I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
1146
TABLEl T I I I .
Boiling Range of Fraction,
COMPARISON O F CYCLOP.4RAFFIN CONTENT O F
FRACTIONS
cs AND HEAVIER
Vol. 44, No. 5
for dicycloparaffins produced a laigc change in the refractivity interccpt determination for paraffins.
(By mass spectrometer analysis and refractivity intercept method) Volume % of Saphtha Uncorrected Corrected6 MonocyoloDicycloCycloCycloParaffins paraffins paraffinsa Paraffins paraffins Paraffins paraffins
LITERATURE CITED
(1) Bell, M. F., A n a l . Chem., 22, 1005 F. (1950). (2) Brown, R. A , , Zbid., 23, 430 (1951). REFRACTIVITY I N T E R C E P T hf E T H O D MAE^ SPECTROYETER ANALYSIS (3) Brown, R. A,, Taylor, R. C., Mel56 44 0 57 43 57 43 270-306 polder, F. W., and Young, W. S., 306-358 55 40 5 44 56 54 46 Zbid., 20, 5 (1948). 54 40 6 33 67 46 54 358-391 391 and higher 53 38 9 25 75 45 55 14) Cady, W. E., AIarschner, R. F.,and Cropper, W . P., payer presented a t a M a y also include small amounts of tricycloparaffins. b Refractivity intercept of t h e cycloparaffins corrected for the presence of dicycloparaffins indicated 119th Meeting, AM. CHEM.SOC., in mass spectrometer analysis. Cleveland, Ohio, April 1951. ( 5 ) Glaspow. A. 11.. Willinpham. C. B.. a n d Rossini, F. D.; IND. ENG. CHEM.,41, 2282 (1949). (6) Whir, B. J., J. Reseaich Natl. Bur. Standards, 34, 435 (1945). fractions. -4serious discrepancy is shown in Table VI11 between (7) Noller, C. R., and Barusch, h.1. R., IND. ENG.C H m f . , API'AL. ED., the paraffin-cycloparaffin splits in the C Qand heavier distillate 14, 907 (1942). fractions boiling above 306" F. as calculated by the mass spec(8) Podbielniak, W. J., Zbid., 13, 639 (1941). (9) Rampton, H. C., J . Znst. Petroleum, 35, 42 (1949). trometer and refractivity intercept methods. I t is believed t h a t (10) Starr, C. E., Tilton, J. A , , and Hockberger, W. G., END. KNG. much of this difference is due t o the presence of dicycloparaffins CREM.,39, 195 (1947). and possibly tricycloparaffins which are shown in Table VI11 t o (11) Taylor, W. J., Wagman, D. D., Williams, M. G., Pitzer, K. S., as high as 9% in one fraction. The data of Ward and Kurta ( 1 2 ) and Rossini, F. D., J . Rrseurch N a t l . Bur. S t a d w d s , 37, 95 (1946). show that the difference in refractivity intercept between paraffins (12) Ward, A. L., and Kurts. 8. S.,IND. ENG.CHEY.,.ANAL. ED.,20, and dicycloparaffins is several times greater than t h a t between 559 (1938).
paraffins and monocycloparaffins. By correcting the value of the refractivity intercept of cycloparaffins for the presence of dicycloparaffins, a fair agreement was reached between the two methods of analysis for paraffin content. Exact agreement can hardly be expected since a small deviation in the mass spectrometer analysis
RECEIVED for review May 2, 1951. -4CCEPTED December 31, 1961. Presented as p a r t of t h e Symposium cn Composition of Petroleum and Its Hydrocarbon Derivatives presented before t h e Division of Petroleum Cliernistry at t h e 119th Meeting of the . 4 ~ E ~ l c . 4C3H E Y I C A I . EOCIETY, Cleveland, Ohio, April 1951.
