Cyclohexenylacetonitrile for Morphinans

[Pimpo]

Cyclohexylidenecyanoacetic Acid and-Cyclohexenylacetonitrile

Organic Syntheses, Coll. Vol. 4, p. 234

(http://www.orgsyn.org/orgsyn/prep.asp?prep=cv4p0234)

I guess that 1-cyclohexenylacetonitril might easily be converted (LiAlH4?) to 2-(1-cyclohexenyl)-ethylamine, which is in turn useful for making racemorphan/levorphanol (cf.

Post 319328

(blondie: "TITLE Preparation of useful intermediates of ...", Chemistry Discourse)). I saw an Indian company offering both the amine and nitrile on the web today and I figured that they make the amine from the nitrile, I never heard of any other way of making it anyway.

[Rhodium]

Linking to individual preparations at the Organic Syntheses website is easy as soon as you understand their URL logic, see here:

Org. Syn. Coll. Vol. 4, p. 234: 1-Cyclohexenylacetonitrile

(http://www.orgsyn.org/orgsyn/prep.asp?prep=cv4p0234)

Note the last part of the URL: prep=cv4p0234 - corresponds to - Coll. Vol. 4, p. 234

[Pimpo]

Thanks alot for your information Rhodium, I will make use of it in the future. Finally having found a synthesis for cyclohexenylacetonitrile greatly increased my interested in the morphinane series, so I decided to go through the literatue today (even though I'm so f**king tired that I hurt myself badly this morning, toenail turned blue and hurts alot ). Well anyway, I was quite succesful , here it is:

translated excerpts taken from Helvetica Chimica Acta, 23 (1950), p. 1437-1448, O. Schnider and J. Hellerbach - "187. Synthese von Morphinanen (2. Mitteilung)"

(p. 1443-1444)

A. Synthesis of Cyclohexene-1-yl-ethylamine

[..]

cyclohexene-(1)-yl-ethylamine

1. from cyclohexene-(1)-yl acetamide

To a well stirred suspension, cooled to 0 °C, of 139 g cyclohexene-(1)-yl acetamide (1 mol) in 6000 ml absolute ether is added dropwise under nitrogen in portions within ca. 15 min 76 g (2 mol) well powdered LiAlH4. The reaction product [not rather mixture?] is stirred over night and then heated for 15 min. After destroying the excess LiAlH4 with water [dangerous?] the ethereal solution is decanted and the residue washed with ether. After drying the ethereal solution with anhydrous K2CO3 the ether is distilled off and residual cyclohexenyl-ethylamin distilled under water pump pressure. Bp. 75-76 °C (10 mm). Yield 82.5 g (66% of theory).
The hydrochloride melts with decomposition at 160-163 °C
[elemental analysis info omitted]

2. from cyclohexene-(1)-yl-acetonitrile

a) with LiAlH4: To a solution of 42 g LiAlH4 (1.1 mol) in 2000 ml absolute ether is added under nitrogen atmosphere at 0 °C 121 g cyclohexene-(1)-yl-acetonitrile in 800 ml absolute ether with good stirring and stirring is continued for 2 h. After destroying the excess LiAlH4 with water and adding conc. NaOH until the ethereal solution is completely clear the reaction product is worked up as in the example before. Yield 89.5 g (66% of theory).

b) with catalyst: 1 kg cyclohexenyl acetonitrile is dissolved in 2000 ml methanol and reduced in the presence of 250 g Raney cobalt [anybody know about that?] in an autoclave at 60 °C under a pressure of 95 atm of hydrogen. After the uptake of 2 mol hydrogen the reduction is stopped and the catalyst filtered off. The residue that remains after distilling off the methanol is fractionated over a column; after a small forerun the cyclohexenyl ethylamine distills at 75-76 °C (10 atm). Yield ca. 90%.

... will be continued soon, too busy, too tired right now.

[Rhodium]

Very Interesting! Please go on, and definitely continue up to the point where they prepare an active compound from this.

