Skip to main content

Full text of "The Chemistry of Phosphetane Compounds"

See other formats

































































































































































































































































































































































































































































































































































Digitized by the Internet Archive 
in 2018 with funding from 
Public.Resource.Org 


https://archive.org/details/chemistryofphospOObhup 


ILLINOIS INSTITUTE OF TECHNOLOGY 


THE CHEMISTRY OF PHOSPHETANE COMPOUNDS 

BY 

BHUPENDRA C. TRIVEDI 


Submitted in Partial Fulfillment of the 
Requirements for the Degree of 
Doctor of Philosophy in Chemistry 
in the Graduate School of 
Illinois Institute of Technology 


Approved 


GtiltArAAs' 


(Adviser) 


CHICAGO, ILLINOIS 
June, 1970 





TABLE OF CONTENTS 


Page 

ACKNOWLEDGEMENT . ...... . V 

ABSTRACT.....Vi 

LIST OF ILLUSTRATIONS.viii 

LIST OF TABLES. ix 

CHAPTER 

I. SYNTHESIS AND STEREOCHEMISTRY OF PHOSPHETANE 

DERIVATIVES . 1 

1. Introduction . 1 

2. Background.1 

3. Results and Discussion . 5 

a. Synthesis.5 

b. Reactions of Phosphetanium Salts . 11 

c. Reactions of Phosphetane Oxides . 14 

d. Stereochemistry.I 5 

4. Experimental.26 

II. STEREOCHEMISTRY OF NUCLEOPHILIC ATTACK ON 

CYCLIC PHOSPHORUS COMPOUNDS . 49 

1. Introduction.49 

2. Hydrolysis of Phosphinate Esters . 65 

a. Results and Discussion. 65 

b. Experimental. 74 

3. Hydrolysis of Phosphonium Salts . 94 

a. Background. 94 

b. Results and Discussion .. 97 

c. Experimental. 110 

iii 

























TABLE OF CONTENTS - CONTINUED 


Page 

III. SYNTHESIS AND REACTIONS OF «-HALOMETHYLPHOSPHORUS 

COMPOUNDS. 128 

1. Introduction. 128 

2. Results and Discussion.13° 

3. Experimental.142 

IV. NUCLEAR MAGNETIC RESONANCE STUDIES ON PHOSPHETANE 

DERIVATIVES. 156 

1. Introduction. 136 

2. Results and Discussion. i57 

3. Experimental.l^ 8 

REFERENCES. 171 

VITA. 178 


iv 














ACKNOWLEDGMENT 


The author wishes to express his sincerest appreciation to Dr. Shel¬ 
don E. Cremer for generous assistance, helpful guidance and encouragement 
during the entire course of this research. 

Congenial environment provided by colleagues and faculty members is 
gratefully acknowledged. 

The author is indebted to his parents and other members of the family 
without whose love and encouragement, this work would not have been possi¬ 
ble . 


v 









ABSTRACT 


A series of new phosphetane derivatives (oxides, phosphinic acids, 
phosphinate esters, acid chlorides, etc.) was prepared by cyclization of 
polymethylated alkenes with phosphorus trichloride in the presence of an¬ 
hydrous aluminum chloride. In addition, a number of 1-substituted 2,2,3, 

4,4-pentamethylphosphetane oxides were prepared by the reaction of the 
corresponding organometallic compound with l-chloro-2,2,3,4,4-pentamethyl- 
phosphetane 1-oxide. Each phosphetane oxide was conveniently and stereo- 
selectively reduced to the respective phosphetane and the latter was qua- 
temized to give the phosphetanium salts. Ring opening reactions of phos¬ 
phetane oxides and ring expansion reactions of phosphetanium salts in se-r 
veral systems were also investigated. A novel ring opening reaction of 
l-chloro-2,2,3,4,4-pentamethylphosphetane-1-sulfide with sulfuryl chloride 
was observed. 

The stereochemistry of nucleophilic substitution of a phosphinate es¬ 
ter was investigated by a detailed deuterium labelling experiment. Reten¬ 
tion of configuration in the reaction could be rationalized by assuming the 
formation of trigonal bypyramid which could under go psuedorotation. Kine¬ 
tics of the alkaline hydrolysis of phosphinate esters was examined. The 
rates in each case were second-order; first-order in each component. The 
rate acceleration (compared to the open-chain analogues) in these esters 
was explained on the basis of ease of intermediate formation resulting from 
strain in the ring. 


vi 










» 






Isomer crossover in phosphetanium salts was found to be catalysed by 
alkali. Pseudorotation of the initial intermediate was invoked as a ra¬ 
tionale for the observed phenomenon. Phosphetanium salts were found to 
decompose with base stereospecifically to yield a predominance of one isomer 
of the oxide. The alkaline decomposition of phosphetanium salts and one 
phospholanium salt was investigated. The reactions were found to be third- 
order overall, first-order in the concentration of the phosphonium salt 
and second-order in alkali. Enhanced rates of hydrolysis were also observed 
in these cases, which corroborated the strain theory. 

Reactions of <*-halomethylphosphorus compound were studied to explore 
their potential as precursors for small-membered phosphorus heterocycles. 

A novel rearrangement of “-halomethylphosphines in the presence of acid was 

investigated in detail. 

31 

P nuclear magnetic resonance data were obtained for several phos- 
phetane compounds. In the cases studied, no correlation could be obtained. 


vi 1 


























LIST OF ILLUSTRATIONS 


Figure Page 

1-1. X-ray Data on 1-Phenyl-1,2,2,3,4,4-hexamethylphosphe- 

tanium Bromide ( 27) . 20 

1-2. Ring-puckering in l-Phenyl-2,2,3,4,4-hexamethlyphosphe- 

tanium Bromide ( 27) . 21 

11-1 - Pseudorotation in a Trigonal Bipyramid..51 

II-2. Methyltetrafluorophosphorane and Dimethyltrifluoro- 

phosphorane.53 

II-3. Dimethyltrifluorophosphotane.55 

II-4. Tetramethylenetrifluorophosphorane.56 

II-5. Tetramethylenetrifluorophospherane.57 

II-6. Projection of Balaban's 20-Vertex Graph 

(as given by Mislow).58 

II-7. Pseudorotation in Cyclic Phosphate Esters . 60 

II-8. Pseudorotation in Cyclic Phosphinate and 

Phosphonate Esters . 63 

II-9. Transition States for Apical and Equatorial 

Departure.21 

11-10. Plot of log(k_/T) vs. 1/T for the Alkaline 

Hydrolysis of ( 11) .82 

II-11. Plot of log(k /T) vs. 1/T for the Alakline 

Hydrolysis of ( 15) .93 

11-12. Isomer Cross-over in Phosphetanium Salts 

by Pseudorotation.98 

11-13. Pseudorotation in Phosphetanium Salt . 102 


11-14. Plot of log(A -A) vs. time for the Alkaline 

Hydrolysis of ( 100) .118 

2 

11-15. Plot of k , vs. B for the Alkaline Hydrolysis 

of (100).°..114 


III-1. Working Curve for the Estimation of Acetic Anhydride 

in Acetyl Chloride.183 

31 

IV-1. Plot of Calculated vs. Observed P Chemical Shifts 

of Phosphetanes.162 

IV-2. Pmr Spectrum of l-Chloro-2,2,3,4,4-pentamethylphos- 

phetane (48) . 

— .... 


167 
































LIST OF TABLES 


Table Page 

II-l. Alkaline Hydrolysis of Phosphinate Esters. 

Second-Order Rate Constants at 25°. 71 

II-2. Alkaline Hydrolysis of Phosphinate and 

Phosphinate Esters . 72 

II-3. Alkaline Hydrolysis of l-Methoxy-2,2,3,3,4- 

pentamethylphosphetane 1-Oxide (8) at 25.4°.75 

II-4. Alkaline Hydrolysis of l-Methoxy-2,2,3,3,4-penta- 

methylphospnetane 1-Oxide (8) at 25.4°.76 

II-5. Alkaline Hydrolysis of l-Methoxy-2,2,3.3,4-penta- 

methylphosphetane 1-Oxide (8) at 25.4°.76 

II-6. Alkaline Hydrolysis of l-Methoxy-2,2,3,3,4-penta¬ 
methylphosphetane 1-Oxide (8) at 60°.77 

II-7. Alkaline Hydrolysis of l-Methoxy-2,2,3,3,4-penta- 

methylphosphetane 1-Oxide (8_) at 60°.77 

II-8. Activation Parameters for the Alkaline Hydrolysis of (8) yg 

II-9. Alkaline Hydrolysis of l-Methoxy-2,2,3,4-tetra- 

methylphosphetane 1-Oxide ( 11) at 25°.78 

11-10. Alkaline Hydrolysis of (11) at 25°.79 

II-11. Alkaline Hydrolysis of ( 11) at 39.6° . 79 

11-12. Alkaline Hydrolysis of ( 11) at 39.6° . 80 

11-13. Alkaline Hydrolysis of ( 11) at 60°.80 

11-14. Alkaline Hydrolysis of ( 11) at 60°.81 

11-15. Activation Parameters for the Alkaline Hydrolysis of (1) 83 

11-16. Alkaline Hydrolysis of l-Methoxy-2,2,3,4,4-penta- 

methylphosphetane 1-Oxide ( 86) at 25°. 83 

11-17. Alkaline Hydrolysis of ( 86) at 25°.84 

11-18. Alkaline Hydrolysis of l-Methoxy-2,2,3,3-tetra- 

methylphosphetane 1-Oxide ( 15) at 0°.84 

11-19. Alkaline Hydrolysis of ( 15) at 0°.85 


IX 






































































LIST OF TABLES - CONTINUED 


Table Page 

11-20. Alkaline Hydrolysis of ( 15) at 15° ..... . 86 

11-21. Alkaline Hydrolysis of ( 15) at 15°.87 

11-22. Alkaline Hydrolysis of ( 15) at 25° . ..• * 88 

11-23. Alkaline Hydrolysis of ( 15) at 25°.89 

11-24. Alkaline Hydrolysis of ( 15) at 33.9°. 1 90 

11-25. Alkaline Hydrolysis of (15) at 33.9° . 

11-26. Activation Parameters for the Alkaline 

Hydrolysis of ( 15) .,92 

11-27. Equilibration of Phosphetanium Salts with Base . 101 

11-28. Alkaline Hydrolysis of Phosphetanium Salts 

in 50% Ethanol-Water. Third-Order Rate Constants 

at 25 °.107 

11-29. Solvent Effect on the Hydrolysis of Phosphetanium 

Salts. Third-Order Rate Constants at 25°.109 

11-30. Alkaline Hydrolysis of l-Benzyl-l-phenyl-2,2,3,4,4- 

pentamethylphosphetanium Bromide (94) in 50% Ethanol- 
Water at 25°.~T.US 

11-31. Alkaline Hydrolysis of ( 94) in 50% Ethanol Water at 50°. no 

11-32. Alkaline Hydrolysis of ( 94) in 50% Ethanol-Water at 50°. 

11-33. Alkaline Hydrolysis of ( 94) in 50% Ethanol-Water at 25°. 117 

11-34. Activation Parameters for the Alkaline Hydrolysis of 

( 94) in 50% Ethanol-Water. 117 

11-35. Alkaline Hydrolysis of 1-Phenyl-l,2,2,3,3- 

pentamethylphosphetanium Bromide ( 102) in 50% 

Ethanol-Water at 25°.118 

11-36. Alkaline Hydrolysis of ( 102) in 50% Ethanol-Water 

at 25°.118 

11-37. Alkaline Hydrolysis of l-Benzyl-l-phenyl-2,2,3,3- 

tetramethylphosphetanium Bromide ( 100) at 25.2° .... 119 

11-38. Alkaline Hydrolysis of (100) at 25.2° . ... .119 


x 



























































































LIST OF TABLES - CONTINUED 


Table Page 

11-39. Alkaline Hydrolysis of (100) . , 12Q 

11-40. Alkaline Hydrolysis of (100) at 39.6° . 120 

11-41. Alkaline Hydrolysis of ( 100) at 60° . 

11-42. Alkaline Hydrolysis of ( 100) at 60° ^ 21 

11-43. Activation Parame for the Alkaline 

Hydrolysis of (100).122 


11-44. Alkaline Hydrolysis of 1-Benzyl-l,2,2,3,4,4- 

hexamethylph . 122 

11-45. Alkaline Hydrolysis of 1,l-Dibenzyl-2,2,3,4,4- 
pentamethylphosphetanium Bromide ( 26) . in 50% 

Ethanol-Water at 25°. 223 

11-46. Alkaline Hydrolysis of 1-Phenyl-l,2,2,3,4,4- 
hexamethylphosphetanium Bromide (27) in 50% 

Ethanol-Water at 25°. 223 

11-47. Alkaline Hydrolysis of l-Benzyl-l-phenyl-2,2, 

3.4.4- pentamethylphosphetanium Bromide at 25°. 224 

11-48. Alkaline Hydrolysis of C94) in 50% Ethanol- 

Water at 25°.124 

11-49. Alkaline Hydrolysis of l-Benzyl-l-phenyl-2,2,3,3- 
tetramethylphosphetanium Bromide ( 100) in queous 
Solution at 25°.225 

11-50. Alkaline Hydrolysis of (100) in 50% Ethanol-Water 

^ 25°. 125 

11-51. Alkaline Hydrolysis of l-Benzyl-l-phenyl-2,2,3, 

3.4- pentamethylphosphetanium Bromide ( 101) in 

Aqueous Solution at 25°.126 

11-52. Alkaline Hydrolysis of ( 101) in 50% Ethanol- 

Water at 25°.126 

11-53. Alkaline Hydrolysis of 1-Phenyl-l,2,2,3,3-pen- 
tamethylphosphetanium Bromide ( 102) in 50% 

Ethanol-Water at 25°. 127 

11-54. Alkaline Hydrolysis of 1-Benzyl-1-phenyl- 

phospholanium Bromide ( 104) in 50% Ethanol- 

Water at 25°.127 


xi 



































LIST OF TABLES - CONTINUED 


Table Page 

III-l. Relative Rates of Nucleophilic Substitution in °:-Halomethyl 

Compounds. 136 

III-2. Results of NMR Integration of Labelling Experiments for 

Chloromethyldiphenylphosphine . 151 

III- 3. Kinetics of Halogen Exchange of the Reaction of 

Ph 2 P(0)-CH 2 Cl 17.155 

31 

IV- 1. P Chemical Shifts of Phosphetanes, Relative to 

85% H„PO..161 

3 4 

31 

IV-2., P Chemical Shifts of Phosphetanium Salts Relative 

to 85% H 3 P0 4 .165 

31 

IV-3. P Chemical Shifts of Phosphinate Esters Relative to 

85% H 3 P0 4 .166 


XI1 










CHAPTER I 


SYNTHESIS AND STEREOCHEMISTRY OF PHOSPHETANE DERIVATIVES 

1. Introduction 

Small-membered phosphorus heterocycles have attracted considerable 

1 2 

attention, not only in the area of their synthesis > but also from 

2 

the point of view of their stereochemistry, chemical reactivity , and 
physical properties. The chemistry of phosphetane (phosphacyclobutane) 
systems has been investigated in detail; interestingly, a number of the 
reactions seem to occur by initial attack at the phosphorus site. These 
phosphetanes, in general, have contributed significantly to the knowledge 
of phosphorus chemistry because they undergo reactions and possess phy¬ 
sical properties which are unique to the four-membered ring system and 
are quite different than those observed in open-chain analogues. 

2. Background 

The general methods for the preparation of these phosphorus hetero- 

2,3 

cycles have been reviewed . However, it is pertinent to briefly 

describe the various methods of synthesizing four-membered phosphorus 
heterocycles. 

4 

Kosolapoff has measured the pK of a four-membered cyclic phos- 
phinic acid. The synthesis involving an intramolecular Grignard reaction, 
resulted in a very poor yield (0.1%) and therefore was of little prepara¬ 
tive value. 


1 


































































2 


HO) P-CH -CH -CH Br 
5 2 2 2 2 


M 

g 

Ether ^ 


C 2 H 5° 





0 


n 


II jJ 


HO- 



A general method for the synthesis of phosphorus heterocycles has 
been described by Wagner a so dium alkylphosphinide is 

allowed to react with an a,w - dihaloalkane. Phosphetane systems result 
when n=3 in the following equation: 

RPHN + X-(CH_)-K-> R-P< >(CH 0 ) _ 

a 2 n 2'n-2 

n= 2 to 7. 


Unfortunately, this method has been reported in the patent literature 
only, and the experimental details of these synthesis have not been giv¬ 
en. 

g 

Green has described the synthesis of a four-membered ring in a 
tricyclic system (1). The preparation consists of the conjugate addition 
of methylphosphonous dichloride to norbornadiene. However, attempts to 

9 

repeat this reaction have been unsuccessful 



(1) 














3 


The synthesis of a highly methylated phosphetane derivative has 
been described by McBride and co-workers The preparation was accom¬ 

plished by the treatment of 2,4,4-trimethyl-2-pentene with a complex of 
phosphorus trichloride and anhydrous aluminum chloride in methylene chlo¬ 
ride solution. The propesed mechanism involves electrophillic attack of 
the phosphorus moiety (PC1+) on the double bond of the alkene. The re¬ 
sultant secondary carbonium ion (2) undergoes a Wagner-Meerwein rearrange¬ 
ment in which one of the terminal methyl groups tertiary carbonium ion 
(3) as indicated in the following scheme. The latter cyclizes to produce 
the phosphonium salt (4) which on treatment with water gives the acid 
chloride (_y . The intermediacy of (4) has been verified in the nmr by 
Cremer 9 . 




I H 
-C-C 

I 

CH 3 



PC1„ + 

3 © 

PC1 2 



A1C V 
+ ici; 


CH CH 

1 *3 I *3 

\>© 1 

> HC— C — CH— C—CH 

3 I * 

CH :PC1 


( 2 ) 




(5) 


(3) 


(4) 


























































4 


The above preparation has subsequently been extended by Cremer and 
11 

Chorvat for the synthesis of a number of substituted phosphetane 

compounds by using phenyl-phosphonous dichloride (6^ a) and methylphos- 
phonous dichloride (6^ b). A variety of substituted alkenes were employed 
as well. The illustration below summarizes their results: 






C-C—R 1 

H 


A1C1 3 

+ R 4 PC1 2 -* 

(6a) R 4 =Ph 





(6b) R 4 =CH 3 


i) R 1 =R 2 =CH 3 , R 3 =H, R 4 =Ph 

ii) R 1 =R 2 =CH 3 , R 3 =H, R 4 =CH 3 

iii) R 1 =CH 3 , R 2 =H, R 3 =CH 3 ,R 4 =Ph 

iv) R 1 =R 2 =H, R 3 =CH 3 , R 4 =Ph 

v) R 1 =R 2 =R 3 =H, R 4 =Ph 


Recently, Zyablikova and co-workers have described the prepara¬ 

tion of phosphetane oxides by the treatment of phenyl bis-(chloromethyl) 
phosphinate with the sodium salt of diethyl malonate in the presence of 
sodium ethoxide. The product is the result of a double displacement rea¬ 
ction: 


0 

II 

pho-p; 


CH 2 C1 

CH 2 C1 


0 

+ NaOi(COOEt) 2 NaQEt > PhO-P<^^XCOOEt) 2 



























5 


The corresponding carboxylic acid was produced by hydrolysis and decarbo¬ 
xylation of the product. Another recent example of the syntheses of a 
phosphetane derivative has been provided by Berglund and Meek . Ring 
formation was achieved by treating 2-chloromethyl-2-methyl-1,3-dichloro- 
propane with sodium diphenylphosphide in liquid ammonia to produce a sub¬ 
stituted phosphetanium salt: 


2NaP(Ph) 2 + 


NH r 


ch 3 c(ch 2 ci) 3 


■>Ph 2 P 



CH 


CL 


CH„P 


3. Results and Discussion 

a) Synthesis . This phase of the project consisted in the synthesis 
of new phosphetane compounds and involved the study of the chemical re¬ 
activity and the physical properties of these as well as other reported 
compounds. The method of McBride which was extended by Cremer and 
Chorvat , was utilized for the preparation of various new acid chlo¬ 
rides. Thus, addition of 3,4,4-trimethylpentene-2 to a previously pre¬ 
pared complex of PC1 3 / A1C1 3 in methylene chloride solution followed by the 
decomposition of the resultant solution with water yielded 1-chloro-2,2,3, 
3,4-pentamethylphosphetane 1-oxide (7). Treatment of the latter with so¬ 
dium methoxide gave the corresponding phosphinate ester (8). Both (7) and 
(8) could be hydrolyzed to the acid (9) with alkali: 










6 


CH 3~ 


CH CH 
I ^ I 5 

c - c = 

l 

CH 


CHCH, 


(i) pci 3 /aici 3 


CH 2 C1 2 


(ii) H 2 0 


■> 



OCH, 


(i) NaOH 

P=0 

^ CL (ii) HCL 


(7) 


(^N^OH 
(ii) HCL 



(9) 





Similarly, l-chloro-2,2,3,4-tetramethyl 1-oxide ( 10) was prepared 
starting with 4,4-dimethyl-2-pentene-2; the resultant acid chloride ( 10) 
was converted to the corresponding ester (11) and acid (12): 


CH H 

i 3 I 

CH — C -C == CHCH 


(i) PC1 3 /A1C1 3 


CH 2 C1 2 


3 I 


CH, 


3 (ii) H 2 0 





(12) 



OCH. 


Ql) 

















































7 


It may be noted here that although compound ( 11) contains three asy¬ 
mmetric centers, predominantly one isomer of the product was obtained 
(determined by nmr). Thus, the ring closure seems to occur in a highly 
stereoselective manner. 

Because of the sensitivity of l-chloro-2,2,3,3-tetramethylphosphetane 
1-oxide ( 14) produced from 2,3,3-trimethylbutene-l) to water, it was not 
possible to isolate this material. The ester derivative ( 15) was readi¬ 
ly obtained by treatment of crude ( 14) with methanol-triethylamine (15). 
The acid ( 13) was obtained by the alkaline hydrolysis of ( 15) . The acid 
in turn could be converted to the acid chloride ( 14) with thionyl chlo¬ 
ride . 


ch 3 CH 



i) PC1 3 /A1C1 3 


ch 3 — c —c =ch 2 



H 


H 


soci 2 /<j>h 




0 


0 



OH 



\ 


CL 


(13) 



CH 3 OH/NEt 3 


(14) 


0 



OCH 


3 


(15) 


















































The cyclization reaction described above, was extended to the syn¬ 
thesis of l-phenyl-2,2,3,4-tetramethyl-phosphetane 1-oxide ( 16) as well 
as a fused system (19). The former was prepared by the treatment of 4, 
4^dimethylpentene-2 with PhPCl^/AlCl_ . Reduction of (16) with tri- 

O - 

chlorosilane pyridine yielded the phosphetane (17) which was quaternized 
to give the salt ( 18) : 




CH— CH 3 


(i) PhCl /A1C1 

(ii) H 2 0 



(16) 


v Ph 

(17) 



(18) 


HSiCi 

Pyridine ^ 


The fused system ( 19) was synthesized by the following procedure; 


the synthesis of the precursor is also illustrated: 











































9 



(i) PhPCl 2 /AlCl 3 
CH) H 2 0- 



(i) Red 11 

_N 

(ii) CH^Pn 



cii) 


Cremer and Chorvat had investigated the reaction of phenyllithium 
with (5) and isolated the corresponding phenylphosphetane oxide (20) with 
the open-chain oxide ( 21 ): 


(£) 


PhLi 




CH„ 

CH_ 

CiL 

0 

— i 3 

i 3 

1 3 

II .Ph 

+ Ph- C — 

C — 

c — 

P 


H 

1 

CH 3 

^OH 


(20) (21) 


The reaction of organometallic reagents with the acid chloride (5) was 
generalized to include other alkyl and aryl substituents at the phosphorus 
heteroatom. Treatment of (5) with the appropriate Grignard reagent or 
organolithium compound gave a number of phosphetane oxides; the reaction 
proceeded with high stereoselectivity to give only one isomeric product 
in each case: 



R = -CH 2 Ph, 1-naphthyl, 1-phenanthryl, 

9,-phenanthryl, -CH^ (9), t-Butyl (14) 
m l £ - chlorophenyl, (14), £ - fluorophenyl (14). 





























The stereospecific reduction of the phosphine oxides with trichlo- 
rosilane/triethylamine gave the corresponding phosphetanes. Quatemiza- 
tion of the phosphetanes with methyl bromide gave the quaternary salts: 


10 


\ 






HSiCl /NEt 

o ---^ 

S' 




v R 


n r 


CH 3 Br 


-> 



One stereoisomer of the salt (24) (R=benzyl) was prepared ( 12) by 
the reduction of the methylphosphetane oxide ( 22) followed by quatemiza- 
tion of the phosphine ( 23) with benzyl bromide (for stereochemistry see 


P-15 ) 



HSiCl, 


( 22 ) 


'CH„ 



PhCH 2 Br 



>CH 2 Ph 


(24) 


CH, 


The synthesis of the benzylphosphetane oxide ( 25) made the preparation of 
the other isomeric salt possible, as shown below. Two isomers of the p- 
deuteriobenzyl salts were also prepared by the same method: 


PhCTLM Br 
2 g 




HSiCl. 


(25) 


^7 

P--CH 2 Ph 


(24)R=CH 3 or 


H ( 26 )pDPhCH 2 - 

V 

PhCH-Br 




CT 'CH PhDp_ 


CH PhDp. 


CH^Ph 


D _,CH.PhD 

W rP 

XH 2 Ph Br 


(26) 





































































11 


These salts have been very useful for studying the participation of pseu¬ 
dorotation in these systems. 

B. Reactions of Phosphetanium Salts . The alkaline decomposition 

of the phosphonium salt ( 24) was also investigated by Cremer and Chrovat*^. 

The initial structure ( 28) for the product was found to be incorrect and 

17 

was later assigned the new structure ( 29) by Cremer by using a deterium 
labelling experiment. Cremer treated l^p-deuterophenyl-1,2,2,3,4,4-hexa- 
methyl-phosphetanium bromide (30)with alkali and observed that the deu¬ 
terium atom in the product was at the allylic position as shown in struc¬ 
ture (29b) rather than the vinylic position as would be the case if struc¬ 
ture ( 28) was correct. Treatment of ( 30) with NaOD gave the corresponding 
dideuterio compound which showed no allylic protons in nmr: 



























































12 


Recently, Tripett and co-workers have arrived at similar conslusions. 

The unique rearrangment reaction observed from the decomposition of 
the salt ( 27) has been extended to the synthesis of various tri and tetra¬ 
cyclic phosphorus compounds. Compound ( 31) for instance, was decomposed 
to the tricyclic phosphine oxide (32). Aromatization of the latter by 
Pd/C dehydrogenation ( 19) would result in the cleavage of one of the bonds 
to yield either of the two isomeric phosphine oxides; compound ( 33a ) would 
result from the cleavage of a P-C bond whereas ( 33b) would result from the 
breakage of a C-C bond. 



The rearrangement during aromatization is not without precedent ( 20) : 


0^-00 

Since there is little difference between the bond energies of the P-C bond 
(80 kcal.) (21) and the C-C bond (77-80 kcal.) (22); it would be difficult 




























13 


to predict which of the products ( 33a or b) would predominate. 

In any event, these compounds are potential precursors for steroid¬ 
like molecules containing phosphorus atoms at either the 1- or 4-position; 
to date H phosphasteroids" are an unknown class of compounds. The tetra¬ 
cyclic model precursor ( 36) of the steroidal skeleton was prepared in the 
following manner; 


Ph-CH 2 COOH + 



Br 



COOH 



Quinoline 


0 Br 

Ph-CH=CHPh— 




(Cis or Trans) 
(i)HSiCl^HEt 3 

(ilTCH^Er 

(iii)OH 


> 




















14 


C. Reactions of Phosphetane Oxides . Alkaline decomposition of ter¬ 
tiary phosphine oxides results in the formation of a phosphinic acid and 
a hydrocarbon ( 23) : 

R 3 P=0 + OH ->R 2 P(0)-0H + RH 

As would be anticipated, the leaving group which stabilizes a negative 
charge should be the easiest to remove (24). 

PhCH^ > Ph > CH^ > higher alkyls. 


However, alkaline treatment of 1-phenyl-2,2,3-trimethyl^phosphetane 1-oxide 
( 37) resulted in the ring-opened compound ( 38) : 



OH 


-> 

methyl 

cellosolve 


CH„ CH 0 
I 3 | 3 , 

H — C — C — P — Ph 
I I I 
CH„ CH„ OH 


(38) 


(37) 


Similarly, 1-phenyl-2,2,3,3-tetramethylphosphetane 1-oxide (39) gave 
an open chain phosphinic acid ( 40) on treatment with alkali: 



(39) 


OH_ > 

methyl 

cellosolve 


CH, CH, 0 

I 3 | 3 | 

CH,- C — C — P — Ph 

3 I I I 

CH CH OH 

5 5 (40) 


The results above are surprising, since on the basis of carbanion sta¬ 
bility, one would anticipate the formation of benzene. The ring-opening 
during alkaline decomposition, therefore, reflects the strain in the four- 
membered ring. 













































15 




© 

Br 


(27) 


Stereochemistry . The stereochemistry involved in the preparation of 
the four-membered ring compounds as well as that of the reactions of these 
heterocyclic compounds has stimulated considerable research in this labo¬ 
ratory and other ^>14,17^ Cremer and co-workers first prepared the 
phosphetane oxide ( 20) and assigned the cis- (1-phenyl and 3-methyl) geo- 

O 

metry to the high melting isomer (126-127 ). The other isomer which melts 

O 

at 117-118 has also been isolated. The original (tentative) assignment 

was based on the fact that the oxide ( 20) underwent stereoselective re- 

25 

duction with trichlorosilane/pyridine to give a phosphetane (41). The 

latter was assigned cis- geometry on the basis of nmr studies as well as 

26 

thermodynamic equilibrium work . The phosphetane ( 41) was subsequently 

quaterized with methyl bromide to yield a salt which was submitted for 

X-ray analysis. The 3-methyl and 1-phenyl groups were shown to have a trans 
27 

relationship (*). Since the quatemizati on step is known to proceed with 

28 

retention of configuration at the phosphorus, the assigned geometry of 
the phosphetane was not substantiated. Hydrogen peroxide oxidation of the 
phosphetane(41) gave back the starting isomeric phosphetane oxide (20). 

29 

Since this oxidation is known to proceed with retention of configuration 
it was concluded that the oxide ( 20 - high melting) possessed the trans 
geometry. 


(*) Prof. Trefonas, Private Communication. Prof. Trefonas has revealed 
that the geometry is trans ; although it is shown cis in this publi 

cation27 














I 



















; 
























































16 


The rigorous assignment of the phosphetane oxide however, has been 

subject to some confusion. An original communication of Caughlan and 

30 ° 20 

Haque referred to the oxide (m.p. 127 ) as "cis" but pictorially 

represented the substituents in the trans configuration. Caughlan has 

recently disclosed that the " cis" assignment was incorrect and that the 

29 

molecule is the " trans" isomer . Trippett has made several unwarrented 
assumptions with respect to Caughlan’s X-ray work and has written revers¬ 
ed geometry in several publications. 

30 27 

The above mentioned X-ray data of Caughlan and Trefonas nec- 
cessarily led to the following stereochemical course of the reduction and 
oxidation reactions. As illustrated below, the reduction with trichlo- 
rosilane as well as the oxidation with hydrogen peroxide must proceed 
with retention of configuration to account for the X-ray results: 



Recently, Haque has performed an X-ray analysis on compound (^)and 
has shown that the chloro-substituent at the phosphorus and the methyl 
group at position 3 have a trans relationship. 



( 5 ) 






































































17 


The acid chloride (5) can be converted to the phenyl-phosphetane oxide 

o 

(20,m.p. 127 ) by treatment with phenyl-lithium or phenylmagnesium bro- 
2 

mide . Since the oxide (20)has been shown to be the trans isomer, the 
reaction with organometallic reagents must proceed with retention of 
configuration at the phosphorus: 



In view of the above discussion, it seems reasonable to conclude that 

treatment of the trans acid chloride (5) with sodium methoxide should also 

33 

yield the trans ester ( 42) via a retention mechanism (*). This as- 

34 

signment is consistent with that of Bergesen's who claimed to have i- 
solated both the isomeis of the analogous ethyl ester. He assigned the 
stereochemistry on the basis of refractive index, desity and nmr data. 



27 30 

The X-ray investigations , mentioned above, have not only pro¬ 
vided the geometrical relationship of substituents but also data on the 

bond-lengths and bond angles in the phosphetane system. Fig. 1 represents 

27 

one such system . The strain in the molecule is evident from the inter- 

O 

nal angles in the ring in which the angles are = 82.6 , P-C2-C^ = 

O O 

83.9 , and P-C.-C = 85.4 . The angles between the substituents external 


(*) This has been substantiated by a detailed labelling study (p.67 ). 























18 


to the ring at the phosphorus is nearly tetrahedral (1Q9.1 ) The bond 
length data, however, is rather striking. The P-C bonds in the ring are 

O 

slightly longer (1.90 and 1.94 A) than the reported values in other systems 
(1.78-1.84 A) 21 . 

O 

The C-C bond lengths for the ring (1.60 A) are longer than theoreti¬ 
cal values; this also appears to be the case for the C-CH^ bonds at posi¬ 
tions 2 and 4. The bond length C^-CH on the other hand is very short 

O 

(1.36 A). This shortening of the C-C bond distance has been shown not to 
be due to crystal packing. 

Another interesting feature of the ring system is the puckering of 
the four-membered ring. The phosphetane ring has been shown to be nonpla- 

O 

nar with a dihedral angle of 24 . This has also been found to be the case 

32 

in compound (£) . All pertinent data are shown in the Fig. 1. The puck¬ 

ering seems to be in the predictable direction in Fig. 2 because in that 
conformation, the nonbonded interactions would probably be minimal. 

