4518
2. Luz
AND
B. SILVER
[CONTRIBUTIONPROM THB WEIZMANN INSTITUTE
OF
1'01. 83
SCIENCE, REHOVOTH,
ISRAEL]
The Acid Catalyzed Exchange ob Phosphorus Bonded Hydrogen in Aqueous Solutions of Dialkyl Phosphonates, Studied by Nuclear Magnetic Resonance BY 2. Lvz
AND
B. SILVER
RECEIVED MARCH 22, 1961 T h e exchange of the phosphorus-Gonded hydrogen with deuterium in D,O has been studied for dimethyl-, diethyl- and di-n-propylphosphonate.' The ki:rctics W:LS followed by means of intensity nieasurernents on the proton magnetic resonance spectra of these compounds. The exchange was found to be acid-cat:ilyzed, the rate law for the reaction being: rate = kH[phosphonate] [Hi! k,[phosphonate]. From R comparison o f the data obtained v i t h those found for the oxidation of the same compounds, it is suggested t h a t the phosphite' form of the dialkyl phosphonates serves as a common intermediate f o r both the exchange and osidntion reactions.
+
Ir,troduction
Experimental
Little work has been done on the exchange of phosphorus bonded hydrogen. Martin used Eainan spectroscopy to study the exchange of deuterium with phosphorous acid and Fox3 used infrared spectroscopy to follow the exchange of deuterium with clibutylphosphonate in butyl alcohol-d. However in neither case mas it possible to make an exact kinetic analysis of the exchange reaction. Jenkins and Yost4 used tritium labelling to study the exchange reaction in hypophosphorous acid and gave a detailed kinetic analysis of their results. In the present work a study was made of the exchange of phosphorus bonded hydrogen in dimethyl-, diethyl- and di-12-propylphosphonate. Use was made of the fact that the intensity of a proton magnetic resonance line is proportional to the concentration of the relevant proton in the samp! e studi cd. Interpretation of Spectra.-The nuclear magnetic resonance spectra of the dialkyl phosphonates consist of a group of lines due to the alkyl groups and a doublet due to the phosphorus bonded hydrogen, with a spin-spin interaction of about 700 cycles.6 The components of the doublet (hereafter referred to as the P-H lines) are isolated from the other lines of the spertrum. If H20 is present, its absorption line appears near those of the alkyl groups. The intensities of the P-H lines are directly proportional to the concentration of the phosphorus bonded hydrogen in the observed sample. The progressive replacement of this hydrogen by deuteriuni results in a corresponding decrease ir? the intensity of the P-H lines, due to the fact that the deuterium resonance in the same field is a t a much lower frequency. The facts outlined above form the basis for a method of following the kinetics of hydrogen exchange in the dialkyl phosphoiiates. The method is applicable only if several spectra can be recorded during the half-life o€ the reaction.
Dialkyl %?hosphorrates.-Dimethyl- and di-n-propylphosphonates were made from the corresponding alcohols and plrclsphorus t r i ~ h l o r i d e . ~Diethylphosphonate was obtailled commercially (Albright and Wilson). All phosiihonates were doubly distilled before use; h.p. dimethyl-, 57-58" (9 mm.)(lit.* 5 t ~ 5 8 "(10 mm.)), diethyl-, 7'2" nim.)(lit.D72-73" (9 mm.)), di-n-propyl-, 93" (9 mm.)(lit. 91' (9-10 mm.)). DC1 Solution.-A stock solution of 3.64 iV DCl was prepared by passing DC1 gas into D20. DCl was preparedLI from benzoyl chloride and DzO. Spectrometer.-The n.m.r. spectrometer has been drscribed previously.12 Procedure.-Preliminary experiments on solutions of dialkyl phosphonates in DpO showed t h a t the exchange rate increased with increasing acidity. I n order t o study the kinetics quantitatively the following general procedure was adopted. Solutions of dialkyl phosphonate in D20 (-99%,) containing varying amounts of DCl (0.2-1.0 ill) were prepared, the recording of spectra being commenced as soon as possible after the preparation. The exchange process ryas followed by observing the decrease in intensity of one component of the P-H doublet (see Fig. 1). Since a change in the radio frequency (r.f.) intensity will in itself result in a change in the absolute intensity of a resonance line, use was made of a n arbitary measure of intensity, t i z . , the ratio of the heights of the P-H line t o one of the lines of the alkyl groups recorded immediately afterwards. Thip ratici, giving the relative intensity of the P-H line, is independent of a n y long term variations in the r.f. field, since the intensity of the alkyl lines is unaffected by the exchange reaction. T h e reactions were followed till a t least the half-life. T h e experimental error in estimating the relative intensity varied between E1-157~depending on the absolute intensity of the P-H line. The solvent used contained approximately 1o/c H20 giving a corresponding absorption line. During the exchange reaction an increase in the intensity of this line is ohserved and in principle this increase could also be used t o follow the exchange kinetics, However, because the relative changes in the water line are less than those in the P-I3 lines, this method was found t o be niuch less sensitive and therefore was not used.