[Pimpo]

Thanks for your attention, Rhodium. The rest of the synth basically follows DopaMan's outline 

https://www.thevespiary.org/rhodium/Rhodium/chemistry/levorphanol.html

, but of course there are a lot more details given:

... continuing from p. 1444 - p. 1446, all elemental analysis info omitted

Synthesis of N-Methyl-morphinane

phenylacetic acid-(cyclohexenyl-ethyl)-amide

To 125g Cyclohexenyl-ethyl amine (1 mol) in 250 ml absolute benzene is slowly added with cooling in ice and stirring 85 g (0,55 mol) phenylacetyl chloride in 170 ml absolute benzene. The reaction product [?] is allowed to stand for 30 min at RT and is then warmed for another 30 min on a water bath. After filtering off the Cyclohexenyl-ethyl amine hydrochloride the benzene solution is washed with sodium bicarbonate and the benzene is distilled off; the residue (123 g), which solidifies, consists of phenylacetic acid-(cyclohexenyl-ethyl)-amide. Mp 69 - 71 °C from benzene-petrolether. For analysis the product is distilled at high vacuum[is that the right word ? I mean like really low pressure.].

1-benzyl-2-methyl-1,2,3,4,5,6,7,8-octahydro-isochinoline

243 g (1 mol) phenylacetic acid-(cyclohexenyl-ethyl)-amide are heated with 730 g P2O5 in 1200 ml benzene for 45 min and then allowed to stand for 2 h at RT. To the reaction product[?] is added ice with strong cooling and the aqueous phase is separated from the benzene, which contains small amounts of the starting material. The base is precipitated from the aqueous phase by addition of solid KOH at 0 °C and extracted with ether and dried. After distilling off the ether there remain 178g of a light yellow oil, to which, dissolved in 80 ml aceton, there is added a solution of 200 g methyl bromide [yuck! the iodide might do too?] in 300 ml anhydrous acetone, and this is allowed to stand for 12 h.

The residue, which remains after removing the acetone in vacuo, is dissolved in little water and extracted with ether. The aqueous solution of the quaternary salt, after addition of 56 g KOH in methanol, is reduced in the presence of Raney-Nickel in an autoclave at 50 atm H2. The reduction product is dissolved is dissolved in benzene, after filtering off the catalyst and removing most of the methanol in vacuo, and then washed with water. After the usual workup the 1-benzyl-2-methyl-1,2,3,4,5,6,7,8-octahydro-isochinoline distills at 106 - 108 °C (0.01 mm). There is obtained 85.8 g of a yellow oil, the hydrochloride of this melts at 195 °C and does not depress the mp in a mixture with benzyl-methyl-octahydro-isochinoline HCl obtained by another method [1,2].

N-methyl-morphinane

The cyclization of the benzyl-N-methyl-octahydro-isochinoline to N-methyl-morphinane is done as known [1] and yields a base of bp 110 °C (0.04 mm), phosphate mp 141 - 145 °C.

C. Synthesis of 3-Hydroxy-N-alkyl-morphinanes [this is racemorphan]

p-methoxy-phenylacetic acid-(cyclohexenyl-ethyl)-amide

a) To 125 g (1 mol) cyclohexenyl-ethylamine in 300 ml benzene is added dropwise at 0 °C with good stirring within 30 min a solution of 92.3 g (0,5 mol) thionyl chloride in 92 ml benzene. The reaction starts imidiatelly and is finished after 2 h standing at RT. After workup as described for phenylacetic acid-(cyclohexenyl-ethyl)-amide the p-methoxy-phenylacetic acid-(cyclohexenyl-ethyl)-amide melts at 81 - 82 °C, after crystallization from petrolether. There are obtained 123 g, i.e. 90 % of theory.

b) the same product is obtained in virtually theoretical yield when equimolar [better yield and no waste of precious amine!] amounts of cyclohexenyl amine and p-methoxy-phenylacetic acid are refluxed in xylene, until the calculated amount of water has been removed with a trap [you know what I mean].

1-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-isochinoline

273 g (1 mol) p-methoxy-phenylacetic acid-(cyclohexenyl-ethyl)-amide is refluxed in 1000 ml benzene with 307 g POCl3 for 3 h. After cooling the reaction product [?] is poured on ice and the aqueous phase is made alkaline with ca. 30 % NaOH. The precipitated base is extracted with ether and the ethereal solution washed with water and dried. The crude 1-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-isochinoline, which remains after distilling off the ether, is dissolved in the five-fold amount of methanol
and reduced at normal pressure in the presence of 50 g Raney-Nickel.

After the calculated amount of hydrogen has been taken up, the catalyst is filtered off, and the alcoholic solution made slightly acidic with 48 % HBr. Then the methanol is removed in vacuo and the residue recrystalized from water. There are obtained 170 g (50 % of theory) 1-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-isochinoline hydrobromide, mp 198 - 199 °C.