Solubility of Hydrogen, Oxygen, Nitrogen, and Helium in Water AT ELEVATED TEMPERATURES H . A. PRAY, C. E. SCHWEICKERT, AND B. H . R/IIYVIvICH1 Battelle iMemorial I n s t i t u t e , Columbus 1 , Ohio
THE 1
i n c i d n g application of high temperaturesand pressures
to various processes has made a knowledge of the solubilities
of compressed gases in water necessary for purpoises of engineering design. A survey of the literature has revealed that considerable data are available on the solubilities of gases under partial pressures of more than 25 atmospheres and a t relatively low temperatures. Data in the region from about 5 to about 25 atmospheres and from about 125' F. t o temperatures near the critical point of water are very meager and incomplete. A determination of the solubilities of oxygen, hydrogen, helium, and nitrogen in water a t temperatures from 128" to 650" F. and a t presmres up to about 500 pounds per square inch absolute was,therefore, undertaken. EXPERIMENTAL PROCEDURE. For determining the solubilities of gases in water, the apparatus shown in the schematic diagram (Figure 1) was used. A typical example of the use of this apparatus is as follows: Valves A and B are closed, valves C and D are opened, and the 3-liter bomb contained in the rocking autoclave, E, is evacuated by means of the vamum pump, F. Valve A is then opened and about 1500 ml. of distilled water are admitted to the bomb from 1
Present address, Naval Ordnance Testing Station, Inyokern, Calif.
INLET
Figure 1. Diagram of Solrability Apparatus
May 1952
INDUSTRIAL AND ENGINEERING CHEMISTRY
1147
MRTIAL PRESSURE, P S I A
Figure 4. Figure 2.
Solubility of Oxygen i n Water with Varying Pressure
0. Authors
A . Froblioh
e t 01.
Solubility of Hydrogen in Water with Varying Pressure A. B. C.
Ipatieffand Teodorovish (2)
Authom Wiebe and Gaddy (3)
(I)
TEMPERATURE,
Figure 3.
Solubility of Oxygen in Water with Varying Temperature
Figure 5.
OF.
Solubility of Hydrogen in Water with Varying Temperature
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
1143
PARTIAL
Figure 6.
Figure 8.
Solubility of Helium i n Water with Varying Pressure A.
0.
Figure 7.
PRESSURE, PS.1 A
Vol. 44, No. 5
Solubility of Nitrogen in Water with Varying Pressure A.
R.
Frohlich et QZ. ( I ) Wiebe and Gaddr ( 5 )
Wiebe and Gaddy ( 4 ) Authors
Solubility of Helium i n Water with Varying Temperature
the water buret, G. Valve A is then closed and the system is evacuated to a pressure corresponding to the saturated vapor pressure of water a t room temperature. Valve C is then closed and the rocking autoclave, which is heated electrically, is set in motion and brought to some predetermined temperature. The
Figure 9.
Solubility of Nitrogen i n Water with Varying Temperature
temperature of the autoclave is controlled by a Leeds & Northrup Micromax temperature controller and recorder, N . The teluperature inside of the bomb is determined by means of a chromelalumel thermocouple inserted in a stainless steel well inside of the bomb. The electromotive force of the thermocouple i s deter-
May 1952
INDUSTRIAL AND ENGINEERING CHEMISTRY
is closed and the sampling line side of valve B is opened. About 25 ml. of water containing an amount of gas which depends on the partial pressure and water temperature are then collected over mercury in the measuring buret, L, which is surrounded by a water thermostat. The gas and water volumes are then measured and the needle valve, M , is opened. This allows the water that remained in the sample line, because of capillary action, to drain into the buret. After the total amount of gas and water have been determined, dry air is blown in through the inlet a t N and allowed to escape through M . This procedure dries all of the sampling lines in preparation for the next sample. The solubility of the gas in water is then determined by reducing the gas and water volumes, as found in the measuring buret, to cubic centimeters of gas a t normal temperature and pressure per gram of water,
TABLE I. SOLUBILITY OF OXYGEN IN WATER OZ P a r t i d Pressure Lb./'Sq. In.' Abs. Souroe
Cc. OP/G. of Water
Average
Maximum Deviation
Probable Error
77' F. 140 295 370
(1)
.. ,. ..
.. .. ..
0.28 0.56 0.70
, .
..
... ... ...