As mentioned earlier (p.3), a symmetrical intermediate (4) is formed 
during the synthesis of the acid chloride (^) . Stereoselective hydrolysis 

gives a predominance of the trans isomer of (5). This stereoselectivity 

35 

can be rationalized on the basis of the following assumptions : 

i) a pentacovalent intermediate is formed by an apical attack 
of water which is similar to that formed during the alkaline 
hydrolysis of phosphonium salts (see p.98), 

ii) the intermediate phosphorane is capable of undergoing pseu¬ 
doration (see p.98), 

iii) the leaving group departs from an apical position only. 

Attack of a molecule of water on the symmetrical intermediate (4), 

































































19 


followed by loss of a proton gives the phosphorane intermediate (43) 
Pseudorotation of the latter would yield two other phosphoranes (44) and 
(45) both of which are capable of undergoing decomposition by the loss of 
chloride ion from the apical position, followed by the loss of a proton to 
yield (5). In intermediate ( 44) , the solvated hydroxy group is on the same 
side of the ring as the 3-methyl group. An interaction of these two groups 
would result in the faster decomposition of ( 44) than (45)(i.e.k^ in 

the scheme below): 



CX^Cl 


A similar explanation can be advanced for the stereoselective formation of 
one predominant isomer of the phenylphosphetane oxide ( 20) during an alkaline 
decomposition of the corresponding benzylphosphetanium salt. This reaction 
occurs without ring opening; there is no departure of the apical P-C bond. 

The above argument staisfactorily explains the formation of tne isomer 

34 

of the acid chloride (5J). Bergesen has described the isolation of the 
single isomer of (5). McBride ^ in his original paper has made similar 
claims. Thionyl chloride treatment of the acid ( 46) rendered a mixture of 
isomeric acid chlorides (5) which are quite distinct in the nmr ^ . This 
is more reasonable in view of the tautomerization of the acid in solution: 












20 



bO 

•H 

PL, 


o 



































21 



Fig. 1-2. Ring-puckering in 1-phenyl-1,2 , 2 ,3,4,4-hexamethylphosphetanium bromide 
(27) . The cis (l-phenyl+3-methyl) is shown. 




























To obtain conclusive proof for the stereochemical course of various 
reactions, it was critical to synthesize the cis isomer of the acid chlo¬ 
ride (5). One possible approach was based on the following sequence of 
reactions: 



(5) (47) (48) 


36 

Following the method of Griffin , the acid chloride (5) was heated with 

phosphorus pentasulfide to obtain a mixture of isomeric thioacid chlorides 

(47)♦ The latter was reduced to the chlorophosphetane ( 48) with tri-(n- 

37 

butyl) phosphine . This reaction is an equilibrium process and the pro¬ 
duct must be removed as it is formed in order to drive the equilibrium to 

• u*. 36 
the right : 


R P = S + (nBu) P: . . - > R P: + (nBu) P=S 

J J J 

the chlorophosphetane (48)(cis:trans::1:2) was distilled by passing a cur- 

36 

rent of dry nitrogen gas through the reaction mixture. Trippett has 
recently reported the synthesis of (48) by another route. This was accom¬ 
plished by the reduction of (£) with diphenylsilane to yield a mixture of 
products in an unspecified yield: 













23 



(48) 


In an attempt to convert the thioacid chloride ( 47) to a mixture of 

39 

isomeric acid chlorides (5) with sulfuryl chloride , an unexpected pro¬ 
duct was obtained (49). The ring open product was characterized by its ir, 
nmr, and elemental analysis. 


(47) 


S0 2 C1 2 


H 3 C 


*v 


* 


ch 7 ch„ s 

I 3 I 3 II 
CH — C — P 
I 

CH„ 


Cl 

Cl 


(49) 


The formation of ( 49) can be visualized to have occurred by either 

40 

a radical or a polar mechanism. In the radical mechanism, a chlorine 
atom could abstract an atom of hydrogen from one of the “-methyl groups. 
The resulting radical ( 50) would undergo ring-opening to yield the phos¬ 
phorus radical ( 51) which would subsequently abstract a chlorine atom to 
give the product (49): 


r* 

H — CH, 
Cl 




CH. 


N 


-> 


(47) 


Cl 




(50) 


Cl 


FS0 2 C1 

P=s 

cl 


(49) 


41 + 

The polar mechanism would involve an attack by (S0 2 C1) ion as shown 

below: 


\ 


s 4- S0 2 CL 




CH. 


C 1 


\ 


C 1 


© 


P=S 

Cl 


( 49 ) 
































24 


An attack at the sulfur atom of the thioacid chloride may also be envisioned, 
which is in keeping with the fact that the acid chloride (^) does not react 
with sulfuryl chloride. 

As pointed out earlier, the chlorophosphetane ( 48) was obtained as 

a mixture of isomers in which the predominant isomer was trans . The assign- 

1 31 

ment was based on H and p nmr data. Treatment of a 1:2 ( cis : trans) 
mixture of ( 48) with phenyllithium gave a 2:1 ( cis : trans) mixture of pheny- 
lphosphetane ( 41) . This conclusively shows that nucleophilic substitution 
at phosphorus in this compound proceeded with inversion of configuration: 


H 



(48) 


:P V (1:2) 

^CJ (cis: trans) 



(41) 
38 


Similar conclusions have been reported by Trippett . Quaternization of 
the phosphetane ( 48) would be expected to yield a salt analogous to the in¬ 
termediate (4) formed during the cyclization reaction (p. 3). Treatment 
of (48) with methyl bromide followed by recrystallization from acetonitrile 
gave a compound which showed the absence of chlorine in the elemental an¬ 
alysis. The product gave a satisfactory analysis for a compound in which 
the chloro substituent had been replaced by a hydroxyl group. In order to 
verify the proposed structure (53), the compound was treated with ethyl- 
diazoacetate. The products of the reaction were identified as ethyl brom- 
oacetate (by glpc) and 1,2,2,3,4,4-hexamethylphosphetane oxide (54). 





















































25 


The addition of the elements of hydrobromic acid to ethyl diazoace¬ 
tate supports the structure ( 53) , although a pentacovalent structure ( 55) 

42 

cannot be completely ruled out. Recently, Petrov and co-workers have de¬ 
monstrated that a hydroxyphosphorane ( 56) could be converted to the cor¬ 
responding methoxyphosphorane ( 57) by treatment with diazomethane: 




(55) 


N CHCOOEt 



P—OCILCOOEt 

XI 2 

H 3 C Br 


By consideration of this analogy, it may be argued that if the compound had 
structure ( 55) , the corresponding alkoxy compound would have been formed on 
treatment with ethyl diazoacetate. However, it should be emphasized that 
little is known about hydroxyphosphonium salts ( 58) or hydroxyphosphoranes 
(59) and the analogy drawn here may not be valid. 


© 

P—OH 
0 

Br (58) 



( 59 ) 
































26 


In any event, the stage at which incorporation of the hydroxy group occurs 
is not clear. It is likely that it occurred either during the work-up or 
through the moisture in the solvent during quaternization (ether) or re- 
crystallization. 

3. Experimental 

Nuclear magnetic resonance spectra were recorded on a Varian Associ¬ 
ates Model A-60 spectrometer with tetramethylsilane as an internal standard. 

31 1 

The P- H decoupling experiments were performed on an NMR Specialties Mo¬ 
del HD-60A heteronuclear spin decoupler in conjunction with A-60 spectro¬ 
meter. The infrared spectra were recorded on Perkin-Elmer Model 21 and 
237 spectrophotometers. Microanalyses were performed by Alfred Bernhardt, 
Microanalytisches Laboratorium, Mulheim (Ruhr), West Germany. All boiling 
points and melting points (Thomas-Hoover apparatus) are uncorrected. Sol¬ 
vents during the reaction work-up were removed with a rotating evaporator. 
Reactions involving trivalent phosphorus compounds were conducted and pro¬ 
cessed under an atmosphere of nitrogen. 

Synthesis of l-Chloro-2,2,3,3,4-penatmethylphosphetane 1-Oxide (7); 

To a mixture of 13.15 (0.1 mole) of anhydrous aluminium chloride and 13.75 
(0.1 mole) of phosphorus trichloride in 95 ml of methylene chloride in a 
3-necked round bottomed flask equipped with a mechanical stirrer, an addi¬ 
tion funnel, a thermometer, a nitrogen inlet and an anhydrous calcium chlo¬ 
ride guard-tube, was added 11.2g (0.1 mole) of 3,4,4-trimethyl-2-pentene 

O 

in 10ml of methylene chloride at 0-5 c with stirring over 90 min.. The mix¬ 
ture was stirred overnight and then quenched by pouring over 200g of ice 
with stirring. The two layers were separated, the aqueous layer was ex¬ 
tracted with methylene chloride, and the combined organic layers were 





























































27 


washed with ice-cold water. The organic layer was dried over anhydrous 
sodium sulfate and evaporated to give 18.5 g brown semi-solid. Recrystal- 

O 

lization from benzene-petroleum ether (30-60 ) gave 9.8g (50%) of white 

O 

crystalline solid, m.p. 114-119 . The nmr (DCCl^) showed peaks at x=8.54 

(3H, d, 3 J pccH =22 cps), t= 8.72 (3H, d, 3 J pccH =22 cps), x=8.77(3H, 2d, 3 PCCH= 

3 2 2 

23.5 cps, J H cch =7 * 5 c P s )» t= 6.79 (lH,2q, J pcH =14.5 cps, J h CH =7 *° c P s )> 

x=8.92 (6H,s). 

Synthesis of 1-Hydroxy-2,2,3,3,4-pentamethylphosphetane 1-Oxide (9) 
from (7). A mixture of l.Og (0.0051 mole) of acid chloride (7) and 10ml 
of 1M sodium hydroxide solution was stirred overnight. The resulting 
mixture was extracted twice with benzene and then the aqueous layer was 
acidified with concentrated HC1. The solution was extracted with methy¬ 
lene chloride and the organic layer was dried over anhydrous sodium sul¬ 
fate. Evaporation of the solvent gave 0.8g pale yellow oil which solidi- 

O 

fied on cooling, m.p. 103-104.5 . The nmr (DCCl^) exhibited peaks at 
T =8.94 (3H, 2d, 3j pcch =21 - 5 cps, 3j hcch =7 -° cps), x=8.91 (3H,d, 3 J pccH = 

19.5 cps), t= 8.78 (3Hd, 3j pcch =19 *° cps), T =8.99 (6H,s), x=7.0-7.7 (1H, 
ring, m), x=1.19 (lH,s). 

O 

The compound was sublimed three times in vacuo at 100 (0.05mm) to ob¬ 
tain the analytical sample, a white powder, m.p. 104-104.5°. 

Anal. Calcd. for C 0 H 0 P : C, 54.53; H,9.72; Found: C, 54.29; H, 9.60. 
Synthesis of l-Methoxy-2,2,3,3,4-pentamethylphosphetane 1-Oxide (8J) 
from (7). A solution containing 0.1 mole of sodium methoxide was prepared 
from 2.3g (O.lg atom) sodium metal in 100 ml absolute methanol. To this 
stirred solution at room temperature was added a solution of 19.5g (0.1 mole) 









28 


of acid chloride (7) in 100 ml methanol over 1 hr*. The temperature of the 
reaction mixture was maintained near 20° with an ice-water bath. The mix¬ 
ture was stirred overnight, the solvent evaporated, and the residue was washed 
with 400 ml methylene chloride in portions and filtered. Evaporation of the 
filtrate gave 23.Og crude liquid. A bulb to bulb distillation ill vacu o 
(30-40° at 0.2 mm) gave 18.5g liquid that solidified on standing. The hy¬ 
groscopic compound was recrystallized from petroleum ether (30-60°) to give 
a crystalline solid m.p. 105-107°. The nmr (benzene) showed peaks at:T = 

6.39 (3H, d, 3 J pocH =10.0 cps),x=9.06 (3H, d, 3 j pcch = 19.6 cps),t=8.80 (3H, 
d, 3 JpccH =19 * 5 cps),T=8.91 (3H, 2d, 3 Jp CCH =21.5 cps. 3j j| CCH =7 - 5 cps),x = 

9.21 (6H,s), t =7.37 (1H, 2q, 2 J pcH =14.5 cps, 2j hch =7 -0 cps). 

Synthesis of l-Chloro-2,2,3,4-tetramethylphosphetane 1-Oxide ( 10) . To 
a mixture of 13.15g (0.1 mole) anhydrous aluminum chloride and 13.75g (0.1 
mole) phosphorus trichloride in 95 ml of methylene chloride in a 3-necked 
round bottom flask equipped with mechanical stirrer, an addition funnel, a 
thermometer, a nitrogen inlet and an anhydrous calcium chloride guard-tube, 
was added 9.8g (0.1 mole) 4,4-dimethyl-2-pentene in 10 ml of methylene chlo¬ 
ride at 0-5° with stirring over 105 min. . The mixture was stirred overnight 
at room temperature and then quenched by pouring over 200 g of ice with stir¬ 
ring. The two layers were separated, the aqueous layer extracted with methy¬ 
lene chloride, and the organic layers were dried over anhydrous sodium sulfate, 
and then evaporated to give 17.5g solid. Recrystillization from petroleum 
ether (30-60°) gave a white crystalline solid m.p. 99-103°, 20.5-24.Og (56-66%). 
The nmr (benzene) showed peaks atx=7.52 (2H, m), t=8.95 (3H, d, J=22.0 cps), 
t=9.00 (3H, d, J pccH = 2 3.5 cps), t=9.25 (3H, 2d, 3j pcch =14.0 cps), 3 J HCCH = 7 -0 
cps), t=9.44 (3H, d, 3j hcch =7 -° cps). 














































29 


Synthesis of l-Methoxy-2,2,3,4-tetamethylphosphetane 1-Oxide ( 11) . A 
solution of 0.1 mole of sodium m$thoxide:was prepared by dissolving 2.3g 
(O.lg atom) of sodium metal in absolute methanol. To this stirred solu¬ 
tion was added a solution of 18.05g of acid chloride ( 10) in 100 ml of 
methanol at room temperature for 1 hr.. The temperature of the exothermic 
reaction was maintained near 20° with an ice-water bath. A white preci- 
piate formed during the addition. The mixture was stirred overnight, the 
solvent evaporated and the residue was washed with 400 ml of methylene 
chloride, in portions and filtered. Evaporation of the solvent from the 
filtrate gave 18.5g crude viscous liquid. Distillation in vacuo gave a 
clear liquid b.p. 62.0-62.5° (0.1 mm) which solidified on cooling (85%). 

The nmr (benzene) showed the liquid to be a single isomer with peaks at 
t=6 .42 (3H, d, 3 J pccH =10.0 cps), t= 7.30-8.00 (2H, m), t= 8.98 (3H, d, 3 J 
PCCH =19 '° c P s )> t = 9.05 (3H, d, 3j pcch “ 18 - 5 cps), t= 8.99 (3H, 2d, 3 J pGCH = 
20.0 cps), 3 J hcch = 7.0 cps), T =9.25 (3H, broad d, 3j hcch =6 * 5 c ps)• 

Synthesis of 1-Hydroxy-2,2,3,4-tetramethylphosphetane 1-Oxide ( 12) . To 
l.Og of acid chloride ( 10) was added 10 ml of 10% sodium hydroxide solution 
and the mixture was stirred overnight. The resultant mixture was extracted 
with benzene and then the aqueous layer was acidified with concentrated HCI 
The solution was extracted with methylene chloride, the combined organic 
layers were washed with water and dried over anhydrous sodium sulfate. 
Evaporation of the solvent gave 0.82g of oil that solidified on standing. 
Recrystallization from hexane gave 0.75g of white powder-like solid m.p. 
71-74° (94%). The nmr (DCCI^) showed peaks at: T =8.78 (3H, d, 3j pcch = 

19.5 cps, t =8.85 (3H, d, 3 J pCGH =19.0 cps),x =8.80 (3H, 2d, 3 j pcch = 20.0 cps, 

3j PCCH =7 -° c P s )> t=9 -° ( d > 3j HCCH =7 '° cps ),t =7.30-8.00 (1H, m),x=1.76 
(1H, s). Anal . calcd. for: C^H^^O^P: C,52 17, H, 8-76. Found: C,5196; H, 


8.90. 








































































30 


Synthesis of l-Methoxy-2,2,3,3-tetramethylphosphetane 1-Oxide ( 15) . 

To a mixture of 66.7g (0.5 mole) of anhydrous aluminum chloride and 68 . 78 g 
(0.50 mole) of phosphorus trichloride in 200 ml of methylene chloride at 
0-5° was added with stirring, 49.Og (0.5 mole) of 2,3,3-trimethyl-l-butene 
in 200 ml of methylene chloride over 3 hr. under anhydrous conditions. 

The mixture was stirred overnight and then the reaction was quenched by 
pouring it over 400g of ice. The two layers were separated, the aqueous 
layer extracted with methylene chloride and the combined extracts were 
dried over anhydrous sodium sulfate. Evaporation of the solvent gave 83g 
of crude viscous liquid. 

To the stirred solution of 83g of the above crude liquid in 325 ml of 
anhydrous benzene was added 45.2g (0.45 mole) of trieth lamine followed by 
15.3g (0.5 mole) of methanol in 125 ml of benzene over 1 hr.. After the exo¬ 
thermic reaction ceased, the mixture was allowed to reflux for 2 hr.. A 
crystalline precipitate formed which was filtered, and washed with benzene. 
Evaporation of the solvent gave a pale brown liquid which turned to a semi- 
solid on standing. A bulb to bulb distillation (25-30° at 0.1 mm) gave 

40.lg of plates m.p. 50-52° (23%) over all. The nmr (benzene) showed peaks 

3 3 

at t=6 .45 (3H, d, J pocH -H*0 cps), t= 9.00 (3H, d, J pCGH =19.0 cps), t= 
8.88 (3H, d, =15.5 cps), t=9.05 (3H,s), t=9.03 (3H,s), t=7.81 (1H, 

r LLn 

d, 2j p CCH =15 * 5 C P S )> t= 7.90 (1H, d, 2 J pcH =14.5 cps). 

Compound ( 15) was recrystallized from petroleum ether (30-60°) and sub¬ 
limed in vacuo at 25-30° (0.2-0.4 mm) to obtain plates, m.p. 52-55°. Anal . 

Calc, for C 0 H OP: C,54.53; H,9.73; P, 17.58. Found: C, 54.59; H,10.02; 

O 17 2 

P,17.56. 

Prepatation of 1-Hydroxy-2,2,3,3-tetramethylphosphetane 1-Oxide ( 13) . 

A mixture of l.Og (0.57 mole) of the ester ( 15) and 10 ml of 10% sodium hy- 


































































31 


droxide was stirred overnight. Washing the solution with benzene followed 
by acidification with concentrated HC1 gave an oil which was extracted with 
methylene chloride. The organic layer was washed with water and heated over 
anhydrous sodium sulfate. Evaporation of the solvent gave 0.92 (100% of 
solid, m.p. 175-183°)The nmr (DCCl^) exhibited peaks at: x=8.83 (6H, d, 

3j PCCH =20 '° CpS ^ T=8 * 88 ( 6H > 2s )> t= 7.61 (2H, d, 2 J pcH =15.5 cps). 

Preparation of l-Chloro-2,2,3,3-tetramethylphosphetane 1-Oxide ( 14 ). 

To 14.lg (0.087 mole) of the acid ( 13) was added 11.9g (0.1 mole) of thio- 
nyl chloride. The resulting mixture was heated to reflux for 4 hr.. The 
excess of thionyl chloride was evaporated and the residue was distilled in 
vacuo at 66° (0.2 mm) to give 12.lg (77%) of the acid chloride (14). The 

3 

nmr (benzene) showed peaks at t=8.75 (3H, d, *JpQ{ =2 ® C P S )> T= 8.85 (3H, 

3 2 

d, JpcchT 25 C P S )» T= 8 ‘^ 2 C^H, broad s,), and t=7.27 (2H, d, J pcH = 

16.0 cps) (*). 

Preparation of l-Phenyl-2,2,3,4-tetramethylphosphetane 1-Oxide ( 16) . 

To a homogenous solution prepared from 35.8g (0.2 mole) of phenylphospho- 
nous dichloride and 26.7g (0.2 mole) of anhydrous aluminum chloride in 
280 ml of methylene chloride at 0-5°, was added 19.6g (0.2 mole) of 4,4- 
dime thyl -2 -pentene in 60 ml of methylene chloride over 2 hr. with stirring 
and under anhydrous conditions. The mixture was stirred overnight and the 
reaction was quenched by adding 200 ml of ice-water dropwise to the stirred 
mixture of 0-5° over 2 hr.. The two layers were separated, the aqueous layer 
extracted with methylene chloride and the combined extracts were washed with 
water and dried over anhydrous sodium sulfate. Evaporation gave 47.lg of a 


"£*■) This peak is an apparent doublet but is probably a more complex 

system. 





















































































































32 


crude viscous liquid. Distillation of the liquid in vacuo gave 10.Og li¬ 
quid, b.p. 110° (0.2-0.4 nun) which solidified on cooling (22.4%). The nmr 
(CCl^) showed one isomer with peaks at: x=1.9-2.6 (5H,m), t=9.14 (3H, d, 

3j PCch =19 -° c P s )> t=8.71 (3H, d, 3 j pcch = 17-0 cps), t= 8.70 (3H, 2d, 3 J 

PCCH =18,5 cpS> J HCCH =7 * 5 CpS ' )> t= 8.90 (3H, 2d, Jp CCCH =l cps, J HCCH = 

7.0 cps), t= 6.30-7.20 (lH,m), x= 7.30-8.10 (lH,m) 

Synthesis of l-Phenyl-2,2,3,4-tetramethylphosphetane (17).A complex 
of 4.9g (0.062 mole) pyridine and 8.7g (0.065 mole) trichlorosilane in 50 
ml benzene at 0-5° was prepared under anhydrous conditions and under a 
flow of nitrogen. To this mixture was added with stirring, 4.5g (0.02 m) 
of the phosphine oxide ( 16) in 60 ml of benzene at 0-5° for 2 hr.. The cold 
bath was removed and the mixture was allowed to reflux for 3.5 hr.. The mix¬ 
ture was stirred overnight and then quenched by adding 100 ml of 20% sodium 
hydroxide solution dropwise at 0.5° until two clear layers were obtained. 

The aqueous layer was separated, extracted with benzene and the combined 
layers were washed with water and saturated sodium chloride solution. Dry¬ 
ing of the organic layer over sodium sulfate followed by evaporation gave 
2.5g (60%) of a liquid with a characteristic phosphine odor. 

Preparation of 1-Methyl-lphenyl-2,2,3,4-tetramethylphosphetanium Io - 

dide ( 18) . The above liquid was dissolved in 20 ml of anhydrous ether and 
treated with 5 ml of methyl iodide and allowed to stand overnight. The pre¬ 
cipitate was filtered, washed with ether and then immediately recrystalli¬ 
zed from acetonitrile-ether acetate to give 2.30g of pale yellow needles 
m.p. 209-213° (decomposition) (32% over all). The nmr (DCC1 3 ) showed peaks 
at: t= 7.35 (3H, d, 2 J pcCH =13.5 cps), x=8.37 (3H, d, 3j pcch =21 -0 cps), x= 
8.74 (3H, d, 3 J pCGH =21.0 cps), x=8.81 (3H, 2d, 3j hcch =7 -° cps, 4j pcCH = 

1.0 cps) x=8.44 (3H 2d, 3 J pcCH =22.0 cps, 3j hcch =7 -0 cps), x=5.50-6.25 
(1H, m), x=6.50-7.30 (1H, m), x=1.50-2.50 (5H, aromatic). 































33 


Anal , calcd. for C 14 H 22 IP: C, 48.29; H, 6.37; I, 36.45. Found: C, 48.36; 

H, 6.51; I, 36.45. Found: C, 48.56, H, 6.31. 

Preparation of 1-Methylene-2,2-diraethylcyclohexane . To 184g (0.515 
mole) of triphenylmethylphosphonium bromide suspended in 1.1 of ether was 
added 400 ml of 1.3 M n-butyilithium in pentane solution at 0°. The orange 
colored solution was stirred for several hours at room temperature. 

To the above solution at 0° under nitrogen was added 50g (0.397 mole) 
of 2,2-dimethylcyclohexanone in 200 ml of ether over 4 hr. with stirring. 

A precipitate formed with the contact of the two solutions. The stirring 
of the reaction mixture became more difficult as the reaction progressed and 
so the mixture was allowed to stand overnight. The reaction was quenched 
by pouring the mixture over ice and the precipitated solid was filtered. 

The filtrate was separated, and the ether layer was dried over anhydrous 
sodium sulfate. Evaporation of the ether followed by distillation of the 
crude liquid in a spinning band column gave 20g liquid at 139-142° at at¬ 
mospheric pressure. (40%). The ir of neat liquid showed a strong absorption 
at 1640 cm" 1 . The nmr (neat) showed peaks at: t=5.38 (2H, s, vinyl), t= 
7.82 (2H, m, allyl), t= 8.52 (6H, m, alkyl), x=8.98 (6H, s, methyl). 

Preparation of 1,6-Dimethyl-7-phenyl-7-phosphabicyclo^-(4,2,0)-octan e 
1-Oxide . To a homogenous solution of 25g (0.14 mole) of phenylphosphorous 
dichloride and 20g (0.15 moel) of anhydrous aluminum chloride in 500 ml of 
methylene chloride was added a solution of 19g (0.153 mole) of the above al- 
kene in 500 ml of methylene chloride at 0° over 4 hr. with stirring. The 
mixture was poured over ice, the layers were separated and the methylene 
chloride layer was washed with saturated sodium chloride solution and dried 
over anhydrous sodium sulfate. Evaporation of the solvent gave a viscous li¬ 
quid which was distilled b.p. 155-160° (0.1 mm) to obtain 10.6g (28%) of a 
colorless liquid. 



























34 


Preparation of 1,6,7-Trimethyl-7-phenyl-7-phosphonia-bicycle(4,2,0) " 
octane Bromide (19). To a stirred solution of 10.Og (0.04 mole) of the phos¬ 
phine oxide described above in 300 ml of benzene under nitrogen, was added 
5.1g (0.05 mole) of triethylamine followed by 6.9g (0.05 mole) of trichlo- 
rosilane. The nixture was heated to reflux for 15 hr., cooled and quenched 
by adding 80 ml of 20% sodium hydroxide solution dropwise with stirring. 

The layers were separated, and the benzene layer was washed well with sa¬ 
turated sodium chloride solution and dried over anhydrous sodium sulfate. 

The solvent was evaporated and the resulting material was dissolved in ben¬ 
zene-ether, and treated with excess of methyl bromide in a pressure flask 
for 2 days. The resulting glassy solid was recrystallized from acetonitrile- 
ethyl acetate to give 3.0g of a solid, m.p. 191-194°. The nmr (DCCl^) showed 
the compound to be a single isomer having peaks at: x=1.25-2.60 (5H, m, 
aromatic), t= 6.60 (2H , 2d,Jp^=13.5 cps, cps), x=7.20 (3H, d, 

2jpcH = 14.0 cps), x=8.0-9.0 (14 H, broad absorption). 

The compound (19) was recrystallized several times for the analytical 
sample m.p. 192-195°. Anal . calcd. for C^H^PBr: C,58.71; H, 7.39, P,9.39; 
Br, 24.48. The compound has been submitted to Prof. Trefonas for X-rays in¬ 
vestigation . 

Preparation of l-Benzyl-2,2,3,4,4-pentamethylphosphetane 1-Qxide ( 25) . 

A Gringnard reagent was prepared from 25.2g (0.224 mole) of benzyl chloride 
and 4.75g (0.2g atom) of magnesium turnings in 150 ml of anhydrous ether. 

The above Grignard reagent was added to 38.9g (0.2 mole) of acid chlo¬ 
ride (5) in 500 ml of ether at 0-5° with stirring over 2 hr.. The mixture 
was stirred for an additional one half hr. and then the reaction was quench¬ 
ed with a saturated solution of ammonium chloride. A white solid separated 
which was dissolved by adding an excess of ether. The organic layer was se- 











































































35 


parated, washed with 5% sodium bicarbonate and then dried over anhydrous 
sodium sulfate. The ether solution on evaporation gave a white solid 
which was recrystallized from cyclohexane to obtain 30g of a crystalline 

solid, m.p. 187-188° (60%). The nmr (DCCl^) showed peaks at: 1=2.40-2.92 

2 3 

(5H, m, aromatic), x=6.75 (2H, d, J pc ^=10.5 cps), x=8.78 (6H, d, 

15.5 cps), x=8.85 (6H, d, 3,J pccH =17 • 5 cps) , x=9.12 (3H, dd, ^ J p CC ^ H = 

3 3 

1.8 cps, J hCCH = 2 ‘ 3 cps), x=8.02-8.50 (1H, m, ^pcch = 2 ‘ 5 c ps) . Anal . 

calcd. for C, 72.00; H, 9.20. Found: C, 72.16; H, 9.08. 

Reduction and Quatemization of (25); Preparation of 1-Benzyl-1,2 ,2, 

3,4,4-hexamethylphosphetanium Bromide ( 24) . To a complex prepared from 
4.05g (0.03 mole) of trichlorosilane and 3.5g (0.035 mole) of triethylamine 
in 90 ml of benzene was added a solution of 7.5g (0.03 mole) of the oxide 
( 25) in 120 ml of benzene over 10 min. at 0-5° with stirring under an at¬ 
mosphere of nitrogen. The cold bath was removed and the mixture was allowed 
to reflux for 3 hr.. The reaction was quenched by adding a 20% sodium hy¬ 
droxide solution at 0-5° until two clear layers were obtained. The organic 
layer was separated, the aqueous layer was extracted with benzene and the 
combined benzene layers were washed with saturated sodium chloride solution 
and then dried over anhydrous sodium sulfate. Evaporation of the solvent 
gave a white solid which was dissolved in 20 ml of ether and then treated 
with 5 ml of methyl bromide in a sealed flask for 24 hr.. The precipitated 
solid was filtered and recrystallized from acetonitrile to obtain 7.2g of 
needles, m.p. 207-214° (80%). Repeated recrystallization raised the ra.p. 
to 213-218°. The nmr (DCCl^) showed peaks at: x=2.23-2.97 (5H, m, aromatic), 
x=5.23 (2H, d, 2 J pcH =14.5 cps), x=6.75-7.25 (1H, m), x=7.82 (3H, d, 2 J pcH = 

13.5 cps), x=8.31 (6H, d, 3 J pccH =18.5 cps), x=8.60 (6H, d, 3j pcch 19 - 5 cps), 

3 4 

x=8.98 (3H, dd, J HCCH = 7 - 0 cps, J PCCCH =1 °ps). 











































36 


Preparation of the Dibenzyl Salt ( 26) . The phosphine was prepared as 
described above and then treated with 5 ml of benzyl bromide and the mix¬ 
ture was allowed to stand in a sealed flask for 4 days. The precipitate 
was filtered, washed with ether and then recrystallized from acetonitrile 
to give 6.3g of crystalline needles m.p. 219.5-222.50. (50%). The nmr 
CF^COOH) showed peaks approximately at: t=2.70-3.50 (10H, m, aromatic), 
t=6 .30 (4H, 2d, each 2 J pCH =12.5 cps), x=8.60 (6H, d, 3 pcqi = 1 9.3 cps), t = 

8.65 (6H, d, 3 J pcCH =19.8 cps), x=9.08 (3H, dd, 3 J HCCH =7.0 cps), 4 J pccCH = 

1 cps). Anal . calcd. for C^^QBrP: Br, 19.71. Found: Br, 19.66,19.85. 

Preparation of 1-(1-Naphthyl)-2,2,3,4,4-pentamethylphosphetane 1-Oxide . 

A Grignard reagent was prepared from 25g (0.1 mole) of 1-iodonaphthalene 
and 2.5g (0.1 g. atom) of magnesium turnings in 200 ml of ether. This so¬ 
lution was added to a solution of 22g (0.11 mole) of the acid chloride (5) 
in 300 ml of ether with stirring in a three-necked flask under nitrogen at 
0° over 40 min.. The reaction was allowed to start at room temperature for 
6 hr.. A viscous oil formed at the bottom of the mixture. The reaction was 
quenched with a saturated solution of ammonium chloride at 0°. An insolu¬ 
ble precipitate was formed which was filtered and recrystallized from etha¬ 
nol-water and subsequently from cyclohexane to give 3.5g (11.2%) of the 
product m.p. 177-181°. 

The ether layer was evaporated and the residue was treated with pe¬ 
troleum ether (30-60%). The ppt. was collected, recrystallized and sublimed 

to give 0.6g of additional product m.p. 179-182°. The nmr (DCCl^) showed 

3 3 

peaks at: x=1.55-2.68 (7H, m) , x = 7.40 (1H, 6, ^HCCiT^*^ C P S > ^PCCH = 

3.0 cps), x=9.01 (3H, 2d, 3j hcch =7 -° C P S » 4j pccCH =1 ' 2 cps ^’ t= 8.40 (6H, 
d, 3 J pGCH =16.5 cps), x=8.75 (6H, d, 3 J pcCH =18.0 cps). Anal . calcd. for 


C 0 H OP: C,75.49; H,8.10. Found: C, 75.45; H, 8.17. 
lo 23 


































































































































37 


Preparation of 1-(1-Naphthyl)-1,2,2,3,4,4-hexamethylphosphetanium 

Bromide (31). To a solution of 8.3g. (0.029 mole) of the above phosphe- 
tane oxide in 170 ml of benzene was added 3.2g (0.032 mole) of triethyla- 
mine followed by 4.5g (01132 mole) of trichlorosilane. The mixture was 
heated to reflux for 2 hr. under nitrogen atmosphere, cooled and treated 
with 20% sodium hydroxide solution until distinct layers were obtained. 

