NOTES
April, 1949 Polar Effects of Organosilicon Substituents in Carboxylic Acids' BY L. H . SOMMER, J. R. GOLD,G. M . GOLDBERG AND N. S. MARANS
The acid strengths of carboxylic acids containing organosilicon substituents close to the carboxyl are of interest in connection with an elucidation of the polar effects of such substituents in organic molecules. We wish to report the synthesis and acid strengths of four silicon-containing carboxylic acids. TABLEI Acid
I 11"
Formula
1509
TABLEI1 DISSOCIATION CONSTANTS FOR RCHzCOIH AT 25" R K X 106 WKlrostio (CHahSi 0.60 0.34 (CHa),SiOSi(CH& .60 .34 (C6H5) (CHdzSi .54 .31 (CH3)3SiCH* 1.24 .71 (CHdsC 1.00 .57 CHs 1.34" .77 H 1.75 Value found in this work, 1.32 X 10-5.
According to current theory,6 the acid-weakening effects of organosilicon substituents in acids I, 11, and I11 show that these substituents exert a n inductive efect of electron-releaserelative to alkyl and 133 hydrogen. The data for acid IV demonstrates that this ef207 fect diminishes rapidly in a saturated carbon chain. Additional theoretical aspects will be dis193 cussed in a later publication. 145 We thank Dr. A. W. Hutchison for valuable ad147" vice and Dr. N. C. Cook for the t-butylacetic acid.
M. p , , Silicon, yo Neut. equiv. OC. Calcd. Found Calcd. Found
(CHa)sSiCHzCO%H 40 (CHa)sSiOSi(CHa)zCHtCOzH 17 I11 (CeH6)(CH3)2SiCHzCOzH 90 IVb (CH&SiCH2CHzCOzH 22 * Acid I1 has n z a 1.4149. ~ (65 mm.), n Z a1.4279. ~
2 1 . 2 21.3
132
2 7 . 2 2 6 . 7 206 1 4 . 4 14.3 194 1 9 . 1 19.1 146 Acid IV has b. p.
(5) For an excellent summary see Johnson and Gilman, "Organic
Acids I, IT and I11 were synthesized in yields of Chemistry," Vol. 11, John Wiley and Sons, New York, N. Y . 1943, 88, 85, and 69%, respectively, by carbonation of pp. 1842-1845. the Grignard reagents derived from the appropri- THEWHITMORE LABORATORY AND PHYSICS ate chlorides.2 The Grignard reagents were ScnooL OF CHEMISTRY STATECOLLEGE added to an excess of Dry Ice in the form of chips THEPENNSYLVANIA COLLEGE, PA. RECEIVED FEBRUARY 25, 1949 and the reaction mixture allowed to stand for STATE twelve hours. Subsequent to treatment with 5% hydrochloric acid, ether was removed from the Grignard Reactions. X'IX.l Preparation of 2,acids by vacuum distillation a t temperatures be4,4-Trimethyl-3-pentanone low 25'. These acids were not distilled. They BY CARLYLE J. STEHMAN,~ NEWELLc. COOKAND FRANK were purified by two recrystallizations from hepc. WHITMORE3 tane.3 Acid IV was synthesized from chloromethyltrimethylsilane by the malonic ester synthesis The limitations of the Grignard reaction in the (66% yield) and from iodomethyltrirnethylsilane synthesis of highly branched compounds are well by the acetoacetic ester synthesis (63% yield). known. Such a synthesis, the reaction of t-amylThis acid was purified by distillation (65 mm.). magnesium chloride with isobutyryl chloride, has Determination of the acid strengths was made previously given little or no k e t ~ n e . ~ . ~ by a potentiometric method using a Leeds and Copper vessels for Grignard reactions have been Northrup 7662-A1 p H Indicator A ~ s e m b l y . ~in use in this Laboratory and others for many Stability of acids I-IV under the conditions used years. Crane, Boord and Hemes noted slightly in this procedure was demonstrated by prepara- better yields in copper than in glass. It has now tion of the sodium salts and regeneration of the been found that t-amylmagnesium chloride reacts acids. with isobutyryl chloride in copper reactors to The dissociation constant of t-butylacetic acid give 2,4,4-trimethy1-3-pentanonein yields up to was determined for comparison with acid I. 87%. Propionic acid was run as a check on the accuracy Further studies on the copper-catalyzed reacof the method a t the concentrations used. tion of acid chlorides with Grignard reagents are in progress. (1) Paper XXI in a series on organosilicon chemistry. For paper XX see THISJOURNAL, 70, 3512 (1948). (2) (a) Chloromethyldimethylphenylsilane, b. p. 138" (50 mm.), n'On 1.5212, Si, 15.1% (calcd. 15.2), was synthesized in 72% yield from chloromethyldimethylchlorosilnne and phenylmagnesium bromide; (b) the Grignard reagent derived from chloromethylpentamethyldisiloxane has recently been reported; Bluestein, THIS 70, 3068 (1948). JOURNAL, (3) The non-isolation of trimethylsilylacetic acid (I) in a previous preparation resulted from the different procedure used in working up the product; Whitmore, Sommer, Gold and Van Strien, THIS JOURNAL, 69, 1551 (1947). (4) For a recent application of this method see Nerman. Taylor, Hodgson and Garrett, ibid., 69, 1784 (1947).
Experimental For preparation of the t-amylmagnesium chloride, and coupling of the Grignard reagent with isobutyryl chloride, (1) XVIII, Whitmore and Sloat, THIS JOURNAL, 64, 2968 (1942). (2) Present address: Monsanto Chemical Co., Dayton, Ohio; research work performed during academic leave of absence from Monsanto. (3) Deceased. 6S, 643 (1941). (4) Whitmore, 6l al., THISJOURNAL, (5) W. A. Morher, Doctor's Dissertation, 1940, The Pennsylvania State College. (6) Crane, Boord and Henne, THISJOURNAL, 67, 1237 (1945).
