682
T H E JOURNAL OF INDUSTRIAL A N D ENGINEERING CHEMISTRY
Portland cement-magnesite mixtures had a bad color, absorbed a large amount of water and shrank on drying; in general, they resembled more closely the cement t h a n t h e composition floors, Plaster of Paris gave compositions which dusted badly, lacked hardness. and set too quickly. Likewise, gypsum which had been heated t o 400’ C. gave samples which set too quickly and were no harder t h a n t h e samples made from ordinary plaster of Paris. Samples prepared with t h e addition of a cement made of molecular quantities of magnesia, magnesium sulfate and plaster of Paris, said t o form a white, quick-setting strong cement, also set quite quickly and were very hard, b u t the material rubbed off easily. I n the effort t o find the material which would be a suitable binder for fillers or one t h a t could well be used as a surface material, a number of experiments were made with asphalt, “Redmanite,” gelatin-glycerol mixtures, “Galalith,” viscose and vegetable resins. N o results of commercial importance were obtained. Finally, it may be mentioned t h a t in 1912 a section of a floor, 1 2 ft. b y 16 f t . , was laid in a factory room which was in constant use; this floor was p u t down in t h e center of the room and was subjected t o a great deal of wear. It was scrubbed every day with alkaline cleansing material and hot water. ,4t t h e end of 7 mos. t h e floor was reported t o be entirely set arid as good as terrazo. Further experience warranted t h e conclusion t h a t t h e flooring, t h e composition of which was identical with t h a t laid in 1911, noted above, was more desirable t h a n a floor constructed from a hard, nonpliable, highly conductive substance like terrazo. PRACTICAL CONCLUSIONS
The successful working of composition flooring of the type considered does not depend entirely upon t h e chemical composition, b u t also upon the mechanical manipulation. The exercise of constant care is necessary in laying t h e floor and consequently only skilled workmen should be employed in practice; in particular, composition flooring should not be over-troweled, and t h e necessary precautions should always be taken in laying floors over surfaces which are subjected t o heat. T h e composition should be dependent upon the uses t o which t h e floor is t o be p u t ; b u t a formula is given which, when suitably modified along the lines indicated and expertly used, will comply with the conditions usually encountered in practice. ~ I E L L OIASTITUTE N OF I A D U S T R RESEARCH I~L PITTSBCRGH
EMULSIFIED OR CUT PETROCEUM By CHARLES K. FRANCIS Received April 25, 1916
.I large quantity of emulsified or “cut” oil is produced from many wells in t h e Cushing pool of Oklahoma. Some of t h e pipe lines have refused t o accept this oil claiming t h a t it contains too much sediment which does not separate well on standing, and the refineries complain t h a t it has a bad action on t h e stills. I t is the author’s intention t o present part of t h e details of a n investigation which was undertaken
VOI. 8 .
NO.8
for t h e purpose of determining the nature of these emulsified oils and t o suggest methods for properly treating them. The oil selected for special study was produced from t h e Bartlesville sand and had a specific gravity of 0.8279 (39.1’Be.). After I O min. in the centrifuge, which made 14jo r. p. m . , t h e total residue appeared t o be 1.6 per cent. Subsequent examinations proved t h a t this residue contained b u t a very small amount of solid matter, t h e larger part being water. The solid substance consisted of minute particles which, when measured under t h e microscope, were found t o vary from l , l q o 0 mm. t o l / 2 0 mm. in diameter, or from 98 millionths t o 19 ten-thousandths of a n inch. A portion of t h e oil was filtered and washed with casing head gasoline, gravity 84’ BC. This treatment left a n insoluble residue, equivalent t o 0 . j O j per cent, consisting of 0.277 per cent nonvolatile mineral matter and 0 . 2 2 8 per cent organic and volatile substances. AXALYSIS05% RESIDUEFROM C U T OIL FROM THE BARTLESVILLE SANU SUBSTANCE Per cent Based on Residue Oil Sand.. . . . . . . . . . . . . . . . . . 2 . 6 9 0.007 0.037 Gypsum.. . . . . . . . . . . . . . . 12.35 0.019 Calcium chloride.. . . . . . . I ,60 0 002 Magnesium chloride.. . . . 0.91 Salt . . . . . . . . . . . . . . . . . . . 7%45 0.209 ~