The two layers were separated, the organic layer was washed with water, 
dried over anhydrous sodium sulfate and evaporated to give the phosphine 
which was dissolved in ether and treated with an excess of methyl bromide 
in a sealed flask. After standing for several days, the product was fil¬ 
tered to give 10.3g solid m.p. 248-254°. The nmr (trifluoroacetic acid) 
showed the salt to be a mixture of isomers with peaks at: x=9.00 (3H, 2d, 

3j hcch = 7- ° cps 4 j pccch = 1 ' 3 cps )> T=7 * 74 ( 3H > d > 2 j pch = 13, ° CpS ' ) ’ T= 

7.10 (1H, 0, 3 J pcGH =2.0 cps, 3 J HCCH =7.0 cps), t= 8.42 (6H, d, % CCH =20.5 
cps), t= 8.60 (6H, d, 3 JpcQj = 19*0 cps), t= 1.70-2.80 (7H, m, aromatic). 

The product was recrystallized from acetonitrile-ethyl acetate to give 
the analytical sample. Anal . calcd. for C-^H^P Br: C, 62.47; H, 7.17. 
Found: C, 62.44; H, 7.42. 

Alkaline Decomposition of the Salt (31). A solution of 3.8g (0.01 mole) 
of the salt ( 31) was treated with l.Og of sodium hydroxide in 3 ml of water 
at 0° with stirring. The product was extracted in benzene and the benzene 
solution was washed with water, dried over anhydrous sodium sulfate and was 
evaporated to a viscous oil. The nmr (DCCl^) exhibited peaks at: x=6.50- 
6.90 (2H, m, allyl), 1=3.70-4.70 (2H, m, vinyl), t=8.20 (1H, m), x=8.62 
(3H, d, 3 J pccH =12.0 cps), x=8.72 (3H, d, 3 J pccH =14.0 cps), x=8.74 (3H, 
d, 3 J pcCH = 15 -° cps), x=8.89 (3H, s), x=9.30 (3H, s). 

The product from the above reaction was dissolved in 150 ml of decalin 











































































38 


and heated with lg of 30% palladium on charcoal at reflux for 4 days, while 

a stream of nitrogen was bubbled through the reaction mixture. The mixture 

was cooled, filtered, and evaporated (40°, 0.1 mm) to give a solid residue 

(1.2g). Recrystallization of the product from cyclohexane gave 900 mg. of 

the pure compound m.p. 173-176°. The ir (nujol mull) of the product showed 

-1 

a strong absorption at 1150 cm . The nmr (DCCl^) showed peaks at: t=2.00 
-2.70 (6H, m, aromatic), t=8.28 (3H, d, 2 J pCH =12.0 cps), x=8.91 (3H, d, 

3j HCCH =7 -° Cps) ’ T=8 - 60 ( 3H > s )> t=8.72 (3H, si, x=8.78 (3H, d, 3 J pcCH = 

14.0 cps), x=8.72 (3H, d, 3 ^pcch = ^’ 0 C P S ) • Anal, calcd. for: C^R-^OP: 

C, 75.96; H, 8.39; P, 10.31. Found: C, 75.83; H, 8.52; P, 10.42. 

Synthesis of precursors for (34) . 

Preparation of 2-Phenyl~2-bromocinnamic-Acid . In a 500 ml three-necked 
round bottomed flask equipped with a condenser, a mixture of 80.g (0.43 mole) 
of <3-bromobenzldehyde, 58.8g (0.43 mole) of phenylacetic acid, 43. 6g (0.43 
mole) of triethylamine and 64.7g (0.63 mole) of acetic anhydride was re¬ 
fluxed for 3 hr. (temperature of the mixture 164-165°). A fresh 10 ml por¬ 
tion of triethylamine was added and the mixture was allowed to reflux for an 
additional hour. The reaction was cooled and the resulting brown viscous 
liquid was dissolved in 500 ml of benzene. The benzene solution was slowly 
treated with 20% sodium hydroxide solution in a separatory funnel. On shak¬ 
ing the latter, three layers were formed. The thick bottom layer was the 
solution of the salt of the desired product. Separation, acidification with 
6M HC1, with cooling, gave a precipitate which was dissolved in benzene. The 
benzene solution was washed with water, dried over anhydrous sodium sulfate 
and evaporated to give a solid. Recrystallization of the solid from ethyl 
acetate (using Norit) gave lOOg of crystalline solid m.p. 178-182°. 









































39 


Preparation of 1-Bromostibene . A mixture of 50g (0.166 mole) of 2- 
phenyl-2-bromocinnamic acid, 244 ml of quinoline and 4.0g of copper chro¬ 
mite catalyst was heated in a round bottomed flask with a condenser to 
2100-225° and maintained at that temperature for 80 min. during which evo¬ 
lution of carbon dioxide occurred. After the evolution of the gas had 
ceased, the mixture was cooled. Removal of quinoline in vacuo gave 48.7g 
of a dark brown liquid. Distillation of the liquid residue gave a pale 
yellow liquid, b.p. 130-135° (0.6-0.7 mm) (29.Og). The product was dis¬ 
solved in 75 ml of ether, washed with 10% HC1 followed by saturated sodium 
bicarbonate solution. 

The ether solution was dried over anhydrous sodium sulfate. Distilla¬ 
tion of the product gave 19.5g of the pale yellow liquid, b.p. 115-118° 
(0.2-0.3 mm) (44%). 

Preparation of 1-Bromophenanthrene . In a quartz flask with cooled 
copper tubing inside for cooling, was placed 3-5g of the above 1-bromo- 
stilbene and 100-200 mg iodine in 2 1 of cyclohexane. The solution was 
irradiated using Hanovia lamp (450 watts) surrounded by a corex filter. 

The temperature of the solution was maintained between 25-30°. The pro¬ 
gress of the reaction was followed by taking aliquots and observing the 
disappearance of 960 cm * peak of stilbene in the ir. After the peak di¬ 
minished (34-36 hr.), a fresh portion of the stilbene and iodine was added 
and the process continued until a total of 19.5g of stibene had been added. 
Evaporation of the solvent gave a crude semi-solid. Recrystallization of 
the product from ethanol yeilded 7.55g 1-bromophenathrene, m.p. 105-108° 
Recrystallization from acetonitrile raised the m.p. to 110-112°. (109.5 - 
110°) (43). 
































. 




























40 


Preparation of l-(l-Phenanthryl)-2,2,3,4,4-pentamethylphosphetane 1 - 

Qxide (34) . A solution of 5.9g (0.023 mole) of 1-bromophenanthrene in 30 ml 
of warm benzene was added dropwise to 0.32g (Q.046g atom) of cut lithium 
wire in 40 ml of anhyd ether. The mixture which gradually turned deep brown 
in color was allowed to reflux for 3 hr. and then allowed to stand for 5 hr. 
at room temperature. A test with Gilman's reagent was positive. 

The above organolithium compound was added to 10.5g (0.054 mole) of the 
acid chloride (5) in 250 ml of ether with stirring at -40° over 15 min. The 
mixture was stirred at room temperature overnight and then quenched with a 
saturated solution of ammonium chloride. An insoluble precipitate was formed 
which was filtered and recrystallized from ethanol to yield 0.89g of a so¬ 
lid, m.p. 195-205°. The ether-benzene filtrate was stirred with 100 ml of 
10% sodium hydroxide solution (to remove any unreacted acid chloride) and 
subsequently washed with water, dried and evaporated to give 0.40g of addi¬ 
tional product, m.p. 201-204°. The ir (1160 and 1178 cm” 1 , P[0] ) of each 
of the products was identical. The compound was sublimed at 150° (0.1 mm) 
and recrystallized from ethanol-water and cyclohexane to give flat, silvery 
needles, m.p. 202-204°. The nmr (DCCl^) showed peaks due to the major iso¬ 
mer at: t=1.50-2.70 (9H, m, aromatic), t=8.43 (6H, d, 3j pcch =16 *° cps), 
t= 8.72 (6H, d, 3 J pcCH =17.0 cps), T=8.90 (3H, 2d, 3 J HCCH =7.0 cps , 4 J p(XCH = 
1.8 cps), t=8.10 (1H, m). Anal . Calcd. for C 22 H 25 0P: C,7854; H, 7.49; P, 
9.21. Found: C, 78.26; H, 7.70; P,9.42. 

Preparation of l-(l-Phenanthryl)-l,2,2,3,4,4-hexamethylphosphetanium 

Bromide (35). The 1.25g (0.0037 mole) of the oxide ( 34) in 50 ml of benzene 
was added 500 mg of triethylamine followed by lg (O.oo74 mole) of trichlo- 
rosilane. The mixture was stirred for 5 hr. at room temperature and then 
brought to reflux for 4 hr.. The cooled mixture was treated with 50 ml of 









41 


20% sodium hydroxide solution. The organic layer was separated and washed 
with a saturated sodium chloride solution, dried over sodium sulfate and 
evaporated. The residue was dissolved in ether and treated with 5 ml of 
methyl bromide in a sealed flask. After 2 days, 1.34g (87% overall) of 
solid product was filtered. The salt was recrystallized from ethanol-ethyl 
acetate to produce a solid, m.p. 267-269°. The nmr (DCC1_) showed peaks 

v) 

at: t= 0.75-2.92 (9H, m, aromatic), x=7.21 (3H, d, 2 J pCH =12.5 cps), t= 

8.15 (6H, d, 3 j pcqj = 20.0 cps), x=8.60 (6H, d, ^J pcCH =18.5 cps), x=8.75 

3 

(3H, d, J HC ch =7 -° c P s )> t= 6.50 (1H, m). Anal . Calcd. for C^^H^gBrP: C, 

66.50; H, 6.79; Br, 19.24. Found: C, 66.51; H, 6.69; Br, 19.16. 

Alkaline Decomposition of (35). A solution of 1.09g (26 mmoles) of 
( 35) in 20 ml of water was covered with 20 ml of benzene-chloroform (1:1) 
and stirred in an ice-bath. To the above solution was added dropwise 4 ml 
of 1 N sodium hydroxide solution over 15 min.. The mixture was stirred at 
room temperature for another 20 min. and the organic layer was separated, 
dried over anhydrous sodium sulfate and evaporated to a crude liquid. The 
nmr (DCCl^) showed peaks at: x=1.8-2.75 (6H, m, aromatic), x=.60-4.70 
(2H, m, vinyl), x=6.08-6.47 (2H, m, allyl), x=8.56 (3H, d, 2 J pcp{ =13.0 cps), 

x=8.73 (3H, d, 3 J pcCH =15.0 cps), x=8.92 (3H, s) , x=9.33 (3H, s), x=9.08 

(3H, d, 3j hcch =7 -° C P S )- 

Preparation of (36). The crude product above was dissolved in 150 ml 
of decalin and treated with 500 mg of 30% Palladium over charcoal. The mix¬ 
ture was heated to reflux for 4 days during which a stream of nitrogen gas 
was bubbled through it. The resulting mixture was cooled, filtered and the 
catalyst was washed with benzene. The combined organic filtrates were con¬ 
centrated in vacuo at 90° (10 mm) to reduce the total volume to 10 ml; cool¬ 


ing of the solution gave 310 mg of white precipitate which was filtered. 




















































42 


m.p., 211-216° (35% yield from the last two steps). The nmr (DCCl^) showed 
peaks at: t=Q. 50-2.77 (8H, m, aromatic), t= 8.28 (3H, d, 2 J =11.5 cps), 

PLri 

t= 8.93 (3H, d, 3 j hcch = 7.0 cps), T =8.56 (3H, s), x =8,72 (3H,s), T =8.77 

(3H, d, 3 J pccH =14.0 cps), t=8 .72 (3H, d, 3 J pccH =16.0 cps), x=7.99 (1H, 


.31 


partially hidden). The above assignments were verified by P decoupling 
experiment. 

The product was recrystallized from benzene-petroleum ether (30-60°) 
several times and sublimed at 175° (0.1 mm) to give the analytical sample 
m.p. 250-252°. Anal. Calcd. for C 23 H OP: C, 78.83; H, 7.77; P, 8.84. 

Found: C, 79.00; H, 7.87; P, 8.80. Found: C, 78.83; H, 7.70. 

Preparation of 1-(9-Phenanthryl)-2,2,3,4,4-pentamethylphosphetane- 

1-Oxide . A Grignard reagent was prepared from 53g (0.206 mole) of 9-bro- 
mophenanthrene and 5.5g (0.23g atom) of magnesium turnings in 500 ml ether- 
benzene . 

The above Grignard reagent was added to a solution of 60g (0.3 mole) 
of the acid chloride Q5) in ether solution over 2 hr. at 0° with stirring, 
under an atmosphere of nitrogen. The Gilman test was negative after 2 hr. 
The reaction was quenched with saturated ammonium chloride solution at 0° 
and the mixture was stirred overnight. The precipitate was filtered. The 
organic layer was treated with 100 ml of 10% sodium hydroxide solution with 
stirring for several hours, dried and evaporated. The residue was treated 
with ether-petroleum ether (30-60°) to give a solid. Both products were 
recrystallized from ethanol-water (Norit) to obtain 21.7g of the purified 
solid, m.p. 206-213°. (22%). The nmr (DCCl^ showed peaks due to the ma- 

3 

jor isomer at: t=1.20-2.55 (9H, m, aromatic), t=7.42 (1H, o, J pccH =3 ' 0 C P S » 
3j HCCH =7 '° cps ^ T=8 - 41 ( 6H > 3j PCCH =16 -° cps )» t=8.75 (6H, d, 3j pcch = 


17.5 cps), t= 9.07 (3H, 2d, J 


PCCCH 


=1.6 cps, J hcch = 7.0 cps). 




























































43 


The compound was recrystallized several times from ethanol-water and 
then sublimed at 210° (0.05 mm) to give the analytical sample. Anal . Calcd. 
for C 22 H 25 OP: C, 78.54; H, 7.49. Found: C, 78.43; H, 7.47. 

Reduction of the Oxide and Quatemization . A suspension of 16.8g 
(0.05 mole) of the above oxide in 500 ml of benzene was heated until the so¬ 
lid dissolved. To this solution at room temperature, was added 6g (0.66 
mole) of triethylamine and 8g (0.06 mole) of trichlorosilane. The mixture 
was stirred at room temperature for 1 hr. and then refluxed for 4 hr.. 

The cooled reaction mixture was treated with 125 ml of 20% sodium hydro¬ 
xide solution. The organic layer was separated, washed with water, dried 
over anhydrous sodium sulfate and evaporated to give a solid residue, m.p. 
105-108°. The residue was redissolved in ether and treated with an excess 
of methyl bromide. After allowing the reaction to stand for 24 hr., the 
salt was filtered to yield 18.Og of the product, m.p. 265-269°. Recry¬ 
stallization from ethanol-ethyl acetate gave white crystals, m.p. 270-273°. 

The nmr (DCCl^) showed peaks due to the major isomer at: t=0.83-2.70 

3 3 

(9H, m, aromatic), t=6.48 (1H, o, J hCCH = ^'^ c P s > J PCCH = ^'^ C P S )> t=7.18 
(3H, d, 2 J pCH =13.2 cps), t= 8.08 (6H, d, 3 J pcCH =20.5 cps), i=8.42 (6H, d, 
3 JpCCH = 18.5 cps), x=8.89 (3H, d, broad). Anal. Calcd. for C^^H^gBrP, 19.24. 
Found: Br, 19.56. 

Attempts to convert the above salt to the steroid precursor were un¬ 
successful. Although, the decomposition did occur, a glassy product was 
obtained. The product gave glassy masses after treatment of the crude pro¬ 
duct with Pd/ C in decalin. 

Alkaline decomposition of l-Phenyl-2,2,3-trimethylphosphetane 1-Oxide 

(37) to (38). To a mixture of 2.2g (0.01 mole) of the oxide ( 37) and o.8g 
(0.02 mole) of sodium hydroxide in 10 ml methyl cellosolve (filtered from 









































































44 


ferrous sulfate) was added 4-5 drops water, and the mixture was refluxed 
for 2 hr.v The solvent was evaporated, and the resulting brown semi-solid 
was treated with 15 ml water. The solution was extracted with two 20 ml 
portions of benzene. The aqueous layer was acidified with concentrated 
HC1 which precipitated an oily liquid. The mixture was extracted with me¬ 
thylene chloride, and the combined extracts were dried over anhydrous so¬ 
dium sulfate and evaporated to obtain 2.7g of a brown, oily liquid that 
solidified on scratching. Recrystallization from methanol-water gave 1.8g 
(75%) of acid, m.p. 92.5-96°. The nmr (DCCl^) showed peaks at: x=2.08- 
2.75 (5H, m, aromatic), t=8.00-9.00 (1H, m), t=9.02, (6H, d, 3j pcch = 

17.0 cps) x=9.12 (6H, d, ^J^££ H =7.0 cps), x=2.60 (1H, s). Anal . Calcd. 

for C 12 H lg 0 2 P, 1/2 H 2 0: C, 61.26; H, 8.57. Found: C, 61.68,; H, 8.23. 
Alkaline decomposition of l-Phenyl-2,2,3,3-tetramethylphosphetane 1- 

Oxide ( 39) to (40). To a mixture of 2.36g (0.01 mole) of the phosphine ox¬ 
ide ( 39) and 0.8g (0.02 mole) of sodium hydroxide in 10 ml methyl cello- 
sol ve was added 4-5 drops of water and the mixture was refluxed for 2 hr.. 
The solvent was evaporated and the residual brown liquid was dissolved in 
15 ml water. The equeous solution was extracted with methylene chloride 
and then acidified with concentrated HC1. The precipitate was filtered to 
obtain 2.3g of a pale yellow powder. Recrystallization from methanol-water 
yielded 2.2g (86.5%) of the acid in two crops, each with m.p., 181-184°. 

The nmr (DCCl^) showed peaks at: x=2.06-2.75 (5H, m, aromatic), x=8.98 
(6H, d, cps), x=9.01 (9H, s), x*1.70 (1H, s). Anal . Calcd. for 

C 13 H 21°2 P: C ’ 64>98 > H ’ 8 - 81, Found: C > 64.82; H, 8.66. 

Reaction of l-Chloro-2,2,3,4,4-pentamethylphosphetane 1-Oxide (5) with 

Phosphorus Pentasulfide . In a flask equipped with a water condenser fitted 
to a calcium chloride guard-tube, a mixture of 5.4g (0.027 mole) of the acid 
































































































































45 


chloride (5) and 2.2g (0.009 mole) of phosphorus pentasulfide was heated 
to 150 for 4 hr.. The mixture melted, darkened and a white solid deposi¬ 
ted in the cold part of the condenser. The contents of the flask were 
sublimed in vacuo at 40° (0.2 mm) to produce 4.4g of crystalline solid with 
an adhering oil. Recrystallization from petroleum ether (30-60°) gave a 
crystalline solid ( 47) with a faint characteristic odor, m.p. 107-113.5°, 
yield 4.0g (69%). Repeated recrystallization raised the m.p. to 124-125.5°. 
The nmr (benzene) showed the compound to be a mixture of two isomers (2:1). 

3 

The predominant isomer showed peaks at: t=8.88 (6H, d, Jp ( _,^ H =22.0 cps), 

x=8.86 (6H, d, 3 J pcCH =25.0 cps), t= 9.44 (3H, 2d, 3j hcch = 7 - 0 C P S > 4j pccCH = 

1.0 cps), t= 8.01 (1H, m); and the minor isomer showed peaks at: 8.92 (6H, 

d, 3 J pccH =23.0 cps), t= 8.79 (6H, d, 3 J pcCH =25.0 cps), x=9.37 (3H, 2d, 

3 4 

Jhcch =7 *° c P s > J PCCCH =1 ‘° C P S )> t= 8.01 (1H, m). Anal. Calcd. for C g 

H 16 C1PS: C, 45.60; H, 7.65. Found: C, 45.94; H, 7.25. 

Conversion of l-Chloro-2,2,3,4,4-pentamethylphosphetane 1-Sulfide 
(47) to l-Chloro-2,2,3,4,4-pentamethylphosphetane (48). A mixture of 25g 
(0.12 mole) of the compound ( 47) and 47.2g (0.234 moel) of tri-(n-octyl) 
phosphine was heated at 190-210° with stirring under a nitrogen atmosphere 
for 12 hr. in a three-necked flask equipped with a reflux condenser, a 
drying tube and a thermometer. The flask was connected to a distillation 
head. A stream of nitrogen (10-20 bubbles/ min.) was passed into the hot 
reaction mixture and the distillate over 40-93° was collected in an ice cold 
receiver (some solid that deposited in the condenser was collected by melt¬ 
ing). The distillate was redistilled in vacuo at 36-38° (0.1-0.05 mm) using 
a 12" Vigreaux column to obtain a transparent colorless liquid, 12.6g (60%). 
The nmr (neat) showed the liquid to be a mixture of two isomers (2:1). The 

3 

major somer (trans) exhibited peaks at: x=8.77 (6H, d, J p ^ c ^=2.10 cps). 



























































































































46 


t= 8.88 (6H, d, J pcCH =8.0 cps) , x=9.33 (3H, 2d, 4 J pcCH =l cps, ^^=7.0 

C P S )» x=7.22 (1H, 2q, ^J pr;rH =l. 25 cps , 3 , J^q^ = 7*0 cps). The minor iso¬ 


mer showed peaks at: x=8.70 (6H, d, J PCCH =20.0 cps), x=8.85 (6H, d. 


PCCH 


=6.0 cps), x=9.29 (3H, 2d, J 


=1 cps, j hcch =7 -0 cps), x=7.84 


(1H, 2q, J PCCH = 2 * 5 cps, J 


PCCCH 

HCCH =7,0 c P s )* Anal• Calcd. for CgH^^ClP: C, 


53.80; H, 9.03. Found: C, 53.91; H, 9.07. 

Preparation of 1,1,2,3-T etramethyl-3-butenylthiophosphonyl Dichloride 

(49). To a solution of 8.5g (0.04 mole) of the phosphine sulfide (47) in 
100 ml of anhydrous benzene and 10 ml of petroleum ether (30-60°) at 0-5° 
was added 5.4g (0.04 mole) of sulfuryl chloride in 50 ml of benzene over 

2 hr. with stirring under a nitrogen atmosphere. The mixture was stirred 
at room temperature for 24 hr.. A yellow oil separated at the bottom. Eva¬ 
poration of the solvent followed by distillation of the residue from a 12" 
Vigreaux column in vacuo gave a colorless liquid, b.p. 49-51° (0.03-0.04 mm), 
8.7g (89%). The nmr (neat) showed peaks at: x=5.0-5.20 (2H, vinyl), x= 

6.60-7.30 (1H, m,allyl), x=8.58 (3H, d, 3 j pcch = 28.2 cps), x=8.55 (3H, d, 

3 3 

JpCCH =2 ^-8 cps), x=8.72 (3H, d, ^hCCH =7 '' 2 c P s ) » x=8.22 (3H, broad s). 

Anal. Calcd. for C 0 H lc Cl_PS: C, 39.19; H, 6.17; P, 12.63; Cl, 28.92. 

- o lb L 

Found: C, 39.23; H, 6.23; P. 12.49; Cl, 28.95. 

Preparation of a 1:9 ( cis : trans) Mixture of Isomers of 1-Phenylphos- 
phetane ( 41) was accomplished by Reduction of 1:19 ( cis : trans) of the 
Oxide ( 20) Stereospecifically with HSiCl^/Pyridine by the Method Described 
by Chrovat (2). 

Stereospecific Oxidation of Phosphetane ( 41) with t-Butyl Hydroperoxide . 
To a stirred solution of 2.0g (0.011 mole) of the above prepared phosphetane 
(41) in 5 ml of benzene at 0-5° under nitrogen was added a solution of 0.91g 
t-butyl hydroperoxide (90%) in 15 ml benzene over 45 min.. The mixture was 



















































































































































































47 


stirred overnight without cooling. Evaporation of the solvent gave a crude 
solid (2.2g - 100%) of the crude oxide. A small amount of the product was 
sublimed in vacuo at 80° (0.2 mm). The nmr (benzene) showed the product 
to be a mixture of isomers ( cis : trans ::11:89) of ( 20) indicating that the 
oxidation reaction resulted in predominant retention of the configuration 
at phosphorus. 

Quatemization of Chlorophosphetane (41) with Methyl Bromide . To l.Og 
(0.0056 mole) of the chlorophosphetane in 5 ml of benzene was added 2 ml 
of methyl bromide and the contents were allowed to stand in a sealed flask 
for 4-days. A white precipitate formed. The ppt was filtered, and washed 
with benzene to obtain 1.3g of a crystalline solid (47.5%). Recrystalliza¬ 
tion from acetonitrile gave a solid, m.p. 225-242° (decomposition). The 

3 

nmr (DCCl^) showed peaks at: for the major isomer x=8.57 (6H, d, JpccH = 

3 

18.0 cps), x=8.66 (6H, d, J pc ^ H =21.5 cps); minor isomer x=8.49 (6H, d, 

3 3 

JpCCH =1 8-5> cps), x=8.67 (6H, d, J p( -,^=22.0 cps), peaks common to both the 

2 3 

isomers at: x=7.82 (3H, d, J^^=12.5 cps), x=8.97 (3H, 2d, ^pccH = ** 3 C P S > 

=7.0 cps). Anal. Calcd. for C_H on P0Br: C, 42.36; H, 7.90; P, 12.14; 
HLLH y Zu 

Br, 31.32. Found: C, 42.53; H, 7.89; P, 12.12; Br, 31.20; Cl, 0.00. 

Treatment of the Hydroxyphosphetanium Salt with Ethyl Diazoacetate . To 
about 300mg of the salt covered with 10 ml of anhydrous ether was added 1 ml 
of ethyl diazoacetate. A slow evolution of gas occurred which became rapid 
on crushing the solid. After the effervescence ceased, the solution was in¬ 
jected into the glpc and the volatile products were identified as ether,ethyl 
diazoacetate and ethyl bromoacetate by comparison of retention times x to an 
authentic sample. The analysis was performed on the Varian A-90 instrument 
on a 20% SE 30 column. The remaining solution was evaporated under vacuum 
and the residual solid was identified as 1,2,2,3,4,4-hexamethylphosphetane 









1-oxide ( 22) on the basis of the nmr of the product and that of an authentic 













CHAPTER II 


STEREOCHEMISTRY OF NUCLEOPHILIC ATTACK 
ON CYCLIC PHOSPHORUS COMPOUNDS 


1. Introduction 

Organophosphorus compounds are subject to nucleophilic displacement 
reactions at the phosphorus site. Depending upon the geometry of the 
transition state, the reaction may lead to retention, recemization or in¬ 
version of configuration at the phosphorus. Available evidence points to 

44 

inversion of configuration in most cases and so the transition state 
for bimolecular reaction may be written as follows: 



where X is the leaving group and N the nucleophile. In addition, there 
are reactions which are assumed to proceed through stable intermediates. In 
these cases, a pentacovalent intermediate may be formed which may either 
have trigonal-bypyramidal ( 60) or a tetragonal pyramidal geometry (61), 

3 

depending upon whether the phosphorus in the intermediate is sp d 2 or 

z 

3 

sp d^2 ^2 hybridized (21). The geometry and the behavior of such interme¬ 
diates may be explained on the basis of pentacovalent phosphorus compounds. 


49 





































50 




Calculation of the relative energies of structure ( 60) and ( 61) have 

45 46 

shown ( 60) to be more stable than ( 61) , * and indeed all PX^ compounds 

known to date (where X is any monovalent substituent) have a trigonal-bi- 
pyramidal geometry. Evidence for this was not obtained until recently, 
since early electron diffraction data was inconclusive ^, and inference 
of the geometry was based mainly on spectroscopic evidence. It must also 
be pointed out here that in a PX^ molecule of trigonal-bipyramidal geome¬ 
try, the axial bonds should theoretically be longer than the equatorial 


bonds 


48 



a= axial or apical bond 
e= equatorial bond 
x= F 


(62) 


47 


Braune and co-workers were the first to obtain the electron di¬ 
ffraction data on pentafluorophosphorane (2). They found that all the 
phosphorus-fluorine bonds in this compound were of equal length. This is 
not only contrary to the theoretical expectation but to spectroscopic and 

chemical evidence as well (see p. 52 ). Reinvestigation of the same com- 

49 

pound (62) by Hansen and Bartell (1.534 A ) were indeed found to be 

shorter than the axial bonds (1.577 A°). Prior to the latter results, all 

the structural assignments were based on spectral data. Thus, Gutowsky 

50,51 on the basis of ir data had concluded that PF^ had a trigonal-bipy- 

52,53 


ramidal geometry. Holmes and co-workers 


studied the temperature 





























* 





























51 


19 


F nmr and chlorine quadrupole resonance spectra of various compounds of 


the general formula PCI F where n=l,2,..5 and observed that all the com- 
pounds studied had trigonal-bipyramidal geometry. Recently, evidence for 

the geometry of various phosphorane compounds has been obtained through 

v • . . 54 

X-ray investigations 

Berry 55 first proposed a possibility of equilibration of substituents 
in a trigonal-bipyramidal PX^ molecule and coined the term "pseudorotation" 
for the pnenomenon. This is defined as an intramolecular process in which 
a trigonal-bipyramidal molecule is transformed by deforming bond angles in 
a way, so that it appears to have been rotated by 90° about one of the in¬ 
teratomic bonds. 


3 

/N 


y 



In fig. 1, substituent 4, which is towards the viewer ( along the Y- 
axis) remains fixed; whereas the apical substituents 1 and 2 are pushed 
backward and the equatorial substituents 3 and 5 are pulled forward so as 
to produce a tetragonal pyramid in which group 4 is at the apex. Continu¬ 
ing this process, a second trigonal bipyramid is produced which appears to 
have been produced by rotating the first trigonal bipyramid about the bond 
along the Y-axis in which groups 3 and 5 are apical and 1,2 and 4 are equa¬ 
torial. The substituent about which a pseudorotation occurs is referred to 
as a 'picot' group (substituent 4 in the fig. 1). The tetragonal pyramid 

occurs as a transition state in this process because it is relatively higher 
46 


m energy 
























































































































52 


The positional exchange of substituents in trigonal-bipyramidal mo¬ 
lecules by an intramolecular process was demonstrated in the classical rea- 
search of Gutowsky, Schmutzler and Muetterties. Gutowsky showed that 
the 19 F nmr of PF 5 exhibited only one doublet ( 19 F signal split by phos¬ 
phorus, spin=l/2). This observation o the equivalence of fluorine atoms 

in PF^ can be best explained by assuming one of the following: 

19 

i) a fortuitous overlap of two F signals, 

ii) valence-orbit hybridization, 

iii) a rapid positional exchange of substituents in a trigonal 
bipyramid. 

55 

The suggestion of accidental degeneracy was ruled out by Berry on 

the following grounds: Downs and Johnson 56 have shown that phosphorus pen- 
tachloride exhibits a difference in the exchange rates with isotopically 
labelled chlorine at the equatorial vs. the axial sites. Berry argued that 
this shows a definite difference in the reactivity and hence, as a corollary; 
in the electron density at the two reaction sites. This example of differ¬ 
ence in the electron density at the two sites in PC1 5 can be easily extra¬ 
polated to PF 5 and so overlap of the signals could have not occured. 

The hypothesis of valence-orbit hybridization was first advanced by 

C *7 

VanWazer and co-workers . The calculations carried out the explain spec¬ 
tral data had some underlying assumptions which, as the authors pointed out 
were critical to the hypothesis advanced. It did not gain widespread sup¬ 
port . 

This leaves Berry's hypothesis of pseudorotation to be tested for its 
generality. Considerably more came to be known about pseudorotation as a 
result of the investigations by Schmutzler and Muetterties. For example, 
the nmr of methyltetrafluorophosphorane ( 63) shows only one kind of proton 






















































































53 


and one kind of fluorine atom, a phenomenon which can be easily explained 

by invoking pseudorotation. However, dimethyltrifluorophosphorane ( 64) 

shows two kinds of fluorine atoms in the ratio 2:1 but only one type of 
58 

methyl proton . This can be explained by assuming that the two methyl 
groups occupy equatorial positions wheras one of the three fluorines occu¬ 
pies and equatorial site and the remaining two occupy apical positions. 
(fig. 2). The substituents do not equilibrate; which implies an absence 

of pseudorotation. A tetragonal pyramid could rationalize the nmr obser- 

59 

vations. However, Hansen and Bartell , in their electron diffraction 
investigations on (63) and (64), have not only shown that the molecules 
are trigonal bipyramids but also, that the methyl groups in both the sys¬ 
tems occupy equatorial positions. The equivalence of the fluorine atoms 
in ( 63) can be easily rationalized by assuming that the methyl groups ser¬ 
ves as a pivot in each pseudorotation allowing it to remain in the equa¬ 
torial position. This has been shown in fig. 2, in which the fluorine at¬ 
oms are numbered for convenience. 





F 4 (fig. 11 —2) 

Similar evidence for pseudorotation in the cyclic system tetramethylenetri 

fluorophosphorane ( 65) was obtained by Schmutzler during the temperature 
19 58 

dependent F nmr investigation on the compound. However, he was un¬ 


able to obtain any evidence for the equilibration of fluorine atoms in pen- 
tamethylenetrifluorophosphorane ( 66) (see p.55 ). The phenomenon of pseu¬ 
dorotation has also been observed in compounds which contain other hetero- 


















































































54 


69 

atoms, e.g., phenylarsenic tetrafluoride (PhAsF^) 

From the above discussion, it is not clear why ( 64) does not show any 

evidence of pseudorotation. In this case, for a pseudorotation to occur, 

one of the methyl groups must occupy an apical position. In order to ex- 

58 

plain the non-equivalence of fluorine atoms in (64), Muetterties sug¬ 
gested that more electronegative groups preferentially occupy apical posi¬ 
tions whereas less electronegative groups prefer equatorial positions. This 
allows for the placement of the methyl groups in equatorial positions in 
both compounds ( 63) and ( 64) . 