1-(p-methoxybenzyl)-2-methyl-1,2,3,4,5,6,7,8-octahydro-isochinoline

From a solution of 338 g (1 mol) methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-isochinoline in water the free base is precipitated with conc. NaOH and dissolved in 600 ml MeOH. To the solution is added 90 ml of ca. 40 % formaline and this is allowed to stand for 2 h at RT and then reduced after adding 100 g Raney-Nickel. The reduction stops after the calculated amount of hydrogen is taken up. After filtering off the catalyst methanol and water are removed with addition of benzene in vacuo and the residual oil is distilled at high vacuum. The 1-(p-methoxybenzyl)-2-methyl-1,2,3,4,5,6,7,8-octahydro-isochinoline distills at 117 - 119 °C (0.008 mm). The distillate is dissolved in acetone and the calculated amount of oxalic acid is added. There are obtained 235 g of oxalate, which is recrystalized from water or alcohol, mp 163 - 164 °C. Yield 65 % of theory. Hydrochloride mp 149 - 151 °C. The mixed mp with the hydrochloride from Schnider & Grüssner was not depressed.

3-oxy-N-methyl-morphinane

The oxalate is converted to 3-oxy-N-methyl-morphinane with 100 % H3PO4 as described in [1] by Schnider & Grüssner, giving the yield stated there, mp 251 - 253 °C; hydrobromide mp. 192 - 194 °C; tartrate mp 147 °C. From the mother liquor there can be obtained in minute amounts a compound of the same ["bruttoformel" = same percentage of elements, I think this might be the corresponding isomorphinane reported in other literature]. The base melts at 202 - 203 °C, hydrobromide mp 282 - 283 °C.

[The rest is about dioxy-morphinans and ethers of morphinans, this is omitted, as far as I know these are not or not very active.]

refs:

[1] 1. Mitt. O. Schnider & A. Grüssner, Helv. Chim. Acta 32, 821 (1949).
[2] R. Grewe and co-workers, Naturwiss. 33, 333 (1946) [couldn't get this, anybee want to try?]; Z. angew.Chemie, 59, 194 (1947) [I got this one too, interesting read, but NO experimentals]; A. [?] 564, 161 (1949).



Great article, unfortunately some steps (e.g. autoclave) seem to bee somewhat a problem for all but the best-equipped bees. Well let's see what comes from this. I'll post something on morphinane SARs soon. I can tell you that much, there are some that let levorphanol look like light-beer compared to whiskey , at least in animals. Not harder to make, most likely too.

[Pimpo]

This is from

MEDICINAL CHEMISTRY
A Series of Monographs
Volume 5: George DeStevens (Ed.). Analgetics. 1965
ACADEMIC PRESS New York and London
pages 142-143

A book that I cited before (cf.

Post 456856 (missing)

(Pimpo: "2-aminoindane as analgetic", General Discourse)). It is definately on of the best concerning analgesics, which have read!

TABLE III

ANALGETIC ACTIVITY OF N-METHYLMORPHINANS WITH SUBSTITUENTS IN THE AROMATIC PORTION

                         Mouse, subcutaneous (mg/kg)
                         ---------------------------
R              Salt           ED 50       LD 50
----------------------------------------------------
H(a)           H3PO4          11.3        92
3-OH(b)        tartrate       0.5         365
3-OCOCH3(b)    tartrate       2.2         -
3-OCH3(b)      HBr            3.0         260
3-OH, 2-CH3(b) HCl            inactive    200
Morphine       HCl            2.1         576
Codeine        HCl            14.0        270