0.15 0.31 0.46
0.01 0.00 0.01
t0.01to -0.01 f O . O O to -0.00
0.18 0.28
0.01 0.01
f O . O 1 to -0.01 $0.02 to -0.01
0.64
0.04
f 0 . 3 5 t o -0.17
0.91 1.35
0.02 0.06
+0.01 to -0.02 t 0 . 3 3 to -0.26
1 71
0.03
+0.09 t o -0.17
0.63 0 . 6 0 0 . 6 9 0.62 ., 1.39 1 45 1 . 4 3 2.22 2:14 2.34 2:02 2:24 650' F.
0.63 1.42 2.19
0.01 0.02 0.04
-I-0.06 to -0.03 f0.03 to -0.03 + 0 . 1 5 t o -0.17
1.22 1.85 2.29 2.96 2.56 2.99
0.01
+0.04to -0.05
..
325' F. 200 loo 300
Authors Authors Authors
.
0 . 14 0 . 1 6 0 . 16 0.31 0.31 0.31 0.47 0.47 0 . 4 5
.. .. ..
t 0 . 0 1 to -0.01
400' F. 100 150
Authors Authors
0.19 0 . 1 8 0 . 1 7 0.30 0.28 0.27
100
Authors
200 300
Authors Authors
0 . 6 8 0.63 0.47 0 . 9 9 0.92 0 . 8 9 1.52 1.68 1.37 1.24 1.80 1 . 6 3
.. .. 5000 P.
400
Authors
0 . 6 2 0.51 0.69 0 . 5 6 0 91 1.18 1:OQ
..
t .
1:70
.. .. .. , .
..
600' P.
loo 300 104 175 205 280 2 89 309 a A t standrsrd
Authors Authors Authors
Authors 1.17 Authors 1.85 Authors 2.39 Authors 2.96 Authors 2.51 Authors 2.99 temperature and
1.26
..
2128
2:19
2:61
, ,
, .
..
..
.,
..
.. .. ,
.
.. .. .. . .
.
0:03
+o. lo'to' -0.10
0:03
+O.O5'to' - 0 . 0 5
EXPERIMENTAL RESULTS
...
..
The present experimental results and certain previously reported data for the solubility of hydrogen, oxygen, helium,
pressure.
mined with the aid of a Leeds & Northrup Semi-Precision potentiometer, H . The thermocouple was calibrated against the melting points of metals obtained from the Bureau of Standards for that purpose. The temperature during any one run is held constant to within 2" F. After the temperature inside of the bomb reaches the deRired temperature, the outlet valve, A , leading to the mercury U-tube is opened and the saturated vapor pressure of water is determined with the deadweight gage, J. The mercury U-tube isolates the material in the bomb from the oil in the dead-weight gage. The pressure, as determined by the dead-weight gage, is checked against the vapor pressure of water given in the steam tables. The gas under investigation is then admitted to the bomb by opening one side of valve C. The pressure is allowed to build up to slightly over the desired partial preseure at which the solubility of the gaa is to be determined. Valve C is then closed and the system is allowed to come to equilibrium. This is done by keeping the autoclave in motion overnight at the desired temperature. The temperature is then readjusted, if necessary, and the pressure of the gas is readjusted to ita desired partial preasure. When the temperature and pressure have remained constant for about 2 hours, a sample is taken. This is done by circulating ice water through the condenser, K , and then opening the purging line side of valve B. After about 25 ml. of water have been taken off,the purging line valve
1149
TABLE11. SOLUBILITY OF HYDROGEN IN WATER Ha Partial Pressure Lb./Sq. In: Abs. Source
CO.Hia/G. of Water 75' F.
300 367 200 300 350
(8) Authors Authors Authors
..
..
. .. ,
.. ..
.. ..
..
.
,
125°F. ' ' 0.38 0.32 0 . 3 2 0 . 3 1 0 . 3 7 0.30 0.41 0.42 0.40 0 . 4 1 0 . 4 7 0 . 4 4 0 . 4 4 0.44 0.'46
.:'
Average
Probable Error
Maximum Deviation
0.01 0.00 0.01
+ 0 . 0 5 to - 0 . 0 3 + O . O l to -0.01 +0.02 t o - 0 . 0 1
0.32 0.44 0.33 0.41 0.45
300' F. 100 200 300 375 500
(3 h (8)
(8)
::. . ::. . :. :. :. .: :: : ' .. ..