To invoke electronegativity as the basis of positional occupancy can 
be easily questioned, since steric factors alone may explain the observa¬ 
tions. Triechel and co-workers have advanced an interesting case against 
this argument. They showed that HPF^ and DPF^ show rapid positional equi¬ 
libration of the substituent fluorines between -90 and -50°. On the other 
19 

hand, the F nmr of H^PF^ and D 2 PF 3 show two different kinds of fluorine 
atoms in the ratio 2:1 at -46°. The downfield signal (intensity=2) arises 
from the two apical fluorine atoms in the equatorial position. By warming 
up to -15°, the peaks are observed to broaden and finally coalesce to a 
doublet supporting the view that the fluorine atoms experience a similar 
averaged environment by an intramolecular process. The P-F coupling re¬ 
mained constant indicating an absence of any significant contribution from 
an intermolecular process. The equatorial preference of hydrogen and deu¬ 
terium conclusively rules out the positional preference due to steric fac¬ 
tors alone. 

The selectivity in positional occupancy and its relation to the elec¬ 
tronegativity of the substituent can be rationalized on theoretical grounds 
62,66. ph OS phorus in a trigonal bipyramid is sp 3 d hybridized. Of these 




























































55 


five spatially oriented hybrid bonds, the s- character is concentrated 

2 

in equatorial bonds (sp hybridization), rendering the equatorial bonds 
more electronegative. The axial-bonds do not have any s- character (pd 
hybridization) and so are more electropositive. As a result, more elec¬ 
tronegative substituents prefer axial positions whereas less electrone¬ 
gative (more electropositive) substituents prefer equatorial sites. 

On the above basis, it is easy to understand the nonequivalence of 
fluorine atoms in (64). As shown in fig. 3, for the equilibration to oc¬ 
cur, one of the methyl groups must occupy an apical position to produce 
an energetically unfavorable intermediate: 



19 

On the other hand F nmr of compound ( 65) shows that all three fluorine 

atoms in the molecule are equivalent. Since H-F and P-F couplings remain 

constant, any contribution due to an intermolecular process can be ruled 

out. An intramolecular process such as pseudorotation may therefore be 

responsible. However, this cannot occur unless one of the P-C ring bonds 

is placed in an apical position-an energetically unfavorable situation. 

6 7 

Schmutzler has investigated this case in detail by studying the tempera¬ 
ture dependent nmr of the compound. The A for the process was estimated 
to be 7 kcal/mole. He proposed that one of the P-C bonds in the system can 
occupy an apical position since the energy of such a state is lowered due 
to the strain present in a five-membered ring. Thus, he proposed contri¬ 
butions from structure shown in fig. 4. 












56 




Recent X-ray data shows that the internal C-P-C angle in a five-membered 
ring is about 95° 6^ ,6 9^ t hus, contribution due to structures such as 

(64) in which the ring spans two equatorial positions can be neglected be¬ 
cause the C-P-C angle will be forced to expand to 120°. A more conveni¬ 
ent path for the pseudorotation to occur involves intermediates such as 

( 65) in which the ring spans an equatorial-apical relationship. This can 
be seen easily in fig. 5, where the substituents have been labeled: 

On the basis of this information, Muetterties formulated the following 
rules which have become known as Muetterties' Rules ^. 

i. An intramolecular exchange of substituents in a trigonal 
bipyramid molecule can occur through pseudorotation, 

ii. Electronegative substituents prefer apical positions, 

iii. The preference in (ii), can be limited due to internal 
strain in the molecule. 

In the light of the above generalizations, the application of the 
pseudorotation hypothesis can be readily understood. 

After following through a few steps of pseudorotation in a particular 
system, it becomes obvious that the determination of the possibility for 
one of the steps to occur could be quite complicated. This could be less 
difficult when there are geometrical constraints (such as a small ring in 
the molecule) involved. An attempt has been made to simplify this process 

























































































































u. 




(65) 





















58 


rr 

to 



to 

(N 


4-1 

O 

01 

P P 

P p P 

jz o 6 
h x: o 

Ph-H 



0) 

P 

• 

O 

P 

to 

X 

P 

r- 

P. 

P 


a> 

P 

o 

X -P 

r—t 

p 

P 

t/1 



•H 

4-t 

0) 

2 

O 

-rH 

X 

•h| 

to 

rO 

PI 

CM 


•H 


p 

6 

• 

<D 

P 

X 

> 

P 

p 

• H 

p 

p 

o 


p 


a> 

p 

01 

rC 

•H 

P 

P 

■P 



P 

-p 

o 

P 

a. p 


P 


P 

P 

h3 

P 

ID 

P 

P 


X bo lO 

0) -H 
p ttn3 

P <f> p 

0) CTl P 

> 

i p ^r 

o P - 

(N rt to 


in m t/i 

— ?H 

P T3 <D 
P P A 
Xi (j s 
c5 3 

f-H rH 

$ c/l bt> 
»—t P 
4-1 P .H 
O -H P 
P 


P P 

bo 

o E 

P 

•H ^ 

• H 

P z 

p 

o 

*H 

P X 

P 

•1—i p 

P 

O T3 

P 

P P 

O 

a- -h 

O 

| 


HH 


h-i 



bO 

•rH 

U-. 








































59 


71 73 71 

by various workers * . Cram has given a diagram that can be utili- 

zed for the open-chain system. Mislow and co-workers have described a 
very elaborate topological representation of the stereochemistry of the 
displacement reactions in phosphonium salts and similar systems. The re¬ 
quirement is that the reaction passes through a phosphorane intermediate. 

He has shown that the stereochemical outcome of such displacement reactions 
can be conveniently followed with the aid of a modified version of Bala- 
ban's 20-vertex graph which resembles the carbon skeleton of hexasterane 
(fig. 6). The eighteen vertices of the figure represent the eighteen 
stereoisomers of intermediate phosphorane trigonal bipyramid containing 
five different ligands. Two of these ligands are the termini of a ring 
system which are incapable of occupying the two apical positions. The 
pathways for the interconversion of isomers by pseudorotation are repre¬ 
sented by twenty-four edges in fig. 6. A very interesting feature of the 
diagram is that in a represen tion, three planes intersect at the 
six-fold axis; each of the planes divides the graph into two subsets of 
nine vertices. Each subset represents the nine phosphoranes resulting from 
an external nucleophilic attack on a phosphorus atom of the given enatio- 
mer of a cyclic phosphonium salt. As a result, the positions of the in- 
terraidiate phosphorane within sectors defined by two intersecting planes 
gives the stereochemical outcome of the reaction. The figure is a general 

one and too complex to be described here. Wislow has described a number of 

73 

applications of this topological representation 

Although the convenient device discussed above will not be specifi¬ 
cally used in the discussion which follows, it is valuable for following 
the course of a reaction in these rather complicated systems. At this 
point therefore, a few systems which undergo nucleophilic substitution to 






























































o© 


60 






Fig. II-7. 






















61 


yield a phosphorane intermediate capable of undergoing pseudorotation will 
be discussed. 

74 

Westheimer found that the hydrolysis of methyl ethylene phosphate 

( 67) with ring opening proceeds nearly 10 times faster than trimethyl 

phosphate (68). More important is the fact however, that the acid catalyzed 

exocyclic cleavage of themethoxy group occurs 10^ times faster than ( 68) . 

75 

Haake and Westheimer also observed that acid catalyzed hydrolysis of 

18 

hydrogen ethylene phosphate ( 69) results in a rapid 0 exchange in the 
unreacted ester. The enormously fast rate of ring opening can probably 
be rationalized in terms of the ring-strain in the cyclic ester. Diffi¬ 
culty is encountered in explaining the almost equally fast rate of exo¬ 
cyclic hydrolysis without ring opening. Apparently, the strain in the 
ring cannot possibly explain the rapid hydrolysis external to the ring. 

This problem of accounting for this unusual behavior will be dealt with 
in some detail because it will invoke some basic assumptions which will 

be necessary for this work. 

75 

Westheimer advanced an argument that there ought to be a single 
mechanism which could account for all the observations. He made the follow¬ 
ing assumptions: 

i. the reaction proceeds through a trigonal-bipyramidal inter¬ 
mediate with the ring spanning one apical and one equatorial 
position, 

ii. the entering and leaving groups do so from an apical position 
(extended principle of microscopic reversibility). Justifi¬ 
cation for this argument is dealt with in detail ( p.69 ). 








































































































62 


With the aid of these assumptions and the possibility of pseudorotation 

of the intermediate phosphorane, Westheimer extended the following expla- 
35 

nation . Addition of a molecule of water on (69) should result in the 
formation of the intermediate (70). This process is favored because of 
the ring strain in the ester (see fig. 7). Protonation of this interme¬ 
diate will give ( 71) which can then ring open by an apical departure. The 
protonation at the hydroxy oxygen will give another intermediate ( 72) 
which cannot give the observed 18^ exchange if rule (ii) above were fol¬ 
lowed. Pseudorotation of ( 72) to yield ( 73) followed by departure of the 
molecule of water from the apical position will explain the 18^ exchange. 

The above discussion will help rationalize both the fast ring opening 

18 

as well as fast rate of 0 exchange and the dydrolysis external to the 

ring. The rationale is as follows: the ring spans one apical and one 

equatorial position in the intermediate which involves a C-P-C ring angle 

of 90°. Formation of the intermediate results in the reduction of the 

strain in the starting material. Since this intermediate can both hydro- 

18 

lyze with ring opening as well as with 0 exchange; the fast reactions 
can be explained on the basis of the strain in the molecule because for¬ 
mation of the intermediate is facilitated due to the strain. It is assumed 
that the water and methanol molecules are not very different since ex¬ 
change involves entering and leaving of the former whereas the exocyclic 
hydrolysis requires entering of the former and departure of the latter. 

Recently, Boyd has carried out molecular orbital calculations on 
cyclic and acyclic phosphate esters. He verified the above explanation 
for the fast rate of hydrolysis of the cyclic ester. The energy barrier 
for pseudorotation of the intermediate phosphorane was estimated to be 
about 12-15 kcal/mole as an upper limit. 
























































































63 



Fig. II-8. 






















64 


This hypothesis can be further applied to the hydrolysis of the phos- 

tonate ( 74) and the phosphinate ( 75) esters, both of which involve a five- 

membered ring (fig. 8). In case of (74), it was observed that the hydro- 

77 

lysis resulted in almost exclusive ring cleavage . This can be explain¬ 
ed on the basis of the positional preference rules of Muetterties. An 

attack by a molecule of water on ( 74) will preferably give the intermedi- 

77 

ate ( 76) which can hydrolyze with ring opening . Intermediate ( 76) can 

also pseudorotate to give either ( 77) or (78). The formation of ( 77) can 

be ruled out because it will result in the expansion of the C-P-C ring 

angle to 120° and thus will be energetically unfavorable. Intermediate 

( 78) on the other hand is not favored because it has an electronegative 

oxygen atom in an equatorial position whereas an electropositive methylene 

group in the apical position as in the intermediate ( 79) . The energy for 

such a process may be so high that the reaction may not even go through 

a phosphorane intermediate. This has been confirmed by Aksnes and co- 
78 

workers who showed that (75)hydrolyzes with about the same rate as the 
six-membered ring phosphinate ester ( 80) as well as the open-chain analo¬ 
gue. 

The restriction of the positional preference due to the strain in 

the molecule as mentioned in Muetterties* rules can be further supported 

by the fact that the highly strained phosphine oxide (1) undergoes a rapid 
18 79 

0 exchange . Most phosphine oxides fail to undergo exchange or do so 
extremely slowly. Once again, a phosphorane intermediate undergoing pseu¬ 
dorotation may be proposed as shown below: 



( 1 ) 













































































































































65 


70 80 81 

Westheimer * * has demonstrated a very interesting example of 

pseudorotation during the hydrolysis of the bicyclic esters (81,82). One 
of the ester groups in these compounds hydrolyzed very rapidly compared 
to other open-chain or cyclic phosphinate esters. He argued that by 
analogy with bicycloheptane derivatives the bond angle at the 7- posi¬ 
tion of the bicyclic systems must be constricted. As a result of this 
angular strain, the hydrolysis could occur through the formation of a 

phosphorane intermediate followed by pseudorotation of the intermediate. 

69 

Recent X-ray data reported by Lipscomb corroborated the assumption of 
the strain in the molecules; the internal C-P-C angle was shown to be 87°. 
Thus, in these cases also, the relief of the strain achieved in going 
from the starting material to the intermediate is sufficient to overcome 
the positional preference. 




2. Hydrolysis of Phosphinate Esters 

Results and Discussion . It is apparent from the previous discussion, 
that internal strain in five-membered ring systems has, in some cases, led 
to the verification of the pseudorotation hypothesis. 

The four-membered ring compounds, due to larger strain (C-P-C = 83°) 

(page 17, present a very interesting system for testing the theory of 
pseudorotation and its consequences. 
































































































66 


The most important assumption is that the molecule is sufficiently strained 

so that the barrier for the placement of the P-C ring bond in the apical 
position of the intermediate may be overcome by the relief of strain ac¬ 
hieved in going from the starting material to the intermediate. It is 

70 

therefore critical to demonstrate that the reactions under investiga¬ 
tion proceed through intermediates that are capable of undergoing pseudo¬ 
rotation. 

Nucleophilic reaction at a phosphoryl center is known to proceed 

through a transition state and results in overall inversion of configura- 

44 

tion at phosphorus . The mechanism, by analogy with the nucleophilic 
substitution at a carbon center, is called Sj.2P (Substitution Nucleophi¬ 
lic Bimolecular at Phosphorus). A classical demonstration of inversion 

82 

at a phosphoryl center was given by Green and Hudson . They showed 

14 

that the optically active C labelled phosphinate ester ( 83) equibrated 
with its unlabelled analogue in methanolic sodium methoxide - (eq. 1): 


Ph 0 
\H 
P — 

/ 



14 

OCHj ♦ 



(83) 


Ph 0 



OCHj ♦ 


14 

OOlj 


( 1 ) 


14 

The rate of racemization was shown to be twice the rate of loss of C 

83 

label from the ester. Similarly, Green and Hudson have shown that nu¬ 
cleophilic reactions on thiol esters ( 84) and phosphorochloridates ( 85) 
also occur with inversion of configuration. 







67 


R — 


? 



0 

II 

R— P — 


R 


Cl 


(84) 


(85) 


It was of interest to determine whether the stereochemistry of the above 
open-chain systems applied to the four-membered ring phosphinate esters. 
For this purpose, the reaction of sodium methoxide with the cis and trans 
isomers of estei' ( 86) was examined. 




V R 


( 86 ) 


: 3 CH., and R, ( trans) 
b^ : 3 CH^ and R, ( cis) 


A methanolic solution of ester ( 86 a) and a mixture of 86_ a and b 

(2:3) (R = OCH^ in each case) were separately treated with sodium metho- 

xid° solution in evacuated, sealed nmr tubes. Heating to 50° did not 

show any significant change in the ratio of isomers as determined by nmr 

integration. Methanolic solutions of ( 86 a) and ( 86 a and b) (2:3) 

(R=0CD 3 ) were similarly treated with sodium methoxide. Exchange of OCD^ 

by OCH^ took place in each isomer. Since exchange of isomers occurred with 

out any isomer crossover, it was concluded that nucleophilic substitution 

with methoxide ion on the phosphinate ester ( 86) proceeded with retention 

33 

of configuration 

The observation above, can be rationalized in terms of pseudorotation 
of the pentacovalent intermediate. Attach of methoxide ion on ( 86 a R=0CD^) 
along the P-C bond from the apical position leads to intermediate ( 87) All 

































































66 


arguments used for (86 a) are equally applicable to (86 b). Pseudorota¬ 
tion of ( 87) with oxygen as a pivot will lead to ( 88) , while employment 
of the deuteriomethoxy group as a pivot will yield (89). Both ( 88) and 
(89) are rotated in order to aid in visualization. Intermediate (88)can 



decompose with the -OCD^ group departing from the apical position (micro¬ 
scopic reversibility) to yield the same isomeric trans ester. Intermedi¬ 
ate ( 89) will be higher in energy than ( 87) and (88)(which should be simi¬ 
lar in energy) since ( 89) has a negatively charged oxygen atom in an api¬ 
cal position. According to Muetterties rules ^,0 is less electronega¬ 
tive than the neutral 0 atom and should preferably occupy an equatorial 
position. Furthermore, ( 89) can undergo pseudorotation to yield either (87) 
or ( 90) which subsequently could decompose to yield isomer crossover which 
was not observed. 

The same argument can be applied to the case of equatorial attack and 

equatorial departure. The methoxide ion can attack along the P-C^ ^ on< ^ to 

(*) 

yield the mirror images of ( 88) and ( 90) depending upon whether the 
attack occurs from the same or the opposite side of the 3-methyl group. 

(*) The attack along the P-C^ bond will produce the object of the mirror 

image and therefore no optical acitivity is generated during the process. 









































































69 


Pseudorotation of the mirror image of ( 88) with 0~ as pivot yields the 
mirror image of ( 87) which on departure of the -OCD^ group from an equa¬ 
torial position should render the starting trans ester. The major reason 

for ruling out equatorial attack and departure is based on the considera- 

84 

tion of the principle of least motion . Stated in terms of Rice and 
81 

Teller , "those elementary reactions will be favored that involve least 

change in atomic positions and electronic configuration." The transition 

state for apical and equatorial departure (or attack) are shown in fig. 9. 

In the following argument the angles in the phosphorane intermediate are 

assumed to be the same in an idealized trigonalbipyramid; whereas those in 

30 

the ester are assumed very similar to the phosphine oxide (R=Ph) , For 

apical departure, bonds P-R^. and P-R^ will have to change by about 20° 

(fig. 9 ) .On the other hand, during the equatorial departure bonds 

P-R r will be altered 20° whereas bond P-R. will have to be transformed 
b 6 

60-70°. Although both paths should lead to the same results, only an api¬ 
cal in-out mechanism is favored because the molecule will go through the 
least deformation in this process. 




(Fig. II-9) 


Having demonstrated that nucleophilic substitution on the phosphinate 
ester ( 86) proceeded via an intermediate that can undergo pseudorotation, 
it seemed appropriate to investigate the rates of alkaline hydrolysis of 
these esters. In the event that the reaction goes through an intermediate, 
it would be expected that hydrolysis external to the ring will be fast. 




































































70 


Table 1 shows the rate data in the case of esters (8), (11), ( 15 ) and 

( 86) . A comparison shows that phosphinate ester ( 15) hydrolyzes 2300 times 

faster than ( 86) . The unsymmetrical tetramethyl substituted ester (H) an< ^ 

and pentamethyl ester (8) hydrolyze at about the same rate but do so about 

as fast as (86). Structurally, (8_) and ( 86) each have one additional methyl 

group in the 2-position relative to (15)and (11), respectively. However, 

the ratio of the rate constants are: (11)/(86) = 4, whereas (15)/ (8) =1100. 

The reason for the difference seems to be steric in origin. To understand 

the nature of the proposed steric effect, examination of data in table 2 
86 

is necessary . It can be seen that the estimated realtive rates in the 

phosphonate esters ( 91) when R=Et, i-Pr and t_-Bu are 200: 7:1. On the 

other hand, the same change in the phosphinate esters ( 92) R= Etji^Pr, t-Bu) 

produces the relative rates of hydrolysis are 10 5 :500:1. Trippett et al ., 
86 

have argued that the dramatic rate differences between the substances 
with only one bulky group and those with two such groups are due to the 
steric hindrance in the transition state leading to the intermediate phos- 
phorane. They further argued that, of two bulky t-Bu groups in the mole¬ 
cule, one must occupy an equatorial position. As a result, the steric hin¬ 
drance in phosphinates with two bulky groups is more effective in blocking 
the approach of the in-coming nucleophile and consequently decelerates the 
rate of hydrolysis in these compounds. 

The argument outlined above seems to be valid in the cyclic systems in 
which the reaction must go through an intermediate as shown earlier. Its 
application to the open-chain systems is questionable since the available 

evidence indicates that nucleophilic substitution in such open-chain systems 

82 

goes through a transition state rather than an intermediate 

In any event, Trippett*s hypothesis of steric blocking of nucleophi- 



























71 


Table II-1. Alkaline Hydrolysis of Phosphinate Esters 
Second-Order Rate Constants at 25°. 


Compound 


k 2 (M^SecT 1 ) 


( 8 ) 

(IT) 

(15) 

( 86 ) 


2.51+ 0.06 x 10 
1.96+^ 0.065 x 10 
1.51+ 0.012 x 10 
5.85+ 0.13 x 10" 


Compound 


K 2 (M" 1 Sec7 1 ) at 45° 


( 15) 5.9 x 10 *(Extrapolated) 

(93) 3.9 x 10 _1 (Ref. 82) 



OCH. 


( 8 ) 



OCH, 


( 11 ) 



^ OCH 

(15) 



\ 


(86) 0CH 3 



(93) OCH 3 











































72 

Table II-2. Alkaline Hydrolysis of Phosphonate and Phosphinate 
Esters - Second Order Rate Constants 


RP ('• 0) (OEt) 2 (91_) 


R 

Et 


i-Pr 


t-Bu 


k 2 (M' 1 Sec." 1 ) 

7.4 x 10" 3 (120°) * 

4.4 x 10~ 4 
6.0 x 10~ 5 


R 2 P(:0) OEt (92_) 70° 120° 


Et 

2.6 

x 10“ 4 



i-Pr 

9.8 

x 10' 7 

4.1 

x 10' 5 

t=Bu 



8.0 

x 10" 8 


* Calculated 


Ref.: S. Tripett and W. Hawes , Chem. Comm., 577 (1968). 
































73 


lie attack during the formation of the intermediate can be applied to the 
cyclic phosphinates. This can be clearly seen in the relative rates of 
( 15) and ( 86) . There are obviously two effects which are working in op¬ 
posite directions: a) ring strain would lead to facile formation of the 
intermediate, which in turn would be reflected in acceleration of the rate 

of hydrolysis; b) steric blocking of the attacking reagent (noepentyl ef- 
87 

feet) would result in deceleration of the reate of hydrolysis. In the 
case of (86)therefore, the rate of hydrolysis is faster than for (92)(R= 
t-Bu); however, the dramatic effect of rate acceleration is reflected in 
the compound ( 15) in which steric hindrance is not able to overcome the 
effect due to the ring strain. 

Another interesting feature in table 1 is the relative rates of com¬ 
pounds (8) and ( 11) ; compound (8) which has an additional methyl group in 
position 3 hydrolyses slightly faster than (11). The reason for this small 
rate acceleration is not entirely clear. The same is true for ( 15) and ( 93) 

One possible explanation is based on the conformational differences in 
the compounds. In ( 11) there is only one methyl group at position 3 which 
because of the puckering of the ring may stay as far from the reaction site 
as possible. In (8), on the other hand, there are two methyl groups in po¬ 
sition 3 and as a result, one methyl group is always near the reaction site 
(buttressing effect). This would raise the energy of the ground state and 
would thus account for the increased rate of hydrolysis. The AH^ values 
reflect this possibility since in (8_), less energy would be required to 
reach the intermediate as shown below: 




















74 


88 

It should be mentioned that Bergesen , has claimed to have isolated two 
isomeric acids derived from the ester ( 86 R=Et). It is difficult to under¬ 
stand the claim of isolating isomeric acids in view of the possible tauto- 
merization of the acid in solution: 


V_ 1 




\ 


l 


% 

''OH 


V_ 




^OH 

*i>0 




86 

The rates given in this work as well as those of Trippett seem to be 
compatible with the trans isomer of the ester. 

b. Experimental . The alkaline hydrolysis of the esters (8), ( 11) and 
( 86) was followed by a titration procedure. Equal volumes of the required 
ester and alkali solutions were separately equilibrated in a double cham¬ 
bered flask in a constant temperature bath. The solutions were mixed. A 
5 ml aliquot was withdrawn at intervals and quenched with a known volume 
of standard hydrochloric acid solution. The excess of the acid was tit¬ 
rated against a standard sodium hydroxide solution using phenolphthalein 
as an indicator. 

In the case of the ester (15), the course of the reaction was followed 
by conductivity measurements. In a constant temperature bath, 0.04 molar 
solutions of the ester and alkali were separately equilibrated. Equal vo¬ 
lumes of the solutions were mixed in a conductivity cell in a constant 
temperature bath. The resistance of the reaction mixture was determined 
at intervals with a conductivity bridge (model RC IB, Industrial instru¬ 
ments) . The calculations were performed as shown below (89): 
t=(l/ak) [(R - R 0 )/R q ] [ R/(R„- R)] - 1/ak 


where t=tirae 







































































































































































75 


a = initial concentration 

Ro = Resistance at time = o 

R = resistance at time = t 

R = resistance at time 
00 

and 

k = rate constant. 


Table II-3. Alkaline Hydrolysis of l-methoxy-2,2,3,3,4-pentamethyl- 
phosphetane 1-oxide (8)at 25.4! A =0.03776 M. B = 
0.02925 M. o o 


# time 
hr. 


Cone. NaOH 
B 


log AB /A B 
6 o o 


1 

0.278 

2.906 

X 

-2 

10 

0.004 

2 

6.0 

2.337 

X 

(N 

1 

O 

rH 

0.2240 

3 

18.0 

1.730 

X 

io ' 2 

0.0595 

4 

24.0 

1.409 

X 

io " 2 

0.0895 

5 

42.0 

1.028 

X 

10“ 2 

0.1443 

6 

66.0 

6.695 

X 

io - 3 

0.2345 


k 2 = 2.66 x 10 4 +_ 7.1 x 10' 6 M7 1 Sec. 
k 2 = 2.66 ^ 0.071 x 10~ 4 M? 1 Sec." 1 
















76 



Table 11-4. 

Alkaline Hydrolysis of 
phosphetane 1-oxide (8) 
0.0196 M. 

1-methoxy-2,2,3,3,4-pentamethy1- 

at 25.5°. A =0.0372 M., B = 
o o 

# 

Time 

hr. 

Cone. NaOH 

log AB /A B 

1 

9.0 

1.428 x 10~ 2 

0.07ol 

2 

15.5 

1.153 x 10' 2 

0.1235 

3 

36.0 

7.043 x 10’ 3 

0.2648 

4 

45.0 

6.347 x 10' 3 

0.2978 

5 

59.5 

4.644 x 10' 3 

0.4014 

6 

70.5 

3.909 x 10" 3 

0.4615 

7 

79.5 

3.367 x 10" 3 

0.5153 

k 2 

= 2.273 x 10~ 4 
= 2.27 + 0.055 

+ 5.5 x 10-6 M. 1 Sec. 
x 10" 6 M.' 1 Sec. ~ 1 

-1 



Table II-5. 

Alkaline Hydrolysis 
phosphetane 1-oxide 
0.0260 M. 

of 1-methoxy-2,2,3,3,4-pentamethy1- 

(8) at 25.4°. A -0.0413 M.,B = * 

— o o 

# 

Time 

hr. 

Cone. NaOH 

log B q A/A o B 



1 

9.0 

1.834 

X 

10- 2 

0.0626 

2 

18.0 

1.358 

X 

io' 2 

0.1265 

3 

24.5 

1.138 

X 

io" 2 

0.1688 

4 

45.0 

6.850 

X 

10- 3 

0.3086 

5 

54.0 

5.960 

X 

io: 3 

0.3512 

6 

69.0 

4.528 

X 

io’ 3 

0.4402 

7 

80.0 

3.754 

X 

io’ 3 

0.5042 


k 2 = 2.59 x 10" 4 + 5.25 x 10" 6 M.' 1 Sec. 
= 2.59 + 0.0533 x 10“ 4 M.' 1 Sec.' 1 . 



















77 


Table II-6. Alkaline Hydrolysis of l-methoxy-2,2,3,3,4-pentamethyl- 
phosphetane 1-oxide (£)at 60°. A =1.11 x 10 _1 M. , B = 
5.35 x 10*3 m. Pseudo-unimolecular reaction. 


ft Time 
min. 


Cone. NaOH log B 

B, M. 



1 

8.0 

4.401 

X 

10- 3 

-2.3564 

2 

16.0 

3.648 

X 

ID’ 3 

-2.4379 

3 

24.0 

3.110 

X 

io- 3 

-2.5072 

4 

40.0 

2.318 

X 

IQ’ 3 

-2.6349 

5 

56.0 

1.868 

X 

io’ 3 

-2.7286 

6 

80.0 

1.340 

X 

io’ 3 

-2.8729 

7 

104.0 

1.183 

X 

io* 3 

-2.9270 


k' 2 = 2.09 ^0.17 x 10" 3 M. 1 Sec. 


Table II-7. Alkaline Hydrolysis of l-methoxy-2,2,3,3,4-pentamethyl- 
phosphetane 1-oxide (8) at 60°. A =1.11 x 10 - 1 M.,B = 
5.35 x 10- 3 M. Pseudo-unimolecular reaction. 


# Time Cone. NaOH log B 



1 

8.0 

4.421 

X 

io- 3 

-2.3545 

2 

16.0 

3.668 

X 

l(f 3 

-2.4355 

3 

24.0 

3.159 

X 

io’ 3 

-2.5005 

4 

40.0 

2.347 

X 

io' 3 

-2.6295 

5 

56.1 

1.819 

X 

io" 3 

-2.7402 

6 

80.0 

1.340 

X 

io- 3 

-2.8729 

7 

104.0 

1.066 

X 

io- 3 

-2.9722 


k 2 - 2.23 x 10" 3 +_ 1.36 x 10' 4 M.' 1 Sec. 
= 2,23 + 0.140 x 10“ 3 M. _1 Sec.' 1 


-1 



















Table II-8. 


78 


Activation Parameters for the Alkaline Hydrolysis of 
l-methoxy-2,2,3,3,4-pentamethylphosphetane 1-oxide (8) 
in water. 


Temperature 

°T 


k 2 (Average) 
M' 1 Sec." 1 


AH^ AS^ 

kcal/mole c.u. 


298.5 


2.51 x 10 


333.1 


2.16 x 10 


11.6 


-36.1 


Table II-9. Alkaline Hydrolysis of l-methoxy-2,2,3,4-tetramethyl- 

phosphetane 1-oxide ( 11) at 25.1°. A =1.254 x 10“1 M., 
Bq= 5.001 x 10 -3 m. Pseudounimolecular reaction. 


# 

Time 

10 3 Sec. 

Cone. NaOH 

B, M. 

log B 



1 

10.0 

3.835 

X 

10- 3 

-2.4162 

2 

15.0 

3.355 

X 

io- 3 

-2.4744 

3 

20.0 

2.845 

X 

io- 3 

-2.5459 

4 

26.0 

2.480 

X 

io' 3 

-2.6055 

5 

33.0 

2.105 

X 

icf 3 

-2.6768 

6 

41.0 

1.771 

X 

io’ 3 

-2.7518 



1.98 x 10~ 4 +_ 7.1 x 10 
1.98 + 0.071 x 10" 4 M. 


M." 1 Sec. 
Sec. 1 















79 


Table 11-10. Alkaline Hydrolysis of l-methoxy-2,2,3,4-tetramethyl- 
phosphetane 1-oxide 111) at 25.1°. A = 1.254 x 10 -1 
M., B q = 5.001 x 10-5 Pseudounimolicular reaction. 


# 

Time 

10 3 Sec. 

Cone. NaOH 

B, M. 

log B 



1 

14.0 



3.387 

X 

ID’ 3 

-2.4702 

2 

22.0 



2.730 

X 

10- 3 

-2.5638 

3 

27.0 



2.365 

X 

10- 3 

-2.6262 

4 

30.0 



2.209 

X 

10- 3 

-2.7011 

5 

36.0 



1.990 

X 

10- 3 

-2.7011 

6 

40.0 



1.792 

X 

10- 3 

-2.7466 

7 

46.0 



1.521 

X 

10- 3 

-2.8179 



-4 


-6 


-1 -1 


k 2 

=1.94 x 

10 

+ 6.00 

x 10 

M. 

Sec. 





-4 

-1 


-1 



=1.94 + 

0.06 

x 10 

M. Sec. 




Table II-11. Alkaline Hydrolysis 

phosphetane 1-oxide 

M., B =5.035 x 10-3 
o 


of 1-methoxy-2,2,3,4-tetramethyl- 

(11) at 39.65°. A = 1.218 x 10" 1 
— o 

M. Pseudounimolecular reaction. 


# Time Cone. NaOH log B 

10 2 Sec. B, M. 


1 

18.0 

4.474 

X 

10- 3 

-2.3493 

2 

36.0 

3.805 

X 

ID’ 3 

-2.4196 

3 

54.0 

3.254 

X 

ID’ 3 

-2.4876 

4 

72.0 

2.812 

X 

ID’ 3 

-2.5510 

5 

108.0 

2.104 

X 

IQ’ 3 

-2.6770 

6 

144.0 

1.711 

X 

10' 3 

-2.7668 


k 2 = 6.32 x 10 4 + 2.46 x 10~ 5 M. 1 Sec. 


= 6.32 + 0.25 x 10 


-1 

























80 


Table 11-12. Alkaline Hydrolysis of l-methoxy-2,2,3,4-tetramethyl- 
phosphetane 1-oxide (11) at 39.65°. A =1.218 x 10-1 

M-> B q = 5.035 x 10~3 Pseudounimolecular reaction. 


# 

Time 

10 2 Sec. 

Cone. NaOH 

B, M. 

log B 



1 

18.0 

4.523 

X 

10- 3 

-2.3446 

2 

36.0 

3.825 

X 

io- 3 

-2.4173 

3 

54.0 

3.225 

X 

IQ’ 3 

-2.4914 

4 

72.0 

2.792 

X 

io’ 3 

-2.5541 

5 

109.0 

2.124 

X 

io' 3 

-2.6729 

6 

144.0 

1.711 

X 

io- 3 

-2.7668 


k 2 = 6.325 x 10~ 4 +_ 2.59 x 10 -5 M. 1 Sec. 
= 6.33 + 0.26 * 10~ 4 M.' 1 Sec. -1 


Table 11-13. Alkaline Hydrolysis of l-methoxy-2,2,3,4-tetramethyl- 

phosphetane 1-oxide ( 11) at 60°. A =1.218 x 10"! M., 

B =5.035 x 10“3 M. Pseudounimolec8lar reaction, 
o 


Time 
min. 


