Sorry HumbleStudent for me taking somewhat long to reply
.
Thanks for posting the above synth, I read about this, too, but didn't really consider it worthwhile posting, mainly because of the difficulty making the 5,6,7,8-tetrahydroisochinoline (reaction with POCl3 at elevated temperature + pressure involved, synth can be found at: R. Grewe + A. Mondon, Chem. Ber. 81, 279 (1948)). Also again a very good = very expensive pump is needed for cleanup and catalytic hydrogenation is utilized. Furthermore p-MeOPhMgX (the grignard reagent used here) is rather too reactive causing the grignard-reaction to give worse yields than with ordinary PhMgX. Therefore the method is considered to be more suitable for (only) N-substituted morphinans, which haven't been investigated thoroughly enough in my oppinion.
So here are the translated excerpts concerning the Pictet-Spengler synthesis:
- Unfortunately the exact reference was lost, but the article by Hans Hennecka (Bayer Research Laboratories) is from 1953, starting at page 110. The journal, I think, is Angewandte Chemie.
starting from p. 126:
'4. Synthesis of 3-Hydroxy-N-methyl-morphinan
p-Methoxyphenyl-glycidic acid methyl ester (XXXVIII) [of course the 2,3-substituted compound, note that the
methyl ester is formed]
To synthesize this glycidic acid ester it is not necessary to work in ether with sodium [31], dry sodium methylate or with sodamide. Rather it is possible, as already R. B. Loftfield described, to effect the condensation of anisic aldehyde with chloroacetic acid [ethyl] ester in methanolic sodium methylate solution. To compensate for the accompanying side reaction of formation of methoxy acetic acid [methyl/ethyl?] ester [Trying to use bromoacetic acid ethyl or methyl ester instead of the chloro compound resulted in complete failure of the reaction. SWIM supposes that with the more reactive bromo compound this side reaction becomes the main reaction, as sodium bromide is immidiatelly formed upon dripping the aldehyde/ester mix into the methanolic solution. The bromoacetic acid ethyl ester is a terrible lachrymator and a carcinogen anyway!], 1 1/2 mol each of chloroacetic acid [ethyl] ester and sodium methylate is used per mol of anisic aldehyde.
A sodium methylate solution containing 69 g (3 gramm atom) sodium (prepared by dissolving 70 g sodium in 1 L methanol and titration after filtration through glass filter) is cooled to - 10 °C and treated within 5 - 6 hours with moderate stirring with a mixture of 272 g (2 mol) freshly distilled anisic aldehyde and 326 g (3 mol) of acid free chloroacetic acid [ethyl] ester. A nearly colorless to slightly yellow paste, that can still be stirred well is formed. After completing the addition stirring is continued for 2 hours at - 10 °C and the mixture is allowed to stand overnight without renewing the cooling. The next morning the mix is slightly acidified by addition of little GAA and the paste is transfered to 6 L of H2O that has been slightly acidified with GAA. After standing 1/2 h the crystal solid is filtered with suction, washed with water and air-dried. Yield 360 - 390 g, 86.5 - 93.7 % of theoretical yield, mp. 60 - 62 °C; the ester can be distilled at high vacuum without decomposition, bp. 145 °C / 0.7.
[elemental analysis omitted]
1-p-Methoxybenzyl-10-hydroxy-decahydro-isoquinoline (XXXIII)
a) with p-Methoxyphenyl-acetaldehyde bisulfite compound
[omitted, worse yields and more hassle than b)]
b) with p-Methoxybenzyl-glycidic acid [methyl] ester
In 20,145 L of dil. aq. HCl, containing 3.95 mol of HCl, 465 g (3.72 mol) of cyclohexenylethyl amine is dissolved, 774 g p-methoxyphenyl-glycidic acid methyl ester is added and stirred for 48 h at 80 - 90 °C until CO2 evolution ceases. pH of solution: 3 - 4. After cooling down, the solution is decanted from a small amount of tarry residue and the now colorless solution is clarified by stirring with a little bit of animal charcoal, filtrated and the now clear solution is overlayed with a little bit ether and base is set free with conc. K2CO3 solution. The, at first oily, deposit soon solidifies to fine, nearly colorless crystals that mainly collect at the interface of the ether layer. The thereby obtained raw 1-p-methoxybenzyl-10-hydroxy-decahydro-isoquinoline is filtered with suction after 2 h standing, washed with water and then with a little bit of ether and dried. Raw yield 413 g, 40.3 % of theoretical yield. After recrystallization from dil. methanol (1800 ml MeOH, 2000 ml H2O) 373 g of mp. 152 °C is obtained. Identical with the product from a) by mp. and mixed mp.