(1) T h e norncnclalure in this article follows t h a t used in a recent review by G. 0. Doak a n d L. D. F r e e d m a n , Chem. Revs., 61, 31 (1961). (2) R. Bruce M a r t i n , J . A m . C h n . Soc., 81, 1874 (1959). (3) R . B. F o x . " X R L Report." 5242, Jan. 8, 1959. (4) W. A . Jenkins a n d D. hl. Yusl, J . Inorg. Nuclear Chenr., 11, 297 (1959). (5) J. A. Pople, W. C. Schneider a n d H. J. Bernstein,"High-Resolution Nuclear Magnetic Resonance," McGrsw-Hill Book C o m p a n y , Inc., New York, I\. S . , 1959, C h a p . 19. (8) H. Finegold. An%. N . Y. A r o d . Sci., 70, (4), 885 (1958).
'8,
Results Isotopic exchange reactions, in the absence of appreciable isotope effects, show first order kinetics as measured by the disappearance of one isotopic species from one of the reactants.13 The expression (7) H. MuCombie, S. C . Saunders a n d G . J. Stacie, J . Chem. .Sot 380 (1945). (8) T. hlilobendzki aud 1.Sachnowsky, Chenz. P o l s k . , 15, 31 (1917). (9) Vv', Strecker a n d R . Spitaler, Bcr., 69, 1764 ( l n 2 6 ) . (10) A. E. Arbuzov, ibtd., 38, 1171 (190Fij. (11) H. C. Brown and C. Groot, J , A m . Chem. SOL.. 64, 222R (1942). (12) E. Grunwald, A. Loemenstein and S. hieihoom, J . C4ein. P h y s . , 2 7 , 630 (1957). (13) "Radioactivity Applied t o Chemistry," Editors. A. C . Wahl aiid N. A. Bonner, John Wiley a n d Sons. I b c , , New York, N. Y.,1931, C h a p . 1.
4519
ACIDCATALYZED EXCHANGE OF PHOSPHORUS BONDED HYDROGEN
Nov. 2G, 1961
n L
_-
- il stream of dry n a t e ~ . ~ -The ~ reactions of S206F2 with several addi- nitrogen with a slight excess of fluorine in the presence of a tional substances have now been studied, and it has silver difluoride catalyst in a “catalytic reactor’I6 a t 155” Z , was purified by prolonged pumpbeen found that the peroxide acts in three ways: produced the S ~ O ~ Fwhich (1) as an oxygenating agent through the addition of ing a t -78” t o remove the contaminants silicon tetrafluosulfuryl fluoride and fluorine fluorosulfonate. I n a 15 oxygen to the central atom of the reacting species, ride, hr. run using a nitrogen flow of 8 liters per hour and a catalyst e.g., with CO, PF3, SOF2, COClp, CC14 or S; (2) as a contact time of about 15 minutes, a 96 g. sample of product fluorosulfonating agent, e.g., with HgO, KC1, IC1, was obtained. It contained 97% peroxydisulfuryl difluoC5Fs, CzF4, or SF4; (3) as an oxygenating and fluo- ride. I n another run, 26.1 g. of sulfur trioxide yielded 31.6 g. F P 9770 theoretical). rosulfonating agent by adding oxygen and fluoro- of S ~ O ~(yield, Storage of the compound in a glass vessel at -78” was sulfonate groups to the central atom, e.g., SOClF, satisfactory. Although the glass container apparently was Mo or Mo(CO)~. Pyrosulfuryl fluoride (S206F2) not attacked a t room temperature, a nonvolatile oily mateis obtained as a product of reactions of types 1 and rial was slowly formed in the S206F2. Apparently this did not introduce contaminants into the S206F2 distilled from 3. the vessel. T h e yellow oil was not identified. Experimental Materials.-The Swarts reaction’ provided a general Preparation of Peroxydisulfuryl Difluoride.-Peroxydisulfuryl difluoride can be prepared by reaction of fluorine with a n excess of sulfur trioxide a t about 250”, by the combination of fluorine fluorosulfonate with sulfur trioxide a t 3OOo4 or by t h e electrolysis of fluorosulfonic acid.5 T h e catalytic fluoronation of sulfur trioxide vapors by fluorine in the presence of a heated catalyst of copper ribbon coated (1) D e p a r t m e n t of Chemistry, University o€ Idaho, Moscow, I d a h o . (2) John E. Roberts a n d George H. Cady, J. din. Chein. Sac., 81, 4166 (195Q). (3) John E Roberts a n d George H .C a d y , ibid , 82, 3 3 2 , 3.53 (1960). (4) F. B . Dudley a n d G. H. C a d y , i b i d . . 79, 513 ( 1 9 5 7 ) . 15) F. B. Dudley, Thesis, University of New England. Australia
(1960).
method for the conversion of chlorides to the corresponding fluorides by reaction of the former with antimony(II1) fluoride in the presence of antimony(V) chloride as a catalyst. Thionyl fluoride and thionyl chlorofluoride were obtained using thionyl chloride while phosphorus(II1) fluoride was prepared from phosphorus(II1) chloride. Tetrafluoroethylene was produced by the pyrolysis of Teflon (polytetrafluoroethylene) in a n iron vessel at 550’. Sulfur tetrafluoride was used directly from a cylinder supplied by E. I . du Pont de Nemours and Co., Inc. All other materials were of reagent grade. (ti) K . B . Kellogg and G. H. C a d y , J. A m . C h e m Sac., 7 0 , 3086 (1948). (7) H .S . Booth a n d F. C . Mericola, i b i d . , 62, 040 (1040).
