Preparation and pyrolysis of some N -protected derivatives of aminomethylene Meldrum’s Acid

The syntheses of new N -protected aminomethylene Meldrum’s acid derivatives have been achieved and their gas-phase pyrolyses under flash vacuum pyrolysis (FVP) conditions have been studied. The pyrolyses were dominated by polymerisation and radical-cleavage pathways rather than by intramolecular cyclisation mechanisms


Introduction
An inspiring young chemist, Meth-Cohn, Over all Lakeland fell-sides would roam, But unlike Katritzky (Or even Suschitzky?),He always could find his way home.This selfsame 'young' fellow called Otto, Was of voce which none could call sotto, For his sixty-fifth year, We wrote this paper here, (Though not everything did what it ought-to.) In previous papers we have shown that flash vacuum pyrolysis (FVP) of N,N-disubstituted aminomethylene derivatives of Meldrum's acid 1 provides a straightforward synthetic route to N-alkyl-or N-aryl-3-hydroxypyrroles 2 (R 1 = alkyl or aryl), by a hydrogen transfer-cyclisation mechanism from the initial methyleneketene intermediate (Scheme 1). 1,2Because of their electron rich nature and sensitivity to oxidation, 3-hydroxypyrroles are often difficult to obtain by other means.However, this route is not directly applicable to the preparation of N-unsubstituted hydroxypyrroles 2 (R 1 = H), since the pyrolysis takes

Scheme 1
Our route to the precursors involves the reaction of methoxymethylene Meldrum's acid (MMMA) 3 with an appropriate protected amine (Scheme 1, R 1 = protecting group) in acetonitrile solution. 1Reaction conditions are mild (room temperature, 5-10 min) and methanol is the only co-product.Silyl protection of amines using classical silylation reagents is often unsuccessful because of the extreme hydrolytic instability of the products.However, the use of N-tbutyldiphenylsilyl derivatives has been advocated; 6 these are generally stable under neutral and basic conditions and require 80% acetic acid for deprotection.Compounds 4-6 were made by standard methods, 6,7 and were reacted with MMMA 3 in acetonitrile, but the only products which could be isolated, generally in excellent yield, were the N-unsubstituted compounds 7-9.These were easily identified from the characteristic doublet at ca.δ H 8.2 (J 15 Hz) due to the methylene proton which shows trans-coupling to the adjacent NH.Compounds 7 and 9 have been previously made by direct reaction of MMMA 3 with the appropriate amine. 3The facile loss of the protecting group under essentially anhydrous, mildly basic conditions was most surprising, and may be connected with the sterically congested nature of the expected products, which are vinylogous amides rather than amines.It is perhaps significant that when the reaction was carried out using a non-polar solvent (cyclohexane), only recovered starting materials were obtained, even after extended reaction times.
In principle, weak bonds from nitrogen to heteroatoms (N or O) may be cleaved under reductive conditions. 8We therefore targeted the hydrazone 10 and the hydroxylamine drivative 11 as potential pyrolysis precursors.The N-benzyl substituent was chosen because of the high reactivity of benzyl-type hydrogen atoms to the transfer-cyclisation process, 1 and so the known N-benzylhydrazine 12 9 was synthesised and reacted with MMMA 3 to give the pyrolysis precursor 10 in 44% yield.The major product from the pyrolysis of the hydroxylamine derivative 11 at 500, 550 or 600 °C was again an insoluble polymer.A minor, soluble component showed a number of resonances in the region δH 3.45 -5.30 but no constituent compound could be characterised; it was clear that no pyrrolone had been formed.It is possible that methoxygroup migration on the methyleneketene energy surface (which is known to be a facile process 10,11 ) leads to iminoketene intermediates, which polymerise on warming.Matrix isolation experiments would be required to confirm this hypothesis.

