S e p t . , 1916
T H E JOLRiVAL OF I N D C S T R I A L A N D ENGINEERING CHEMISTRY
The average rate of feeding t h e charge t o the furnace was I j kg. per hr., ranging between' I O a n d 2 0 kg. per hr. T h e best reductions were obtained when t h e furnace was kept well filled with t h e charge. T h e material used was finely pulverized barytes (98+ per cent B a S 0 4 ) (material used in paints b y W. P. Fuller C o . ) , as large quantities of t h e precipit a t e d barium sulfate could not be obtained. Six tests were made using city gas a n d one using oil for fuel. T h e main object in these tests was t o obtain a high reduction of t h e barium t o t h e water-soluble sulfide; high efficiency of t h e reducing agent was. a secondary consideration. One test '( 5-7) below, however! was conducted for t h e purpose of obtaining a higher fuel efficiency t h a n t h e others. This test showed t h a t almost as high a reduction could be obtained with much less fuel consumption. Xone of these tests furnished conclusive d a t a for calculating t h e fuel t h a t would be required for effective reduction in a commercial sized kiln. TABLEVII-REDUCTION OF COMMERCIAL PULVERIZRD BARYTES IK CEMENT KILN CAARACTBR OF C USED Pulverized Pulverized Pulv. Coke Charcoal None Coke Coke TESTNo 5-1 5-2 5-3 5-4 5-5 5-6 5-7 Per cent n charge . . . . . . . . . . . . . . . 20 15 20 16 0 18 13 RateCharge:Kg.perhr ... 10 10 15 15.2 12.5 16 18.T Gas Consumption, Cu. ft. 5.4 5.6 6.1 6.3 6.2 oil(a) 3.8
1
hrs . . . . . . . . . . . . . . . . . . 25 10 5 5 4 2 4 Av. Temp. ( " C . ) : Exit Gases . . . . . . . . . . . . . . . . 450 460 420 400 510 ... 370 4 v . Tem Middleof Kiln 850 880 880 860 900 . . . 840 Max. A m p . in Kiln (approx.) . . . . . . . . . . . . . 10.50 1050 1100 1050 llOO+ 1360 1050 4 v . per cent CO in Furnace Gases (middle).. 3.8 4.2 3.1 4.2 6.2 1.5 ANALYSES OF PRODUCT OF TEST: (b) Per cent of Total Ba: , Water-soluble. 89 87 90 89 69 84 82 Acid-soluble . . . . . . . . . . 98 97 98 98 78 96 90 Insoluble (BaSOa). . . . . 2 3 2 2 22 4 10 ( a ) Product of Test 5-5. Clinkered and unreduced Bas04 left in interior of the clinker. ( b ) Rate of oil consumption, 3 gals. per hr.
.
...
........
SUMNARY
I-The d a t a obtained in t h e five series of reduction tests outlined above indicate t h a t t h e maximum reduction of barium sulfate t o sulfide was obtained a t t h e higher temperature, i. e., around 1000' C . , a n d when t h e reductions were effected in a n indirect fired furnace, e . g., in t h e muffle furnace, 1 5 or 16 per cent carbon gave the highest fuel efficiency consistent with completeness of reduction of t h e barium t o t h e sulfide. 11-In reductions effected in a direct fired furnace ( a cement kiln, multiple hearth roasting furnace, or a shaft furnace where hydrogen, hydrocarbons, or where t h e products of combustion of t h e fuel, water a n d carbon dioxide, were brought into contact with t h e sulfate or sulfide) there were formed a larger proportion of barium compounds insoluble in water, t h a n were formed in a furnace indirectly fired. Thus, although t h e barium compounds in t h e best products from a direct fired furnace were 90 t o 95 per cent soluble in acid, yet t h e barium present as t h e watersoluble sulfide was not more t h a n 85 t o 87 per cent of t h e total. 111-By effecting t h e reductions rapidly a t high tempeiatures, i. e . , above 1000' C . , t h e prbportion
777
of these water-insoluble barium oxides a n d carbonates was less t h a n t h a t formed i n reductions a t lower t e m peratures over a longer period of time. IV-Below 750' C. t h e reduction b y carbon or reducing gases was too slow t o be considered commercial. BUREAUOF MINES, WASHINGTON
THE THERMAL DECOMPOSITION OF THE ETHANEPROPANE FRACTION FROM NATURAL GAS CONDENSATE By J.
