(i).
OXIDATIVE CHEMISTRY OF PHOSPHOROTHIONATE
AND RELATED COMPOUNDS
A thesis
submitted in fulfilment of
the requirement for the degree of
Doctor of Philosophy
from
Department of Organic Chemistry
School of Chemistry
University of New South Wales
by
Jeong Han Kim
March 1992 (ii)
DECLARATION
This is to certify that the work described in this thesis was carried out by the author during the period March 1988 to March 1992, in Department of Organic Chemistry, University of New South Wales, Australia under the supervision of Dr. Robert F. Toia and Associate Professor Michael J. Gallagher. It is hereby declared that this thesis has not been submitted in part or in full to any other university or institution for the any degree or diploma.
Date: .l.$:;.J.... ~.t't.~. (iii)
ACKNOWLEDGEMENTS
I would like to express my deepest gratitude and appreciation to my supervisors,
Dr. Robert F. Toia and Associate Professor Michael J. Gallagher for their help, guidance and continuous encouragement throughout the study.
My special thanks also go to Professor J. E. Casida for the valuable extensive study in Pesticide Chemistry and Toxicology Laboratory in U. C. Berkeley and to
Professor C. K. Park in Seoul National University for his guidance and encouragement.
Many thanks also to Dr. J. Hook for his great help with 3lp NMR, to the staff members of the Department of Organic Chemistry in University of New South Wales and to my fellow students for their help and friendship.
Great acknowledgement is made to the Australian government for financial support in the form of a Commonwealth Postgraduates Research A wards.
I would like to express my appreciation and love to my dear wife, family and relatives for their patience, encouragement and support. (iv)
ABSTRACT
The oxidation of cyclic phosphorothioates has been studied to investigate the stereochemical changes which occur during the oxidatively - induced displacement reactions. The stereochemistry of the starting materials, i.e. cis- and trans-5- hydroxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3,2-dioxaphosphorinane was confirmed by single crystal X-ray crystallography of their acetate and 3,5-dinitro benzoate derivatives, respectively.
All of the oxidation studies were carried out with varying molar ratio of MCPBA n (m-chloroperbroic acid) to substrate and variety of solvents were used. Product formation on oxidation of the above cis- and trans- isomers in CDCl3 occur with substantial retention of configuration. With 1-methyl-4-phospha-3,5,8-trioxabicyclo
[2,2,2] octane-4-sulfide (BPS) only 1-methyl-4-phospha-3,5,8-trioxabicyclo [2,2,2] octane-4-oxide (BPO) was obtained. In CH30H, however, both isomers and BPS reacted similarly to give the cis- and trans-methyl phosphates by non stereospecific phosphorylation of methanol. The major intermediate noted during the oxidation was extensively studied by spectroscopic methods and assigned as the monocyclic sulfenate. Minor products and non-phosphorus products were also assigned by MS and
GC-MS studies. In aqueous acetone, an acyclic hydrogenphosphonate was commonly observed from both isomers and BPS.
With 5,5-dimethyl dioxaphosphorinanes as comparison compounds, it was found that the hydroxyl group of the cis- and trans- isomers does not affect the overall reaction in CH30H although it may play a role in stabilizing the intermediates.
From model studies with acyclic phosphorothioates, disulfides were found as phosphorylating intermediates in CH30H. Using an 180-containing thiolate as the starting material the mechanistic pathway was investigated in the formation of the methyl (v) ester from oxidation in CH30H and it appeared that product fonnation proceded via a rearragement process.
In an attempted synthesis of the major intennediates by the reaction between cis trans- isomers and S(hCh only BPO was obtained. Furthe(more the reaction between BPS and S(hCh yielded trans-monocyclic chloridate which was confirmed by X-ray analysis and the mechanistic details of their reaction were postulated.
In the alcoholyses of BPO and BPS stereospecific ring opening to the trans isomer was observed at early stage and this was confirmed by X-ray analysis. It was also shown that the trans- and' cis- isomers equilibrated through the intermediate acyclic derivatives. (vi)
ABBREVIATIONS
BPO 1-Methyl-4-phospha-3,5,8-trioxabicyclo [2,2,2] octane-4-oxide
BPS 1-Methyl-4-phospha-3,5,8-trioxabicyclo [2,2,2] octane-4-sulfide
MCBA m-Chlorobenzoic acid
MCPBA m-Chloroperbenzoic acid
MIBO 2-Methylthio-4H-1,3,2-benzodioxaphosphorin-2-oxide
1MP Trimethyl phosphite
TMPO Trimethyl phosphate
1MPS Trimethyl phosphorothionate (vii)
TABLE OF CONTENTS
Title page i
Declaration 11
Acknowledgements iii
Abstract iv
Abbreviations vi
Table of Contents Vll
Introduction 1
Results and Discussion
Syntheses of Cyclic phosphorothioates and related compounds. 35
Structural assigment of the isomers of dioxaphosphorinanes 41
Oxidation of cis- and trans-5-hydroxymethyl-5-methyl-2-thiono-r-
2-ethoxy-1,3,2-dioxaphosphorinane [cis- and trans-(116)] by MCPBA. 58
Oxidation of BPS by MCPBA. 77
Oxidation of 5,5-dimethyl-2-thiono-2-ethoxy-1,3,2
-dioxaphosphorinane ( 133). 86
Spectroscopic investigation of the intermediate E in the oxidation of
BPS by MCPBA in CD3OD (CH3OH). 88
Oxidation of Triethyl phosphorothionate (TEPS)
and trimethyl phosphorothionate (TMPS) 105 (viii)
Oxidation of the sulfenate (EtO)iP(O)SOCH3 (146) by MCPBA
112
Reaction of trimethyl phosphite (TMP) with Bis(diethoxy phosphinyl)
·disulfide. 114
Oxidation of bis(diethyl phosphinyl)disulfide (130) by MCPBA. 115
Oxidation of Q,Q-diethyl-S-phenyl phosphorothiolate-1SO
(lSO-thiolate, 186) by MCPBA in CH3OH. 117
Oxidation of 5-hydroxymethyl-5-methyl-2-thiono-2-ethoxy
-1,3,2-dioxaphosphorinane (116) by SO2Cl2 in CDCl3_ 121
Reaction of BPS with SO2Cl2. 125
Alcoholysis of cyclic phosphates 132
Conclusion 144
Experimental 146
References 219 Appendices
Appendix A: X-Ray data of rrans-5-acetoxymethyl-5-methyl
-2-thiono-2-ethoxy-1,3,2-dioxaphosphorinane (119). 236 Appendix B: X-Ray data of trans-5-(3,5-Dinitro)benzoyloxymethyl-5. -methyl-2-thiono-r-2-ethoxy-1,3,2-dioxaphosphorinane ( 122). 240
Appendix C: X-Ray data of trans-5-chloromethyl-5-methy-2-thiono
-r-2-ethoxy-1,3,2-dioxaphosphorinane ( 195). 245
Appendix D: X-Ray data of trans-5-hydroxymethyl-5-methyl
-2-oxo-r-2-methoxy-1,3,2-dioxaphosphorinane (118). 248 (ix)
Appendix E: X-Ray data of trans-5-hydroxymethyl-5-methyl
-2-thiono-r-2-methox y-1,3 ,2-dioxaphosphorinane (203). 252 INTRODUCTION 1
The chemistry of phosphorothioates has been the subject of much attention, particularly with respect to defining the mechanism of action of insecticidally active
I compounds. Of the approximately 106 organophosphorus pesticides in current use, phosphorothioates account for more than two thirds1 and in biological systems these are converted to metabolites which can be more toxic or less toxic than the parent compounds. These processes are termed "B ioacti vation" and "Detoxification", respectively. In many instances, these transformations are oxidatively induced, and these can be grouped into four broad classes of reactions, as illustrated below :
(a) oxidative desulfuration of the thiophosphoryl group.
s 0 a I 1 R- P-R1 R- P-R I I R~ R2
(b) conversion of a sulfide to a sulfoxide or sulfone. 0 0 t t ·1 1 1 R-S-R R-S-R R-S-R J. 0
(c) hydroxylation of an alkyl or aryl group.
(d) N-oxidation of a phosphoramidate. 2
Of these transformations, oxidative desulfuration is particularly interesting because it normally results in a bioactivated compound. Oxidative desulfuration of the thiophosphoryl group has been shown to occur for a wide range of thiono-pesticides2 including phosphorothionates such as parathion (1), fenitrothion (2), phosphorothiolothionates such as dimethoate (3), and malathion (4) and phosphonothiolothionates such as fonofos (5).
(2) (3) (1)
(5) (4)
Relative to the insecticidal activity of these compounds, the conversion is accepted as an activation process since the parent thiophosphoryl compounds are poor inhibitors of acetylcholinesterase whereas the phosphoryl derivatives are good inhibitors. To illustrate this point, LDso (rat, oral) data for selected phosphorothioates
and their corresponding oxidative desulfuration products are given in Figure 1. These
transformations can be brought about chemically, photochemically through exposure to
sunlight, in animals by microsomal mixed function oxidases (mfo), or in plants by
peroxidases. 3 3
(0]
parathion (1) paraoxon
1.4 m50 3·3
s 0 D u (CH30)2P-S - CH- C02~Jis [0] (CH30)2P - S - CH-C02~H5 I I CJ½C02~H5 CH2C02~Hs
malathion (4) malaoxon
LDso 2600 308
[0]
· fonofos (5). fonofos oxon
LD50 14·7 2.8
Figure 1. Selected phosphorothionate pesticides, their oxidative desulfuration products and rat LDso values (oral, mg/kg).2 4
Oxidative desulfuration has received extensive chemical study, and a number of oxidants have been used to effect the transformation. These include peroxytrifluoroacetic acid,4 potassium permanganate,5 nitric acid,6 dimethyl sulfoxide7(or selenoxide8), ozone,9 dinitrogen tetroxide,10 hydrogen peroxide,11-13 and m-chloroperbenzoic acid (MCPBA) (6).11,14-34 However, if the results from the chemical studies are to be considered relevant to the biological system then the chemical model should mimic the biological oxidizing system as closely as possible. Specificially, it should show the same stereospecificity (retention of configuration) when chiral compounds are involved, and the newly introduced oxygen atom should come directly from the oxidant rather than from water.27 The oxidation by MCPBA has been considered to generally meet these requirements and the pattern of oxidation products was found to be similar to that of metabolic transformation. Therefore, the MCPBA-oxidation system has been widely used as a biomimic model for oxidation of not only phosphorothionates, 17,34 but also phosphorothiolates 22,23 and dithioates.14,20
More recently, in an attempt to improve the biomimetic model for relevance, the oxidations have been studied in aqueous solvents. In these studies magnesium monoperoxyphthalate (MMPP) (7) has been used in place of MCPBA because of the low solubility of the latter compound in aqueous systems.35 This led to a series of interesting observations relative to the influence of solvent on the reaction, and these are described later (see page 21).
( ©(co; )Mg-6ffiO COtf 2
(7) (6)
In addition to the formation of the phosphoryl products the reactive, short lived intermediates which occur during the course of the transformation in the oxidation of 5 phosphorothioates are also interesting. Some of these are potent phosphorylating agents22,24 and are considered to be responsible, at least in part, for producing the adverse biological effects associated with phosphorothioates.36
In an investigation of the oxidation of parathion (1) using microsomal mixed function oxidases from rat liver, fortified with NADPH and in an atmosphere of 1802,
Ptashne and coworkers found that the resulting paraoxon (8) was isotopically enriched.37 Based on this result they proposed a three membered cyclic intermediate (9) which could rearrange with loss of sulfur to form the oxon (8). This intermediate (9) could also account for the diethyl phosphorothioic acid (10) which would arise as a result of nucleophilic attack by H20 following an initial dissociation (Figure 2).
[ (C,H5Q½ ~'lo-@-N02 ]
(1) / (9) l
0 II + -o -@-NO:, (C2H5O)2 - P - O
(8)
s II (CiffsO)iP -OH
(10)
Figure 2. Proposed mechanism for the mixed function oxidase metabolism of parathion (1).37 6
It should be noted that the oxygen transfer mechanism as proposed by Ptashne and coworkers had earlier been suggested for the peroxytrifluoroacetic acid oxidation and for the microsomal oxidation of a number of aromatic compounds including naphthalenes. In these studies the usefulness of the peracid oxidation as a model system for the metabolism of aromatic compounds by mixed function oxidases was also verified.38,39 In comparative studies on the oxidation of dyfonate [fonofos (5)] with
MCPBA, and with mixed function oxidases from rat liver as the oxidants, McBain and coworkers 14,40 obtained the same end product in both cases. This was assigned as the cyclic oxygenated intermediate (11). However, this proposed structure proved to be incorrect and other workers showed the compound to be phenylethoxy(ethyl)phosphinyldisulfide (12).15 It is of interest to note that this corrected structure did provide indirect evidence for the existence of the proposed cyclic intermediate (11), i.e. compound (12) can be accounted for by rearrangement of (11). s-o-@ CH3CI½,-\~I - s 0
OCH2CH3 (11) (12)
The stereochemistry of the phosphorothioate - phosphate transformation is complex and shows interesting results depending on the oxidants used. For example, in a study of the oxidation of a chiral phosphine sulfide with potassium permanganate,
Horner and WinklerS observed that the reaction proceeded with retention of configuration whereas with dimethyl selenoxide, 8 dimethyl sulfoxide 7 and hydrogen peroxidel2 inversion of configuration was noted. In an independent study, Michalski et al. 10 found that the oxidation of chiral phosphine sulfides with dinitrogen tetroxide proceeded with retention of configuration but with nitric acid complete inversion resulted. In the oxidation of cyclic compounds a further variation was noted; viz. when dioxaphosphorinanes were oxidized with a variety of oxidants all products formed with retention of configuration.6-8,12 Retention of configuration was also observed in the 7 oxidation of a thiophosphinate by hydrogen peroxide and MCPBA,11 and in the oxidation of a phosphonothionate by hydrogen peroxide. 12
As a result of studies on the MCPBA and peroxytrifluoroacetic acid oxidation of the diastereomeric .S,-menthyl methylphenylphosphinothioates (13) Herriot 16 proposed that two distinct reaction mechanisms were operating (Figure 3). In particular, he found that with MCPBA the oxidation proceeded with retention of configuration while with peroxytrifluoroacetic acid the product formed with inversion, and he concluded that the transformation was pH-dependent. Thus, in neutral or only weakly acidic media
(MCPBA) a three membered ring intermediate (14) was considered to be involved (path a), whereas in strongly acidic media the starting material was protonated and attacked I directly by the otherwise weakly nucleophilic oxidant (path b). Intermediate (15) would subsequently lose a proton, followed by losses of elemental sulfur and a carboxylic acid to form (16), a sequence of events consistent with the product being formed with inversion of configuration. Pseudorotation of the pentacoordinate intermediate (17) could conceivably occur but this is expected to be relatively slow since the methyl and phenyl groups should prefer to remain in the equatorial positions (Figure 3).16
The MCPBA oxidations of a wide variety of organophosphorus compounds were investigated by Bellet and Casida 17 and they considered the "phosphorus oxythionate" (18), arising from the addition of oxygen to the sulfur of the thiophosphoryl group, to be the important intermediate. Other plausible intermediates such as (19) and (20) were also suggested, as well as the five membered cyclic adduct
(21) and the diradical (22).17 R I 0 0 0 0-C-OH 6 R l)s R 's I a;-5'-..p 'p-( 0 \ o, 1/s· 'P 'p/ p Rl' ,XR.2 Rl / 'XR.2 Rl / 'xa2 ...... , / ......
(18) (19) (20) (21) (22) 8
.;o ,, o· s s [OJ U I + (path a) /i'- 4·--·--·,.,..~,
(13) (14)
l. H• (path b) ~{SI
H
@o-!.-o 0 11,ct:OH @111, .. ;_Q H C,I I 3 0 I COAr ! (15)
11 @ • .. f-o~ -S,ArCOOH liJC'0 I 0 I COAr ~ (16) (17)
Figure 3. Proposed mechanism of the oxidation of S-menthyl methylphenylphosphinothioate.16 9
Taking account of the findings of earlier workers, Bellet and Casida 17 proposed three pathways (Scheme 1) for the MCPBA oxidation of various phosphorothionates
(23).
R.R l ,R 2 = alkyl aryl etc
I MCPBA T
(b) [ R 'i/~") l R1/'XR2
(18) oxon (24) ~ R'-.~Op Rl /'SXR 2 (26) 9XR2 + eXR~ 1.oue -He / [SI R 0 • (25) 'p/'f'e + [9SXR2] / '-.:·· R1 0 cleavage producu I ,.------othercompounds pyrophosphorus compoWlds
Scheme 1. Alternative possible pathways for the MCPBA oxidation of I phosphorothionates.17 10
In each pathway the formation of the phosphorus oxythionate intermediate ( 18) was considered as the initial step and its further reactions were dependent upon the nature of the substituent groups attached to the phosphorus centre. Thus, in Scheme 1 pathway (a) desulfuration of (18) results in direct formation of the corresponding oxon
(24); in pathway (b) the phosphorothioic acid (25) is accounted for from either cleavage of the P-X bond followed by nucleophilic attack of H2O at phosphorus, or by direct attack of H2o on the phosphorus oxythionate ( 18). The cleavage products could also undergo further reactions to form phosphinothioyl derivatives (26) or pyrophosphorus compounds. If X is nitrogen or sulfur then pathway (c) appears to predominate in which intramolecular rearrangement can account for the observed phosphinothioyl derivative (26). Further degradation in this case would also lead to the corresponding phosphorothioic acid and pyrophosphorus compounds. It is noteworthy that the latter pathway had earlier been proposed by Fahmy and Fukuto 18 to explain the formation of
2,2-dimethyl-2,3-dihydrobenzofuranyl-7-N-methyl-N-(dimethoxyphosphinylthio) carbamate (27) as one of the peracid oxidation products of 2,2-dimethyl-2,3- dihydrobenzofuranyl-7-N-methyl-N-(dimethoxyphosphinothionyl) carbamate (28).
0 s 0 0 11 II II II O-C-N-P-OCH3 0-C-N- S - P-OCH3 I I I I CH30CH3 CH3 OCH3
(27) (28)
In a study of the oxidation of ( +) and (-)-O-ethyl O-(2-nitro-5-methylphenyl) N isopropyl phosphoroamidothionate (29) with rabbit microsomal fractions, Ohkawa et al. found that the reaction proceeded with retention of configuration.41 11
Similarly, when the oxidation of each enantiomer of fonofos (5) was investigated using either the mixed function oxidase system 42,43 or MCPBA 19, the oxon product (30) was also found to form with retention of configuration. However, in
contrast, the rearrangement product (12) formed predominantly with inversion of
configuration. The involvment of the three membered ring intermediate (11) was
speculated to occur as indicated previously (Figure 4). The formation of the disulfide
(12) can be rationalised by two routes; either from nucleophilic displacement of the .S, phenyl group by attack of the peracid to produce (31) followed by subsequent attack of
the ~-phenyl group on sulfur of (31) and loss of m-chlorobenzoic acid, or by rearrangement of (11). In either case the stereochemical outcome of the transformation is the same. s I ·P -@ [OJ FJ'll's Q -
(11) 0 (SJ.._ ./ ~ II 7 ¥~ a.:·r,s-@ ao l-[-,-@] (30) ~-s-@ o s~ ~P\'.''IEt 0 O& II ~-·•1tEt - C-0-0 ' • S -@ --°"---(r:;\_----:::g~• V O& Cl f;2J c,o- (12) (31) Cl
Figure 4. Alternative oxidation pathways for fonofos (5). 19 12
In the oxidation of O,O-diethyl S,-4-chlorophenyl phosphorodithioate (32) with
MCPBA, Miyamoto and Yamamoto 20 provided evidence for the formation of disulfide
(33) by carrying out the peracid oxidation in the presence of H21BO. Again, the
intermediate cyclic 5-coordinate species (34) was proposed (Figure 5). In the absence
of, or at low levels of H2O, (35) or (33) can be formed through the loss of sulfur or rearrangement respectively, but at high levels of H2O, (34) is hydrolysed to (10) and
(36). Furthermore, disulfide (33) could be formed by oxidative coupling of (10) and
I (36) and the other disulfide (37) could be formed by oxidative coupling of two
molecules of (36).
~~ (EiO)iP - S """\Q;- Cl (32)
(35)
(10) (36) · (37) I (33) +
Figure 5. Products from the oxidation of O ,O-diethyl S..-4-chlorophenyl phosphorodithioate (32).20 13
In organophosphorus compounds containing several sulfur atoms, it is the differing chemical environments which determine the relative ease of oxidation. For example, with organophosphorus compounds containing a thioether functionality, the sulfur center has been observed to undergo ready oxidation leading, sequentially, to the sulfoxide and sulfone. These products are stable and can be readily isolated. This oxidation can be effected with chemical oxidants, but also occurs both in animals, where it is presumably mediated by the mixed function oxidase system, and in plants by a photocatalytic process.44 Sulfide oxidation in plants is particularly interesting since many of the systemic insecticides contain a sulfide group, and because the insecticidal activity generally increases with increasing oxidation state of the sulfur atom, i.e., sulfide < sulfoxide < sulfone. It should be noted, however, that these metabolities are not as active anti-acetylcholinesterases as are the products from oxidative desulfuration of the thiophosphoryl group.
Sulfoxides are also relatively stable metabolically and their formation is generally faster than their further oxidation to the sulfone. The sulfoxide can be effectively translocated in plants a characteristic which, at least in part, accounts for the systemic activity. For example, vamidothion (38), is readily oxidised to the sulfoxide but further oxidation to the sulfone was not observed.45 For fensulfothion (39), oxydisulfoton
(40) and aphidan (41), the sulfoxides were more insecticidally active than the parent sulfides and they have been used as systemic insecticides.2
The oxidative metabolism discussed above has been observed with a variety of I insecticides including abate (42)46, carbophenothion (43), demeton (44), fenthion (45), phenamiphos (46), disulfoton (47), thiometon (48),2 and phorate (49).47 The effects of these transformations are not always predictable. For example, with abate (42), which undergoes oxidation of the thioether group followed by oxidative desulfuration, the insecticidal activity was found to decrease with the increase in the oxidation state of the thioether centre. This result was attributed to changes in physiochemical parameters, in particular the decrease in lip ophilicity of the oxidized metabolite.46 14
As noted above, when phosphorothioates which contain both a thioether group
and a thiophosphoryl group are oxidised, the thioether group is more readily oxidized
than the thiophosphoryl group. With fenthion (45)48 and demeton-O (50)49 the
thioether group can be oxidized to both the sulfoxide and sulfone derivatives without oxidation of the thiophosphoryl group. The phosphorothiolate linkage is also more resistant to oxidation than the thioether group. For example, Fukuto and coworkers49
observed that oxidation of demeton-S (44) with potassium permanganate gave O,O
diethyl S:,ethyl-2-sulfonylethyl phosphorothiolate (51) as the major product. Other
examples which further illustrate this order of oxidation include phorate (49)48 and
carbophenothion (43).50
o CH3 11 I (CH30)iP-S-C}½CHzSCH-C-NHCH3 II 0 (38) (39)
S 0 II t (CH3CH20)zP-S-ClJiCH2S CH2CH3
(41) (40)
(42) (43)
0 u ~ ...«\\- (CH3CH20)i-P-SCH2CH2S CH2CH3 (CH30)zP.O -Q---· SCH3 CH3 (44) (45)
s II (CH3CHzO)iP-SCH2CH2SCH2C H3
(47) · 15
s s II II (CH30)2-P-SCH2CH2S CH2CH3 (CH3CH20)2-P-S CHzS CH2CH3
(48) (49) 0 0 s II i II (CH3CHzO)i-P-S-CHzCH2S CHzC H3 (CH3CH20)2-P-OCH2CH2S CH2CII3 J. (50) 0 (51)
The oxidative chemistry of phosphorothiolates has also received substantial attention. In general the sensitivity of the sulfur atom to oxidation in compounds of this class is dependent on electron density considerations. For example, ~he thiolate sulfur atom of barium .S.-n-butyl thiophosphate (52) was noted to be oxidised very readily, and this was attributed to the high electron density of the ions.51 This explanation was also believed to apply to neutral phosphoramidothiolates such as (53) on the basis of the availability of electron density from the nitrogen atom.21
Compound (53) (methamidophos) is also of interest for another reason. It is an important insecticide which has been used in agriculture for many years. It exhibits high insecticidal activity but it is a poor anti-acetylcn:o linesterase in vitro. 52 This has raised questions about its bioactivation, a problem which has been studied by several researchers. 21,52,53
(53) 16
Eto and coworkers21 observed that the anti-acetylcholinesterase activity of methamidophos (53) increased substantially on oxidation with MCPBA. This suggested that an activated intermediate may be involved as a phosphorylating species, and the formation of Q,O-dimethyl phosphoramidate (54) on MCPBA oxidation of methamidophos (53) in the presence of methanol supported this idea. Based on such evidence these workers proposed that the initially-formed active intermediate was the sulfoxide, O,~-dimethyl phosphoramidothiolate-~-oxide (55), and that this intermediate might also rearrange to a sulfenylphosphoryl anhydride (56).21
0 0 II ~ ?i II CH30-P-NHz CH3S - P - OCH3 CH3 - S - 0 - P - OCH3 I I I OCH3 NH2 NH2
(54) (55) (56)
In an investigation of the peracid oxidation of a series ~-alkyl phosphorothiolates (57), Segall and Casida 22,23 reported a new class of phosphinyloxy sulfonates (58) whose formation was rationalized by a novel rearrangement process of the phosphorothiolate-S,.-oxide (59) (Scheme 2). For example, oxidation of the insecticide profenofos (60) with 5 equivalents of MCPBA in chloroform (Figure 6) gave the sulfonate ester (61), which could be isolated by rapid extraction of the solution with aqueous sodium bisulphite and sodium bicarbonate at O °C. Although the sulfoxide was not observed directly, its intermediacy was accepted on the basis of the formation of diethyl phosphate (62), effectively as the sole product, when the oxidation was carried out in ethanol.22 These workers also proposed that the phosphorothiolate-£ oxide was the most potent phosphorylating species known at that time.
Further oxidation studies with various alcohols as solvents showed that the outcome of the competition between rearrangement and phosphorylation is determined by the nature of the nucleophile; specifically the steric bulk of the alcohol.24 17
R'- ;?O [OJ ... /P...... _ [ Rl/p'-?OR, --[ R, P. ~-_o Rl SR2 MCPBA 9 l ...... _R2 Rl/ '-. S/.!... R2 ] (57) (59) i 0 0 '-. ,f' II R,11 II MCPBA [ R O O ] MCPBA [ R, ho ] p-O-S-R2 /p( s-R2 R1/ P'-o/ s...... _R2 Rl 0/ 1/ II R 0
(58)
Scheme 2. Proposed mechanism of formation of the phosphorus oxysulfonate. 22
CH3CH20 O '-.4 o'P'\. SCHzCHzCH3
Br Cl
(60) I
[0~ ELOH
(62) (61)
Figure 6. Alternative products from oxidation of profenofos (60) in different solvents. 18
More extensive studies on the bioactivation of phosphorothiolates were carried out by Wing and coworkers54,55 who observed that several of these esters were bioactivated by the mixed function oxidase system to more potent anticholinesterases. This study was extended to consider stereochemical factors. For profenofos ( 60) the (-) enantiomer was found to become a better inhibitor of acetylcholinesterase whereas the
( +) isomer was deactivated under the same conditions of in vitro oxidative metabolism. 54
Following these initial studies much further effort has been expended in attempting to directly observe that proposed reactive intermediate species. Direct observation of the phosphorothiolate-S.-oxide intermediate was provided by Thompson and coworkers 25 who investigated the MCPBA oxidation of (SCH3-13C)-labelled methamidophos (63) by 13C NMR spectroscopy. They noted the formation of a pair of doublets at 6 36.37 and 6 4 7 .11 in the proton decoupled Be NMR spectrum, and these resonances were attributed to the diastereomeric S.-oxides (64). They also noted that treatment of the oxidation mixture with trimethyl phosphite resulted in regeneration of methamidophos (63), a reversion consistent with the oxygen scavenging properties of the phosphite.
(MeO)JP
(63) (64)
In a further study seeking to identify reactive intermediates, Bielawski and Casida 26 investigated the oxidation of O,O-diethyl O-phenyl phosphorothionate (65), the isomeric S.-phenyl phosphorothiolate (66), and the related phosphorodithioate (67). They utilised MCPBA and 2-(phenylsulfonyl)-3-(4-nitrophenyl)oxaziridine (68) as
oxidants, and probed the reaction in a variety of solvents. They found that oxo analogs 19 were the major product of oxidation in non-hydroxylic solvents, but in primary alcohols such as methanol the major product was, in each case, diethyl methyl phosphate (69).
This arises from phosphorylation of methanol by the intermediate, but with different leaving groups being involved. A number of minor peaks were noted in the the 3lp
NMR spectrum ranging from 27 - 34 ppm upfield from the starting materials and these were assigned to transient species, including the three membered ring phosphoxathiirane
(70) from the phosphorothionate (65), and the phosphorodithioate (67) and the phosphorothiolate-S-oxide (71) from the phosphorothiolate (66).26 However, in none of these cases was definitive proof obtained for the assignments. It was clear from these studies that the reactive intermediates are sensitive to steric bulk since, in the reaction with alcohols, the reaction occurred more readily with methanol and n-propyl alcohol than with isopropyl alcohol or t-butyl alcohol. ELVO 00""'-s-@
(66)
@-so,-l'at -@-m,
(68) (69)
(70) (71)
In addition to 31p NMR studies, other spectroscopic methods including IR, UV,
Raman spectroscopy and EPR have been employed by various workers for studies in
this area. For example Field et al., 34 Swinson27 and Swinson et a[.28 used many of
them to investigate the intermediates formed in the MCPBA - mediated oxidation of a
series of phosphorothioates and phosphoroamidates. From the 31 P NMR studies the 20 life times of the intermediates were ascertained to vary from a few minutes to many hours depending on the structure of the starting compounds. The major 31 P NMR peaks for all intermediates in all oxidations appeared in the same region, about midway between the position of starting material (P = S) and products (P = 0). This was taken to indicate that all the intermediates have closely related structures and it was suggested that they are tetracoordinate and have phosphonium polysulfide structures of the type
R3PSx (72, Scheme 3) or perhaps R3POSx, UV spectra also afforded support for the polysulfide structure, and EPR spectra gave no indication of homolysis. In addition Raman, 3lp NMR and UV spectra were consistent with the long-term presence of bisphosphonium species (73) (Scheme 3).
[OJ + Biol- SSH (76)
or(72)- 2S'C (72) \25·c sulfur donor -r e.g. [R3POS] R3PSxS0(2 or 3) c:yclo - Sn+ R3PS x -n (75) R3P0,(75)erc. ~ i[O)~
(where R = various combination of O-Alkyl, O-Aryl and N(H)- Aryl)
Scheme 3. Proposed reaction pathways for the oxidation of phosphorothioates and
phosphoroamidates. 28 21
Based on this evidence, Swinson et a/.28 proposed a reaction mechanism which
is more detailed than the others previously suggested27 ,34 (Scheme 3). Thus, in
Scheme 3 the oxygenated intermediate (74) could lead to phosphorus species containing
a growing polysulfide chain (e.g. 72) which could lose elemental sulfur (75) in either
paths A and B. In path C, the oxygenated intermediate might reasonably be expected to
produce hydrodisulfide (76) as proposed in the metabolism of thionophosphorus
compounds36 in the presence of a biomacromolecule containing an SH group.
As noted earlier in this introduction, to more closely approximate physiological
conditions and to improve the peracid model for biological relevance, oxidations of
phosphorothionates in aqueous media were studied by Wu et a/.35 using the water
soluble magnesium monoperoxyphthalate (MMPP) (10) as oxidant. Under these
conditions dialkyl hydrogenphosphonates (77) was observed in up to 70% yields, a
surprising result since the phosphorus centre was reduced relative to the starting
material, in an oxidatively-induced reaction. It was further noted that
hydrogenphosphonate formation was facilitated by the presence of a good leaving group
in the starting material and a plausible mechanism was proposed as indicated in the
Scheme 4 (Rl = the leaving group).
18 1 R°" ~ S H2o P.7 [OJ [R:)> ] [R~ ( ~HS-OH ] Rif '--Rl R 'R1 Rif Rl (78) (79) i R1H
18 18 R°' '9'0 H20 R°' '9'0 [0] R°'-. 0 0 • [ 18 ] R6'- H -H2S04 Rif"P(S_ OH Rif;s-oHII (TT) 0 (80)
Scheme 4. Formation of hydrogenphosphonate from the oxidation of . phosphorothionates. 35 22
In this scheme, the initial oxidation product, the phosphoxathiirane (78) is attacked by water yielding the five-coordinate intermediate (79) which subsequently forms the phosphoro(thioperoxoic) acid (80) by cleavage of the leaving group. In support of this suggestion, quantitative incorporation of 180 was observed into the hydrogenphosphonate when the oxidation was carried out in H2180. The thioperoxoic acid can be oxidized further and its hydrolysis yields the hydrogenphosphonate (77).
The key intermediate in this proposal, the thioperoxoic acid (80), was first observed by
Segall et al. 32 directly by 3tp NMR in the oxidation of phosphorothioic acid (81) in ethanol (Scheme 5). In this case (82) and its isomer (83) were the major initial products observed. Hydrogenphosphonate (84) was also noted as a final product, but it occurred only in minor amounts. The major products were the ethyl ester (85) and phosphoric acid (86).
Rl 1 '-?o R ,.,,,, OH P. R20/ 'sH . R20/~S
(81) MCPBA j[OJ l-[O)
1 [OJ [ R' 'eo ] ....._ [ R1 'I OH ] [ R .___ P.,<;O :;;' 'eo ~ R2o/ 's~H R2o/ 'soH --- 2 r· J - R2o/~l - R :o/ 'osH ] (83) (88) (82) (87)
<;H50H !C:C,H,OH . CzH50S03H ~S03H-+ S02 + H20 ~[SJ
R,1 ,~o Rl ,~o p, Rl ,?o P. R2o/ 'ff P. R2a/ 'oH R20/ H · 'ex: 2 5 (84) (86) (85)
Scheme 5. Proposed reaction mechanism for the oxidation of phosphorothioic acid (81).32 23
In the scheme proposed by Segall et al. 32 it is noteworthy that intermediates with sulfurylating as well as phosphorylating reactivity were observed. Thus, rearrangement of phosphoroxathiirane (87) may lead either to starting material (81) by loss of oxygen, or to the oxysulfonate (83) which in turn yields phosphoric acid (86) and elemental sulfur. The isomeric thioperoxoic acid (82) is also available via (87) and can be further oxidised to intermediate (88). This in turn yields the ethyl ester (85) from reaction with ethanol at the phosphorus center, whereas analogous reaction at the sulfur center gives the hydrogenphosphonate (84).
In contrast to the formation of hydrogenphosphonates from oxidation of phosphorothionates in aqueous media, oxidation of phosphorothiolates under the same conditions produced only the hydrolysed product. This is consistent with the reports from other workers. For example, Yang et al. 33 reported the detoxification of the nerve agent Q.-ethyl £-[2-( diisopropy lamino )ethyl] me thy lphosphonothiolate(89) by oxidatively-induced hydrolysis (Figure 7).
0 0 0 II [0] II I EtO - P - SCH2CH2N(i-C3H7)i .. EtO - P - SCHzCHzN(i-C3H7)2 I I + CH3 CH3 (89) i (90)
0 0 0 0 0 II I I I II 3 (OJ HOS0 Cl½CH2N (i-C3H7)i + ELO - P - OH EtO- P- SCH2CH-,N (i-C3H7)i 2 I + -+ + I H02 CH3 CH3 (91) (92)
Figure 7. Proposed reaction mechanism for the oxidation of 0-ethyl £-[2-
(diisopropylamino)ethyl] methylphosphonothiolate (89). 24
In their Scheme, with MCPBA the thiolate (89) was first converted to the N oxide (90). This subsequently underwent oxidation at sulfur to give (91), a compound undergoing immediate hydrolytic cleavage of the P-S bond to produce the non-toxic ethyl methylphosphonic acid (92). These workers concluded that the hydrolysis of the thiolate (89) occurred only after the sulfur was oxidized and the alternative mechanism, in which oxidation of the sulfur is preceded by hydrolysis at the P-S bond, was ruled out.
Bicyclophosphates (BPs) (93) with suitable R substituents are highly toxic compounds to mammals56,57 which act as noncompetitive y-aminobutyric acid antagonists.57-59 They are also potential occupational hazards.56,60-62 A number of
BPs have been synthesized for structure/activity studies57,63,64 and the most toxic analogs are t-butyl bicyclophosphate (R = t - C4H9,q3) and its thiono analog, t-butyl bicyclophosphorothionate (94).63,65,66 The LD50 of injected t-butyl bicyclophosphate is 0.036 mg/kg for mice compared with 10 mg/kg for American cockroaches 66 and 32 mg/kg for adult houseflies.67
Monocyclophosphates such as (95) are of biological interest since they may be considered as protoxicants. Although they are not toxic themselves they have the potential to be converted to toxic derivatives, viz. to the BPs (93), in vivo. Toia and Casida29 found that (95) with R' = 0-aryl or .S.-alkyl were converted to BPs (93) in vivo, or on MCPBA oxidation. In the latter situation with R' = .S.-alkyl (95) almost quantitative formation of the BP (93) was obtained depending on the starting isomer.
(93) (94) 25
In further studies Toia and Casida30 carried out the oxidation of several 4-alkyl
monocyclophosphates by MCPBA in different solvents and found that the behaviour of
the individual stereoisomer is solvent dependent. Thus, when either isomer was
oxidized in chloroform (ethanol free) the corresponding BP (93) was obtained in
quantitative yield. However, in methanol, the isomer (96) found to be more toxic in I mice yielded the BP (93) quantitatively, whereas the other (less toxic) isomer (97) gave
the methyl phosphate (98). These data were used in the initial stereochemical
assignment of the isomers, and these, along with the stereochemical details of the
transformations, are presented in Scheme 6.