Cone. NaOH 
B, M. 


log B 


1 

8.0 

4.333 

X 

10" 3 

-2.3632 

2 

16.0 

3.560 

X 

io' 3 

-2.4486 

3 

24.45 

2.905 

X 

io' 3 

-2.5369 

4 

32.0 

2.572 

X 

io- 3 

-2.5898 

5 

48.0 

1.966 

X 

io- 3 

-2.7054 

6 

64.0 

1.575 

X 

io- 3 

-2.8027 


k = 2.44 x IO -3 + 1.42 x 10~ 4 M.' 1 Sec.' 1 

= 2.44 + 0.14 x 10“ 3 M.' 1 Sec.' 1 


Sec 















81 


Table 11-14. Alkaline Hydrolysis of l-methoxy-2,2,3,4-tetramethyl- 
phosphetane 1-oxide (11) at 60°. A =1.218 x lO - * M., 

B =5.035 x 10~3 M. Pseudounimolecular reaction. 


# Time 
min. 


Cone. NaOH log B 

B, M. 


1 

8.0 

4.196 

X 

o 

1 

04 

-2.3772 

2 

16.0 

3.531 

X 

10 3 

-2.4521 

3 

24.0 

3.002 

X 

10- 3 

-2.5226 

4 

32.0 

2.562 

X 

10- 3 

-2.5915 

5 

48.0 

2.015 

X 

h-* 

O 

1 

04 

-2.6957 

6 

64.0 

1.614 

X 

o 

1 

04 

-2.7921 


k 2 = 2.32 x 10' 3 +_ 1.1 x 10" 4 M." 1 Sec. 
k 2 = 2.32 ^0.11 x 10" 3 M.' 1 Sec.' 1 


Table 11-15. Activation parameters for l-methoxy-2,2,3,4-tetramethy1- 
phosphetane 1-oxide (11). 


Temp. ° c. : 

25.1° 

: 

39.65° 


• • 

O' 

o 

o 


k 2 ,M. -1 Sec. -1 : 

1.94 x 

10" 4 : 

6.32 x 

io' 4 

: 2.44 

x 10' 3 



-4 


-4 

: 2.32 

-3 


1.98 x 

10 : 

6.32 x 

10 

x 10 


= 13.32 + 0.2 kcal 
S* = 


-30.7 + 1.2 cal. 
















82 



Fig. 11-10 

















83 


Table 

11-16. 

Alkaline of l-methoxy-2,2,3,4,4-pentamethylphosphetane 
1-oxide (86) at 25°. 

Time 

hr. 


B 

Cone. NaOH 

log B A/A B 

0 0 


1.5 


2.103 x 10" 2 

0.0037 

8.5 


2.028 

0.0097 

24.0 


1.794 

0.0303 

30.0 


1.742 

0.0355 

48.5 


1.539 

0.0582 

57.5 


1.474 

0.0666 

77.5 


1.322 

0.0883 

103.2 


1.128 

0.1225 

144.0 


9.545 x 10" 3 

0.1614 

192.0 


7.804 

0.2122 

A = 3.355 x 
0 

B = 2.158 x 
0 

10 2 

io' 2 

k 2 = 5.88 x 10' 5 
5.88 +_ 0.069 

M. -1 Sec." 1 , 
x 10" 5 M." 1 Sec.' 1 

Table 

11-17. 

Alkaline Hydrolysis of 1- 
phosphetane 1-oxide (86) 

methoxy-2,2,3,4,4-pentamethyl- 
at 25°. 

Time 

hr. 


Cone. NaOH 

B 

log B A/A B 
o o 


24.0 


1.629 x 10“ 2 

0.0203 

96.0 


1.246 

0.0561 

120.0 


1.177 

0.0649 

192.0 


8.92 x 10" 3 

0.1106 

240.0 


7.460 

0.1446 

312.0 


6.330 

0.1786 

A = 0.258 

0 

M 

k 0 = 5.72 x lO -5 

M.' 1 Sec." 1 

B = 0.01946 
o 

M 

z 

+ 2.01 

x 10' 6 



k_ = 5.72 + 0.20 

x 10" 5 M." 1 Sec." 1 


Sec 


-1 



























84 


Table 11-17. Alkaline Hydrolysis of l-methoxy-2,2,3,4,4-pentamethyl- 
phosphetane 1-oxide ( 86) at 25°. 


time 

hr. 

Cone. NaOH 

B. 

Log B A/A B 

6 O O 



24.0 

1.629 

X 

10~ 2 

0.0203 

96.0 

1.246 

X 


0.0561 

120.0 

1.177 



0.0649 

192.0 

8.920 

X 

lO' 3 

0.1106 

240.0 

7.460 

X 


0.1446 

312.0 

6.330 



0.1786 


A = 0.0258 M . c ,,, in -5 M -1 _ -1 

o = 5.72 x 10 M. Sec. 

B = 0.01946 M 1 ^ in -6 

o +_ 2.01 x 10 

k 2 = 5.72 +_ 0.20 x 10" 5 M. _1 Sec. 












85 


Table 11-19. Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide ( 15) at 0°, by conductivity. 


Time R -R(R ro -R) 

Sec. ohms 


100 

150 

250 

166 

500 

182 

750 

196 

1000 

207 

1500 

228 

2000 

246 

2500 

267 

3000 

280 

3500 

298 

4000 

307 

5000 

329 

6000 

349 

CO 

545 


A = B 2.00 10' 2 M. 

o o 


0.3797 

0.4380 

0.5014 

0.5616 

0.6124 

0.7192 

0.8227 

0.9604 

1.0566 

1.2065 

1.2900 

1.5230 

1.7810 


= 3.10 x 10" 2 M.' 1 Sec.' 1 

+_ 2.79 x 10" 4 

= 3.10 +0.028 x 10' 2 M.' 1 Sec. 














86 


Table 11-20. Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide ( 15) at 15°, by conductivity. 


Time Resistance -R/ (R^ -R) 

Sec. R ohms. 


100 

122 


0.4729 

200 

135 


0.5510 

300 

145 


0.6170 

400 

153 


0.6740 

500 

161 


0.7352 

600 

167 


0.7840 

700 

173 


0.8357 

800 

180 


0.9000 

1000 

194 


0.9588 

1250 

205 


1.1714 

1500 

218 


1.3457 

2000 

238 


1.6760 

2500 

252 


1.9687 

3000 

262 


2.2203 

3500 

274 


2.5894 

4000 

281 


2.8383 

4500 

291 


3.2697 

5000 

297 


3.5783 

oo 

380 



A = B 0.02 M. 

o 0 

k 2 

9.05 x 10 2 
+ 8.24 

M.' 1 Sec.' 1 , 
x 10" 4 


7.05 + 0.82 x 10 2 M." 1 Sec. 













87 


Table 11-21. Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide ( 15) in water at 15°. A q =B o = 


Time Resistance -R/ (R^ -R) 

Sec. ohms 


100 


124 

.4844 

200 


133 

.5385 

300 


143 

.6034 

400 


152 

.6667 

500 


157 

.7040 

620 


167 

.7840 

700 


174 

.8447 

800 


179 

.8905 

1000 


192 

1.0213 

1200 


202 

1.1348 

1400 


213 

1.2754 

1600 


222 

1.4051 

2000 


237 

1.6573 

2400 


247 

1.8571 

2800 


259 

2.1405 

3600 


277 

2.6893 

4200 


286 

3.0426 

00 


380 


k 2 

9.09 

9.09 

x 10' 2 + 6.13 x 10" 4 M." 1 Sec." 1 

+ 0.061 x 10" 2 M.' 1 Sec." 1 
















88 


Table 11-22. Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide (15) at 25.1°, by conductivity. 


Time Resistance -R/ (R^ -R) 

Sec. R ohms 


75 

108 



0.5714 

125 

115 



0.6319 

175 

121 



0.6875 

2225 

129 



0.7679 

300 

137 



0.8563 

400 

147 



0.9800 

500 

156 



1.1064 

600 

161 



1.1838 

800 

179 



1.5169 

1000 

189 



1.7500 

1200 

199 



2.0306 

1500 

210 



2.4138 

2000 

223 



3.0135 

2500 

235 



3.7903 

00 

297 




A = B 0.02 M 

0 0 

k 2 * 

1.44 

1.44 

X 

+ 

+ 

10" 1 M. 1 Sec.' 1 

1.2 x 10" 3 

0.012 x 10' 1 M.~ 


Sec 


-1 

















89 

Table 11-23. Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide ( 15) at 25.1°, by conductivity. 


Time 

Sec. 

Resistance 

R ohms 

-R/ (R„ -R) 



100 

193 

0.5288 

220 

205 

0.5989 

300 

220 

0.6509 

400 

232 

0.7117 

500 

2422 

0.7658 

600 

256 

0.8477 

800 

275 

0.9717 

1000 

291 

1.0899 

1400 

322 

1.3644 

1700 

340 

1.5569 

oo 

558 


A = B = 0.01 M. 
o o 

k 2 = 1.58 + 0.012 

x 10' 1 M." 1 













90 


Table 11-24. 

Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide (15) at 33.9° , by conductivity. 

Time 

Sec. 

Resistance 

R ohms 

-R/ (R„ -R) 


100 

174 

0.6237 

150 

181 

0.6654 

200 

191 

0.7290 

300 

205 

0.8266 

400 

220 

0.9442 

500 

232 

1.0498 

600 

244 

1.1675 

700 

263 


800 

263 

1.3842 

1000 

281 

1.6337 

1500 

312 

2.2128 

2000 

335 

2.8390 

2500 

351 

3.4412 

3000 

366 

4.2070 

00 

453 


A = B = 0.01 M 

0 0 

k 2 = 2.68 x 

+ 

10' 1 M." 1 Sec." 1 

3.35 x 10" 3 


268 + 0 

.034 x 10" 1 M." 1 Sec." 1 


Sec 


-1 
















91 


Table 11-25. Alkaline Hydrolysis of l-methoxy-2,2,3,3-tetramethyl- 
phosphetane 1-oxide ( 15) at 33.9° , by conductivity. 


Time 

Sec. 

Resistance 

R ohms 

-R/ (R„ -R) 



100 

172 

0.6078 

150 

181 

0.6606 

200 

190 

0.7170 

250 

197 

0.7636 

300 

203 

0.8056 

400 

219 

0.9280 

500 

232 

1.0404 

600 

243 

1.1462 

700 

254 

1.2637 

800 

262 

1.3575 

1000 

281 

1.6149 

1500 

314 

2.2270 

2000 

334 

2.7603 

2500 

351 

3.3750 

3000 

366 

4.1124 

00 

455 


A = B = 0.01 M 
o o 

k 2 

2.64 x 10' 1 M." 1 Sec. -1 


+ 2.74 x 10 


-3 


2.64 + 0.027 x 10 -1 M. -1 


Sec. 


-1 



















92 


Table 11-26. Activation Parameters for Alkaline Hydrolysis of 1-me- 
thoxy-2,2,3,3-tetramethylphosphetane 1-oxide ( 15) in 
water. 


Temperature k 2 M. * Sec.”'*' AH^ AS^ 

OK kcal/M e.u. 


273.1 

3.10 

x 10 2 
-2 

273.1 

3.31 

x 10 

-2 

288.1 

9.05 

x 10 

-2 

288.1 

9.09 

x 10 

-1 

298.2 

1.44 

xlO 

-1 

298.2 

1.42 

x 10 

-1 

307.0 

2.68 

x 10 

-1 

307.0 

2.64 

x 10 


9.63+0.36 -29.9+1.30 
















93 












94 


3. Hydrolysis of Phosphonium Salts 

90 

a. Background . Hoffmann first studied the thermal decompo¬ 
sition of quaternary ammonium hydroxide. He observed that tetramethylam¬ 
monium hydroxide yields trimethylamine and methanol on heating: 

(ch 3 ) 4 n + oh- A (ch 3 ) 3 n + ch 3 oh 

However, if one of the alkyl groups in the salt possesses a B-hydrogen 

atom, the products of the reaction are a tertiary amine, an alkene and 

„ 91 

water : 

(C 2 H 5 ) 3 fi-CH 2 -CH 3 + OH' ^ C 2 H 5 ) 3 N + CH 2 =CH 2 + H 2° 

An extensive study by Hoffman on the B-elimination led to the well-known 

92 

Hoffmann degradation which is so widely used in organic chemistry . 

In contrast to the ammonium hydroxides, tetraethylphosphonium hydrox- 

93 

ide thermally decomposes to yield triethylphosphine oxide and ethane 

CSV/ + 0H ‘ ( c 2 h 5 ) 3 p=o ♦ c 2 h 6 

The generality of the reaction was established by Hoffman. 

94 

Letts and co-workers found that the decomposition of tetrabenzyl- 

phosphonium hydroxide could be carried out at comparatively low temperature 

95 

to yield tribenzylphosphine oxide and toluene. Meisenheimer and Ingold 
96 studied the reaction in detail with variously substituted phosphonium 
salts and observed that the ease of elimination of substituents parallels 
the stability of the corresponding anion. The following order of decreas¬ 
ing ease of elimination of substituents was observed: 
benzyl> phenyl> methyl> B-phenethyl> ethyl> higher alkyls 
The study of this preferential elimination of groups was the basis of the 





























































































































95 


isolation of an asymmetric phosphine oxide 


97. 


(PhCH 2 ) 4 f 

0 

(PhCH 2 ) 2 PR- 


i) Red n 

ii) R1 


(PKCH 2 ) 3 P=0iiy^li!_ (PhCH 2 ) 3 ?R 
(PhCH 2 ) 2 ?R R' I" —-» PhCH 2 P \ R - 


The foregoing does not imply that the olefin formation or similar 

reactions analogus to those in ammonium salts are unknown in phosphonium 

98 

salts. Hey and Ingold have shown that elimination of an alkene could 
be a predominant pathway for the decomposition of phosphonium salts with 
proper substituents: 

p h nH - 

X CH - CH„P (nBu) —-* Ph^C = CH_ + (nBu)_P + H.O 

y Z Z> Z Z Z> Z 

Ph 

Analogous elimination of the phosphine moiety as the major reaction 

has also been observed in the following cases described by Horner and 
99 

Hoffmann : 

?(ch 2 oh) 4 + 0H~ -> p(ch 2 oh) 3 + ch 2 o + h 2 o 

£(ch 2 ci) 4 + OH"-> p(ch 2 ci) 3 + ch 2 o + HCI 


It might be expected that the yield of the olefin could be increased 

98 

by using a stronger base such as ethoxide ion. Hey and Ingold studied 
the decomposition of quaternary phosphonium ethoxide. They observed that 
only a small increase in the yield of olefin was obtained in the compounds 
with favorable structure. The predominant path was the formation of an 

alkane. 


Thus, the alkaline decomposition of phosphonium salts generally leads 























































































96 


to a phosphine oxide and a hydrocarbon. A systematic investigation of 
this reaction has been reported recently ^6,100^ McEwen and co-workers* 00 
measured the kinetics of the base catalyzed decomposition of benzylethy- 
methylphenylphosphonium chloride in aqueous sodium hydroxide solution: 

They observed that the reaction was third-order overall, first-order in 
the concentration of the phosphonium salt and second-order in the hydroxide 
ion concentration i.e.. 

Rate = k* [ P + ] [OH] 2 


Alkaline decomposition of a series of meta and para substituted benzyl- 
tribenzylphosphonium halides in water-1,2-dimethoxyethane solution were 
studied by McEwen and co-workers* 01 * * 02 . In addition to the observation 
that all phosphonium salt decomposition reactions were third-order, the 
following conclusions were drawn: 


i♦ the relative ease of elimination of various benzyl groups para- 
lell their stability as anions, 

ii • the relative ease of elimination of a group is influenced by the 
nature of the non-departing groups, 

iii• a good correlation with values in an application of the Hammett 

equation was obtained. 

98 

Hey and Ingold had arrived at almost the same conclusions except that 

the dependence of the rate on the non-departing group was not recognized. 

102 

On the basis of the above data, McEwen and co-workers proposed the 
following step-wise mechanism: 




+ OH 


R.P-OH 

4 

















i > 







































97 


k 

. „ 2 . 

R.P-OH + OH «- R.P-0 + H o 0 

4 k_ 2 4 2 

- k 3 9 

R.P-0 R ,P = 0 + R W 

4 4 

R" + H 2 0 ——» RH + OH" 

A steady state approximation, assuming the rate of change of the concen¬ 
tration of intermediates [R^POH] and [R^P-O] to be zero leads to the fol¬ 
lowing expression: 

k 2 k 3 [OH"] 

k -l (k -2 * k 3 J + k 2 k 3 [0H ’ ] 

For the reaction to be third-order the following condition must be ob¬ 
served: 

k 2 k 3 « k _l( k _2 * V 

Steps 2 and 3 may be discrete or concerted. 

b. Result and Discussion . The step-wise mechanism postulated above 

includes the formation of intermediates in the equilibrium process. This 

98 

is contrary to the original assumption by Ingold . He assumed the reac¬ 
tion to be second-order overall, in which the attack of hydroxide ion was 
the rate determining step. The equilibrium step can be easily tested in 
a system in which pseudorotation of the intermediate phosphorane followed 
by reversal of the new intermediate would lead to isomerization of the 
starting material. As yet, there have been no reports in which such a 


— d[P ] 


dt 


= k x [P ] [OH'] 



















































98 


racemization of the starting material has been shown to occur. This has 
been mainly due to the fact that the intermediates in such cases probably 
have a high energy barrier for pseudorotation. The phosphetanium salts 
are unique for two reasons: 

i) due to the strain in the ring, the formation of the intermediate 
should be facilitated, 

ii) pseudorotation of a properly substituted phosphorane intermediate 
may involve low energy barriers. 

The scheme in fig. 12 represents the minimum number of different isomeric 
phosphoranes the s>stem would have to go through, in order to change from 
the trans to the cis isomer of a hypothetical phosphetanium salt and vice 
versa: 




Thus, a minimum of three pseudorotations and four intermediate phospho¬ 
ranes are required to change the configuration of the system from trans 
to cis and reverse. 

It is conceivable therefore, that the cis isomer of a phospetanium 
salt can be converted to the trans isomer. This possibility of isomer 


interconversion can be verified very conveniently by nmr spectroscopy. 






























































































99 


The chemical shift of at least one of the groups in the compounds inves¬ 
tigated was sufficiently different to demonstrate that interconversion 
did (or did not) occur. Various isomeric compositions (9:1), (1:1) (1:4) 
of l-benzyl-l-phenyl-2,2,3,4,4-pentamethylphosphetanium bromide ( 94) in 
deuteriochloroform were treated with one drop of IN sodium hydroxide solu¬ 
tion. An almost instantaneous change occured to give a fianl composition 
of 1:4. The results of this and other compounds investigated under simi¬ 
lar conditions are given in table 27. 



CH 2 Ph 


© 


Br 


(94); R=Ph 
(24); R=CH 


a, RjCH^ trans 


^ b_, R,CH^ cis 


103 


However, contrary to these results are the observations of Trippett 

He has claimed that (94a) and (24a) each yield a mixture of isomeric oxides 

on treatment with base. Quenching of the corresponding ylid reaction with 

n-buytllithium was shown to give a mixture of isomeric phosphetanium salts 

in each case; as would be expected on the basis of the equilibration of 

salts. However, Trippett reported that alkaline decomposition of salts 

(94b) and (24b) each yielded a predominance of only one isomeric oxide. 

This led him to suggest that an isomer cross-over in (94a) and (24a) occur- 

2 

red through a ylid intermediate which was proposed to go through an sp d 
planar transition state in order to yield the isomeric ylid. The latter 
would abstract a proton from the solvent to give the isomeric phospheta¬ 
nium salt. He argued that the formation of the proposed planar transition 
state was facilitated due to the interaction between the 2-methyl group 
and the anionic substituent at the phosphorus in the ylid in (94a) and 
(24a) only. The mechanism above would be expected to lead to a complete 


conversion of both (94a) and (24a) to (94b) and g4b), respectively. This 































































































100 


hypothesis can be easily disproved because the same equilibrium is reach¬ 
ed starting with either isomer of the salt (table 27). Furthermore, it 
was found that 1-phenyl-1,2,2,3,4,4-hexamethylphosphetanium bromide ( 27) 
on treatment with sodium deuterioxide in D^O underwent an isomer inter¬ 
conversion faster than the deuterium exchange. If the isomer interchange 
occurred through an ylid intermediate, the rate of deuterium exchange would 

be expected to be faster or equal to the rate of isomer corss-over. The 

103 

planar transition state proposed by Trippett would also be anticipated 
to be highly strained. It should be pointed out here, however, that the 
conditions of Trippett's experiments and those of this investigation were 
not identical. 

Another interesting feature of the phosphetanium salts is the forma¬ 
tion of predominantly one isomer of the phosphetane oxide during the alka¬ 
line decomposition of the corresponding benzyl salts. Thus, various mix¬ 
tures (9:1, 1:1, 1:4) of the cis and trans isomers of ( 94) when treated 

with alkaline sodium hydroxide give a predominance (1:9) of one isomeric 

104 

phosphetane oxide ( 20) ( cis : trans) . Similarly, treatment of either 
pure isomer of ( 24) with alkali gives rise to an 85:15 mixture of oxide 
( 22) with the same isomer of the oxide predominating in each case. The 
rationale for this observation may be analogous to the one proposed for 
the predominant formation of one isomer of acid chloride (5). Thus, 
an addition of a molecule of base to the cis phosphetanium salt leads to 
intermediate ( 95) (fig. 11), whereas phosphorane ( 96) is produced from 
the trans salt. Pseudorotation of these intermediate leads to other tri- 
gonal-bipyramidal structures. However, during the decomposition, the 
steric interaction of the solvated hydroxyl group with the methyl group 
at position 3, leads to a faster decomposition of this intermediate ( 98) 
than that of (97), i.e., k'<< k"(fig. 11). 






















































































101 


Table 11-27. Equilibration of Phosphetanium Salts with Base. 


Compound Solvent Composition (transfers) (*) 

Initial Final 


(94) 

DCC1 3 

9:1,1:1 

1:4 

(94) 

h 2 ° 

9:1,1:1 

1:2.5 

(24) 

DCC1 3 

95:5,5:95 

1:2 

(24) 

H 2° 

95:5,5:95 

1:2 

(27) 

»2° 

95:5 

1:2.5 

(18) 

DCC1 3 

28:1 

3.4:1 

(27) 

d 2 ° 

8.7:1 

1.5:1 


(D-exchange 15%) 


(*) Cis and trans relationship for the substituent at phosphorus with 


respect to 3-methyl. 
















102 




Fig. 11-13. 











103 


In view of the above discussion, the alkaline decomposition of 
optically active benzylethylmethylphenylphosphonium iodine to give the 
phosphine oxide with inverted configuration appears contradictory: 


Ph 


Ph 


C 2 H 5 


CH 3 


/ 


CH 2 Ph 


OH 


0 = P 




C 2 H 5 

CH„ 


This, however, can be easily explained by the apparent lack of pseudoro¬ 
tation in the intermediate phosphorane, since pseudorotation would probab¬ 
ly be energetically unfavorable. 

In this context, the results of Marsi are very interesting. An 
alkaline decomposition of each isomer ( cis and trans) of 1,3-dimethyl-1- 
benzylphospholanium salt led to the oxide with retention of configuration: 









































































104 


Trippett 103 has recently reported the formation of a different iso¬ 
meric composition of oxides ( 20) and ( 22) starting with the different iso¬ 
mers of the corresponding benzyl salts ( 94) and (24), respectively. The 
results are inconsistent with the observation of equilibration of salts 
which would necessarily lead to the same ratio of isomeric oxides, re¬ 
gardless of the stereochemistry of the starting salt. Marsi proposed that 
the attacking hydroxide ion enters from the apical position; but that the 
benzyl anion departs from an equatorial position. Although, this assump¬ 
tion explains the retention mechanism it can be questioned on two grounds: 
the principle of microscopic reversibility and the principle of least 
motion (see also page 67 ) 33,84 both of which require apical entry and api¬ 
cal departure. In order to interconvert the isomers in this case, the 
phosphorane intermediate initially produced, must go through a minimum of 
three pseudorotational steps. It appears therefore, that the five-membered 
ring is not as strained as the four-membered ring because the C-P-C ring 
angle in this case is 95° 68 compared to 83° 27 in the four-membered ring. 
As a result, the energy of placing an electon donating methyl group in the 
apical position may not be overcome by the relief of the strain achieved 
in going from the starting material to the intermediate. Thus, only one 
pseudorotation (e.g., step 2 or 4) occurs which leads to retention of con¬ 
figuration. It is equally probable that the molecule does not go through 
a pseudorotation as a discrete step but only as a part of a decomposition 
step in the transition state. 

Similar to the interconversion of the isomeric salts, the isomer in¬ 
terchange of the oxides can also be brought about by base. In this case, 
however, refluxing of the oxide with IN sodium hydroxide in 95% ethanol 























































































105 


is required . The process is also complicated by the simultaneous pro¬ 
duction of the open-chain acids. Thus, pure (98%) cis l-phenyl-2,2,5,4,4- 
pentamethylphosphetane oxide ( 20) was treated with alkali under the above 
conditions for 14 hr.. The unreacted oxide was a mixture of the trans and 
cis isomers in the ratio (22:78) respectively; as determined by nmr in¬ 
tegration. The interconversion can take place through the pseudorotation 
of the phosphorane intermediate (see scheme belwo) which would result in 
a symmetrical intermediate ( 99) . Decomposition following a pseudorotation 
of ( 99) would give both isomers with equal probability. Since the phos¬ 
phine oxide ( 20) does not have any “-protons, the necessity of a ylid in- 

103 

termediate can be easily ruled out in this case. 



The isomerization of the oxide ( 20) can also be brought about by other 

107 

reagents. Mislow and co-workers have demonstrated recently that in the 
presence of lithium aluminum hydride, ( 20) undergos isomer cross-over. The 
following intermediate was proposed: 


A1 


LI 



4 


(the alkaline decomposition of the phosphonium salts) as shown earlier is 
third-order and passes through a phosphorane intermediate. An inves¬ 
tigation of the kinetics of the alkaline decomposition of the salts was 
undertaken with a two-fold purpose: 












106 


i) to ascertain whether the reactions follow third-order kinetics, 

ii) and to obtain the magnitude of the acceleration of the rate, 

produced on formation of a relatively strain free intermediate. 

The available phosphetanium salts provuded an interesting set of 

substrates. Thus, exocyclic cleavage of the benzyl group could be studied 

in compounds (24), ( 26) , (94), ( 100) , ( 101) . Compounds ( 27) and ( 102) 

were selected in order to investigate the rates of ring opening and ring 

2 

expansion reaction, respectively, Chorvat showed that compound ( 102) is 
hydrolyed by base to yield open-chain compound (103): 


. CH 

pV 3 


OH 


0 


CH. 


V 


c— c — c— 3 

I I x Ph 


Ph 


( 102 ) 


(103) 


Compound ( 27) on the other hand, was shown to undergo ring expansion to 
yield (29a) (see p.ll ). The rate of the alkaline hydrolysis of ( 104) was 
also examined in order to provide a comparison of the phosphetanium salts 
with open-chain systems. Table 28 shows the rates of different compounds 
in 50% ethanol-water at 25°. 

An inspection of table 28 reveals that compounds ( 26) and ( 24) hy¬ 
drolyzed (60) and 15 times faster than is triphenylbenzylphosphonium bro¬ 
mide ( 105) . The latter is substituted with three electron withdrawing 
phenyl groups and would be expected to be hydrolyzed much faster than its 
alkyl analog. Thus, the ease of intermediate formation in the cyclic com¬ 
pounds is reflected in their rates of hydrolysis. 

The dramatic effect of the strain can be seen more clearly in the 

108 

following comparison. Aksnes has shown that the salt with a five- 























































































































107 


Table 11-28. Alkaline Hydrolysis of Phosphonium Salts. Third-Order 
Rate Constants at 25° in 50% Ethanol-Water. Ionic 
Strength 0.1M. 


Compound 



-2 -l 

(M Z Sec. 


) 


(24) 

1.09+0.02 x 10° 

(26) 

4.43+0.08 x 10° 

(27) 

1.54+0.013 x 10' 1 

(94) 

8.13+0.032 x 10 _1 

(100) 

2.097+0.03 x 10 3 

(101) 

1.37+0.023 x 10 2 

(102) 

2.82+0.026 x 10° 

(104) 

1.29+0.018 x 10* 1 


(105) 7.08 x 10" 2 (ref. 


'X 

'T 


109) 


(24) : 
(26): 
(27): 
(94): 


R 1 =CH 3, R 2 =CH 2 Ph 
R 1 =R 2 =CH 2 Ph 

R 1 =CH 3 , R 2 =Ph 

R 1 =CH 2 Ph, R 2 =Ph 


\ 


% — R 
I 

R„ 


1 

Br- 


( 100 ) 


P— Ph 
CH Ph 

Br' 


P — Ph 
CH Ph 
(101) Br' 


— P 


CH, 


(104) 


Br 


+ 

(Ph) P-CH ? Ph Cl 
(105) 


P 


/ \ 

Ph r ch 3 


(107) 






I 

(108) 



Ph^ ^CH 0 Ph (104) 
Br 2 - 


( 106 ) 

© 

(CH 3 ) 3 P-CH 2 Ph 

Br 


( 109 ) 




















































108 


membered ring ( 107) is hydrolyzed 1300 times faster than the salt with a 
six-membered ring ( 108) . Both of these salts have been shown to decompose 
with exclusive exocyclic cleavage. Salt ( 104) is known to undergo exclu¬ 
sive exocyclic cleavage as well. Based on the assumption that the same ar¬ 
guments would hold for compound ( 106) , it would be reasonable to expect 
that ( 104) would also hydrolyze about 1300 times faster than ( 106) . In 

4 

table 28, compound ( 100) which decomposes 10 times faster than the five- 

4 7 

membered ring, salt ( 104) , should react 1300 x 10 10 times faster than 

( 106) . Since the latter rate is probably as fast as that of the open-chain 
system ( 109) . The hypothesis, therefore that the strain in the phospheta- 
nium salts is relieved in going to the trigonal-bipyramidal intermediate 
is corroborated. 

A peculiar solvent effect is demonstrated in table 29. The reaction 
of the phosphonium salt with hydroxide ion involves the reaction between 
two oppositely charged species. In the following equation, a relation be¬ 
tween the rate constant and the dielctric constant of the medium is given 


(109): 


In k=ln k' 
o 


NZ _ , e 
aZ b. 


D R T r* 


where k=calculated rate constant in a medium of infinite dielectric con¬ 
stant, N=Avogadro's number, Z and Z, are the charges of the two ions, e= 
electron charge, D=dielectric constant, R=gas constant, T=absolute tem¬ 
perature and r*=distance between the ions in the activated complex. The 
equation predicts an increase in the rate of the reaction in a medium of 
lower dielectric constant for a reaction between two oppositely charged 
ions. Thus, the acceleration of the rate of hydrolysis in 50% ethanol- 
water is not unanticipated. However, the magnitude of the acceleration 























































































109 


is unexpectedly large and so the effect of the solvent may be more complex. 

The activation parameters for one system ( 100) in aqueous solution are 
given in table 43. Unfortunately, the paucity of such data in other sys¬ 
tems prevents a meaningful conclusion at the present time. Further, the 
activation parameters are overall values and hence, do not reflect on any 
individual step of the reaction. 


Table 11-29. Solvent Effect on the Hydrolysis of Phosphonium Salts 

Third-Order Rate Constants at 25°. Ionic Strength 0.1M. 


Compound 

Solvent 

k 3 (M" 2 Sec. -1 ) 

(Ethanol-water) 
k^ (water) 


(94) 

Water 

6.80 x 10“ 2 




1 

1200 

(94) 

50% ethanol-water 

8.16 x 10 


(100) 

Water 

1.75 x 10 1 




7 

119.8 

(100) 

50% ethanol-water 

2.10 x 10 




-1 


(101) 

Water 

7.70 x 10 




O 

178.3 

(101) 

50% ethanol-water 

1.37 x 10 






















110 


c. Experimental . The alkaline decomposition of phosphonium salts 
was determined, initially by titration methods. Separately prepared so¬ 
lutions of the phosphonium salt (0.005 N) and NaOH (0.005 N), both in 
50% ethanol-water, were equilibrated in a double chambered flask. The 
solutions were mixed and a 5.0 ml aliquot of the mixture was quenched with 
10 ml of 0.005 N HC1 solution. The excess of the acid was titrated against 
a standard NaOH solution to determine the amount of unreacted alkali in 
the reaction mixture. The concentration of the phosphonium salt was cal¬ 
culated on the basis of the stoichiometry (1:1) since: 

R 4 ? = 20H~ -=► R P = 0 + R” + H 2 0 ->■ RH + OH - 

.*. R.f + OH -» Products 

4 

The results obtained for compounds ( 94) and ( 102) are tabulated (Tables 30-36) 

The ionic strength of the solution was not maintained constant in these 

108 2 
runs . A straight line was obtained by plotting 1/C vs time showing 

the reaction to be third-order. The reactions were followed to at least 
60% completion. No correlation was obtained for a second-order reaction. 

Similarly, in the case of compound ( 100) , the third-order rate con¬ 
stants were determined in the presence of 0.2M KC1 solution in water. The 
rate constants for three different temperatures were obtained which were 
subsequently used for computing the activation parameters. 

After having ascertained that the reactions are third-order, the me¬ 
thod of following the kinetics of these reactions was changed. According 

to the third-order equation: 

* 2 + 

Rate = k^ A B where A = Cone of P salt 


B = Cone of NaOH 




























































Ill 


If the concentration of the base is not allowed to change appreciably, the 

equation becomes: 

Rate: = (k_ B 2 ) A = k , A 
3 obs 

•k 

From the value of k , at different base concentration, k„ can be deter- 

obs 3 


* 7 

, , = k 

obs 3 


mined from the relation: 

k 

c 

A pseudo first-order plot was obtained by following the decrease in the 
concentration of the phosphonium salt. This could be conveniently accom¬ 
plished with the aid of uv spectrophotometer. In a separate experiment 
the wave length at which a maximum change in absorbance for a compound 
occurred, was determined. The need for the determination of extinction co¬ 
efficient was obviated because of the independance of pseudo first-order 
reaction on the concentration of the species in question. A plot of log 
(A -A ) vs time where A^ is absorbance at time t and A that at time infin- 
ity; gave a straight-line whose slope was equal to 1/2.303 x k Deter¬ 
mination of was performed at various alkali concentration. The alkali 

solution was prepared from a standard stock solution; e.g., for the prepa¬ 
ration of 0.1 N NaOH in 50% ehtanol-water solution: 10 ml of 1 N NaOH di¬ 
luted with exactly 50 ml of absolute ethanol followed by diultion to exact¬ 
ly 100 ml with distilled water at 25°. To make the total ionic strength 
to 0.2 M, 0*58454 g of sodium chloride was dissolved in the solution prior 
to dilution. 

