Acid catalyzed alcoholysis

Get Complete Project Material File(s) Now! »

As with the 2,8-N-aryl derivatives, both N-alkyl substituted substrates reacted with the HCl-containing methanol according to the reaction shown in (Scheme 2.1), yielding the corresponding salts of the products1 5d and 5f. The regioselectivity of the alcoholysis of the N-aryl substituted derivatives was explained in terms of the basicity of the different nitrogen atoms. In the N-alkyl derivatives all the amide nitrogen atoms carry only alkyl substituents and should not differ much in basicity. 15N NMR studies of the P-N bonding in cyclic amides demonstrated however that the endocyclic nitrogen is always more shielded than the exocyclic nitrogen atom2 .
Following the arguments developed in an important contribution from Von Philipsborn and Müller3 , our NMR results suggest that the nitrogen incorporated into the ring should always be more basic, therefore the bridgehead N atom in compounds 3d and 3f should be more basic than the other two nitrogen atoms, each located in a single phospholidine ring. However, if the observed selectivity for 3d (R=Et) and 3f (R=PhCH2) depended only on the basicity of the different nitrogens, it would require a pKa difference of as much as approximately two units between the bridgehead nitrogen and the other two nitrogens. It has to be remembered, however, that in the pentacoordinate intermediate (the transition state) of the substitution, the location of the bridgehead N(5) atom is in the favourable apical position (apical departure). It seems therefore that the selectivity of the acid alcoholysis of all substrates 3 is determined by both the pre-equilibrium protonation step and by stereoelectronic effects.

Base catalyzed alcoholysis

Both prepared N-alkyl substrates 3d and 3f yielded exclusively the eightmembered cyclic products in the base-catalyzed methanolysis, the same as the products obtained from methanolysis in the presence of HCl. This result suggests that the cleavage of the P-N bond by MeO- , follows a mechanism different from that operating for the N-aryl derivatives, and is driven by the cleavage of the more strained P-N(5) bond in the bicyclic system. The difference between the N24 University of Pretoria etd – Laurens, S (2005) aryl and N-alkyl substrates was also observed for their reactivity in the related 5→6 rearrangement (see 2.3.3). 2.2.3 Rearrangement of the alcoholysis product Both N,N’-dialkyl substituted compounds 5d and 5f rearranged to compounds 6d and 6f, but much slower than in the N,N’-diaryl analogues. [Scheme 2.2] The rearrangement reaction was much less clean for N-alkyl than for the N-aryl derivatives. Additional signals to those of the rearrangement products appear in the 31P NMR spectrum at higher conversions.
In agreement with the trend observed for the N-aryl substrates, 5d (R=Et) was found to be ca. five times less reactive than 5a. The rearrangement of 5f (R=PhCH2) was very slow and the reaction did not allow us to isolate and characterize the product 6f. In the first experiment (entry 5, Table 2.1) 3f was used as a precursor for 5f. When 3f was dissolved in MeOH containing NaOMe, it solvolized relatively fast to 5f; after four days the solution contained no 3f, 98% of 5f and 2% of the product 6f. The 5→6 interconversion yielded very slowly the final product with an approximate half-life of 960 hours. In the absence of the base (MeO- ) the rearrangement was too slow to allow any rate determination.

N,N-Bis(2-chloroethyl)-N’,N”-dibenzylphosphoric triamide, 1f

A solution of N,N-Bis(2-chloroethyl)phosphoramidodichloride (0.256 g,1.0 mmol) in ether (20 ml) was added dropwise with stirring at -70°C to a solution of benzylamine (0.439 g,4.1 mmol) in ether (10 ml). The mixture was kept at -70°C for 2 h, allowed to warm up to room temperature and stirred for another 140 h. The precipitate (benzylammonium chloride) was filtered off and washed with ether (20 ml). The combined ethereal solution was washed with water (2 × 20 ml) and cooled to 0°C (without drying). Two layers separated, which, after 4 days, yielded the crystalline product at the interface of the layers. Colourless crystals, 0.385 g (97%); mp 83-85 °C.