Total.. . . . . . . . . . . 100.00
__
0.274
There is b u t a trace, less t h a n 0 . j per cent, of water in this oil, SO t h e quantity of salts present is more t h a n sufficient t o saturate it. The common method for testing a n oil for suspended matter, a t the well or while in t h e stock t a n k , is t o place some of it on a piece of glass a n d then look through it while holding i t t o t h e light. An experienced gauger can b y means of this simple method quite accurately estimate t h e quality of an ordinary oil, b u t t h e test gives no positive information concerning t h e nature of any sediment which may be detected. Sometimes particles of water and gas may be thought t o be mineral matter; indeed, such mixtures often have t h e appearance of suspended mineral matter. Oils holding water and gas in suspension soon settle. whereas oils truly cut or emulsified have been allowed t o stand for weeks without clearing satisfactorily. The nature of any substance held in suspension in petroleum may be quickly determined b y means of a microscope. Good oils are represented in Figs. I and 11-the predominating objects are water and gas; b u t , under the same magnification of t h e microscope, a cut oil, t h a t is, emulsified, will appear as shown in Figs. I11 and 11’. The mineral matter in these oils is chiefiy common salt. I n Pig. V several of the more prominent crystals of this substance are indicated b y a n X beneath. This photograph illustrates how the material shown in Fig. TI1 changed through evaporation while exposed t o the air for a few minutes, and brings out t h e character of t h e mineral matter. In the lower right-hand corrier of t h e photograph a crystal of salt may be seen encased in a globule of water. The analysis and t h e microscopical examination confirm each other as t o t h e nature of t h e solid material present in this oil. T h e amount of mineral residue,
Aug., 1916
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
per cent, is surprisingly small when compared with t h e 1.6 per cent found by means of the centrifuge, which is so extensively used for making this test. A residue of 0 . 2 per cent would not be sufficient t o cause the oil to be rejected b y the average gauger representing t h e pipe line, h u t t h e chemical properties of the salts are very serious if their behavior when 0.2
683
Many experiments have been made t o devise a method which will cause emulsified petroleum t o separate. A patented process has been describedl b y A. T. Beasley which is being used in California, consisting of a combination of electricity and hot water. "The action of t h e electricity is t o create a very strong electrostatic field in which the infinitely small particles
~
J PTC. I
x
m.3
c~~~ nIL--SHowNc
heated is considered. The well-known action of chlorides, especially magnesium chloride, in boiler waters may be used t o illustrate this: h4gCL H20 = MgO 2HCI Hydrochloric acid (HCI) may be formed in the same way, certainly t o some extent, when calcium chloride is heated. T h e scale forming and corrosive action
+
+
Pic. I1 X ,372 WATSR A N " G A S
of water will be formed into chains from electrode t o electrode. If t h e voltage be sufficiently high the fine films of nonconducting oil between t h e water particles will be punctured, bringing the entire chain together in t h e form of one comparatively large drop. This drop is now free water and is deposited." The system is composed of three units-a water trap, .
F.
Io
,..-.+ ..
@& . :
. , ..
'
6
4
. "{*..
?c;; Y
. e>..
..
4'
.
0 .
i" I
.f -r> P l C . I V x 372 w,LIBP A N D M ~ N ~ R A lA ~ rLr s a of the group of salts reported in the analysis has often a n electrical treater, and an automatic oil and water been a grave matter in steam boilers; so then, oils separator. T h e oil in t h e treater is maintained a t containing them would in a short time seriously a temperature which may vary with the viscosity of injure the condensing tubes and t h e stills in the re- t h e oil and t h e weather conditions, b y means of n finery. On a basis of 4000 bbls. of oil treated each steam coil. No doubt t h e heat has much t o do with day, the amount of salt accumulating daily would be t h e breaking down of the oil globule when this method 2400 Ibs., about 240 Ibs. of calcium chloride and 24 is used. Ibs. of magnesium chloride. I Oil Ape. April. 1911; see Redwoorl. 1. 318. FlG.