Equal volumes of both the phosphonium salt solution and NaOH solution 
(containing the requisite amount of NaCl to make up the required ionic 
strength) were mixed after equilibrating them separately at 25° for at 
least 20 min.. The resultant solution had half the ionic strength and half 
the base concentration of the original base solution (In the example above. 









































112 


the final Cone. NaOH = 0.05 N, ionic strength = 0.1 M.. The concentration 

-4 

of the phosphonium salt was kept to near 5 x 10 M (after mixing) so that 
its contribution to the ionic strength could be neglected. The uv determin¬ 
ations were performed on the double beam Cary 11 spectrophotometer with the 
cell compartment jacketted with and connected to tubes through which water 
at 25° was circulated. No reference was used except for the initial setting 
of the instrument in which case matched cells containing the solvent of the 
reaction was used. The sweep was performed at a fixed wave length by running 
the chart at a known constant speed. The reading at time =°° was taken after 
about 10 half lives for each reaction. An exponential plot fo the change 
in absorbance vs time thus obtained from the plot was utilized for computing 
the values of logCA^-A^). The k^g was determined by using a least-squares 
program designed for Wang 300 electronic calculator (except where indicated 
in which case a graphical determination of k was made). The third-order 

rate constants were similarly computed from the one parameter least-squares 

2 

operations on k ^ vs B . 























































113 



To 8D 120 160 200 240 280 320 

Fig. 11-14. time sec. 

Alkaline Hydrolysis of l-benzyl-l-phenyl-2, 2, 3, 3-tetramethylphosphefeanium bromide (100) 
Concentration of NaOH 0.02 M. Pseudo first-order plot of log (Af.-A^) vs time, in wateTTt 25° 







114 



sq° M 


Z 0T x 


Alkaline Hydrolysis of l-benzyl-l-phenyl-2, 2, 3, 3-tetramethylphosphetanium bromide (100) 
at 25° in water. Plot of k Q bs vs [ Bl 



















115 


Table 11-30. Alkaline Hydrolysis of 1-benzyl-1-phenyl-2,2,3,4,4- 

pentamethylphosphetanium bromide ( 94) in 50% ethanol- 
water at 24.9° by titrations, Ionic strength not con¬ 
stant. A =B =0.005 M. 
o o 


Time Cone. NaOH l/C 2 -l/Co 

Sec. B. M. 


165 

2.98 x 10" 3 

0.726 x 10 

240 

2.66 

1.013 

310 

2.43 

1.294 

400 

2.20 

1.666 

800 

1.73 

2.941 

1200 

1.46 

4.291 

1800 

1.18 

6.682 

2400 

1.10 

7.800 

3000 

0.94 

10.917 


k 3 = (1.798+0.055)x/0 2 M _2 Sec. 
















116 


Table 11=31. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,4,4- 

pentamethylphosphetanium bromide ( 94) in 50% ethanol- 
water at 24.9° by titration. Ionic strength not con¬ 
stant. A B =0.005 M. 

o o 

Time 
Sec. 


Cone. NaOH 
B M 


1/C 2 - 1/Co 


75 

3.70 x IO -3 



3.30 x 10 4 

150 

3.10 



6.410 x 10 4 

250 

2.61 



1.068 x 10 5 

400 

2.19 



1.685 x 10 5 

800 

1.65 



3.273 x 10 5 

1200 

1.40 



4.702 x 10 5 

1800 

1.18 



6.782 x 10 5 

3000 

0.92 



10.680 x 10 5 

k 3 = 1.88 + 0.034 

2 -2 -1 
x 10 M. Sec. 




Table 11-32. 

Alkaline Hydrolysis of 1-benzyl-l 
pentamethylphosphetanium bromide 
water at 5°. Ionic strength not 
5.00 x 10“^M. 

-phenyl - 
(94) in 
constant 

2,2,3,4,4- 

50% ethanol- 

. A B = 
o o 

Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #1) 


1/B 2 - 1/Bo 


300 

4.110 

X 

10- 3 

1.920 

X 

io 4 

600 

3.730 

X 

ID' 3 

3.880 

X 

io 4 

900 

3.460 

X 

ID' 3 

4.353 

X 

io 4 

1200 

3.210 

X 

IQ’ 3 

5.720 

X 

io 4 

1800 

2.900 

X 

IQ' 3 

7.890 

X 

io 4 

2400 

2.580 

X 

IQ’ 3 

1.102 

X 

10 5 

3600 

2.300 

X 

IQ’ 3 

1.490 

X 

io 5 

5500 

1.990 

X 

ID’ 3 

2.120 

X 

io 5 

7000 

1.770 

X 

ID’ 3 

2.792 

X 

io 5 


k 3 = 1.89+ 0.0092 x 10' 


M. 2 Sec. 


-1 

















117 


Table 11-33. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,4,4- 

pentamethylphosphetanium bromide ( 94) in 50% ethanal- 
water at 5°. Ionic strength not constant. A =B = 

5.00 x 10‘ 3 M. ° 0 


Time 

[B] Cone. NaOH 

M 

(Run #2) 

1/B 2 - l/Bo 



400 

3.960 

X 

10- 3 

2.380 

X 

10 

1000 

3.330 

X 

10- 3 

5.020 

X 

10 

2000 

2.760 

X 

ID" 3 

9.130 

X 

10 

3000 

2.400 

X 

10 - 3 

1.336 

X 

10 

4500 

2.100 

X 

io- 3 

1.868 

X 

10 

6000 

1.850 

X 

ID’ 3 

2.522 

X 

10 

9000 

1.600 

X 

io' 3 

3.506 

X 

10 

15000 

1.300 

X 

io' 3 

5.517 

X 

io ] 


k 3 = 1.91 +_ 0.037 x 10 1 M." 2 Sec. 


Table 11-34. 


Temperature 
° K 


Activation Parameters for Alkaline Hydrolysis of 1- 
benzyl-1-phenyl-2,2,3,4,4-pentamethylphosphetanium 
bromide (94) in 50% ethanol-water. 

— — T T 

k 3 M Sec. AH t AS t 

kcal.deg ^ e.u. 


278.2 

278.2 

298.1 

298.1 


1.91 x 10 1 
1.88 x 10 1 
1.798 x 10 2 
1.88 x 10 2 


+ 13.2 














118 


Table 11-35. Alkaline Hydrolysis of 1-phenyl-1-2,2,3,3-pentamethy1- 

phosphetanium bromide (102) in 50% ethanol-water at 25°. 

A =B =0.005 M. 
o o 


# 

Time 

Sec. 

Cone. NaOH 

B M. 

1/B 2 



1 

1,800 

4.4 x 10 

5.834 x 10 

2 

4,500 

3.45 

8.420 x 10 4 

3 

14,400 

2.44 

1.680 x 10 5 

4 

21,600 

2.11 

2.246 x 10 5 

5 

36,000 

1.72 

3.380 x 10 5 


k 3 = 4.075 +0.059 M7 2 Sec. 


Table 11-36. Alkaline Hydrolysis of 1-phenyl-1-2,2,3,3-pentamethyl- 

phosphetanium bromide (102) in 50% ethanol-water at 25°. 

A =B =0.005 M. 
o o 


# 

Time 

Sec. 

Cone. NaOH 

B M 

1/C 2 





-3 

4 

1 

1,800 

4.14 x 10 

5.834 x 10 

2 

4,500 

3.49 

8.210 x 10 4 

3 

14,400 

2.45 

1.666 x 10 5 

4 

21.600 

2.10 

2.268 x 10 5 

5 

36,000 

1.73 

3.341 x 10 5 


k 3 = 4.046 + 0.083 M. 2 Sec. 






















119 


Table 11-37. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 

methylphosphetanium bromide (100) in aqueous solution 

at 25.2°. Ionic strength 0.2M. A =B =4.714 x 10 ^ M. 

& oo 


Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #1) 

1/B 2 



600 

3.875 

X 

to' 3 

6.66 x 10 4 

1194 

3.423 

X 

to 

1 

o 

r—t 

8.535 x 10 4 

2253 

2.991 

X 

i<T 3 

1.118 x 10 5 

3295 

2.590 

X 

10- 3 

1.491 x 10 5 

4390 

2.364 

X 

I—* 

o 

1 

w 

1.789 x 10 5 

5504 

2.210 

X 

10- 3 

2.047 x 10 5 


k 3 = 1.43 + 0.004 x 10 1 M." 2 Sec. 


Table 11-38. Akkaline Hydrolysis of 1-benzyl-1-phenyl-2,2,3,3-tetra 

methylphosphetanium bromide ( 100) in aqueous solution 

at 25.2°. Ionic strength 0.2M. A =B =4.714 x 10“^ M 

o o 


Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #2) 

1/B 2 



1104 

3.516 

X 

IQ" 3 

8.09 x 10~ 4 

2237 

2.930 

X 

to 

1 

o 

r—H 

1.165 x 10' 5 

4276 

2.354 

X 

10- 3 

1.805 x 10“ 5 

5539 

2.179 

X 

10- 3 

2.106 x 10~ 5 

6760 

1.943 

X 

io“ 3 

2.649 x 10“ 5 

7690 

1.922 

X 

t— 1 

O 

1 

to 

2.707 x 10' 5 


k 3 = 1.50 ^0.07 x 10 1 M. -2 Sec. 























120 


Table 11-39. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 
methylphosphetanium bromide ( 100) in aqueous solution 
at 39.6°. Ionic strength 0.2M. A 0 = B 0 =4.714 x 10 -3 m. 


Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #1) 

1/B 2 



588 

3.197 

X 

10- 3 

9.784 

X 

10 4 

1092 

2.673 

X 

IQ’ 3 

1.399 

X 

10 5 

1629 

2.395 

X 

nr 3 

1.743 

X 

io 3 

2205 

2.169 

X 

ID' 3 

2.126 

X 

10 s 

2714 

1.943 

X 

IQ' 3 

2.649 

X 

!0 5 

3172 

1.893 

X 

ID' 3 

2.794 

X 

io 5 

3590 

1.799 

X 

ID’ 3 

3.090 

X 

io 5 

4477 

1.604 

X 

IQ' 3 

3.887 

X 

10 s 


k 3 = 3.63 ^0.11 x 10 1 M. ~ 2 Sec. 


Table 11-40. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 
methyphosphetanium bromide ( 100) in aqueous solution 
at 39.6°. Ionic strength 0.2M. A q =B o = 4.714 x 10"^ m. 


Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #2) 

2 

1/B Z 



595 

3.21 x 10‘ 3 

9.705 

X 

10 

1200 

2.615 x 10' 3 

1.462 

X 

10 

1800 

2.275 x 10" 3 

1.932 

X 

10 

3605 

1.770 x IO -3 

3.192 

X 

10 

4207 

1.688 x 10" 3 

3.501 

X 

10 

4841 

1.627 x 10" 3 

3.778 

X 

10 


















121 


Table 11-41. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 

methylphosphetanium bromide ( 100) in aqueous solution 

at 60°. Ionic strength 0.2M. A =B =4.714 x 10“3 pj 

oo 


Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #1) 

2 

1/B 



80 

3.480 

X 

NO 

1 

o 

f— { 

8.257 

X 

io 4 

130 

3.120 

X 

10- 3 

1.027 

X 

io 5 

175 

2.820 

X 

1<T 3 

1.257 

X 

io 5 

235 

2.645 

X 

io' 3 

1.429 

X 

io 5 

357 

2.285 

X 

-3 

10 

1.915 

X 

io 5 

409 

2.210 

X 

io~ 3 

2.047 

X 

io 5 

468 

2.090 

X 

io- 3 

2.289 

X 

10 s 


k 3 = 1.85 +_ 0.045 x 10 2 M. 2 Sec. 


Table 11-42. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 

methylphosphetanium bromide ( 100) in aqueous solution 

at 60°. Ionic strength 0.2M. A =B = 4.714 x 10-3 M. 

o o 


Time 

Sec. 

[B] Cone. NaOH 

M 

(Run #2) 

2 

1/B 



83 

3.410 

X 

io- 3 

8.600 

X 

10' 

161 

2.915 

X 

io’ 3 

1.177 

X 

10 

233 

2.615 

X 

io' 3 

1.462 

X 

10 

299 

2.417 

X 

io’ 3 

1.712 

X 

10' 

362 

2.250 

X 

1 <T 3 

1.975 

X 

io : 

429 

2.115 

X 

io~ 3 

2.235 

X 

10 

490 

2.022 

X 

io- 3 

2.446 

X 

10' 

600 

1.857 

X 

io- 3 

2.900 

X 

10' 


k 3 = 1.96 +_ 0.012 x 10 2 M." 2 Sec. 



















122 


Table 11-43. 


Temperature 


o 


K 


Activation Parameters for 1-benzyl-1-phenyl-2 , 2,3,3- 
tetramethylphosphetanium bromide ( 100) in aqueous so¬ 
lution. Ionic strength 0.2M. 

k (Average) AH^ AS^ 

5 -2 -1 -1 -1 

M Sec. kcal.deg M e.u. 


298.2 1.47 x 10 1 

312.6 3.48 x 10 1 14.76 -5.55 

333.0 1.89 x 10 2 + 1.57 + 0.67 


Table 11-44. Alkaline Hydrolysis of 1-benzyl-1,2,2,3,4,4-hexamethyl- 
phospetanium bromide (24) in 50% ethanol-water at 25%. 
Ionic strength 0.1M. 


# Cone. NaOH 

(OH"). M. (OH") 2 k obs’ Sec ' 


1 

5.0 x 10~ 2 

2.5 x 

io" 3 

2.74 x IO -3 

2 

6.25 x 10" 2 

3.906 

x 10~ 3 

4.03 x 10~ 3 

3 

7.5 x 10" 2 

5.625 

x IO -0 

5.22 x 10" 3 

4 

8.25 x 10" 2 

6.806 

x IO - ' 5 

8.09 x 10" 3 

5 

1.0 x 10' 1 

1.0 x 

io -2 

1.09 x 10“ 2 

6 

1.0 x 10" 1 

1.0 x 

io" 2 

1.109 x 10' 


k = 1.09+^ 0.03 M. “Sec. 

















123 


Table 11-45. Alkaline Hydrolysis of 1,l-dibenzyl-2,2,3,4,4-pentame- 
thylphosphetanium bromide (26) in 50% ethanol-water at 
25°. Ionic stren th 0.1M. 


# Cone, of NaOH 

(0H‘), M. (OH - ) 2 k obs’ SeC - 


1 

5.0 x 10 -2 

2.5 x 

-13- 

10 

1.094 

X 

icf 2 

2 

5.0 x 10 -2 

2.5 x 

10- 3 

1.041 

X 

io -4 

3 

6.25 x 10 -2 

3.096 

x 10 -3 

1.080 

X 

io -2 

4 

7.5 x 10 -2 

5.625 

x 10 -3 

2.637 

X 

io -2 

5 

1.0 x 10 -1 

1.0 x 

io -2 

4.334 

X 

io’ 2 


k 3 = 4A5 1 0.079 M. 2 Sec. 


Table 11-46. Alkaline Hydrolysis of 1-phenyl-l,2,2,3,4,4-hexamethyl- 
phosphetanium bromide ( 27) in 50% ethanol-water at 25°. 
Ionic strength 0.1M. 


# 

Cone, of NaOH 







(OH - ), M. 

(OH') 

2 

k i 

obs 

,Sec. ■*" 


1 

5.0 x 10 -2 

2.5 x 

io -3 

3.65 

X 

io -4 

2 

5.0 x 10 -2 

2.5 x 

io' 3 

3.53 

X 

io -4 

3 

7.5 x 10 -2 

5.625 

x 10 -3 

8.54 

X 

10 -4 

4 

7.5 x 10 -2 

5.625 

x io -3 

8.62 

X 

io -4 

5 

8.75 x 10 -2 

7.656 

x 10 -3 

1.20 

X 

IO' 3 

6 

1.0 x 10 -1 

1.0 x 

io -2 

1.54 

X 

io' 3 


k 3 = 1.54 +_ 0.013 x 10 -1 M. -2 Sec. 















124 


Table 11-47. Alkaline Hydrolysis of 1-benzyl-1-phenyl-2,2,3,4,4-phos- 
phetanium bromide (94) in aqueous solution at 25%. Ionic 
strength 0.1M. 


# Cone. NaOH 

(OH - ), M. (OH - ) 2 k obs ,Sec. 1 



1 

5.0 

X 

10 -2 

2.50 x 10 -3 

1.54 x 10 -4 

2 

7.5 

X 

io -2 

5.625 x 10 -3 

3.995 x IO -4 

3 

7.5 

X 

io -2 

5.625 x 10" 3 

3.995 x 10 -4 

4 

8.0 

X 

io -2 

6.40 x 10 -4 

4.35 x 10 -4 

5 

9.0 

X 

io -2 

8.10 x IO"* 5 

5.30 x IO" 4 

6 

9.0 

X 

io -2 

8.10 x 10 -3 

5.50 x 10" 4 




-2 

-3 

-4 

7 

8.5 

X 

10 

7.225 x 10 

4.83 x 10 


k 3 = 6.80 -^0.10 x 10 2 M. 2 Sec. 



Table 

11-48. 

Alkaline Hydrolysis of 1-benzyl- 
tamethylphosphetanium bromide in 
25°. Ionic strength 0.1M. 

1-phenyl-2 , 2 ,3,4,4-pen- 
50% ethanol-water at 

# 

Cone. 

NaOH 







- 2 


-1 


(OH ) 

, M. 

(oh r 

k u 
obs 

,Sec. 




-3 

-5 


-3 

1 

5.0 x 

10 

2.5 x 10 

2.28 

x 10 



-3 

-5 


-3 

2 

5.0 x 

10 

2.5 x 10 

2.12 

x 10 



-3 

-5 


-3 

3 

7.5 x 

10 

5.625 x 10 

4.50 

x 10 



-2 

-4 


-3 

4 

1.0 x 

10 

1.00 x 10 

8.20 

x 10 



-2 

-4 


-3 

5 

1.0 x 

10 

1.00 x 10 

8.37 

x 10 



-2 

-4 


-2 

6 

1.5 x 

10 

2.25 x 10 

1.79 

x 10 


k 3 = 8.13 +_ 0.032 x 10 +1 M. -2 Sec. 















125 


Table 11-49. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 
methylphosphetanium bromide ( 100) in aqueous solution 
at 25°. Ionic strength 0.1M. 


# Cone. NaOH 

(OH"), M. (OH") 2 k obs ,Sec. 


1 

2.0 

X 

10 2 

4.0 x IO -4 

6.893 x 10' 3 

2 

2.0 

X 

io " 2 

4.0 x 10" 4 

6.725 x IO -3 

3 

2.5 

X 

io - 2 

6.25 x 10~ 4 

1.09 x 10“ 2 

4 

3.0 

X 

io " 2 

9.0 x 10“ 4 

1.603 x IO -2 

5 

4.0 

X 

io " 2 

1.6 x 10" 3 

2.787 x 10' 2 

6 

5.0 

X 

io - 2 

1.6 x IO -3 

4.422 x 10“ 2 

7 

5.0 

X 

io ~ 2 

2.5 x IO -3 

4.399 x 10' 2 


k 3 = 1.76 +_ 0.0055 x 10 1 M. 2 Sec. 


Table 11-50. Alkaline Hydrolysis of l-benzyl-l-phenyl-2,2,3,3-tetra- 
methylphosphetanium bromide (100) in 50% ethanol-water 
at 25°. Ionic strength 0.1M. 


# 

Cone. NaOH 

_ 2 

-1 


(OH ), M. 

(oh r 

k , .Sec. 
obs 









-3 

—6 



-3 

1 

1.0 

x 

10 

1.0 x 10 

2.31 

X 

10 




-3 

-6 



-3 

2 

1.6 

X 

10 

2.56 x 10 

5.02 

X 

10 




-3 

-6 



-3 

3 

1.6 

X 

10 

2.56 x 10 

5.14 

X 

10 




-3 

-6 



-3 

4 

2.0 

X 

10 

4.0 x 10 

7.80 

X 

10 




-3 

, ~ “6 



-2 

5 

3.0 

X 

10 

9.0 x 10 

1.87 

X 

10 




-3 

-6 



-2 

6 

3.0 

X 

10 

9.0 x 10 

1.94 

X 

10 


k 3 = 2.097 +_ 0.031 x 10 3 M." 2 Sec. 



















126 



Table 11-51. 

Alkaline Hydrolysis of 1-benzyl-1-phenyl-2 , 2 ,3,3,4-pen- 
tamethylphosphetanium bromide (101) in water at 25°. 
Ionic strength 0.1M. 

# 

Cone, of NaOH 

(OH - ), M. 

(OH') 2 k obs' SeC - 1 



1 6.0 x 10 

2 6.0 x 10 

3 7.0 x 10 

4 8.0 x 10 

5 8.0 x 10 

6 1.0x10 

7 1.0 x 10 


k 3 = 7 ' 70 l 0 * 092 x 10_1 M." 2 Sec. 


3.6 

X 

10- 3 

2.73 

X 

Hf 3 

3.6 

X 

io- 3 

2.73 

X 

io- 3 

4.9 

X 

ID' 3 

3.68 

X 

io- 3 

6.4 

X 

io’ 3 

5.20 

X 

io' 3 

6.4 

X 

io’ 3 

5.00 

X 

io' 3 

1.0 

X 

io" 2 

7.85 

X 

io' 3 

1.0 

X 

io" 2 

7.42 

X 

io -J> 


Table 11-52. Alkaline Hydrolysis of 1-benzyl-1-phenyl-2,2,3,3,4-pen- 
tamethylphosphetanium bromide ( 101) in 50% ethanol-water 
at 25°. Ionic strength 0.1M. 


# Cone. NaOH 

(OH~), M. (OH - ) 2 k obs' SeC - 






-3 

-5 

2.09 


, -3 

1 

4.0 

X 

10 

1.66 x 10 

X 

10 




_3 

-5 



-3 

2 

6.0 

X 

10 

3.60 x 10 

4.33 

X 

10 




-3 

-5 



-3 

3 

8.0 

X 

10 

6.40 x 10 

9.03 

X 

10 




-3 

-5 

9.17 


, -3 

4 

8.0 

X 

10 

6.40 x 10 

X 

10 




_? 

-4 

1.43 


. -2 

5 

1.0 

X 

10 

1.0 x 10 

X 

10 

6 

1.2 

X 

io -2 

1.44 x 10" 4 

1.93 

X 

io" 2 


k = 1.37 + 0.023 x 10 2 M.~ 2 Sec. 

3 - 





































































127 


Table 11-53. Alkaline Hydrolysis of 1-phenyl-1,2,2,3,3-pentame- 

thyl phosphetanium bromide ( 101) in 50% ethanol-water 
at 25°. Ionic strength 0.1M. 


# Cone. NaOH 
(OH"), M. 


(OH') 2 


k , ,Sec. 
obs 


-1 


1 2.0 x 10 

2 2.5 x 10 

3 2.5 x 10 

4 3.0 x 10 

5 3.0 x 10 

6 3.5 x 10 

7 4.0 x 10 


4.0 x 10 
6.25 x 10" 
6.25 x 10" 
9.0 x 10" 4 
9.0 x 10" 4 
1.225 x 10 
1.6 x 10" 3 


1.09 x 10 
1.74 x 10 
1.77 x 10 
2.57 x 10 
2.50 x 10 
3.33 x 10 
4.52 x 10 


k = 2.82 +_ 0.026 



Table 11-54. Alkaline Hydrolysis of 1-benzyl-l-phenylphospholanium 
bromide ( 104) in 50% ethanol-water at 25%. Ionic 
strength 0.1M. 


# 

Cone. NaOH 

(OH"), M. 

(OH') 2 

k , ,Sec. ■*" 
obs 



-2 

_3 

-4 

1 

5.0 x 10 

2.5 x 10 

2.886 x 10 


_2 

-3 

-4 

2 

6.25 x 10 

3.906 x 10 

4.94 x 10 


-2 

_3 

-4 

3 

7.5 x 10 

5.625 x 10 

6.89 x 10 


_1 

-2 

-3 

4 

1.0 x 10 

1.0 x 10 

1.297 x 10 


-1 

-2 

-3 

5 

1.0 x 10 

1.0 x 10 

1.313 x 10 


-1 

,, -2„ -1 


k 7 

= 1.29 + 0.0177 x 10 

M. Sec. 


3 

— 

























CHAPTER III 


SYNTHESIS AND REACTIONS OF <=-HALOMETHYLPHOSPHORUS 


COMPOUNDS 


1. Introduction : 

Three and four-membered phosphorus heterocycles have been synthesized 

by using several different routes. The various methods by which this can 

2 

be accomplished have been reviewed . In general, one can envision cycliza- 
tion by the following general procedures in which the position of the ring 
closure is denoted by dotted lines: 



R-P X 

n 


IT 


nx 


X 


XE 


3 

Significant use of the first two approaches have been made . In contrast 


relatively little attention has been directed towards the methods illustra¬ 
ted in approaches III and IV. To date, procedure III has not been employed 
for the synthesis of the four-membered ring, although it has been used for 
the synthesis of larger rings There is only one example of procedure 

IV described in the literature; this was recently reported by a group of 

12 

Russian workers . In this example, a four-membered phosphorus heterocycle 

(110) has been generated by treating phenyl bis-(chloromethyl) phosphinate 

(111) with diethyl malonate in the presence of sodium ethoxide according to 
the scheme: 


128 





























129 


9 CH 9 C1 
PhOP ^ 


x ch 2 ci 

( 111 ) 


+ CH 2 (COOEt) 2 1E - t - -» 



(COOEt) 2 


( 110 ) 


It is apparent from the above discussion that the potential of proce¬ 
dure IV has not been fully realized. A likely starting material for such an 
approach would be a compound with an “-substituent adjacent to the phospho¬ 
rus atom. In order to give a ring closure, the “-substituent should be mo¬ 
derately good leaving group. A possible example is illustrated in a com¬ 
pound with the general formula: 


CXCI Vn K 


(Z) 


3-n 


in which X and Z are replacable groups such as halogen and Y is oxygen, sul¬ 
fur, a lone pair, etc.. A number of “-halomethylphosphorus compounds fall 
in this general category. 

The synthesis of the “-halomethylphosphorus compounds poses a special 
problem. Many potential starting materials are derived from the condensa¬ 
tion of carbonyl compounds with primary or secondary phosphines in the pre¬ 
sence of an acid or a base 

0 

i i roi i 

RPH 2 + 2R CHO -> RP (CHOHR j 2 . L J> R - P (CHOHR ) 2 


Recently, significant progress has been made in this direction ’ 

One limitation of the reaction is that it involves at least three steps. 
Further, the condensation is reversible and thus certain reagents (e.g. PCI , 
S0C1 2 ) cannot conveniently be employed for the purpose of converting the 












130 


hydroxy compound to the halide derivative The procedure of Hellmann 
which involves base decomposition of °=-halomethylphosphonium salts to yield 
^-halomethylphosphine also suffers from limitations; low yields of the pro¬ 
duct are obtained in a mixture of compounds 


Ph£(CH 2 Cl) 3 Cl 


0 CH_C1 
II / 2 

PhP 


OH 


^ Ph 2 P(CH 2 Cl ) 2 + 


21 




+ CH 2 0+ hci 


51% 


CH 2 CH 2 CL 


A similar result was found in this laboratory (p. 148) : 


(Ph) 2 ^—CH 2 CL 
CH 3 


0 

0H~ H 

-- - ■ > Ph 2 P-CH 3 + Ph 2 PCH 3 


19*/. 


677 


Results and Discussion: 


The purpose of the present investigation was to synthesize compounds 
of the general formula: 

X 

(XCRJ P(Z)_ 

2'n 3-n 

in order to circumvent some of the synthetic difficulties described above. 
The synthesis of a compounds with the general formula where n=l was investi¬ 
gated an a working model. A model compound was deemed necessary in view of 
the unknown chemical behavior of “-halo substituents in these molecules. In¬ 
itial studies were directed toward the introduction of substituents by the 
replacement of Z. Chloromethylphosphonyl dichloride ( 112) provided a con¬ 
venient starting material for these studies. It may be mentioned here that 
although the end result of the synthesis of the small-membered ring by me¬ 
thod 4 was not achieved during the course of this research, a number of use¬ 
ful substrates were prepared and several unprecedented reactions were inves- 












































131 


tigated. 

Nucleophilic substitutions at the phosphoryl center are quite well- 
known . Treatment of ( 112) with phenyllithium resulted in the recovery 

of only polymeric material: 

0 0 
il . II 

C1 2 P-CH 2 C1 + 2 PhLi Ph 2 P-CH 2 Cl 

A more detailed examination of the compound (112) reveals that there is more 
than one possible site for attack. For example, the following reactions are 
conceivable: 

i) Nucleophilic substitution at phosphorus, 

ii) Nucleophilic attack at carbon, 

iii) Removal of the proton “-to phosphorus and subsequent reactions 
of the resulting carbanion, 

iv) Halogen metal exchange and subsequent reactions of the carbarion. 

These reactions are illustrated below: 

0 0 
ii PhLi 

C1 2 PCH 2 C1 Ph 2 PCH ? Cl i) 

0 

II 

Ph 2 PCH 2 Ph ii) 

0 

II <© 

Ph 2 PCHCl- iii) 

0 

II © 

Ph 2 PCH 2 - iv) 

In order to avoid the side reactions, the phosphorus compound (113) in 
its lower oxidation state, e.g., the phosphine was selected (*). Chlorome- 


(*) The acidity of the “-protons as well as the susceptibility of phosphorus 
nucleophilic attack is considerably reduced in phosphines compared to 
the phosphine oxide. 










132 


thylphosphonous dichloride ( 113) was successfully converted to diphenylchlo- 
romethylphosphine ( 114) in 60-65% yield by treatment with phenyllithium. 
Although the phosphine ( 114) could not be purified (due to its decomposi¬ 
tion in the absence of solvent), it was readily converted to the correspond¬ 
ing phosphine oxide ( 115) , phosphine sulfide ( 116) and the methiodide salt 
(117): 


(0) Ph 2 P(0)CH 2 Cl 


C1„PCH„C1 PhL -~ > Ph„PCH„Cl Ph 2 P(S) CH 2 C1 


'2 2 


(113) 


2 2 


(114) 


CH 3 I Ph 2 ?-CH 2 C1 ■ 


CH, 


(115) 


(116) 


(117) 


Similar results were obtained on treatment of ( 113) with phenylmagnesium 
bromide. 

The evidence that the “-chloromethyl group in the phosphine ( 113) con¬ 
stitutes a potential site of attack by a nucleophile (see pl38) came from 
the isolation of a small amount of the vinyl compound ( 118) formed from the 
reaction of ( 113) with an excess of phenyllithium followed by oxidation of 
the crude product with hydrogen peroxide. Treatment of chloromethyldiphenyl- 
phosphine ( 114) with phenyllithium followed by oxidation also gave the same 
compound in low yields. The formation of compound ( 118) may be rationalized 
by the following mechanism: 





































133 


Ph 2 P-CH 2 Cl —k 1 > Ph 2 P-CHCl 


Ph 2 PCH 2 Cl 


* Ph 2 P-CH 2 -CHClP-Ph 2 


PhLi 


-ft,Cl' 


Ph 2 P-CH=CH-PPh 2 

1 (0) 

Ph 2 P(0)-CH=CH-P(0)Ph 2 

(118) 

Examination of the conformations of the intermediate phosphine reveals that 
a predominant formation of the trans isomer should occur, as was found. This 
is because the intermediate ( 119) trans isomer would involve less steric 
interaction of the bulky substituents than (120) which would lead to the 


cis isomer: 



Cl 



A carbene mechanism could also rationalize the results. 

During the synthesis of “-chloromethyl phosphorus compounds, a one 
step method for the preparation of bis-(chloromethyl) phenylphosphine 
oxide (121) was achieved in 50% yield using phenylmagnesium bromide: 

0 0 


cip(ch 2 ci) 2 

( 122 ) 


PhM Br 11 

g , PhP(CH 2 Cl) 2 


( 121 ) 



























































134 


The requisite phosphinyl chloride ( 122) was conveniently prepared in 56% 
yield by the chlorination of the commercially available bis-(hydroxymethyl) 
phosphinic acid ( 123) with phosphorus pentachloride: 


0 

II 

HOP(CH 2 OH) 
( 125) 



0 

II 

C1P(CH 2 C1) 2 

( 122 ) 


In view of the requirement of a good leaving group, bis-(iodomethyl) 
phenylphosphine oxide ( 124) was prepared. The iodination was achieved under 
milder conditions than described by Yagupolski These workers treated 

( 121) with sodium iodide in refluxing ethylene glycol for 6 hr.: 


0 

II 

PhP(CH Cl) 


0 

Nal 'I 

2 v, PhP(CH 2 I) 

Acetone 


2 


( 121 ) 


(124) 


The possibility of nucleophilic substitution at the^-carbon site in these 
compounds, as shown above, was the basis of the attempted reaction of com¬ 
pound ( 121) with azide,thiophenoxy and of ( 124) with cyanidetions. The 
starting material was recovered in each case. An attempt to prepare a four- 

membered ring containing both sulfur and phosphorus, patterned after a pro- 

117 

cedure by Newmann , did not yield any cyclic product: 

0 0 

II s II 

PhP(CH 2 I) 2 > PhP<> S 

Attempted cyclization with disodium EDTA as well as FeCl^/^Hj-MgBr 
also failed to yield any cyclic product; only the reduction dimethylphenyl- 
phosphine oxide ( 125) was isolated: 


















































135 


0 


FeC1 3 


0 



ll 


II 


PhP (CH 2 C1) 2 


C 2 H 5 MgBr 


PhP(CH 3 ) 2 + 


(125) 



Na 2 EDTA 


* PhP(CH 3 ) 2 


0 


II 


0 

+ No PhP<] 


etc 


(125) 


Some of the above results are quite surprising in view of the know nu- 


120 

oleophilic substitution in similar compounds . However, Hojo and Yo- 
121 

shida have studied the reactivity of ^-chloromethyl sulfoxide (126) and 


sulfone ( 127) . The rate of displacement of chloride by iodide in these 
compounds was compared to that of n-butyl chloride. It was observed that 
although “-chloroacetophenone reacted 10^ times faster than n-butyl chloride, 
“-chloromethyl sulfoxide ( 126) and sulfone ( 127) reacted one-hundred times 
slower than n-butyl chloride. 