READ Classification of Photoreactor on the Basis of the State of the Catalyst

1-Oxo-2,8-diethyl-2,5,8-triaza-1λ5

-phosphabicyclo[3.3.0]octane (3d). A mixture of the phosphoric triamide described above (1.00 g, 3.6 mol), NaH (prewashed with benzene, 0.400g 16.6 mmol) and Bu4NBr (0.27 g, 0.18 mmol) was stirred in benzene (100 ml) at room temperature for two days; the reaction progress was monitored by recording directly the 31P NMR spectra of samples of the solution. An additional amount of NaH (0.400 g) was added and the stirring was continued for another two days. The 31P NMR spectrum demonstrated full conversion and formation of a single phosphorus-containing product. The benzene solution was decanted, the residue was washed several times with benzene and the combined benzene solution was evaporated under reduced pressure. The crude product (viscous oil, 0.886 g, >100%) was purified by bulbto-bulb distillation (oven temp. 200 °C/0.06 mmHg) yielding 0.475 g (65%) of the almost pure product; second bulb-to-bulb distillation (oven temp. 150-200 °C/0.07 mmHg) afford pure 3d (0.400 g, 58%) as a colorless viscous oil.

Acid catalyzed alcoholysis

The first reaction studied for the new bicyclic compound 11a was the acid catalysed solvolysis of the amide bond. A similar substitution reaction was expected like for the phosphoryl derivatives because the P-N bond is very unstable under acidic conditions20 in a variety of amides derived from phosphorus acids. When the thiobicyclic compound 11a was treated with methanol containing one mol equivalent of HCl, we observed the appearance of a new signal, only 5 ppm up field in the 31P-NMR spectrum of the crude reaction mixture, with complete disappearance of the signal of the substrate. Reaction was completed in less than five minutes. Under acidic conditions the product was stable in solution and could be isolated as a white solid after evaporation of methanol. 1 H-NMR spectrum indicated that it is the substitution product, the eight membered ring compound 15 (Scheme 3.11). The isolated product was not pure enough to give a reliable 13C NMR spectrum. Therefore the crude product was analysed by GC-MS without further purification to confirm the structure of the compound. The product 15 of the acid catalyzed reaction is stable as the neat hydrochloride salt. The hydrochloride salt was dissolved in chloroform and treated with aqueous K2CO3 to liberate the free amine. After decanting the aqueous layer, the organic layer was dried and without evaporation of the solvent, the 1 H-NMR of the sample was recorded. 1 H-NMR spectrum of the free amine 15’ did not correspond to the 1 H-NMR spectrum of the eight-membered ring compound 15 and 13C NMR was not very useful because of the impurity of the sample.

INDEX :

Abstract
Opsomming
1. Introduction
- 1.1 General Background
- 1.2 Objectives
- 1.3 References
2. N-Alkylderivatives of 1-oxo-2,8-disubstituted-2,5,8-triaza-1λ5 – phosphabicyclo[3.3.0]-octane
- 2.1 Introduction
- 2.2 Results and Discussion
- 2.2.1 Acid catalyzed alcoholysis
- 2.2.2 Base catalyzed alcoholysis
- 2.2.3 Rearrangement of the alcoholysis product
- 2.3 Experimental
- 2.4 References
3. Thio analogues of 1-oxo-2,8-diaryl-2,5,8-triaza-1λ5 –phosphabicyclo[3.3.0]- octane
- 3.1 Introduction
- 3.2 Results and Discussion
- 3.2.1 Synthesis
- 3.2.2 Acid catalyzed alcoholysis
- 3.2.3 Base catalyzed alcoholysis
- 3.2.4 Rearrangement of the alcoholysis product
- 3.2.5 Derivatives of 1-thio-2,8-diaryl-2,5,8-triaza-1λ5 – phosphabicyclo[3.3.0]-octane
- 3.3 Experimental
- 3.3.1 Synthesis of new thio analogues
- 3.3.2 GC-MS Analysis
- 3.4 References University of Pretoria etd – Laurens, S (2005)
4. Structural Analysis
- 4.1 NMR Analysis
- 4.1.1 Introduction
- 4.1.2 Experimental
- 4.1.3 Results and discussion
- 4.2 Crystal Structure Analysis
- 4.2.1 Introduction
- 4.2.2 Experimental
- 4.2.3 Results and discussion
- 4.3 References
5. Theoretical Calculations
- 5.1 Introduction to Molecular Modeling
- 5.2 Experimental
- 5.3 Results and Discussion
- 5.4 References
6. Conclusion
- 6.1 References
Publications originated from this work