.. .
,
.
, , ,
..
,
.,
,
,
,
, ,
.,
,
,
0.13 0.28 0.40 0.52 0.70
345' F. 100 200 300 375 500
0.15 0.30 0.43 0.66 0.75
.. ..
390" F. 100 200 300 376
0.18 0.34 0.52 0.68
.. ..
435O F.
100
(8)
200 300 375 500
(a) (8)
100 200 300
Authors Authors Authors
yj
. .. . . . , .. . , , , ., . . .. .. , , .. . . .. .. . . , . . . . .. . . . . . , ,. . . , , .. , . , . ,
500' F. 0 . 4 5 0 . 3 8 0 . 3 1 0.32 . . 0 . 9 8 0.91 0.84 0.92 , , 1.24 1.23 1.28 ,,
100 200 300
Authors Authors Authors
0.39 0.91 1.25
..
600" 0 . 6 8 0.61 0.68 0.62
..
.
. .
.. ..
0.02 0.10
..
+0.06 to -0.08
4-0.07 to -0.09
0.01
+0.03t0 - 0 . 0 2
0.01
+0.03 to -0.04 +0.09 to - 0 . 0 8 t0.05 to -0.03
F.
1 . 2 4 1.31 1.34 1.41 . . 2.05 2.06 1.98 1.94 . 650' F.
100 Authors .. .. ., .. Authors . .. . . .. 115 Authors ,. .. ., .. 120 Authors . .. .. ,. 125 At standard temperature and pressnre.
0.22 0.49 0.75 0 94 1.26
.. .. , . ..
,
.
., .,
, ,
.. ,
,
.
0.65 1.32 2.01 1.40 1.63 1.68 1.74
0.02 0.02
11.50
INDUSTRIAL AND ENGINEERING CHEMISTRY , TABLE IT'. h'l Partial Pressure, Lb./Sq. In. Abs.
Source
294 367
(1) (6)
588
(1)
367
(6)
Vol. 44, No, 5
SOL~BILITY O F NITROGEN x i i WATER Cc.
of Water dverage
Nna/G.
.,
Maximum Deviation
.. .. ..
... .,. ..
F.
77'
. . . . . . . . . . . .
.,
Probable Error
0.28 0.35 0.55
..
1220 F.
. . . . . .
0.27
167" F. 367
(6)
. . . . . .
...
0.25
2120 F. 3 67
(6)
. . . . . .
0.26
, .
0.44 1.24
0.01 0.02
5 0 03 t o -0.03 +O.lO to -0.06
0.56
0.55
0.00
+O 01 to - 0 01
2:26
2:32
0.'02
+0,08'tb'-0,06
150 400
Authors -4uthors
5000 F. 0.47 0.43 0.41 1.21 1 . 3 4 1.18
150 300 400
Authors Authors Authors
0.56 0.54 1.56 2.40 2:29
600' F.
standard temperature and pressure.
6
TABLE v.
HENRY'S
Temp.,
E".
Oxygen
32 75 77 122 125
11:o
6:56
..
.. ,.
8 :7 2
..
212 300 325 346 390 400 435 500 GOO 650
and nitrogen are listed in Tables I, 11, 111, and IV, and are also shown graphically in Figures 1 to 9. The probable error shown in column 5 of tjhe tables was calculated from the usual formula
..
8:oO I
.
.. ..
9.30 Si34 6.98
6: 2177 1.74 1.22
x,FOR
Gas, K X 10-6 Hydrogen Helium
,.
167
Figure 10. Solubility- Constant versus Temperature near the Critical Point
LAW COSSTANT,
4:98 2.90
1.86 0.885
VARIOUSG a s ~ s a Nitrogen
20.40
...
2 i 75
:
13:io
, . .
16.M
IS: io ...
...
lS:& 18.00
8.04
...
...
...
3:56 2.20
4:18
...
2.38
...