I11
x
83
BrnYLSIPlED oIL-sIIOwYINo
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
684
In t h e Oklahoma fields it is t h e common practice t o treat t h e emulsified oils with water, the temperature of which is maintained a t 80 t o 125' F., depending upon t h e quality of t h e oil a n d t h e weather conditions. This treatment is conducted in a t a n k (Fig. V I ) , placed between t h e flow t a n k and t h e stock tanks. I t is simply one unit in addition t o t h e usual
Vol. 8, No. 8
pansion causes t h e breaking of the oil film; then the fine particles of salt are dissolved by t h e extra water present and carried t o t h e bottom of t h e stock tank within a short time after t h e treated oil is delivered there. Some salt is removed by t h e water during the treatment, consequently t h e water in t h e treating t a n k should be renewed occasionally or it may become saturated with t h e salt and have no action on t h e oil being passed through it. OKLAHOMA AORICTILTURALA N D MZCHANICAL COLLBOB STILLWATER
THE DECOMPOSITION OF HYDROCARBONS AND THE INFLUENCE OF HYDROGEN IN CARBURETED WATER GAS MANUFACTURE By M. C. WHITAXERA N D E. H. LHs~la Received June I , 1916
(Conduded from our previous issue) T H E DECOMPOSITION O F P A R A F F I N H Y D R O C A R B O N S
equipment on a lease. The illustration indicates all the essential features of this method. The flow of oil is continuous, being controlled so t h a t a constant stream is distributed upward through t h e warm water, passing out t o t h e stock tanks. Many samples of crude oil from t h e Cushing field, which have been examined here, have had an initial boiling point of 130' F.; hence this treatment, even
n
54s
r 4
. II
I
I
lL-=lO!===-
.
I
I
L Fie. V I - T a a r n r ; ~ T ~ v wFOE B M U L S I P IO C n~
when conducted a t the higher temperature, should not seriously affect t h e gravity of the oil. There is a large quantity of salt in some wells in t h e Cushing pool. and it is this substance, as noted above, which is mainly t h e cause of t h e emulsified oils. In the warm water treatment, the slight ex-
Figs, g to 14 indicate t h e proportions of t h e various components obtained when t h e oil was decomposed alone. Fig. 9 shows t h a t at 621' C. t h e gases are composed of about I O per cent hydrogen a t all oil rates above 5 cc. per minute. The reactions which contribute t o t h e formation of this hydrogen, named in order of their probable importance, are: (I) CiHs CpH, H, (2) C,H, T ,C2Ha H* (3) CHd -+ C 2H2 ( 4 ) C2H2 --+ Z C H? The conditions are those' which are known t o he favorable t o t h e condensation of acetylene. The methods for t h e determination of acetylene are so unsatisfactory t h a t no effort was made t o determine t h e proportion of acetylene in t h e gases formed. Other investigators have found the amount small in similar experiments. The rapid increase in t h e proportion of hydrogen with decreasing oil rates below 5 cc. per minute is thought t o he due t o a marked increase in t h e extent t o which reactions (3) and ( 4 ) take place. A t 7 2 3 ' C. only a slightly larger proportion of hydrogen is formed than a t 621O C., indicating t h a t reactions ( 3 ) and ( 4 ) do not become rapid u p to.this temperature. A very marked increase in the hydrogen production takes place when t h e temperature is elevated t o 825' C., chiefly due t o a n increase in reactions (3) and ( 4 ) . a s is evidenced by t h e very larae amount of free carbon which was iiberated and which tended t o stop t h e furnace tube. N o trouble was occasioned by this carbon deposition a t 621 or 7 2 3 ' C. Fig. 1 1 shows t h e proportion of illuminants, or components removed by 2 0 per cent fuming sulfuric acid, in t h e oil gases made a t t h e three temperatures. The percentage of illuminants is highest in t h e gases made a t 621' C. a t all oil rates, and remains practically constant a t 5 2 per cent a t all oil rates above I O cc. per minute. The proportion of illuminants formed a t 7 2 3 ' C. is higher than a t 825' C. except possibly a t high oil rates. I t is thought, however, t h a t t h e proportion of illuminants a t high oil rates would not he greatly different a t any of these temperatures.
+ + + +
