June 5, 1939
C O N D U C T I V I T Y O F SODIUM [CONTRIBUTION FROM T H E
2iS7
TRIPHENYLXETHIDE IN E T H E R
DEPARTMENT OF
CHEMISTRY,
DUKEUNIVERSITY]
Conductivity of Sodium Triphenylmethide in Ether. Non-conductivity of the Sodium Derivatives of Certain Active Hydrogen Compounds' BY DOUGLAS G. HILL,JOHN BURKUS,~ SETAMAHAKIAN LUCKAND CHARLES R. HAUSER RECEIVED DECEMBER 1, 1958 The conductivity of sodium triphenylmethide in ethereal solution increases as the concentration is increased. The sodium derivatives of ketones and certain other active hydrogen compounds prepared by means of this reagent fail to conduct. Possible explanations for these results are suggested.
Schlenk and Marcus3 have observed that sodium triphenylmethide in ether solution conducts slightly in 0.0376 molar solution but not in 0.0051 molar solution. Swift3" has made precise measurements of the conductance in more dilute ether solution. He found a small conductance a t molar solution, decreasing with increasing concentration as expected for weak electrolytes. -4 minimum was hI and an increase observed between to a t higher concentration. We have extended such measurements over a higher range. In Table I are given the molar conductances of sodium triphenylmethide a t various concentrations a t 26'. The two results of Schlenk and Marcus (at Z O O ) are included in this table. It can be seen that in this range the conductance increased as the concentration of the reagent was increased. In Fig. 1 is plotted the log of the molar conductance against the log of the concentration, including the earlier work.
the simplest of which would be the sodium ion and anion I (equation 2(C6E15)3C-Sa+
Na+
+ (C&)3CPl'aC(CQH6)3-
(1)
I
The conductivity of lithium triphenylmethide in ether was found to be somewhat less than that of sodium triphenylmethide. Thus, the lithium reagent showed a molar conductance of only 0.007 ohm-' a t a concentration of 0.0239 molar and no measurable conductance (specific conductance < 1.5 X lo-; ohm-') a t 0.0125molar, whereas the sodium reagent exhibited appreciable conductance even a t lower concentration (see Table I). This lower conductivity of the lithium triphenylmethide probably indicates somewhat less formation of ionized complexes. No measurable conductivity was observed with potassium triphenylmethide in 0.005 molar ethereal solution, which was the maximum concentration obtained with this reagent. Since Schlenk and Marcus3 reported no conductivity for sodium triTABLE I phenylmethide a t approximately this concentraCONDUCTAXE OF ETHEREAL SODIUM TRIPHENYLMETHIDE tion, determination of the relative conductivities of Molar Molar the potassium and sodium reagents was not realConcentration, conductance, Concentration, conductance, mole/liter ohm-' mole/liter ohm-' ized. 0,0051" 0" 0.0482 0,0488 Incidentally, potassium triphenylmethide was ,0090 ,0180 ,0601 .0960 considerably less soluble in ether than sodium tri,0157 ,0244 .0902 .202 phenylmethide, although the latter reagent was .0261 ,0254 .1503 ,417 more soluble than lithium triphenylmethide. ,0376" ,048" ,2283 .T92 Non-conductivity of Sodium Derivatives of Certain Active Hydrogen Compounds.-Sodium tria Data from Schlenk and Marcus, ref. 3. phenylmethide reagent is a sufficiently strong base A similar increase in conductance with increase to effect the essentially complete ionization of the in concentration has been observed by Fuoss and a-hydrogen of ketones. For example, this reagent Kraus4 with tetra-isoamylammonium nitrate in di- converts a molecular equivalent of acetophenone oxane and by Evans and Pearsons with Grignard to its s3dium derivative accompanied by the disreagents in ethyl ether. The former system4showed charge of the characteristic red color of the reagent an initial decrease in conductance a t very low (equation 2). concentrations and the latter, a decrease a t etlier CH3COC6H6 -+ very high concentrations. However, over much (Cd%)&Na of the concentrations studied the curve resembled (red reagent) that given in Fig. 1. (CBH6)aCH Sa(CHiCOCaH5) ( 9 ) Fuoss and Kraus concluded that the ion pair of (colorless) the salt has a very low degree of dissociation and In contrast to sodium triphenylmethide reagent, that it was in equilibrium with associated, ionized such sodio ketones failed to conduct measurably in complexes which conduct. On applying this mul- approximately 0.2 molar etheral solutions. This tiple ion theory to the present system, the conduct- was readily determined by adding the ketone in ance would be assumed to be due not to the sodium small portions to a molecular equivalent of the reatriphenylmethide itself but to ionized complexes, gent. Thus, on adding acetophenone, methyl iso(1) Supported in p a r t b y t h e Duke University Research Council. butyl ketone, or methyl isovaleryl ketone, the con(2) Du P o n t Fellow, 1952-1953. ductance decreased roughly in proportion to the (3) W. Schlenk a n d E. Marcus, Bey., 47, 1664 (1914). ,
+
+
(3a) E. Swift, J r , , TEIISJOURNAL, 60, 1403 (1938). (4) R.M.Puoss a n d C. A. Kraus, i b i d , 56. 2387 (1933). ( 5 ) W.V. Evans and R. Pearsoo, i b i d . , 64, 2865 (1942).
