (11.3 cubic meters) per minute, partial oxidation took place and the temperature of the oil rose, owing to heat of reaction, from material were obtained, which was about 15 per cent by volan initial temperature of 500" to 750' F. (260' to 398.9' C.), ume of the total aqueous layer. Samples from each of the and a liquid consisting of aqueous and oil layers distilled over fractions and of the still residue were taken as material to be into the receiver. The process could be made continuous, and identified. the operating conditions maintained within narrow limits by feeding fresh oil into the still a t the proper rate, no external Identification of Compounds of Aqueous Layer heating being necessary as the heat of reaction was sufficient.
R
(0
The issuing gases contained about 1.5 per cent carbon dioxide, no oxygen, 0.5 per cent of combustibles, the rest being nitrogen. It was found necessary to draw off a certain amount of s!udge
* Received July 2, 1931.
(0
The first step in the investigation of the water distillate was the determination of the boiling-point ranges and other properties of the fractions of concentrated organic material from the water distillate. The result? are shown in Tables I and 11.