The ^-chloromethyldiphenylphosphine oxide ( 115) has also been found to 

exhibit a similar lack of reactivity towards nucleophilic displacement of 

chloride by iodide. The oxide ( 115) was found to react 30-40 times slower 

than n-butyl chloride (Table 1). The difference in the reactivity of «- 

chloromethyl carbonyl, sulfonyl, sulfinyl and phosphoryl compounds cannot 

be rationalized on the electronic effects since the latter are very similar 

121 

in these compounds . The large difference between “-chlorocarbonyl com¬ 
pounds and the other heterocompounds can be rationalized on the basis of a 

87 

steric effect, commonly known as a " neo -pentyl effect" . Thus, all the 

substituent groups (e.g., carbonyl, phosphoryl,sulfinyl and sulfonyl), due 

to their electron withdrawing nature, facilitate the attack of the nucleo- 

2 

phile at the <*-carbon atom. However, the carbonyl compound with sp geometry 



















136 


provides less steric hindrance to the attacking reagent than sulfones and 
phosphineo/oxides, since they have tetrahedral geometry. As a result of 
this geometry, the latter are very similar to neo- pentyl chloride. This 
leads to steric blocking of the attacking reagent as illustrated below: 


Table 111-1. Relative Rates of Nucleophilic Substitution in Chloro- 
methyl Derivatives. 


Compound Relative Rate 

50° 80° 


n-Butyl Chloride 

Ph-C0-CH 2 C1 

Ph-S0-CH 2 C1 

Ph-S0 2 -CH 2 C1 

Ph 2 -P0-CH 2 C1 


1 1 
<1.05 x 10 5 
< 0.01 
< 0.02 


<0.03 



0 

II 

Ph — P— Ph 

I 

X-CH ---N 


Ph 

I 

0 - S - 0 

X - - - CH- 2 - - N 


This, however, is not the complete explanation since, the lack of re- 

121 

activity of the sulfoxides cannot be rationalized on the same grounds 

In the course of this investigation, an interesting reaction was ob¬ 
served during an attempt to oxidize the chloromethylphosphine ( 114) to the 
corresponding oxide ( 115) . The reaction was thought to be sufficiently un¬ 
usual as to merit a more detailed study. Thus, treatment of (114) with hy¬ 


drogen peroxide in acetone yielded the chloromethylphosphine oxide ( 115) : 




































































137 


Ph 2 PCH 2 Cl 



Ph 2 P(0)CH 2 Cl 


Acetone 


(114) 


(115) 


On the other hand, oxidation in glacial acetic acid, followed by evapora¬ 
tion of the solvent, treatment with water and extraction of the product 
with benzene afforded diphenylmethylphosphine oxide ( 128 ): 


0 


CH COOH 



Ph_PCH„Cl 



(114) 


(128) 


On further examination, it was found that hydrogen peroxide was not involved 
as a reactant since the same product ( 128) was obtained by heating ( 114) 
with glacial acetic acid at 80° followed by the previous work-up procedure. 
The replacement of the chlorine atom by hydrogen was suggestive of some 
type of a hydride transfer mechanism. Based on this mechanistic assumption, 
the phosphine ( 114) was treated with acetic acid-d^ followed by work-up H 2 0 
to obtain a nearly quantitative incorporation of one atom of deuterium in 
the oxide ( 128) . A control experiment (see Table 2) with acetic acid/D 2 0 
and acetic acid/H 2 0 showed that the deuterium atom transferred during the 
process was the one from the acetic acid-d^: 


0 


II 


Ph 2 PCH 2 Cl 


+ d 3 ccood 


A Ph 2 P-CH 2 -D 


(*) The purpose of the oxidation was to obtain the corresponding oxide which 

was to be used in the model study. 



















138 


The course of deuterium incorporation was conveniently followed by nmr 
integration. 

Related to this observation is the important work of Kabachnik, Grif¬ 
fin and Goldwhite. Most of their research involved a similar displacement 
of chloride, but under basic reaction conditions. It would be pertinent to 

review some of these studies related to results here. 

122 

Kabachnik and co-workers treated tris-(chloromethyl) phosphine 

( 129) with sodium ethoxide and obtained bis-(ethoxymethyl)methylphosphine 
oxide ( 130) instead of the expected tris-(ethoxymethyl)phosphine ( 131) : 


0 


( 129) 
122 



(130) 


(131) 


Kabachnik proposed that a "pseudoallylic" rearrangement was responsible 
for the reaction, since there is an analogy between the following two elec¬ 
tronic structures: 


1 

Cl— CH - CH=C 
\ 

(132) 


Z 7 - 

Cl—CH —P — 


The pseudoallylic rearrangement may be visualized to occur as follows: 

<4\ I EtOH * 

C1CH 0 P: + OEt —* CH =P-0Et ->CH„— P= 0 + (Et) o 0 

2, 2 , 3 , 2 

123 

Griffin and co-workers showed that in thh presence of base, chlo- 
romethylphosphinic acid ( 133) rearranged to methylphosphonic acid ( 134) : 

















139 


ci-ch 2 


0 

II 

-P-0 

I 

H 


OH 




(133) 



0 

OH 


However, contrary to the pseudoallylic rearrangement proposed by Kabachnik 
122 

, Griffin showed by isotopic labelling, that the predominant reaction 

124 

took the following course : 


*C1- CH 


0 
II 

o p -0 

2 i\e 

-H OH 


* 



0 

It 

-P-0 

I 

OH 


Thus, most of the deuterium in the starting phosphinic acid ( 133) was 

found in the methyl group of the product, conclusively demonstrating that 

the hydride transfer occurred through an intramolecular process. 

A similar base-induced rearrangement of chloromethylphosphine ( 135) 

to methylphosphinic acid (136) has been reported by Goldwhite and co-workers 
125. 

0 

OH II ^ 

G1CH_PH„ —— ► CILPC 


On the basis of the analogy to reactions of polyfluorophosphonous dichloride, 
Goldwhite proposed a phosphaalkene ( 137) as an intermediate: 


ci-ch 2 ph 2 


OH ^ 

-HC1 


ch 2 =ph 



OH 

I 

ch 3 -p-h 

(137) 


H O j? OH 

- £ —> ch -p; 

H 


This analogy may not be exactly parallel since nucleophilic attack on tri- 
valent phosphines is known to occur and because the hydride migration shown 
below, would essentially give the same results: 


Cl-C 


4|-h 

OH 


H 



0 

It / 

CH 3 P x 


OH 

H 


> ch 3 -p-oh 


















140 


However, no isotopic labelling experiments were performed to determine 
whether the mechanism involved an intramolecular hydride transfer or a 
phosphaalkene intermediate. 

During the rearrangement of ( 114) , some experimental difficulty was 
encountered because of its instability, especially in the absence of sol¬ 
vent. A more satisfactory compound was selected for carrying out quantita¬ 
tive work, namely tris-(chloromethyl)phosphine ( 129) which could be distil¬ 
led under vacuum with slight decomposition. Similar to ( 114) , the phosphine 
( 129) also led to incorporation of approximately one atom of deuterium in 
the oxide product ( 138) on treatment with acetic acid-d^: 

0 

II 

(C1CH ) P + D CC00D -> D-CH -P(CH Cl) (138) 


Approximately equivalent amounts of phosphine ( 129) and glacial acetic acid 
were sealed in an nmr tube under vacuum and heated for 27 hr. at 131°. The 
products were found to be acetyl chloride (91.6 mole %), acetic anhydride 
(8.4 mole %) and bis-(chloromethyl)methylphosphine oxide ( 138) : 


p(ch 2 ci) 3 + ch 3 cooh —% ch 3 cooci 

91.6% 


+(ch 3 co) 2 o 

8.4% 


0 

II 

+ CH 3 P(CH 2 C1) 2 
(138) 


This led to the following mechanism: 


Cl-CH 0 -P- 
1 I 



ci-ch 2 



CH 3 -P=0 + CH 3 C0C1 





CH 3 C00 


( 139 ) 




















141 


The production of the small amount of acetic anhydride may be rationalized 
by the attack of acetic acid or its anion on the pentacovalent intermediate 
(139): 


H 

'©-CH 0 -P< - 

2 A 

(139) , 0 - C -CH. 

CH 3 8 ‘ 


* 


0 

II 

CH„ —■ P — + (CH„CO) _0 

O j o i 


The above evidence, of course, does not rule out the possibility of a 
pseudoallylic mechanism illustrated below: 


l 0 

*C1~CH 2 -P: + OC-CH 3 


i ° 

CH =P-0-C-CH T 
l | 3 

+ Cl" 


CH 3 COOH 


ch 3-p=° ♦ ^ch 3 Jo) 

CH 3 C0C1 +(CH 3 C0) 2 0 


However, one would predict that the formation of acetic anhydride in 

the pseudoallylic mechanism, would be significant because in the early 

stages of the reaction, the concentration of chloride ions would be much 

less than that of acetic acid. In any event, definite proof for determining 

which mechanism operates may possibly be obtained by carrying out the ex- 

37 

periment in the presence of external Cl ions. The pseudoallylic mecha- 

37 

nism should lead to almost 50% incorporation of Cl in the acetyl chloride, 

37 

whereas the proposed mechanism requires no incorporation of Cl since the 
formation of acetyl chloride is intramolecular. This assumes lack of ex¬ 
change of chloride bn with acetyl chloride. 







































142 


3. Experimental . 

Synthesis of Chloromethyldiphenylphosphine(114) . To a stirred solu¬ 
tion of 15.2g (0.1 mole) of chloromethylphosphonous dichloride ( 113 ) in 
200 ml petroleum ether (30-60°) was added 90 ml of 2.58M phenyllithium 
(2.2 equivalent) in 3:1 benzene ether over 115 min. at -65° under a nitro¬ 
gen atmosphere. A pale yellow precipitate slowly accumulated as the addi¬ 
tion progressed. The mixture was stirred for an additional 40 min. and 
then warmed to -10-0° by removing the cold bath. The reaction was quenched 
with 100 ml of ice water while stirring and the mixture was allowed to stir 
for an additional 3 hr.. The two layers were separated, the aqueous layer 
was extracted with ether and the combined organic layers were washed with 
water and dried over anhydrous sodium sulfate. Evaporation of the solvent 
(see pJ.32 ) gave 31.6g mobile liquid which contained some solvent. The nmr 
of the crude liquid (benzene) showed peaks at: x=1.83-2.50 (10H, m, aromatic), 
t= 5.95 (2H, d, 2 J pcH =6.0 cps). 

Preparation of Chloromethyldiphenylphosphine Oxide ( 115) . The crude li¬ 
quid obtained above was dissolved in 60 ml acetone and oxidized by adding 
16 ml of hydrogen peroxide (30%) with stirring at 10°. The acetone was ev¬ 
aporated and the residue was dissolved in 30 ml benzene. The two layers 
were separated and the organic layer was dried over anhydrous sodium sulfate. 
Evaporation of benzene gave 22.8g of a viscous liquid wiich solidified on 
scratching. Recrystallization from methanol-water followed by sublimation 
at 120° (0.2 mm) gave 20g of a crystalline solid m.p. 130° forms murky li¬ 
quid which clears at 136-137° (136-137°) (114) (50%). The nmr (DCC1 ) showed 

2 

peaks at: t= 1.90-2.80 (10H, m, aromatic), x=5.95 (2H, d, J pCH =6.3 cps). 

























































143 


Treatment of (114) with Sulfur . To the crude phosphine prepared from 
15.2g (0.1 mole) of ( 113) and 90 ml phenyllithium as outlined on p. 142 , 
was added 6.4g (0.2g atom) sulfur and 160 ml of benzene and the resulting 
mixture was allowed to reflux for 3 hr. under a nitrogen atmosphere. The 
mixture was cooled, filtered, and the solvent was evaporated to obtain 
25.6g of semi-solid. The product was extracted with hot cyclohexane and 
recrystallized by concentration of the solvent to give 5.6g of yellow cry¬ 
stals, m.p., 76-78° (21.6%. The nmr (DCCl^) showed peaks at t=1.90-280 

2 

(10H, m, aromatic), x=5.85 (2H, d Jpcn = 6*0 cps). Anal . Calcd. for 

C 13 H 12 C1PS : C ’ 58 * 55 ; H > 4 * 54 ; P > n - 62 ' Found: 58.34; H, 4.47; P, 11.82. 

Quaternization of Phosphine (114) with Methyl Iodide . To a solution 
of the crude phosphine ( 114) (obtained from 22.7g ( 113) and phenyllithium) 
in 250 ml anhydrous ether, was added 6-10 fold excess of methyl iodide and 
the mixture was stored in the dark for 48 hr.. The precipitate was filtered, 
washed with ether and then recrystallized from methanol-ether (using Norit) 
to obtain 48.4g crystalline solid ( 117) m.p. 203-205° (with decomposition) 
Yield 67.5%. The nmr (CF^COOH) showed peaks at: t= 1.20-1.86 (10H, m, aro¬ 
matic), t= 4.74 (2H, d, • 0 C P S ) > t= 7.06 (3H, d, 2 Jp CH =15.0 cps). 

Anal . Calcd. for C^H^Cl : C, 44.65; H, 4.01; I, 33.69; P, 8.22. Found: 

C, 44.80; H, 4.11; I, 33.72; P, 8.39. 

Reaction of (115) with Phenylmagnesium Bromide . A solution of phenyl- 
magnesium bromide was prepared from 23.5g (0.16 mole) of bromobenzene and 
4.0g (0.165g atom) of magnesium turnings in 150 ml of ether. 

To a stirred solution of 3.8g of chloromethylphosphonous dichloride 
(113) in 100 ml of petroleum ether (30-60°) at -65° was added the above 


Grignard reagent over 2 hr. under an atmosphere of nitrogen. The mixture 




















































144 


was stirred for an additional 10 min. and quenched with 50 ml of distilled 
water. Gradual warming of the mixture to room temperature gave a thick 
emulsion which was cleared by adding a saturated solution of ammonium chlo¬ 
ride and an excess of ether. The two layers were separated, the aqueous 
layer extracted with ether, and the combined organic layers were washed 
with water and dried over anhydrous sodium sulfate. Evaporation of the 
solvent gave a semi-viscous liquid which was dissolved in ether and treated 
with an excess of methyliodide. The mixture was stored in the dark for 
48 hr.. The precipitate was filtered, washed with ether and recrystallized 
from methanol-ether to give 6.3g (67%) of crystalline solid ( 117) m.p. 203- 
206°. A mixture m.p. with the authentic compound did not show any depression. 
The nmr was identical to that of an authentic compound ( 117) . 

Preparation of Bis-(Chloromethyl)phosphinyl Chloride (122) . To a vi¬ 
gorously stirred suspension of 625g (3.0 mole) of phosphorus pentachloride 
in 1500 ml of chloroform in a three-necked flask was added at 15°C 126g 
(1.0 mole) of bis-(hydroxymethyl)phosphinic acid ( 123) over 35 hr.. The 
acid in the addition funnel was occassionally warmed to prevent crystalli¬ 
zation. The resulting clear solution was stirred for an additional 7 hr.. 
Evaporation under vacuum followed by distillation of the liquid gave 102.4g 

(56.0%) of a colorless liquid b.p. 75-80 (0.3-0.4 mm) (98-101/1 mm) ( 126) . 

2 

The nmr (CCl^) showed peaks at: t=5.83 ( d, ^PCH = ^‘^ C P S )• 

Treatment of (122) with Phenylmagnesium Bromide, Preparation of Bis- 

(Chloromethyl)phenylphosphine Oxide (121) . Phenylmagnesium bromide was pre¬ 
pared from 47.lg (0.3 mole) of bromobenzene and 7.2g (0.3g atom) of magnesium 
turnings in 175 ml anhydrous ether. 

The above Grignard reagent was added to a solution of 36.3g (0.2 mole) 
of bis-(chloromethyl)phosphinyl chloride ( 122) in 200 ml of ether at -65° 













145 


with stirring under an atmosphere of nitrogen over 75 min.. A white preci¬ 
pitate formed which made the mixture difficult to stir. A 100 ml portion 
of ether was added and the mixture was allowed to stir for an additional 

1 hr.. The reaction was quenched with 150 ml of water and then the mixture 
was allowed to warm up to room temperature with stirring (to break the lumps 
of the solid). 

Filtration and recrystallization from methanol (using Norit) gave 

22.8g (50%) of needles, m.p. 142-143° (141-142°)(116). The nmr (DCC1 ) 

2 

showed peaks at: x=1.80-2.60 (5H, m, aromatic), t= 6.00 (4H, d, Jp£ H = 

7.0 cps). 

Preparation of Bis-(Iodomethyl)phenylphosphine Oxide (124) . To a hot 
solution of l.lg (0.005 mole) of the oxide ( 121 ) in 25 ml of acetone was 
added 5.8g (0.04 mole) of sodium iodide in one portion and the mixture was 
allowed to reflux for 24 hr.. A white precipitate appeared. After cooling, 
the mixture was poured over lOOg of ice. The precipitate was filtered, 
washed with cold water, and dried in air to give 1.5g (75%) of a pale yel¬ 
low crystalline solid, m.p. 169-170°. Recrystallization from benzene-pet¬ 
roleum ether (30-60°) raised the m.p. 171-173° (172-173°) ( 116) . The nmr 

(C F^COOH) showed peaks at: t= 2.10-2.90 (5H, m, aromatic), x=6.54 (4H, d, 

2 

J p CH =6 * 5 C P S )' 

Attempted Reaction of (121) with Sodium Azide . A mixture of 1.12g 
(0.005 mole) of oxide ( 121) , l.Og of sodium azide and 10 ml of DMF was 
heated to reflux for 24 hr.. After cooling, the mixture was poured over lOOg 
of ice. The precipitate was filtered to give l.Og of the starting material. 
(Recovery 89%). 















146 


Attempted Reaction of (124) with Sodium Cyanide . A solution of 4.06g 
(0.01 mole) of ( 124) in hot methanol was added to a solution of l.Og (0.02 
mole) of sodium cyanide in 6 ml of water and the mixture was heated to re¬ 
flux for 12 hr., cooled and filtered to obtain dark brown filtrate. Eva¬ 
poration of the solvent gave a dark brown semi-dolid. The residue was ex¬ 
tracted with chloroform, and the combined organic layers dried over anhy¬ 
drous sodium sulfate and evaporated to yield 2.4g (60%) starting material, 
characterized by its ir and nmr. 

Attempted Reaction of (124) with Sodium Sulfide . A sample of sodium 

117 

sulfide was recrystallized and dried by the procedure of Newman 

To 2.1g (0.052 mole) Of ( 124) in 30 ml of ethyl cellosolve was added 
to a solution of 1.25g of anhydrous sodium sulfide in 18 ml ethylene glycol 

over 14 min.. The mixture was allowed to stir for 70 hr. and the solvent 

was removed under vacuum (90° at 0.05 mm). The residue was dissolved in 
chloroform, the organic layer was washed with water and dried over anhydrous 
sodium sulfate. Evaporation of the solvent gave a polymeric material that 
could not be characterized. 

Attempted Reaction of (121) with Sodium Thiophenolate . A solution of 
sodium thiophenolate was prepared by treating 2.2g (0.02 mole) of thiophenol 
with a solution of 0.46 (0.02g atom) of sodium in 40 ml of ethanol followed 
by refluxing the mixture for one half hr. 

To the above solution was added 1.12g (0.005 mole) of the oxide ( 121) 
in portions. The mixture was refluxed for 3 hr; a precipitate formed during 

this period. The solvent was evaporated and the residue extracted with car¬ 

bon disulfide and evaporated, to produce a resinous mass which could not 


be characterized. 

















































147 


Attempted Cyclization (121) with FeCl^/C^H^MgBr. A solution of ethyl- 
magnesium bromide was prepared from 32.5 g (0.3 mole) of ethyl bromide and 
7.5g (0.315g atom) of magnesium turnings in 200 ml ether. 

To a stirred solution of 11.2g (0.05 mole) of bis-(chloromethyl)phenyl- 
phosphine oxide ( 121) in 100 ml of ether in a four-necked round bottomed 
flask equipped with a water condenser, were added simultaneously, the 
above Grignard reagent and a solution of 0.7g of anhydrous ferric chloride 
in 50 ml of ether over 75 min. A dark brown metal deposited on the side of 
the flask. The exothermic reaction was stirred overnight and then quenched 
by adding a slurry of 200 ml of 2M HC1 saturated with ammonium chloride and 
ice, until two clear layers were obtained. The ether layer was separated, 
washed with water dried and evaporated; no residue was obtained. The aque¬ 
ous layer was extracted with methylene chloride and the combined organic 
layers were dried over anhydrous sodium sulfate and evaporated to give 

7.0g (91%) of pale brown solid. Recrystallization from cyclhexane gave 

128 

6.0g crystalline solid m.p. 102-103 clearing at 105-107° (110°) . The 

nmr (DCCl^) showed peaks at: x=1.90-2.70 (5H, m, aromatic), t=8.30 (6H, 

2 

d, Jp ( _ ;H =13.0 cps) . The compound was identical to dimethylphenylphosphine 
oxide ( 125) . 

Attempted Cyclization of Bis-(Iodomethyl)phenylphosphine Oxide (124) . 

A mixture of 4.04g (0.01 mole) of (124), 1.3g (0.0435 mole) of disodium 

EDTA, 3.45g (0.086 mole) sodium hydroxide, in 6 ml of water, 0.28g of so¬ 
dium iodide, 2.56g (0.039g atom) of zinc powder and 20 ml of 95% ethanol 
was heated to reflux for 1 hr. in an oil bath at 120°. The mixture was 
cooled, 150 ml of 95% ethanol added. The reaction was then heated to reflux 
and the mixture was filtered hot. The solvent was evaporated, water was 
added to the residue followed by extraction with chloroform. 























148 


The chloroform layer was dried over sodium sulfate, and evaporated 

to give 2.2g of liquid that solidified on cooling. Recrystallization from 

cyclohexane gave 1.4g (94%) of white needles, m.p. 100-102 (murky liquid 

128 

clearing at 105-107°) (100°) . The nmr of the compound was identical 

to that of an authentic sample of (125). 

Preparation of txans-1,2-Vinylenebis(diphenylphosphine) Dioxide (118) . 

The crude phosphine ( 114) was prepared from 7.6g (0.05 mole) of ( 115) and 
50 ml of 2.58 M phenyllithium as described earlier and aas evaporated to 
16.6g of liquid. The above liquid was dissolved in 25 ml of ether and 
treated with 25 ml of 2.58 M phenyllithium (3:1, benzene:ether) at -30° 
over 50 min.. The mxiture was allowed to warm up to 15° and quenched with 
25 ml of water. The two layers were separated, the aqueous layer was ex¬ 
tracted with benzene and the combined organic layers were dried over anhy¬ 
drous sodium sulfate. Evaporation of the solvent gave 13.0g of the crude 
phosphine which was dissolved in 25 ml of acetone and oxidized with 30% hy¬ 
drogen peroxide solution until a positive starch-iodide test was obtained. 

The solvent was evaporated and the residue was treated with 100 ml of ben¬ 
zene when an insoluble solid precipitated. Filtration gave l.lg (5.1%) of 

product. Recrystallization from 95% ethanol gave 0.62g of a crystalline 

q 127 

solid m.p. mmp. 306-309° (306-309) . The nmr (CF^COOH) identical to that 

127 

of an authentic compound (Provided by A. Aquiar of Tulane Univ. ). 

The benzene layer was dried and evaporated to give 11.lg of ( 115) char¬ 
acterized by mmp. with an authentic sample. 

Alkaline Decomposition of Chloromethylmethyldiphenylphosphonium Iodide 

(117) . A mixture of 4.0g (0.01 mole) of powdered ( 117) , 20 ml of water, 30 ml 


of benzene and 5 ml of methyl iodide was stirred under an atmosphere of 










































































149 


nitrogen. To this mixture was added 15 ml of 5% sodium hydroxide solution 

and the stirring was continued for 48 hr.. A precipitate was formed which 

was filtered and washed with a small amount of benzene to yield 0.65g (19.CU) 

of a solid which was recrystallized from acetonitrile-ethyl acetate to give 

a crystalline solid, m.p. 248.5^249.5° (241°) (mmp)The nmr (F^CCOOH) 

2 

showed peaks at: x = 1.90-2.46 (10H, m, aromatic), t=7.40 (6H, d, 

14.0 cps). The sample was identical to an authentic sample of dimethyldi- 

129 

phenylphosphonium iodide 

The layers in the filtrate were separated, the layer was extracted with 
benzene and the combined benzene layers were washed with water and dried 
over anhydrous sodium sulfate. Evaporation of the solvent and recrystalla- 
tion of the residue from cyclohexane gave 1.55g (67.5%) of white needles, 
m.p. 109-110.5° (109-111°) 13 °. The nmr (DCC1 3 ) showed peaks at: x=2,00- 
2.75 (10H, m, aromatic), x=8.00 (3H, d, 2 J pCH =13.3 cps.). The compound 
was characterized as diphenylmethylphosphine oxide ( 128) by mmp. 

Preparation of Diphenylmethylphosphine Oxide (128), Reaction of (114) 
with Acetic Acid . To the crude phosphine ( 114) containing benzene, prepared 
from 15.2g of ( 113) and phenyllithium as shown earlier (p. 142 ) was added 
15 ml of glacial acetic acid and the mixture was heated under an atmosphere 
of nitrogen for 3 hr. at 60-80°. The solvent was evaporated in vacuo . The 
thick oily residue was treated with water and extracted with benzene severa! 
times; the combined benzene layers were washed with water and dried over an 
hydrous sodium sulfate. Evaporation gave 9.8g of a viscous liquid. The 
liquid was extracted with hot cyclohexane (Norit) and concentrated to give 
white needles m.p. 110-110°(109-111°) . The yield of the pure diphenyl¬ 

methylphosphine oxide ( 128) was 3.4g (47.7%). The nmr (DCCl^ was identica 




























































150 


to that of an authentic sample. 

Control Experiments on Deuterium Exchange . Approximately 2.0g of an 
aliquot of the phosphine ( 114) (prepared as on p.142) was treated with an 
equivalent amount of acetic acid-d^ and the mixture was heated at 60-80° 
for 3 hr.. The volatile material (solvent, acetyl chloride, ect.) was eva¬ 
porated and the residue was treated with a slight excess of D^O and then 

allowed to stand for 10 min. The resulting suspension was processed as 
on p. 149 # 

In an analogous manner, a 2.0g aloquit was treated with acetic acid 
and then quenched with D2O. The resulting mixture was processed as on 
P-149 . The results of the deuterium exchange reactions are summarized 
in table III-2. 

Tris-(chloromethyl)phosphine (129) was prepared by the method of 

124 

Hoffman_by an alkaline decomposition of tetrakis-(chloromethyl) phos- 

phonium chloride. The latter was synthesized as follows . To a suspension 
of 150g (0.72 mole) of phosphorus pentachloride in 400 ml carbon tetra- 
c loride was added 27.5g (0.14 mole) of tetrakis-(hydroxymethyl)phospho- 
nium chloride through a wide-mouth addition funnel with stirring over 5 
min.. The mixtu e was allowed to reflux for 7 hr.. On cooling, a cry¬ 
stalline solid precipitated. The mixture was diluted with an equal volume 
of ethyl acetate, and filtered and recrystallized from methanol-ethyl 

acetate to give 28.5g (75%) of a crystalline solid, m.p. 202-203° (192- 

131 ? 

193°) . The nmr (CF^COOH) showed peaks at: t= 5.26 (8H, d, Jp CH _ 

5.7 cps). 

Treatment of tris-(Chloromethyl)phosphine (129) with Acetic Acid , 
Preparation of (L38) . To 3.06g (0.05 mole) of glacial acetic acid in a 
round bottomed flask was added 4.5g (0.025 mole) of the phosphine (129), 






































151 


Table III-2. Results of the NMR Integrations on the Product of the 
Reactions. 


1) Ph_P-CH 0 C1 + CH COOH 


2) Ph P — CH_C1 + CD COOD -> 

d 2 o 

3) Ph P—CH Cl + CH COOH - > 

d 2 o 


Reaction Integration Values 
in mm. 

Ar Aik 


Proton Ratio 
Ar Aik 


Average(*) 


% 


error 



179.75 

53.35 

10 

2.985 



193.3 

54.75 

10 

2.815 


1 

178.55 

52.65 

10 

2.945 

-2.77 


222.35 

63.80 

10 

2.84 



190.15 

39.05 

10 

2.00 



197.70 

42.00 

10 

2.14 


2 

211.70 

44.15 

10 

2.09 

+2.98 


215.50 

44.00 

10 

2.04 



392.90 

125.00 

10 

3.18 


3 

206.5 

66.05 

10 

3.18 

+6.17 


(*) % error was estimated as follows: 

P=(Integration value of Aromatic Protons)/10 
Q=(Integration Value of Alkyl Protons)/3(2) 


0 P-Q 

% error = —~ 


X100 





















152 


and the resulting mixture was heated to 90-100° for 6 hr under an atmos¬ 
phere of nitrogen. Acetic acid and the volatile products were removed 
under vacuum. The residue was treated with 10 ml of water and extracted 
with benzene. 

The benzene layer was dried and evaporated to give 1.5g (37%) of an 

oily material that solidified on standing, m.p. 41-43° (clears at 45-45°). 

122 

(49-50°) . The nmr (DCCl^) of ( 138) showed peaks at: t=6.11 (4H, d, 

2 2 

JpCH = 6* 5 cps), t=8.26 (3H, d, J pcH =13.2 cps). Similar results were ob¬ 

tained with acetic acid-d^. The amount of dueterium incorporation was es¬ 
timated to be 0.86+0.069 atom/molecule. The lack of complete corporation 
of one atom of deuterium in this compound may be due to the decomposition 
of ( 129) to give HC1. 

The Product Analysis of the Reaction Acetic Acid with tris-(chlorome- 

thyl)phosphine (129) . Equivalent amounts of ( 129) and glacial acetic acid 
were mixed and sealed at 131° for 27 hr.. The nmr was reexamined and each 
peak was integrated. The peaks were identified as acetyl chloride and 
acetic anhydride by adding small amounts of pure acetyl chloride and acetic 
anhydride to the opened nmr tube and watching the respective peaks increase 
in intensity. (Acetyl Chloride t=7.80. Acetic Anhydride t=8.11). 

A working curve was constructed by weighing known amounts of acetyl 
chloride and acetic anhydride and estimating the relative intensity of the 
peaks due to these compounds. The plot of Mole % composition by integra¬ 
tion vs Mole % of acetic anhydride is shown in Fig. 1. The amount of acetic 
anhydride in the reaction mixture was estimated to be 8.4 mole % on the 
basis of this plot. 

The rate of halofen exchange of the reaction: 

Ph 9 P(0) - CH Cl + 1 ~ Ph P(0) - CH I + Cl 
was determined by the method of Conant . A sample of 0.001 mole of chlo- 

romethyldiphenylphosphine oxide was weigned in an ampule. To this was 













% acetic anhydride in acetyl chloride by integratio 



g.HI-1. Working curve for the determination of concentratio 
of acetic anhydride in acetyl chloride. 



















154 


added 5 ml of 0.04 M potassium iodide solution acetone. The reactants 

were essentially unreactive at room temperature. The ampule was sealed 

and heated in a constant temperature bath at 80°. After a given length 

of time, the tube was removed from the bath and the contents of the tube 

were quantitatively transferred to a small separatory funnel containing a 

132 

mixture of ice and concentrated HC1 . The mixture was extracted with 
three 5 ml portions of chloroform. The aqueous solution was transferred 
to an Erlenmeyer fork, a 5 ml portion of choroform was added and the mix¬ 
ture was titrated against 0.00301 N KIO^ solution from the burette. The 
titration took about 10 min. and the end point was marked when the yellow 
color of the chloroform layer just disappeared. A blank titration was 
performed with 5 ml of 0.04 M potassium iodide in acetone. The rate con¬ 
stant for the reaction was calculated using the following formula by Con- 

132 
ant : 

i 1 i 5-z 
k " t x4b log S(l-z) 

where b=concentration of potassium iodide in moles per liter, z=the frac¬ 
tion of KI which has reacted in the time t measured in hr., the value of 
z is the difference in the actual and blank titrations divided by the blank 
titration. The table DU-3 represents the values of the rate constants ob¬ 
tained for (115). 

























155 


Table III-3. Kinetics of Halogen Exchange for the Reaction: 

PH 2 P(0)-CH 2 C1 + I~ -* Ph 2 P(0)-CH 2 I + Cl 

(115) 


At 80° 


# 

Time 

hr. 

Reading 

k (Calcd.) 1 

1 . m. hr. 