1-p-Methoxybenzyl-N-methyl-10-hydroxy-decahydro-isoquinoline (XXXV)
A solution of 20 g XXXIII in 100 ml of technical formic acid (ca. 84 %), prepared in the cold, is treated with 8 g of 30 % formaldehyde solution and heated for 12 h on a water-bath after standing for 2 h. The excess formic acid is evaporated i. v., the residue is dissolved with water, the base set free with K2CO3 solution and the oil that deposits is dissolved with ether. The oily base that is obtained upon evaporating the ether can be distilled at high vacuum. Bp. 160 - 165 °C / 0.01; highly viscous oil, that crystalizes upon dissolving in little pet ether. mp. 65 - 66 °C, picrate mp. 153 - 156 °C (from MeOH).
[elemental analysis info omitted]
1-(p-hydroxybenzyl)-N-methyl-delta 9, 10-octahydro-isoquinoline (XXXVII)
11.1 g 10-hydroxy-precursor (XXXV) is boiled in 100 ml constant boiling HBr for 20 min, until the evolution of gas is finished. Upon pouring the cooled down solution into ice water a stiff paste deposits that becomes crystalline upon standing. It is filtered with suction, washed with a little bit of ice water and recrystallized from MeOH/ether. Thereby 3.7 g of the hydrobromide of XXXVII with mp. 235 - 237 °C is obtained.
[elemental analysis info omitted]
Gives no [mp] depression with the salt of equal constitution from Schnider & Grüssner [15].
3-Hydroxy-N-methyl-morphinan (XXIX)
150 g XXXV is dissolved in 1500 ml of constant boiling HBr (d = 1.49 - 1.50) and boiled for 6 h. Then excess HBr is evaporated from the slightly yellow solution i. v. at 60 - 80 °C, the syrup obtained is dissolved in 300 ml hot methanol and the thereby obtained solution is diluted with 3 L H2O with stirring. Benzene-[1-]butanol 1 : 1 is added on top in a sep funnel and the base is set free with conc. NH3 and the voluminous non-crystalline deposit is driven over into the benzene-butanol layer, which can easily be done. The layer is removed, the aqeous phase is again extracted with benzene-butanol and the pooled extracts are washed first with water and then with conc. NaCl solution, filtrated through a double filter and then the solvents are evaporated i. v. at 50 - 60 °C. The remaining syrup, which contains crystals, is boiled in 150 ml methanol until all syrupy substance has been dissolved. The colorless crystals obtained upon cooling down are filtered with suction, washed with little methanol and then with ether. Raw yield 41.2 g, 31 % of theoretical amount; mp. 250 - 251 °C (from methanol).
Identical by mp. and mixed mp. with the 3-hydroxy-N-methyl-morphinan obtained by Schnider & Grüssner [15]
The dark brown mother liquor of 3-hydroxy-N-methyl-morphinan is boiled down to a syrup i. v. at 40 - 50 °C and vigorously shaken with 500 ml ether in a flask until a fine, seemingly still amorphous, separation of the syrup has been effected. After prolonged standing in a sealed flask a light brown crystalline precipitate is formed, which is filtered with suction and washed with ether. Raw yield 38 g; mp. 187 - 193 °C. By careful fractionating crystallization from methanol 20 - 25 g of pure compound with mp. 197 - 198 °C can be obtained [this isomer of unclear constitution is stated to be ineffective as an analgesic!]
References:
[15] O. Schnider, A. Grüssner, Helv. chim. Acta 32, 821 (1949).
[31] ethyl ester: K. W. Rosenmund, H. Dornsaft, Ber. dtsch. chem. Ges. 52, 1740(1919); see also DRP. 591 452 (21. 3. 1930) I. G. Farbenindustrie AG'
Sounds sweet, eh
? Hard to say, how well the conversion from octahydro-isoquinoline/10-hydroxy-decahydroisoquinoline to morphinan really works, but it does look possible! The Pictet-Spengler route seems to be perhaps the most elegant and promissing one, IMHO. This field takes a lot of dedicated and frustration resistant work, but it might be worth the fuss in the end
.