E. ZANETTIAND E. H. LESLIE Received August 3, 1916
I n a previous paper' one of us presented t h e results obtained b y t h e thermal decomposition of t h e propane-butane fraction from natural gas condensate. As a continuation of t h a t work i t seemed desirable t o s t u d y t h e lower fraction, t h a t containing chiefly ethane a n d propane, which fraction can likewise be found on the market a n d is used mainly for lighting a n d oxygen welding. This fraction comes in cylinders under 500 t o 1000 lbs. pressure, a n d owing t o t h e great difference in boiling points between ethane and butane. -93 a n d + I O , is little contaminated b y butane. At 7 j o lbs. a n d t h e 'temperature of 2;' t h e vapor pressure of butane as calculated from Burrell a n d Robertson's formula2 Log P = -1633/T f 1.7j log T -0.01094 T 7.590 is I j 4 5 m m . , a little above 2 atmospheres. Since t h e pressure in t h e cylinder used was 7 5 0 lbs. above atmospheric, a t 25' t h e amount of butane present in t h e issuing gas would be about 4 per cent. If we consider t h a t t h e vapor pressure of t h e butane must be considerably diminished b y t h e fact t h a t it is dissolved in the liquid ethane a n d propane, t h e amount of impurity in t h e gas from this source becomes small. T h e very high pressure in some of these cylinders would indicate t h a t there is also some dissolved methane in t h e liquid ethane. The critical temperature for ethane is +34' a n d t h e critical pressure j 0 . z atmospheres. As a t 25' t h e pressures are often above t h e critical, t h e only other hydrocarbon present would be methane. This matter is dwelt on a t present as it further bears out t h e observation made b y one of us i n connection with t h e propane-butane fraction t h a t t h e a r o m a t i c h y d r o c a r b o n s obtained f r o m these f r a c t i o n s are built up from a l i p h a t i c c o m p o u n d s of lower carbon content t h a n benzene. There is no possibility of aromatics having been obtained here b y t h e splitting off of t h e benzene ring from phenyl paraffins as no such compound is known t h a t would boil a t t h e temperature of liquid ethane. under atmospheric pressure, or conversely under t h e pressure of liquid ethane a t ordinary temperature. If such compound were present in this gas its vapor pressure would be so small in comparison with t h a t of ethane a n d propane t h a t it would constitute b u t a minimal fraction of t h e issuing gas a n d it could in no way account for t h e yield of t a r obtained in these experiments--2.; cc. of t a r
+
1
2
THISJOURNAL, 8 (1916). 674. J. A m . C h e w SOL.,37, 2190.
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
778
Vol. 8, No.
c)
60
30
J0
10
10 I
I
600‘ FIG
1
65-0‘
I
I
700‘
750’
1 8W
1 850’
I
9W’
I
I
950’ lO00“c
I-DECOMPOSITION OF
ETHANE-PROPANS MIXTURE 3 0 CAT4LYZBR
(liquid) per cu. ft. of gas used. The cylinder of gas used showed a pressure of 750 a t 2 j’ which pressure remained constant, vc.ithin the limits of accuracy of the gauge used. during these experiments. Analysis of the gas showed it t o be composed chiefly of ethane and propane. No C 0 2 mas present and only less than 0 . j per cent “unsaturated.” The apparatus and method of procedure were exactly as described in t h e case of the propane-butane fraction. The gas was passed a t a measured rate through a heating chamber, t h e t a r “fog” formed precipitated electrically and samples oE t h e gas analyzed for “unsaturated” and hydrogen. The results of these analyses are plotted in Fig. I. As in t h e case of the butane-propane fraction. t h e percentage of “unsaturated” increases gradually t o a maximum in the neighborhood of 7 5 0 ’ ~ decreasing again above t h a t temperature t o a minimum a t about gjo’. The content of hydrogen increases slowly a t first, a marked increase in its rate taking place above 7jO’. The beginning of the decrease in “unsaturated” formation and of the increase in hydrogen formation are coincident, as noted in t h e case of t h e propane-butane fraction with t h e appearance of t h e t a r “fog.” The “unsaturated,” as shown by fractionation of the bromides, consisted chiefly of ethylene and propylene with small amounts of butene. The presence of butenes, which occurred in a very small amount, is doubtless due t o the decomposition of t h e butanes, which, as pointed out above, must occur in the gas in small amounts. The t a r formation occurred a t about 7 j O o . The yield was very much smaller than in the case of the butane-propane fraction, amounting t o only 2 . j cc. per cu. ft. of gas used. This might be expected from the fact t h a t the average molar weight of the gas is much smaller t h a n in t h e other fraction. How much of this aromatic formation is contributed b y t h e ethane and how much by the propane is a question which we cannot discuss a t the present moment. T h a t t h e ethane does contribute t o some extent can not be doubted since Bone and Coward1 have found it t o be t h e case, though only t o a slight extent. The very small percentage of the lower boiling aromatics and 1
J . Chem. SOC.,93, 1197.