1 

193 

3.5 

ml 

2.89 x 

io “ 2 

2 

295 

2.0 

ml 

2.70 x 

io " 2 

3 

365 

1.1 

ml 

2.96 x 

io " 2 

Av. k 2 = 

2.85 x 10“ 2 1. m" 1 . 

hr’ 1 

• 



n-BuCl 

k 2 


50° 

0.0415 

l.m .hr 


60° 0.117 

80° 


0.944 (extrapolated) 



















CHAPTER IV 


NUCLEAR MAGNETIC RESONANCE STUDIES ON 
PHOSPHETANE DERIVATIONS 

1. Introduction . 

The development of organic chemistry in the last two decades is marked 

by the deluge of nmr data in the literature 133,134. Although, most of 

1 19 31 13 

these data concern H and F systems; both P and C (each with spin = 

135 136 

1/2) have received a considerable attention * . The chemical shifts 

and coupling constants provide a useful information of the various struc- 

1 19 

tural features and stereochemistry of the molecules. The H and F nu¬ 
clei are monovalent and hence do not possess an intrinsic stereochemistry. 
31 13 

The P and C nuclei on the other hand, because they are polyvalent, 
have stereochemistry due to substituents around them. Thus, nmr spectros¬ 
copy of molecules containing these nuclei provides an additional source 

of information. 

31 

P magnetic resonance spectroscopy has made a significant contrihu- 

137 

tion to the early development of nmr spectroscopy . This was primarily 

31 

due to the fact that the isotope P occurs with 100% natural abudance, 

which allowed for measurements, even with instruments of low resolution. 

As a result, the early development of nmr theory was not seriously hamper- 

137 

ed due to the inavailability of instruments of high resolution 

31 

The routine use of a typical P instrument is limited due to the low 

sensitivity of the phosphorus nucleus to nmr experiments, ca. 6.6% of that 

138 

for proton at the same field . This requires a high concentration of 
the samples. Such limitations have led to the development of techniques 


156 






1 




















157 


by which chemical shifts and coupling constants for these nuclei can be 
calculated. The technique is commonly known as "nuclear Magnetic double 
resonance" or" heteronuclear spin decoupling" . 

The technique mentioned above has had wide applications in the analy¬ 
sis of an nmr spectrum. The basic principle is that the mnltiplet struc¬ 
ture arising from the coupling of certain nuclei can be eliminated by de¬ 
coupling the spins of these nuclei from the remainder of the spin system 
140 

. As a result, a simplified spectrum is obtained. The technique of 

double resonance, therefore, is a valuable tool for the determination of 

139 141 

coupling constants and theri relative signs * 

The method of double resonance has also been utilized to obtain chemi¬ 


cal shifts of various nuclei, e.g., C 


13„ 141 3U 142 


, 15.. 143 _ 

and N . The 


method was first employed by Royden 
quency of ^CH^I. 


141 


for determining the resonance fre- 


2. Results and Discussion . 

The advantage of the above technique can be applied to the organo- 

31 

phosphorus compounds in this investigation. Since P chemical shifts can 
be obtained while observing the corresponding nmr spectrum, concentrated 
solutions are unnecessary. Further, most of the compounds in this inves¬ 
tigation have moderately resolved nmr spectra and are readily interpreta- 

K1 11 
ble 

31 

The P chemical shift of all the compounds were obtained by the me- 
144 

thod of Stothers . The chemical shifts of some phosphetanes relative 
to 85% phosphoric acid are shown in table 111-1. 

Attempts have been made to correlate phosphorus chemical shifts with 
different parameters. Correlations based on electron donating or with¬ 
drawing effects of the substituents at phosphorus do not seem to exist in 































158 


137 

the phosphorus compounds . Attempts have also been made to correlate 

31 1 *7 n 

P chemical, shifts with Taft values without success . It may also 

be mentioned here that Taft correlations on the basicity of the phosphe- 

2 

tanes have also been unfruitful 


The available evidence indicates that 


the chemical shifts of phosphorus nuclei depend on at least two parameters, 

namely; i) the ionic character of the bond, and ii) hyperconjugation of 
145 

the protons . The failure of Hammett and Taft correlations in these 

systems can be readily understood on the basis of the above variables. In 

any event, some correlations have been possible and have led to several 

generalizations 145,146^ Empirical equations for calculating chemical 

shifts in tertiary and secondary phosphines are shown below: 

3 

Appm = 62 - Z 0n P for tertiary phosphines where 0 P = empirically 
n=l determined group contribution 


APPm = 99 - Z 0 p for secondary phosphines n=number of substitu- 
n=l ents 


Good correlations are obtained by ploting the calculated vs. observed chemi¬ 
cal shifts 

Grim and co-workers re ported a new parameter in which the effect 

of hyperconjugation was an important factor. The group contributions were 

calculated from the following relation: 

Cr = 21 - 148 + 3T 

c c 

where 8 and T are the number of 8 and Tcarbon atoms of a substituent, on 
c c 

phosphorus. An examination of the equation reveals that substituents of 
a <*-hydrogen by a methyl group leads to a significant negative contribu¬ 
tion to the chemical shift, i.e., leads to deshielding of the phosphorus 
nucleus. In the following equation: 


















































































































159 


r 2 p - cr 2 h r 2 p = cr 2 H + 

progressive substitution of hydrogens reduces the hyperconjugative shield¬ 
ing of the nucleus. 

31 

P chemical shifts of the several phosphetanes are shown in table 1. 

The calculated values based on Grim's group contributions are also shown 

where applicable. The plot of the calculated vs. observed chemical shifts 

fails to show adequate correlation (Fig. 1). 

The effect of hyperconjugation, however, may be qualitatively used 

to provide a rationale for the observed phosphorus chemical shifts. Thus, 

compound ( 141) which has two hydrogens in the ring has the most shielded 
31 

P nucleus in the series investigated. Substitution of “-hydrogens by 
a methyl group leads to deshielding of the nucleus as anticipated, e.g., 
( 141) > ( 17) , ( 140) > ( 142) . Similar effects of hyperconjugation in ( 23) 

vs. ( 143) can also be seen. However, the hyperconjugative contribution due 
to the ring substituents in the molecule and that due to the exocyclic 
substituents are not necessarily parrallel, e.g., phosphorus in ( 140) with 
two “-hydrogens in the ring is significantly more shielded than that in 
( 23) which has three “-protons in the exocyclic part of the molecule. 

The effect of m- and p- substituents on the aromatic ring is not sig¬ 
nificant. Thus, chemical shifts for compounds ( 144) , ( 145) and ( 146) are 
very similar to that for the unsubstituted phosphetane (41). 

A very interesting feature of the table 1 is the chemical shift dif¬ 
ferences between the trans and the cis isomers of some of the phosphetanes: 

148 

i.e., the cis isomer is more shielded than the trans . Katz has reported 


65 ppm difference in the two isomers of ( 147) : 






































































160 


P-Ph 

(147) 

This, however, appears to be an isolated case. The dependence of chemi- 

13 

cal shifts of a heteronucleus on stereochemistry has been noted in C 

149 

nmr. Stothers has shown that the carbinol carbons bearing axial and 
equatorial oxygen functions in derivatives of cis and trans 4-t_-butylcyc- 
lohexanol have substantial chemical shift differences. The carbon posses¬ 
sing an axial substituent was shown to be 4-5 ppm more shielded than the 
one with an equatorial substituent: 






























































161 


Table IV-1. 


71 

P Chemical Shifts of Phosphetanes relative to 85% 

W 

6 ref P(0Ci y3 ■ - 140 ' 5 


Compound 

v Sample 

v Ref 

g Correction 

631 P ppm 
observed 

Calcd. 



cps 

cps 





6156 

9338 

+2.25 

-5.8 

-12 



(140) 



^(141) 

6038 

8867 

+2.70 

-21.4 

-15 




_/Ph 

6267 

8870 

+2.40 

-31.0 

-29 



P (17) 

V 

' (142) 

_ 

5945 

8747 

+2.25 

-23.0 

-26 



_Vh , t t . 








6865(trans) 
6389(cis) 

8914 

8914 

+2.03 

+2.03 

-54.2 

-36.6 




(141) 

p 

Ph 

-43 






1 

(23) 

5543 

8225 

+0.62 

-29.5 

-25 




6783(trans) 
6055 (cis) 

8204 

8204 

+0.91 
+ 1.48 

-81.3 

-50.6 




(143) 

> 

-69 



> 

7655(trans) 
7162 (cis) 

9683 

9683 

+2.09 

+2.13 

-55.0 

-34.6 




C(144) 

> 




Ph-F(p) 








_ 

6094(trans) 
5684(cis) 

8174 

8174 

+2.00 

+2.00 

-53.0 

-36.0 




(145) 




Ph-Cl(p) 

6347 (trans) 
5925(cis) 

8419 

8174 

+2.00 

+2.00 

-53.2 

-36.0 




(146) 




x 

Ph-Cl(m) 

9453(trans) 
8884 (cis) 

8765 

8717 

+ 2.33 
+2.30 

-165.5 

-145.0 



—i 

(48) 

{ 











































































-70 


162 



Observed P Chemical Shifts, ppm. 

O 1 

Fig. IV-1 . Plot of Calculated vs Observed P Chemical Shifts of Phosphetanes. 












163 



The contribution to the chemical shift of the substituent not bonded 
to the nucleus is questionable, since the relative position of the nucleus 
with respect to the substituent does not change significantly (see also ref 



Thus, the differences of the chemical shifts in isomers must arise either 

from the substituents or the lone pair of electrons. In the case of phos- 

31 

phetane oxides and phosphetanium salts, no differences in P chemical 

shifts was observed between the isomers. This leads one to conclude that 

the orientation of the lone pair of electrons is probably a major cause 

of the chemical shift difference between the isomers of phosphetanes. In 

31 

any event, these differences in P resonance in isomers ( cis and trans) 
of phosphetanes are empirically useful in making isomer assignments. 

Attempts to make correlations in phosphonium salts have been reported 


150 


Once again, electron withdrawing or donating effects fail to give 

150 


correlation with available data . Hyperconjugative can be qualitative¬ 
ly used to rationalize the observed chemical shifts in some open-chain 

systems, but not enough data is available to give any quantitative predic- 

+ ...... 150 

tability 

31 

The qualitative correlation of P chemical shifts in phosphetanium 

salts may also be attempted on the basis of hyperconjugation arguments. 

150 


Thus, phosphorus in compound 


is more shielded than in (27) as antici- 










































































164 


pated (table IV-2). Similar arguments, however, fail to hold for the exo- 
cyclic substituents in case of compounds (22), (94)> and ( 149 ). The ex¬ 
pected trend on the basis of hyperconjugation is: 

(149)> (27)> (94) 
where as the observed order is: 

(27)> (94)> ( 149) 


The chemical shifts of phosphinate esters are shown in Table IV-3. The 
values for various compounds are so close to each other, that correlations 
on the basis of hyperconjugation (or any other effect for that matter) can 
not be used. 

Proton magnetic resonance spectroscopy has also provided a valuable 

2 11 

tool for the stereochemical assignments in the phosphetane compounds * , 

and has been successfully employed for following the stereochemical course 

26 33 

of reactions of these compounds “ 5 . Fig. IV-2 shows a partial pmr spec¬ 

trum of a mixture of isomeric chlorophosphetanes ( 48) . The corner proton 
at position 3 in the trans isomer (1,0 and 3-methyl, trans) (see p.16). 

is less shielded than that in the cis isomer. The assignment is consistent 
31 

with the P data. The corner proton with phosphorus in the cis isomer 

is observed. Since, in this case as well as in others phosphetanes for the 
3 3 

comer proton, JpccH( trans ) JpcCH^ c ^ s ^ seems t0 hold the relation may 
be a good guide-line for the isomer assignments. 

Another application of isomer assignments is possible in the phenyl- 

phosphetane (148). Cremer and Chorvat found that the compound did not 

2 

isomerice on heating . The assignment of the geometry of the 1,phenyl. 


2-methyl relationship was not made. In light of the additional information 
below, a tentative assignment may be made. 















165 


31 

Table IV-2. P Chemical Shifts of Phosphetanium Salts Relative 
to 85% H^PO^. Reference <SP(OCH^)2=140.5 ppm. 


Compound 

vSample 

cps 

vReference 

31 

^Correction 6 P ppm. 



© 


Ph 


(27) 


/ 


P x ( 148) 
Ph 



(149) 


Ph 


(94) 


(150) 


4119 


5491 


7573 


3926 


7215 


6097 


7712 


9485 


5977 


9560 


+0.25 -58.65 


+0.77 -48.3 


+0.98 -61.8 


+1.12 -55.5 


+0.72 -43.0 




































166 


Table IV-3. Chemical Shifts of Phosphinate Esters Relative to 

85% H 3 P0 4 .<5 Reference (CH PC1 2 ) = 191 ppm. 


Compound 


vSample vReference 6Correction 63/p 

Observed 


(11) 

5404 

8840 

-0.9 

-50.5 

(15) 

5367 

8850 

-0.8 

-48.5 

(8) 

5464 

8874 

-0.9 

-49.8 

(86) 

5532 

8895 

-1.1 

-53.7 

(93) 

5459 

8887 

-0.9 

-50.8 



OCH, 


( 8 ) 



0 

0CH 3 


( 11 ) 



0 

GCH 3 


(15) 



OCH, 


( 86 ) 



OCH. 


(93) 































167 


























168 


Gagnaire has described the influence of bond orientation around 

phosphorus on the coupling constants in trivalent phosphorus compounds 

The dependance of the magnitude as well as the sign of the coupling constants 

on orientation has also been shown. Fig. IV-3 represents the curve of 
2 

JpCH vs - f( a ) i n which “ is the dihedral angle subtended by the two 
planes defined by the P, C and H atoms and by the C-P bond together with 
the three-fold axis of the bonds around phosphorus (assuming a regular py¬ 
ramidal arrangement, with a bond angle value of 100° for all compounds. 

152 

On the basis of this curve, Gagnaire has made the following 
geometrical assignment: in the case of phenylphosphetane (149) : 



2 

The observed coupling constant Jp^ in phenylphosphetane ( 148) is +7 cps. 
This suggests that the phenyl group and the methyl group are on the same 
side of the ring, i.e., the phenyl and the methyl are cis as represented 
in the illustration below: 


+_ 7 cps 

( 148) 

3. Experimental . 1 'Ph 

The proton magnetic resonance spectra of the compound were obtained on 
a Varian A-60 instrument. For the decoupling experiment, a suitable peak 
for each compound in the pmr was observed while irradiating the sample 
with another frequency in the vicinity of 24.3 mcps provided by an HD-60A 
oscillator (NMR Specialties Co.). The frequency was modulated with an au¬ 
dio oscillator and was transmitted through the crystal, to the sample. De- 














































169 



Fig. IV-3. 







































170 


coupling was evident from the collapse of the splitting pattern due to 
phosphorus, i.e., a doublet — a singlet, etc.. The gain in the decoupler 
was minimized in order to reduce the error in the frequency measurement. 
The frequency of decoupling was observed in the counter while allowing a 
stream of dry nitrogen to cool the sample. This was necessary to over¬ 
come the heating effect during irradiation. The chemical shifts of the 

144 

phosphorus nuclei were calculated using the following relations : 


<5 Sample 


vsample- vref. 
24.3 


SUncorrected = ^sample - 6ref. 

6ppm = 6uncorrected +_ 6correction 

correction in ppm is the difference in the chemical shifts of the hydrogen 

of the compound under investigation and the proton shift of the reference 

compound. The Correction factor Scorrection is to be added or subtracted 

depending upon whether the proton signal (used for decoupling) of the sam- 

pel is at a higher or lower field than the signal for the reference materi- 
144 


al 






171 


REFERENCES 

1. S. E. Cremer and R. J. Chorvat, Tetrahedron Lett., 413 (1968. 

2. R. J. Chorvat, Ph.D. Thesis, Illinois Institute of Technology, Chicago, 

Illinois, 1968. 

3. G. Markl, Angew. Chem. Internat. Ed. Engl., £, 1023 (1965). 

4. G. M. Kosolapoff and R. F. Struck, J. Chem. Soc., , 3739 (1957). 

5. R. I. Wagner, U.S. Patent 3,086,053 (1963); Chem. Abstr., 59 , 10124 

(1963). 

6. R. I. Wagner, U.S. Patent 3,086,056 (1963); Chem. Abstr., 60^, 559 

(1964). 

7. R. I. Wagner, L. D. Freedman, H. Goldwhite and D. G. Rowsell, J. Amer. 

Chem. Soc., 89^, 1102 (1967). 

8. M. Green, J. Chem. Soc., 541 (1965). 

9. S. E. Cremer, Private Communication. 

10. J. J. McBride, Jr., E. Jungermann, J. V. Killhefer and R. J. Clutter, 

J. Org. Chem., 27, 1833 (1962). 

11. S. E. Cremer and R. J. Chorvat, J. Org. Chem., 32 , 4066 (1967). 

12. T. A. Zyablikova, A. R. Panteleeva and I. M. Shemegron, Izvest. Akad. 

Nauk S.S.S.R. Ser. Khim., 373 (1969) (Russ). 

13. D. Berglund and D. W. Meek, J. Am. Chem. Soc., 90 , 518 (1968). 

14. C. H. Chang, Ph.D. Thesis, Illinois Institute of Technology, Chicago, 
Illinois (1969). 

15. A. M. DeRoos and H. J. Toet, Rec. Trav. Chim. Pays-Bas, 78 , 27 (1959). 

16. S. E. Cremer and R. J. Chorvat, Tetrahedron Lett., 413 (1968). 

17. S. E. Cremer, Chem. Commun., 1132 (1968). 

18. S. Trippett and W. Hawes, J. Chem. Soc., C^, 1465 (1969). 

19. R. 0. House, " Modern Synthetic Reactions" , W. A. Benjamin, New York, 

N. Y. (1965). 

20. C. S. Marvell and L. A. Brooks, J. Am. Chem. Soc., 63, 2630 (1941). 

21. R. F. Hudson, " Structure and Mechanism in Organophosphorus Chemistry" , 

Monograph 6, Academic Press, New York, N. Y. (1965). 







172 


22. J. Hine, "Physical Organic Chemistry", McGraw-Hill, New York, N. Y., 

(196 ). 

23. L. Horner, H. Hoffmann and H. G. Wippel, Chem. Ber., 91 , 64 (1958). 

24. K. Sasse in " Methoden Per Organischen Chemie" , Vol. XII, Part I, ed. 

by E. Muller, George Thieme Verlag, Stuttgard, (1963). p. 239. 

25. L. Horner and W. D. Balzer, Tetrahedron Lett., 1557 (1967). 

26. S. E. Cremer, R. J. Chorvat, C. H. Chang and D. W. Davis, Tetrahedron 

Lett., 5799 (1968. 

27. L. M. Trefonas and C. Moret, J. Am. Chem. Soc., 91 , 2255 (1969). 

28. M. J. Gallagher and I. D. Jenkins, " Topics in Stereochemistry" , Vol. 

3, N. L. Allinger and E. L. Elliel, Ed., John Wiley and Sons, Inc., 

New York, N. Y., (1 68). 

29. D. W. Denney and N. G. Adin, Tetrahedron Lett., 2569 (1966). 

30. C. N. Caughlan and M. Haque, Chem. Commun., 1228 (1963). 

31. C. N. Caughlan, Private Communication. 

32. M. Haque, Private Communication. 

33. S. E. Cremer and B. C. Trivedi, J. Am. Chem. Soc., 91_, 7200 (1969). 

34. K. Bergesen, Acta. Chem. Scand., 21 , 1587 (1967). 

35. F. H. Westheimer, Accounts Chem. Res., 70 (1968). 

36. E. Uhing, K. Rattanbury and L. Maier, J. Am. Chem. Soc., 83 , 2299 (1961). 

37. H. E. Ulmer, L. C. D. Greenweghe and L. Maier, J. Inorg. Nucl. Chem., 

20_, 82 (1961). 

38. S. Trippett and D. J. H. Smith, Chem. Commun., 855 (1969). 

39. R. Collin and G. Shreder, German Patent, 1,056,606 (1958), Chem Zent., 
130 , 17390 (1959). 

40. H. L. Roberts in " Inorganic Sulfur Chemistry" , G. Nickless, Ed., El¬ 
sevier Publ Co., Amsterdam (1968) Chapter 12. 

41. M. S. Kharasch and H. C. Brown, J. Am. Chem. Soc., 61, 2142 (1939). 

42. Zh. L. Eutikhov, N. A. Razumova and A. A. Petrov, Proc. Acad. Sc. 

USSR (English Tmas.) 181 , 746 (1968). 















173 


43. W. E. Backman and C. H. Boatner, J. Am. Chem. Soc.., 58 , 2194 (1936). 

44. A. J. Kirby and S. G. Warren, " The Organic Chemistry of Phosphorus" , 

Monograph 5, Elsevier Publ. Co., Amsterdam, (1967). 

45. R. J. Gillespie, J. Chem. Soc., 1002 (1962). 

46. R. J. Gillespie, J. Chem. Soc., 4672 (1963). 

47. H. Braun and P. Pinnow, Z. Physik. Chem., B-25 , 239 (1937). 

48. F. A. Cotton and G. Wilkinson, " Advanced Inorganic Chemistry" , Inter¬ 
science, New York, N. Y. (1962). 

49. K. W. Hansen and L. S. Bartell, Inorg. Chem., 4 _, 1775 (1965). 

50. H. S. Gutowsky and A. D. Liehr, J. Chem. Phys., 20 , 1652 (1952). 

51. H. S. Gutowsky, D. W. McCall and C. P. Slichter, J. Chem. Phys., 

21 _, 279 (1953). 

52. R. R. Holmes, J. E. Griffith and R. P. Carter, J. Chem. Phys., 41 , 

863 (1964). 

53. R. R. Holmes, J. E. Griffith and R. P. Carter, Inorg. Chem., 3_,1748 
(1964). 

54. F. Ramirez, W. C. Hamilton and S. J. LaPlaca, J. Am. Chem. Soc., 87_, 
127 (1965). 

55. R. S. Berry, J. Chem. Phys., 32 , 933 (1960). 

56. J. J. Downs and R. E. Johnson, J. Am. Chem. Soc., 77 , 2098 (1955). 

57. J. R. VanWazer and J. H. Letcher, J. Chem. Phys., 45 , 2926 (1966) 

58. E. L. Muetterties, W. Mahler and R. Schmutzler, Inorg. Chem., 2_, 

613, (1963). 

59. K. W. Hansen and L. S. Bartell, Inorg. Chem., 4 _, 1777 (1965). 

60. E. L. Muetterties, W. Mahler, K. J. Parker and R. Schmutzler, Inorg. 
Chem., .3, 1298 (1964). 

61 P. M. Triechel, R. A. Goorich and S. B. Pierce, J. Am. Chem. Soc., 

89_, 2017 (1967). 

62. R. S. Drago and P. C. VanderVoom, J. Am. Chem. Soc., 88 , 3255 (1966). 

63. H. Bent, Chem. Rev., 61 , 275 (1961). 

64. D. P. Craig, R. S. Nyholm, A. Maccoll and L. E. Sutton, J. Chem. 

Soc., 332 (1954). 














65 

66 

67 

68 

69 

70 

71 

72 

73 

74 

75 

76 

77 

78 

79, 

80. 

81. 

82. 

83. 

84. 

85. 

86 . 

87. 

88 . 


174 


R. J. Gillespie, Can. J. Chem., 38, 818 (1960), 39, 318 (1961). 

R. J. Gillespie, J. Chem. Soc., 332 (1954). 

R. Schmutzler, Inorg. Chem., 2_, 613 (1963). 

R. Alver and B. H. Holtedahl, Acta. Chem. Scand., 21, 359 (1967). 

W. N, Lipscomb and Y. H. Chiu, J. Am. Chem. Soc., 91^, 4150 (1969). 

F. H. Westheimer and R. Kluger, J. Am. Chem. Soc., 91_, 4143 (1969). 

D. J. Cram, Private Communication. 

J. D. Dunitz and V. Prelog, Angew. Chem. Intemat. Ed. Engl. 7, 725, 
(1968). 

K. Mislow, K. E. DeBruin, K. Naumann and G. Zon, J. Am. Chem. Soc., 

91, 7031 (1969). 

F. H. Westheimer and F. Covitz, J. Am. Chem. Soc., 85_, 1773 (1963). 

F. H. Westheimer and P. C. Haake, J. Am. Chem. Soc., 8Z_, 1102 (1961). 

D. B. Boyd, J. Am. Chem. Soc., 9^, 1200 (1969). 

F. H. Westheimer and E. A. Dennis, J. Am. Chem. Soc., 88^, 3432 (1966) 

G. Aksnes and K. Bergesen, Acta, Chem. Scand., 20 , 2508 (1966). 

D. Samuel and B. L. Silver, Advan. Phys. Org. Chem., 3_, 123 (1965). 

F. H. Westheimer, F. Kerst, D. G. Lee and E. A. Dennis, J. Amer. Chem. 
Soc., 89^_ 3918 (1967). 

F. H. Westheimer, F. Covitz, E. A. Dennis, R. Kluger and L. David 
Williams, J. Am. Chem. Soc., 91 , 6066 (1969). 

R. F. Hudson and M. Green, Proc. Chem. Soc., 307 (1962). 

R. F. Hudson and M. Green, Angew. Chem. Intemat. Ed. Engl. 2, 17, (1963) 

J. Hine, J. Org. Chem., 31 , 1236 (1966). 

F. 0. Rice and E. Teller, J. Chem. Phys., 6_, 489 (1938), 7_, 199 (1939). 

S. Trippett and W. Hawes, Chem. Commun., 577 (1968). 

F. G. Bordwell and G. D. Cooper, 73 , 5184 (1951). 

K. Bergesen, Acta. Chem. Scand., 21, 1587 (1967). 




































































175 


89. A. A. Forst and R. G. Pearson, "Kinetics and Mechanism", John Wiley 
and Sons, Inc., New York, N.Y. (1961). 

90. A. W. Hoffmann, Ann., 78_, 253 (1851). 

91. A. W. Hoffman, Ber., 1£, 459, 659 (1881). 

92. A. W. Hoffman, Ann., 79, 11 (1851). 

93. A. W. Hoffmann, and A. Cahours, Ann., 104 , 1 (1857). 

94. E, A. Letts and N. Collie, Trans. Roy. Soc. Edinb.,30^, 181 (1885). 

95. J. Meisenheimer, Ann., 449 , 218 (1926). 

96. C. K. Ingold and G. W. Fenton, J. Chem. Soc., 2342 (1929). 

97. J. Meisenheimer, Ber., 44 _, 356 (1911). 

98. C. K. Ingold and L. Hey, J. Chem. Soc., 531 (1933). 

99. H. L. Homer, H. Hoffmann, H. G. Wippel and G. Hassel, Chem. Ber., 

91_, 52 (1958). 

100. W. E. McEwen, K. F. Kumli, A. Blade-Font, M. Zanger and C. A. VanDer 
Werf, J. Am. Chem. Sco., 8(3, 2378 (1964). 

101. W. E. McEwen, K. F. Kumli and C. A. VanDer Werf, J. Am. Chem. Soc., 

81_, 3806 (1959). 

102. W. E. McEwen, G. Axelrad, M. Zanger, and C. A. VanDer Werf, J. Am. 

Chem. Soc., 87_, 3948 (1965). 

103. S. Trippett, J. R. Corfield and J. R. Shutt, Chem. Commun., 789 (1969). 

104. S. E. Cremer, R. J. Chorvat and B. C. Trivedi, Chem. Commun. 769 (1969). 

105. W. E. McEwen, K. F. Kumli and C. A. VanDer Werf, J. Am. Chem. Soc., 

81_, 3805 (1959). 

106. K. L. Marsi, J. Am. Chem. Soc., 91 , 4724 (1969). 

107. K. Mislow, K. Naumann and _P. D. Henson, J. Am. Chem. Soc., 91, 5646, 

(1969). ~ 

108. G. Aksnes and K. Bergensen, Acta. Chem. Scand., 19^, 931 (1965). 

109. G. Aksnes and J. Songstad, Acta. Chem. Scand. 16, 1426 (1962). 

110. G. Markl, Z. Naturforsch, 18b , 84 (1963). 

111. K. A. Petrov, V. A. Parshina and V. A. Gaidaniak, Zhur, Khim., 31, 

3411 (1961). 









































176 


112. S. A. Buckler and M. Epstein, Tetrahedron, 18^, 1231 (1962). 

113. S. A. Buckler and V. P. Wystrach, J. Am. Chem. Soc., 83 , 168 (1961). 

114. H. Hellmann and Jorg Bader, Tetrahedron Lett., 724 (1961). 

115. K. D. Berlin and M. E. Peterson, J. Org. Chem. 32 , 125 (1967). 

116. L. M. Yagupolskii and P. A. Yufa, J. Gen. Chem. USSR. Engl. 28 , 

2880 (1958). 

117. M. S. Newman, J. R. Leblanc, H. A. Karnes and G. Axelrad, J. Am. 

Chem. Soc., 86 , 868 (1964). 

118. D. E. McGreer, Can. J. Chem., 38, 1638 (1960). 

119. C. H. DePuy, G. M. Dappen, K. L. Eilers and R. A. Klein, J. Org. 

Chem., 29_, 2813 (1964). 

120. N. Krutzkamp, E. Schmidt-Samoa and K. Herberg, Angew Chem. 77_, 

1138 (1965). 

121. M. Hojo and Z. Yoshida, J. Am. Chem. Soc., 90 , 4596 (1968). 

122. M. I. Kabachnik and E. N. Tsvetov, Proc. Acad. Sc. USSR., Chem. Sec., 
Engl. 143 , 211 (1962). 

123. C. E. Griffin, E. H. Uhing and A. D. F. Toy, J* Am. Chem. Soc., 83 , 
2299 (1961). 

124. C. E. Griffin, E. H. Uhing and A. D. F. Toy, ibid , 87 , 4757 (1965). 

125. H. Goldwhite, B. Fontal and D. G. Rosell, J. Org. Chem,, 31, 2424 

(1966). 

126. K. Moedritzer, J. Am. Chem. Soc., 8_5, 4381 (1961). 

127. A. Aguiar and D. Daigle, J. Am. Chem. Soc., 86 , 2299 (1964). 

128. J. Meisenheimer, J. Casper, M. Horing, W. Lauter, L. Lichtenstadt 

and W. Samuel, Ann., 449 , 213 (1926). 

129. A. E. Arbuzov, J. Russ. Phys. Chem. Soc., 42 , 549 (1910). 

130. D. E. Seyferth, D. E. Welch, and J. K. Heeran, J. Am. Chem. Soc., 

86_, 1100 (1964). 

131. H. Hoffmann, J. Am. Chem. Soc., 52_, 2995 (1930). 

132. J. B. Conant and W. R. Kimer, ibid ., 46 , 232 (1924). 

133. J. A. Popple, W. G. Schneider and H. J. Bernstein, "High Resolution 

Nuclear Magnetic Resonance", McGraw-Hill, New York, N.Y., (1959). 















































































177 


134. F. A. Bovey, " NMR Data Tables for Organic Compounds 1 *, Vol. 1, Inter¬ 
science , New York, N.Y. (1967). 

135. J. W. Emsley , J. Feeney and L. H. Sutcliffe, " High Resolution Nu¬ 
clear Magnetic Resonance Spectroscopy 1 *, Pergamon, Oxford, (1966). 

136. F. A. Bovey, " Nuclear Magnetic Resonance Spectroscopy" , Academic 
Press, New York, N.Y., (1969). 

137. "Topics in Phosphorus Chemistry", Vol. 5, *^P NMR, Academic Press, 
New York, N.Y., (1967). 

138. P. C. Laute^ber, "N uclear Magnetic Resonance Spectra of Elements 
Other Than H and F in Determination of Organic Structures by 

Physical Methods" , Vol. 2, F. C. Nachod and W. D. Phillips, ed., 
Academic Press, New York, N.Y., (1962). 

139. D. Baldeschweiler and E. W. Randall, Chem. Rec., 81 (1963). 

140. F. Bloch, Phys. Rec., 102, 104 (1959). 


141. V. Royden, ibid , 96_, 543 (1954). 

142. J. D. Baldeschweiler, J. Chem. Phys., 36 , 152 (1962). 

143. D. W. Turner, J. Chem. Soc., 847 (1962). 

144. J. B. Stothers and J. R. Robinson, Cand. J. Chem., 967 (1964). 

145. S. 0. Grim, and W. McFarlane, Nature, 208, 995 (1965). 


146. L. Maier, Helv. Chim. Acta., 49^, 1718 (1966). 

147. S. 0. Grim. W. McFarlane and E. F. Davidoff, J. Org. Chem. 32 , 781, 
(1967) . 

148. T. J. Katz, C. R. Nicholson and C. A. Reilly, J. Am. Chem. Soc., 

89_, 4301 (1967) . 

150. S. 0. Grim, W. McFarlane, E. F. Davidoff and T. J. Marks, J. Phys. 
Chem., 70_, 581 (1966). 

151. D. Gagnaire, J. P. Albrand and J. B. Robert, Chem. Commun. 1469 (1968). 

152. D. Gagnaire, J. P. Albrand, J. Martin and J. B. Robert, Bull. Soc. 

Chim. Fr., 40 (1969). 



















178 


VITA 


Name: Bhupendra Chandrashanker Trivedi 

Bom: September 18, 1942; Godhra, Gujrat (India) 

Education: 

The New Era High School, Godhra (India 
1951-1958 

B.Sc. Degree, M. S. Univeristy, Baroda (India) 
1958-1962 

M.Sc. Degree, M. S. University, Baroda (India) 
1962-1964 

Illinois Institute of Technolgoy, Chicago, Ill. 
1965-1970 

Publications: 

Stereochemistry of Base Decomposition of Phos- 
phetanium Bromides. S.E. Cremer, R.J. Chorvat 
and B.C. Trivedi , Chem. Commun., 769 (1969) 

Stereochemistry of Nucleophillic Attack at a 
Phosphinate Ester. S.E. Cremer and B.C. Trivedi , 
J. Am. Chem. Soc., 91 , 7200 (1969). 

The Chemistry of Phosphetanium Salts. S.E. Cre¬ 
mer, R.J. Chorvat and B.C. Trivedi , Presented 
at the Third Great Lakes Regional Meetings, 
American Chemical Society, Dekalb, Illinois, 

June 1969. 

Societies: 


American Chemical Society 
Phi Lambda UpsiIon 






















Bound By 
BURGMEIER 
Book Bindery, Inc. 

CHICAGO, ILL. 60606