2,
a K = where PA = partial Pressure of gas, A , in pounds per 3 ~ u a r 0 NA inch and N A = mole fraction of gas, A , in solution.
.-." LA' -\in(n. - 1) I
probable error = 0.6745
TABLE 1x1. He Partial Pressure, Lh./ Sq. In. Abs.
SOLUBILITY O F HELIUM IX WATER
Cc. He"/G. of Water-
Source
Average
Probable Error
;\laximum Deviation
0.23
..
...
0.22
I .
32' F. 367
(4)
. . . . . . . . . . 77' F.
367
(4)
. . . . . . . . . .
...
167' F. 367
(4)
. . . . . . . . . .
0.24
325' F. 100 200 300
Authors Authors huthors
0.20 0.31 0.39
0 . 2 0 0.19 0.31 0.30 0.37 0.37
100
Authors Authors Authors
0.42 0.58 1.36 1.15 1.31 1.74
0.38 0.60 0.13 1.07 1 41 1:64
200 300 400 500
Z .
Authors Author8
200 300 400
Authors Authors 'Authors
500
Authors
, .. .. ,
.
,
.
,, ,
.
600' F. 0.42 0 . 3 8 0.40 0.64 0.59 0 . 8 2 0.81 0:87 0.87 1 . 0 3 1 . 0 3 1 32 1 . 4 7 1:86 1.70 1:85
0.92 0 . 9 5 0.89 1.64 1.61 1.70 2.44 2.37 2.59 2.84 2.37 2.72 3 1 2 8:07
.kt standard temperature and preeaure.
0.20 0.31 0.38
0.00 0.00 0.00
i O . 0 0 to -0.01 i o . 0 0 t o -0.01 -0.01 to - 0 . 0 1
0.40 0.60
0.01
+ 0 . 0 2 to - 0 . 0 2 i O . 0 4 to -0.03
0:99 1.38 1.76
0.03 0.03
0:05
+0.37 tb'10.37 +0.09 t o -0.07 + 0 . 1 0 t o -0.12
0.01
600' F. 0.93 1.66 2.31
,,
,.
0.92 1.66
0.01
0.01
LO.03 to -0 04 +O.O4to -0 05
3:i3
2:82
2.'49 2.99
0:05 0 06
+0.35'tb'-0.18 + 0 . 2 3 t o -0.27
,.
which gives an indication of the precision of the data, where Z A = the arithmetical sum of the deviations and n = the number of measurements. It is interesting to observe from the tables and graphs that the solubilities of these gases in water increase Kith increasing temperature, in the high temperature range, where= it has been estahlished that at temperatures in the region from about 20" to ZOO" F. the solubilities decrease with increasing temperature. Within experimental accuracy, the solubilities of hydrogen, oxygen, helium, and nitrogen appear to be linear functions of pressure over the range investigated. The resulting straight lines (Figures 2, 4, 6, and 8) show that the solubilities in question follow Henry's law and may be predicted within the limits of engineering accuracy over B fairly wide range of temperatures and pressures from the Henry's law constants, which are listed io Table V.
May 1952
INDUSTRIAL AND ENGINEERING CHEMISTRY
At and above the critical temperature of water (705" F.), the gases must be infinitely soluble and the Henry's law constant must become relatively very small. The constants for the four gases in the region of the critical temperature are plotted in Figure 10. It is apparent that in this region the solubility constants tend to converge and approach very low values at 705" F., thus indicating that the data are consistent with the fact that the gases are miscible in all proportions at the critical point for water.
11.51
LITERATURE CITED
(1) Frolich et al., IND. ENG.CHEM.,23, 548 (1931). (2) Ipatieff and Teodorovish, J. Gem. Chem. (U.S.S.R.), 4,No.3,396
(1934). (3) Wiebe and Gaddy, J. Am. Chem. SOC.,5 5 , 947 11933). (4)Ibid., 56, 76 (1934). (5)Ibid., 57, 847 (1935). RECEIVED for review July 14, 1960. ACCEPTED December 31, 1951. This work was csrried out under contract with the Atomic Energy Ccmmission, Contract 7405-eng-92.