R 0-P_,.sit' [OJ 8 /~o a-G'P-O Cll30H OH (93) (96) o-p~~R ¾ Scheme 6. MCPBA oxidation products of monocyclophosphorothiolates.30 Abdullah 31 observed analogous results in the oxidation of each isomer of .S. methyl monocyclophosphorothiolate (R = CH3) and he also applied the reaction to the preparation of salioxon (99) by oxidation of MTBO ( 100) in methanol. However in this case the reaction lacked stereospecificity since when optically active MTBO (100) was used as starting material only racemic salioxon (99) was obtained. 26 0 MCPBA ©(j-OCll3 MeOH MTBO (100) salioxon (99) Although the "Introduction" of this thesis is concerned primarily with the oxidative transformation of organophosphates, hydrolysis reactions are in many ways linked and their importance warrants comment. For example, the hydrolysis of organophosphorus pesticides occmswidely in biosystems and in the environment and the process generally results in their detoxification. Therefore, as a generalisation, the biological activity of a compound is also related to the ease with which it is hydrolysed. Most of the toxic organophosphorus esters can be detoxified quickly by hydrolysis in an alkaline solution 68,69 a reaction which involves nucleophilic attack of hydroxide ion at the phosphorus centre. Phosphate esters are generally hydrolysed 2 to 20 times faster than the corresponding phosphorothionates. This is because oxygen is more electronegative than sulfur thereby making the compound more reactive by creating more positive charge on phosphorus. For example, Ruzika et al. 70 studied the hydrolysis of many pesticides by GC and reported that in ethanol at 70°C, the half life of paraoxon (6) is 28.0 hours and the thionazin-O-analog (101) is 8.2 hours while that of parathion (1) is 43.0 hours and that of thionazin (102) is 29.2 hours. The substituent attached to phosphorus also influences the rate of hydrolysis. Methyl esters are generally more unstable than ethyl and higher alkyl esters because the alkyl groups release electron density to phosphorus by the inductive effect (methyl < ethyl< propyl). For example, the half life of demeton-S-methyl (103) is 7.6 hr and 27 parathion-methyl (104) is 8.4 hr whereas demeton-S (44) is 18.0 hr and parathion (1) is 43.0 hr.70 (101) 0 s II II ~ (CH30)2 PSC2H4S---C2H5 (CHJO)i-P-O~N02 (103) (104) Generally, phosphorothiolate esters are much more reactive than the corresponding phosphate esters even though the electronegativity of sulfur is lower than oxygen. This is rationalized by the following considerations : (i) the polarizability of the sulfur atom,71 (ii) the reduced p1t-d1t contribution in the P-S bond,72 (iii) the lower bond strength of the P-S bond,73,74 Thus, Murdock et al. 15 reported that the S-aryl thiolate isomer of paraoxon is hydrolyzed 22 times faster than paraoxon in 0.0lN NaOH solution, and Hudson et al. I 73 observed that diisopropyl methylphosphonodithiolate (105) is hydrolyzed 25,000 times faster than the corresponding oxygen ester in alkali. However, Heath 71 reported that Q,S,-diethyl O-p-nitrophenyl phosphorothiolate (106), which is an impurity in the technical preparation of parathion, is hydrolyzed at the p-nitrophenyl ester linkage 479 times faster than paraoxon. This suggested that if there is a more acidic substituent than the thiol group in the molecule then it is the more acidic linkage which is hydrolyzed. 28 (105) (107) For some phosphorus esters with a hetero atom at the ~-position of an 0- or£ alkyl ester group, the £-alkyl analog is hydrolyzed more slowly than O-alkyl analog. For example, Fukuto et al .49 observed that demeton-S (44) is hydrolysed at half the rate of the phosphate analog of demeton (50), and amiton (107) is more slowly hydrolyzed than even its thiono isomers.76 This variation was attributed to a different hydrolysis mechanism involving an intramolecular process as follows:. + ,. X Y-R II /" OH ~ · (RO)1 PO- + CH2 - CH2 ~ X=OorS Y=SorNR For .S.-alkylthio (or amino)-ethyl phosphorothiolates, in contrast, the hydrolysis usually occurs by P-S bond cleavage i.e., the C-S bond is more stable than the P-S bond as indicated below. ,S_-Alkyl groups of phosphorothiolates are such poor electrophilic alkylating agent that they are some hundred times less active towards nucleophiles than O-alkyl groups of corresponding phosphate esters.77 29 0 , (ROh P" - S - CH2CH2 - Y-R + NaOH ... Cyclic compounds are generally much more reactive than the acyclic analogs because of the ring strain which results in a lowered p7t - d7t contribution in the P-0 bonds and consequently makes the phosphorus atom more electropositive.78,79 This is illustrated in Figure 8 for the 5-membered ring phosphate versus the acyclic analog. Thus, the hydrolysis of five membered ring phosphates is much faster than corresponding acyclic compounds (Table t 80-83). CH2...... __ / --....., -.39 o -.60 Cllz r1.15 1.50 ".. p+ .34 1 +.30 :----,.__ CH3 -.40 O . I ~l.l6/CH3 "'~ CH3"-. I 1.54 o...... 01Y\\ "-0 / 0 -.33 -.36 -.50 O-CH3 (Signed numbers : net atomic charges; unsigned numbers: overlap populations) Figure 8. Electronic structures of cyclic and acyclic phosphates.78 The enhanced rate of reaction also extends to six-membered ring compounds. For example, Eto et al. 84 found that cyclic phosphate salioxon (99) is hydrolyzed, by fission of the endocyclic aryl ester bond, as shown below ,much faster than paraoxon. 0 O HO II 0 CH30-P::o/ II I I + CH30- P-0 -o~ 0 ~ I - OH (99) 30 Table 1. Relative Rates of Hydrolysis of Cyclic Phosphorus Esters. Diesters k(rel) Neutral esters Ring opening Hydrolysis k(rel) of exocyclic ester, k(rel) #o [o\~ 108 [o\~ #o 8 10 -106 cfp'\.OH cfP '-ocH 3 \0 \~0 ~ 3-10 c#o 106 -1 -cfp'\.OH P'-oc 2H5 0 II 1 (CH30)JP=O 1 1 (CH30)2POH Besides ring opening, the hydrolysis of the exocyclic ester linkage is also greatly enhanced in five membered ring phosphates (Table 1), and Dennis and Westheimer85 explained this in terms of "pseudorotation" of the trigonal bipyramidal intermediate (Figure 9). Thus, the nucleophile approaches the phosphate ester on the back side of one of the endocyclic P-O bonds. This and the newly formed bonds are apical bonds, and the other P-OC bond and P-O- (formerly phosphoryl) became the basal bonds of a trigonal bipyramidal intermediate. The intermediate undergoes pseudorotation easily, provided the phosphoryl oxygen serves as "pivot" remaining in a basal position. The apical oxygen prefers to be protonated, and then move away from phosphorus. 31 For the phosphonate, in contrast to the phosphate, such an enhancement is not observed (Table 1). This is presumably because it is too energetically unfavourable for a carbon atom to occupy an apical position and consequently the pseudorotation is restricted. ro o I (,t. ,~,,_-:;- OCH3 pseudo O --'P -OCH3 CtO~·· OCH3 -- %~.1-- --rotation 0 !-~0H2 Q I 0 + I "OH2 OH2 Jt H + rOH 0 1} (,t, ~- ,~-P--:;- OCH3 O---'P-OCH3 i~1' 0 l.~.. - 0 H OH, OJ H l ~ [OH Fl (? ...... -:P 0-P O ~'OH ~'OH 0 0 ring opening exocyclic ester hydrolysis Figure 9. Pseudorotation of the trigonal bipyramidal intermediate.2 32 Further variations are also noted. For example Fukuto et al .86 observed that hydrolysis of the exocyclic ester occurs preferentially to ring opening in the ethylene cyclic phosphate of p-nitrophenol (108) as illustrated below: ,-roo~P-0 N02 CI (108) Accelerated hydrolysis rates have also been reported for the bicyclic phosphate (BPO) (109) by Aksnes and Bergesen87, and Gorenstein and coworkers have shown that part of this rate enhancement in cyclic compounds is probably due to the stereoelectronic effect. 88-94 The stereoelectronic effect in this reaction is believed to be due to T\o (oxygen lone pairs) H 6* P-O (P-O anti bonding orbital) orbital mixing which facilitates P-O bond cleavage or formation. Fanni et al. 95 found that the rate of hydrolysis of BPO (109) is 5.2 x 103 times that of the acyclic triethyl phosphate (110) while the BPS (bicyclicphosphorothionate, 111) is 8.1 x 102 times that of acyclic triethyl phosphorothionate (112). This rate enhancement was explained in terms of stereoelectronic effects in the transition state intermediates. x-o (109) X=S(111) x-o (110) x-s (112) 33 In Figure 10, for the bicyclic compound (A) the two equatorial endocyclic oxygen centers have lone electron pairs which are approximately antiperiplanar to the breaking apical P-O bond. This interaction is suggested to be mainly responsible for the enhancement of the rate of cleavage of this bond relative to its acyclic analogues triethyl phosphate (110) and phosphorothionate (112) since in this cases the freezing of one conformation is required to place two lone electron pairs antiperiplanar to the breaking P-O bond (B) and this conformational restriction will be entropically disfavoured.96--98 (B) (A) Figure 10. Stereoelectronic effects in bicyclic and acyclic compounds.95 Ozoe et al. 99 studied the rate constants for the hydrolysis of several bicyclic phosphates related to their monocyclic compounds and found that the thiono ester was approximately 9 times as stable as the corresponding phosphate in alkaline solution. 0-P~#Y x8 tx = Et, t-Bu, Bu) --... 0/ ~OH f½O OH \y =0, S The relatively high reactivity of the phosphate ester is explained in terms of (i) the high electronegativity of the phosphoryl oxygen and (ii) the strain in the characteristic "cage" structure (The POC bond angle= l 15"Cl00 ). 34 In their solvolysis studies of bicyclic compounds, these workers99 also observed that BPO (109) undergoes a two-step methanolysis. However, the the. stereochemistry o~monocyclic intermediate was not assigned. MeOH HOr /OMc 0-P=o HO "OMc (109) In overview, many facets of the oxidation and hydrolytic chemistry of various classes of phosphorus-containing esters has been established, but numerous other questions remain to be answered. In the present work, the oxidation of bi- and monocyclic phosphorothioates has been studied to investigate the stereochemical changes which occur during the oxidatively-initiated displacement reactions. Structural ellucidation of some of the reaction intermediates which may occur and which are potent phosphorylating agents is also probed using various spectroscopic methods. The thesis further considers non-oxidative displacement reactions for a better understanding of reaction mechanisms, and also comparesthe oxidation reactions of simple cyclic and acyclic phosphorothioates. RESULTS AND DISCUSSION* *In a number of the oxidation reactions spectra were taken at different times such that the changes in the reaction mixture could be observed (The changes are described in the appropriate experimental section.) and the term " intermediate " is used for a compound which appears and disappears during the period for which the reaction was monitored. For convenience oxidations in CH30H or CD30D are generally discussed in same sense, except where otherwise specified. 35 Syntheses of Cyclic phosphorothioates and related compounds. Substituted 1,3,2-dioxaphosphorinanes have been employed by a number of workers to study the stereochemistry of substitution at phosphorus.101 •114 These compounds have distinctive chemical shifts ( 1H and 3 lP) with 31P- 1H coupling constants which often enable the conformation of these systems to be established, at least to a degree of certainty. Thus, whether inversion or retention of configuration occurs at the phosphorus centre can quite easily be determined. For example, 2- substituted-5-chloromethyl-5-methyl-2-oxo-1,3,2-dioxaphosphorinane (113) is conformationally immobile and the cis- and trans- isomers are easily distinguished by the characteristic chemical shifts of the methyl and chloromethyl protons l09,l lO,l l2. Furthermore, a number of these compounds are crystalline and their crystal structures I may also be determined. 115 (113) As a result of investigations using these types of compounds it was established that for reaction at the phosphorus centre, retention of configuration at phosphorus occurred in the presence of electron withdrawing ligands, or a lewis acid. 101,112,114,116 This was rationalized in terms of the increase in the electropositivity of phosphorus, which increases the possibility of back bonding between the attacking nucleophile and the phosphorus thereby leading to a trigonal-bipyramidal intermediate (114). Hence, retention was observed in the presence of nucleophiles such as the alkoxide ion. On the other hand, inversion of configuration was observed in cases where the leaving group was weakly bonded to the phosphorus centre, i.e. for a good leaving group such as chloride ion, or when nucleophiles are involved which are poor back bonders. In this 36 situation an intermediate such as (115) would result, and inversion of configuration occurs. o· 0 RO,,,,, I+ 11 P-X Nu---······ p ·-······· X R0,,,,-1 _.. ·· \ Nu RO OR (114) (115) Nu ; Nucleophile X ; Leaving group In the present work, to study the stereochemistry of phosphorothionate oxidation-displacement reactions several selected bicyclic compounds and some monocyclic 1,3,2-dioxaphosphorinanes were synthesized, and their structures and stereochemistry unambiguously assigned. 5-Hydroxymethyl-5-methyl-2-thiono-2- ethoxy-1,3,2-dioxaphosphorinane (116) and BPS were prepared by reaction of 2- methyl-2-hydroxymethylpropane-1,3-diol with the corresponding dichlorophosphorothionate or trichlorophosphorothionate, respectively. As reference compounds, BPO, 5-hydroxymethyl-5-methyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinane (117) and 5-hydroxymethyl-5-methyl-2-oxo-2-methoxy-1,3,2- dioxaphosphorionane (118) were also synthesized by the reaction of the appropriate trichloro- and dichlorophosphates and 2-methyl-2-hydroxymethylpropane-1,3-diol. All the reactions were carried out under dry nitrogen or argon in the presence of dry pyridine (Scheme 7). These reactions proceeded smoothly to give the products in reasonable yield and, as expected, the crude dioxaphosphorinanes consisted of a pair of geometric stereoisomers, which could readily be evidenced by NMR spectroscopy (Figure 11). Thus, the 5-methyl protons and 5-hydroxymethyl protons appear as sharp singlets for each of the two isomers in the 1H-NMR spectrum, and the 3 l P NMR 37 specrum gave singlets for each of the isomers, as anticipated. The integral (1H NMR) showed that these isomers were present in an approximately 1: 1.2 - 1.5 molar ratio, with an excess of the more polar isomer as defined by relative Rf values on silica gel tlc (the upper band was designated as the less polar isomer and the lower band was designated as the more polar isomer). The 3lp NMR resonance of the more polar isomer always appeared at higher field than the less polar isomer. ~~P=O ~0/ (BPO) ~Q Cl3P=O"" OH OH OH ~Q Cl, P _, OEt"' Cl,... :.::.o "' (118) Scheme 7. Synthesis of cyclic phosphorothionates. 38 More polar isomer - Less polar isomer ------.. I I I I • I I I • I ' I I I I I I I I 4.2 3.6 3.0 2.4 1.8 1. z • 6 PPM More polar isomer J Less polar isomer \ 60 40 20 0 -20 Figure 11. Spectra (1H and 3Ip) of crude 5-hydroxymethyl-5-methyl-2-oxo-2-ethoxy 1,3,2-dioxaphosphorinane ( 117). 39 The formation of bicyclic compounds (BPO or BPS) as by-products was also noticed in the preparation of the dioxaphosphorinanes, probably via the transesterification as shown below. These bicyclic esters are evidenced by the occurrance of a distinctive doublet from their ring protons at around 4.5 ppm (lH-NMR) [The dioxaphosphorinane was evidenced by the appearance of a multiplet in the region of 3 - 5 ppm (1H NMR) which are the resonances characteristic of the dioxaphosphorinane ring protons]. Pyridine was chosen as the base in these condensation reactions rather than triethylamine because the latter compound has resonances in the 1H NMR spectral region of interest (i. e., 3 - 5 ppm). The geometric isomers of the crude dioxaphosphorinanes from each synthesis were separated by preparative tic using a modification of the system previously described for (118).117. Since the isomers in each case have very similar Rf values, even in the optimum solvent system found for separation, each plate required as many as 7 developments. Apart from the limited recovery of the products, the major problem associated with multiple development of the plates is the resultant diffusion of the product bands. Thus, in the present case a little overlapping of the two bands of interest was always observed, and even at the extremities of the plate the separation between the bands was very small. A funher complication arose since the compounds could not be visualised on the plate under the U.V. light. This meant that the progress or the result of the separation between or after successive developments of the plates was difficult to 40 monitor. A further concern was the possibility that the phosphates would undergo intramolecular transesterification (shown above) on the silica to form BPO (or BPS) and for this reason separations and recoveries were effected as rapidly as possible. The above difficulties not withstanding, the method was satisfactory and the separated isomers were crystallized as colourless needles or plates, generally from acetone and light petrol (b.p., 60-80 °C). One exception was the less polar isomer of (116); this could not be crystallized and remained as a sticky colourless oil. Since (116) was the starting material for the oxidation study its acetate (119), benzoate (120), tosylate (121) and 3,5-dinitrobenzoate (122) derivatives were prepared in attempt to obtain crystals suitable for single crystal X-ray analysis such that the actual stereochemistry of each isomer would be known. Derivatization was readily achieved by reacting ( 116) with an excess of the appropriate reagents in the presence of pyridine. C No difficulties were encountered in any of these esterifications and the refions proceeded in reasonable yields (Scheme 8). The geometric isomers of the acetate (119) were readily separated by column chromatography. Other separations (for 120,121 and 122) were achieved using preparative tic with multiple developments. The advantage of the benzoate (120), tosylate (121) and 3,5-dinitrobenzoate (122) was that the separation of isomers during preparative tic could be monitored between successive developments of the plate by visualising the plate under U.V. light. The separation of the isomers of (122) was found to be more difficult due to tailing of the more polar isomer. The separated isomers of (119), (120), (121) and (122) were crystallized from acetone and light petrol (b.p. 60-80°C) as colourless needles or plates, except for the less polar isomers of (120) and (121) which were obtained as amorphous solids. Of these, less polar isomer of (119) and more polar isomer of (122) were suitable for X-ray analysis, and these results is discussed later. 41 /OEt o-p~ / s 0 0 (120) 0 I CH3c_ O CH3C"" b (122) (121) Scheme 8. Derivatization of (116). Structural assigment of the isomers of dioxaphosphorinanes The 1H-NMR spectra of the dioxaphosphorinanes (116, 117 and 118) had some characteristic feactures which proved useful in discerning between the isomers of each compound; in particular, the chemical shift of the protons attached to substituents at the 5-position of the dioxaphosphorinane ring.109,110,112 42 The common feacture in the lH-NMR spectra of all the compounds examined was the sharp singlet corresponding to the methyl protons of the 5-methyl group (Figure 11). The more polar isomers of (116), (117) and (118) gave this singlet at 0.2 - 0.4 ppm to higher field than the corresponding singlet of the less polar isomer. The 5- hydroxymethyl group provided a similar marker; the sharp singlet of the methylene protons of the more polar isomer was observed at 0.2 - 0.4 ppm to lower field than the corresponding peak for the less polar isomer. For assigning structure and conformation, the axial and equatorial ring protons which absorbed in the 3 - 5 ppm region (1H NMR) proved to be most useful. These an generally appeared atABX system by virture of their coupling to phosphorus (Figure 12). Assignment of the chemical shift to the axial and equatorial protons in the 1H NMR spectrum of each compound was based on the observed coupling values of these protons to phosphorus (JPOCH), and on anisotropy arguments. Axial protons Equatorial protons 4.5 4.3 4. 1 3.9 PPM Figure 12. Spectrum of ring proton region of less polar isomer of (117). 43 The Karplus relationship between JPOCH and dihedral angle (0) between the protons and the phosphorus atom predicts that JPOCH approaches a minimum as 0 approaches 90°, and JPOCH reaches a maximum as 0 approaches 0° or 180°.118 Thus, in a six membered ring, the coupling between an axial proton and phosphorus will be substantially smaller than the coupling of phosphorus to an equatorial proton (Figure 13). The relative position of axial and equatorial protons and their chemical shift differences were found to be very useful to assign cis- or trans- stereochemistry to the isomers of each compound. Generally, the phosphoryl ~lithas a strong tendency for the equatorial position. 119•121 Hence, the structural difference between the isomers is the position of the hydroxymethyl group relative to the 2-alkoxy group. Therefore, for the purposes of this study a molecule is defined as having cis-stereochemistry when the 2-alkoxy group and the hydroxymethyl group are on the same side of the dioxaphosphorinane ring, and trans-stereochemistry when they are on opposite sides of the ring (Figure 14 ). Ha Ha ... --· .... , J POCHc \ \ I I I I I He He J POCHa < J POCHc Figure 13. Relationship between dihedral angle and POCH-coupling constants. 44 Trans -isomer Cis-isomer Figure 14. Definition of stereochemical tenninology as used in this study Molecular models clearly showed that the distance between the hydroxyl group and the axial protons in the cis-isomer is substantially smaller than the distance between the hydroxyl group and the equatorial protons in the trans-isomer. As a result, the deshielding effect of the hydroxyl group on the axial protons of the cis-isomer should be greater than for the equatorial protons of the trans-isomer. Thus, the axial protons are expected to resonate further downfield than the equatorial protons, as observed for the less polar isomers of ( 117) and ( 118) (Figure 12). On the basis of the above considerations and 1H-31P coupling values, when the chemical shift differences were distinct the less polar isomers of the dioxaphosphorinanes were assigned the cis-stereochemistry However, some exceptions were observed based on the coupling constants. For example, for the less polar isomer of (116) the aniosotropic effect was insufficient to shift the axial protons to lower field than equatorial protons in CDCl3 solution (Figure 15-A), but interestingly, in CD30D solution the deshielding effect was enhanced greatly and the situation was reversed (Figure 15-B). Presumably this must be due to the deshielding effect of methanol which may aggregate around the hydroxyl group of the molecule of interest by hydrogen bonding. To confirm the above argu ments and assignt.~~nts, the coupling constants and chemical shifts were calculated from the spectrum oiless polar (cis-) isomer of (116) in 45 CD3OD (Table 2), and the spectrum was then simulated. The result (Figure 15-C) clearly demonstrated that the calculations and argu ments are in agreement for the assigned stereochemistry. Table 2. Chemical shifts (6, ppm) and ]-values (Hz) of the less polar isomer of (116) in CD3OD (measured). Axial protons (Ha) Equatorial protons (He) 6 JporHa lHaHe 6 ]p()("He JHaHe 4.54 10.89 -11.0 4.22 15.96 -11.0 Since the equatorial protons generally absorb at lower field than the axial protons a ixsix membered ring, 122 and the chemical shift difference between the axial and equatorial protons in the more polar isomer was found to be very small, it is more likely that the hydroxyl group has very little or no deshielding effect on the equatorial protons in the more polar isomer, i.e. the isomer with the trans-stereochemistry. Trans-(116) had a very small chemical shift difference between the axial and equatorial protons to give a so called "deceptively-simple ABX" spectrum. This has been reponed with other n dioxaphosphoritnes 123•124 (Figure 16). In the cis isomer (less polar isomer) the chemical shift difference between the axial and equatorial protons was larger and the axial protons showed absorbance at lower field than the equatorial protons. With the cis- and trans-isomers assigned on the basis of the chemical shifts and coupling constants of their ring protns, the singlets from the 5-substituent can be used, in turn, as a simple and absolute marker for distinguishing between the cis- and trans- 46 isomers of the dioxaphosphorinanes; the singlet for the 5-methyl protons of the trans isomer occurs at higher field than the singlet for the 5-methyl group of the cis-isomer. The 3lp NMR spectrum also presented a clear picture; the trans-isomer of the pair always resonated at higher field. Axial protons: co) , Equatorial protons: (x> X O X. 0 (A) (B) 0 X )(. X 0 0 )( )( 0 ' 4.3 ... 4.2 4.1 4.5 4.4 4.3 4.2 0 (C) 0 ')( I( 0 )( l( ____;V ------..Ju "' --- Figure 15. Spectra (1H) of cis-(116) in CDCl3 ( spectum A), in CD30D ( spectrum B ), and the simulated spectrum ( spectrum C ). 47 Ring protons: (A) A - 4.3 4.2 4.1 Figure 16. Deceptively-simple ABX spectrum of trans -(116) ( ring proton region) The assignments of these cis- and trans-isomers were subsequently confirmed by X-ray crystallographic analysis of the less polar isomer of the acetate (119) and of the more polar isomer of the 3,5-dinitrobenzoate (122) (Figure 17 and 18, respectively). The less polar isomer of the acetate (119) was found to have the acet yl group and the ethoxy group on the same side of the ring, i.e. defined as cis-geometry (Figure 17), and the more polar isomer of the 3,5-dinitrobenzoate (122) (Figure 18) showed the ethoxy group and 3,5-dinitrobenzo1I group to be trans. In both cases, the thiophosphoryl ; 1~~as found to be in an equatorial position. These results were in accord with the assignments made previously to the phosphorothiolates on the basis of their different toxicities, 125 and which was also confirmed by X-ray analysis of more polar isomer of 2-thioethyl-2-oxo-5-hydroxymethy 1-5-isopropyl-l ,3,2-dioxaphosphorinane ( 123) and less polar isomer of 2-(p-nitrophenoxy)-2-oxo-5-hydroxymethyl-5-methyl-1,3,2- dioxaphosphorinane (124). 126 48 N02 \ 1 SEt - 0-P~ io / ~o 0-P~ 0 / ~o 0 (123) (124) Figure 17. X-ray crystal structure and numbering of atoms of the cis-isomer of 5- acetoxymethyl-5-methyl-2-ethoxy-2-thiono-1,3,2-dioxaphosphorinane (119). (X-ray data are listed in appendix A). 49 0(8) Figure 18. X-ray crystal structure and numbering of atoms of the trans-isomer of 5- (3 ,5-dinitrobenzoxy )me thy1-5-meth y1-2-ethoxy-2-thiono- l ,3,2-dioxaphosphorinanes (122). (X-ray data are listed in appendix B). 50 In addition to the above mentioned compounds some simple acyclic and 5,5- dimethyl-2-alkoxy-2-thiono-1,3,2-dioxaphosphorinanes and their oxo analogues were also prepared as model systems for study, as follows (Scheme 9,10): (EtO)JP: + S (EtO)JP=S (112) (MeO)JP: + s (MeO)JP=S (125) (EtO) p ... o Et3N • (EtO) P... o (69) 2,a + MeOH 2 'oMe Ao 0 Et N. EtO-P~ + MeOH 3 EtO-P~ (126) 'c12 (MeO)i (EtO) p,-,O + H2o (EtO)iP:'.:gH (127) 2 'c1 s EtOH + :PC1 (EtO)iP:'.:~ (EtO)iP:'.:iH 3 Et3N (128) (10) Et3N j PhSSPh (EtO)iP:'.:~Ph (129) 0 0 II 11 2(Et0)3P=S + S02Cl2 (EtO)iP-S-S-P(EtO)i (112) (130) 0 0 II 11 2(Me0)JP=S + S02Cl2 --- (MeO)iP-S-S-P(MeO)i (125) (131) Scheme 9. Synthesis of acyclic phosphorothioates and related compounds 51 Trialkyl phosphorothionates (112) and (125) were prepared in high yield by the addition of sulfur to the corresponding phosphites. O,Q-diethyl Q-methyl phosphate (69) and Q,Q-dimethyl Q-ethyl phosphate (126) were synthesized by the reaction of methanol with the corresponding monochlorophosphate and dichlorophosphate without any difficulties, and Q,Q-diethyl phosphate (127) was obtained from simple hydrolysis of diethyl chlorophosphate. The reaction between ethanol and phosphorus trichloride yielded Q,Q -diethyl hydrogenphosphonate (128), which gave diethyl phosphorothioric acid (10) by addition of sulfur, or Q,Q-diethyl S-phenyl phosphate (129) on addition of diphenyldisulphide. The bis-phosphoryl disulfides (130) and (131) were prepared by the reaction of S02Cl2 with the trialkyl phosphorothionates (112) and (125). 5,5-dimethyl dioxaphosphorinanes (132), (133) and (134) were prepared by the reaction of 2,2-dimethyl-1,3-propanediol and the corresponding dichlorophosphate or dichlorophosphorothionate or phosphorus oxychloride in the presence of pyridine. Hydrolysis of 5,5-dimethyl-2-chloro-2-oxo-dioxaphosphorinane (134) by excess water yielded the corresponding acid (135), while the reaction with water and pyridine gave p_yrophosphate the hit , · 1 (136). 5,5-dimethyl-2-oxo-2-methoxy-1,3,2-dioxaphosphorinane (137) was obtained by the reaction of methanol and (134). The reaction of phosphorus trichloride, ethanol and 2,2-dimethyl-1,3-propanediol yielded 5,5-dimethyl-2-oxo-2- hydro-1,3,2-dioxaphosphorinane (138), which in turn gave the £-phenyl derivative (139) on addition of diphenyldisulfide. All of the reactions above proceeded smoothly and no particular difficulties were encountered. One interesting point was noticed during the synthesis and characterization of the hydrogenphosphonates (128) and (138) which was related to their chemical shifts and J-values (31 P). Generally, hydrogenphosphates have a large coupling between phosphorus and proton (P-H), a feature considered as the most characteristic marker for these compounds. It was found that althought the Jp_H values were relatively similar, the chemical shift values varied with the solvent system but did not correlate with the 52 particular solvent used (Table 3). Some literature127 values for (138) are also available for comparison, e.g. in aqueous ethanol 6 -8 with JP-H = 690 Hz.and in propanol 6 -9 and -5 with JP-H = 670 Hz and 660 Hz. Caution must therefore be exercised using chemical shift alone for structural assignment . o' ,. OEt )C ,P~s O (133) a, P,.oa: (u a- :::::s N v-OH P03, EtOH )CO, ,. H ----- oP:::.:o /\_OH (138) (132) ,.SPh q ,.OH o' ,.a O, )C ,P~o ------)C ,P~o )C ,P~o 0 0 0 (134) (139) (135) "o~o q /OJ< )C P-0-//0 P-.. I II " 0 o 0 (137) (136) Scheme 10. Synthesis of 5,5-dimethyl cyclic phosphorothioates and related compounds. 53 As discussed earlier, the axial and equatorial protons of the dioxaphosphorinanes generally appeared as an ABX system by virture of their coupling to phosphorus in addition to their coupling to each other. In some cases, however, the appearance of these signals was not straightforward. For example, the spectrum of the trans-isomer of (118) (i.e. trans-118) showed unusual multiplicities in which each peak of the equatorial protons appeared as a doublet of doublets of triplets while the axial protons appear as a doublet of quartets. These arise in part, because of the smaller coupling of the axial protons to phosphorus than the equatorial protons (Figure 19). Such multiplicities were also observed with (132), (137) and (139), and these features have also been reported by other re searchers 117' 123 with different dioxaphosphorinanes but without detailed investigation. Table 3. Chemical shifts (6, ppm) and J-values (Hz) for hydrogenphosphonates in CH2Cl2 and aqueous acetone. Compound (128) Compound (138) CH2Cl2 Aq. acetone CH2Cl2 Aq. acetone 6 JpH 6 JpH 6 Jptt JPOCH 6 Jpff JPOCH 7.8 691 10.8 717 3.2 669 22 6.8 694 21 54 Axial protons Equatorial protons /OMe --Pi I ~o 0 OH 4 • .s 4.2 4, I ,.o Figure 19. 1H NMR spectrum of trans-(118): ring proton region. Clear results were not obtained from decoupling of .each of the axial and Shlft equatorial protons, presumably because of the small chemicalffferences which make it difficult to achieve selective irradiation. Therefore, simulation of spectra was tried by manipulating the J-values calculated from the a~tual spectrum (Table 4) in particular for h,4, h,5, J3,4 and J3,5 (Figure 20). Since long range coupling (W coupling) was expected between H3 and H5 and between H2 and H4, a value of h,4 = J3,5 = 1.5 Hz was used. h,5 and J3,4 were set at 0. However, the resulting simulation gave only a doublet of doublets for each of the axial and equatorial protons (Figure 20-A). A more reasonable spectrum was simulated by using h,4 =h,s= J3,4 = J3,5 = 1.5 Hz and J 1,4 =J 1,2 = 1.5 Hz ( Figure 20-B). 55 (A) (B) Equatorial protons Axial protons Figure 20. Simulation of spectra of trans-(118): ring proton region: (A) h,4 = J3,5 = 1.5 Hz, (B) h,4 = h,5= J3,4 = J3,5 = 1.5 Hz and J1,4 = J1,2 = 1.5 Hz A distinctive result was also obtained with 5,5-dimethyl-2-ethoxy-2-oxo-1,3,2- 4- dioxaphosphorinane (132). In this case the axial protons appeared aslbroad doublet (Figure 21-A). However, decoupling of the axial 5-methyl group removed the coupling to the axial protons l28 and a well separated doublet of doublet of triplets resulted while the equatorial protons remained as in the case of trans-(118) as a doublet of doublets of triplets (Figure 21-B). Again, the simulation is based on the J-values and chemical shifts calculated from the actual spectrum (Table 4) (Figure 21-C). 56 Equatorial protons (A) Axial protons -- .., , ...... , ,. ,.. ,.. ,., ,., .s - .. (B) ..... ,.. ,., 1.1 ... ,., ,., --,.. ,.. ,.. .. .s (C) Figure 21. Decoupling experiment and simulated spectra for (132). 57 The conclusion from these observations is that fast equilibration of the ring conformations between chair and twisted chair occurs and seems to put the ring protons in magnetically equivalent environments. ~o---P / 0 Table 4. Chemical shifts (0, ppm) and J-v~lues (Hz) for ring protons of trans-( 118) and (132). Protons Axial protons Compounds 6 JPOCHa 1HaHe J* 0 JPOCHa JHaHe J* Trans-(118) 4.04 2.12 -11.5 1.5 4.21 21.62 -11.5 1.5 (132) 4.04 3.9 -11.0 1.3 3.88 19.9 -11.0 1.3 J* = coupling values for apparent triplets. 58 Oxidation of cis- and trans-5-hydroxymethyl-S-methyl-2-thiono-r- 2-ethoxy-1,3,2-dioxaphosphorinane [cis- and trans-(116)] by MCPBA. (A) In CDCI3, With the stereochemical features of the starting materials established, their oxidative chemistry was then investigated. With MCPBA as oxidant the starting materials cis-( 116) and trans-( 116) disappeared quickly and gave a mixture of products via some intermediates. The reactions proceeded with substantial retention of configuration at phosphorus. The representative spectra from the oxidation of trans and cis- (116) are shown in Figure 22. On oxidation of trans-(116) (Figure 22-A) trans-(111), which formed with retention of configuration, was observed as a major product. Cis-(117) whose formation involves inversion of configuration was obtained as a minor product and the cyclized product, BPO, was also noted. On oxidation of cis (116) (Figure 22-B) Cis-(117) was formed as the major product; trans-(111) and a trace of BPO was observed. A control experiment was carried out by subjecting a mixture of cis- and trans (111) to similar reaction conditions for the oxidation of cis- and trans-(116). No change was observed in the isomer ratio indicating that the cis- .and trans-(111) are formed directly during the oxidation process and not by isomerization in the reaction media. Furthermore, the formation of BPO was not observed in the control experiment, thereby indicating that BPO must be formed during the Ol!iffit10"by intramolecular transesterification cf (116). 10Et~OH OEt ~ 1 0Et o/P~s o-P~ , o/P~0 0 / S 0 OH O OH Trans-(116) Cis-(116) Trans-(111) Cis-(117) 59 (A) / Trans-(116) Trans - (117 ) .~ Cis-(117) \ BPO 60 I Cis-(116) (B) Cis -(117) ,/ ~ Trans -( 117)-.. \. 60 80 60 40 20 0 -20 Figure 22. 3lp spectra from oxidation of trans- and cis-(116) in CDCl3 (final mixture). Spectra of starting materials are shown in insets. A mechanism for the oxidation of trans -( 116) is suggested in Figure 23 in which the starting material is initially converted to the oxygenated intermediate (140) I which can yield the major product, trans-(111), by direct desulfuration (pathway B). The same intermediate (140) can also yield BPO by desulfuration (pathway C) via a further intermediate (141) which may be formed by intramolecular transesterification (pathway A). 1 0Et D--;;P~S 0 OH Trans-( 116) [0] 0Et 1 0-P-01 0Et] A [ . 0.."- / OJ B 0 0/p~O B [~o/ 's' - -{S~;:.-r,t OH OH (141) Trans-(117) (140) C Cis-(117) (142) (BPO) Figure 23. Proposed Mechanism for oxidation of trans -(116) in CDC13. 