TABLE IV

ANALGETIC ACTIVITY OF (-)-3-HYDROXY-N-SUBSTITUTED MORPHINANS

                             Mouse, subcutaneous (mg/kg)
                             ---------------------------
N-Substituent          Salt        ED 50      LD 50
--------------------------------------------------------
H                      HBr         60.2       222
CH3                    tartrate    0.5        365
CH2CH3                 HBr         9.5        243
CH2CH2CH3              HBr         inactive   530
CH2(CH2)2CH3           tartrate    1.1        -
CH2(CH2)3CH3           HCl         0.4        158
CH2(CH2)4CH3           HCl         0.5        -
CH2Ph                  HBr         inactive   -
CH2CH2Ph               HBr         0.14       -
CH2COPh                HCl         0.10       > 400
CH2CH(OH)Ph            HCl         0.09       225
CH2(CH2)2Ph            HBr         25.4       > 400
CH2(CH2)3Ph            tartrate    3.6        > 400
CH2CH2-cyclo-C6H11     HCl         2.6        > 750
CH2CH2-4-pyridyl       tartrate    0.09       500
CH2CH2-4-piperidino    H2SO4       inactive   -
CH2CH2-N-morpholino    salicylate  70.1       > 600
CH2CH2-2-furyl         HCl         0.01       250
CH2CH2-2-thienyl       HCl         0.02       502
CH2CH2-Ph-p-NH2        base        0.02       111
CH2CH2-Ph-m-NH2        HCl         0.04       176
CH2CH2-Ph-o-NH2        base        0.7        665
CH2CH2-Ph-p-OH         HCl         0.2        675
CH2CH2-Ph-p-OCH3       tartrate    0.2        > 400
CH2CH2-Ph-p-NO2        HCl         0.04       > 600
CH2CH2-3,4-methylene-  HCl         0.07       > 800
dioxyphenyl
CH2CH2-Ph-p-SCH3       base        0.05       > 600

My conclusion: The 3-hydroxy substituent on the aromatic ring seems to be by far the most active substituent on this portion of the molecule. We can clearly see, that there are a lot of possibilities to make more effective yet less toxic derivatives by changing the N-substituent. I think the synthesis wouldn't be much harder in some intersting cases.

[josef_k]

Wow, this is interesting numbers. That book seems like it might be worth looking up.
I wonder why the O-acetyl version of racemorphan is less potent?

[Pimpo]

josef_k, the book is of course out of date, but on the opiod families known at the time it is really quite extensive and a great source of references.
Concerning the activity of levorphanol compared to its acetyl derivative: I am too rather surprised by this, but one factor that might be important here is that levorphanol is less subject to metabolization than is morphine (compare p.o. and i.v. doses, ratio much smaller in levorphanol than in morphine), so that protection by an acetyl group is not necessary.
Given the existence of highly active (and legal?) derivates it seems like the morphinanes might be even the most interesting group of synthetic opiods, particularly concerning oral activity. The synthesis isn't too easy, but at least for the 3-OH derivatives it doesn't require elevated pressure. One step that looks pretty expensive and possibly dangerous is the reduction of cyclohexenyl acetonitrile. That exotic cobalt catalyst doesn't look very promising and LiAlH4 is the only alternative stated. I might add here that the Helv. Chim. Acta article states (p. 1439): "[..] by the reduction of nitrile XII with sodium and alcohol or with sodium and liquid ammonia cyclohexyl-ethylamine is always formed."

[algebra]

Really nice info, shame that there is no easy alternative,  (like Na/R-OH or a birch if these are easy...) to reduce the nitrile without also completely reducing the ring - why is there always some obstacle like LAH or 100 atm on some obscure hydrogenation catalyst that lurks in the interesting stuff...., arrrgh - frustration,   .

Has anybee ever come across a specific lit-ref for the birch reduction of phenethylamine? Is ethylamine considered electron donating? - this is required of the ring substitution group to get the correct dihydro alignment in a birch. note also that a birch will give two unsaturated bonds and not one as in cyclohexenyl-ethylamine - the extra bond is reduced at a subsequent stage of the levorphanol synth. corrections welcome - working from slightly impaired memory here...

[Assholium]

Ber., 92, p. 649, (1959)
20 g phenethylamin werden in 51 ccm absol Athanol gelost und zu 400-450 ccm flussigen Ammoniak gegeben, die sich in einem offenen Dewar-Gefab befinden. Unter Ruhren fust man inner halb weniger minuten 20 g. Natriumstuckchen hinzu, welche die Grosse von Reiskorhen haben. Nach beendeger Zugabe Lasst man wahrend  mehrer Stunden das Ammoniak aus dem offenen Dewar-Gefab werdampfen, fugt troten waise 250-300 ccm H2O hinzu und ext. mit Ether -> 18 g, Sdp (12) = 80 C.

[algebra]

Assholium, thank you   . can any kind bee read german and provide a translation?  What is the orientation of the dihydro ring? - is the birch reducing the pea to 2(2,5-cyclohexadienyl)) ethylamine or 2-(1,4 cyclohexadienyl) ethylamine\.  Sorry Pimpo did not intend to lead off of the topic,   ...