Selection of Surface Active Agents for Detergent Applications SUSPENDING POWEK AND MICELLAR SOLUBILIZATION A. M. MANKOWICH Paint & Chemical Laboratory, Aberdeen Proving Ground, Aberdeen, Md.
S
ELECTION of surface active agents (sirfactants) for specific metal cleaning applications is usually made on a trial and
error basis, using laboratory detergency tests which have been correlated with field test results. Experience plus the fragmentary available information on such properties as molecular structure, type and chemical stability of surfactant, surface and interfacial tensions, and Draves and Clarkson sinking times ( a ) may be used &s empirical screening media. It is imperative that all possible combinations of compatible types of surfactants be studied, since it is not possible at present t o predict synergistic combinations. Builder action varies with type of agent and soil, which necessitates investigation of additional combinations, The development of B laboratory detergency test that can be correlated with field results is a problem. An obviously considerable amount of research must be performed even when only a limited number of surfactants are tested. This process has t o be repeated for every detergent application in which the soil and/or substratum is varied. The time-consuming, trial and error approach in determining the suitability of various surfactants for use i n cleaning coinpounds intended for the removal of specific soils from specific substrata is viewed unfavorably in this laboratory. It is believed that the selection of surfactants for specific detergent applications can be accomplished more scientifically and economically by determining the fundamental or prime factors in the detergency process; studying each factor to establish numerical criteria for the various types of surfactants and builders under varying pH, concentration, and temperature conditions; and classifying soils with reference to the numerical values of the prime factors, obtaining cofactors (minimum numerical criteria for each prime factor necessary t o accomplish soil removal in specific soil-surface combinations). Once the prime factors for various types of surfactants and the cofactors for the soils are determined, an inspection of the data will indicate appropriate surfactants or combinations of surfactants, as well as suitable temperature, pH, concentration, and builder data. This is done by selecting the surfactant, or combination of surfactants, whose prime factors have equal or greater numerical ratings than the corresponding cofactors. The principle advanced is that detergency in a specific soil-surface application is accomplished only if the numerical ratings of all the cofactors are equaled or exceeded by the prime factors of the cleaning solution. The need for utilizing more than one surfactant for a specific detergency application will be readily apparent in those
cases where an otherwise! satisfactory agent is deficient in one or more prime factors. When i t is necessary to select two or more surfactants, the maximum numerical rating for each prime factor of the combination is estimated by adding the respective ratings of the individual surfactants involved. Preliminary work indicates that the numerical rating of a prime factor approaches a de&nite maximum which is not exceeded, regardless of the number of surfactants combined for a specific detergent application. This maximum is usually the highest value attained by any of the individual surfactants investigated. However, Bdditiveness of numerical ratings of a combination of surfactants is obtained until the maximum for the prime factors is reached. It is intended to cover the determination of cofactors and classification of soils and substrata in a later paper. An advantage of the proposed method of selection is that it will indicate a number of surfactants or combinations of surfactanta for each soil-surface application, thus permitting a final selection on an economic basis. No attempt is made to evaluate the relative importance of the various prime factors, and the method does not depend on such evaluation. The proposed method is not t o be confused with the speculations ($1)that detergency will eventually be calculated from a formula containing weighted physicochemical factors. Assuming that these speculations are realized eventually, they will not aid in the selection of surfactants for specific detergent applications. The formula will only give a n index of detergency. The proposed method indicates the minimum physicochemical requirements for a detergent application, together with a number of surfactants or combinations of surfactants capable of meeting the requirements. It is important to realize that the proposed method indicates synergistic combinations of surfactants. PHYSICOCHEMICAL FACTORS OF DETERGENCY
Present-day knowledge indicates that detergency is the resulta n t of many factors (I, 22, 20, $2). It is understandable, therefore, t h a t attempts t o correlate detergency with one factor only have been unsuccessful. Reich and Snell (18) emphasize a further error made in attempting t o evaluate detergency with one factor. Detergency is dependent on three groups of variables (13): those originating in the soil, in the surfactant, and in the surface. A factor involving only one of the groups, such as surface tension, or two of the groups, such as micellar solubilization, is basically not in correlation with detergency.