61 By another intramolecular transesterification (pathway D) intermediate (142) can be produced from intermediate (140), and this in turn can give cis -(117) by desulfuration (pathway E). Cis -(117) could also arise by pseudorotation of the pentavalent species (140) ( pathway F ). Pathway G on the other hand, for the interconversion between (141) and (142) involves either intramolecular transesterification or phosphorylation of ethanol. In the oxidation of cis-( 116), however, the formation of BPO (via pathways A and C, or D, G and C) was negligible, probably because of decreased accessibility of the internal -OH group for transesterification. Pseudorotation to the inversion product trans-(111) was more significant than in the oxidation of trans-( 116). In both oxidations 3lp NMR spectra demonstrated that considerable amounts of various intermediates were formed at different times. These then disappeared at I differing rates associated with the concomitant formation of products (Table 5). Table 5. Product distribution (%) on oxidation of cis- and trans-(116) at two times. Starting material Trans-( 116) Cis-(116) Products Cis- Trans- BPO Cis- Trans- BPO (117) (117) (117) (117) Stage-A* 9 58 8 57 14 Final Mixture 12 73 15 70 30 Trace * The stage at which starting material was no longer present. 62 Many oxidation studies have been carried out with cyclic chiral compounds to allow the stereochemistry of the transformation to be studied. The isomeric 4-methyl- 1,3,2-dioxaphosphorinanes (143) were oxidized with several oxidants but only retention of configuration was observed.6-8,10,l2,l29 More interesting results were obtained by Toia and Casida 29,30 who investigated the oxidation of each diastereomer of £-alkyl 1,3,2-dioxaphosphorinanes (96,97) in different solvents. In chloroform, both isomers produced BPO quantitatively but in methanol the cis-isomer yielded the trans- methyl phosphate (98) only, and the trans-isomer gave the BPO. ~ ,R ~/P~x 0 R=methoxy X = lone pair, S, Se, 0 (143) (96, 97) (98) To investigate if there was any involvement of BPO in the formation of the isomer mixtures of cis- and trans -(117) in the oxidation of cis - and trans -(116), the oxidations of cis- and trans-5-acetoxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3,2- dioxaphosphorinane [i.e. cis -, trans -(119)] were investigated (Figure 24). In these cases the formation of BPO was prevented by the derivatization of the hydroxymethyl group. In addition, to examine the effect of concentration of the oxidant, two different levels of oxidant (1 and 3 molar ratios) were used. With a 1 : 1 molar ratio of oxidant to substrate, some starting material [cis - or trans -(119 )] remained in the final mixture. With more oxidant (3: 1 molar ratio), the starting material was completely consumed and two product peaks, assigned as cis- and trans -5-acetoxymethyl-5-methyl-2-oxo-2- ethoxy-1,3,2-dioxaphosphorinanes [cis-, trans-(144)] were observed. The distribution c. of products as a fu1ion of the amount of oxidant is shown in Table 6. 63 Generally, these oxidations are quite stereoselective giving the corresponding phosphate isomers, i.e. cis - and trans -(144) from cis - and trans-(119), respectively, in more than 90% yield. For trans-(119), the use of more oxidant resulted in more inversion product and this may reflect the more vigorous conditions. From the present experiments, the formation of the alternative isomer as a minor product was observed but without any formation of BPO. This suggests that the BPO is not involved in the formation of the minor isomer, and that reaction involving pathways A, C, D, E and G in Figure 23 can be excluded. Thus, in Figure 24 the minor isomer cis - (144) must be formed directly from starting material trans- (119) via an intermediate (145) which allows a small degree of isomerization, probably through pseudorotation. Table 6. Products distribution (%) in the oxidation of the isomeric acetates ( 119). Oxidation Cis-(119) Trans-(119) i~" Cis-(119) Cis-(144) Trans- Trans- Cis-(144) Trans- [O] Ratio•, (144) (119) (144) 1 : 1 7 91 2 9 4 87 1: 3 0 98 2 0 6 94 * substrate: oxidant (MCPBA) ratio 64 ,.,OEt ,.,OEt] oo/P~s [OJ O-P-0 [~ /\I 0 s OAc OAc Trans-( 119) (145) Pseudorotation OAc 0-P~OEt / "'o 0 Cis-(144) Figure 24. Possible mechanism for the oxidation of trans -( 119) in CDCl3, {B) In CH3.QH Since trans-( 116) was oxidized to the phosphate with retention of configuration when CHCl3 was used as solvent, and Toia and Casida 29,30 reported different products from the oxidation of .S,-alkyl dioxaphosphorinanes (96) and (97) in methanol and in CHCl3, the oxidation of the trans -(116) in CH3OH was investigated. In the present work the reaction was monitored for about 78 hours and representative spectra are shown in Figure 25. The intermediates A and Chad very short life times (about 10 mins under the conditions used) (Figure 25-A). Intermediates D and E subsequently appeared and after approximately 70 min were the major intermediates in the reaction mixture (Figure 25-B). At this time (70 min) intermediate B had disappeared and the starting material had declined to 27% of the total mixture. When the starting material I disappeared only the intermediates D and E were left and they diminished as products appeared. The variations of the starting material trans -(116), intermediate B and the major intermediates (D and E) are plotted in Figure 26 to the point where the starting material 65 had disappeared. On addition of a second portion of MCPBA (at 80 min) only intermediate B reappeared suggesting that it is directly related to trans -(116) and it may be an oxygenated intermediate such as (140). In the final mixture (Figure 25-C) five products were identified [cis -(118), trans -(118), BPO, cis -(117), and trans -(117)] and two unknown minor by-products were also observed. These assignments are all based on direct comparisions with authentic samples. The observation of cis -( 118) and trans -(118) is of interest because it suggests that phosphorylation of CH3OH occurred during the reaction without any stereospecificity to give them in a nearly 1 : 1 ratio. Trans-(118) Cis-(118) (%) 100 - 80 IJ Trans-(116) -·-7)0--- Intcrmcdia1e B Intcrmcdiatc D and E 60 ------· 40 ----..... ------20 0 -'------....,;;==------'!F---_,,;a;....___ 0 20 40 60 80 100 Mins Figure 26. Changes with time in composition of the reaction mixture on oxidation of trans-(116) in CH3OH. 66 Trans-(116) (B) / (A) E / D ~ Tran-(118) __.. Cis-( 118) ___,, (C) Cis-(117) Trans-(1170 BPO Minor by-products _.,,.._.. 70 50 30 10 0 -10 Figure 25. 3lp NMR spectra from the monitoring of oxidation of trans-(116) in CH30H. 67 The major intermediates D and E may be isomeric compounds related to cis - (118) and trans -(118) because the difference in chemical shifts of 0.3 ppm between them is the same as that between cis-(118) and trans-(118). However, although the exact relationship i.e. between peak D and cis -( 118) and between the peak E and trans ( 118) was not established, intermediate D must have same stereochemistry as cis -(118) and intermediate E as trans-( 118). The BPO was an expected product as were the desulfurized compounds cis -(117) and trans -( 117). A mechanism is proposed in Figure 27 for this reaction, and the major intermediates D and E are, because of their chemical shifts (B 24.4 and <524.1, respectively) which correspond to the phosphorus-sulfur single bond region, considered to be (149) which has the SOCH3 group attached to phosphorus. Furthermore Bielowski and Casida 26 reported (EtO)2P(O)SOCH3 (146) as one of t\..eproducts from the oxidation of O,O-diethyl Q-phenyl phosphorothionate (65) in methanol. In Figure 27 pathways A, B,C, D, E and F accounting for the cis -and trans - when (117) and BPO also appear to operate similarly• the oxidation of trans-(116) was effected in CDCl3. The major intermediate (149) can be produced by rearrangement (pathway H) of intermediate (148) which was formed from intermediate (140), (141) or (147) by phosphorylation of CH3OH (pathways G, K or M). In pathway L intermediate (147) can be formed by cleavage of the P-OEt bond and it [i.e. (147)] can phosphorylate CH3OH to give intermediate (148). Alternatively attack of methanol on sulfur (pathway N) would lead directly to (149). Oxidation of (149) (pathway I) to intermediate (150) which, in turn, can phosphorylate (pathway J) CH3OH 22,24,26 generates cis- and trans -(118). These products can also be formed directly from intermediate (149) by desulfuration of the sulfenate group (SOCH3) (pathway S)l30. Intermediate (150) can produce BPO by phosphorylation of the internal hydroxyl group (pathway R) or it can be oxidized further to intermediate (152) after a novel 68 rearrangement 22 to (151) (pathway 0). The intermediate (152) can also give cis-and trans-(118) by phosphorylation of CH30H (pathway Q), or yield BPO again by phosphorylation of the internal hydroxyl group (pathway P). (147) lM [~·1] p (148) lH [~/::-J (149) Cis-(117) s 1 ! o o (~.:~] _Q_[~·::""'°J-1 '-P/:~] CH20H CHiC)H l CH2OH (150) (151) j1 I ~ ~.::- Q [ B/::r] CH20H CH20H (118) (152) Figure 27. Proposed mechanism of product formation on oxidation of trans-(116) in 69 To investigate any differences in the oxidations between the trans- and cis isomers, cis-(116) was also studied. The final spectrum from the reaction is given in Figure 28. It was found that the course of the reaction and the final products were similar for both starting materials in every aspect, including the appearance of the intermediates, by-products and products, which comprise cis -(118), trans-(118), BPO, cis-( 117) and trans -( 117) (Figure 28). The variation of cis -(116), major intermediates D and E and intermediate B with time is presented graphically in Figure 29. The regeneration of intermediate B was also observed, as previously noted, on a second addition of MCPBA at 95 min. Cis -(118) and trans-(118) were the major products in a I : 1. 1 ratio; again the same as that noted on °1"rtti.0,"of the trans -isomer . D_..... Tran-(118) ('E / Cis-(118) "--- Minor by-products ~ Cis-(111) ¥" Trans-(111) v BPO Figure 28. 31 P NMR spectrum showing products from the oxidation of cis-(116) in CH3OH. Intermediates are shown in inset. 70 Notwithstanding the above, some minor differences are also evident. One small unknown peak was observed at 6 7.4 which was absent in the oxidation of trans -( 116) and the ratio between cis-( 117) and trans -( 117) was 2. 7 : 1 compared with 1 : 2.6 in the o~on of trans -(116). This was not surprising because cis -(116) was expected to give more of the product with the same stereochemistry [cis-(117)] than the inverted product [trans -(117)]. (%) 100 Cis-(116) 80 ----o-• Intermediate B ------Intermediate D and E 60 ____ ...... 40 ------x 20 , , 0 /' 0 20 40 60 80 100 120 Mins Figure 29. Change in composition of the reaction mixture for the oxidation of cis-(116) in CH30H. One further difference worthy of note is the ratio between the major intermediates D and E; peak D was bigger than peak E throughout the reaction while the reverse is true in the oxidation of the trans-isomer (Figure 28). This suggests again that the two intermediate peaks represent stereoisomers, and that trans -(116) leads to more 71 trans- intermediate (peak E) and cis -(116) leach.to more cis- intermediate (peak D). However, even though the ratio between the two major intermediates D and E was different depending on the stereochemistry of the starting material, the major products, cis -(118) and trans-(118), showed a 1 : 1.1 ratio. This suggests that major intermediates D and E both react with methanol in a non-stereospecific manner. These transformations and mechanistic details are sulllII»rized in Figure 27 for trans-(116). I Oxidation of trans-( 116) in methanol was also carried out with a lesser amount of oxidant to study the fate of the intermediates under conditions where they might not be oxidized further (Figure 30). The reaction initially proceeded as previously to give major intermediates D and E and the minor peak F. However, in the final mixture two new products (peaks X and Y) were observed and these are presumably derived from decomposition of D, E and F. This result suggests that when excess oxidant is present, these intermediates are further oxidized leading to products resulting from phosphorylation of methanol. DandE ~ F I 30 20 70 50 30 10 0 -10 Figure 30. 3Ip NMR spectrum showing products from the oxidation of trans-(116) in CH3OH with less amount of oxidant. Intermediates are shown in inset. 72 To investigate the relative rate of oxidation of each isomer, and the possibility of actual isolation of the intermediates D and Ea mixture of cis - and trans -(116) was also oxidized in CH30H. Prior to addition of oxidant the cis - and trans -isomers were present in a 1 : 1 ratio. Within 5 minutes of addition of MCPBA, the cis : trans ratio was 1 : 3 and this slower rate of oxidation of the trans- isomer suggests that the sulfur is less accessible to the oxidant, possibly a reflection of conformational differences. When all the starting material had disappeared and intermediates D and E were the biggest components in the reaction mixture (30 min) the methanol was removed and EtOAC added. The solution was then washed with Na2S20& solution to destroy any unreacted MCPBA. However, the intermediates D and E also disappeared and small new peaks (6 73.5 and 6 63.8) appeared along with usual products; i.e.,cis -(118), trans -(118,) cis -(117), trans-(111) and by-products. The occurJnce of new peaks may reflect a side reaction resulting from the reducing agent. However, this result doti . I suggest that the intermediates D and E cannot be isolated by work up and extraction procedures. {C) In aqµeous acetone. To study the oxidations in aqueous environment each isomer of (116) was oxidized in aqueous acetone. On oxidation of trans-( 116) acyclic hydrogenphosphonate (156) (JP-H = 673 Hz) was observed as the major product together with the desu~rized product trans -(117) (Figure 31). In addition, the acid (153) was formed as a minor product with traces of the BPO and cis -( 117) also observed. 73 (153) Trans-(154) Cis-(154) (155) On monitoring the above reaction, three intermediates were observed; the short lived li:a9intermediate at o30.3, two hydrogenphosphonates, trans-(154), (JP-H = 695 Hz) and (155) (hydrogenphosphonate, JP-H = 711 Hz) (Figure 31-A) and their conversion to (156) was also noted (Figure 31-B, C). When the variation of these three hydrogenphosphonates i.e., trans-(154), (155) and (156) was plotted (Figure 32) it appeared that trans-(154) and (155) are not interconvertible [i.e., they are formed by distinct pathways, but they lead to a single product (156)]. (%) 100-.------ 80 ----0 ------60 ------,, o--- I ,, Trans-(154) 40 ,, D , ,c (155) -----o---- (156) 20 5 10 15 20 25 Hours Figure 32. Variation of hydrogenphosphonates trans-(154), (155) and (156) with time. 74 (A)SMins (155) '-.. Trans -(154) Trans-(116) o30.3 ; \ ' .. .. 4 (156) (B)I Hour ' ..,...... , ',. ,_ ~ ,-1--.._..,. __.,,.._._,-,;.._,IJ (156) (C) Trans-( 117) 21 Hours (Final mixture) " ./ (153) \.._ 80 60 40 20 0 -20 Figure 31. Representative 31p spectra from the oxidation of trans -(116) in aqueous acetone. 75 Hydrogenphosphonate (156) could be formed by the hydrolysis of both trans (154) and hydrogenphosphonate (155) because this is an anticipated reaction in the . aqueous environment. Considering the behavior of these hydrogenphosphonates and the findings of other workers32,35 a reaction mechanism is proposed as shown in Figure 33. The formation of cis- and trans-(111) and BPO is the same as in the oxidation in CDCl3. However, for the hydrogenphosphonates the initial oxidation product, the phosphoxathiirane (140), is attacked by water yielding the five-coordinate intermediate (157), which subsequently forms the phosphoro(thioperoxoic)acid (158). The phosphoro(thioperoxoic)acid (158) can be further oxidized to produce trans -(154), which in turn hydrolyses to (156). The rearrangement of phosphoro(thioperoxoic)acid (158) to another phosphoxathiirane (159) can yield the oxysulfonate (160), which in turn gives phosphoric acid (153) and elemental sulfur. On the other hand, the five coordinate intermediate (157) may be hydrolysed to the acyclic intermediate (161), which is oxidized further to (162) and subsequently yields the (155). Finally (155) is hydrolysed to (156), the end product of this reaction. a Analog~ results were obtained from oxidation of the cis-(116) usinghimilar reaction procedure, but with a different resultant product distribution. Specifically more desulfurized products occurred [i.e., cis -(117) and trans -(117)] at the expense of some of (156); BPO was not detected. 76 -{S~..o (BPO) 1 [---0 (157) II (O] +H20l-H2S04 0 SOH .....-P~?& H [~/:] ' J s 0 " (160) [~/~~ ID (155) IH'/1120 hs1 +H,0 l-H,SO, ,H H+/H2O ~0/P~ ,oo ~oo/P~o 0 0 So--t" OH OH ID (156) (153) Trans-(154) Figure 33. Proposed mechanism for the oxidation of trans-(116) in aqueous acetone. 77 Oxidation of BPS by MCPBA. Oxidations of BPS were carried out in a non-hydroxylic solvent (CDCh) with molar ratios of MCPBA: BPS varying from 5 : 1 to 0.5 : 1 to examine the effect of oxidant on any intermediates formed, and on the final product distribution. [-0 With excess oxidant (5 : 1 and 3 : 1), BPS was rapidly converted to BPO. As previously reported by Swinson et al. 28 no intermediates were observed. With 1 equivalent of oxidant, BPS was not completely converted to BPO, and 20% remained in the reaction mixture. This behaviour suggests that some oxidant was consumed in oxidising the intermediates, even though they are not detectable, or in oxidizing the released sulfur. With a 0.5 : 1 ratio of oxidant to substrate only 19% conversion to BPO was observed. When MCPBA was added to a solution of BPS in CD3OD, in contrast to the situation when CDCl3 was used as solvent no sulfur precipitate was observed. Generally the course of the reaction was analogous to the oxidation of monocyclic compounds cis- and trans-(l l 6) in CH3OH and the same resonances were noted (Figure 78 34). The most significant difference was that the intermediate E was observed, with little intermediate D, and neither cis- nor trans-(1 l 7) was formed. E Trans-( 118) / D Cis -(118) " '---- Minor BPO by-products / 25 PPM ...,.._ •••"''°re •• ( rt •• I ...... , 40. 1jt' ••t tat 111• ,,,.,.., J\illP,C ...t,'1 1 1 j 1 1 11 j 1 1 j I I I I j I I 4 j I I I I j I I I I j I I j I j I I I I j I I I j I I I I j I I I j I •• I j; j i 1; 1 j; u u j;;:; j 65 60 55 so 45 40 35 30 25 20 15 10 S O -s -10 -15 PPM Figure 34. 31 P NMR spectrum of final mixture of oxidation of BPS in CD3OD. Intermediates are shown in inset. BPS was completely consumed after 80 mins and the major intermediate E was at its maximum intensity (37 %) at that point. Thereafter, the major intermediate E was steadily converted to products over a period of 46 hours (Figure 35). The rate of formation of specific products, cis-(118), trans-(118) and BPO for the above reaction is shown in Figure 36. In the early stages (20 mins) trans-(118) formed faster than cis-(118) compared to the rates at the later stages. At 10 mins BPO accounted for about 34% of the oxidation products, and did not increase much more. This suggests that it can be formed quickly by direct desulfuration of BPS at high concentrations of the oxidant, i.e., during the early stages of the reaction. The oxidation displayed little stereoselectivity and a ratio between cis-(118) and trans-(118) of 1 : 1.15 was observed in the final reaction mixture. 79 (%) 100- ,.• C BPS ,· 80 ----o---· Intermediate B ,.,· Intermediate E ,.,. ------Products; ( 118) and BPO ,· 60 -·-·-·•·-·-·· ,.,. ,· ').;:~ 40 ___,,._.,,,.,., ....V'lll • ..,,..6 ·-·-·-·-·,:,~1ii1·~ ..... -- ~ ..a: ...... ------' ., ------', 20 .,·' , ~-- ',' ., //, , ' ' ' ,;4:.-::.•-----o-.- ---- ' 0 0 20 40 60' 80 50 Mins Hours Figure 35. Time dependent changes in 31 P NMR resonances on oxidation of BPS in (%) 20,------ D Cis-(118) 10 ----o---· Trans-(118) BPO _____ ... - -,r------.... - ---- 0 -F------.--r---,---,-----,-...---,---T------1 0 20 40 60 80 100 Mins Figure 36. Time dependent product formation during the oxidation of BPS. It is possible that methanolysis of BPO to cis-(118) and trans-(118) may occur after it had formed from oxidation of BPS. However, a control reaction was done and no reaction between BPO and methanol could be detected. Furthermore, no 80 isomerization between cis-(118) and trans-(l I 8) was observed nor were they converted into BPO. From these results it is clear that cis- and trans-(118) were formed not by methanolysis of BPO or isomerization between them but from reaction of the intermediates occurring in the oxidation of BPS with the solvent. Further, BPO itself is formed directly from BPS and not by intramolecular transesterification of cis-(118) or trans-(118) in the reaction media. On comparing the oxidation of BPS with the oxidation of the monocyclic isomers cis-(116) and trans-(116) in methanol, the mechanisms are virtually the same in terms of intermediates and products cis-(118) and trans-(118) and the ratio between them. The major variation is the ring opening step and the appearance of only one major intermediate (i.e., intermediate E). The monocyclic isomers cis-(116) and trans-(116) reacted much faster with MCPBA than BPS and, again, this is considered to reflect the steric constraints of the latter. Swinson et al. 28 reported that the intermediates in the oxidation of several phosphorothionates in acetone occurred with 3 IP chemical shifts about midway (13 - 33 ppm) between the position of starting material (P=S) and the products (P=O). Similar results were obtained in the present studies in methanol. However, in methanol the major intermediates had appreciable lifetimes and were easily detected. This was in contrast to the result in acetone where they 28 were difficult to observe, even when the oxidation was performed below -5 °C. Swinson et ai.28 also suggested that the intermediates are tetracoordinate or have a phosphonium polysulfide structure of the type R3PSx or perhaps R3POSx as described earlier in Scheme 3. In the present studies this appears not to be the case since the nucleophilic reactivity of methanol leads to the alkoxy esters, as found in the oxidation of phosphorothiolates 24 , 26 phosphoramidothiolates 21 and phosphorothionates26. In the last case 26 , the sulfenyl ester (146) was observed as one of the products from oxidation of 0,0-diethyl £-phenyl phosphorothionate (65) in methanol. 81 Taking account of the results so far, together with the previous studies, possible mechanisms can be suggested for the oxidation of BPS in meµianol as shown in Figure 37 and the reaction pathways can be discussed as in presentedr'" igure 27. --{S~P=S IQ"" (BPS) [--G ~o/P:o CH20H (150) I (118) (152) Figure 37. Reaction scheme for the oxidation of BPS in CD30D. 82 In considering this oxidation further, when the amount of oxidant was decreased a new peak at o 68.4 (Figure 38) was noted in the final mixture. This must be derived from decomposition of intermediate E, which presumably was oxidised in the previous experiment by the higher concentration of oxidant. An identical peak was observed when trans-(116) was oxidized under the same conditions (Figure 30). This result confirmed that the oxidation of intermediate E should occur prior to phosphorylation of methanol which ultimately yields cis- and trans-(118). These studies also suggested that the peak Y in Figure 30 arise from the intermediate F because it was only observed when intermediate F was also observed. X " (BPS l " 70 50 30 10 0 -10 Figure 38. Oxidation of BPS in CD3OD with less amount of MCPBA. (C) In agueous acetone Since hydrogenphosphonates were observed in the oxidation of monocyclic cis and trans -(116), oxidations of BPS were carried out in aqueous acetone to investigate, in detail, the formation of hydrogenphosphonates from the bicyclic starting material. 83 As have been observed with the oxidation of the monocyclic isomers cis- and trans-(116) in aqueous acetone, the hydrogenphosphonate (156) was found in the final product mixture, along with BPO and the acid (153). However, there were also some key differences, specially the formation of both cis- (154) and trans-(154) indicating that the reaction is nonstereospecific, and the absence of hydrogenphosphonate (155). By monitoring of the reaction with time it was found that the cis-(154) and trans-(154) also converted to (156) (Figure 39). Therefore, both the monocyclic or the bicyclic system yields the same acyclic hydrogenphosphonate ( 156). (%) 50-r------ 40 30 C (156) -----o---- Cis-(154) 20 1e Trans-(154) 10 0 +:,,..._...--,--~=::;::::;:::~~L.r------r-...,...--~_J 0 2 4 6 8 10 12 Hours Figure 39. Variation of hydrogenphosphonates with reaction time. As discussed earlier, two pieces of evidence suggests this final product to be an acyclic compound. First, acyclic analogues usually have chemical shifts slightly downfield relative to the cyclic compound, and second this compound, unlike the 84 starting isomers, did not show any diastereoisomerism. The absence of hydrogenphosphonate (155), which was observed in the oxidation the cis- and trans isomers of (116) in aqueous acetone (Figure 33) confirmed that the assignment was correct because such a compound can not be formed in this oxidation of BPS. A reaction mechanism is suggested in Figure 40 and most of the pathways can be explained as dicussed earlier for Figure 33, including the formation of BPO and the acid (153). However the key difference is the formation of the bicyclic thioperoxoic intermediate (163) and its subsequent ring opening. The formation of acid (153) as a hydrolysis product of BPO was excluded because in the control experiment with BPO under same reaction conditions no other products were observed. Two more experiments were carried out, in aqueous CH3CN and DMF, to investigate any differences in the formation of hydrogenphosphonates as dertermined by the solvent. Surprisingly, in aqueous CH3CN, no hydrogenphosphonate was € evidenced, and the phosphate was the major product. In aqueous DMF, the reation proceeded similarly to that in aqueous acetone to yield a hydrogenphosphonate but the product mixture was dominated by BPO. Thus it appears that hydrogenphosphonate formation is solvent dependent and that aqueous acetone is the best medium to optimize its formation. 85 -G:P=S o (BPS) l[OJ -{S~--[-{S~ hsi 0/P:H 0 0 00 (153) Figure 40. Proposed mechanism for the oxidation of BPS in aqueous acetone. 86 Oxidation of S ,5-di methyl-2-th i ono-2-ethoxy-1,3,2-dioxa phosphorinane (133). To evaluate the influence of the internal hydroxyl group in trans- or cis-(116), or in BPS after ring opening, the results from oxidation of these compounds were compared with those from oxidation of 5,5-dimethyl-2-thiono-2-ethoxy-1,3,2- dioxaphosphorinane (133). This starting material offers several further advantages since no stereoisomerism is involved and it has better solubility in methanol than BPS. Also, since the products do not contain the hydroxymethyl group GC analysis is more feasible and, finally, it is relatively easy to prepare the required related reference compounds for confirming product assignments. The oxidations were carried out in different solvents as were the other oxidations and, as expected, only 5,5-dimethyl-2- oxo-2-ethoxy-1,3,2-dioxaphosphorinane (132) was obtained from reaction in CDCl3. However, in contrast to the previous studies no intermediates were observed thereby suggesting that the hydroxymethyl group, perhaps through intramolecular hydrogen bonding, may exert a stabilizing influence in such species (116). The above reaction was repeated in methanol and the results compared with trans-(116) selected as a representative reference [as discussed earlier, the same results were obtained from trans- or cis-(116) or BPS]. Interestingly, the reactions of 5,5- dimethylcyclic thiophosphate (133) and trans-(116) are very similar. The only significant difference was the numbers of products, which is accounted for in terms of the formation of diastereomers, and the overall product distribution. This suggests that 5-hydroxymethyl group does not influence the reaction, per se. When the levels of oxidant were reduced oxidation of (133) in methanol gave similar results to those obtained from the oxidation of trans-(116) as shown in Figure 30. Thus, two by-product peaks at o 68.2 and o 66.5 were now obtained also. An t:.ke. attempt was made to identify these two by-products by GC-MS analysis oy(reaction 87 e mixture since they are considered to be derived from the intermediate of intest. However, only the major products (137) and (132) could be confirmed, and no useful information about the other materials was forthcoming. In aqueous acetone the oxidation of 5,5-dimethyl cyclic thiophosphate (133) the proceeded similarly to the oxidation of cis-, or trans-( 116) inisame medium. Thus, in the final mixture hydrogenphosphonate (165, Jp_H = 671 Hz) was observed together with the desulfuration product (132) and the acid (135). By monitoring the reaction, it was observed that the hydrogenphosphonates (164) and (138) were converted to (165) and the latter conversion was confirmed by separate control experiments with authentic (138). A hydrogenphosphonate analogous to (165) was also observed as an end product and as the only hydrogenphosphonate from the oxidation of trans or cis - (116) or BPS in aqueous acetone. The reaction pathway involved is analogous to that shown earlier in Figure 33 relative to assigning the hydrogenphosphonate. o o o, ,....OE.t \ ..... 0B: o, ,.O:Me )C ' ,....H )C l~s >C /~o )C ,P~Q ,P.::::::o 0 0 0 0 (138) (133) (132) (137) ,....OB: ,....H o, ,....OH o-P-H o-\~O )C ,P~o >C \) >C OH 0 00 00 (135) (164) (165) 88 Spectroscopic investigation of the intermediate E in the oxidation of BPS by MCPBA in CD3 0D (CH30H). From monitoring of the oxidation of trans- and cis-(116) in CH30H by 3lp NMR the increase and decrease of intermediatesD and E were observed as was the formation of products. At this point these intermediates were considered to be isomeric tile monocyclic sulfenates (149). Since intermediate E was formed asimajor intermediate, with little D, in the oxidation of BPS in CD30D further detailed spectroscopic studies were conducted to investigate its structure. !Al_31 P and 1H NMR monitoring. The oxidation of BPS in CD30D was monitored by 31 P and 1H NMR. The peaks arising from intermediate E could be observed in both 31P and 1H NMR and in a series of experiments the reaction was monitored with only one or two minutes delay between taking sequential 31P and 1H measurements. After 3 hours of reaction no BPS remained and intermediate E (31P NMR) was a major component in the reaction mixture (Figure 41-A). Some minor intermediates, cis (118), trans-(118) and BPO, and minor by-products were also observed. In the 1H spectrum (Figure 41-B), BPO, cis-(118) and trans-(118) were observed. The complicated pattern of peaks between o4.5 and O3.9 are from the ring protons of cis and trans-(118). If resonances for BPO, cis- and rrans-(118) are substracted from the spectrum, it is apparent that peaks P and S are from intermediates, and therefore ol'\ primarily from the intermediate E. It was found that some of these resotes overlapped 89 with peak N, the resonance corresponding to the ring protons of trans -(118). Therefore, the major intermediate E may have same trans configuration as trans -( 118). In view of chemical shifts of peaks S, P and N, it was concluded, again, that the intermediate E has a monocyclic structure with the methylene hydroxy and alkoxy substituents trans to each other. (A) E 0 I Ii I I I I I I I I I I I I,, I I I I I I I I I I I I I I I I I 65 s'o ' ' 5'5° ' 5'0 ' '4•5 ' ' 4'0' ' I I I I I I I I I I I I I 35 30 25 20 15 10 5 0 PPM ·5 -10 ··15 (B) Cis -(118):(o) Trans-(118) : ex> s )( p BPO :(A) )C 0 0 N• A 4 \., _J Ring protons (o, x) 4.& 4.4 4.2 4.0 3.0 3.6 3.4 3.2 3.0 2.8 2.& 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 .8 PPM , Fgure 41. Spectra (3lp, lH) from the monitoring of the oxidation of BPS in CD3OD. 90 (B). Solvent substitution The intermediate E from the oxidation of BPS with MCPBA in CD30D gave cis and trans-(118), and the monocyclic structure (149) is now postulated for it. If correct, 2H NMR monitoring of the fate of -SOCD3 group can potentially give useful information about the structure of this intermediate and the reaction pathway involved (Scheme 11 ). For example, if the oxidation is initiated in CD30D and then this solvent is replaced with CH30H at the stage when there is a maximum concentration of ~ intermediate E without starting material, its sub~uent reactions can be evaluated. --G:P=S (BPS) ~C ~OCH3 r---~0/p~o + CH20H (118) Scheme 11. Reaction scheme applicable to the approach with 2H-NMR monitoring. As a preliminary test of this approach, 2H-spectra were taken with the known products from oxidation of BPS in CD30D. Four peaks were observed as products, but unfortunately they were broad and overlapped to the extent that this method could not be used to provide useful information. As noted earlier, by monitoring the oxidation of BPS in CD30D slow conversion of intermediate E to products, mainly the monocyclic isomers cis- and trans- 91 (118), could be observed. However, it was not clear whether E was phosphorylating methanol to give those products, or if E simply decomposed to products directly. If the intermediate E phosphorylated methanol, then CD30D should be phosphorylated when the reaction is carried out in CD30D. To test this hypothesis a sample of the reaction mixture in CH30H was divided into two portions (A and B) at the stage where no starting material remained but a substantial amount of E was still present. Both samples were evaporated to dryness, then portion A was taken up in CD30D, and portion B I taken up in CH30H, and the reaction allowed to continue. The experiment is shown schematically in the following diagram. -{S~P=S (BPS) MCPBA ICH30H Portion BI-·-----'-----~·! Portion A I l 1 Evaporation Evaporation ICH30H jCD.JOO H3C~ 1 0CH3 o / p ~ 0 0 CH20H (118) Integral C-CH3 = Integral P-OCH3 [ CH3+CH3 = OCH3+0CH3 ] (118) Integral C-CH3 > Integral P-OCH3 [CH3+CH3 > OCH3+0CD3 ] A considerable amount of the intermediate E (20%) was present in the reaction L mixture (without BPS) at the point it was dFvided into two portions. Figure 42 shows 92 the comparison for the final 1H spectra from portions A and B. In the CH3O H experiment (portion B) the ratio between integral of C-Cfu and P-OCfu was 1 : 1.1 while in the CD3OD experiment (portion A) the ratio between C-Cfu and P-OCfu was 1 : 0.8. Although the integration differences are small, the spectra are clean and these results therefore suggest that the products from the portion of the reaction mixture containing the CD3OD contain deuterated methoxy groups. This in tum, suggested that the intermediate E does not decompose to products; rather it phosphorylates methanol directly to form cis - and trans -(118). Portion A P-OCH3 I Cis-(118): {o) 6 3.90 Trans-( 118): (x) X '---.. BPO: (A) " 0 0 ~ Ring protons ( o, x) 4.6 4.4 4.2 4.0 3.8 3.6 3.4 3.2 3.0 2.B 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 .B PPM Portion B X. :x 0 ~ Ring protons (o,x) 4.6 4.4 4.2 4.0 3.B 3.b 3.4 3.c? 3.0 2.8 2.b 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 .B PPM Figure 42. Comparison of 1H NMR of portion A and B. 93 An unknown resonance (B 3.90) was also observed and its intensity in portion A was also smaller than in portion B. This suggests that it is assignable as a methoxy group. This was subsequently shown to be the methyl resonance for the methyl ester of a MCBA (166) by GC-MS analysis of the reaction mixture offlater experiment, and in the present experiment by the addition of authentic (166) which gave enhancement of that resonance. Thus peak (B 3.90) was from methoxy (C-OCfu) group of (166) and was formed by the reaction of MCBA (167) and solvent (CH3OH) O~ .,,,OH o~ .,,,oMe 6 Cl 6 Cl (167) (166) {C). 