600‘ FIG.
bSOo
700‘
11-DECOMPOSITION OF
750’
800’
85.0”
980“
QSO’
/#00%
E T H A N E - P R O P A N E M I X T I I R E , L-SING COPPER A S CATALYZER
the unusually high one of naphthalene is worthy of notice. Only about 3 cc. out of 20 cc. of t a r came over below I Z j ’ and about I cc. more below zoo’. T h e rest was naphthalene and pitch. The naphthalene formed choked t h e condenser coil several times during a run and could be noticed in large flakes a t t h e end of t h e quartz t u b e when cleaning out the apparatus between experiments. The effect of copper as catalyzer was not marked in any way. The results when plotted (Fig. 11) were, both for the “unsaturated” and for hydrogen. very similar t o those obtained without a catalyzer. The aromatic formation proceeded in much t h e same way. With iron as catalyzer (Fig. 111) the same sharp drop in t h e unsaturated content a t about 710’ was noticed as with the propane-butane fraction. the hydrogen content increasing very rapidly as t h e ‘(unsaturated” decrease. No aromatic formation was noticeable beyond the formation of a slight amount of “fog.” Much carbon was deposited on the catalyzer necessitating its frequent renewal. EXPERIMEXTAL
31ATERIAL-The material was a “liquid gas” from West Virginia in a steel cylinder under 7 j o lbs. pressure a t 2 j ’. The results of t h e explosion with oxygen are given in Table I. -4nalyses I and I1 were made a t t h e beginning, I11 and IV a t the end of the experiments presented. T h e gas had in addition 0.4 per cent “unsaturated” and about 0 . j per cent hydrogen. There was no C O z . T.4BLE I---.!NALYSIS
ANALYSIS: I Volume gas. .. . . . . . . . . . . . . . 1 2 . 3 Contraction.. , . . . . . . . . . . . . 3 4 . 4 l’olurne CO?., . . . . . , . , . . . 2 9 . 9
OF GAS
I1 12.3 34.0 29.6
111
IV
11.7 32.0 30.9
12.2 33.2 31.j
A P P A R A T U S A X D PROCEDURE-The apparatus and procedure were the same in all respects as used for t h e propane-butane fraction and their description is unnecessary. The rate of gas flow through the chamber was the same, 0 . 4 j cu. f t . per hr. G A S ANALTsIs--The gas analysis was conducted over water. For the determination of t h e “unsaturated,” bromine was used. The hydrogen was determined by combustion over copper oxide. The results are given in Table 11.
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
Sept., 1 9 1 6
779
111--The action of copper a n d iron as catalyzer has been studied. Iron prevents aromatic formation a n d favors t h e decomposition of t h e hydrocarbons into carbon a n d hydrogen. F u r t h e r work upon these topics is now in progress in this laboratory.
60
50
(10
DEPARTMENT OF CHEMISTRY, COLUMBIA UNIVERSITY N E W YORK CITY
30
THE SOLUBILITY OF LEUCITE IN SULFUROUS ACID 20
By J. SCHROEDER Received May 19, 1916,
JO
FIG.111-DECOMPOSITION O F ETHANE-PROPANE MIXTURE, USINGIRONA S CATALYZER
THE “UNSATURATED”-The gases Coming between 800 a n d 8jo’ were passed into bromine till t h e latter became colorless. After drying, 2 j cc. were distilled a n d t h e following fractions collected: 5 cc. below 1 2 7 ’ ; 9 cc., 1 2 7 - 1 3 2 ’ ; j cc., 1 3 2 - 1 4 2 ’ ; 5 cc. residue. On allowing t h e residue t o .stand a small amount of crystals separated out. Pressing from t h e mother liquor a n d crystallizing from alcohol t h e y were obtained in t h e form of small, transparent, prismatic crystals melting a t I 16 O . AROMATICS-The t a r obtained was quite fluid a n d after filtering showed a specific gravity of 1 . 0 7 5 . Twenty cc. were distilled i n a small flask. Three Above cc. came over below 1 2 5 O , I cc. below 200‘. t h a t , naphthalene began t o solidify in t h e condenser. T h e distillation was continued under reduced pressure, about 8 cc. coming over a n d solidifying t o a yellowish
TABLE 11-ANALYSIS OF GASEOUS PRODUCTS WITH AND WITHOUT CATALYZER No CATALYZER c _ _ -
UnsatH T p p . urated Per C. Per cent cent 580 630 680 730 775 800 825 850 875 925 975
2.8 7.8 18.9 28.0 31.0 27.9 24.2 18.0 15.2 7.2 3.0
3.4 5.8
8.8
15.3 19.6 24.0 26.1 34.0 35.3 46.4 58.8
COPPERAS CATALYZERIRONAS CATALYZER ,-&_-
7
UnsatH Temp. urated Per C. Per cent cent 580 630 680 730 750 775 800 825 875 925 975
1.7 8.1 19.9 30.2 31.5 30.0 25.4 18.2 11.0 2.6 2.8
1.3 4.0 10.1 15.6
.... , ...