Issue in
In addition, the secondary derivatives 18 and 19 were also synthesised (see Experimental section) but again both gave polymeric material on pyrolysis.In the case of the hydrazine 18, it was hoped that, by analogy with the cyclisation of unsaturated hydrazinoesters 20 discovered by Chuche and co-workers (Scheme 3), 12 unusual pyrazole derivatives might be formed via analogous ketene intermediates (c.f.21 and 22).The failure of our cyclisation suggests that the presence of the electron withdrawing group may be essential for the success of the hydrazinoester process.
Hydrogenolysis of N-benzyl groups is a common means of deprotection which has been used in the alkoxypyrrole series. 13Because pyrolysis of N-alkyl-Nbenzylaminomethylene Meldrum's acid derivatives always leads to N-alkyl-2-phenyl-3hydroxypyrroles via hydrogen transfer from the benzyl CH 2 group, 1 we hoped that the dimethylbenzyl group would provide required protection (Scheme 4).
N,α,α-Trimethylbenzylamine 23 was made by the literature method, 14 and was reacted with MMMA 3 to give the product 24 in 45% yield.Unfortunately the only product from the pyrolysis of 24 at 600 °C was α-methylstyrene 25, which is clearly formed by an initial radical cleavage of the C-N bond.Pyrolysis at lower temperatures (500 and 550 °C) gave only 24 and 25 in differing proportions.We therefore briefly investigated the temperature profile of the pyrolyses of Nmethylbenzylamine 26 (which gives bibenzyl), N-α-dimethylbenzylamine 27 (which gives styrene) and N,α,α-trimethylbenzylamine 23 (which gives α-methylstyrene) (Figure 1, series 3, 2 and 1 respectively).All of these decompositions are presumably initiated by a radical C-N cleavage mechanism (Scheme 6).The temperatures for 50% conversion to products for 26, 27 and 23 are >800 °C, 790 °C and 690 °C respectively.It is clear that the second C-methyl group in 23 has a greater effect in lowering the temperature for radical cleavage than the first C-substitution.In addition, the temperature for quantitative conversion of 23 to products was >800 °C (by comparison with quantitative decomposition of the Meldrum's acid derivative 24 at ca. 600 °C) which suggests that the aminomethylene Meldrum's acid moiety stabilises the co-formed N-centred radical more effectively than a simple methylaminyl function.
In conclusion, we have had no success in our attempts to extend our hydroxypyrrole synthesis to N-unsubstituted examples by using the range of different N-protected precursors described here.In subsequent work the use of amide (lactam) precursors gave a restricted access to these targets and these results will be reported in a later paper.

Silylation of primary amines 6,7
To a stirred solution of the appropriate amine (freshly distilled and dried over potassium hydroxide pellets, 2 mmol) in dry acetonitrile (6 cm 3 ), under nitrogen, was added triethylamine (0.304 g, 3.5 mmol) and t-butyldiphenylsilyl chloride (0.550 g, 2 mmol).The mixture was stirred at room temperature for the time quoted below and the solvent was removed under reduced pressure.The residue was dissolved in a mixture of 4:1 hexane/ethyl acetate (25 cm 3 ) and washed with aqueous sodium hydrogen carbonate (1M, 3 x 25 cm 3 ).The combined aqueous washings were back extracted with 4:1 hexane/ethyl acetate and the combined organic layers were dried (MgSO 4 ).The solvent was then removed under reduced pressure to give the desired product as an opaque oil.The following compounds were prepared in this manner.The amine used and reaction time are given for each example in parentheses.(17) and 77 (8).In all of the reactions, the use of "wet" acetonitrile, or "wet" amines resulted in a dramatic reduction in yield of the desired product and the production of tbutyldiphenylsilanol,mp 61-62 °C (from toluene) (lit., 15   (17), 137 (30), 121 (52), 78 (57) and 77 (100).
In an alternative method, 1 N-(t-butyldiphenylsilyl)-1-phenylethylamine 6 (0.179 g, 0.5 mmol) was dissolved in cyclohexane (10 cm 3 ) and 5-methoxymethylene Meldrum's acid 3 (0.093 g, 0.5 mmol) was added with stirring.The mixture was heated under reflux for 24 h and the solvent was removed under reduced pressure to give a yellow solid.Examination of the crude product by 1 H NMR spectroscopy showed only unreacted starting material.
A slightly different procedure was used for the hydroxylamine derivatives.The hydroxylamine hydrochloride (10 mmol) was suspended in acetonitrile (150 cm 3 ), and triethylamine (1.01 g, 10 mmol) was added.The mixture was stirred at room temperature for 30 min before addition of methoxymethylene Meldrum's acid 3 (10 mmol), then stirred for a further 2.5 h at room temperature.The solvent was removed under reduced pressure, the residue was dissolved in dichloromethane and washed with dilute hydrochloric acid (1 M).The organic fraction was dried (MgSO 4 ), and concentrated to give the product.The following derivatives were made in this way: 2,2-Dimethyl-5-(N-methyl-N-methoxyaminomethylene)-

Issue in Honor of Prof. Otto Meth-Cohn ARKIVOC 2000 (v) 806-819 ISSN 1424-6376 Page 808
© ARKAT USA, Inc N,O-Dimethylhydroxylamine is commercially available as its hydrochloride salt; without isolation of the free base, this salt was allowed to react with MMMA 3 in dilute solution in acetonitrile, in the presence of one equivalent of triethylamine, to give 11 in high yield after aqueous work-up.These conditions should be generally applicable to the reactions of other unstable amines, provided they are available as salts.

Issue in Honor of Prof. Otto Meth-Cohn ARKIVOC 2000 (v) 806-819 ISSN 1424-6376 Page 814
13and13C NMR spectra were recorded at 200 or 250 Mz and 50 or 63 MHz respectively for solutions in deuteriochloroform.Coupling constants are quoted in Hz. Mas spectra were obtained under electron impact conditions.