2-D CHeteronuclear correlation) NMR Since intermediate E from the oxidation of BPS in CD3OD or CH3OH was of long half-life, a 2-D (heteronuclear correlation) NMR experiment was carried out because this technique allows the direct correlation of a 3 l P resonance with its coupled protons. From the spectra (3lp and lH) taken after 10 mins, the sweep width (515.07 Hz for 1H, 62.5 ppm for 3lp) and 01 (offset, 5692.937 Hz for 1H and 1082.03 Hz for 31P) were calculated for the part of the spectrum of greatest interest. In addition a JrocH = 11 Hz was used as the average coupling constant for P-O-CH protons. (See "Experimental" part for more detailed parameters). In the 2-D spectrum obtained (Figure 43), several correlated peaks were found and the peak at B = 25.0 (3lp) x B 4.18 - 4.20 (lH) from intermediate E was clearly observed. In particular, the proton chemical shifts observed corresponded to the region of the ring protons of trans-(118), but these overlapped with other resonances. By I 94 taking a spectral slice through the column at 6 25.0 (31 P), the proton spectrum shows up an AB quartet (Figure 44) clearly indicating the presence of ring protons at the same area of the ring proton of trans -(118). 1HNMR • "'-1-+--+---+---+---+---+----+----i~-H 0 t-1--+---l---l---l---+---+-----+----l--H; 0 ~-1-.!-.---+-----1---l---1---+---+-----+----l~ 0 • ~-1-+---+-----1---l---+--+--+---+--t-i:: 0 ~-1--+---+--+.~-+---+--+--+--t----,-;_: 0 ~-1-+---+---1---l---+---+---+---+--f-gi 0 i:i--1-+---+---l---l---+---+---+---+--t-l 'Vo 'V :a: ~-l-+---1----1--1---+---l--+--+--+i! 0 UI 0 ,l-l,---1---11--+--+---+--+--f---H; 0 I Ul-l-+i..--1---+---+---+---+---+---+----t-i, 0 I o-1--1--+---1---+---+---+---+----t--1- o I iA-l-+---1----h---l--+--+---+--+--+~~ .o 'V I :a: w w .. ~ • ...... ~ OI • 1\1 0 a, OI • N 0 Figure 43. 2-D Spectrum of oxidation of BPS in CD30D. 95 t• •. 6 •. s •.• ..3 •. 2 •. l •. o 3.q 3.9 3.7 3.6 PPM Figure 44. 1H-spectrum corresponding to the intermediate E. Three conclusions can be drawn by summarizing the results so far. First, the major intermediate E has a single bond between phosphorus and sulfur (P-S), as judged by its 3Ip chemical shift (O 25.0). Second, Eis a monocyclic compound with a trans configuration similar to isomer product (2); however, a dimer (or polymer) cannot be ruled out. Third, except for the ring protons, no other proton containing-group is coupled to phosphorus, or if it is the coupling value is considerably different from the normal value approximately 1 lHz. In the present experiment the NMR resonances with significantly different coupling values would not be observed. However, structure (168) can be considered for intermediate E in addition to the one (149) already postulated. Several approaches were now taken to further define this material. 96 HO 1 SOCH3 0-P-S-S-p/? 0~ O /P~o / II 0 0 O OH OH (149) (168) (D) Direct analysis of the reaction mixture by chemical ionization mass spectromett:y . Mass spectrometry can be a good probe for the study of intermediates,13I especially using chemical ionization techniques. By careful comparisons of the spectra from the reaction mixture at various times with those of the final mixture, peaks appearing and disappearing can be assigned to intermediates. Relative to the present case, and for convenience, the spectra of the final product mixture will be discussed first. In the 3lp NMR spectrum of final product mixture five major product peaks were observed, viz; cis- and trans-(l l 8), BPO, and two minor by-products. Thus cis and trans-(118); m/z 200 (M+l), 228 (M+29), 240 (M+41)) and BPO; m/z 165 (M+l), 193 (M+29), 205 (M+41)) were observed in MS spectrum. A series of ions at m/z 297 (M+l), 325 (M+29), 337 (M+41) were also observed and in the total ion chromatogram (Figure 45) the ion current for m/z 297 showed a different time maximum from the other ions confirming it to arise from different substance(s). By comparing the 3Ip NMR spectrum and the MS spectrum this material can be two minor by-products and regarded as isomers because only one ion series was observed for them with two 3Ip chemical shifts values. These isomers were tentatively assigned as (169). ~/·~~ CH2OSO2OCD3 (169) 97 A OXI ,P 00 lP:1 0 C 3 oov file JJK314 tBl.7-tBl.7 ••u.~~~ M 100 200 300 400 500 ~~8 88no MCBA !6 '--- i50eo (118) ~~o Methyl ester U8 ofMCBA igo ~lio36 1;0 ~ao no iio ,~fgo ~00 £0 ~o 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Figure 45. Total ion current chromatogram. For the studies on intermediate E, the 3 lP NMR spectrum was taken after 1 hour and 10 mins to ensure that it was the biggest resonance. That reaction mixture was then analysed by MS (Cl mode) using the direct insertion probe. In the resultant series of spectra MCBA(167) was predominant; however, some peaks at m/z 313 - 315 were observed after 5.16 min and these increased to a maximum at 5.78 min (Figure 46). By 6.36 min they were no longer observed. Assuming that it is stable to MS conditions these new peaks can be considered to be derived from E since all of the peaks below m/z 300 are assignable to known compounds. However, these data were not useful for the further assigning structure. 98 From the MS analysis of the reaction mixture discussed above, MCBA was predominant since exce.ss MCPBA was used. To overcome this problem less MCPBA (1 : 1 molar ratio) was used, and the reaction was refluxed to promote the formation of the major intermediate in a shorter time. This approach was rationalized since it had been previously observed that the major intermediate quickly formed from BPS, but its conversion to products was slow. However, refluxing enhanced the rate of all the reactions and only 11 % of the major intermediate was observed after 10 minutes. After 20 minutes, the major intermediate diminished further with a corresponding increase of products cis- and trans-(118). However, a new peak at B 68.5 also formed, presumably from direct decomposition of the major intermediate, as in Figure 38. In conclusion, this approach, also, was not found to be useful for further MS analysis since it did not provide any further information. file >JK310 BPS MCPBA OXIDATION OIP:150 CI 3000V 5mgBPS/25mg Scan 531 Bpk Ab 164737, 5,78 min. 110 157 100 90 14000 80 12000 70 10000 0 50 8000 139 ''\ 0 6000 30 4000 20 2000 Figure 46. MS spectrum from the monitoring of oxidation of BPS in CD3OD. 99 CE} GC-MS analysis. To ascertain whether intermediate E and its related products in the oxidation sequence could be analysed by GC-MS, GC analysis of the reaction mixture was first carried out. The biggest component, as anticipated, was MCBA (167) and of the other components B increased throughout the time of the measurements (30 hr). Component A showed interesting fluctuations, i.e. it maintained a reasonable intensity at the early stages but disappeared at the end of the reaction .. This time dependency was similar to that of intermediate E observed by 3 l P NMR monitoring of the reaction and it was assumed that this peak derived from it1 probably by decomposition under the GC conditions. When analysed by GC-MS, (167) was confirmed as MCBA and the component B was identified as the methyl ester of MCBA (166). Both compounds gave common fragment ions of m/z 139 (base peak) and m/z 111 (Figure 47). Unfortunately, no useful information was obtained about component A . 0~ .,,..OH o~ ..,,..o.Me rn/z = 139 &0~(17) &a (166) (167) ~ o&~+ ~ m/z = 156 m/z = 170 1' /. a l· 0-0 (28) Cl a rn/z= 111 Figure 47. Fragmentation pattern of (166) and (167) 100 (f) Effect of Trimethyl phosphite (TMP). Trimethyl phosphite (TMP) has been used as a reducing agent for peroxides 72 and has been shown to readily remove oxygen from sulfones and sulfoxides.21,25,132 It is also known to abstract sulfur from such substrates as Sg, thiols, disulfides, episulfides and sulfenyl halides.133 Applicable to the present work, Borecka et a[.130 reported quantitative desulfuration of sulfenates to give esters as illustrated below. Swinson et al. 131 carried out the oxidation of trialkyl thiophosphates in DMSQ-d6 and observed TMPS (trimethyl phosphorothionate) too, by addition of TMP to the reaction mixture. ~ 9 (EtO)iP-SOMe + (MeO)JP: (EtO)iP-OMe + (MeO)JP=S (TMP) (TMPS) Since the structure of intermediate E was postulated to have either the sulfenate (-SOCH3) or disulfide linkage (-S-S-), TMP was added to the reaction mixture at the middle stage of the reaction in which a considerable amount of E is present, and then the mixture analysed for TMPS and trimethyl phosphate (TMPO). The latter is expected from the oxidation of TMP by the excess MCPBA and/or deoxygenation of a sulfone or sulfoxide if the sulfenate sulfur had been oxidized further by MCPBA. The various reactions possible are shown in Scheme 13. A complication which had to be taken into account was that TMPS, which could be formed by the reaction of TMP with E, may be oxidized to TMPO by MCPBA before the oxidant is reduced to MCBA by TMP. Representative 3 l P NMR spectra from the experiment are shown in Figure 48. Before the addition of TMP a small amount of BPS was noted to remain and a new peak at B 12.8 was observed together with other peaks which were usually observed in these ;j oxidation~including the intermediate E (Figure 48-A). The peak(<> 12.8) was identified as a hydrogenphosphonate by its large coupling constant OP-H = 712 Hz). Excess TMP was added until the TMP peak was observed in the spectrum and then the reaction mixture was allowed to stand (24 hours). 101 1 ~O 1 S0Me ~o 0Me (MeOhP=S r-·-~o/P~o + r- -~o/P~o TMP (TMPS) CH20H CH20H (149) (118) MCPBA TMP 0~ SOMe(?) ~o ,SOMe 1 (Me0)3P=O o /P~O TMP r--~o/P~o + (TMPO) 0 CH20H CH20H IMCPBA (150) (149) (CH30hP (TMP) Scheme 12. Reaction scheme. In the final spectrum several interesting resonances were noted. Cis-(118), trans-(118) and BPO as well as that for TMPS was observed, clearly suggesting that the intermediate E has a sulfenate or a disulfide structure. The occurrence of both TMPO and (Me0)2P(O)H were also confirmed. The peak at 6 36.4 was not identified but it appears not to be related to E because it was not observed when the latter disappeared on addition of TMP at the early stages of addition; moreover oxidation of TMPS did not give such a peak in a control reaction. Since it was possible that the TMPS might be formed by trans-sulfuration between TMP and BPS, a control reaction was also carried out. However, this experiment gave a negative result. Therefore the rationale for the formation of TMPS is reliable and the experiment confirmed the postulated structures for the intermediates. 102 (A) E o 12.8 (118) J BPS BPO .,,/ 65 60 55 50 45 40 35 30 25 20 15 10 PPM 5 0 -5 (B) (Me0)2P(O)H '- TMPO / o36.4 TMPS Figure 48. Effect of addition of TMP to the reaction mixture. {31 P NMR spectra) 103 (G) Investigation of formation of hydrogenphosphonate. To further investigate the peak at 6 12.8 (hydrogenphosphonate, J = 712 Hz) in Figure 48 a series of further spectra were recored. The same peak was observed for more than 11 hours and it was even more significant than intermediate E after 10 minutes of reaction. However, after 1 hour E became the major intermediate and in the final mixture (85 hours) the five peaks which are the usual products observed from oxidation of BPS in CD30D were noted in earlier experiments. On the basis of the coupling constant (712 Hz) and chemical shift (6 12.8) this hydrogenphosphonate was assigned as the acyclic hydrogenphosphonate (170). A similar hydrogenphosphonate (155) (JPH = 711Hz) was found on oxidation of the monocyclic isomers cis- and trans ( 116) in aqueous acetone. The formation of (170) was rationalized as shown in Figure 49; viz acid catalized methanolysis of (154) can yield the acyclic hydrogenphosphonate (170). -{S~P=S-- - (BPS) (154) CH30H lH' Figure 49. Formation of (170). 104 The reaction was repeated and once the starting material (BPS) had disappeared (3 hr) water was added to the reaction mixture to investigate any changes of the peaks, especially of the intermediate E and of the hydrogenphosphonate (170). It was observed that intermediate E decreased by 30% within 5 minutes of addition of water; without added water a comparable decrease would require about 1.5 hours. After 4 hours only a trace of E remained, again a process which usually took more than 20 hours when water had not been added. At the end of 4 hours hydrogenphosphonate (170) had increased by 1.8 fold. The addition of water also caused a significant change in the ratio between the stereoisomers of the final product dioxaphosphorinanes i.e. cis-(118) and trans-(118). This was 1 : 1 (cis : trans) before the addition of water, 1 : 2.2 after addition, and finally 1 : 2.6 at the point where (170) disappeared (84 hr). In summary, the addition of water enhanced the decomposition of intermediate E and resulted in a higher yield of (170). The cyclic product trans-(118) was also increased, but cis-( 118) showed negligible change. The predominant formation of trans-(118) from the E again supported the trans geometry suggested for Eby the 3lp, 1H NMR and 2-D heteronuclear NMR experiments described earlier. 105 Oxidation of Triethyl phosphorothionate (TEPS) and trimethyl phosphorothionate (TMPS) As discussed so far, oxidation of the monocyclic isomers cis-(116) and trans (116) and bicyclicphosphorothionate (BPS) in chloroform or methanol yielded interesting results but the structure of intermediate E remained in question. Therfore, to further probe the intermediates involved in the oxidation both TEPS and TMPS were oxidized in CDCI3 or CH3OH as model systems. (A) In CDCl3 In CDCl3 total conversion to the corresponding phosphate, TEPO and TMPO respectively, occurred in both cases, accompanied by a sulfur precipitate. (EtO>JP=S MCPBA (TEPS) MCPBA (MeO)JP=S (MeO)3P=O + [S] (TMPS) (fMP()) A more complex situation resulted when TEPS was oxidized in CH3OH since several products or intermediates were detected. Only a trace of TEPS was left after 5 minutes (Figure 50) and after 20 minutes three of the peaks( 6 29.9, 28.2 and 26.5) disappeared. The rest of them remained and were considered as final products. 106 Of these remaining resonances the peak at 6 -0.9 was identified as TEPO by 31 P NMR, GC and GC-MS. In the mass spectrum an ion at m/z 155 was considered to form by elimination of C2H3. A further loss of C2H4 gave m/z 127 and this fragmented again to give m/z 99, which is the base peak (Figure 51). In addition diethyl methyl phosphate (69) and dimethyl ethyl phosphate (126) and TMPO were also tC: idefed using the same procedures described for TEPO. 6-0.9 I 6 o.3 I 6 29,9 628.2 6 1.5 \ (( 626.S '-.. / 625.2 / 6 2.6 I/ 623.5 '-to. , (621.7 70 50 30 10 0 -10 Figure 50. Oxidation ofTEPS in CH3OH. (31 P NMR) C2H5O- P.~ OH C2H50/ 'oH m/z 155 (TEPO) rn/z 182 + + HO-P.~OH C2HsO- P.~ OH + C2H4 HO/ 'oH ----- HO/ 'oH rn/z 99 m/z 127 Figure 51. Proposed mass spectral fragmentation pathways of TEPO. ...... ____ _,_.___-____ _ ...... ,.,...... -...... • • ~t.- ... - : 107 Product (69) must be formed by phosphorylation of CH3OH during the course of the oxidation and its mass spectral fragmentation pattern was similar to that of TEPO. For dimethyl ethyl phosphate (126) an ion at m/z 127 (M-C2H3) was observed although the molecular ion (m/z 154) was absent. For TMPO an ion at m/z 110 was observed as the base peak, formed by loss of CH2O from the molecular ion (m/z 140); this is common fragmentation pathway for methyl phosphates. Of the P-containing products identified in this reaction mixture, TEPO and (69) were expected but (126) and TMPO were not since they represent further ester exchange. Bis (diethoxy phosphinyl) disulfide (130,621.7) was also noted, unequivocally confirmed by comparison of 3lp chemical shif~and spiking with authentic compound into the reaction mixture, as was bis-(dimethoxy phosphinyl) disulfide ( 131, 6 25.2). This latter product was also produced as the only product in the oxidation of TMPS in CH3OH. On the basis of these assignments the product at 6 23.5 was considered to be bis (methoxy, ethoxy phosphinyl) disulfide (171) since it and the other two compounds (130) and (131) are separated by regular chemical shift differences (1.7 ppm). This would only occur if they have a close structural relationship,, e.g., varing only in the number of methoxy groups. A possible reaction mechanism which rationalizes these observation is presented in Figure 52. The resonances at 6 30.0, 28.2 and 26.5 may be assigned as the intermediates (172), (173) and (174) because they have short life times and were separated by regular chemical shift differences of 1.7 ppm, as observed previously for (131), (171) and (130). These results are interesting when compared to the work by Bielawski et a/.26 who showed (69) and diethyl sulfenate (146) as the products in the oxidation of diethyl phenyl phosphorothionate (65) in CH3OH. 108 (EtOhP=S ( TEPS) MCPB1 [ (E Q\..P.....-~ ] -~--- (EtO)JP=O t FJ -..... o ...... [S] • ( TEPO ) (l~~eOH ~ 0 O ~ II II (EtOh- ...- S] (EtO)iP-S-S-P(EtO)i [ MeO/ P...._b (130) */ 0 0 / (173) EtO, ~-s-s-i-OEt ~ l?([Ol/MeOH) MeO...- ....._OMe IMeOH [~~ O t II (171) '* (EtO)iP-OMe EtO- p/~ ] (69) ? ( [0]/MeOH) " [ ! (MeO)i _.. '-o (174) ~ ~ ~ 0 0 (MeO)iP-OEt [S] II II (126) !MeOH (MeO)iP-S-S-P(MeO)i (131) [ (MeOl,P::J] / [S] ~ (MeO)JP=O (TMPO) 0 0 II II (RO)iP-S-S-P(OR)i Figure 52. Proposed mechanism for the oxidation of TEPS in CH30H. 109 Based on the literature the phosphoroxathiirane (18) appears to be generally accepted as a first intermediate in the peracid oxidation of phosphorus compounds such as (23). However, it has not been isolated and its detection is difficult since it has high reactivity and a short life time. As an alternative structure to the' phosphorooxathiirane (18), the 1,3-dipolar structure (20) has also been considered and if this is the case it might be trapped as the dimethyl maleate adduct (175) as shown below. With this possiblity in mind, and as a model experiment, TEPS was oxidized by MCPBA in dimethyl maleate as solvent. However, only TEPO, the rapid desulfuration product, was obtained. The result suggested that dimethyl maleate was either not sufficiently reactive to trap the intermediate (20), or that the intermediate is not involved. This approach, therefore, was not pursued further. R--....._ p-9-S J R1__.... -- XR2 (23) (18) (20) R, R 1, R2= Alkyl, Alkoxy, Aryl (175) 110 (D) In agueous acetone. To further investigate the oxidation of phosphorothioates in aqueous acetone, the acyclic analogue of BPS was studied. One treatment with 4 equivalents of MCPBA TEPS was converted, within 1 hr, to a mixture of diethyl hydrogenphosphonate (128) and TEPO. A simple transient resonance (o 29.8) was noted, which presumably represented an intermediate in the transformation. After one week the diethyl hydrogenphosphonate (128) was totally hydrolysed; first to ethyl hydrogenphosphonate (176) and then to phosphonic acid (177). It is of interest to note that ethyl hydrogenphosphonate (176) currently finds use as a fungicide under the name of Fosetyl, 1 and that its presumed mode of action is via hydrolysis to phosphonic acid. 0 0 II II F.tO-P-H HO-P-H I I OH OH (176) (177) With less oxidant TEPS did not react completely and disulfide (130) was identified by direct comparison with authentic material was observed as an intermediate. In the final reaction mixture diethyl phosphorothioic acid (10) was also present, presumably arising from the disulfide, as well as unreacted disulfide (130), diethyl hydrogenphosphonate and TEPO. Segall et al. 32 reported the formation of hydrogenphosphonates (128) in the oxidation of a phosphorothioic acids (10), and suggested that the reaction mechanism involved the isomeric phosphoro(thioperoxoic) acids (Scheme 5). On the basis of that report the intermediate at 8 29.8 in the present study is assigned as diethyl phosphoro(thioperoxoic) acid (178), and a reaction sequence is proposed as shown in Figure 53. 111 (EtO)JP=S ( TEPS) MCPBAl /OH .. HO/H+2 (EtOhP~ [ (EtO>Je::: b] -...::::.::[S] .. (EtOhP=O s (TEPO) (10) / (172) 0 0 [ (EtO)~~OH II II J (EtO)iP-S-S-P(EtO)i (178) (130) MCPBAj ? [ (EtO)iP~OH] MCPBA /~ SOH J H2o (EtOhP~/H [(EtOhP~O~ ------.. 0 (128) Figure 53. Proposed reaction sequence for the oxidation of TEPS in aqueous acetone. In this scheme it is proposed that the oxygenated intermediate (172) can produce TEPO by direct desulfuration. Structure (178) is assigned as the intermediate which is observed at o 29.8. Subsequently, (178) is oxidized further to yield diethyl hydrogenphosphonate (128). As an alternative route it also leads to the disulfide (130), which in turn may produce diethyl hydrogenphosphonate (128), or decompose to diethyl phosphorothioic acid (10). The latter product appears to be of greater significance when the oxidant is limiting. T,his differs from the proposal of Segall et al. 32 in that the formation of the disulfide is now accounted for. 112 Oxidation of the sulfenate (EtO)iP(O)SOCH3 (146) by MCPBA in CD30D Since the intermediate E in the oxidation of BPS and the monocyclic isomers cis and trans-(116) in CH3OH was suspected to be the sulfenate or disulfide and since diethyl sulfenate (146) was found as product in the oxidation of diethyl S.-phenyl phosphorothionate (65) in CH3OH,26 direct oxidation of (146) was carried out in CD3OD to determine any relationship of the sulfenate to the intermediate E. Representative spectra from this reaction are shown in Figure 54. As reported by Borecka et al. l30 sulfenate (146) was unstable and contained minor unidentified impurities as well as diethyl methyl phosphate (69) formed by desulfuration of (146). With the addition of the first portion of MCPBA, the starting material disappeared completely to produce a new peak at o22.7 and increase of (69) was noted (Figure 54- A). In the lH-coupled 3lp spectrum the peak at o 22.7 was observed as a quintet (JrocH = 9.1 Hz) indicating that the diethyl substituents were still present in the molecule. With this information, and on the basis of chemical shift and coupling values, this product was assigned as bis(diethoxy phosphinyl) disulfide (130). More MCPBA converted (130) to diethyl deuteromethyl phosphate (69-d3) (Figure 54-B). A control experiment showed quick decomposition (15 min) of the sulfenate (146) in CD3OD at room temperature in the absence of MCPBA to many products different from those with MCPBA. Based on the above results the reaction pathway is postulated as shown in Figure 55. In this sequence (Figure 55) the phosphate (69) and the disulfide (130) arise after the first addition of MCPBA and (69) is formed by subsequent decomposition of the unstable sulfenate (146). Compound (130) arises from an oxidative coupling reaction. With the second addition of MCPBA the disulfide (130) is oxidized and the reactive species phosphorylates CD3OD to yield (69-d3). 113 (A) (130) (69) ~ ../ ; (B) (69),(69-d3) 65 60 !55 50 45 40 35 30 25 20 15 10 5 0 -5 -10 -15 PPM Figure 54. 3lp NMR spectra from oxidation of (146) in CD30D. 0 II (EtO}iP-SOMe (146) Al ? ~ 9 (EtO)iP-S-S-P(EtO}i + (EtO}iP-OMe (130) (69) 9 (EtO)iP-OCD3 (69-d3) Figure 55. Proposed reaction pathway for oxidation of sulfenate (146) in CD30D. 114 Both the sulfenate and disulfide structures were proposed earlier in this thesis for intermediates D and E which arise in the oxidation of BPS and the monocyclic isomers cis- and trans-(116). However, in the acyclic system disulfides were observed in the reaction mixture and sulfenate converted to disulfide, which appeared to be an intermediate for funher oxidation. Therefore it appeared that the disulfide structure for D and E is the more reasonable at this stage and the three disulfides observed in the oxidation of TEPS could be the intermediates which were noted when MCPBA was limiting. Reaction of trimethyl phosphite (TMP) with Bis(diethoxy phosphinyl) disulfide. Disulfides were observed as intermediates in the oxidation studies with TEPS and TMPS, and also from oxidation of the sulfenate. Therefore, on the assumption that I a disulfide is the common intermediate in the oxidation of both acyclic and cyclic thioate compounds its reactions with TMP were investigated. In CH3OH, disulfide (130) rapidly reacted with TMP to give three new compounds. These were TMPS, diethyl methyl phosphate (69) and diethyl phosphorothioic acid ( 10). A reaction mechanism is proposed (Figure 56) in which the phosphite acts as a nucleophile leading to cleavage of the disulfide. The intermediate (179) phosphorylates the solvent to produce the mixed ester (69) and TMPS. 115 0 0 ~o II II TMP II + + (EtO)iP'-. (EtO)iP-S-S-P(EtO)i [ (EtO)iP-S-P(OMe0 h~ (10) s- (130) (179) MeOH 0 II (EtO)iP-OMe + (MeO)JP=S (69) (TMPS) Figure 56. Proposed reaction mechanism on the reaction of TMP with disulfide (130). Oxidation of bis(diethyl phosphinyl)disulfide (130) by MCPBA. Since bis(diethyl phosphini~) disulfide (130) was found as one of the products in the oxidation of TEPS, and ashntermediate from the oxidation of sulfenate, this disulfide was oxidized with MCPBA in CH3OH under same conditionsand diethyl methyl phosphate (69) was obtained as the only major product. From GC analysis, MCBA (167) and its methyl ester (166), which were the common by-products in all oxidations of BPS and the monocyclic isomers (116), were also detected. No change was observed in a control experiment involving the disulfide (130) and MCBA in CH3OH, thereby verifying that the transformation was oxidatively induced. 116 Oxidation of the disulfide (130) in aqueous acetone was also investigated to see whether diethyl hydrogenphosphonate (128) formed from disulfide (130) since both products were observed in the oxidation of TEPS in aqueous acetone. In this case, however, only one intermediate peak (o 16.9) was noted and diethyl hydrogenphosphonate (128) and diethyl phosphoric acid (127) were both formed. The latter product was not observed in the oxidation of TEPS in aqueous acetone. In the control experiment (no added oxidatant) diethyl phosphorothioic acid (10) and diethyl phosphoric acid (127) were observed and these are probably formed by hydrolysis of disulfide ( 130) (Figure 57) . ~ f? (EtOhP-S-S-P(EtOh (130) ..,...... H ..,...... OH ..,...... OH (EtOhP.:::::- (EtOhP.:::::- (EtO)iP.:::::- 0 0 S (128) (127) (10) Figure 57. Products formed from the oxidation of disulfide (130) in aqueous acetone. 117 Oxidation of 0,0-diethyl-S-phenyl phosphorothiolate-18Q (180 - thiolate, 186) by MCPBA in CH30H. Segall and Casida22 noted that .S.-alkylphosphorothiolates (180, Figure 58) react with MCPBA to form phosphinyloxysulfonates (182) possibly by rearrangement of the phosphorothiolate S-oxide (181) (Pathway A). The oxysulfonate (182) was reported to give methyl ester (183) in methanoI.26 In the present study, however, oxidation of the disulfide (130) in aqueous acetone gave diethyl hydrogenphosphonate (128) as one of the products and in view of this observation another possible mechanism (Pathway B) can be postulated. In particular, pathway B may operate in which the phosphorothiolate S-oxide (181) rearranges to phosphite (184) via attack by the phosphoryl oxygen on the I S=O funtionality rather than sulfuryl oxygen on the P=O funtionality. The phosphite (184) would then be oxidised further to the oxysulfonate (182) which would be expected to give methyl ester (182) in methanol as also arises in pathway A. The phosphite (184), which is in fact a mixed anhydride, should phosphorylate methanol to give the phosphite ester (185) which would readily oxidise further to the methyl ester (183). In aqueous media the phosphite (184) may also hydrolyse to produce hydrogen phosphonate (188), and this would be relatively insensitive to further oxidation. To investigate these alternative mechanisms, 0,0-diethyl-S,_-phenyl phosphorothiolate-•so (186) was prepared as shown below. 3lp NMR showed 88 % of 180 was incorpfrated in (186). By using this as substrate for the oxidation study the mechanism can be probed. Pathway A will give the product as diethyl methyl phosphate-18O (187); pathway B will give diethyl methyl phosphate (69), which does not contain 180. 180 180 ~ PhSSPh ~ (Et0)3P: + H2180 --(EtO) P (EtOhP\_ 2' H SPh (186) 118 R10- ,;::::;-0 R20,.....P..._ H (188) [0] [ R10-p- 0Me7 R20,..... J (185) Figure 58. Proposed mechanism for the oxidation of thiolate. 119 When this oxidation was carried out, only one product was found from oxidation of the 180-thiolate (186) in methanol and it was determined as 180-diethyl methyl phosphate (187) by GC-MS analysis. This result suggests that the originally proposed pathway A (Figure 58) is followed. In support of this conclusion the ratio of 180 to 160 is the same in both starting material ( 186, 129) and product ( 187, 69) within the limits of error of the 31 P NMR and MS experiments. 180 1/0 -:;::::.180 -:;::::. -:;::::.O (Et0)iP...... _ (Et0)iP...... _ (EtO)iP...... _ (EtO)iP...... _OMe SPh SPh 0Me (186) (187) (69) (129) The above conclusion not withstanding, the oxidation chemistry of phosphite (184) in pathway B was investigate relative to the formation of hydrogen phosphonate I (188). In principle, hydrolysis of (184) would lead to the latter class of compound. However, with the phosphorothiolates (129) and (139) as substrates in aqueous acetone only phosphates (127) and ( 135) were obtained as presented below thereby indicating that hydrolysis is not involved. MCPBA (EtO)iP~~Ph (EtO)iP:gH Aqueous acetone (129) (127) o, ,.,SPh MCPBA )C ,P~Q ------ Aqueous acetone 0 (139) (135) These observations could be rationalized as shown in Figure 59 in accord with with the earlier work by Bielawski and Casida26 who argued that (189) formed as a result of water present in the solvent. 120 R...... _ -:;:::-0 p [0] .. R1...... - -.. SPh R, R1, = Alkyl, Alkoxy, ~ Aryl [:~p::~SPJ R...... _ -:;:::-0 R 1,...... P-._ OH [0] (189) R...... _ -:;:::-0 [ J R i.....- P-._ O~Ph 0 (190) Figure 59. Formation of ( 189) on oxidation of thiolate. Recently, Yang et a/.33 studied the oxidative hydrolysis of O,S-diethyl methylphosphonothiolate (89) by MCPBA in aqueous solvents in which O-ethyl methyl phosphonate (92) was obtained (Figure 7). They concluded that the hydrolysis of (89) occurred only after the sulfur was oxidized to the sulfoxide (91) because this step was found to be much faster than the further oxidation of sulfur. They therefore concluded that formation of the sulfonate ester such as ( 190) in Figure 59 was not part of the reaction sequence. 121 Oxidation of 5-hydroxymethyl-5-methyl-2-thiono-2-ethoxy-1,3,2- dioxaphosphorinane (116) by S02Cl2 in CDCl3. Up to this point the intermediate E from the oxidation of BPS or cis- and trans ( 116) in CH30H had been assumed to be a sulfenate (149) or disulfide (168) and the latter was favored. Disulfides were also found to be intermediates in the oxidation of acyclic compounds. To further investigate this, the possiblity of the synthesis of the monocyclic sulfenate or disulfide was tried with the reaction between the trans- (116) and S02Cl2 using the 1· ·1 method reported by other researchers. 130, 134, l35 However, the reaction of trans-(116) with S02Cl2 gave an unexpected and interesting result since BPO was obtained as the only product (Figure 60). This product was isolated and fully characterized. A mixture of cis and trans-( 116) again gave the same result without any difference in the reaction rate between the cis and trans isomers. This represents a rapid and effective intramolecular displacement with retention of configuration (trans isomer to BPO) and inversion of configuration (cis isomer to BPO) (Figure 61). Trans-(116) /OEt (BPO) 0-P~/ ~s 0 Cis-(116) Figure 61. Formation of BPO from the raction between (116) and S02Cl2. 122 Trans-( 116) S. D 4.5 4. D 3. S 3. 0 2. S 2,D l.S I.D PPM ..-----EtCl s. 0 4.6 4. 2 3.8 3, 4 3,0 2. 6 2.2 ,.a , • 4 I. D • 6 Figure 60. 1H NMR spectra of starting material (A) and product (B) from the reaction of trans-(116) with s0ic12. 123 To study the mechanism of this transformation two compounds were chosen; 5,5-dimethyl-2-thiono-2-ethoxy-1,3,2-dioxaphosphorinane ( 133) and 5- hydroxymethyl-5-methyl-2-oxo-2-ethoxy-1,3,2-diozaphosphorinane ( 117) since these would allow the influence of the hydroxymethyl group and the effect of the sulfur atom to be determined. With ( 117) and S02Cli, no BPO was formed; rather the hydroxyl group reacted with S02Cli to give chloro-sulfonyl derivative (191). The other substituent groups were not altered. ~O /OEt !-~o/~o CH20S02Cl (117) (191) With 5,5-dimethyl dioxaphosphorinane (133) and S02Cl2, the disulfide (192, 8 12.7) was isolated as a major product. With this result in hand, it was now concluded that neit.her intermediate E nor D had the disulfide structure since they had substantially different chemical shifts. which would not result from the introduction of a remote hydroxyl funtionality. From 31p monitoring of the reaction, it could be seen that both the disulfide (192) and sulfenyl chloride (193) were present in the initial stages of the reaction. Excess S02Cl2 converted all of the disulfide (192) to the sulfenyl chloride (193) (Figure 62), a result similar to that observed by Krawczyk and Skowronska l35 for bis(4-methyl-2-oxo- l ,3,2-diozaphosphorinane) disulfide ( 194) 124 + RO + SOi 0 O'p-S-S-p// OJ< + RCl , I/ , O o 0 (192) Figure 62. Reaction of (133) with S02C!i. or In contrast to these results, no detectable disulfide =-= the sulfenyl chloride such as (193) or chlorosulfenyl compound such as (191) was observed in the formation of BPO from isomeric ( 116) under similar reaction conditions. Thus, the cyclization to BPO must occur at a very early stage of reaction via intermediates whose short life time precludes their detection by 3Ip NMR, respectively. The reaction mechanism for BPO formation is therefore proposed as shown in Figure 63. 125 er ~O /OEt r- . ~O/pl>S S02Cl2 CH20H (116) -{S~P=O (BPO) + EtCl + S02 + SCI Figure 63. Proposed mechanism for the formation of BPO from the reaction between ( 116) and S02Cl2. Reaction of BPS with S02Cl2. In view of the above results (BPO formation) and as a further variant BPS was reacted with S02Cl2. Two products were obtained. One was BPO (minor) and the other (major) was a single isomer of 5-chloromethyl-5-methyl-2-oxo-2- chlorodioxaphosphorinane (195). This was fully characterised, and its geometry was elucidated by a single crystal X-ray analysis (Figure 64). The chloromethyl group and the chlorine on phosphorus were shown to be in a trans relationship . The chlorine is axial and the phosphoryl oxygen equatorial, consistent with the known general 126 stereochemical preference for this latter functionality in cyclic esters 119,120. In a control study BPO did not react with S02Cl2. This suggests that the chloridate (195) forms directly from BPS in a stereospecific manner, and not via any intermediate oxidation product. The mechanism of formation of BPO and ( 195) from BPS is proposed as shown in Figure 65. --{S~>=s (BPS) OH- Figure 65. Proposed mechanism for the formation of (195) from the reaction between 127 Figure 64. X-ray crystal structure of (195) and numbering of atoms. 128 The final positional parameters for (195) are listed in appendix C and the numbering of the atoms is depicted in Figure 64. The anomeric effectl36-l40 causes the flattening of the ring around phosphorus, and this can be seen from the ring torsion angles (Table 5, appendix C) or by comparison of the angle Al and A2 in Table 6, (appendix C). Bond lengths and angles are also shown in appendix C and they are normal, as compared to other related substituted 1,3,2,-dioxaphosphorinanes, 5,5- dimethy 1-2-chloro-2-oxo-( 134) 141, 5 ,5-dimeth y1-4-isopropy 1-2-chloro-2-oxo( 196) 142 and 2-bromo-5-bromomethyl-2-oxo-5-methyl-1,3,2-dioxaphosphorinane (197) 143, 144. (Figure 66). 5-Chloromethyl-5-methyl-2-oxo-2-chlorodioxaphosphorinane ( 195) has been widely used to study the stereochemistry of nucleophilic displacement at phosphorus.101, 103, 104, 106-110, 113, 114 Compound (195) was made via a Michaelis- Arbuzov reaction, by slow addition of a carbon tetrachloride solution of the bicyclic phosphite (198) to sulfuryl chloride. It is noteworthy that the a:;;~11:e:11ei B53&E!l=il:IJ a19:i:i:13Fi:e:o:i:i:te. reverse addition leads to a polymeric, intractable material. 113 ~C /Cl l.~ ~o/~o Cl (195) 129 Chlorine gas has also been usedl03,108,I 45 to effect this transformation but the S02Cl2 reaction gave a purer product. In earlier work, the structure of the chloridate ( 195) was established from analysis, spectra and considerations of its mode of formation. 113 It was assumed that it was stereochemically analogous to 2-bromo-5- bromomethyl-2-oxo-5-methyl-1,3,2-dioxaphosphorinane ( 197), a compound prepared by adding bromine to the phosphite, and whose configuration was established by X-ray analysis.143·144 A trans relationship was noted between the bromine and bromomethyl substituents. The present study is the first crystal structure determination of the chloro compound. 