[Rhodium]

I think the text could be translated as follows:

20g phenethylamine was dissolved in 51 mL anhydrous ethanol and added to 400-450 mL anhydrous liquid ammonia in an open Dewar container. With stirring, 20g rice-sized sodium pieces was added over a couple of minutes. After the addition was finished, the ammonia was allowed to evaporate over a few hours, whereafter 250-300 mL water was added and the product extracted with ether. Yield 18g, bp 80°C/12 mmHg.

[Assholium]

Algebra,

 I check this article (Rudolf Grewe und Hans-Werner Otto, Die Synthese von Hexahydroisochinolinen) and really, they wrote about 2-(1,4-cyclohexadienyl)-ethylamine as product from liq. NH3/Na reduction.
 Of course, this amine can be condensed with 4-MeO-phenylacetic acid to amide and then resulting amide can reduced to octahydroisoquinolune.
 Usual reduction of PEA with Li in liquid ammonia or primary amine (common route described by Berkeser: JACS, 77, 3230, 6042 (1955)) gaves 2-(1-cyclohexeneyl)-ethylamine.

[Pimpo]

First a little typo :

1-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-isochinoline

273 g (1 mol) p-methoxy-phenylacetic acid-(cyclohexenyl-ethyl)-amide is refluxed in 1000 ml benzene with 307 g POCl3 for 3 h. After cooling the reaction product [?] is poured on ice and the aqueous phase is made alkaline with ca. 30 % NaOH. The precipitated base is extracted with ether and the ethereal solution washed with water and dried. The crude 1-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-isochinoline, which remains after distilling off the ether, is dissolved in the five-fold amount of methanol
and reduced at normal pressure in the presence of 50 g Raney-Nickel.

would of course be 3,4,5,6,7,8-hexahydro-isochinoline. This is then reduced to octahydro.


Now the first question. While idly reading here and there in the "Organikum", which is sure well-known to any bee interested in German chem literature, I noticed that it stated the reduction of 1 mol of nitrile to require 1/2 mol LiAlH4 and the reduction of a 1° amide to require 1 mol. The book actually encouraged the reader ("warum?") to think about these ratios and that is what I did. I came up with these equations (including hydrolysis):

R-CONH2 + LiAlH4 + 3 H2O --> RCH2NH2 + LiOH + Al(OH)3 (1 mol required, 2 mol used!)

R-CN + 1/2 LiAlH4 + 2 H2O --> RCH2NH2 + 1/2 LiOH + 1/2 Al(OH)3 (1/2 mol required, 1 mol used!)

So why does the synth use twice the required amount for both the amide and the nitrile? I know that a slight excess of LiAlH4 is commonly used, but so much? I mean, there are no other functional groups that could react with the LiAlH4. Might this be a mistake, or did they want to fool their competitors?

Second question. Could 1-(p-methoxybenzyl)-3,4,5,6,7,8-hexahydro-isochinoline be reduced to octahydro with LiAlH4? It is an imine and I see no groups interfering.

Third question. This is only directed to those who have read the synthesis thoroughly. Could p-methoxy-phenylacetic acid-(cyclohexenyl-ethyl)-amide be condensed to 1-(p-methoxybenzyl)-3,4,5,6,7,8-hexahydro-isochinoline with P2O5 instead of POCl3 just like the plain phenylacetic acid-(cyclohexenyl-ethyl)-amide to 1-benzyl-3,4,5,6,7,8-hexahydro-isochinoline? Might SOCl2 possibly be used instead of POCl3?

Answers to these questions would really bee helpful!

Thanks, Pimpo

P.S. algebra, I think we shouldn't give up on this one. Looks so damn tasty ! I just read about Urushibara catalysts, might they be applied here? The U-Co looks suitable for the nitrile! Really have to learn more about those buggers.

[slappy]

2-Cyclohexylethylamine can also be made from phenethylamine (PEA) by hydrogenating with PtO2 or Pt black in AcOH at 20atm H2.

[Assholium]

2-Cyclohexylethylamine can also be made from phenethylamine (PEA) by hydrogenating with PtO2 or Pt black in AcOH at 20atm H2
  
2-Cyclohexylethylamine is almost useless here, for this rxn (Bishler-Niepiralski cyclization with formation 3,4,5,6,7,8-hexahydroisoquinoline) we need one or two double bonds in cyclohexane ring.

