HER~ERT C. BROWN
14.52
The straight-line relationships shown in Figs. 4 and 5 are taken. as ample evidence that the reaction proposed in equation (1) represents the precipitation reaction a t G O O . Table I shows, for three runs, the differences between the concentrations of aluminate iori as calculated from the carbon dioxide absorbed and from the analysis I.)f the solutions for total aluTABLE I DIFFERENCESBETWEEN CONCENTRATIONS OF ALUMINATE TON AS CALCULATED FROM TOTAL ALUMIKAAND FROM CARBON DIOXIIIEABSORBEII .rime, hr.
1 2 3 4 5
0,007 013
,013
022 ,013
..
7
...
,017 , 0 10 ,012
8
... .011
,012 .n13
6
Av.
,012 ,015
,005
n'II.SO3
( J . 002
008 007 . .
,003 004
[COXTRIBUTIOX FROM T H E
Summary
.. ...
7
~~~
0 . 0oii OCl6
mina. The differences suggest, as a rough approximation, that the solubility of gibbsite is about 5 X low3mole per liter in sodium aluminate solutions of PH 11.20 to 11.45 and ionic strength 0.0 to 0.7. Acknowledgment.-The investigation was made at the suggestion of R. L. Copson, Chief of Chemical Research and . Engineering. Gray Zucas and E. R. Rushton assisted in the experimental work A method was developed for measurement of the pH of alkaline solutions containing carbon dioxide without their contamination by silica. 2. The equation 2A1022H30+ -+ .A1203. 3H20 was found to represent the mechanism of the precipitation of alumina from sodium aluminate solutions by carbon dioxide a t 60'. 3 . The solubility of gibbsite in sodium aluminate solutions of ionic strength 0.6 to 0.7 and pH 11.20 to 11.45 was estimated to be about 5 X mole per liter.
Excess of AlOr- a s calculated f r o m t o t a l AlKh coilcentration over t h a t calculated from COa ahsorption, ~ _moles,']itcr _ .~ ______ 1 1 20 p H 11 30 PFI 11. l i
fix
\?d. HS
I.
+
]>AM, A L A B A M A
DEPARTMENT O F CHEMISTRY
RECEIVEDMARCH9. 1945
O F WAYXE UNIVERSITY]
Studies in Stereochemistry. VII. The Effect of F-Strain on the Relative Base Strengths of Ammonia and the Ethylamines BY HERBERT C. BROWX In previous papers of this series, the need of considering both polar and steric factors in theorizing has been pointed out. For example, if only the polar effect of the methyl group (+I) is considered, it would be predicted that the strength of ammonia and the methylamines as bases should increase regularly with the number of methyl groups. However, this order, NH3< CH3NH2< (CH3)zNH < (CH3)3N (Fig. I , curve A), is not observed. Instead, operation of the steric factor causes B-strain,l which slightly reduces the strength of dimethylairline arid markedly reduces that of trimethylaniiiie. As a consequence, the order of base strength observed with the aid of acids with relatiirely low F-strain factors, such as hydrogen chloride ant1 triniethylboron, is NH3< (CHa)3Y< CH3NH2< (CH3)zNH (Fig. 1, curve B). Reference acids with higher F-strain factors should cause the observed order to change successively to C, D, E, and, finally, to F (Fig. 1). Actually, the order (CH3).!S< (CH3)?NH< ?;Ha < CH3N€12 (curve E) is obtained with the highly hindered reference acid, tri-t-butylboron.2 It should be possible to obtain order F, a decrease in base strength with iiicreasiiiK number of I I ) B r o a r i , U L i r t h o l u m . ~J~~ : c !7'