1 Br 0-P~/ ~o 0 (197) The stereochemical course in ring cleavage reactions of phosphites has also been extensively studied, 104, 146-150 A wide variety of ring compounds of type (199) have been used 103,107,108,119,145,151-153 in which the R2 and CH2X substituents have a trans relationship. R1B-P~R2 / "o 0 X (199) 130 In solution, the conformational mobility of these types of compound was indicated by NMR studies. 11 9,152 In particular, when R2 = Me or Brand X = halide, the favoured conformation possesses axial P-Me (Br) and CH2X groups. In analogues where R2 = Ph3C and X = halide, at least two conformers appear to be present in approximately equal amounr.119 Alkyl groups on phosphorus when X = halide apparently provide sufficient steric bulk to accomodate an equilibrium between the two possible chair conformations. In contrast, NMR evidence suggests145 that when R1= Me, R2 = NC5H 10 and X = Cl, the dominant conformer (200) possesses equatorial NC5H10 and CH2Cl groups in solution. 103,I08,I 45 The opposite conformation was postulated when R1 = Er.107. p O-P- NC5H10 / 0 (200) (201) The conformation shown in (200) was confirmed for the closely related cyclophosphamide (201) by X-ray analysis.146 In the present work the formation of the Michaelis-Arbuzov product (195) from BPS (Figure 65) suggests that the reaction can be extended to include the thiophosphoryl compounds as well as phosphite. 131 1.465 1,487 (134) ( 196) CH(Me)i 0 -1.4-s1- -<:.,_ 1 go "'- ~~:; o~~o ,~b1/p~ 1.471 y ··/;,"),..._ / p'Z.oo;, 0 / Br 1.465 o ""<- a (197) (195) 1N~3-~O 11?,~4-~o 105 .0 P 112.0 106.6 p 111.6 \'2.1•0Q4.~-•v \ 7.o.17'os .a"--.···· '----o · "-a ---o ., "-a (134) (196) CH(Meh a 12?0~6-~o 11~~5.3~0 llO~ p 115.5 I 10~ p 111.5 \ \9·/ 104.0··- \\ Figure 66. Bond lengthes and angles of halogenated 1,3,2-dioxaphosphorinanes. 132 Alcoholysis of cyclic phosphates Since there are a number of mechanistic similarities between hydrolysis and oxidation reactionsand to better understand the oxidation reactionsJ the alcoholysis of some cyclic phosphates and phosphorothionates were carried out in different solvents. The reactions were monitored by 31 P NMR (Figure 67). Methanolysis of BPO gave three products which were identified as cis- and trans-5-hydroxymethyl-5-methyl-2-oxo-r-2-methoxy-1,3,2-dioxaphosphorinane [cis and trans-(118). respectively ] and acyclic phosphate (202) (Figure 67-C). By monitoring the reaction with 31 P NMR the stereochemical changes that occur during the reaction which were not described by other authers 99 could be clearly demonstrated. During the reaction trans-( 118) was formed first followed by the appearance of acyclic phosphate (202) and following cis-( 118) (Figure.69). The variation of products with time is plotted in Figure 68. (%) 120 ~------ 100 C BPO -----N---- Trans-(118) o Cis-(118) 80 -·-·-·•-·-·· Acyclic Phosphate (202) .. x.. , , .. x,. 60 .. , ,,:-.----·- ·•·-·-·-·-•·-·-·-·-· -·-·-·-·-.·-·-·-·-•·-·-·-·-· 40 /11 .,. ______,· .---·------11------,c 20 • ----o~--~oo------0-0----00---00 0 -f'K-.....;.--.T-_---,~-----.---.....------~ 0 50 100 150 200 Hours Figure 68. Changes of products during the methanolysis of BPO. 133 (A)2 Hours BPO I Trans-( 118)--- {B) 13 Hours Acyclic phosphate (202) ./ (C) 180 Hours (Final mixture) Cis-(118) '-- . I 1 9 e 1 s s 4 3 ~ P~H -1 -2 -3 -4 -5 -6 -7 -B -9 Figure 67. 31 P NMR spectra for methanolysis of BPO. 134 -{S~p.o _o_M_e·--~/~:- (BPO) Trans-(118) / OH ,OMe 0/P~ 0 0 Cis-(118) Figure 69. Reaction pathway in the Methanolysis of BPO. From the NMR data it appears that the initial ring opening of the BPO to the trans-isomer (118) is stereospecific (Figure 67-A). Thus, the reaction proceeds directly to the acyclic phosphate via the trans-isomer until all the BPO is consumed The obseivation of cis-(118) as a product is interesting because it did nofarallel the formation of the trans -( 118). Rather it is obseived at a later stage of the reaction when the BPO has almost disappeared (Figure 67-B). Thus cis-(118) may result from (i) direct SN2 (P) substitution with inversion at phosphorus of trans isomer (118) or (ii) re-closure of acyclic phosphate (202) to form a ring followed by re-opening of the ring. To further probe the pathway involved in the formation of cis-(118), trans-(l 18) and acyclic phosphate (202) were each reacted with methoxide. In each case the same mixture was obtained as noted above from the methanolysis of trans-(118), and the products appeared in the same sequence (Figure 70). However, in the methanolysis of acyclic phosphate (202), trans-and cis-isomers (118) were formed at the same time. This indicates that ring closure of the acyclic system must be involved in the reaction and presumably an equilibrium is reached (Table 7). In the above experiments, the trans 135 isomer was always formed in greater amount than the cis-isomer which suggests that it is the thermodynamically more stable of the two. 'e/~7 Trans-(118) + + y Figure 70. Scheme for methanolysis of trans-(118) and aC4clic phosphate (202). Table 7. Ratio between the products from the methanolysis. (%) . I "' Trans-(118) Cis-(118) Acyct phosphate ~ts (202) BPO 1.7 1 2.6 Trans-( 118) 2.0 1 3.5 Acycluc phosphate 1.6 1 3.5 (202) 136 On the basis of the findings to this point, the reaction mechanism for the methanolysis of BPO is postulafd as shown in Figure 71. °'. ( OMe· [~°'. +/ OMe ] ~~P~-,------~~'6- (BPQ) l /,OMe ~\+ ,--..._ P -o· I\ 0 OMe CH Cis-(118) Trans-(118) Figure 71. Proposed mechanism for the methanolysis of BPO It should be noted that such ring opening and re-closure process are known. For example, Wadsworth et ai. 11 2 reported that these reaction occur with 2-substituted- 5-chloro-methyl-5-methyl-2-oxo-1,3,2-dioxaphosphorinanes. 137 Ethanolysis of BPO was also carried out. The sequence of the product formation and product distribution [1.4: 1: 3.1 (trans: cis: acyclic)] was similar to that found for methanolysis. oo ~OH f ;OEt -~/P~ + o-p'oo o,...... P\ o 0 0 / ~ + OH O O OEt ID To probe the influence of the thiophosphoryl group the alcoholysis reactions were repeated with BPS. Apart from a much slower reaction similar results were obtained to those from BPO, i.e. in the formation of three products, in their sequence of formation and ratio between them . OH ;OMe 0.. ,OMe OMe f - '-. ~0-P ' ,...... P-S __t;;;:_O_ P=S - / ~ + + ~er' 0 s In the present studies the stereospecificity of the initial ring opening of the bicyclic systems (BPO, BPS) to the trans-isomers in the alcoholysis reactions was confirmed by an X-ray crystallographic study, specifically of trans-5-hydroxymethyl-5- methyl-2-oxo-r-2-methoxy-1,3,2-dioxaphosphorinane [trans-(118), Figure 72] and its 2-thiono-analogue (203, Figure 73) which were isolated from methanolysis of BP6 and BP5: respectively. The trans relationship is clearly observed between the methoxy oX~i>en group and hydroxymethyl group, and the phosphoryl ~ takes the expected equatorial orientation. A similar stereochemical observation was reported for the hydrolysis of BPS by Farini et al _95 A number of stereochemical studies using 1,3,2- dioxaphosphorinanes have been undertaken and much of this reviewect.154-156 138 Recently, Gallagher 157 rewiewed this area including other ring systems, and the application of 31 P NMR to problem solving in this field. With the above two X-ray results available, some comments can be made about the conformation of related monocyclic molecules. In terms of conformational tendencies, the P=O bond has a strong tendency to occupy the equatorial position in phosphate esters ll9-l21 and the axial alkoxy group is favoured for 2-alkoxyphosphates 119, 124, 158-l6l, However, for 2-alkyl 160-165 or 2-amino phosphates 160, 161. 164 the P=O bond goes axial Similar tendencies to these are found in 2-thiono-2-substituted derivatives 166. Y=OR; X=O Y=R, N2R; X=O Y=P(Ph)i; X=O Y=OR, R, Cl ; X=O In solid state, X-ray crystallographic studies 95, l l5, 141-144, 167-209 showed that generally the 1,3,2-dioxaphosphorinane takes up a chair conformation, somewhat flattened at the phosphorus center. However, a half chair (chaise longue) conformation was found 184, 185 for trans-2-triphenylmethyl-2-oxo-4,6-dimethyl-1,3,2- dioxaphosphorinane (204) while for cis-2,5-di+butyl-2-oxo-1,3,2-dioxaphosphorinane (205) a boat conformation was reported l 75. These conformational preferences presumably arise from the effects of the steric bulk of the triphenylmethyl and t-butyl group. Bu-t t-Bu -= ---0...... _ /CPh3 I ---~--"-----o-- P=: 0 ___----0...... -P'-O / (204) (205) 139 0(4) Figure 72. X-ray crystal structure of trans-( 118) and the numbering of the atoms. 140 0(3) Figure 73. X-ray crystal structure of (203) and the numbering of the atoms. 141 These 1,3,2-dioxaphosphorinane systems are also well suited for studing the influence of the anomeric effect 136-140 upon molecular geometry. This effect accounts for the stabilizing interaction of lone-pair orbitals with antibonding s* or p * orbitals at adjacent atoms 210, 211 by 'back donation' from oxygen lone-pair electrons into an antibonding s* or p* orbital. Such an interaction is largest if the effective axis of the lone pair orbital is in an anti position with respect to the bond into which electrons are being transfered. It is also accentuated if this bond is polar and electron-withdrawing. Thus, the conformer with the largest number of lone pairs in an antiperiplanar position to electronegative groups is the most stable. The methoxy group is assumed to be more electornegative than the P=O, P=S and P=Se group, and thus the conformation with methoxy group in an axial position is most stable. The molecules investigated in the present study (Figure 72 and 73) show that the phosphate ring adopts a chair conformation with the methoxy group in an axial position and the phosphoryl group in an equatorial position in agreement with the anomeric effect. A similar conformation was found for the 5,5-dimethyl-2-methoxy-2-selenide l99, -2-oxide 177 (137) and- 2-sulfide 205 (133) derivatives; the opposite conformation was reported for -2-methyl-2-oxide (206) derivatives 207. (206) 142 The final positional parameters for trans-(118) and (203) are listed in appendix D and E respectively, and the numbering of the atoms is depicted in Figure 72 and 73. The attachment of electronegative atoms causes a loss of electron density around the phosphorus atom. The anomeric effect compensates for this loss by back donation of electrons from oxygen lone-pair into antibonding s* or p* orbitals. This causes a flattening of the ring around phosphorus, which can be seen from the ring torsion angles or by comparing the angles Al and A2 in Table 6 in each case. Bond lengths and angles are also listed in appendix D and E and they are normal compared with those of other related 5,5-dimethyl-1,3,2-dioxaphosphorinanes (Figure 74). The POC bond angles approximate 120°, indicating that the oxygen atoms are sp2-like and the C-O bonds in the phosphate ring are significantly longer than the C-O bond for the methoxy group or the C-O bond for hydroxyl group; the latter are those normally found in ethers and alcohols. This is a typical feature of anomeric centres as a result of the anomeric delocalization l39 and similar elongations have been found in other anomeric systems such as acetals 139, 2l2 and cyclic sulfites 213 . Figure 72 and 73 shows that the methoxy group is directed away from the six-membered ring (see angle B, Table 6, appendix D and E) therby avoiding severe steric interaction that would otherwise occur with axial protons on C(3) and C(l). 143 ,---0 I b. 0 1.450 ~ p . ~b1/ p ~/.J'. Y 6s 1.470 1.472 0 0 (137) - ,.,.t Me OH 1.464 Trans-(118) (203) Me Me' ---0 s ---0 0 116.7~12.v 118.o~4-~ 1os%P~6.7 105.7 p 116.0 1 \ \1,1/ ,01.3 "-. '----\\ho/io6.~ 0 -~-0 119.7° (137) 122.5 Me Me ----0 s o ~~o 118.~12.~ 111.1~12-~ _10~ P 115.5 I /c p 116,2 \ \?,.4/ ,02.3' __ _ \\1·9/ 99.~ ---o ., "-o ---O 121.8° 120.6 Trans-(118) (203) Me Me Figure 74. Bond lengthes and angles of related 2-methoxy-1,3,2-dioxaphosphorinanes. 144 Conclusion In summary, the oxidation of bicyclic, monocyclic and acyclic phosphorothioates gave similar products which were solvent dependent In a non-hydroxylic solvent (CDCl3) desulfuration was the main reaction and the phosphate was obtained as major or the only product. For the oxidation of cis- and rrans-5-hydroxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3,2-dioxaphosphorinane in CDCl3, the stereochemistry was usually maintained. The degree of inversion or cyclization to BPO varied dep~nding on the stereochemistry of the starting material. When the above cis- and trans-isomers were oxidized in CH30H methyl phosphates were formed, non stereospecifically, as major products by phosphorylation of the solvent. Similar reaction processes were noticed in the oxidation of BPS in CH30H (or CD30D) and many intermediates appeared during the course of reactions for the both cases. In the case of cis- and trans- isomers two of these were major which for the BPS only one was prominent. In the latter case formation of the intermediate involved stereospecific ring opening to a monocyclic species as the major intermediate which was assigned as trans-5-hydroxymethyl-5-methyl-2-oxo-r-2-methylsulfenyl- 1,3,2-dioxaphosphorinane. However, regardless of the stereochemistry of the major intermediates, the products obtained were a mixture of methyl phosphate isomers. In aqueous acetone , an acyclic hydrogenphosphonate was commonly observed from cis and trans-isomers and BPS. This probably formed via hydrolysis of some hydrogenphosphonate intermediates which occur during the reaction With 5,5-dimethyl dioxaphosphorinanes as comparison compounds, it was found that the hydroxyl group of the cis- and tran;- isomers does not affect the overall reaction in CH30H although it may play a role in stabilizing the intermediates. 145 From the oxidation studies on the acyclic compounds in CH30H disulfides were also found as intermediates which can phosphorylate CH30H to produce methyl phosphates. Alternative reaction pathways were also considered and the rearrangement t:: processe involved was differenJiated using a 180-thiolate analogue. From the oxidation of these starting materials in aqueous solvents, hydrogenphosphonates were 6- observed, and their sufsequent hydrolysis to acyclic compounds occured was also noted. However, for thiolates oxidative-hydrolysis in same solvent system gave the phosphoric acids. The reactions between S02Cl2 and cis- and trans- isomers showed cyclization to BPO. With BPS the stereospecific ring opening to give trans- chloridate was observed. i. The mechanistically related alcoholysts reactions of BPO and BPS were also studied. Stereospecific ring opening to trans-isomer was observed again and the trans and cis-isomer equilibrated via acyclic derivatives . With all the reagents used, the first ring opening of bicyclic compounds was dnti found to be stereospecificito lead to the trans- isomer. Various of the stereochemical aspects of the starting materials and products were also confirmed by X-ray analyses of appropriate compounds. EXPERIMENT AL 146 General Experimental Details. Melting points (m.p.) were determined on Kofler Hot Stage Microscope melting a point apJtatus and are uncorrected. Micro analyses were carried out by Dr. P. Pham, Microanalytical unit, School of Chemistry. All syntheses were carried out in an inert atmosphere (either nitrogen or argon ) unless otherwise specified. Solvents were dried before use. With chloroform the trace of ethanol were first removed by washing with H2SO4 and water. Analytical and preparative t.l.c. utilized layers of silica gel GF 254 type 60 (Merk) of 0.25 and 1.0 mm thickness, respectively. The plates were visualised either under U.V. light, with iodine vapour, or for the compounds containing the P-S linkage by spraying with an acidified (HCl) aqueous PdCl2 solution (0.5% ). Infra red (IR) spectra were recorded with either an Hitachi EPI-G grating spectrometer or a Perkin Elmer 580B spectrometer. The bands were measued in cm- 1 and described as strong (s); medium (m); weak (w) ·or broad (br). lH Nuclear magnetic resonance (lH NMR) spectra were recorded either on a Bruker AM 500 (500 MHz), AM 300 (300 MHz), or CXP-300 (300 MHz) spectrometer n with CDCl3 as inte*11 standard unless otherwise specified. Chemical shifts are quoted on the a scale, and the abbreviations s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet) and br (broad) have been used. 13C Nuclear magnetic resonance (13C NMR) spectra were recorded with a Bruker AM 500 (125.77 MHz) with CDCl3 as interal standard unless otherwise specified. 31p Nuclear magnetic resonance (31p NMR) spectra were recorded either on a Bruker AM 300 (121.47 MHz), or CXP-300 (121.47 MHz) spectrometer generally with ·...... ~-,--·- ...... ,,_ ___ , ..-.. ~-- .. ~--- ·. ,-~~------147 broad band proton decoupling, and with 85% H3PO4 in D2O as exteral standard unless otherwise specified. Some chemical shifts from the quoted literatures have been referenced to different standards, or used the earlier convention of opposite sign relative to 0 ppm. Mass spectra were taken on either a AEI-MS12 spectrometer (operating at 70 eV, electron impact mode) attached to a VG-display digispec data acquisition system computer or a Hewlett-Packard 5985A spectrometer (chemical ionization mode utilized CB4 as the reagent gas). 148 Preparation of 5-hydroxymethyl-5-methyl-2-thiono-2-ethoxy- 1,3,2-dioxaphosphorinane (116). A solution of 1,1,1-tris (hydroxymethyl)ethane (3.6 g, 0.03 mol) and pyridine (4.84 ml, 0.06 mol) in CH3CN (200 ml) was added dropwise to a stirred, cooled (ice bath) solution of Q-ethyl phosphorodichloridothionate (5.37 g, 0.03 mol) in CH3CN (30 ml). On completion of the addition, the reaction mixture was stirred (1 hr) at room temperature, refluxed (2 hr) and then the solvent evaporated. The residue was taken up in CHCl3, washed with dilute HCl solution (10 ml x 2, 5 % w/v), NaHCO3 (10 ml x 2, 5% w/v), brine (10 ml x 1) and then dried (Na2SO4). On evaporation of the solvent the product, which contained a mixture of isomers, was obtained as a pale yellow oil (4.42 g, 65.2%). The mixture of isomers was separated by preparative t.l.c. (1 mm plates, 7 times developed in CHCl3/petrol. ether (5:1)). The less polar isomer (Rf= 0.47) was obtained as a colorless oil, and more polar isomer (Rf = 0.35) as fine needles from acetone/petroleum ether. Less polar isomer: Cis-5-hydroxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3.2- dioxaphosphorinane. Analysis (C7H15O4PS), Found(%); C, 37.46 H, 6.87 Calculated(%); C, 37.17 H, 6.64 I MS, m/z (%): 226 (M+, 100), 195 (54), 167 (45), 165 (28), 149 (43), 143 (58), 115 (66). 31p NMR (CDCl3): 8 64.4 149 1H NMR (CD3OD): o4.54 (d,d. J = 10.91 Hz, 10.91 Hz, 2Hax. CfuOP), 4.37 (d,q. J = 10.31 Hz, 7.07 Hz, CH3C.fuOP), 4.22 (d,d. J = 15.96 Hz, 11.07 Hz, 2Heq, CfuOP), 3.65 (s, 2H, CfuOH), 1.53 (t, J = 7 .07 Hz, 3H, CfuCH2OP), 1.26 (s, 3H, CH3). 13C NMR (CDCl3): o 73.1 (d, J = 6.5 Hz, CH2OP), 65.0 (d, 4.7 Hz, CH3~H2OP), 64.7 (s CH2OH), 37.5 (d, J = 5.3 Hz, C~). 16.8 (s, CH3), 15.9 (d, J = 6. lHz, CH3CH2OP). IR (neat): 3450 (s.br), 2990 (s), 2950 (m), 2900 (m), 1466 (m), 1398 (m), 1242 (w), 1203 (m), 1165 (m), 1103 (m), 1060 (s), 788 (m), 1033 (s), 1000 (s), 975 (s), 920 (m), 850 (s), 660 (s). Spectrum simulation for the Less polar isomer: Cis-5-hydroxymethyl-5-methyl-2- thiono-r-2-ethoxy-1,3,2-dioxaphosphorinane (116). The lH spectrum for the ring protons was simulated with the PANIC VERSION 820601.1 progrom on ASPECT 3000 computer using the chemical shifts and coupling constants reported above. When the simulated spectrum agreed with the actual spectrum the interpretation of chemical shift and coupling data was considered to be corret. More polar isomer: Trans-5-hydroxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3.2- dioxaphosphorinane m.p.: 79-81 °C H, 6.85 Calculated(%): C, 37.17 H, 6.64 150 MS, m/z (%): 226 (M+, 10), 143 (100), 115 (79). 3Ip NMR (CH3OH): 6 62.6 1H NMR (10%benzene/CDC13): 6 4.38 (d,q. J =10.27 Hz, J = 3.19 Hz, 2H CH3CfuOP), 4.26 (m, 4H, ClliOP), 3.88 (s, 2H, ClliOH), 1.54 (t, J = 7.05 Hz, 3H, Cli3CH2OP), 0.97 (s, 3H, Cfu). 13C NMR (CDCI3): 6 72.8 (d, J = 8.1 Hz, CH2OP), 64.5 (d, J = 4.2 Hz, CH3CH2OP), 63.5 (s, CH2OH), 36.8 (d, J = 5.7 Hz, Cf), 16.0 (s, CH3), 15.8 (d, J = 6.5 Hz, CH3CH2OP), IR (KBr): 3264 (s.br), 2990 (m), 2960 (m), 2903 (m), 1478 (m), 1469 (m), 1206 (m), 1153 (m), 1061 (s), 1041 (s), 999 (s), 974 (s), 945 (s), 920 (s), 844 (s), 828 (s), 812 (s), 795 (s), 661 (s), 650 (s). Preparation of 5-hydroxymethyl-5-methyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinane (117) 5-Hydroxymethy1-5-methy 1-2-oxo-2-ethoxy- l ,3,2-dioxaphosphorinane was prepared from the condensation of 1,1,1-tris(hydroxymethyl)ethane (2.4 g, 0.02 mol), pyridine (3.22 ml, 0.04 mol) with O-ethyl phosphorodichloridate (3.26 g, 0.02 mol) using the general procedure described for the thiono analog. The product, which was obtained as a pale yellow oil, consisted of a mixture of isomers (3.60 g, 85. 7%) which were separated by preparative t.l.c. (1 mm plates, 7 times developed in CHCl3'EtOAc (4: 1)). Both the less polar (Rf = 0.53) and more polar (Rf = 0.42) isomers were crystallized from acetone/petroleum ether as colorless plates. 151 Less polar isomer: Cis-5-hydroxymethyl-5-methyl-2-oxo-r-2-ethoxy-1.3,2- dioxaphosphorinane. m.p.: 87-89 °C Analysis (C7H15O5P), Found(%): C, 40.10 H, 6.94 Calculated(%): C, 40.00 H, 7.14 MS,m/z (%): 211 (M+l, 6), 183 (16), 180 (22), 179 (24), 152 (24), 151 (24), 127 (52), 99 (100) 3Ip NMR (CDCI3): 6 -6.1 1H NMR (CDCl3): 6 4.38 (d,d, J = 5.88 Hz, 11.10 Hz, 2Hax· CfuOP), 4.15 (d,q, J = 8.13 Hz, 7.17 Hz, 2H, CH3CfuOP), 4.01 (d,d, J = 11.25 Hz, 18.22 Hz, 2Heq·, CfuOP), 3.46 (s, CfuOH), 1.36 (t, J = 7.08 Hz, 3H, Cli3CH2OP), 1.14 (s, CH3) 13C NMR (CDCI3): 6 74.7 (d, J = 6.1 Hz, CH2OP), 64.5 (s, CH2OH), 63.97 (d, J = 5.6 Hz, CH3CH2OP), 37.4 (d, J == 5.2 Hz, C4), 17.3 (s, CH3), 16.1 (d, J = 6.2 Hz, CH3CH2OP) IR (Nujol): 3384 (s. br), 1486 (m), 1393 (m), 1340 (m), 1274 (s, br), 1188 (m), 1074 (s), 1030 (s, br), 976 (s), 912 (s), 875 (s), 834 (s), 642 (s), 540 (m), 488 (m) More polar isomer: Trans-5-hydroxymethyl-5-methyl-2-oxo-r-2-ethoxy-1,3,2- dioxaphosphorinane. m.p.: 49-52 °C Analysis(C7H 1sOsP), Found (% ): C, 40.16 H,7.42 152 Calculated(%): C, 40.00 H, 7.14 MS,m/z (%): 211 (M+l, 4), 180 (32), 152 (25), 127 (32), 99 (100) 3 lP NMR (CDCl3): 6 -6.6 1H NMR (CDCl3): 6 4.23-4.03 (m, 4H, CfuOP), 3.75 (s. 2H, CH2OH), 1.37 (t, J = 7.07 Hz, CfuCH2OP), 0.85 (s, 3H, Cfu) 13C NMR (CDCl3): 6 73.6 (d, J = 6.8 Hz, CH2OP), 63.8 (d, J = 5.4 Hz, CH3.QH2OP), 62.8 (s, CH2OH), 37.3 (d, J = 5.2 Hz, C4), 16.2 (d, J = 6.2 Hz, ,CH3CH2OP), 15.5 (s, CH3) IR (Nujol): 3416 (s, br), 1288 ( s, br), 1268 ( s, br), 1244 ( s, br), 1205 (s), 1159 (s), 1076 (s, br), 1061 (s, br), 1033 (s, br), 1003 (s, br), 977 (s, br), 937 (s), 915 (s), 860 (s), 831 (s), 636 (s), 600 (m), 549 (s), 521 (s), 479 (s) Preparation of S-hydroxymethyl-5-methyl-2-oxo-2-methoxy-1,3,2- dioxaphosphorinane (118) 117 A solution of 1,1,1-tris(hydroxymethyl)ethane (2.40 g, 0.02 mol) and pyridine (3.5 ml, 0.02 mol) in CH3CN (100 ml) was added dropwise to a stirred, cooled (ice bath) solution of O-methylphosphorodichloridate (3.3 g, 0.02 mol) in CH3CN (10 ml). . On completion of the addition, the reaction mixture was stirred (2 hr) at room temperature, refluxed (5 min.) and then the solvent evaporated. The residue was dissolved in CHCl3, washed with dilute HCl solution (10 ml x 1, 5% w/v), NaHCO3 (10 ml x 1, 5% w/v), brine (5 ml x 1) and then dried (Na2SO4). Evaporation of the solvent gave the product isomers as a pale yellow oil (1.3 g, 33%). A further quantity of product was recovered from the aqueous layer by back-extraction with EtOAc (0.4 g, 10 %). 153 MS, m/z (%): 197 (M+l, 100), 225 (M+29, 8), 237 (M+41, 4) 3lp NMR (CDCl3): 6 -5.0, -5.7 Preparation of S-methyl-S-hydroxymethyl-2-oxo-2-hydroxy- 1,3,2-dioxaphosphorinanes (153).9 5 Aqueous NaOH (8 ml, 2.0 M) was added to a solution of BPO (204.3 mg) in dioxane (15 ml). After stirring for 3 hr at room temperature the solution was adjusted to pH 6 with dilute HCl (10 %) and the solvent evaporated. The residue was taken up in absolute ethanol, filtered, and the filtrate was evaporated and washed with acetone to give the title compound as a white solid (208.1 mg, 91.8 %). 3lp N.M.R. (CH3OH): 6-1.5 Preparation of 1-methyl-4-phospha-3,S,8-trioxabicyclo [2,2,2] octane-4-sulfide (BPS).214 A solution of 1,1,1-tris(hydroxymethyl)ethane (12.0 g) and pyridine (24.2 ml) in CH3CN (100 ml) was added dropwise to a stirred cooled (ice bath) solution of thiophosphoryl chloride (18.0 g) in CH3CN (50 ml) and the reaction mixture was stirred I overnight. After evaporation of the solvent the residue taken up in CHCl3, washed with dilute HCl (5 %, 20 ml x 1), water (30 ml x 1), and brine(20 ml x 1), then dried (Na2SO4) and the solvent evaporated to yield the product as a white solid (5.26 g, 29.3 %). m.p.: 226-227 °C (lit. 215: 224-225 °C) 3lp NMR (CHCl3): 6 57.3 154 13C NMR (CDCl3): 6 78.6 (d, J = 7.7 Hz, CH20P), 33.5 (d, J = 40.2 Hz, C4), 13.7 (s, CH3) Preparation of 1-methyl-4-phospha-3,5,8-trioxabicyclo [2,2,2] octane-4-oxide (BP0).214 The title compound was prepared as described for BPS by using 1,1,1-tris (hyclroxymethyl)ethane (1.2 g), pyridine (2.4 ml) and phosphoryl chloride (1.53 g). After the work-up procedure product was obtained as a white solid (792.9 mg, 48.4 %). m.p.: 249-250 °C (lit. 215: 249-250 °C) 3lp NMR (CHCl3): 6-7.3 13C NMR (CDCl3): 6 78.8 (d, J = 5.2 Hz, CH20P), 34.2 (d, J = 38.9 Hz, C4), 12.8 (s, CH3) Preparation of 5-acetoxymethyl-5-methyl-2-thiono-2-ethoxy- 1,3,2-dioxaphosphorinane (119). Acetic anhydride (2.0 g) was added to a stirred, cooled (ice bath) solution of a mixture of the desired isomer of 5-hydroxymethyl-5-methyl-2-thiono-2-ethoxy- 1,3,2-dioxaphosphorinane (2.0 g) in pyridine (20 ml). After stirring for 24 hr at room temperature the reaction mixture was poured into an ice/water mixture (100 ml) with vigorous stirring. The aqueous solution was extracted with CHCl3 (30 ml x 3), washed with saturated NaHC03 (10 ml x 2), brine (10 ml x 1), then dried (Na2S04) and the solvent evaporated to yield the product isomers as a brownish oil (1.52 g, 64.1 %) The mixture of the isomers was separated by column chromatography with petroleum ether/ether as eluents. Both isomers were obtained as white crystalline solids. 155 Less polar isomer: Cis-5-acetoxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3,2- dioxaphosphorinane. m.p.: 89-92 °C Analysis (C9H17O5PS), Found(%): C, 40.25 H, 6.56 Calculated(%): C, 40.30 H, 6.34 MS,m/z (%): 268 (M+, 38), 195 (26), 185 (28), 149 (36), 143 (100) 3lp NMR (CH2Ci2): B 63.7 lH NMR (CDCl3): B 4.27 (d,q. J = 10.53 Hz, 7.09 Hz, 2H, CH3CfuOP), 4.19 (d, J = 13.8 Hz, 4H, CfuOP), 4.08 (s, 2H, CfuO), 2.10 (s, 3H, C(O)Cfu), 1.37 (t, J = 7.16 Hz, 3H, Cli3CH2OP), 1.08 (s, 3H, Cfu) 13C NMR (CDCl3): B 170.5 (s, C(O)CH3), 72.7 (d, J = 6.2 Hz, CH2OP), 65.5 (s, CH2O), 65.2 (d, J = 4.8Hz, CH3CH2OP), 36.1 (d, J = 5.3 Hz, C4), 20.7 (s, C(O)CH3), 16.8 (s, CH3), 15.9 (d, J = 6.4 Hz, .CH3CH2OP). IR (KBr): 3452 (m), 2988 (m), 2903 (m), 1741 (s), 1488 (m), 1389 (m), 1267 (s), 1245 (s), 1223 (s), 1058 (s), 1037 (s), 1006 (s), 994 (s), 972 (s), 841 (s), 812 (s), 675 (m), 660 (s). More polar isomer: Trans-5-acetoxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3,2- dioxa,phosphorinane. m.p.: 102-105 °C Analysis (C9H17O5PS), Found(%): C, 40.24 H, 6.56 Calculated (% ): C, 40.30 H,6.34 156 MS rn/z (%): 268 (M+, 17), 185 (23), 143 (100), 115 (43). 3lp NMR (CDCl3): B 61.1 lH NMR (CDCl3): B 4.24 (s, 2H, CfuO), 4.22 (q, J = 3.29 Hz, 2H, CH3CfuOP), 4.16 (m, 4H, CfuOP), 2.10 (s, 3H, C(O)Cfu), 1.38 (t, J = 7 .08 Hz, 3H, CfuCH2OP), 0.90 (s, 3H, Cfu) 13C NMR (CDCl3): B 170.6 (s, {:(O)CH3), 76.3 (d, J = 8.6 Hz, CH2OP), 65.0 (s, CH2O), 64.6 (d,J = 4.5 Hz, CH3CH2OP), 35.4 (d, J = 5.72Hz, C4), 20.7 (s, CH3), 16.4 (s, C(O).CH3), 15.9 (d, J = 6.8 Hz, CH3CH2OP). IR (KBr): 3442 (m, br), 2976 (m), 2940 (m), 2900 (m), 1745 (s), 1476 (s), 1471 (s), 1405 (m), 1388 (s), 1374 (s), 1248 (s), 1169 (s), 1057 (s), 1050 (s),1027 (s), 1000 (s), 975 (s), 923 (s), 840 (s), 829 (s), 811 (s), 804 (s), I 660(s), 451 (s). Preparation of 5-benzoyloxymethyl-5-methyl-2-thiono-2- ethoxy-1,3,2-dioxaphosphorinane (120). Benzoyl chloride (1.1 ml) was added dropwise to a cooled (ice bath), stirred solution of a mixture of isomers of 5-hydroxymethyl-5-methyl-2-thiono-2-ethoxy- 1,3,2- The mixture was separated ( 1 mm plates, 2 developments in petroleum ether/ether (1:1)) and the less polar isomer (Rf= 0.74) was obtained as a colorless oil, which slowly solidified as an amorphous solid ; the more polar isomer (Rf= 0.59) was obtained as colorless crystals from acetone/ petroleum ether. 157 Less polar isomer: Cis-5-benzoyloxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3 ,2- dioxaphosphorinane. m.p.: 42-43 °C Analysis (C14H19O5PS), Found(%): C, 51.18 H, 6.03 Calculated(%): C, 50.91 H, 5.76 MS, m/z (%): 330 (M+,8), 105 (100) 3lp NMR (CDCl3): 6 64.7 lH NMR (CDCl3): 6 8.02, 7.60, 7.47 (m, 5H, aromatic), 4.33 (s, 2H, CH2O), 4.37-4.22 (m, 6H, CfuOP, CH3CfuOP), 1.37 (t, J = 7.09 Hz, 3H, CfuCH2OP), 1.19 (s, 3H, CH3). 13C NMR (CDCl3): 6 166.0 (s, C(O)Ph), 1335, 129.6, 129.5 129.6 (m, Ph), 72.8 (d, J = 6.2 Hz, CH2OP), 65.9 (s, CH2O), 65.3 (d, J = 5.0Hz, I CH3CH2OP), 36.5 (d, J = 5.3 Hz, C4), 17 .1 (s,CH3), 16.0 (d, J = 6.8 Hz, CH3CH2OP). IR, (KBr): 2983 (s), 2942 (m), 2896 (m), 1725 (s), 1468 (s), 1454 (s), 1401 (m), 1377 (m), 1316 (s), 1274 (s), 1179 (s), 1165 (s), 1116 (s), 1058 (s), 1030 (s), 1001 (s), 968 (s), 843 (s), 826 (s), 714 (s) More polar isomer: Trans-5-benzoyloxymethyl-5-methyl-2-thiono-r-2-ethoxy- 1,3 .2-dioxaphosphorinane. m.p.: 134-135 °C Analysis (C14H19O5PS), Found(%): C, 50.82 H, 5.82 Calculated(%): C, 50.91 H, 5.76 158 MS, rn/z (%): 330 (M+, 4), 105 (100) 3lp NMR (CDCl3): B 59.5 lH NMR (CDCl3): B 8.03, 7.58, 7.45 (m, 5H, aromatic), 4.34-4.19 (m, 6H, CH.zOP, CH3CfuOP), 4.51 (s, 2H, CH.2O), 1.40 (t, J = 7.09 Hz, 3H, CH3CH2OP), 0.99 (s, 3H, Cfu). 13C NMR (CDCl3): 6166.1 (s, C(O)Ph), 133.3, 129.6, 128.53 (m, Ph), 72.8 (d, J = 8.3 Hz, CH2OP), 65.3 (s, CH2O), 64.7 (d, J = 4.5 Hz, CH3,CH2OP), 36.0 (d, J = 5.5 Hz, C4), 16.7 (s, CH3), 15.90 (d, J = 6.3 Hz, CH3CH2OP). IR (KBr): 2992 (s), 2970 (s), 2900 (s), 1728 (s), 1478 (s), 1458 (s), 1325(s), 1286 (s), 1123 (s), 1056 (s), 1037 (s), 999 (s), 972 (s), 920 (s), 848 (s), 828 (s), 805 (s), 705 (s), 654 (s). Preparation of 5-tosyloxymethyl-5-methyl-2-thiono-2-ethoxy- 1,3,2-dioxaphosphorinane (121). p-Toluenesulfonyl chloride (440.0 mg) was added portionwise to a cooled (ice bath), stirred solution of a mixture of isomers of 5-hydroxymethyl-5-methyl-2- thiono-2-ethoxy-1,3,2-dioxaphosphorinane (519.0 mg) in pyridine (1.5 ml) and the solution was stirred (30 min, ice bath), stored at +5 °C for 16 hr, and then at room temperature for 2 hr. The excess reagent was destroyed by addition of ice then the mixture was extracted with CHCl3 (100 ml), dried (Na2SO4) and the solvent evaporated to give a yellow oil (640 mg, 76.1 %). The product isomers were separated by preparative t.l.c. (1 mm plates, 2 developments in petroleum ether/ether (1:1)). The less polar isomer (Rf= 0.67) was obtained as an amorphous solid from CHCl3 and the more polar isomer (Rf= 0.55) as colorless crystals from acetone/petroleum ether. 159 Less polar isomer: Cis-5-tosyloxymethyl-5-methyl-2-thiono-r-2-ethoxy-1,3,2- dioxaphosphorinane. m.p.: 43-45 °C Calculated(%): C, 44.21 H, 5.53 MS, rn/z (%): 380 (M+,50), 225 (50), 155 (58), 91 (100) 3lp NMR (CDCl3), 6 65.0 1H NMR (CDCl3): 6 7.79, 7.38 (m, 4H, aromatic), 4.23-4.04 (m, 6H, CH2OP, CH3Cli2OP), 4.00(s, 2H, C!:!2O), 2.46 (s, Ph-CH3), 1.33 (t, J = 7.06 Hz, 3H, CfuCH2OP), 1.00 (s, 3H, CH3). 13C NMR (CDCl3): 6 145.3, 132.2, 130.02, 127.9 (m, Ph), 71.8 (d, J = 6.0 Hz, CH2OP), 70.6 (s, CH2O), 65.3 (d, J = 5.0 Hz, CH3CH2OP), 36.5 (d, J = 4.9 Hz, C4), 21.6 (s, Ph-CH3), 16.3 (s, CH3), 15.9 (d, J = 6.7 Hz, CH3CH2OP) IR (KBr): 2985 (s), 2943 (m), 2901 (m), 1598 (s), 1470 (s), 1455 (s), 1398 (m), 1376 (s), 1296 (m), 1215 (m), 1191 (s), 1178 (s), 1097 (s), 1058 (s), 1032 (s), 1006 (s), 983 (s), 968 (s), 922 (m), 848 (s), 836 (s), 813 (s), 792 (s), 777 (s), 673 (s), 560 (s), 555 (s). More polar isomer: Trans-5-tosyloxymethyl-5 methyl 2-thiono-r-2-ethoxy-1,3,2- dioxaphosphorinane. m.p. 87-89 °C Analysis (C14H21O6PS2), Found(%): C, 44.34 H, 5.71 160 Calculated(%): C, 44.21 H, 5.53 MS, m/z (%): 225 (M-155, 100), 197 (78) 3lp NMR (CDCl3): 6 60.7 lH NMR (CDCl3, 500 MHz): 6 7.78, 7.37 (m, 4H, aromatic), 4.15 (d,q. J = 10.35 Hz, 7.llHz, 2H, CH3Cli2OP), 4.12 (s, 2H, C!:!2O), 4.10 (d,d. J = I 11.55 Hz, 3.63 Hz, 2Hax, CfuOP), 4.00 (d,d, J = 11.50 Hz, 21.72 Hz, 2Heq, CfuOP), 2.45 (s, 3H, Ph-Cfu), 1.34 (t, J = 7.07 Hz, 3H, CfuCH2OP), 0.91 (s, 3H, CH3). 13C NMR (CDCl3): 6 145.4, 131.9, 130.20 128.0 (m, Ph), 71.5 (d, J = 8.4 Hz, CH2OP), 70.3 (s, CH2O), 64.7 (d, J = 3.8 Hz, CH3CH2OP), 35.9 (d, J = 5.4 Hz, C4), 21.7 (s, Ph-CH3), 16.0 (s, CH3), 15.8 (d, J = 6.5 Hz, ,CH3CH2OP). IR (KBr): 3440 (m,br), 2987 (m), 1603 (m), 1473 (m), 1362 (s), 1194 (s), 1181 (s), 1101 (m), 1062 (s), 1038 (s), 1006 (s),976 (s), 923 (m), 848 (s), 818 (s), 794 (s), 672(s), 573 (s), 559 (s). Preparation of 5-(3,5-Dinitro)benzoyloxymethyl-5-methyl-2- thiono-2-ethoxy-1,3,2-dioxaphosphorinane (122). A solution of 3,5-dinitrobenzoyl chloride (1.7 g) in CH3CN (2 ml) was added dropwise to a cooled (ice bath), stirred solution of a mixture of the isomers of 5- hydroxymethyl-5-methyl-2-thiono-2-ethoxy-1,3,2-dioxaphosphorinane ( 1.46 g) and pyridine (1.0 ml) in CH3CN (10 ml). The reation mixture was then stirred overnight at room temperature. Work up, as described for the benzoyl derivative, gave the product as a brown oil (2.26 g, 83.1 %) which was further purified by flash column chromatography (petroleum ether/ether). The mixture of isomers was separated (1 mm 161 t.l.c. plates) by developing once in petroleum ether/ether (2:3) then twice in petroleum ether/ether (1:1). The less polar (Rf= 0.71) and more polar (Rf= 0.53) isomers were obtained as off-white crystals and colorless needles, respectively, from acetone/petroleum ether in each case. Less polar isomer: Cis-5-(3,5-Dinitro)benzoyloxymethyl-5-methyl-2-thiono-r-2- ethoxy-l ,3,2-dioxaphosphorinane. m.p.: 167-169 °C Analysis (C14H17O9N2PS), Found(%): C, 40.19 H, 4.01 N,6.60 Calculated (% ): C, 40.00 H, 4.08 N, 6.67 MS, m/z (%): 420 (M+,10), 195 (100) 3lp NMR (CDCl3): 6 65.5 1H NMR (CDCl3): 6 9.27, 9.13 (m, 3H, aromatic), 4.53 (s, 2H, CfuO), 4.41- 4.22 (m 6H, CfuOP, CH3CfuOP), 1.38 (t, J = 7.07 Hz, 3H, C!!3CH2OP), 1.16 (s, 3H, Cfu) 13C NMR (CDCl3): 6 162.2 (s, C(O)Ph(NO2)2), 148.8, 133.1, 129.3, 122.8 (m, Ph(NO2)2), 72.3 (d, J = 5.5 Hz, CH2OP), 67.9 (s, CH2O), 65.7 (d, J = 4.8 Hz, CH3CH2OP), 36.6 (d, J = 4.9Hz, C4), 16.8 (s, CH3), 16.0 (d J = 6.8 Hz, CH3CH2OP) IR (KBr): 3452 (m, br), 3101 (m), 2978 (m), 1741 (s), 1635 (m), 1552 (s), 1466 (s), 1350 (s), 1276 (s), 1173 (s), 1157 (s), 1079 (m), 1062 (s), 1036 (s), 992 (s), 924 (m), 840 (s), 824 (s), 724 (s). 162 More polar isomer: Trans-5::{3.5-Dinitro)benzoyloxymethyl-5-methyl-2-thiono-r-2- ethoxy-1.3.2-dioxaphosphorinane. m.p.: 153--154 °C Analysis (C14H17O9N2PS), Found(%): C, 40.06 H, 4.25 N,6.75 Calculated(%): C, 40.00 H, 4.08 N, 6.67 MS,rn/z (%): 420 (M+, 20), 195 (100) 3lp NMR (CDCl3): 6 61.0 1H NMR (CDCl3): 6 9.25, 9.14 (m, 3H, aromatic), 4.66 (s, 2H, CH2O), 4.31- 4.18 (m, 6H, CfuOP, CH3CfuOP),l.40 (t, 7.0Hz, 3H, C!i3CH2OP), 1.02 (s, 3H, Cfu) 13C NMR (CDCl3): 6 162.3 (s, C(O)Ph(NO2h), 148.8, 133.3,129.48, 122.8 (m, Ph(NO2)2), 72.7 (d, J = 8.6 Hz, CH2OP), 67.7 (s, CH2O), 64.9 (d, J = 4.5 Hz, .C,H3CH2OP), 35.8 (d, J = 5.7 Hz, C4), 16.6 (s, CH3), 15.9 (d, J = 6.3 Hz, CH3CH2OP). IR (KBr): 3436 (m, br), 3095 (m), 2987 (m), 2897 (m), 1736 (s), 1630 (s), 1548 (s), 1466 (s), 1402 (m), 1348 (s), 1330 (m), 1283 (s), 1165 (s), 1059 (s), 1040 (s), 999 (s), 969 (s), 919 (s), 847 (s), 813 (s), 722 (s). Preparation of 5-acetoxymethyl-5-methyl-2-oxo-2-ethoxy- 1,3,2-dioxaphosphorinane (144). , Acetic anhydride (2.0 g) was added to a stirred, cooled (ice bath) solution of a mixture of the isomers of 5-hydroxymethyl-5-methyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinane (1.4 g) and pyridine (1.0 g) in CH3CN (10 ml). After stirring overnight at room temperature the reaction mixture was poured into ice water, then 163 worked up in the normal manner. Evaporation of the solvent gave the acetate as a colorless oil (1.28 g, 7 4.4 %). Analysis (C9H17O0'), Found(%): C, 42.54 H, 7.00 I Calculated(%): C, 42.86 H, 6.75 MS, m/z (%): 252 (M+,3), 225 (8), 183 (23), 180 (60), 165 (18), 152 (26), 151 (23), 127 (100), 99 (84). 