( 6 ) T h e slope of t h e rising portion of t h e curve in Fig. 1 is c r e a t e r than t h e one half expected for dimer I, and might indicate more highly associated complexes.
2 78h
L).
G.HILL,J. BCKKUS,S. -\I. LUCK.\ND C. I 2.3 x IO6 ohms). At the concentration used all the sodio ketones were soluble. Zook and Rellahan7 have presented evidence that a t least certain sodio ketones exist as dimers in ethereal solution. Since these dimers do not conduct, we suggest that they exist as ring structures such as 11. Such a ring structure would not be possible with sodium triphenylmethide, which conducts. 0 . ..hTa...CH?-C-R
II
II
K-CLL~CH?.. .Sa...O 11
0--Sa
c0
I
CsH,C=CIH --CCH3 I11
Similarly, on adding benzoylacetone to a tnolecular equivalent of sodium triphenylmethide reagent, the red color was discharged and the conductance became immeasurably small. The sodio P-diketone, which precipitated after 15 minutes, is suggested to have the six-atom ring structure 111, although no evidence for such a monomer appears to have been reported. Recently’ bimolecular rate constants were obtaiiied for the Claisen type of acylation of sodioacetophenone with ethyl acetate in dilute ethereal solution. This condensation appeared to involve ( i ) 13, 1). Zook and \ V L Kellahan, THISJ O U R N A L . 79, 881 (19.57). (8) D. G. Hill, J . Burkus and C. R . H a u s e r , ibid.. 81, 602 (1959).
This reaction has now been found to take place without producing a detectable amount of intermediate ions. Thus, no conductance was observed on adding ethyl acetate to a molecular equivalent of the sodio ketone in ethereal solution. The sodio p-diketone I11 precipitated after about 30 minutes. Also, phenylacetonitrile, 2,2-diethyldecanoicacid, 2-ethyl-2-butyldecanoic acid, ethanol, octanol-1 and phenylacetylene were added to molecular equivalents of ethereal sodium triphenylmethide reagent, In each case the conductance became immeasurably small, although in the first two cases the bridge was unsteady as though the resistance was near the 2.5 X IO6 ohms limit of the instrument. Precipitates were formed exccp t with the two carboxylic acids. I t is interesting that even the sodium salt of 2ethyl-2-butyldecanoic acid, which is fairly soluble iri ether,gfailed to conduct. The state of aggregation of this salt or of the other salts was not determined. Experimental Sodium triphenylmethide was prepared in ethyl ether solution essentially a s described prcviciusly.’O The cmiceiitration of thc solution W:IS determined by decompositioii of appropriate samples with water and titration with standard acid. Conductivity of Sodium Triphenylmethide (Table I, Fig. I).-The solvent, ethyl ether, dried over sodium, showed no conductivity. Etliereal solutions of tliis reagent wer? contained in :I closed conductivity crl! o f the usual type with platinized platinum clectrudes. Tlie cell constant (0.360) \vas determined with KCl sc~lutiotiiii conductivity- water. Resistaiice was measured a t % ” by a conductivity bridge m a n u f x t u r e t l by Illdustrial Initruinents Inc., x i t h which a m mum resistaiice of 2.5 x 1176 ohms could be dctermii Tlie cell was filled with the triphen‘-lniethide solution. a n d dilutions were rn.ide i i i :i riry-box (over PZO6) in an a t mosphere of purified nitrogen. The concentration of the diluted solutioiis w:is tleterinined as described above after the resistance ineasurenierits were made. Treatment of Sodium Triphenylmethide with Active Hydrogen Compounds.-The active hydrogen compounds were purified by appropriatc procedures, and weighed samples t o ;L iii~ilecularequivalent o f approximatel!. i n 1 tril,lie!!?.liiietliitlc reagent in the cell i n the , Tile rrsult:, (tre give11 i l l t h e discussicill. ~ ~ I - K I I A MS. , C. ~
.
19) See C . 12. Hauser and W. J . Chambers, i b i t ! , 78, 9837 ( 1 (10) C. R . Hauser and €3. Hudson. “Organic Reactions,” Vul. I , John \Viley and Sons, Inc.. S e w York. N. Y., 19-12, Chap. ‘J.
