A multinuclear NMR study of some cyclic phosphonic diamides

IH,13C, 15N and31p NMR data are presented for four cyclic phosphonic diamides. In tenns of useful structural infonnation it is found that the 31p chemical shifts and IJCP_15N) couplings are the most sensitive parameters to variations in molecular structure.


Introduction
Previously we have used multinuclear Nrv1R techniques in a nwnber of molecular structure investigations.For those molecules which contain nitrogen atoms it has invariably been the case that nitrogen Nrv1R is the most sensitive structural tool for our studies [1].We have also used 31p Nrv1R in a study of tautomeric equilibriwn in some iminophosphines [2].As an extension of this earlier work we now propose to use IH, BC, lSN and 31p Nrv1R in a structural investigation of some cyclic phosphonic diamides.These compounds contain P-N bonds and it will be interesting to observe, whether the lsN or the 31p Nrv1R data is the more sensitive to variations in molecular structw"e.

Result and discussion
The lH and BC Nrv1R data for the compounds studied are presented in Table 1.The IH signals were assigned on the basis of their multiplicity and relative intensities.Thus, the high frequency broad doublets were assigned to the NH groups, with the coupling constant 2Jrlp-IH) = 17-19 Hz.The doublet at 6.5 ppm in compound 4, with the big coupling constant IJ~lp_lH) = 579.4 Hz is assigned to the H atom of the PH group by comparison with literatw"e data [3].The BC signals were assigned by analysis of the chemical shifts and BC_1H coupling constants obtained from the BC Nrv1R fully coupled spectra.The chemical shifts of the Cl, C2 and C3 atoms are similar for all of the compounds studied and are similar for published compounds [4].The doublets at 26.3 ppm and 28.1 ppm in compound 1 were assigned to the -C'H 2 -C"H 2 -exocyclic group.The signal at 26.3 ppm has a typical coupling value of 1 J rlp-BC) = 122.6Hz and the signal in 28.1 ppm has the smaller coupling of 2J~lp_13C) = 3.6 Hz.The low frequency peaks in compounds 3 and 4 are assigned to the exocyclic-OCH3 groups.The I3C NMR data of the phenyl exocyclic group in compound 2 has very similar coupling values to those oftriarylphosphine oxide [5].Only one coupling constant 1 J(31p_ I3CI) = 164.0Hz is higher when compared with the corresponding value for triarylphosphine oxide (104 Hz) [5].The lsN and 31p NMR data are presented in Table 2.The ISN chemical shifts appear in the range typical for phosphonic amide [6], and the IJ(ISN_IH) couplings are typical in value for amino groups.The 31p chemical shifts for the compoWlds 1-3 are characteristic values for phosphonic amides [6][7] and the 31p chemical shift for compound 4 has a nonstandard value for a pentavalent phosphorus atom as found for phosphonic amides [3].Consequently we may conclude that for the purposes of structural investigation of the compounds studied the 31p NMR data chemical shifts and IJ~lp_ISN) couplings are much more sensitive than the result produced by IH, I3C and ISN NMR.

Experimental
The compounds were studied in DMSO-~ solution, with a typical concentration ca.0.1-0.5 m.The solvent deuteriwn signal was used as the internal lock.The solvent peaks were used as a reference for the IH and I3C NMR data (~ = 2.49 ppm; Be = 39.5 ppm).Neat nitromethane was used as the external reference for the ISN measurements ( ~) = 0.0 ppm), and 85% H 3 P0 4 was used as the external reference for the 31p NMR spectra.
IH NMR spectra were obtained on a Varian BB-200 spectrometer at 199.975 MHz.The I3C measurements were recorded on a Varian BB-200 spectrometer at 50.289 I\1Hz, with a relaxation delay of 2s, an acquisition time of 1 s and a flip angle of 45°.The 3Ip measurement were recorded on a Varian BB-2oo spectrometer at 80.950 MHz, with a relaxation delay of 2s, an acquisition time of 0.8s and a flip angle of 45°.
The lsN measurement were recorded on a Broker AM 500 at 50.698 MHz using typical parameter values in the Inept program.
Substrates P[N(CH3h12OCH3 (A) and P[N(CH3h12CJIs (B) were synthesized by standard procedures [8,9].Acetonitrile and CH 2 Cl 2 were dried over P20S and distilled, triethylamine was dried over KOH and distilled, benzene was distilled over Na, HN3 was synthesized in benzene by a standard procedure [10].Reaction of A with CH2=CHCO~t (1).Compound (A) (ca.) 0.01 m) was dissolved in a small amount of acetonitrile and CH 2 =CHC0 2 Et (0.02 m) was added.After 7 days the precipitate was filtered and recrystallized from acetonitrile.
Oxidation of B (2). CompoWld (B) (ca.O.Olm) was dissolved in a small amount of acetonitrile.The mixture was stirred under dry O 2 for 48 h.The precipitate was filtered and recrystallized from benzene.
Oxidation of A (3). Compound (B) (ca O.OIm) was dissolved in a small amount of acetonitrile.The mixture was stirred Wlder dry O 2 for 24 h.The solvent was evaporated in vacuo.The oil was dissolved in a small amount of CH 2 C1 2 and the compoWld was separated on a colwnn (silica gel, CH 2 CIJ. Reaction of A with lIN j (4).CompoWld (A) (ca O.Olm) was dissolved in benzene and HN3 in benzene (O.04m) was added.The mixture was stirred for 7 days.The precipitate was filtered, dissolved in benzene and triethylamine (0.02 m) added.The solution was filtered and the solvent was evaporated in vacuo.The residue was dissolved in a small amoWlt of CH 2 Cl 2 and the compoWld was separated on a colwnn (silica gel, CH 2 CI 2 ).

Table 1 IH
All measurements were taken as solutions in DMSO~; (b) 15N NMR chemical shifts are reported with respect to neat nitromethane as an external standard; ( c) 31p NMR chemical shifts are reported with respect to 85% H 3 P0 4 as an external standard.
(a) All measurements were taken in DMSO solutions and the solvent peaks used as internal standards, the values are recalculated for 1MS (0.0 ppm).