3lp NMR (CDCl3): B-5.8, -7.3 lH NMR (CDCI3, 500 MHz): B4.24 (s, 2H, CfuO), 4.14 (m, 8H, CfuOP), 3.97 (s, 2H, CfuO), 2.08 (s, 6H, C(O)Cfu), 1.37 (t, 7.07Hz, 3H, CH3CH2OP), 1.36 (t, J = 6.98 Hz, 3H, CH3CH2OP), 1.16 (s, 3H, Cfu), 0.85 (s, 3H, CH3) IR (Neat): 3556 (m), 3489 (m), 2985 (s), 2944 (s), 2904 (s), 1745 (s), 1472 (s), 1446 (m), 1385 (s), 1369 (s), 1300 (s), 1240 (s), 1168 (s), 1072 (s), 1039 (s), 1011 (s), 976 (s), 952 (s), 921 (s), 853 (s), 830 (s), 634 (m), 540 (m), 517 (s), 489 (s), 473 (s). Preparation of 5,5-di methyl-2-thiono-2-ethoxy-1,3,2- dioxaphosphorinane (133). This compound was prepared using the procedure described for 5- hydroxymethyl-5-methyl-2-thio-2-ethoxy-1,3,2-dioxaphosphorinane by condensing 2,2-dimethyl-1,3-propanediol (10.5 g) and 0-ethyl phosphorodichloridothionate (18.5 g) in the presence of pyridine (16.1 ml). After workup involving a simple wash with each solution, a colorless liquid was obtained which solidified on cooling. Further purification by flash column chromatography (ether/petroleum ether) gave the product as a white solid (8.1 g, 38.6 % ) 164 m.p.: 58-60 °C (lit.216: 62-63 °C) 3lp NMR (CDCl3): 6 62.2 13C NMR (CDCl3); o 76.9 (d J = 7.0 Hz, CH2OP), 64.3 (d, J = 3.4 Hz, CH3CH2OP), 32.2 (d, J = 5.71Hz, C4), 21.6 (s, CH3), 21.0 (s, CH3), 15.8 (d, J = 6.9 Hz, CH3CH2OP). Preparation of 5,5-dimethyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinane (132). A solution of 2,2-dimethyl-1,3-propanediol (10.5 g) and pyridine (16.0 g) in CH3CN (60 ml) was added to a cooled, stirred solution of O-ethyl phosphorodichloridate (16.4 g) in CH3CN (10 ml) and the reaction mixture was stirred for 2 hr at room temperature. The solvent was evaporated and the residue taken up in CHCl3. After washing with dilute HCl solution (10 ml x 1, 5 % w/v) and drying (N a2SO4), the crude product was purified by column chromatography (petroleum ether/ether) to yield the title compound as~ colorless liquid (13.9 g, 71.6 %). 3lp NMR (CH2Cl2): 6-6.9 (lit.217, 8.5) 13C NMR (CDCl3): 6 77.6 (d, J = 6.6 Hz, CH2OP), 63.5 (d, J = 5.3 Hz, CH3CH2OP), 32.1 (d, J = 5.2 Hz, C4), 21.6 (s, CH3), 20.4 (s, CH3), 16.1 (d, J = 5.8 Hz, CH3CH2OP). 165 Decoupling experiment and simulation of spectrum for 5,5-dimethyl-2-oxo-2-ethoxy- 1,3,2;:dioxaphosphorinane (132). Decoupling of methyl resonance at B 1.21 changed the shape of the axial proton resonances from a broad doublet of doublets (B 4.04, J = 3.9 Hz, 10.7 Hz) to a doublet of doublet of triplets (B 4.04, J = 3.9 Hz, 11.0 Hz, 1.3 Hz). The lH spectrum of the ring protons of 5,5--dimethyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinane was simulated using the same procxedure as for Cis-5- hydroxymethyl-5-methy 1-2-thio-r-2-ethoxy-l ,3 ,2-dioxaphosphorinane () ( axial protons: as reported above; equatorial protons: B 3.88, J = 19.9 Hz, 11.0 Hz, 1.3 Hz). The same spectrum was obtained confirming that the interpretation was correct and that there is a cross ring coupling between the ring protons. Preparation of S,5-dimethyl-2-oxo-2-methoxy-1,3,2- dioxaphosphorinane (137). A CH3OH (5 ml) solution of 5,5...:.dimethyl-2-oxo-chloro-1,3,2- dioxaphosphrinane (205.0 mg) was stirred (2 hr) at room temperature then refluxed for 3hr. Evaporation of the solvent gave the product as a white solid (198.0 mg, 99.0 %). m.p.: 97-98 °C (lit. 216: 94 °C) 3lp NMR (CDCI3): B-5.8 (lit.217: 7.5) 13C NMR (CDCl3): 6 77.8 (d, J = 6.6 Hz, CH2OP), 53.7 (d, J = 5.5 Hz, CH3OP), 32.1 (d, J = 5.6 Hz, C4), 21.6 (s, CH3), 20.3 (s, CH3) lli6 Preparation of S,S-dimethyl-2-oxo-2-chloro-1,3,2- dioxaphosphorinane (134). A solution of 2,2-dimethyl-1,3-propanediol (8.32 g) and pyridine (13.0 ml) in CH3CN (10 ml) was addded to a cooled (ice bath), stirred solution of phosphoryl chloride (1.23 g) in CH3CN (20 ml) and then the stirring continued for a further 2hr at 70 °C. The solvent was evaporated, the residue taken up in CH2Cl2 (100 ml) and then this was worked up in the normal manner to give a white solid. Further purification was by flash column chromatography (petroleum ether/ether) whereby the product was obtained as a white solid. (7.31 g, 52.3 %). m.p.: 103-105 °C (lit. 218,219: 104.5-106 °C) 3Ip NMR (CH2Cli): B-2.0 13C NMR (CDCl3): B 79.1 (d, J = 7.3 Hz, CH2OP), 32.4 (d, J = 6.5 Hz, C4), 21.9 (s, CH3), 19.9 (s, CH3). Preparation of 5,5-dimethyl-2-oxo-2-hydroxy-1,3,2- dioxaphosphorinane (135). An aqueous acetone (8 ml, 50 % v/v) solution of 5,5-dimethyl-2-oxo-2- chloro-1,3,2-dioxaphosphorinane (1.5 g) was refluxed for 2 hr and then stirred overnight. Evaporation of the solvent gave a white solid (1.32 g, 97.8 %). m.p.: 174-176 °C (lit. 218:174-176 °C, lit. 220: 174-175 °C) 3Ip NMR (CH2Cl2): B-3.9 (lit.220:-4.0) 13C NMR (DMSO-d(,): B 76.7 (d, J = 6.4 Hz, CH2OP), 32.0 (d, J = 5.0 Hz, C4), 20.7 (s, CH3) 167 Preparation of 5,5-dimethyl-2-oxo-2-hydro-1,3,2- dioxaphosphorinane (138).221 Phosphorus trichloride (41.3 g) was added to a stirred solution of 2,2-dimethyl- 1,3-propanediol (13.2 g) and ethanol (13.8 g) at 20-25 °C in an N2 atmosphere, then the solution stirred for 1 hr. Removal of the solvent gave a colorless oil which solidified on cooling. Distillation under reduced pressure gave the product as a white solid. m.p.: 47-50 °C (lit.221: 48-50 °C) 3lp NMR (CH2Ch): 6 3.2 13C NMR (CDCl3): 6 76.0 (d, J = 5.7 Hz, CH3OP), 32.1 (d, J = 6.0 Hz, C4), 21.7 (s, CH3), 20.6 (s, CH3). Preparation of bis(S,5-dimethyl-2-oxo-1,3,2- dioxaphosphorinanyl)oxide (136).219 A solution consisting of 5,5-dirnethyl-2-oxo-2-chloro-1,3,2- dioxaphosphorinane (3.70 g), pyridine (1.6 ml), water (400 mg) and THF (15 ml) was refluxed overnight. The solvent was removed in vacuo, the residue taken up in CH2Cl2 (100 ml), and then the crude material worked up in the normal way to give the product as a white solid (2.98 g, 94.6 % ). m.p.: 191-193 °C (lit.219: 188-190 °C) 3Ip NMR (CH2Cl2): 6-20.6 (lit.222:21.5) 13C NMR (CDCl3): 6 79.0 (s, CH2OP), 32.1 (s, C4), 21.8 (s, CH3), 19.8 (s, CH3) 168 Preparation of S,S-dimethyl-2-oxo-2-thiophenyl-1,3,2- dioxaphosphorinane (139). Diphenyl disulfide (2.18 g) was added to a solution of 5,5-dimethyl-2-oxo-2- hydro-1,3,2-dioxaphosphorinane (1.50 g) in benzene (10 ml) and triethyl amine (1.4 ml). After stirring for 40 min. at 80 °C the solvent was evaporated to give the product as a colorless liquid,which was purified by column chromatography (petroleum ether/ether) (1.92 g, 74.4 %). m.p.: 121-122 °C Analysis (C11H15O3PS), Found(%): C, 51.10 H, 6.05 Calculated(%): C, 51.16 H, 5.81 I MS, m/z (%): 258 (M+, 52), 246 (38), 218 (40), 190 (39), 110 (87), 109 (100) 1H NMR (CDCl3): 6 7.64, 7.36 (m, 5H, aromatic), 4.20 (d,d. J = 4.1 Hz, 10.8 Hz, 2Hax, CfuOP), 3.93 (d,d,t. J = 11:22 Hz, 23.73 Hz, 1.5 Hz, 2Heq, CfuOP), 1.28 (s, 3H, CH3), 0.87 (s, 3H, Cfu) 13C NMR (CDCl3); 134.7 (d, J = 5.0 Hz), 129.5, 129.2, 124.8 (d, J = 6.0 Hz)(m, SPh), 78.2 (d, J = 7.3 Hz, CH2OP), 32.5 (d, J = 6.5Hz, C4), 22.0 (s, CH3), 20.4 (s, CH3) IR (KBr): 3432 (w), 2970 (w), 1478 (s), 1442 (w), 1273 (s), 1052 (s), 997 (s), 975 (s), 846 (s), 786 (s), 755 (s), 564 (s), 493 (s), 482 (m) 169 Preparation of triethyl phosphorothionate (112).223-225 Finely powdered sulfur (17.6 g) was added portionwise with stirring to triethyl phosphite (83.1 g) at 100 °C. After further stirring (30 min) at the same temperature the mixture was cooled, filtered and then distilled under reduced pressure to yield a colorless oil. (94.2 g, 95.1 %, b.p., 97-98 °C/21 mm Hg; lit.223: b.p., 117-118 °C/35 mm Hg). 3lp NMR (CH2Cl2): 6 68.1 13C NMR.(CDCI3); 6 63.9 (d, J = 5.5 Hz, CH3CH20P), 15.6 (d, J = 7.5 Hz, CH3CH20P) Preparation of trimethyl phosphorothionate (125). The title compound was preparCrd by the procedure described for triethyl phosphorothionate using sulfur (7 .0 g), trimethyl phosphite (26.3 g) but with refluxing for 2 hr. Trimethyl phosphorothionate was obtained as a colorless liquid. (31.4 g, 94.2 %, b.p., 78-79 °C/21 mm Hg) 3lp NMR (CH2Ci2): 6 73.4 13C NMR (CDCl3): 6 54.2 (d, J = 5.5 Hz, CH30P) Preparation of dimethyl ethyl phosphate (126). A solution of methanol (3.5 ml) and pyridine (6.7 ml) in ether (20 ml) was added dropwise to a cooled (ice bath), stirred solution of 0-ethyl phosphorodichloridate (6.7 g) in ether (20 ml). After stirring for 4 hr the reaction mixture was filtered then the 170 solvent evaporated to yield a liquid which was distilled under reduced pressure to give the product as a colorless liquid. (4.1 g, 65.2 %, b.p., 103-105 °C/23 mm Hg). MS, rn/z (%): 155 (M+l, 3), 127 (100), 109 (50), 95 (20) 3lp NMR (CDCl3): B 1.9 lH NMR (CDC13): B 4.00 (d,q. J = 7.14 Hz, 7.14 Hz, 2H, CH3CfuOP), 3.64 (d, I= 11.04 Hz, 6H, CfuOP), 1.22 (t, J =7 .12 Hz, 3H, CfuCH2OP) 13C NMR (CDCl3): B 63.7 (d, J = 5.6 Hz, CH3.C.H2OP), 53.9 (d, J = 6.0Hz, CH3OP), 15.8 (d, J = 6.5Hz, CH3CH2OP)) IR (Neat): 3550 (m), 3500 (m), 3000 (s), 2975 (s), 2870 (m), 1455 (s), 1400 (m), 1375 (m), 1250 (s,br), 1190 (s), 1020 (s, br), 980 (s), 850 (s), 750 (s). Preparation of diethyl methyl phosphate (69). A solution of methanol (2.50 g), pyridine (3.66 g) and benzene (50 ml) was added dropwise to a stirred solution of diethyl chlorophosphate (8.0 g) in benzene (20 ml) at 8-10 °C. After stirring overnight at room temperature the reaction mixture was filtered and then the solvent was evaporated to give the product as a pale yellow liquid. On distillation under reduced pressure the product was obtained as a colorless oil. (5.85 g, 75.1 %, b.p., 109-111 °C/23 mmHg). 31 P NMR (CH2Cl2): B 0.9 13C N.R (CDCl3): <> 63.7 (d, J = 5.5 Hz, CH3CH2OP), 53.9 (d, J = 5.6 Hz, CH3OP), 16.0 (d, J = 6.2 Hz, CH3CH2OP). 171 Preparation of diethyl bis-phosphoryl disulfide (130))35 Triethyl phosphorothionate (6.50 g) and sulfuryl chloride (2.35 g) in CH2Cl2 (30 ml) were stirred for 3 hr at room temperature. Evaporation of the solvent gave a yellow liquid which was taken up in CH2Cl2 (80 ml), washed with dilute aqueous K2CO3 solution (10 ml x 1,.5 %), brine (10 ml x 1), and then dried (Na2SO4). Removal of the solvent gave the product as a pale yellow liquid (5.3 g, 95.5 %). 31p NMR (CDCl3): o21.3 (lit.135: 19.9) 13C NMR (CDCl3): o 64.8 (d, J = 5.3 Hz, CH3,CH2OP), 15.8 (d, J = 7.1 Hz, ~H3CH2OP). Preparation of dimethyl bis-phosphoryl disulfide (131).I 35 The title compound was prepared from trimethyl phosphorothionate (6.0 g) in CH2Cl2 (30 ml) and sulfuryl chloride (1.8 ml) using the procedure described for the diethyl bis-phosphoryl disulfide. The product was obtained as a pale yellow liquid (5.21 g, 96.1 %). 3lp NMR (CDCl3): o24.2 (lit. 135: 23.5) 13C NMR (CDCl3): o54.51 (d. J = 5.49 Hz, CH3OP). Preparation of diethyl hydrogenphosphonate.(128) 226 A solution of phosphorus trichloride ( 41.4 g) in CCl4 (30 ml) was added dropwise to a stirred solution of ethanol (41.4 ml) in CC4 (45 ml) and the reaction mixture refluxed (30 min.). Air was passed through the reaction mixture after cooling, 172 then it was concentrated and distilled under reduced pressure to yield the product as a colorless liquid (36.3 g, 58.5 %, b.p., 80 °C/21 mm Hg, Iit.226: 73-74 °C/14 mm Hg). 3lp NMR (CH2Cl2): 6 7.8 (d, J = 691 Hz) 13C NMR (CDCl3): 6 61.4 (d, J = 5.3 Hz, CH3QH2OP), 15.9 (d, J = 6.1 Hz, ,CH3CH2OP) Preparation of O,O-diethyl S-phenyl phosphorothiolate (129). Diethyl hydrogenphosphonate (1.38 g) in benzene (10 ml) containing triethylamine (1.4 ml) was treated with a solution of diphenyl disulfide (2.18 g) in benzene (10 ml) and the mixture was stirred (2 hr) at 80 °C. The solvent was evaporated and the residue purified by column chromatography (petroleum ether/ether) to give the product as a colorless oil (1.89 g, 76.8 %). 3lp NMR (CHCl3): 6 23.6 13C NMR (CDCl3): 6 134.5, 134.4, 129.3, 128.90, 126.5, 126.5 (m, SPh), 64.0 (d, J = 5.8 Hz, CH3CH2OP), 15.9 (d, J = 6.9 Hz, CH3CH2OP). Preparation of diethyl phosphate (127). A solution of diethyl chlorophosphate (1.21 g) in aqueous acetone (5 ml, 50 % v/v) was refluxed for 1 hr, then stirred overnight. Evaporation of the solvent yielded a colorless oil ( 1.08 g, 100 % ). 3lp NMR. (CH2C!i): 6 0.7 13C NMR (CDCl3): 6 63.7 (d, J = 5.4 Hz, CH3CH2OP), 16.0 (s, ,CH3CH2OP). 173 Preparation of O,O-diethyl S-phenyl phosphorothiolate- 18Q (186). A mixture of triethyl phosphite (2.3 ml) and 18Q-water (250 mg, 97 atom %- 18Q) was stirred at room temperature for 3 hr. On evaporation of the solvent O,O diethyl hydrogenphosphonate-18O was obtained as a colorless oil (1.35 g, 96.4 %, 31p NMR (CH2Cl2): o 7 .6 (d, J = 692 Hz)) which was dissolved in benzene (10 ml) containing triethylamine (1.4 ml). Diphenyldisulfide (2.20 g) in benzene (10 ml) was then added at 80 °C, the solution stirred for 3 hr and then the solvent was evaporated to give a pale yellow-green liquid, whic~ was purified by column chromatography (petroleum ether/ether). O,0-diethyl S,-phenyl phosphorothiolate-18O was obtained as a colorless oil (1.7 g, 70.8 %). Analysis (C10H15O2PS 18O), Found(%): C, 48.33 H, 6.25 Cailculated (%): C, 48.38 H, 6.09 MS, m/z (%): 248 (M+,96), 220 (31), 192 (25), 111 (68), 110 (100), 109 (56) 3lp NMR (EtoAC): o22.5 1H NMR (CDCl3): o 7.25, 7.34 (m, 5H, aromatic), 4.19 (m, 4H, CH3CfuOP), 1.30 (d, t, J = 0.98 Hz, 7.07 Hz, 6H, CfuCH2OP). 13C NMR (CDCl3): 134.4, 134.4, 129.20, 128.9, 126.5, 126.4 (m, SPh), 63.9 (d, J = 6.0 Hz, CH3CH2OP), 15.9 (d, J = 7.0 Hz, CH3CH2OP). IR (Neat): 3560 (w, br), 3490 (w, br), 2985 (s), 2940 (m), 2910 (m), 1582 (m), 1477 (s), 1442 (s), 1395 (s), 1255 (m), 1225 (s), 1162 (s), 1100 (m), 1020 (s, br), 970 (s, br), 790 (s, br), 750 (s, br), 702 (s), 693 (s). 174 Preparation of O,O-diethyl phosphorothioic acid (10). Sulfur (1.60 g) was added portionwise to a stirred solution of Q,Q-diethyl hydrogenphosphonate (6.90 g), triethylamine (5.10 g) and benzene (10 ml). After stirring for 1 hr, the solvent was evaporated to give the product as a yellow oil (3lp NMR (CH2Cl2): o 57.8). This was dissolved in methanol (8 ml), treated with cone. HCl (5.5 ml), then the solvent removed and ether added to the residue. Following filtration, the organic layer was washed with dil. HCl (5 ml x 1, 5 %), dried (Na2S04), then the solvent removed to give a yellow liquid (6.18 g, 98.5 %), which was distilled under reduced pressure to give a pale yellow liquid (5.41 g, 63.6 % ; b.p., 101-103 °C I 0.05 mm Hg, lit.227:117-119 °C/ 6 mm Hg). 3lp NMR (CDCl3): o58.8 (Iit.228: 57.7) 13C NMR (CDCl3): o64.4 (d, J = 5.2 Hz, CH3CH20P), 15.8 (d, J = 8.0 Hz, CH3CH20P) Preparation of thiophosphoryl chloride.229 A mixture of aluminium chloride (1.0 g), phosphorus trichloride (50 g) and sulfur (12 g) was refluxed (30 min.) then the reaction mixture distilled to give the product as a colorless liquid (54.9 g, 90.4 %; b.p., 125 °Cn60 mm Hg, lit. 229125-126 0 cnro mm Hg) 3lp NMR (CH2Ch): o33.7 Preparation of O-ethyl phosphorodichloridate.230 Anhydrous ethanol (27 .0 g) was_ added dropwise to a solution of phosphorus oxychloride (90.0 g) in dry ether (105 ml) at 5 °C (ice bath) and the reaction mixture was 175 stirred for 2 hr at room temperature. The product was distilled under reduced pressure to give Q--ethyl phosphorodichloridate as a colorless liquid (78.4 g, 90.9 %; b.p., 68- 70 °C/15 mm Hg) 3lp NMR (CH2Cl2): 8 7.4 IH NMR (CDCl3): 8 4.38 (d,q. J = 11.01 Hz, 7.07 Hz, 2H, CH3CH2OP), 1.45 (d,t. J = 1.33 Hz, 7.08 Hz, 3H, CfuCH2OP) 13C NMR (CDCl3): 68.6 (d. J = 9.0 Hz, CH3CH2OP), 15.4 (d, J = 8.8 Hz, CH3CH2OP). Preparation of O-methyl phosphorodichloridate.230 The title compound was prepared from anhydrous methanol (10.6 g) and phosphorus oxychloride (50 g) using the same procedure described for the O-ethyl phosphorodichloridate. The product was obtained as a colorless liquid (33.2 g, 68.9 %; b.p., 52-54 °C/13 mm Hg). 3lp NMR (CH2Cl2): 8 8.4 IH NMR (CDCl3): 8 3.97 (d, J = 16.92 Hz, CH3OP). 13C NMR (CDCl3): 57.3 (d, J = 8.8 Hz, CH3OP). Preparation of O-methyl phosphorodichloridothioate. Thiophosphoryl chloride (50 g) and anhydrous methanol (36.2 g) were mixed dropwise at 19 °C (ice bath). On completion of the reaction, the mixture was washed with cold water (50 ml), dried (Na2SO4) and then fractionally distilled under reduced 176 pressure to give the product as a colorless liquid (30.5 g, 62.8 %; b.p., 59-62 °C/25 mm Hg, lit.231 b.p., 45 °C/l l mm Hg). 31p NMR (CH2Cl2): B 60.7 lH NMR (CDCl3): B 4.00 (d, J = 18.93 Hz, 3H, CfuOP) 13C NMR (CDCl3): 57.4 (d, J = 10.3 Hz, CH3OP) Preparation of diethyl chlorophosphate.232 A solution of phosphorus trichoride ( 46 g) in benzene 80 ml) was added dropwise to a solution of ethanol (46 g) in benzene at 8-10 °C and then the reaction mixture was stirred for 1 hr while maintaining the same temperature. A solution of sulfuryl chloride (45 g) in benzene (80 ml) was added dropwise, also at 8-10 °C, and the reaction mixture was stirred for a further 1 hr at room temperature. Evaporation of the solvent followed by distillation under reduced pressure yielded the product as a colorless liquid (52.4 g, 60.8 %; b.p., 84 °C/21 mm Hg, lit.232;42 °C/0.2 mm Hg) 1H NMR (CDCl3): B 4.26 (m, 4H, CH3CfuOP), 1.40 (t, J = 7.10 Hz, 6H, CfuCH2OP) 13C NMR (CDCl3): 65.6 (d, J = 7 .0 Hz, CH3CH2OP), 15,4 (d, J = 7 .7 Hz, CH3CH2OP) 177 Preparation of methyl m-chlorobenzoate (166). m-Chlorobenzoic acid (1.50 g) was dissolved in a solution of methanoVconc. HCI (40:3, v:v). The reaction mixture was refluxed for 5 hr then the solvent evaporated I to yield the product as a pale brown liquid (1.51 g, 92.6 %). MS, m/z (%): 172 (M+2, 12), 170 (M+, 36), 141 (31), 139 (95), 113 (23), 111 (71), 75 (100), 50 (73) 1H NMR (CDCl3): 6 8.02, 7.92, 7.53, 7.37 (m, 4H, aromatic), 3.93 (s, 3H, CfuO) 13C NMR (CDCI3): c5 165.80 (s, C(O)), 134.46, 132.88, 131.82, 129.63, 127.64 (m, PhCl), 52.32 (s, CH3OP). IR (Neat): 2958 (m), 1735 (s), 1581 (m), 1442 (m), 1299 (s), 1288 (s), 1265 (s), 1130 (m), 751 (s). 178 Oxidations of cis- and trans-5-hydroxymethyl-5-methyl-2-thiono-r- 2-ethoxy-1,3,2-dioxaphosphorinane [cis and trans-(116), respectively] by MCPBA. {Al In CDCI,1 A-1: MCPBA (14.1 mg) was added to a solution of the trans -(116) (5.8 mg) in CDCl3 (0.5 ml) and the reaction monitored by 3lp NMR. The starting material (8 61.6) disappeared in 10 mins and the resonances for two intermediates were observed at 8 23.4 (13 %) and 8 -10.3 (11 %) in addition to products at 8 -5.7 ( 9 %) , 8 -6.7 ( 58 %) and 8-7.0 (9 %). After 3.5 hours the intermediates disappeared and the products were identified by spiking the solution with authentic compounds. The products were as follows: 8 -5.7; Cis -5-hydroxymethyl-5-methyl-2-oxo-r-2-ethoxy-1,3,2- dioxaphosphorinane (117) (12 %) 8 -6.7; Trans -5-hydroxymethyl-5-methyl-2-oxo-r-2-ethoxy-1,3,2- dioxaphosphorinane ( 117) (73 % ) 8 -7.1; BPO (15 %) A-2: The same procedure described in A-1 was repeated with cis-(116) (5.5 mg) and MCPBA (13.0 mg) in CDCI3 (0.5 ml). The cis-(116) (8 64.6) disappeared (10 mins) to give intermediates at 8 23.5 (24 %) and 8 10.3 (6 %) together with the products cis - (117) (57 %) and trans -(116) (13 %). The intermediates disappeared after 5.5 hours and the products were as follows: Cis -(117); (70 % ) Trans -( 117); (30 % ) 179 BPO; trace A-3: The control experiment in which cis- and trans-(111) was treated with MCPBA in CHCl3 showed no change in the isomer ratio over 9 hours. (B). In CH3.OH B-1: MCPBA (60.0 mg, 85% purity) was added portionwise in three equal portions to a solution of trans-(116) (20.0 mg) in CH3OH (2.5 ml) and the reaction monitored by 3lp NMR. Three minutes after the addition of the first portion of MCPBA (20mg) an intermediate at o 31.0 was observed together with a trace of another intermediate at 028.9 (3 mins); the concentration of the starting material (8 61.9) began to decrease. After 9 mins, the intermediate at 6 31.0 was present in trace amount and an increase in the intermediate at 6 28.9 and also of another intermediate at 8 27.3 was noted. The intermediate at 8 28.9 increased to a maximum (28 % of total mixture) after 21 mins whereas the intermediate at 8 27 .3 had disappeared. Intermediate at 8 24.4 and 8 24.1 appeared and they were observed throughout the remainder of the reaction. After 1 hour 10 mins, the peak at 8 28.9 had disappeared and the trans-(116) had decreased to 27 % of total mixture. At that stage the intermediates at 8 24.4 and 8 24.1 were the largest components (41 % of total mixture) with the ratio of 1 : 1.5 between them. Other small intermediates at 8 15.2, 8 14.7 were observed in addition to seven product peaks at 6 1.9, 0.8, -4.7, -5.0, -5.3, -5.8 and -6.2. Addition of a second portion (20.0 mg) of MCPBA resulted in the reappearance and development of the substance giving a peak at 6 28.9 as well as a rapid decline in the starting material. which was consumed within 20 minutes of the second addition of oxidant. The third addition of MCPBA (20.0 mg) was made 1 hour after the second addition and the last spectrum was taken after 78 hours 10 180 mins. In the final mixture 5 out of 7 products were identified by comparison of their chemical shift with authentic compounds : 6 -4.7 ; Cis -5-hydroxymethyl-5-methyl-2-oxo-r-2-methoxy-1,3,2- dioxaphosphorinane ( 118); (33 % ) 6 -5.0; Trans-5-hydroxymethyl-5-methyl-2-oxo-r-2-methoxy-1,3,2- dioxaphosphorinane (118); (34 %) 6 -5.3; BPO; (4 %) 6 -5.8; Cis -5-hydroxymethyl-5-methyl-2-oxo-r-2-ethoxy-1,3,2- dioxaphosphorinane ( 117); (3 %) 6 -6.2; Trans -5-h ydroxymeth y1-5-meth y1-2-oxo-r-2-ethoxy-1,3 ,2- dioxaphosphorinane ( 117); (9 %) Unknown minor by-products; 6 1.9 (10 %), 6 0.8 (7 %) B-2: Trans-(116) (4.7 mg) was oxidized with limiting MCPBA (8.0 mg) in CH3OH (0.5 ml). The starting material disappeared after 45 mins and the reaction proceeded similar to B-1. After 20 hr, products cis- and trans-(118) increased significantly whereas the intermediates had disappeared. Two new peaks also appeared at 6 68.4 and o66.6. The final mixture after 60 hr was as follows: Cis-( 118); (25 % ) Tran-(118); (28 %) Cis-(117); (4 %) Trans-(111); (12 %) Unknowns; 6 68.3 (9 %), 66.6 (3 %), 2.6 (8 %), 1.5 (10 %), -7.2 (trace) 181 B-3: MCPBA (80.0 mg, 85% purity) was added portionwise as specified below to a solution of cis-( 116) (20.0 mg) in CH3OH (2.5 ml) with monitoring by 31 P NMR. On the first MCPBA (20.0 mg) addition, an intermediate at o 31.0 was observed with a trace of another intermediate at o28.9. The latter steadily increased for 8 mins while the intermediate at o31.0 decreased after reaching a maximum at 5 mins. At the above stage (8 mins) the starting material (o 63.5) had decreased to 79 % of total mixture and the other intermediates and product peaks which had been observed in the oxidation of trans-(116) in CH3OH began to appear. After 49 mins, only 23 % of cis-(116) was left and the intermediates (o 24.3 and o 24.0) constituted 35% of total mixture. The intermediate at o28.9 disappeared after 58 mins and the major intermediates at o24.3 I and o 24.0 declined (33%). No further change was observed until before the second addition of MCPBA (20.0 mg) at 1 hour 33 mins, when the intermediate at o 28.9 appeared again. The cis-(116) disappeared within 20 mins of the second addition. After total 10 hours, none of the intermediate at o28.9 was left, but the major intermediates (O 24.3 and o24.0, 20 %) and other minor intermediates (O 22.7, o22.3, o 15.0, o 14.6) were observed in addition to the product peaks which had been observed and identified in the previous experiment B-1. The third portion of MCPBA (60.0 mg) was added after 6 hours 15 mins, and the final spectrum which was taken at 88 hours indicated the following: Cis-( 118); (29 %) Trans-(118); (33 %) BPO; (3 %) Cis-(117); (11 %) Trans-(111); ( 4 % ) Unknowns; o7.4 (2 %), o 1.9 (9 %), o0.7 (8 %) 182 B-4: MCPBA (60 mg) was added to a solution of the above isomeric mixture (116; cis : trans = 1:1, 40 mg) in CH3OH (2.5 ml) and the reaction monitored (3lp NMR) until all of the starting material had reacted. After 5 mins, the starting isomers cis-and trans ( 116) had decreased, but with the cis isomer reacting faster. The cis : trans ratio at this stage was 1 : 3 and the intermediate at B 28.9 was the largest peak in the spectrum. Small resonances considered to represent intermediates were also observed at B 33.2, 31.0, 27.3, 24.4, 24.1, 22.8, 22.4, 15.0, 14.7 as well as products which were also noted in B-1. After 30 mins all the starting materials had disappeared and the intermediates at B 24.4 and B 24.1 became the most intense peaks. The intermediate peak at a 24.1 was twice the size of the B 24.4 peak. The solvent was then evaporated, the residue dissolved in EtOAC (2 ml) and the 3Ip NMR spectrum gave one broad peak at a 24.0. No change was observed over the next 10 mins but on washing with Na2S2O3 solution the intermediates disappeared leaving product peaks as follows; B -5.7; Cis-(118); (27 %) a -6.0; Trans-( 118); (22 % ) B-6.7; Cis-(117); (19 %) B-7.2; Trans-(117); (15 %) Unknowns; B 73.5 (1 %), 65.8 (2 %), 2.6 (3 %), 1.5 (10 %), -9.6 (1 %) (C) In agueous acetone. C-1: MCPBA (31.1 mg) was added to a solution of trans-(116) (9.7 mg) in aqueous acetone [0.5 ml, 1:1 (v/v)]. The starting material (B 62.3) decreased quickly (3 %) within 5 mins to give three major products at B 11.6 (40 %), 7.9 (17 %), and -5.5 (31 % ). An intermediate peak at B 30.0, and a number of minor peaks were noted. The intermediate (B 30.3) disappeared after 10 mins and another peak appeared at B 8.3. In 183 the proton coupled spectrum the three peaks at c> 11.6, 8.3 and 7 .8 appeared as doublets indicating them to be hydrogenphosphonates with J PH couplings of 711 Hz, 695 Hz and 673 Hz respectively. The latter hydrogenphosphonate which showed a further J POCH coupling of 22 Hz This was assigned as trans-5-hydroxymethyl-5-methyl-2-oxo r-2-hydro-1,3,2-dioxaphosphorinane (154), and this was confirmed on comparison with an authentic sample. After 1 hr the hydrogenphosphonate at c> 11.6 became the most intense peak then started to decrease. The hydrogenphosphonate at c> 8.1 continued to increase and trans-(154) disappeared after 4 hr. After 21 hr the final mixture was as follows; c> -4.5; (153); (4 %) c> 5.0; BPO; (trace) c> -5.4; Cis-(117); (trace). c> 5.5; Trans-(111); (13 %) Unknowns; c> 8.0 (82 %, hydrogenphosphonate, JpH = 695 Hz), c> 0.6; (1 %) C-2: The same procedure as described for C-1 was repeated using MCPBA (31.5 mg ) and cis-(116) in aqueous acetone [0.5 ml, 1: 1 (v/v)]. The starting material (c> 63.0) disappeared in 5 mins and the overall changes in the spectrum showed similar tendancies as observed for the trans -isomer. However the expected cis-(154) was not observed and in the final mixture after 18 hr the products distribution was as follows; (153); (6 %) Cis-( 117); (25 % ) Trans-(111); (3 %) 184 Unknowns; o8.0 (62 %, hydrogenphosphonate, JpH = 669 Hz), and o0.5; (4 % ). BPO was not detected. Oxidations of cis and trans-5-acetoxymethyl-5-methyl-2-thiono-r- 2-ethoxy-1,3,2-dioxaphosphorinane [cis and trans-(119), respectively] in CDCl3. A:. MCPBA (3.7 mg) was added to a solution of trans-(119) (5.5 mg) in CDCl3 (0.5 ml) and the reaction monitored by 3lp NMR. The first spectrum was taken after 12 mins at which point the reaction was already complete. No further change was noted during the next 24 hours. In the final mixture three peaks were observed and they were identified by 3lp NMR comparisons with authentic standards as follows:. o 58.5; Trans-(119) (unreacted, 9 %) o-8.8; Cis-5-acetoxymethyl-5-methyl-2-oxo-r-2-ethoxy-1,3,2- dioxaphosphorinane (144); (4 %) o-10.3; Trans -5-acetoxymethyl-5-methyl-2-oxo-r-2-ethoxy-1,3,2- dioxaphosphorinane (144);.(87 %) B: A further oxidation using the same procedure as above but with MCPBA (11. 1 mg) and trans -(119) (5.0 mg) in CDCI3 (0.5 ml) was carried out. The starting material had disappeared by the time that the first spectrum was taken (14 mins), and two product isomers were observed as follows: Cis-(144); (6 % ) Trans-(144); (94 % ) 185 C: MCPBA (3.6 mg) was added to a solution of cis-(119) (6 61.5, 5.4 mg) in CDC13 (0.5 ml). Reaction was completed when the first spectrum was run (12 mins) and three peaks were obseived: Cis-(119); (unreacted, 7 %) Cis-(144); (91 %) Trans-(144); (2 %) D: Same procedure as for C was repeated using more MCPBA (11.6 mg) and cis ( 119) (5.5 mg) in CDC13 (0.5 ml). All of the cis-(119) disappeared within 10 mins to give two resonances as follows: Cis-(144); (98 %) Trans-(144); (2 %) Oxidation of 5 ,5-d i methyl-2-thi ono-2-ethoxy-1,3,2- dioxaphosphorinane {133) {A) In CDClJ MCPBA (50.5 mg) was added to a solution of the cyclic thiophosphate (133) (6 62.5, 52.1 mg) in CDCl3, The starting material disappeared in 5 mins to give cyclic phosphate, 5,5-dimethyl-2-oxo-2-ethoxy-1,3,2-dioxaphosphorinane (132, 6 -7.5) as the only product. 186 {B} In CH.3.OH. B-1: The cyclic thiophosphate (133) (B 63.2, 10.7 mg) was oxidized by MCPBA (49.0 mg) in CH3OH. In 5 mins the starting material disappeared to give major products at c5 -4.4 (10 %) and -5.6 (10 %) and minor products at c5 2.5 (2 %) and 1.3 (2 %). Several intermediates also appeared at c5 29.4 (2 %), c5 27.7 (3 %), c5 24.8 (5 %), c5 24.5 (11 %), c5 24.4 (3 %), c5 22.9 (4 %), c5 19.0 (1 %), c5 15.1 (5 %),B 12.4 (23 %), c5 10.7 (16 %), c5 6.6 (3 %). After 15 mins two intermediates at c5 24.9 and 27.7 disappeared, and the lH-coupled spectrum revealed that the the peaks at c5 12.4 and 10.7 were hydrogenphosphonates (JpH of 710 Hz and 705 Hz, respectively). All intermediates decreased slowly with time and the products increased. The final mixture (48 hr) contained: B -4.2; 5,5-dimethyl-2-oxo-2-methoxy-1,3,2-dioxaphosphorinane (137); (53 %) B -5.4; 5,5-dimethyl-2-oxo-2-ethoxy-1,3,2-dioxaphosphorinane (132); (27 %) Unknowns; B 2.6 (6 %), c5 2.1 (6 %), c5 1.5 (7 %) and c5 1.0 (1 %). B-2: As a reference reaction trans-(116) (11.6 mg) was oxidized with MCPBA (55.2 mg) in CH3OH. The starting material disappeared within 5 mins and the reaction proceeded similarly to B-1. The differences were in the formation of product isomers which were not noted in previous reaction, and their distribution. The final (50 hr) products were; Cis-(118); (25 % ) Trans-(118); (33 %) Cis-(117); (9 % ) Trans-(111); (13 %) _., ______ • •, o • •• ,...... ' • •44• • • • -~·~ JO • •" • 187 Unknowns; 6 9.2 (1 %),6 8.5 (2 %), 6 2.8 (4 %), 6 2.3 (7 %), 6 1.7 (5 %) and 6 1.1 (1 %). B-3: With a less amount of MCPBA (13.5 mg) in CH3OH (0.5 ml) the cyclic thiophosphate (133) (10.8 mg) disappeared after 3 hr. Two major products (137) and (132) were formed and major intermediates were observed at 6 24.8 (11 %) and 6 24.5 (19 %) together with other small intermediates. After 16 hr two new peaks were observed at 6 68.1 and 6 66.4, with decrease of intermediates. The final products were identified by comparision with authentic compounds and GC-MS. (137); (32 %); m/z 165 (M-15, 4 %), 152 (4), 139 (10), 125 (12) and 113 (100) (132); (17 %);m/z 179 (M-15, 5 %), 165 (15), 139 (24), 125 (47), 111(32), and 99 (70). Unknowns; 6 68.2 (22 %), 6 66.5 (12 %), 6 2.5 (7 %) and 6 1.4 (10 %) (C) In agueous acetone. The cyclic thiophosphate (133) (6 62.0, 7.4 mg) was reacted with MCPBA (31.0 mg) in aqueous acetone (0.5 ml, 1: 1 = v/v). The starting material disappeared by the time the first spectrum (5 mins) was taken to give a number of peaks as follows: 6 29.6 (6 %), 6 11.3 (48 %, hydrogenphosphonate of JpH = 712 Hz), 6 7.3 [10 %, 5,5- dimethyl-2-oxo-2-hydro-1,3,2-dioxaphosphorinane (138) of JpH = 694 Hz and JpocH = 23 Hz] and 6 -5.9 (36 %, 134). Another peak at 6 7.8 appeared (20 mins) and it kept increasing (42 % at 18 hr) while (138) and other peaks disappeared. After 6 days the final products were as follows: 6 -4.9; 5,5-dimethyl-2-oxo-2-hydroxy-1,3,2-dioxaphosphorinane (135); (7 %) 188 6 -5.9; (132); (33 %) Unknowns; 6 7.8 (assigned as acyclic hydrogenphosphonate, JpH = 671 Hz, 49 %), 6 4.4 (3 %) and 6 0.5 (8 %). As a control reaction a solution of (138) (7.1 mg) in aqueous acetone (0.5 ml, 1: 1 = v/v) was allowed to stand for 30 hr. The starting material (6 7.2, JpH = 695 Hz, JrocH = 22 Hz) slowly converted to a peak at 6 7.6 (assigned as hydrogenphosphonate, JpH = 670 Hz). Same result was obtained from the repeated reaction with MCBA 18 hr. Oxidations of BPS by MCBPA (A) In CDC12 A-1: MCPBA (25.1mg, 5 molar ratio) was added to a solution of BPS (5.2 mg) in CDCl3 (0.5 ml) and the reaction was monitored by 31 P NMR until the reaction was complete. BPS (6 57.2) was converted to BPO (6 -6.8) within 4 mins. A-2: The same procedure was used with MCPBA (15.8 mg, 3 molar ratio) and BPS (5.3 mg) in CDCl3 (0.5 ml). A similar result was obtained but the reaction rate was slower and a trace of BPS was still observed after more than 1 hour. A-3: MCPBA (5.0 mg in 0.1 ml CHCl3, 1 molar ratio) and BPS (5.3 mg) in CDCI3 (0.5 ml) were reacted as above. In this case only partial (80 % ) conversion of BPS to BPO was observed. 189 A-4: With MCPBA (5.0 mg in 0.1 ml CHCl3, 0.5 molar ratio) and BPS (10.9 mg) in CDCl3 (0.5 ml) only 19 % conversion to BPO was noted. B-1: 31 P NMR monitoring MCBPA (25.0 mg) was added to a solution of BPS (5.0 mg) in CD3OD (0.5 ml) and the reaction was monitored by 31 P NMR until no change in the spectrum occurred on further standing. The starting material (BPS, 6 59.4) was slowly converted via several intermediates (major, 6 25.0) to 5 new substances (6 2.9, 2.4, -3.7, -4.0, -4.3). As the reaction proceeded, BPS decreased steadily until after 80 mins none remained and the major intermediate (6 25.0) had increased to a maximum (37 % of total mixture). Thereafter this intermediate declined to zero over 46 hours. Three of the final products were identified by NMR comparisons (31 P and 1H) with authentic compounds. They were: o-3.7; Cis -(118); (38 %) o-4.0; Trans-( 118); (44 % ) o-4.3; BPO; (8 % ). Unknowns (10%); 8 2.9, o2.4 As a control experiment MCPBA (20.0 mg, 82% purity) was added to a solution of BPO (5.1 mg) in CD3 OD (0.5 ml) and the mixture was monitored by 3lp NMR (44 hr). No reaction occurred during the period of monitoring. 190 Similarly, no charge was observed when a solution of BPO and a mixture (7.0 mg) of the isomers of (118) (7.0 mg) with MCPBA (27.0 mg; 82%) in CD3OD was allowed to stand for 27 hr. B-2: lH NMR MCPBA (25.7 mg) was added to a solution of BPS (5.4 mg) in CH3OH (0.5 ml). When the reaction was complete, the solvent was evaporated to give a whit~ solid containing the 5 products previously observed in the oxidation in CD3OD. However, in the lH-spectrum, two additional peaks were observed at 6 3.90 (s) and 6 3.79 (d, J = 11.00 Hz). The latter peak was identified as the methoxy protons from the mixture of two isomers of (118). The former peak (6 3.90) was not identified and was not removed by 020 exchange. B-3: BPS (4.7 mg) was reacted with a less quantity of MCPBA (10.7 mg) in CH3OH. Within 2 hr BPS had decreased to 9 % and products [trans-(118), 18 %], [cis (118), 24 % ] and (BPO) (1 % ) were noted. A major intermediate was also noted at 6 24.8 (19 %) together with other minor intermediates. After 20 hr the level of BPS decreased further, all of the intermediates disappeared except for that at 6 24.1 and a new peak was observed at 6 68.4. In final mixture (40 hr), the resonance at 6 24.1 had also disappeared and the products distribution was as follows; BPS (unreacted); (2 %) Cis -(118); (32 %) Trans-(118); (38 %)_ BPO; (6 %), 191 Unknowns; o68.4 (13 % ), o2.6 (7 %), o2.1 (2 %). (C) In various agueous solvents C-1: In aqueous acetone. MCPBA (25.7 mg) was added to a solution of BPS (5.1 mg) in aqueous acetone- o- 2.5; (153); (17 %) o-3.2; BPO; (41 %) Unknowns; o9.7 (assigned as acyclic hydrogenphosphonate, 39 %, Jp_H = 669 Hz), o9.5 (3 %). In a control experiment MCPBA (30.0 mg, 82 % purity) was added to a solution of BPO (5.4 mg) in aqueous acetone [0.5 ml 1: 1 (v/v)]. No reaction had occured within 22 hours. 192 C-2: In aqueous CH3CN MCPBA (25.8 mg) oxidation of BPS (5.1 mg) in aqueous (D2O) CH3CN [0.5 ml 1:1 (v/v)] resulted in all BPS (B 58.8) being consumed within 40 min. Several small peaks at B 10.0 - B7 .0 were observed in addition to two major peaks at B -4.4 and B -4.8 (BPO). After 24 hours, five peaks at 8 8.8 (7 %), 80 (8 %) 7.1 (12 %), -4.5 (29 %) and -4.8 (44 %, BPO) were observed; the lH-coupled spectrum showed that no hydrogenphosphonates were present. C-3: In aqueous DMF MCPBA (25.4 mg) oxidation of BPS (5.0 mg) in aqueous DMF [0.5 ml, 1: 1 (v/v)] gave 5 peaks (20 mins) as follows: BPS (B 58.7, 60 %), 7.5 (8 %), 7.0 (6 %), - 4.6 (5 %) and BPO (B -5.3, 22 %). Fifty minutes later a new peak appeared at B 6.9 and an increase in all other peaks was noted except for BPS. After 4 hours, BPS, the peaks at 6 7.5 and B 7.0 disappeared, and only three peaks remained at 6 6.9 (21 %, hydrogen phosphonate, Jp_H = 652 Hz), 8 -4.6 [ 14 %, (154)] and B -5.3 (65 %, BPO). Spectroscopic investigation of the intermediate E in the oxidation of BPS by MCPBA in CD30D (CH30H) .