21.2 33.5 42.0 54.0 62.7
F
A
-
-
-
UnsatH Temp. urated Per C. Per cent cent 580 630 660 710 720 730 770 815 875
6.1 14.9 22.0 31.2 14.2 8.3 5.7 0.3 0.0
3.6 7.9 10.9 22.5 40.9 53.4 60.3 65.8 66.6
mass of crystals. On crystallizing from alcohol t h e y gave a melting point of 79’. T h e residue solidified in t h e flask, forming a black pitch with a strong odor resembling anthracene. Nitrobenzole was obtained from t h e fraction boiling below I 2 j ’, b u t no nitrotoluols could be obtained from i t . SUMMARY
I-It has been shown t h a t mixtures of ethane a n d propane decompose a t high temperatures, giving ethylene, propylene a n d other “unsaturated,” hydrogen a n d aromatics. 11-The percentage of “unsaturated” increases with increasing temperatures to a maximum in t h e neighborhood of 7 jo’, then decreases with increasing temperatures. The aromatic formation begins a t about 7jo’ C. a n d is coincident with a n increase i n t h e r a t e of hydrogen formation.
Leucite belongs t o a class of potash-carrying silicates t h a t has received considerable attention as a possible source of p0tash.l It is a metasilicate of aluminum a n d potassium having t h e formula KA1(SiO3)? or KzOlA12O3.4Si02. On this basis its theoretical composition is: Silica, 55.0 per cent; alumina, 2 3 . 5 per cent; and potash, 2 1 . 5 per cent. From this it is evident t h a t in potash content it is one of t h e richest minerals known. Leucite occurs in t h e more recent volcanic rocks a s embedded crystals, grains, or aggregates of grains. I t is identified especially with t h e lavas of Wt. Vesuvius a n d other localities in Italy, though its occurrence is b y no k e a n s restricted t o these localities. I n N o r t h America it is found in Lower California, New Jersey, Arkansas, British Columbia, a n d especially in several places outlying from t h e Rocky hlountains, notably t h e Leucite Hills of Wyoming. These hills are situated in southwestern Wyoming from I O t o I j miles n o r t h of Bitter Creek, which is a n eastern tributary of Green River a n d is followed b y t h e Union Pacific Railroad. T h e Leucite Hills consist of a number of conical peaks of lava protruded through t h e beds of rocks forming t h e plateau of t h e surrounding neighborhood. Each peak consists of a lava sheet, presenting a n a b r u p t wall from 50 t o I j o ft. high. These deposits are described i n detail b y Schultz and CrossS2 T h e leucite rocks of this neighborhood are of a yellowish gray color a n d often, if not always, of a pronounced cellular structure. The porosity, however, does not reach t h e pumice stage. The leucite inclusions are microscopic crystals 0.03j m m . in diameter. T h e chemical composition of t h e rocks a5 indicated b y t h e analyses of t h e United States Geological Survey is about as follows: Silica, 53 per cent; alumina, 1 1 per cent; ferric oxide, 3 per cent; magnesia, 6.5 per cent; lime, 4. j per cent; soda. I . j per cent; a n d potash, 1 1 per cent.2 T o ascertain t h e possibility of recovering this potash b y leaching, solubility tests were made with sulfurous acid. A sample of leucite rock from Batuku, Celebes, E a s t Indies, containing 9 per cent potash, was obtained through t h e courtesy of Dr. J. P. Iddings of t h e United States National Museum. This was ground a n d separated, b y sifting, into portions of different fineness a n d these portions shaken over night with a solution of sulfurous acid. T h e latter was prepared b y passing sulfur dioxide through water t o approximate satura1 2
Cushman and Coggeshall, THISJOURNAL, 4 (1912), 821. U S Geol Survey, Bull 611.