(Al 3lp and lH NMR monitorin~. MCPBA (27 .5 mg) was added to a solution of BPS (5.4 mg) in CD3OD (0.5 ml) and the reaction monitored by 3 l P and I H NMR ( 1H NMR spectra were taken immediately after 3lp NMR spectra) until the reaction was complete. In the starting mixture, only resonances attributable to BPS were obsered at B 59.4 (3 1P), and two peaks (1H) at 8 4.47 (d, J = 6.48 Hz) and B 0.83 (s). As the 193 reaction proceeded, the 3l P NMR spectrum showed that the peak from BPS decreased while peaks from products cis-(118), trans-(118) and BPO increased correspondingly. The peak at o25.0 (major intermediate) became the most significant peak (24 %) after 90 min ..In the 1H NMR, peaks at o 4.5 - 3.8 (ring protons); o 3.7 and o 3.4 (P OCfuOH) and at o 1.14 and o 0.82 (-CH3) were assigned as characteristic of the isomers (118), and these, together with peaks at o 4.56 (d, J = 6.36 Hz) and 0.86 (s) for BPO increased while peaks at 8 4.47 - 8 4.4 and at o0.94 behaved in the same was as the major intermediate at 8 25.0 (3 1P) in first rising and subsequently falling to zero. In the final mixture, peaks which were observed as minor products at o2.9 and o2.4 in the 3lp spectra correlate with the peaks at 8 3.5 and 8 3.4 at o0.92 and 8 0.91 in the 1H spectra. (B} Solvent substitution B-1: 2H spectra of the product from the above reaction were also taken. For a reference, CDCl3 was added to the reaction mixture. In addition to solvent peaks (CD3OD, o4.37 and o2.72) four broad overlapping products peaks were observed at 8 3.27 (major), and 8 3.18, o 3.15 and o 3.06 (minor). B-2: MCPBA (57 .2 mg) was added to a solution of BPS (11. 1 mg) in CH3OH (1 ml) and the reaction mixture was divided into two portions A and B (0.5 ml each) after 2 hours. The solvent was evaporated from each portion, and then CD3OD (0.5 ml) was added to one portion A and CH3OH (0.5 ml) to the other portion B. With portion A spectra (3lp and 1H) were taken immediately after dissolving in CD3OD and the intermediate (8 25.0, 3lp) was noted (20 %) as one of the major components. The reactions were monitored (3 1P NMR) until oxidation appeared complete. The solvent was evaporated and spectra (1H and 3lp) were taken with portion A and B in both CD3OD to compare the integral of the methyl (C-CH3) and methoxy groups (P-OCH3) of the isomeric product (118). Compounds in the final reaction mixture were identified 194 to be the same as those in previous oxidations by comparison of the chemical shifts ( 1H I and 31P). From the lH spectra, the integrals of P-OCH3 groups was lesser than that of the integrals of C-CH3 groups in portion (B) while in portion (A) the integrals of P OCH3 groups was similar to that of the integrals of C-CH3 groups. (C) 2-D MCPBA (25.8 mg, 99.9%) was added to a solution of BPS (5.6 mg) in CD30D (0.5 ml) and the 31 P and 1 H spectra were taken after 10 mins. From these spectra the parameters for the 2-D correlation experiment were calculated. The 2-D (Heteronuclear correlation) NMR was then carried out for 4 hours [01 = 1082.03, 02 = 5692.937 SWl = 257.5, D1 = 1.5, S1 = 0H, Pl = 26, D0 = .000003, P3 = 11, *P4 = 22, *D3 = .04545, *D4 = .02273, S2 = 12H, RD = 0, PW = 0, DS = 2, and P9 = 75 (* : optimized for Jp_H = 11 Hz)]. In the final 2-D spectrum, several correlated peaks were found as follows; 3lp NMR (6) lH NMR (6) compounds 33.0 3.15 minor intermediate 25.0 4.20-4.18 maior intermediate 3.0 3.96 minor product 2.5 3.92 minor oroduct -4.0 4.11-4.23 trans-( 118) -4.3 4.57 BPO 195 By slicing through the 31 P peak at 8 25.0 the partial proton spectrum consisting of four peaks at 8 4.18, 4.15, 4.14 and 4.10 (AB quartet) which correspond to the ring protons of the major intermediate was obtained. The above spectrum gives the phosphorus decoupled spectrum for protons which are coupled to phosphorus by a coupling constant of -1 lHz (J POCH = 11 Hz). (D) Direct MS analysis MCPBA (25.1 mg) was added to a solution of BPS ( 5.1 mg) in CD3OD (0.5 ml) and the reaction mixture was subjected to direct MS (chemical ionization, Cl4 as the ionizing gas) analysis after 70 mins, and after the reaction had finished. At 70 mins BPS, the major intermediate (8 25.0), significant products [cis (118), trans-(118) and BPO] and other minor peaks (8 15.9, 15.6, 2.9) were observed by 3lp NNR. A small amount of the reaction mixture was placed on the direct insertion probe, the solvent evaporated and the probe insened into the spectrometer. Temperature was elevated from 130° to 250°C in 4 mins and the ion current was monitored (9 min). After 4.16 mins, the peaks from MCBA [M + 1 = 157 (100%), M + 29 = 185, M + 41 = 197] and the deuterated methyl ester of MCBA(167) [M + 1 = 174 (63%), M + 29 = 202, M + 41 = 214] were predominant in the spectrum. Small peaks above 313 m/z were observed at 5.16 mins. These peaks (mainly m/z 313 - 315) increased and then began to disappear after 6.36 mins at which point the peak for MCBA and the methyl ester of MCBA also decreased to zero. After 6.36 mins the peaks from BPO [M + 1 = 165 (50%), M + 29 = 193, M + 41 = 205] and isomer products (118) [M + 1 = 200 (100%), M + 29 = 228, M + 41 = 240] were observed. In the final reaction mixture after the oxidation was complete, five products were observed in the 3 l P NMR spectrum as anticipated; minor products at 8 2.9 and 8 2.4, isomer products (118) and BPO. This final mixture was analysed by MS in the same 196 manner as above and peaks from three compounds were observed as follows: m/z 165 (M + 1, 30%), [193 (M + 29), 205 (M + 41)] for product (BPO); m/z 200 ( M + 1, 100%), [228 (M + 29), 240 ( M + 41)] for isomer products (118) and m/z 297 (M + 1, 45%), [325 (M + 29), 337 (M + 41)] for unknown compound. The latter material must correspond, by exclusion, to the minor products at 6 2.9 and 6 2.4 (3 1P). Thus, these must be isomers of molecular weight of 296 and they were assigned as 5-methyl-5- deuteromethylsulfonylmethy 1-2-oxo-2-deu teromethy 1-1,3,2-dioxaphosphorinane ( 169). {E) GC-MS analysis MCPBA (27.1 mg) was added to a solution of BPS (6.3 mg) in CH3OH(0.5 ml) and the reaction monitored by GC*. After 5 min, MCBA was the major component (Rt = 13.67, 98 %) and other small peaks at 4.37 mins (1 %), 12.10 (trace), 12.63 (trace), 12.95 (trace), 17.94 (trace, BPS) and 18.90 (1 %) were observed. On monitoring the reaction for 30 hr the peak at 4.37 mins was observed to rise and fall, the peak at 12.10 mins increased to 22 % and the MCBA decreased to 78 %. At 3 hours the reaction mixture was analysed by GC-MS and three peaks w~re determined. MCBA (167); Rt= 12.10 mins, m/z 156 (M+, 87%), 158 (M+2, 30%), 139 (M-17, 100%), 111 (M-45, 70%). Methyl ester of MCBA(l68); Rt= 12.10, m/z 170 (M+, 30%), 139 (M-31, 100%), 1 ll(M-59, 56%) Unknown; Rt= 4.37 mins, m/z 179 (M+, 18 %), 180 (M + 1, 4 %), 149 (M- 30, 26 %), 112 (M-68, 100%), 114 (M-66, 35 %), 77 (M-102, 74 %). * GC - parameters : Column: ov-17 (25 m x 0.25 mm) 197 oven: 70 ·c (3 min) ------>90 ·c (1 min) ------>260 ·c (15 min) 5 °C/min 30 °C/min N2; 57mVmin, H2; 44ml/min, air; 35mVmin Detector, FID (280 °C) As a control experiment a solution of MCBA (25.0 mg) in CH3OH (0.5 ml) was allowed to stand at room temperature, and monitored by GC over 24 hours. A trace amount of methyl ester of MCBA (166) was observed just after the preparation of the reaction mixture but no further change occurred up to 24 hours. (f) Effect ofTrimethyl phosphite {TMP) MCPBA (27 .5 mg) was added to a solution of BPS (5.4 mg) in CH3OH (0.5 ml) and a spectrum was taken after 1 hour and 10 mins. Little BPS remained but a new peak at o 12.8 was observed in addition to those peaks for the major intermediate E (o 25.0) and products. The 1H-coupled spectrum showed the new peak (o 12.8) was a hydrogenphosphonate, characterized by its large coupling constant OP-H = 712 Hz). An excess of TMP was added to the reaction mixture until the TMP peak (0141.8) was observed in the spectrum. Products identified were trimethyl phosphorothionate (MeO)JP=S; (TMPS, o 74.6), trimethyl phosphate (TMPO,o 3.7), isomers (118), BPO ,dimethyl hydrogenphosphonate (o 13.5). and unidentified minor products (669.9, 68.5, 36.5). After standing for one day the TMP (o 141.8) resonance disappeared totally, and the peaks at o36.4 and o 13.4 (dimethyl hydrogenphosphonate) increased. As a control experiment TMP (7.3 mg) was added to a solution of BPS (5.3 mg) in CH3OH (0.5 ml) and the reaction monitored by 3lp NMR. No reaction was observed within 18 hours. 198 Other control experiments were carried out using MCPBA (26.9 mg) and TMP (8.3 mg) in CH3OH (0.5 ml). On addition of MCPBA, all of the TMP (6142.4) disappeared immediately and an increase in the resonance for TMPO (6 3.7) occurred. No further change was observed up to 93 hr. The same procedure was carried out with MCBA (21.2 mg) and TMP (8.3 mg) in CH3OH (0.5 ml). Dimethyl hydrogenphosphonate(Jp.H = 710 Hz) formed and increased throughout the monitoring period (93 hours). TMPS (16.0 mg) was oxidized by MCPBA (65.0 mg) in CH3OH (0.5 ml). TMPS (6 73.8) had disappeared within 7 mins to produce 6 peaks at 647.1 (trace), 30.0 (13 %), 25.3 (22 %), 19.4 (6 %), 12.6 (18 %) and 2.7 (41 %). After 12 hours, only two peaks remained and they were identified as follows : 6 25.3 (27 %) : Bis(dimethoxy phosphinyl)disulfide (131, quintet: JpocH = 12.5 Hz) 6 2.7 (73 %): TMPO (G) Monitoring of hydrogenphosphonate MCPBA (28.2 mg) was added to a solution of BPS (5.6 mg) in CH3OH (0.5 ml) and the reaction mixture monitored by 31 P NMR until the reaction was complete. After 10 mins hydrogenphosphonate (612.8) was the most intense peak and intermediate at 629.9 and 6 25.0 were observed. After I hour the peak at 6 29.9 had disappeared and the intermediate at 6 25.0 became the biggest peak (31 % of the total mixture). The hydrogenphosphonate was still present in considerable amount and three other minor intermediates (6 16.0, 15.6, 7.3) were observed in addition to product peaks which comtrised unknowns (6 2.8, 2.3), cis- and trans-(118) and BPO. 199 Same reaction repeated using MCPBA (26.1 mg) and BPS (5.4 mg) in CH3OH (0.5 ml) and the reaction monitored until all the BPS had disappeared (3 hours). The intermediate at 6 25.0 was observed as the biggest peak and hydrogenphosphonate was also observed with small intermediate at 6 16.0, 6 15.6, 6 7.3 and 5 peaks which were the final proudcts in previous reaction. Water (30.1 mg) was added to the mixture and I the reaction was monitored further. Within 5 mins 30% of the major intermediate (0 25.0) decreased and 4 hours later had almost disappeared. The hydrogenphosphonate resonance increased up to 1.8 times its size before the addition of water. Another significant change was the ratio of isomer products (cis: : trans;). Before the addition the ratio was 1 : 1 and changing to 1 : 2.2 after 4 hours. After 15 hours, hydrogenphosphonate decreased from 21 % to 10 %. 84 hours later, it disappeared and the tatio of the isomeric products was 1 : 2.6. Oxidation of triethyl phosphorothionate (TEPS) and triemethyl phosphorothionate (TMPS) {A) In CDCl3 MCPBA (45.0 mg) was added to a solution of TEPS (16.0 mg) in CDCl3 (0.5 ml) and the reaction monitored by 3lp NMR. The first spectrum (5 mins) showed that all of the starting material (o 67.5) had been converted to triethyl phosphate (TEPO, o- 1.7). No by products other than a sulfur precipitate were observed. Similarly TMPS (O 72.8, 12.2 mg) was converted completely to trimethyl phophate (6 7.1, TMPO ) by MCPBA (35.3 mg). 200 (B) In CH_3OH B-1: TEPS (16.0 mg) in CH3OH (0.5 ml) was oxidised with MCPBA (45.2 mg). Within 5 mins most of the starting material (6 68.3) had disappeared and products peaks were observed at 6 29.9 (8 %), 28.2 (8 %), 26.5 (6 %), 25.2 (6 %), 23.5 (6 %), 21.7 (2 %), 2.6 (8 %), 1.5 (13 %), 0.3 (18 %) and -0.9 (25 %). After 20 minutes the first three peaks (6 29.9, 28.2, 26.5) had disappeared. The other 7 peaks remained after 15.5 hours. Three of the seven peaks were identified in the final reaction mixture by comparison of chemical shift (3lp) with authentic compounds or by addition of authentic material: 6 21.7; Bis(diethoxy phosphinyl) disulfide.(130); (6 %) 6 0.3; Diethyl methyl phosphate (69); (24 %) 6 -0.9; TEPO; (20 % ) Unknowns; 6 25.2 (7 %), 23.5 (9 %), 2.6 (15 %), 1.5 (19 %) The final reaction mixture was analysed by GC* and a peak at 7.5 mins was identified as diethyl methyl phosphate (69), one at 9 mins as TEPO and one at 12.5 mins as bis(diethoxy phosphinyl) disulfide (130). MCBA was observed at 18.6 min. Unidentified peaks were observed at 5.0 mins, 6 mins, 10.5 mins, and 11.5 mins. *GC parameters: Column: ov-17 (capillary, 0.25 x 25 m) Oven: 100 ·c (2 mins) ------>150 ·c (5 mins) ------>230 ·c (10 mins) 5 °C/mins 30 "C/mins N2 : 57 mVmin, H2: 44 ml/min, air :35ml/min 201 B-2: The oxidation described above was repeated with MCPBA (37 .5 mg) and TEPS (12.0 mg) in CH30H (0.5 ml) and the reaction mixture was analysed by GC-MS after 20 hours. Six compounds were identified as follows; Trimethyl phosphate (TMPO): 140 (M+, 10%) 110 (100%) Dimethyl ethyl phosphate (126): 153 (M+-1, 3%) 127 (100%) 109 (60%) Diethyl methyl phosphate (69): 168 (M+, 2%), 141 (50%), 113 (100%), 95 (65%) TEPO: 182 (M+, 3%), 155 (53%), 127 (53%), 99 (100%), 81 (47%). Methyl ester of MCBA (166): 172 (M+ + 2, 30%), 170 (M+, 70%), 141 (33%), 139 (100%), 113 (19%), 111 (57%). MCBA (167): 158 (M+ + 2, 25%), 156 (M+, 75%), 141 (33%) 139 (100%), 113 (20%), 111(60%). Thus, the previously unidentified peak in B-1 at o 2.6 (31 P) was TMPO and another unidentified peak at ol.5 (3 1P) was dimethyl ethyl phosphate (126). The Methyl ester of MCBA (166) was also identified. (C) In dimethyl maleate MCPBA (22.0 mg, 3 molar) was added to a solution of TEPS (8.0 mg) in dimethyl maleate (0.5 ml). The starting material (o 69.0) was consumed within 5 mins and TEPO (6 -3.0) was obtained. 202 {P} In agueous acetone. D-1: MCPBA (20.5 mg) added to a solution of TEPS (7.1 mg) in aqueous acetone (0.5 ml 1 : 1, v/v). Within 5 mins TEPS (6 68.1) disappeared and three peaks were observed at 6 29.7 (20 %), 10.8 (23 %) and 0.18 (57 %). The peak at 6 10.8 had a large coupling constant (Jp_H = 717 Hz) indicating to be a hydrogenphosphonate. The peak at 6 29.7 was lost after 1 hour. Spiking authentic compounds into the reaction mixture confirmed that the peak at 6 10.8 was diethyl hydrogenphosphonate (128, 41 %) and the peak at 6 0.18 was TEPO (60 %). After 3 days, another hydrogen phosphonate peak at 67 .0 (Jp_H = 670 Hz) was observed whereas diethyl hydrogen phosphonate decreased. After one week, the latter compound was no longer evident, but two new hydrogenphosphonates were observed at 6 7.0 and 6 4.6. The former was identified as ethyl hydrogenphosphonate ( 176, Jp_H = 679 Hz), the latter as phosphonic acid (177, Jp_H = 666 Hz). D-2: The reaction was repeated with a less amount of MCPBA (40.0 mg) and TEPS (22.1 mg) in aqueous acetone [0.6 ml, 1 : 2 (v/v)]. Two minutes after addition of oxidant, two intermediates at 6 27.0 (40 %), 8.1 (8 %) and TEPO (6 -2.2, 43 %) were observed in addition to unreacted TEPS (3 %). After 7 mins, a peak at 6 20.1 started to increase with a corresponding decrease in the peak at 6 27 .0. Only a trace of the peak at 627 .0 remained after 1.5 hours and the peak at 6 20.1 became one of the major products (30 %). The peaks at 6 20.1 was identified as bis(diethoxy phosphinyl) disulfide (130) by its chemical shift, and the peak at 6 8.1 was diethyl hydrogenphosphonate (128, Jp_H = 722 Hz). After 15.5 hours, (130) decreased (7 %) and diethyl phosphorothioic acid (6 61.9, 13 %) was observed in addition to resonances at 6 20.3 (6 %), 6 8.1 (128, 4 %), TEPO(6 -2.3, 63 %) and other minor peaks. 203 Oxidation of the sulfenate ((EtO)iP(O)SOCH 3 ) by MCPBA in CD30D. MCPBA (excess) was added portionwise to a solution of the sulfenate (146), in CD3OD until the reaction was complete (21 hr), as determined by 3lp NMR. Starting material (6 27.5) contained diethyl methyl phosphate (69, 6 1.4) and trace amount of impurities (6 22.7, 6.0, 1.4, 0.2 and 12.0) which were not identified. Ten minutes after the first addition of MCPBA the starting material had disappeared completely to yield new major peaks at 6 22.7 and 6 1.4. The peak at o22.7 appeared as a quintet (JPOCH = 9. lHz) in the lH-coupled spectrum and it was identified as bis (diethoxy phosphinyl) disulfide (130). More MCPBA was added and the disulfide disappeared to yield a new major peak at 6 1.35 which was assigned as diethyl deuteromethyl phosphate (69-d3). As a control experiment a solution of diethyl sulfenate in CD3OD was allowed to stand at room temperature. After 15 mins, the starting material (6 27 .5) had decomposed giving a cluster of small peaks between o24.2 - 22.4 and at o 1.42. Reaction of trimethyl phosphite (TMP) with bis(diethoxy phosphinyl) disulfide [(EtO)2P(O)S)i]. TMPS (21.3 mg) was added to a solution of the disulfide (130, 11.7 mg) in CH3OH (0.5 ml). The disulfide (o 22.8) disappeared after the addition of TMP to yield product peaks at o74.5 (31 %), o66.1 (37 %) and o 1.1 (32 %). Excess TMP occurred at 6 142.3 and its hydrolysis product, dimethyl hydrogenphosphonate, at 6 13.3 The peak at 6 74.5 was identified as TMPS, o66.1 as diethyl phosphorothioic acid (10) and o 1.1 as diethyl methyl phosphate (69) each by spiking with authentic compounds. These assignment were confirmed by GC-MS: TMPS; m/z 156 (M+, 39%) 126 (21 %), 93 (100%) 204 Diethyl phosphorothioic acid (10); m/z 170 (M+, 20%), 142 (10%), 138 (25%) Diethyl methyl phosphate (69); m/z 168 (M, 1%), 141 (70%), 113 (100%), 95 (58%). Oxidation of bis(diethoxy phosphoinyl) disulfide (130) by MCPBA. MCPBA (32.2 mg) was added to a solution of the disulfide (11.4 mg) in CH3OH (0.5 ml). After 24 hr the starting material (8 22.6) had decreased to 53 % with a concurrent increase of the peak at o0.5 as a major product (diethyl methyl phosphate, 37 % ). GC-analysis confirmed this and also showed the expected presence of MCBA and its methyl ester (166). As a control experiment, a solution of the disulfide (11.4 mg ) and MCBA (24.4 mg) in CH3OH (0.5 ml) were examined. No appreciable change (3lp NMR) was noted on standing for 24 hours at room temperature. (B) In agueous acetone. MCPBA (31.4 mg) was added to a solution of the disulfide (11.7 mg) in aqueous acetone [0.45 ml, 1 : 2 (v/v)] and the reaction monitored (3lp NMR). The disulfide (8 22.5) contained two trace impurities at 8 22.7 and 0.2. Within 5 mins, an intermediate peak at 8 16.9 was observed and the peaks at 810.4 and 0.2 slowly increased. The coupled 3 l P spectrum showed the peak at o 10.4 was diethyl hydrogenphosphonate (Jp.H = 703 Hz) and after 4 hours this was 27 % of the reaction mixture. The peak at o 0.2 at this time had increased to 31 % . After 11 hours the unreacted disulfide (12 %), diethyl hydrogenphosphonate (128, 35 %), diethyl 205 phosphoric acid (127, 6 0.2, 44 %) and ethyl hydrogenphosphonate (176, 6 6.9, 9 % Jp_ff = 672 Hz) were noted. After 44 hrs, the disulfide decreased (4 %) with increase of (127) (56 % ). The diethyl hydrogenphosphonate also decreased (17 %) with an increase of the ethyl hydrogenphosphonate (23 %). For a control experiment a solution of the disulfide (11.7 mg) and MCBA (23.2 mg) in aqueous acetone [0.45 ml, 1:2 (v/v)] was allowed to stand for 24 hours. Diethyl phosphorothioic acid (10) and diethyl phosphoric acid (127) were observed, and these werre associated with a decrease of the disulfide ( 130). Oxidation of Jl.,il.-diethyl-.S,-phenyl phosphorothiolate-18Q (186, 18Q.thiolate) by MCPBA in CH30 H A: MCPBA (34.3 mg) was added to a solution of 18O-thiolate (186, 12.3 mg) in CH3OH (0.5 ml) and starting material (6 25.1) was partly converted (73 %) to diethyl methyl phosphate (6 1.2) in 10 hours. GC-MS analysis indicated diethyl methyl phosphate to contain 18Q; viz. although the molecular weight (M+ = 170) ion was not seen fragment ions were observed at m/z 143 (M- 27, 60%), 115 (M- 55, 100%), 97 (M - 73, 55%) indicating the presence of l8Q when compared to unlabelled compound (see below). The unreacted starting material, MCBA and its methyl ester were also identified B: As a control experiment unlabelled O,O-diethyl-S.-phenyl-phosphorothiolate (129, 11.7 mg) was oxidized with MCPBA (35.3 mg) in CH3OH and only diethyl methyl phosphate (69, 25 %) was observed after 10 hours apart from the unreacted starting material (75 %). The reaction mixture was also analysed by GC-MS for direct comparision with the labelled experiment. Diethyl methyl phosphate (69) gave fragment ions at m/z 141 (M-27, 60%), 113 (M-55, 100%), 95 (M-73, 61 %); the molecular ion (M+ = 168) was not observed. 206 A control experiment with a solution of Q,O-diethyl-,S.-phenyl phosphorothiolate (23.6 mg) in CH3OH (0.5 ml) showed no change after standing at room temperature for 56 hours. 3lp NMR analysis resolved (EtOh P(O)OCH3 (69) from (EtOh P(18Q) OCH3 (187) by 4.6 Hz. It was found that (187) contained 12 % of (69). The same ratio of labelled to unlabelled were found in the starting materials (186) and (129). Oxidation of 0,0-diethyl-S-phenyl phosphorothiolate (129) and a S,5-dimethyl-2-S-phenyl-2-oxo-1,3,2-diox,phos phorinane (139) by MCPBA in aqueous acetone A:. MCPBA (43.7 mg) was added to a solution of (129) (14.4 mg) in aqueous acetone [0.5 ml, 1 : 1 (v/v)] and the starting material (o 25.4) was completely converted to diethyl phosphate (127, o0.4) in 4 hours and 10 mins. In a blank experiment with (129) (23.6 mg) in aqueous acetone no change was observed at room temperature within 15 hours. B: Oxidation of the cyclic thiolate (139) (ol 7 .0, 8.0 mg) also gave only one product d which was 5,5-dimethyl-2-hydroxy-2-oxo-1,3,2-dioxpphosphorinane (135, o-4.8, 86%), together with unreacted starting material (14%) (11 h.). In a control experiment with (139) (6.0 mg) and MCBA (11.6 mg) in aqueous acetone, no change was observed after 11 hours. 207 Oxidation of 5-hydroxymethyl-5-methyl-2-thiono-2-ethoxy-1,3,2- dioxaphosphorinane (116) by S02Cl2 in CDCl3. fil.lH NMR monitoring SO2Cl2 (11.4 mg) was added to a solution of trans -(116) (5.3 mg) in CDCl3 (0.5 ml). All the starting material instantly disappeared and only BPO was observed (1H NMR) as a product. The same result was obtained (1H NMR) as above from the reaction of SO2Cl2 (11.4 mg) and mixture of cis and trans -(116) (14.4 mg). illl_3lp NMR monitoring 31p NMR monitoring was carried out while SO2Cl2 (66.0 mg) was added portionwise (22.0 mg x 3) to a solution of a mixture of isomers of (116) (95.7 mg,) in CDCl3 (0.5 ml). The starting material, cis (0 63.1) to trans (0 60.7) ratio was 1.5 : 1. The isomers appeared to react, forming BPO, at the same rate. Same results were obtained with individual isomers [cis-isomer (7.1 mg) and SO2Cl2 (35.6 mg); trans isomer (6.0 mg) and SO2Cl2 (20.4 mg)] {C) Isolation of product The reaction was repeated on a larger scale with SO2Cl2 (57.0 mg) and the mixture of isomers (56.7 mg) in CDCl3 (0.5 ml) until all the starting material had reacted. The reaction mixture was filtered and evaporated and BPO was isolated as a pale yellow solid (32.3 mg) 208 Reaction of 5-hydroxymethyl-5-methyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinane (117)with SO2Cl2. SO2Cl2 (14.4 mg) and a mixture of cis and trans (117) ( 13. 7 mg) was reacted in CDCl3 (0.5 ml). The starting material disappeared instantly to give products. On a larger scale experiment with SO2Cl2 (14.8 mg) and mixture of isomer (101.4 mg) in CH2Cl2 (1 ml), the product was isolated as a colourless oil (167.7 mg) by column chromatography (silica gel, EtOAC as eluent) and identified as 5-chlorosulfonylmethyl- 5-methyl-2-oxo-2-ethoxy-1,3,2-dioxaphosphorinane ( 191) Analysis (C7H14O7ClPS), Found(%): C: 27.54 H: 4.84 Calculated(%): C: 27.23 H: 4.54 MS, rn/z (%): 401 (M+l, 1)309 (M+,3), 283 (18), 281 (48), 181 (40), 165 (40), 151 (40), 99 (100) 31p NMR (CDCl3): o-5.1, -8.1 1H NMR (CDCl3): o4.67 (s, 2H, CH2OSO2Cl), 4.47 (s, 2H, CfuOSO2Cl), 4.28-4.10 (m, 8H, CH2OP), 1.40 (t, J = 7.21 Hz, 3H, CfuCH2OP), 1.48 (t, J = 8.43 Hz, 3H, CH3CH2OP), 1.17 (s, 3H, Cfu), 0.99 (s, 3H, CH3) IR (Neat): 2990 (s), 1300 (s), 1195 (s), 1170 (m), 1079 (s), 045 (s), 1010 (s), 963 (s), 914 (m), 859 (s), 840 (s). Oxidation of 5,5-dimethyl-2-thiono-2-ethoxy-1,3,2-dioxaphos phorinane (133) by S02Cl2. A: SO2Cl2 (216.6 mg) was added to a solution of the dioxaphosphorinane (308.1 mg) in CH2Cl2 (3 ml). When all starting material had disappeared (tic) the solvent was • 209 evaporated and the product mixture was separated by column chromatography (silica gel , CHCI3 + EtOAC as eluents). The major product was obtained as white crystals (88.8 mg) which was identified as bis (5,5-dimethyl-2-oxo-2-ethoxy-1,3,2- dioxaphosphorinan-2-yl) disulfide (192). m.p.: 125-129 ·c MS, rn/z (%): 330 (M+- 32, 50%), 164 (100) IH NMR (CDCl3): 8 4.33 (d,d, J = 11.17 Hz, 1.19 Hz, 4H, CfuOP), 3.97 (d,d,t, J = 11.38 Hz, 25.08 Hz, 1.31 Hz, 4H, CfuOP), 1.33 (s, 6H, Cfu), 0.91 (s, 6H, CH3) 13C NMR (CDCl3): 8 79.2 (s, CH2OP), 32.5 (d, J = 7.0 Hz, C4), 21.9 (s, CH3), 20.1 (s, CH3) IR (KBr): 2976 (s), 2908 (w), 2887 (w), 1479 (s), 1411 (w), 1378 (m), 1289 (s), 1215 (w), 1053 (s), 1005 (s) 979 (s), 958 (m), 924 (m), 851 (s), 841 (s), 783 (s), 774 (s), 560 (s), 532 (s), 482 (s). B: The reaction was repeated by portionwise addition of SO2Cl2 (22.0 mg x 2) to a solution of the (133) (50.1 mg) in CDCl3 (0.5 ml) with 3lp NMR momitoring. After the first addition of SO2Ch, disulfide (192, 8 12.4 ) and the peak assigned as 5,5- dimethyl-2-oxo-2-chlorosulfenyl-1,3,2-dioxaphosphorinane (193, 8 9.6) were observed together with unreacted starting material (8 61.4). All starting material disappeared after one more addition of SO2C~2 and (193) was observed together with unknown minor peak at 8 -2.8. 210 Reaction of BPS with S02Cl2. SO2Cl2 (22.8 mg) was added portionwise (11.2 mg x 2) to a solution to BPS (5.1 mg) in CDCl3 (0.5 ml) with monitoring (1 H). With the first portion of SO2Cl2 about 70% of the starting material was consumed and small peaks of BPO appeared together with singlets at 6 3.79, 6 1.0 and a multiplet at 6 4.42 - 4.28. The BPS disappeared totally after another portion of SO2Cl2 was added. BPO was observed as minor product and the peaks at 6 3.79 (s), 1.0 (s) and 6 4.42 - 4.28 were from the I major product. This was subsequently isolated and identified as trans-5-chloromethyl-5- methyl-2-oxo-r-2-chloro-1,3,2-dioxaphosphorinane (195) from a larger scale reaction using SO2Cl2 (399 mg) and BPS (315.0 mg) in CH2Cl2 (10 ml). The chlorodioxaphosphorinane (195) was obtained as colourless crystals (159.3 mg) and BPO (minor) as white solid (52.0 mg) after column chromatography (silica gel; CHCl3 / EtOAC as eluents). Trans-5-chloromethyl-5-methy 1-2-oxo-r-2-chloro- l ,3,2-dioxaphosphorinane ( 195) (major product) mp: 58 - 64 ·c. MS, m/z (%): 221 (M+2, 2) 219 (M+, 3), 180 (12), 169 (12), 156 (14), 129 ( 17), 119 (16), 117 (47), 104 (33), 102 (100). 1H NMR (CDCl3): 6 4.36 (m. 4H. CfuOP), 3.79 (s. 2H, CfuCl), 1.00 (s. 3H. Cfu) 13C NMR (CH2Cl2): 8 74.8 (d. J = 8.0 Hz, CH2OP), 46.6 (s. CH2Cl), 37.0 (d. J = 5.3 Hz, C4), 15.9 (s. CH3) 211 IR (KBr): 3437 (m. br), 2982 (w), 2905 (w), 1639 (w), 1482 (m), 1471 (s), (m), 1310 (s), 1113 (s), 1051 (s), 1007 (s), 1991 (s), 918 (s) 883 (m), 855 (s) 794 (s),738( m) 561 (s), 494( s). As a control experiment, BPO (4.8 mg) was treated with SO2Cl2 (11.4 mg) in CDCI3 (0.5 ml). No reaction occurred. 212 Methanolysis of BPO. !A)_3lp NMR monitoring. A solution of NaOMe in methanol (0.05 ml, 0.5 M) was added to a solution of BPO (11.7 mg) in CH3OH (0.5 ml). In 10 mins, the starting material (6 -3.8) decreased to 92 % and trans--5-hydroxymethy-5-methyl-2-oxo-r-methoxy-1,3 ,2- dioxaphosphorinane (118, 6 -3.5, 6 %) was observed, together with an unidentified peak at 6 5.5 (3 % ) which changed little through the reaction. After 2 hours, the acyclic phosphate (202, 6 3.5, 2 %) was observed and this increased to 22 % after 13 hours while the starting material decreased to 4 %. The trans -(118), now 67 %, and cis - (118) (4 %) were also observed. The starting material disappeared after 21 hours and the reaction underwent virtually no further change in the distribution of the products up to 180 hours. In the final mixture (180 hours), the distribution of the three major products, (202), cis-(118) and trans-(118)was 45 %, 17 % and 30 %, respectively. Other small peaks were noted at 6 5.5 (2 %), 4.9 (4 %) and -6.1 (2 %). With a large excess of NaOMe (0.5 ml, 1.0 M) with BPO (5.3 mg), the BPO disappeared in 4 min and the same results were obtained as above after 70 min to give the acyclic phosphate (56 %), thecis isomer (8 %) and the trans isomer (26 %). (B) Isolation of the trans - 5-hydroxymethy-5-methyl-2-oxo-r-methoxy-1,3,2- dioxaphosphorinane ( 118). B-1: Methanolic NaOMe solution (1 ml, 1.0 M) was added to a stirred solution of BPO (100.0 mg) in CH3OH (4 ml). After +O minutes the reaction mixture was adjusted to pH 6 with dilute HCl (10 %), the solvent evaporated and the residue taken up in absolute ethanol and filtered. The filtrate was evaporated to give a white solid which was separated on preparative tlc (1 mm) in acetone-CHCl3 (1: 1). A band of Rf= 0.52 was scraped from the plate, extracted with acetone and evaporated to give the product 213 which was identified as trans-5-hydroxymethy-5-methyl-2-oxo-r-2-methoxy-1,3,2- dioxaphosphorinane (118), as white crystals (66.8 mg). m.p: 95-97°C MS, m/z (%): 237 (M + 41, 5 %), 225 (M + 29, 9), 197 (M + 1, 100) 3lp NMR (CDCl3): -5.9 lH NMR (CDC13): 4.21 (d.d.t, J = 21.64 Hz, 11.49 Hz, 1.44 Hz, 2He, Cfl2OP), 4.04 (d.d.t, J = 1.92 Hz, 11.48 Hz, 1.50 Hz, 2Ha, Cfl2OP), 3.81 (d, J = 10.99 Hz,.3H, CfuOP), 3.76 (s, 2H, Cf4OH), 0.85 (s, 3H, Cfu) 13C NMR (CDCl3): 73.7 (d, J = 6.8 Hz, CH2OP), 62.8 (s, CH2OH), 53.87 (d, J = 5.4 Hz, CH3OP), 37.0 (d, J = 5.4 Hz, C4), 15.4 (s, CH3) IR (KBr): 3433 (s), 3391 (s), 2968 (m), 2904 (m), 1477 (m), 1281 (s), 1207 (m), 1076 (s), 1062 (s), 1035 (s), 1009(s), 987 (s), 918 (m), 861 (s), B-2: In a similar fashion aqueous NaOH (0.4 ml, 1.5 M) was added to a stirred solution of BPO (102.0 mg) in CH3OH (4.0 ml). Work up as described for B-1 again yielded trans-5-hydroxymeth y-5-methy 1-2-oxo-r-2-methoxy-l ,3 ,2-dioxaphosphorinane (118) (64.1 mg). (C) Decoupling and simulation experiments. Decoupling of each ring proton was carried out in an attempt to assertain the cross-ring coupling between ring protons. However, the resulting peaks were too distorted, probably because of the small difference in chemical shifts between the axial and equtorial protons, and the spectrum could not be interpreted. The lH-spectrum of 214 ring protons was simulated with the same procedure as previously described for (132) using the data reported above (B-1), and same spectrum was obtained. (D) Isolation of the acyclic phosphate (202) NaOMe solution (1.0 ml, 0.5M) was added to a solution of BPO (104.6 mg) in CH3OH (5 ml) and the reaction monitored until all BPO had reacted. The solution was evaporated to dryness after the reaction mixture was adjusted to pH6 with dil HCI (10%), and the residue taken up in acetone, filtered then evaporated to yield a colourless oil (141.9 mg) which was separated by preparative tic [1 mm, CHCl3- acetone (1:1) x 1]. The tic band at Rf= 0.30 was scraped and extracted (acetone) to give a colourless oil (47.0 mg), which was identified as acyclic phosphate (202). MS, rn/z (%): 269 (M + 41, 4), 257 (M + 29, 2), 229( (M + 1, 92), 127 (100) 3lp NMR (CDCl3): o4.0 1H NMR (CDCl3): o4.17 (d, J = 9.37Hz, 2H, CfuOP), 3.80 (d, J = 1 l.14Hz, 6H, CfuOP), 3.61 (m, 4H, CfuOH), 0.83 (s, 3H, CH3) 13C NMR (CDCl3): o69.7 (d, J = 5.5 Hz, CH2OP), 67.0 (s, CH2OH), 54.7 (d, J = 5.9 Hz, CH3OP), 41.4 (d, J = 4.7 Hz, C4), 16.4 (s, CH3) IR (neat): 3400 (s,br), 2960 (s), 2870 (s), 1650 (m), 1460 (s), 1250 (s, br) 1180 (s), 1020 (s, br), 860 (s), 780 (s) 215 Methanolysis of trans-5-hydroxymethy-5-methyl-2-oxo-r-2- methoxy-1,3,2-dioxaphosphorinane (118). Methanolic NaOMe solution (0.05 ml, 1.0 M) was added to a solution of trans (118) (7.3 mg) in CH3OH (0.5 ml) and in 10 mins the acyclic phosphate (6 %) was observed together with a decrease in starting material (94 %). The peak for cis-(118) (7 %) appeared after 2 hours. After 14 hours, the starting material had decreased to 30 %, the cis-(118) had increased to 16 % and the acyclic compound (202) to 54 %. Methanolysis of acyclic phosphate (202). Methanolic NaOMe solution (0.05 ml, 1.0 M) was used in the same manner described for the previous reaction with acyclic compound (202, 6.4 mg) in CH3OH (0.5 ml). In 10 mins traces of two peak at 8 -4.0 and -4.2 were observed which grew to 12 % and 12 %, respectively, after 2 hours, while the starting material decreased to 76 %. After 14 hours cis-(118) and trans-(118) increased to 16 % and 27 %, respectively, and the starting material decreased to 57 %. Methanolysis of BPS. f.AL3lp NMR monitoring of reaction. Methanolic NaOMe solution (0.05 ml, 0.5 M) was added to a solution of BPS (5.4 mg) in CH3OH (0.4 ml) and after 10 hours the trans-isomer (8 65.1, 18 %) was observed in addition to starting material (8 59.9, 82 %). The acyclic phosphorothionate (8 73.7, 4 %) appeared after 35 hours and the trans-isomer increased (39 %). The starting material decreased to 58 %. The trans-isomer reached a maximum (55 %) after 87 hours and the cis-isomer also now appeared (8 66.2, 3 %). The final mixture (201 216 hours) consistd of the acyclic compound (50 % ), the cis-isomer (11 %), the trans-isomer (37 % ) and unreacted BPS (2 % ). (B) Isolation of trans-5-hydroxymethyl-5-methyl-2-thiono-r-2-methoxy-1,3.2- dioxaphosphorinane (203). B-1: A methanolic solution of NaOMe (2 ml, 1.0 M) was added to a stirred solution of BPS (100.0 mg) in CH3OH (9 ml). After being stirred for 5 hours, the same procedure described for BPO was followed with separation by preparative tic (1 mm, CHCl3 - acetone (5 : 1) x 1, Rf= 0.52) to give trans-5-hydroxymethyl-5-methyl-2- thiono-r-2-methoxy-1,3,2-dioxaphosphorinane (203) as white crystals (45.5 mg). Recrystallization from petroleum ether/acetone mixture gave colourless crystals which were submitted for X-ray crystallography. m.p.: 87 - 88 ·c Analysis (C6li13O4PS): Found(%), C: 33.93 H: 6.42 Calculated(%), C: 3l96 H: 6.13 MS, m/z (%): 253 (M + 41, 5), 241 (M + 29, 12), 213 (M + 1, 100). 3lp NMR (CDCl3): 6 63.1 1H NMR (CDCl3): 6 4.16 (m, 4H, CfuOP), 3.82 (d, J = 4.52 Hz, 2H, CfuOH), 3.81 (d, J = 13.63 Hz, 3H, CH3OP), 1.79 (t, J = 5.25 Hz, lH, CH2O!:!), 0.88 (s, 3H, Cfu) 13c NMR (CDCl3): 6 72.9 (d, J = 8.6 Hz, CH2OP), 63.4 (s, CH2OH), 54.3 (d, J = 4.8Hz, CH3OP), 36.9 (d, J = 6. lHz, C4), 15.9 (s, CH3) 217 lR (KBr): 3566 (s), 3513 (s), 2958 (s), 2886 (m), 1473 (s), 1463 (s), 1388 (m), 1245 (m), 1205 (m), 1182 (m), 1126 (s), 1065 (s), 1036 (s), 989 (s), 962 (s), 922 (s), 832 (s), 654 (s), 449 (s), 424 (s). B-2: In a similar fashion, aqueous NaOH (2 ml, 1.5 M) was added to a stirred solution of BPS (103.0 mg) in CH3OH (15 ml). Work up as described for the previous reaction (B-1) again gave trans-5-hydroxymeth y1-5-meth y1-2-thiono-r-2-methoxy- l ,3 ,2- dioxaphosphorinane ( 51.1 mg). {C) Isolation of acyclic phosphorothionate. The same procedure as for B-1 was followed using methanolic NaOMe solution (5 ml, 0.5 M) and BPS (103.3 mg) in CH3OH (10 ml). The crude product was obtained as a colourless oil (148.3 mg) which was purified by preparative tlc [1 mm, CHCl3 - acetone (5:1) x 1] (Rf= 0.30, 62.0 mg). Analysis (C7H17O5PS): Found(%), C, 34.09 H: 7.28 Calculated(%), C, 34.4 H:6.97 MS, rn/z (%): 285 (M+41, 3), 273 (M+29, 1), 245 (M+l, 62), 143 (100), 3lp NMR (CDCl3): 8 73.2 lH NMR (CDCl3): 8 4.14 (d, J = 8.50 Hz, 2H, CfuOP), 3.78 (d, J = 13.52Hz, 6H, CfuOP), 3.62 (s, 4H, CfuOH), 0.85 (s, 3H, Cfu) 13C NMR (CDCl3): 8 70.1 (d, J = 5.4 Hz, CH2OP), 67 .2 (s, CH2OH), 54.9 (d, J = 5.4 Hz, CH3OP), 41.2 (d, J = 7.0 Hz, C4), 16.4 (s, CH3) 218 IR (Neat): 3400 (s,br), 2960 (s), 2890 (s), 1470 (s), 1250 (m), 1185 (s), 1030 (s, br), 870 (s), 830 (s, br) Ethanolysis of BPO. A ethanolic solution of NaOEt (0.05 ml, 0.5 M) was added to a solution of BPO (12.6 mg) in EtOH (0.4 ml). In 10 mins the trans- isomer (6 -5.1, 10 %) appeared and the starting material (6 -4.5, 90 %) decreased. After 1 hour, a trace of the acyclic compound (6 1.2) was observed, and it grew to 26 % after 5.5 hours while the starting material disappeared. The trans- isomer decreased (6 -5.1, 70 % ) from its maximum 80 % after 4 hours. The cis- isomer (6 -4.8, 2 % ) was also observed as well as other small peaks [6 3.7 (trace), 0.9 (trace) and -1.0 (2 %)]. In the final mixture (222 hours), the acyclic phosphorothionate, the cis- isomer and the trans- isomer comprised 47 %, 16 % and 22 % of the reaction mixture. The remainder was accounted for by from small peaks at 6 3.7 (6 %), 0.9 (6 %) and -1.0 (3 %). REFERENCES 219 1. Worthing, C.R., Pesticide Manual: World compendium, 8th Ed., British Crop Protection Council, Lavenham Press, Lavenham, Great Britain. 1987 2. Eto, M., Organophosphorus Pesticides: Organic and Biological chemistry, CRC Press, 1974. 3. Knack, J.B., Stahmann, M.A. and Casida, J.E., J. Agric. Food Chem., 10, 154, 1962. 4. Ptashne, K.A. and Neal, R.A., Biochemistry., 11, 3224, 1972. 5. Honer, L. and Winkler, H., Tetrahedron Lett., 175, 1964. 6. Stec, W.J., Okruszek, A. and Mikolajczyk, M., Z. Naturforsch, 26b, 855, 1971. 7. Mikolajzyk, M., Luczak, J., Chem. Ind. , 701, 1974. 8. Mikolajzyk, M., Luczak, J., J. Org. Chem., 43, 2132, 1978. 9. Skowronska, A., Bull. Acad. Pol. Sci., Ser. Sci. Chim., 21, 459, 1973. 10. Michalski, J., Okuszek, A., and Stec, W., J. Chem. Soc., Chem.Commun., 1495, 1970. 11. De'ath, N.J., Ellies, K., Smith, D.J.H. and Trippett, S., Chem. Commun., 714, 1971. 12. Stec, W.J., Okruszek, A. and Michaski, J., J. Org. Chem., 41(2), 233, 1976,. 13. Horner, L., Pure. Appl. Chem., 9, 225, 1964. 14. McBain, J.B., Yamamoto, I. and Casida, J.E., Life Sci., 10(11), 1311-1319, 1971B. 220 15. Wustner, D.A., Desmarchelier, J. and Fukuto, T.R., Life Sci., 11(11), 583, 1972. 16. Herriott, A.W., J. Am. Chem. Soc., 93, 3304, 1971. 17. Bellet, E.M. and Casida, J.E., J .. Agric. Food Chem., 22(2), 207-211, 1974. 18. Fahmy, M.A.H. and Fukuto, T.R., Tetrahedron Lett., 41, 4245, 1972. 19. Allahyari, R., Lee, P.W., Lin, G.H.Y., Wing, R.M. and Fukuto, T.R., J. Agric. Food Chem., 25, 471,1977. 20. Miyamoto, T. and Yamamoto, I., Nippon Noyakai Gakkaishi, 2, 303 - 310, 1977. 21. Eto, M., Okabe, S., Ozoe, Y. and Maekawa, K., Pest. Biochem. Physiol., 7, 367, 1977. 22. Segall, Y. and Casida, J.E., Phosphorus Chemistry : ACS Symposium Series 171, 337-340, 1981. 23. Segall, Y. and Casida, J.E., Tetrahedron Lett., 23, 139-142, 1982. 24. Segall, Y. and Casida, J.E., Phosphorus and Su/fur, 18, 209-212, 1983. 25. Thompson, C.M, Castellino, S. and Fukuto, T.R., J. Org. Chem., 49, 1696, 1984. 26. Bielawski, J. and Casida, J.E., J. Agric. Food Chem., 36(3), 610, 1988. 27. Swinson, J., Ph D thesis, Vanderbilt University (1985). 28. Swinson, J., Field, L., Heimer, N.E., Michalska, D., Muccio, D.D., Polavarapu, P.L., Stone, M.P. and Van Wazer, J.R., Phosphorus and Su/fur, 35, 159-172, 1988. 221 29. Toia, R.F. and Casida, J.E., Toxical. Appl. Pharm., 81, 50-57, 1985. 30. Toia, R.F. and Casida, J.E., Bit>phosphates and their Analogues-Synthesis, Structure, Metabolism and Activity in Procedings of 2nd International Symposium on Phosphorus Chemistry directed Towards Biology, Elsevier, Amsterdam, 465, 1987. 31. Abdullah, A. R., Ph.D. thesis, University of New South Wales, 1986. 32. Segall, Y., Wu, S-Y., Toia, R.F. and Casida, J.E., Tetrahedron Lett., 31(4), 473-476, 1990. 33. Yang, Y-C., Szafraniec, L.L., Beaudry, W.T. and Rohrbaugh, D.K., J. Am. Chem. Soc., 112, 6621-6627, 1990. 34. Field, L., Heimer, N.E., McNeil, R.I., Neal, R.A., Swinson, J., and Van Wazer, J.R., Su/fur letters, 1, 135-141, 1983. 35. Wu, S-Y., Segall,Y., Sanders, M., Toia, R.F. and Casida, J.E., phosphorus, Sulfur and Silicon, 54, 221-224, 1990. 36. Neal, R.A., Rev. Biochem. Toxicol., 2, 131-171, 1980. 37. Ptashne, K.A., Walcott, R.M. and Neal, R.A., J. Pharmacol. Expt. Ther., 179, 380, 1971. 38. Jerina, D.M., Daly, J.W. and Witkop, B., Biochemistry, 10, 366, 1971. 39. Jerina, D.M., Daly, J.W., Witkop, B., Zaltman-Nirenberg, P. and Udenfriend, S., Biochemistry , 9, 147, 1970. 40. McBain, J.B., Yamamoto, I. and Casida, J.E., Life Sci., 10(11), 947, 1971. 41. Ohkawa, H., Mikami, N. and Miyamoto, J., Agri. Biol. Chem., 40(10), 2125, 1976. 222 42. Lee, P.W., Allahyari, R. and Fukuto, T.R., Biochem. Pharmacol ., 25, 2671, 1976. 43. Fukuto, T.R., Pure & Appl. Chem., 50, 1015-1032, 1978. 44. Frehse, H., Pesticide Terminal Residue, Buttersworth, London, 9, 1971. 45. Metcalf, R.L., Enzymic oxidation of toxicants, 151, 1968. 46. Leesch, J.G. and Fukuto, T.R., Pest. Biochem. Physiol., 2,223, 1972. 47. Bowman, J.S., and Casida, J.E., J. Agric. Food Chem., 5,192,1957. 48. Schrader, G., Weinheim Verlag Chemie., West Germany, 393, 1963. 49. Fukuto, T.R., Metcalf, R.L., March, R.B. and Maxon, M.G., J. Econ. Entomol., 48, 347, 1955. 50. Patchett, G.G. and Batchelder, G.H., J. Agric. Food Chem., 8, 54, 1960. 51. Wieland, T. and Lambert, R., Chem. Ber., 89, 2476, 1956. 52. Quistad, G.B., Fukuto, T.R. and Metcalf, R.L., J. Agri. Food Chem., 18, 189, 1970. 53. Robinson, C.P. and Beiergrohslein, D., Pest. Biochem. Physiol., 13, 267, 1980. 54. Wing, K.D., Glickman, A.H. and Casida, J.E., Science, 219, 63, 1983. 55. Wing, K.D., Glickman, A.H. and Casida, J.E., Pest. Biochem. Physiol., 21, 22-30, 1984. 56. Bellet, E.M. and Casida, J.E., Science 182, 1135-1136, 1973. 223 57. Casida, J.E., Eto, M., Moscioni, A.D., Engel, J .L., Milbrath, D.S. and Verkade, J.G., Toxicol. Appl. Pharmacol ., 36, 261-279, 1976. 58. Bowery, N.G., Collins, J.F. and Hill, R.G., Nature (London) , 261, 601- 603, 1976. 59. Korenaga, S., Ito, Y., Ozoe, Y. and Eto, M., Comp. Biochem. Physiol., 57C, 95-100, 1977. 60. Casida, J.E., Chem. Eng. News, 52 (1), 56 , 1974. 61. Petajan, J.H., Voorhees, K.J., Packhan, S.C., Baldwin, R.C., Einhorn, I.N., Grunnet, M.L., Dinger, B.G. and Birky, M.M. Science (Washington D.C.), 187, 742-744, 1975. 62. Kimmerle, G., Eben, A., Groning, P. and Thyssen, J., Arch. Toxicol., 35, 149-152, 1976. 63. Eto, M., Ozoe, Y., Fujita, T. and Casida, J.E., Agric. Biol. Chem., 46(10), 2113-2115, 1976 .. 64. Dabkowski, W., Michalski, J., Radziejewski, C. and Skrzypczynski, Z., Chem. Ber., 1636, 1982. 65. Bowery, N.G., Collins, J.F., Hill, R.G. and Pearson, S., Brit. J. Pharmacol ., 60, 275-276,1977. 66. Milbrath, D.S., Engel, J.L., Verkade, J.G. and Casida, J.E., Toxicol. Appl. Pharmacol., 47, 287-293, 1979. 67. Casida, J.E. and Lawrence, L.J., Environ. Health Perspect ., 61, 123-132, 1985. 224 68. O'Brien, R.D., Chapter 2, Toxic phosphorus esters, Academic Press, London, 1960. 69. O'Brien, R.D., Insecticides Action and Metabolism, Academic Press, New york, 1967. 70. Ruzicka, J.H., Thomson, J. and Wheals, B.B., J. Chromatogr., 31, 37, 1967. 71. Heath, D.F., Organophosphosphorus Poisins (Anticholinesterases and Related Compounds), Pergamon, Oxford, 1961. 72. Kirby, A.T. and Warren, S.G., The Organic Chemistry of Phosphorus, Elsevier, Amsterdam, 1967. 73. Hudson, R.F. and Keay, L., J. Chem. Soc., 3269, 1956. 74. Bencovic,S.J., Hydrolytic reations of inorganic esters in Comprehensive Chemical Kinetics, Vol.10, Bamford, C.H. and Tipper, C.F.H .. , Eds. Elsevier, Amsterdam, 1, 1972. 75. Murdock, L.L. and Hopkins, T.L., J. Agric. Food Chem., 16, 954, 1968. 76. Fukuto, T.R. and Stafford, E.M., J. Am. Chem. Soc., 79, 6083, 1957. 77. Miller, B., Proc. Chem. Soc., 303, 1962. 78. Boyd, D.B., J. Am. Chem. Soc., 91, 1200, 1969. 79. Blackburn, G. M., Cohen, J.S. and Todd, L., Tetrahedron Lett., 2873, 1964. 80. Eberhand, A. and Westheimer, F.H., J. Am. Chem. Soc., 87, 253, 1965. 81. Kumamoto, J., Cox, Jr., J.R. and •Westhemier, F.H., J. Am. Chem. Soc., 78, 4858, 1956 82. Haake, P.C. and Westheimer, F.H., J. Am. Chem. Soc., 83, 1102, 1961. 225 83. Covitz, F. and Westheimer, F.H., J. Am. Chem. Soc., 85, 1773, 1963. 84. Eto, M., Hanada, K., Namazu, Y. and Oshima, Y., Agric. Biol. Chem. ( Tokyo), 27, 723, 1963. 85. Dennis, E.A. and Westheimer, F.H., J. Am. Chem. Soc., 88, 3432, 1966. 86. Fukuto, T.R. and Metcalf, R.L., J. Med. Chem., 8, 759, 1965. 87. Aksnes, G. and Bergesen, K., Acta. Chem. Scand., 20, 2508,1966. 88. Gorenstein, D.G. and Taira, K., J. Am. Chem. Soc., 104, 6130, 1982. 89. Taira, K., Lai, K. and Gorenstein, D.G., Tetrahedron, 42, 229, 1986. 90. Taira, K., Fanni, T. and Gorenstein, D.G., J. Am. Chem., 106, 1521, 1984. 91. Taira, K., Fanni, T. and Gorenstein, D.G., J. Org. Chem., 49, 4531, 1984. 92. Taira, K., Mock, W.L. and Gorenstein, D.G., J. Am. Chem., 106, 7831, 1984. 93. Yang, J.C. and Orenstein, D.G., Tetrahedron Lett., 25, 4627, 1984. 94. Taira, K. and Gorenstein, D.G., Tetrahedron, 40, 3215, 1984. 95. Fanni, T., Taira, K., Gorenstein, D.G., Vaidyanathaswamy, R. and Verkade, J.G., J. Am. Chem. Soc. , 108, 6311-6314, 1986. 96. Gorenstein, D.G.,Findlay, J.B., Luxon, B.A. and Kar, D., J. Am. Chem. Soc., 99, 3473, 1977. 97. Gorenstein, D.G., Luxon, B.A., Findley, J.B. and Momii, R., J. Am. Chem. Soc., 99, 4170, 1977. 226 98. Gorenstein, D.G., Luxon, B.A., and Findley, J.B., J. Am. Chem. Soc., 99, 8048, 1977 99. Ozoe, Y., Mochida, K. and Eto,, M., Agric. Biol. Chem., 46(2), 555-556, 1982. 100. Verkade, J.G., Coordinat. Chem . Rev., 9, 1, 1972. 101.. Bauman, M. and Jr. Wadsworth, W.S., J. Am. Chem. Soc., 100, 6388, 1978. 102. Buchwald, S.L., Friedman, J.M. and Knowles, J.R., J. Am. Chem. Soc. , 106, 4911, 1984. 103. Duff, E. and Trippett, S., Phosphorus, 1, 291, 1972. 104. Edmundson, R.S., J. Chem. Soc. Perkin I, 1660, 1972. 105. Friedman, J.M. and Knowles, J.R., J. Am. Chem. Soc., 107, 6126-6127, 1985. 106. Wadsworth Jr., W.S., J. Org. Chem., 52, 1748-1753, 1987. 107. Wadsworth Jr., W.S., J. Org. Chem., 32, 1603 , 1967. 108. Wadsworth Jr., W.S. and Horton, H., J. Am. Chem. Soc., 92, 3785 , 1970. 109. Wadsworth Jr., W.S., Larsen, S. and Horton, H.L., J. Org. Chem., 38, 256, 1973. 110. Wadsworth Jr., W.S., J. Org. Chem., 38, 2921, 1972. 111. Wadsworth Jr., W.S. and Tsay, Y.G., J. Org. Chem., 39, 984, 1974. 112. Wadsworth, W.G. and Wadsworth Jr., W.S., J. Am. Chem., 105, 1631, 1983. 227 113. Wadsworth Jr., W.S., Ind. Eng. Chem. Prod. Res. Dev., 23, 625, 1984. 114. Wadsworth Jr., W.S. and Wilde, R.L., J. Org. Chem., 41(7), 1264, 1976. 115 Warrent, R.W., Caughlan, C.N., Hargis, J.H., Yee K.C. and Bentrude, W.G., J. Org. Chem., 43(22), 4266, 1978. 116. Gehrke, S.H. and Wadsworth Jr., W.S., J. Org. Chem., 45, 3921, 1980. 117. Wacher, V.J., honour thesis, University of New South Wales. 118. Gorenstein, D.G., Phosphorus 31 NMR-Principles and applications, Academic Press, New York, 1984. 119. White, D.W., McEwen, G.K., Bertrand, R.D. and Verkade, J.G., J. Chem. Soc., B, 1454, 1971. I 120. Geise, H.J., Ree. Trav. Chim. Pays-Bas, 86, 362, 1967. 121. Maryanoff, B.E., Hutchins, R.O. and Maryanoff, C.A., Top. Stereochem. 11, 187, 1979. 122. Jackman, L.M. and Sternhell, S., Application of N.M.R. spectroscopy in Organic Chemistry (Second eddition), Pergammon Press, New York ,1969. 123. Bartle, K.D., Edmundson, R.S. and Jones, D.W., Tetrahedron, 23, 1701- 1711, 1967. 124. Hall, L.D. and Malcolm, R.B, Can. J. Chem., 50, 2092, 1972. 125. Toia, R.F., Unpublished results. 126. Craig, D. and Toia, R.F., Unpublished results. 127. Ovchinnikov, V.V., Gulkin, V.I., Yarkova, E.G., Markova, L.E., Cherkasov, R.A. and Pudovik, A.N., Zh. Obshch. Khim., 48, 2424, 1978. 228 128. Gagnaire, D. and Robert, J.B., Bull. Soc. Chim. Fr., 2240, 1967. 129. Denney, D.E. and Denny, D.B., J. Am. Chem. Soc., 88, 1830. 1966. 130. Borecka, B., Jezierska, B., Bluj, S. and Michalski, J., Bulletin de L'academie polonaikse des sciences. Serie des sciences chimiques, XXII(3), 201-205, 1974. 131. Swinson, J., Field, L. and (in part) Heimer, N.E., Stone, M.P. and Van Wazer, J.R., Phosphorus, Su/fur and Silicon, 44, 39-52, 1989. 132. Amono-Neizer, E.H., Ray, S.K.,Shaw, R.A., and Smith, B.C., J., Chem. Soc., 4296, 1965. 133. Cardogan, J.I.G., Quart. Rev. (London), 16, 223-233, 1962. 134. Michalski, J., Pliszka-Krawiecka, A. and Skowronska, Roczniki Chemii. Ann. Soc. Chim. Polonorum., 37, 1479, 1963. 135. Krawczyk, E. and Skowronska, A., Phosphorus and Su/fur, 9, 189 ,1980. 136. David, S., Eisenstein, 0., Hehre, W.J., Salem, L. and Hoffman, R., J. Am. Chem. Soc., 95, 3806-3807, 1973. 137. Romers, C., Altona, C., Buys, H.R. and Havinga, E., Top. Stereochem., 4, I 39-97, 1969. 138. Hudson, R.F. and Verkade, J.G., Tetrahedron Lett, 3211-3234, 1975. 139 Jaffery, G.A., Pople, J.A., Binkeley, J.S. and Vishveshwara, S., J. Am. Chem. Soc., 100, 378-379, 1978. 140. William, J.O., Scarsdale, J.N., Schafer, L. and Geise, H.J., J. Mol. Struct., 76, 11, 1981. 229 141. L. SILVER and R. RUDMAN, Acra Cryst., B28, 574, 1972. 142. Z. GALDECKI, Acta Cryst., B36, 2807-2809, 1980. 143. Beineke, T. A., Acta Cryst., B25, 413, 1969. 144. Beineke, T. A., Chem. Comm., 860, 1966. 145. Edmundson, R.S. and Mitchell, E.W., J. Chem. Soc. C., 3033, 1968. 146 Clardy, J.C., Mosho, J.A. and Verkade, J.G., J. Chem. Soc. Chem. Comm., 1163, 1972. 147. Edmundson, R.S. and Mitchell, E.W., Chem. Commun., 482 ,1966. 148. Edmundson, R.S. and Mitchell, E.W., J. Chem. Soc. C., 3179, 1971. 149. Berlin, K.D., Hilderbrand, C., South, A., Hellwege, D.M., Peter~on, M., Pier, E.A. and Verkade, J.G., Tetrahedron, 20, 323, 1964. 150. Berlin, K.D., Hilderbrand, C.,Verkade, J.G., and Denner, O.C., Chem. Ind., 291, 1963. 151. Wadsworth Jr., W.S. and Emmons, W.D., J. Am. Chem. Soc., 84, 610 , 1962. 152. Edmundson, R.S. and Mitchell, E.W., J. Chem. Soc., C., 752, 1970. 153. Verkade, J.G., Huttennann, T.J., Fung, M.K. and King, R.W., Jnorg. Chem., 4, 83, 1965. 154. Riddle, F.G., The conformational analysis of Heterocyclic Compounds, Academic Press, 122. 155. Khaikin, L.S. and Vilkov, L., V., Russ. Chem. Rev., 41(12), 1060, 1972. 230 156. Corbridge, D.E.C., The structural Chemistry of Phosphorus., Elsevier Scientific Pub. Co., 407-409, 1974. 157. Gallagher, M.J., 3lp NMR Spectroscopy in Stereochemical Analysis, VCH Publishers, Inc. 297, 1987. 158. Hall, L.D. and Malcolm, R.B., Can. J. Chem., 50, 2103, 1972. 159. Majoral, M.P. and Navech, J., Bull. Soc. Chim. France, 95, 1331, 2609, 1971. 160. Mosob, J.A. and Verkade, J.G., J. Org. Chem., 42, 1549, 1977. 161. Mosob, J., A., Org. Mag. Res., 11, 281, 1978. 162. Yee, K.C. and Bertrude, W.G., Tetrahedron Lett., 2775, 1971. 163. Bertrude, W.G. and Yee, K.C., Chem. Comm., 169, 1972. 164. Bertrude, W.G., Tan, H-W, and Yee, K.C., J. Am. Chem. Soc., 94, 3264, 1972. 165. Dale, A.J., Acta Chem. Scand., 26, 2985, 1972. 166. Dutasta, J.P., Grand, A., Robert, J.B. and Taieb, M., Tetrahedron Lett., 2659, 1974. 167. Kok, A.M.G., Wynberg H., Garcia-Granda, S., Beurskens, P.T. and Smits, J.M.M., Acta Cryst, C44, 2235-2237, 1988. 168. Garcia-Granda, S., Beurskens, P.T., Smits, J.M.M., Kok, A.M.G.and Wynberg H., Acta Cryst., C44, 2233-2235, 1988. 169. Kok, A.M.G, Wynberg H., Parthasarathi, V., Smits, J.M.M., and Beurskens, P.T., Acta Cryst., C43, 1336-1341, 1987. 231 170. Smits, J.M.M., Beurskens, P.T., Parthasarathi, V., Rijk, E.V.A., Kok, A.M.G. and Wynberg, H., Acta Cryst, C43, 1334-1336, 1987. 171. Smits, J.M.M., Beurskens, Kok, A.M.G. and Wynberg, H.Acta Cryst., C43, 1331-1333, 1987. 172. Kok, A.M.G, Wynberg H., Smits, J.M.M., Beurskens, P.T.and Parthasarathi, V., Acta Cryst., C43, 1328-1331, 1987. 173. Jones, P.G., Sheldrick, G.M., Kirby, A.J. and Briggs, A.J., Acta Cryst., I C40, 1061-1065, 1987. 174. Edmunson, R.S and (the late) King, T.J., J. Chem. Soc. Perkin Trans. I, 1943, 1984. 175. Day, R.O., Bentrude, W.G., Yee, K.C., Setzer, W N., Deiters, J.A. and Holmes, R.R., J. Am. Chem. Soc., 106, 103-105, 1984. 176. Day, R.O., Gorenstein, D.G. and Holmes, R.R., Inorg. Chem., 22, 2192- 2195, 1983. 177. Nuffel, P.V., Lenstra, A.T.H. and Geise, H.J., Cryst. Struct. Comm., 9, 733, 1980. 178. Cameron, T.S. and Karolak-Wojciechowska, J.,Acta. Cryst., B33, 2342- 2344, 1977. 179. Cameron, T.S., Galdecki, Z. and Karolak-Wojciechowska, J., Acta. Cryst., B32, 492, 1976. 180. Wagner, R.E, Jensen, W., Wadsworth, W. and Johnson, Q., Acta Cryst. B29, 2160, 1973. 232 181. Wagner, R.E, Jensen, W., Wadsworth, W. and Johnson, Q., Cryst. Struct. Comm., 3, 507, 1973. 182. Wagner, R.E, Jensen, W., Wadsworth, W. and Johnson, Q., Cryst. Struct. Comm., 2, 327, 1973. 183. Saenge, W. and Mikolajczyk, M., Chem. Ber., 106, 3519-3523, 1973. 184. Drew, M.G.B., and Rodgers, J., Acta Cryst., B28, 924, 1972. 185. Drew, M.G.B., Rodgers, White, D.W. and Verkade, J G., Chem. Comm., 227, 1971. 186. Killean, R.C.G., Lawrence, J.L. and Magennis, I.M., Acta Cryst., B27, 189, 1971. 187. Rodgers, White, D.W. and Verkade, JG., J. Chem. Soc. (A), 77, 1971. 188. Ul-Haque, M., Caughlan, C.N. 'and Moats, W.L., J. Org. Chem., 35(5), 1466, 1970. 189. Ul-Haque, M., Caughlan, C.N., Hargis, J. H. and Bentrude, W.G., J. Chem. Soc.(A), 1786, 1970. 190. Murayama, W. and Kainosho, M., Bull. Chem. Soc. lap., 42(7), 1819-1824, 1969. 191. Geise, H.J.,Rec. Trav. Chim. Pays-Bas., 86, 362, 1967. 192. Bukowska-Strzyzewska, M., Michalski, J., Mlotkowska, B. and Skoweranda, J., Acta Cryst., B32, 2605, 1976. 193. Cook, D.S. and White, R.F M., J. Chem. Soc. Dalton Trans, 2212, 1976. 233 194. Bartczak, T.J., Wolf, W., Swepston, P.N. and Zerong, L.L., Acta Cryst., C43, 1788-1790, 1987. 195. Bartczak, T.J., Galdecki, Z. and Wolf, W., Acta Cryst., C42, 244-246, 1986. 196. Bartczak, T.J., Galdecki, Trzezwinska, H.B. and Wolf, W, Acta Cryst., C39, 731-731, 1983. 197. Bartczak, T.J., and Wolf, W., Acta Cryst., C39, 224-227, 1983. 198. Kinas, R., Stec, W.J. and Kruger, C., Phosphorus and Sulfur, 4, 295-298, 1978. 199. Grand, A., Martin, J., Robert, J.B. and Tordjman, I., Acta Cryst., B31, 2523, 1975. 200. Bartczak, T.J., Christensen, A., Kinas, R. and Stec, W.J., Cryst. Struct . Comm., 4, 701, 1975. 201. Bartczak, T.J., Christensen, A., Kinas, R. and Stec, W.J., Cryst. Struct. Comm, S, 21, 1976 202. Fanni, T., Taira, K., Gorenstein, D.G., Vaidyanathaswamy, R. and Verkade, J.G., J. Am. Chem. Soc., 108, 6311-6314, 1986. 203. Bartczak, T.J., Wolf, W. and Cameron, T.S., Acta Cryst., C39, 1467-1469, 1983. 204. Miller, A., Wieczorek, M.W., Karolak-Wojciechowska, J., Mikolazczyk, M. and Ziemnicka, B., Acta Cryst., B37, 1951-1953, 1981. 205. Nuffel, P.V., Lenstra, A.T.H. and Geise, H.J., Acta Cryst, B37, 133-136, 1981. 206. Grand, P.A. and Robert, J B., Acta Cryst., B31, 2502, 1975. 234 207. Dutasta, J.P., Grand, A. and Rohen, J.B., Tetrahedron Lett., 2655-2658, 1974. 208. Edmundson, R.S., Johnson, 0., Jones, D.W. and King, J.J., J. Chem. Soc. Perkin Trans., 2, 69, 1985. 209. Edmundson, R.S., Johnson, 0. and Jones, D.W., Phosphorus, su[fur and silicon, 46, 61-67, 1989. 210. Jaffery,, G.A., Pople, J.A. and Radom, L., Carbohyd. Res., 25, 117-131, 1972. 211. Radom, L., Altona, C. and Pople, W., J., J. Am. Chem. Soc., 94, 2371-2381, 1972. 212. Van de Mieroop, W. and Lenstra, A.T.H., Cryst. Struct. Commu., 7, 577-582, 1978. 213. Petit, G.H., Lenstra, A.T.H. and Geise, H.J., Bull. Soc. Chim. Belg., 87, 659-666, 1978. 214. Chang, W.H., J. Org. Chem., 29, 3711-3712, 1964. 215. Filler, R. and Novar, H. J.Org.Chem. 25, 663, 1960. 216. Edmundson, R.S., Tetrahedron, 20, 2781, 1964. 217. Majoral, J.P., Pujol, R. and Navech, J., Bull. Soc. Chim. Fr., 606-610, 1972. 218. Edmundson, R.S., Tetrahedron, 21, 2379-2387, 1965. 219. Stec, W.. and Zwierzak, A., Can.I. Chem., 45, 2513, 1967. 220. Ramirez, F.,Tsuboi, H., Okazaki, H. and Marecek, J.F., Tetrahedron Lett., 23, 5375-5376, 1982. 235 221. McConnell, R.L. and Coover Jr., H.W., J. Org. Chem. 24, 630, 1959. 222. Stec, W.J., Moddeman, W.E., Albridge, R.G. and Van Wazer, J.R., J. Phys. Chem., 75, 3975, 1971. 223. Zwierzak, A., Tetrahedron, 25, 5177, 1969. 224. Cadogan, J.I.G. and Moulden, H.N., J. Chem. Soc., 5524, 1961. 225 Cadogan, J.I.G., J. Chem. Soc., 3067, 1961. 226. McCombie, H., Saunders, B.C. and Stacey, G.J., J. Chem. Soc., 380, 1945. 227. Pesin, V.G. and Khaletskii, A.M., Zh. Obshch. Khim., 31, 2508, 1961. 228. Kudelska, W. and Michalska, M., Tetrahedron, 42, 629, 1986. 229. Knotz, F., Osterr. Chem-Ety, 50, 128-129, 1949. 230. Saunders, B.C., Stacey, G.J., Wild, F., and Wilding, I.G.E., J. Chem. Soc., 699, 1948. 231. Mcivor, R.A., McCarthy, G.D. and Grant, G.A., Can. J. Chem., 34, 1819, 1951. 232. Sosnovsky, G. and Zaret, E.H., J. Org. Chem., 34(4), 968, 1969. APPENDICES* * Appendices contain X-ray data of appropriate compounds. 236 Appendix A X-Ray data of trans-5-acetoxymethyl-5-methyl-2-thiono-2-ethoxy- 1,3,2-dioxaphosphorinane (119). Cell: Monoclinic, a= 10.618, b = 6.675, c = 21.025A, b = 120.19°, space group P21/c, z=4. Table 1. Non-hydrogen atomic parameters. Esd in parentheses. [Beq(A2) is the isotropic equivalent of the anisotropic temperature factor.] X y z Beq s 0.2148(1) -0.2602(1) 0.1805(0) 5.09(4) p 0.2364(1) 0.0096(1) 0.1575(0) 3.12(3) 0(1) 0.3815(2) 0.0393(3) 0.1551(1) 3.69(7) 0(2) 0.1104(2) 0.0847(3) 0.0810(1) 3.57(6) 0(3) 0.2388(2) 0.1770(3) 0.2108(1) 3.59(7) 0(4) 0.1863(2) 0.5253(4) -0.0451(1) 4.72(8) I 0(5) 0.3467(2) 0.6598(5) -0.0714(1) 6.07(11) C(l) 0.3990(3) 0.2316(4) 0.1274(2) 3.72(10) C(2) 0.2725(3) 0.2679(4) 0.0488(2) 3.39(9) C(3) 0.1298(3) 0.2701(4) 0.0499(1) 3.39(9) C(4) 0.3490(3) 0.1725(5) 0.2888(2) 4.13(10) C(5) 0.3081(4) 0.3248(6) 0.3273(2) 5.25(14) C(6) 0.2686(4) 0.1084(5) -0.0041(2) 4.97(14) C(7) 0.3020(3) 0.4763(5) 0.0287(2) 4.41(11) C(8) 0.2243(3) 0.6169(4) -0.0897(2) 3.86(11) 237 C(9) 0.0961(4) 0.6588(7) -0.1622(2) 5.93(15) Table 2. Hydrogen atom positional parameters. (Thermal parameters equal to those of bonded atom.) X y z HlC(l) 0.4941 0.2337 0.1285 H2C(l) 0.3997 0.3418 0.1607 H1C(3) 0.1307 0.3873 0.0802 H2C(3) 0.0465 0.2876 -0.0019 H1C(4) 0.4476 0.2033 0.2949 H2C(4) 0.3531 0.0346 0.3092 H1C(5) 0.3835 0.3260 0.3818 H2C(5) 0.3058 0.4620 0.3071 H3C(5) 0.2113 0.2933 0.3214 H1C(6) 0.2496 -0.0273 0.0104 H2C(6) 0.1897 0.1381 -0.0557 H3C(6) 0.3647 0.1026 -0.0029 H1C(7) 0.3980 0.4764 0.0301 H2C(7) 0.3054 0.5769 0.0647 H1C(9) 0.1260 0.7244 -0.1959 H2C(9) 0.0459 0.5264 -0.1869 H3C(9) 0.0237 0.7444 -0.1586 Table 3. Bond lengths (A). Esd in parentheses. P-O(1) 1.580(2) O(3)-C(4) 1.459(3) O(1)-C(l) 1.460(3) C(4)-C(5) 1.493(5) 238 C(l)-C(2) 1.537(4) C(2)-C(6) 1.524(4) C(2)-C(3) 1.527(4) C(2)-C(7) 1.531(4) C(3)-O(2) 1.462(3) C(7)-O(4) 1.451(4) O(2)-P 1.570(2) O(4)-C(8) 1.340(3) P-S 1.907(1) C(8)-O(5) 1.190(3) P-O(3) 1.573(2) C(8)-C(9) 1.473(5) Table 4. Bond angles (0 ). Esd in parentheses. S-P-O(1) 111.7(1) C(l)-C(2)-C(6) 111.3(2) O(2)-P-S 114.3(1) C( I )-C(2)-C(7) 104.5(2) S-P-O(3) 116.8(1) C(3)-C(2)-C(6) 110.7(2) O(2)-P-O(1) 105.6(1) C(3)-C(2)-C(7) 109.7(2) O(1)-P-O(3) 105.5(1) C( 6)-C(2)-C(7) 111.6(2) O(2)-P-O(3) 101.7(1) C(2)-C(3)-O(2) 111.3(2) P-O(1)-C(l) 116.5(2) 0(3 )-C( 4)-C(5) 107.9(2) C(3)-O(2)-P 119.3(1) C(2)-C(7)-O( 4) 108.6(2) P-O(3)-C(4) 120.1(2) 0(4 )-C(8)-O(5) 123.2(3) C(7)-O(4)-C(8) 117.4(2) O(4)-C(8)-C(9) 111.4(2) O(1)-C(l)-C(2) 110.5(2) O(5)-C(8)-C(9) 125.4(3) C(l )-C(2)-C(3) 108.8(2) Table 5. Torsional angles (0 ). Esd in parentheses P-O(1)-C(l)-C(2) -58.3(3) O(2)-P-O(3)-C(4) -176.0(2) O(1)-C(l)-C(2)-C(3) 60.5(3) P-O(3)-C(4)-C(5) -169.8(2) 239 C( 1)-C(2)-C(3)-O(2) -56.4(3) C(2)-C(7)-O(4)-C(8) -139.0(3) C(2)-C(3)-O(2)-P 51.5(3) C(7)-O(4 )-C(8)-O(5) -0.4(5) C(3)-O(2)-P-O(1) -43.2(2) C(7)-O( 4 )-C(8)-C(9) -179.5(3) O(2)-P-O(1)-C(l) 46.3(2) 0(1 )..C(l )-C(2)-C(6) -61.8(3) S-P-O(1)-C(l) 171.1(2) 0( 1)-C( 1)-C(2)-C(7) 177.6(2) O(3)-P-O(1)-C(l) -60.9(2) C(6)-C(2)-C(3 )-0(2) 66.2(3) C(3)-O(2)-P-S -166.4(2) C(7)-C(2)-C(3)-O(2) -170.2(2) C(3)-O(2)-P-O(3) 66.7(2) C( 1)-C(2)-C(7)-O( 4) -179.4(2) S-P-O(3)-C(4) 58.8(2) C(3)-C(2)-C(7)-O(4) -62.9(3) O(1)-P-O(3)-C(4) -66.0(2) C(6)-C(2)-C(7)-O( 4) 60.2(3) 240 Appendix B X-Ray data of trans-5-(3,5-Dinitro)benzoyloxymethyl-5-methyl-2- thiono-r-2-ethoxy-l,3,2-dioxaphosphorinane (122). Cell: Monoclinic, a = 15.753, b = 11.868, c = 10.304 A, b = 97.21 °, space group P21/c, z=4. Table 1. Non-hydrogen atomic parameters. Esd in parentheses. [Beq(A2) is the isotropic equivalent of the anisotropic temperature factor X y z Beq s 0.1127(1) 0.0358(1) 0.7958(1) 7.37(3) p 0.1158(0) 0.1937(1) 0.8266(1) 5.10(2) 0(1) 0.1921(1) 0.2309(2) 0.9328(2) 5.35(6) 0(2) 0.1301(1) 0.2657(2) 0.7029(2) 5.06(5) 0(3) 0.0360(2) 0.2493(2) 0.8738(3) 8.58(9) 0(4) 0.3204(1) 0.4142(2) 0.6619(2) 5.25(5) 0(5) 0.4051(2) 0.2730(3) 0.6214(3) 10.67(11) 0(6) 0.2482(2) 0.6976(2) 0.3253(3) 7.30(8) 0(7) 0.2892(2) 0.6735(2) 0.1364(2) 8.14(9) 0(8) 0.5736(2) 0.3165(3) 0.2591(3) 9.61(11) 0(9) 0.5343(2) 0.4292(3) 0.1065(3) 10.45(12) N(l) 0.2905(2) 0.6505(2) 0.2514(3) 5.66(8) N(2) 0.5278(2) 0.3893(3) 0.2099(3) 6.53(9) C(l) 0.2063(2) 0.3512(3) 0.9513(3) 5.80(9) C(2) 0.2216(2) 0.4082(2) 0.8226(3) 4.99(8) 241 C(3) 0.1454(2) 0.3856(2) 0.7199(3) 4.96(8) C(4) -0.0355(3) 0.2036(5) 0.9124(7) 13.16(24) C(5) -0.0820(3) 0.2593(7) 0.9850(6) 14.90(28) C(6) 0.2308(3) 0.5363(3) 0.8443(4) 7.15(11) C(7) 0.3031(2) 0.3586(3) 0.7826(3) 5.32(8) C(8) 0.3729(2) 0.3615(3) 0.5928(3) 5.41(8) C(9) 0.3856(2) 0.4220(2) 0.4700(3) 4.51(7) C(l0) 0.3348(2) 0.5124(2) 0.4238(3) 4.50(7) C(l 1) 0.3475(2) 0.5588(2) 0.3052(3) 4.53(7) C(12) 0.4096(2) 0.5206(2) 0.2319(3) 4.82(7) C(13) 0.4594(2) 0.4325(2) 0.2837(3) 4.89(8) C(14) 0.4490(2) 0.3812(2) 0.3999(3) 4.88(7) Table 2. Hydrogen atom positional parameters. (Thermal parameters equal to those of bonded atom.) X y z HlC(l) 0.2566 0.3654 1.0197 H2C(l) 0.1538 0.3859 0.9846 H1C(3) 0.0928 0.4242 0.7464 H2C(3) 0.1573 0.4207 0.6341 H1C(4) -0.0160 0.1319 0.9641 H2C(4) -0.0737 0.1792 0.8326 H1C(5) -0.1326 0.2182 1.0108 H2C(5) -0.0465 0.2858 1.0720 H3C(5) -0.1041 0.3332 0.9405 H1C(6) 0.1755 0.5687 0.8713 H2C(6) 0.2780 0.5532 0.9157 242 H3C(6) 0.2420 0.5749 0.7621 H1C(7) 0.2953 0.2741 0.7655 H2C(7) 0.3522 0.3687 0.8527 HC(IO) 0.2905 0.5434 0.4761 HC(12) 0.4184 0.5561 0.1459 HC(14) 0.4864 0.3166 0.4366 Table 3. Bond lengths (A). Esd in parentheses. P-O(1) 1.583(2) C(8)-O(5) 1.186(3) O(1)-C(l) 1.454(4) C(8)-C(9) 1.490(4) C(l)-C(2) 1.534(4) C(9)-C(l0) 1.387(4) C(2)-C(3) 1.521(4) C(IO)-C(l 1) 1.378(4) C(3)-O(2) 1.450(3) C(l 1)-C(12) 1.385(4) 0(2)-P 1.573(2) C(l2)-C(13) 1.374(4) P-S 1.900(1) C(13)-C(14) 1.372(4) . P-O(3) 1.551(2) C(14)-C(9) 1.391(4) O(3)-C(4) 1.354(5) C(l 1)-N(l) 1.475(4) C(4)-C(5) 1.292(7) N(l)-O(6) 1.210(3) C(2)-C(6) 1.541(4) N(l)-O(7) 1.213(3) C(2)-C(7) 1.516(4) C(13)-N(2) 1.486(3) C(7)-O(4) 1.464(3) N(2)-O(8) 1.197(4) O(4)-C(8) 1.315(3) N(2)-O(9) 1.182(4) Table 4. Bond angles (0 ). Esd in parentheses. S-P-O(1) 113.1(1) C(3)-C(2)-C(7) 111.3(2) 243 0(2)-P-S 113.8(1) C(6)-C(2)-C(7) 110.8(3) S-P-0(3) 117.8(1) C(2)-C(3)-O(2) 111.2(2) 0(2)-P-0(1) 103.6(1) O(3)-C(4 )-C(5) 121.2(5) 0(1)-P-0(3) 103.9(1) C(2)-C(7)-O( 4) 108.1(2) 0(2)-P-0(3) 103.1(1) O(4)-C(8)-O(5) 124.4(3) P-0( 1)-C( 1) 117.1(2) 0(4 )-C(8)-C(9) 113.0(2) C(3)-0(2)-P 118.2(2) O(5)-C(8)-C(9) 122.6(3) P-0(3)-C(4) 131.2(3) C(8)-C(9)-C(10) 122.1(2) C(7)-O(4 )-C(8) 116.3(2) C( 14 )-C(9)-C(8) 117.0(2) O(6)-N(l)-O(7) 124.6(3) C( 14 )-C(9)-C( 10) 120.8(2) C(l l)-N(l)-O(6) 117.5(3) C(9)-C( 10)-C( 11) 118.2(2) C(l l)-N(l)-O(7) 117.9(3) C(lO)-C(l 1)-N(l) 118.6(2) 0(8)-N(2)-0(9) 123.8(3) N(l)-C(l 1)-C(12) 118.4(2) C(13)-N(2)-O(8) 118.0(3) C(10)-C(l 1)-C(12) 123.0(3) C(l3)-N(2)-O(9) 118.2(3) C(l 1)-C(12)-C(13) 116.4(2) O(1)-C(l)-C(2) 110.9(2) C(12)-C(13)-N(2) 118.6(3) C(l )-C(2)-C(3) 109.2(2) N(2)-C(13)-C(14) 117.9(3) C(l)-C(2)-C(6) 109.4(2) C(12)-C(13)-C(14) 123.5(2) C( 1)-C(2)-C(7) 107.2(2) C(13)-C(14)-C(9) 118.1(3) C(3)-C(2)-C(6) 109.0(3) Table 5. Torsional angles (0 ). Esd in parentheses P-O(1)-C(l)-C(2) -57.3(3) C( 14 )-C( 13 )-N (2)-0(9) 176.2(3) O(1)-C(l)-C(2)-C(3) 57.3(3) O(1)-C(l)-C(2)-C(6) 176.5(3) C(l)-C(2)-C(3)-O(2) -56.3(3) O(1)-C(l)-C(2)-C(7) -63.3(3) C(2)-C(3)-O(2)-P 56.0(3) C(6)-C(2)-C(3)-O(2) -175.7(2) C(3 )-O(2)-P-O( 1) -48.4(2) C(7)-C(2)-C(3 )-0(2) 61.8(3) 244 O(2)-P-O(1)-C(l) 48.9(2) C( 1)-C(2)-C(7)-O( 4) -178.3(2) S-P-O(1)-C(l) 172.6(2) C(3)-C(2)-C(7)-O(4) . 62.4(3) O(3)-P-O(1)-C(l) -58.6(2) C( 6)-C(2)-C(7)-O( 4) -59.1(3) C(3)-O(2)-P-S -171.6(2) 0(4 )-C(8)-C(9)-C( 10) -11.2(4) C(3)-O(2)-P-O(3) 59.6(2) 0(4 )-C(8)-C(9)-C( 14) 171.2(2) S-P-O(3)-C(4) 8.4(5) O(5)-C(8)-C(9)-C( 10) 166.4(3) O(1)-P-O(3)-C(4) -117.5(5) O(5)-C(8)-C(9)-C(14) -11.2(5) O(2)-P-O(3)-C(4) 134.6(5) C(8)-C(9)-C(l0)-C(l l) -176.0(2) P-O(3)-C(4)-C(5) 158.4(5) C(14)-C(9)-C(IO)-C(l 1) 1.6(4) C(2)-C(7)-O(4 )-C(8) -161.0(2) C( 13 )-C( 14 )-C(9)-C(8) 177.3(2) C(7)-O(4 )-C(8)-O(5) 1.1(5) C( 13 )-C( 14 )-C(9)-C( 10) -0.4(4) C(7)-O(4)-C(8)-C(9) 178.6(2) C(9)-C(10)-C(l 1)-N(l) 176.4(2) C(l0)-C(l 1)-N(l)-O(6) 16.2(4) C(9)-C(l0)-C(l l)-C(12) -1.3(4) C(12)-C(l 1)-N(l)-O(6) -166.1(3) N(l)-C(l 1)-C(12)-C(l3) -177.8(2) C(l0)-C(l 1)-N(l)-O(7) -163.9(3) C(10)-C(l 1)-C(12)-C(13) -0.1(4) C(l2)-C(l l)-N(l)-O(7) 13.9(4) C(l 1)-C(12)-C(l3)-N(2) -179.1 (2) C(l 2)-C( 13 )-N (2)-0(8) 176.2(3) C(l l)~C(l2)-C(l3)-C(14) 1.4(4) C(14)-C(l3)-N(2)-O(8) -4.2(4) N (2)-C( 13 )-C( 14 )-C(9) 179.4(2) C(l2)-C(l3)-N(2)-O(9) -3.4(4) C(12)-C( l 3)-C(14)-C(9) -1.1(4) 245 Appendix C X-Ray data of trans-S-chloroymethyl-S-methyl-2-thiono-r-2-ethoxy- 1,3,2-dioxaphosphorinane (195). Cell: Orthorhombic, a= 6.129, b = 11.204, c = 13.191A, space group P212121, Z = 4. Table 1. Non-hydrogen atomic parameters. Esd in parentheses. [Beq(A2) is the isotropic equivalent of the anisotropic temperature factor.] X y z Beq Cl(l) 0.2083(3) 0.1840(1) 0.2017(1) 5.99(4) Cl(2) 0.0231(3) 0.6288(1) -0.0702(1) 5.56(4) p -0.0047(2) 0.3180(1) 0.1814(1) 3.99(3) 0(1) 0.1051(6) 0.4288(3) 0.2304(2) 4.08(8) 0(2) -0.0059(6) 0.3416(3) 0.0656(2) 4.16(8) 0(3) -0.2172(6) 0.2882(4) 0.2207(4) 6.59(12) C(l) 0.3010(8) 0.4806(4) 0.1828(4) 4.50(13) C(2) 0.2685(8) 0.5004(4) 0.0694(3) 3.58(10) C(3) 0.1935(9) 0.3887(4) 0.0192(3) 3.98(11) C(4) 0.4887(9) 0.5382(6) 0.0217(5) 5.86(16) C(5) 0.1008(8) 0.6024(4) 0.0588(4) 3.99(11) Table 2. Hydrogen atom positional parameters. (Thermal parameters equal to those of bonded atom) X y z HlC(l) 0.3381 0.5582 0.2171 H2C(l) 0.4271 0.4239 0.1924, H1C(3) 0.3107 0.3287 0.0238 H2C(3) 0.1612 0.4059 -0.0550 H1C(4) 0.5931 0.4680 0.0289 H2C(4) 0.5507 0.6076 0.0583 H3C(4) 0.4717 0.5565 -0.0516 HlC(5) -0.0318 0.5810 0.1000 H2C(5) 0.1653 0.6780 0.0881 Table 3. Bond lengths (A). Esd in parentheses. P-O(1) 1.552(4) P-Cl(l) 2.007(2) O(1)-C(l) 1.474(6) P-O(3) 1.441 (4) C(l)-C(2) 1.525(7) C(2)-C(4) 1.548(7) C(2)-C(3) 1.490(7) C(2)-C(5) 1.544(6) C(3)-O(2) 1.465(6) C(5)-Cl(2) 1.793(5) O(2)-P 1.551(3) Table 4. Bond angles (0 ). Esd in parentheses. Cl(l)-P-O(1) 105.1(2) C(l)-C(2)-C(3) 110.7(4) O(2)-P-Cl(l) 105.2(2) C(l)-C(2)-C(4) 108.9(4) 247 Cl(l)-P-O(3) 111.5(2) C( 1)-C(2)-C(5) 106.5(4) O(2)-P-O(1) 106.0(2) C(3)-C(2)-C(4) 108.5(4) O(1)-P-O(3) 115.3(3) C(3 )-C(2)-C(5) 112.1(4) O(2)-P-O(3) 112.9(2) C( 4 )-C(2)-C(5) 110.0(4) P-O(1)-C(l) 119.4(3) C(2)-C(3)-O(2) 112.0(3) C(3)-O(2)-P 118.0(3) C(2)-C(5)-Cl(2) 112.6(3) O(1)-C(l)-C(2) 111.6(4) Table 5. Torsional angles (0 ). Esd in parentheses Cl( 1)-P-O( 1)-C(l) 68.6(3) 0( 1)-C( 1)-C(2)-C( 4) -172.3(4) O(2)-P-O(1)-C(l) -42.5(3) 0( 1)-C( 1)-C(2)-C(5) 69.1(5) 0(3 )-P-O( 1)-C( 1) -168.1(3) C( 1)-C(2)-C(3)-O(2) 55.9(5) C(3)-O(2)-P-Cl(l) -66.5(3) C( 4 )-C(2)-C(3 )-0(2) 175.5(4) C(3)-O(2)-P-O( 1) 44.5(4) C(5)-C(2)-C(3)-O(2) -62.9(5) C(3)-O(2)-P-O(3) 171.7(3) C(l )-C(2)-C(5)-Cl(2) -17 4.6(3) P-O(1)-C(l)-C(2) 49.3(5) C(3)-C(2)-C(5)-Cl(2) -53.3(5) C(2)-C(3)-O(2)-P -54.7(5) C( 4 )-C(2)-C(5)-Cl(2) 67.5(5) O(1)-C(l)-C(2)-C(3) -53.0(5) Table 6. Angles (0 ) between some least-squares planes. [O(1)-O(2)-C(3)-C(l)] and [O(1)-P-O(2)]; Al: 37.2 [O(1)-O(2)-C(3)-C(l)] and [C(l)-C(2)-C(3)]; A2: 49.0 248 Appendix D X-Ray data of trans-5-hydroxymethyl-5-methyl-2-oxo-r-2-methoxy-1,3,2 dioxaphosphorinane (118). Cell: Monoclinic, a= 9.491, b = 5.736, c = 19.002A, b = 117.10°, space group P21/c, z=4. Table 1. Non-hydrogen atomic parameters. Esd in parentheses. [Beq(A2) is the isotropic equivalent of the anisotropic temperature factor.] X y z Beq p 0.2072(1) 0.5021(1) 0.1826(0) 3.48(2) 0(1) 0.3503(2) 0.3336(2) 0.2266(1) 3.70(4) 0(2) 0.1751(2) 0.6203(2) 0.2479(1) 3.76(5) 0(3) 0.2368(2) 0.6707(4) 0.1335(1) 6.11(6) 0(4) 0.0564(2) 0.3460(3) 0.1383(1) 4.05(5) 0(5) 0.6115(2) 0.3690(3) 0.4440(1) 5.16(5) C(l) 0.3461(2) 0.1843(3) 0.2881(1) 3.50(6) C(2) 0.3281(2) 0.3285(3) 0.3507(1) 3.12(6) C(3) 0.1781(2) 0.4746(4) 0.3119(1) 3.59(7) I C(4) 0.0359(3) 0.2082(5) 0.0705(1) 5.75(9) C(5) 0.3082(3) 0.1625(4) 0.4089(1) 5.05(8) C(6) 0.4711(2) 0.4898(3) 0.3945(1) 3.80(7) 249 Table 2. Hydrogen atom positional parameters. I (Thermal parameters equal to those of bonded atom.) X y z H0(5) 0.6449 0.2767 0.4096 HlC(l) 0.4474 0.0934 0.3139 H2C(l) 0.2551 0.0740 0.2635 H1C(3) 0.0850 0.3677 0.2898 H2C(3) 0.1724 0.5773 0.3529 H1C(4) -0.0665 0.1218 0.0498 H2C(4) 0.0329 0.3163 0.0279 H3C(4) 0.1253 0.0976 0.0855 HlC(5) 0.2144 0.0596 0.3795 H2C(5) 0.4051 0.0659 0.4366 H3C(5) 0.2905 0.2567 0.4488 H1C(6) 0.4461 0.6004 0.4280 H2C(6) 0.4899 0.5802 0.3545 Table 3. Bond lengths (A). Esd in parentheses. P-O(1) 1.563(1) P-O(3) 1.456(2) O(1)-C(l) 1.464(2) P-O(4) 1.569(1) C(l)-C(2) 1.522(2) O(4)-C(4) 1.446(3) C(2)-C(3) 1.522(2) C(2)-C(5) 1.535(3) C(3)-O(2) 1.464(2) C(2)-C(6) 1.537(2) 0(2)-P 1.562(1) C(6)-O(5) 1.414(2) 250 Table 4. Bond angles (0 ). Esd in parentheses. I O(2)-P-O(1) 106.2(1) O(1)-C(l)-C(2) 111.1(1) O(1)-P-O(3) 112.3(1) C(l)-C(2)-C(3) 109.5(1) O(1)-P-O(4) 107.0(1) C( 1)-C(2)-C(5) 108.7(2) O(2)-P-O(3) 112.6(1) C(l )-C(2)-C( 6) 111.4(1) O(2)-P-O(4) 102.3(1) C(3)-C(2)-C(5) 107.2(2) O(3)-P-O(4) 115.5(1) C(3)-C(2)-C(6) 109.3(1) P-O( I )-C( I) 117.1(1) C(5)-C(2)-C(6) 110.6(2) C(3)-O(2)-P I 18.4(1) C(2)-C(3)-O(2) 112.2(1) P-O(4)-C(4) 120.6(1) C(2)-C(6)-O(5) 113.4(2) Table 5. Torsional angles. Esd in parentheses P-O( I )-C( I )-C(2) -56.2(2) O(1)-P-O(4)-C(4) -70.2(2) 0( 1)-C(l )-C(2)-C(3) 57.7(2) O(2)-P-O(4)-C(4) 178.3(2) C( I )-C(2)-C(3 )-0(2) -55.2(2) O(3)-P-O(4)-C(4) 55.6(2) C(2)-C(3)-0(2)-P 51.2(2) 0( 1)-C( 1)-C(2)-C(5) 174.6(1) C(3)-0(2)-P-0(1) -43.5(1) O(1)-C(l)-C(2)-C(6) -63.3(2) 0(2)-P-0(1)-C(l) 45.9(1) C(5)-C(2)-C(3)-O(2) -173.0(1) 0(3)-P-0(1)-C(l) 169.5(1) C(6)-C(2)-C(3)-O(2) 67.0(2) 0(4)-P-0(1)-C(l) -62.8(1) C( 1)-C(2)-C( 6)-0(5) -69.6(2) C(3)-0(2)-P-0(3) -166.8(1) C(3)-C(2)-C(6)-O(5) 169.3(1) C(3)-0(2)-P-0(4) 68.6(1) C(5)-C(2)-C(6)-O(5) 51.5(2) 251 Table 6. Angles (0 ) between some least-squares planes. [O(1)-O(2)-C(3)-C(l)] and [O(l)-P-O(2)]; Al: 38.7 [O(1)-O(2)-C(3)-C(l)] and [C(l)-C(2)-C(3)]; A2: 50.7 [O(3)-P-O(4)-C(2)-C(6)-C(5)] and [P-O(4)-C(4)]; B: 51.2 252 Appendix E X-Ray data of trans-5-hydroxymethyl-5-methyl-2-thiono-r-2-methoxy- 1,3,2-dioxaphosphorinane (203). Cell: Orthorhombic, a= 6.129, b = 11.205, c = 13.192A, space group Pca21, z=4. Table 1. Non-hydrogen atomic parameters. Esd in parentheses. [Beq(A2) is the isotropic equivalent of the anisotropic temperature factor.] X y z Beq s 0.8824(1) 0.0203(2) 0.1097 5.17(4) p 0.8356(1) 0.1805(2) 0.2367(2) 3.32(3) 0(1) 0.8939(3) 0.1329(5) 0.3548(4) 3.92(8) 0(2) 0.8522(3) 0.4083(5) 0.2167(3) 4.28(9) 0(3) 1.0905(4) 0.4410(10) 0.4727(,5) 7.26(16) 0(4) 0.7144(2) 0.1772(5) 0.2624(4) 4.69(11) C(l) 0.8789(4) 0.2673(8) 0.4541(4) 4.05(12) C(2) 0.9030(4) 0.4774(8) 0.4224(5) 3.78(12) C(3) 0.8363(4) 0.5414(8) 0.3178(6) 4.32(13) C(4) 1.0193(4) 0.5064(10) 0.3885(6) 5.35(16) C(5) 0.8726(5) 0.6136(11) 0.5253(6) 5.68(17) C(6) 0.6580(5) -0.0044(8) 0.2791(6) 5.32(16) 253 Table 2. Hydrogen atom positional parameters. (Thermal parameters equal to those of bonded atom.) X y z HO(3) 1.0781 0.5195 0.5477 HlC(l) 0.9265 0.2289 0.5200 H2C(l) 0.8043 0.2624 0.4809 H1C(3) 0.7605 0.5370 0.3411 H2C(3) 0.8560 0.6782 0.2948 H1C(4) 1.0319 0.6493 0.3744 H2C(4) 1.0334 0.4294 0.3154 H1C(5) 0.9152 0.5745 0.5965 H2C(5) 0.8873 0.7520 0.5044 H3C(5) 0.7965 0.5946 0.5434 H1C(6) 0.5821 0.0250 0.2942 H2C(6) 0.6643 -0.0873 0.2064 H3C(6) 0.6881 -0.0776 0.3475 Table 3. Bond lengths (A). Esd in parentheses. P-O(1) 1.568(4) P-S 1.901(2) O(1)-C(l) 1.462(6) P-O(4) 1.569(3) C(l)-C(2) 1.499(8) O(4)-C(6) 1.436(6) C(2)-C(3) 1.524(8) C(2)-C(4) 1.542(6) C(3)-O(2) 1.474(7) C(4)-O(3) 1.390(8) O(2)-P 1.574(4) C(2)-C(5) 1.538(7) 254 Table 4. Bond angles (0 ). Esd in parentheses. S-P-O(1) 112.6(1) O(1)-C(l)-C(2) 112.2(4) O(2)-P-S 114.1(2) C(l)-C(2)-C(3) 110.1(4) S-P-O(4) 116.2(1) C(l)-C(2)-C(4) 112.2(5) O(2)-P-O(1) 105.1(2) C(l )-C(2)-C(5) 109.6(5) O(1)-P-O(4) 107.6(2) C(3)-C(2)-C(4) 107.7(4) O(2)-P-O(4) 99.9(2) C(3)-C(2)-C(5) 106.4(5) P-O(1)-C(l) 118.1(3) C(4)-C(2)-C(5) 110.8(4) C(3)-O(2)-P 117 .9(3) C(2)-C(3)-O(2) 110.9(4) P-O(4 )-C( 6) 121.8(3) C(2)-C(4)-O(3) 114.4(5) Table 5. Torsional angles (0 ). Esd in parentheses S-P-O(1)-C(l) 69.5(3) C(2)-C(3)-O(2)-P 54.1(5) O(2)-P-O(1)-C(l) 44.8(4) 0( 1)-C( 1)-C(2)-C(3) 56.3(5) O(4)-P-O(1)-C(l) -61.1 (4) 0( 1)-C( 1)-C(2)-C( 4) -63.6(6) C(3)-O(2)-P-S -169.2(3) 0( 1)-C( 1)-C(2)-C(5) 173.0(4) C(3 )-O(2)-P-O( 1) -45.3(4) C( 1)-C(2)-C(3 )-0(2) -56.2(5) C(3 )-O(2)-P-O( 4) 66.1(3) C(4 )-C(2)-C(3)-O(2) 66.3(5) S-P-O(4)-C(6) 52.5(5) C(5)-C(2)-C(3)-O(2) -174.9(4) O(1)-P-O(4)-C(6) -74.8(5) C(l )-C(2)-C( 4)-0(3) -54.5(7) O(2)-P-O(4)-C(6) 175.7(4) C(3)-C(2)-C(4)-O(3) -175.7(5) P-O(1)-C(l)-C(2) -53.8(5) C(5)-C(2)-C( 4)-0(3) 68.3(7) 255 Table 6. Angles (0 ) between some least-squares planes. [O(1)-O(2)-C(3)-C(l)] and [O(1)-P-O(2)]; Al: 38.5 [O(1)-O(2)-C(3)-C(l)] and [C(l)-C(2)-C(3)]; A2: 50.8 [S-P-O(4)-C(2)-C(4)-C(5)] and [P-O(4)-C(6)]; B: 46.4