[Assholium]

Patent US4194044

: Process for preparing 3-phenoxy-morphinans

EXAMPLE 6
N-[2-(1-Cyclohexen-1-yl)-ethyl)-4-phenoxyphenyl acetamide
A mixture of 1.26 g (0.01 mol) of cyclohexen-(1)-yl-ethylamine and 1.9 g of 4-phenoxyphenylacetic acid in 20 ml of xylene was heated at reflux for 16 hrs with removal of the water by means of a Dean-Stark apparatus. After evaporation of the solvent, the crude product was crystallized from cyclohexane-hexane mixture to give 2.6 g (76%) of pure N-[2-(l-cyclohexen-l-yl)-ethyl]-4-phenoxyphenyl acetamide, mp 74°-75".

EXAMPLE 7
(±)-1-(p-Phenoxybenzyl)-1,2,3,4,5,6,7,8-octahydroisoquinoline
A mixture of 2.5 g (0.01 mol) N-[2-(l-cyclohexen-l-yl)ethyl]-4-phenoxyphenyl acetamide, 5.0 ml of POCl3 and 20 ml of toluene was heated at reflux under nitrogen for 2 hrs. After evaporation of the solvent and excess reagent there was obtained (±)-1-(p-phenoxybenzyl)-3,4,5,6,7,8-hexahydroisoquinoline as residue. The residue was triturated with pet ether (2 * 10 ml). The pet ether insoluble hexahydro compound was dissolved in methanol and immediately reduced by portionwise addition of 1.0 g of sodium borohydride. The reaction mixture was stirred at room temperature for 4 hrs. The methanol was distilled off and the residue was partitioned between ether (60 ml) and dilute aqueous ammonium hydroxide. The ether solution was washed with water, dried and evaporated to give 1.6 g (69%) of crude (±)-l-(p-phenoxybenzyl)-l,2,3,4,5,6,7,8-octahydroisoquinoline.
For purification, 1.5 g (0.005 mole) of the above crude (±)-l-(p-phenoxybenzyl)-l,2,3,4,5,6,7,8-octahy-droisoquinoline in 5 ml of acetone was treated with 0.42 g of oxalic acid and allowed to crystallize. The crude oxalate was recrystallized from ethanol (30 ml) to give 1.61 g (84%) of pure (±)-1-(p-phenoxybenzyl)-1,2,3,4,5,6,7,8-octahydroisoquinoline oxalate, mp 176°-178°.

EXAMPLE 5
(±)-1-(p-Phenoxybenzyl)-2-formyl-l,2,3,4,5,6,7,8-octahydroisoquinoline
To a solution of 0.78 g (0.002 mol) of (±)-1-(p-phenoxybenzyl)-1,2,3,4,5,6,7,8-octahydroisoquinoline in 5 ml of chloroform was added 0.42 g of chloral in 1 ml of chloroform dropwise. After the mixture had been stirred at room temperature for 3 hrs it was diluted with IS ml of chloroform and washed with 4 N hydrochloric acid (10 ml) and water (10 ml). After drying, the solvent was removed under reduced pressure to give 0.81 g (96%) of crude (±)-1-(p-phenoxybenzyl)-2-formyl-1,2,3,4,5,6,7,8-octahydroisoquinoline. For characterization, a sample of this compound was distilled, bp 225°-227° (0.08 mm Hq).

EXAMPLE 8
(±)-3-Phenoxy-N-formylmorphinan
(±)-1-(p-Phenoxybenzyl)-2-formyl-l,2,3,4,5,6,7,8-octahydroisoquinoline, 24.5 g (0.07 mol), was combined under stirring with 120 g of phosphoric acid which had been mixed with 1.0 g of concentrated sulfuric acid and the mixture was heated in a nitrogen atmosphere to 70°. The resulting homogeneous solution was kept at 70° for 48 hrs. The mixture was cooled in an ice bath and ice-water was added. The resulting suspension was extracted with ethyl acetate (100 ml). The ethyl acetate solution was washed with water and dried. Removal of the solvent gave 23.3 g (95%) of crude (±)-3-phenoxy-N-formylmorphinan.
For characterization a sample of the morphinan was isolated from the above mixture by preparative tlc on silica gel (eluted with ethyl acetate) and distilled, bp 210°-220°(0.1 mm Hg).

EXAMPLE 9
(± )-3-Phenoxy-N-methylmorphinan
To a suspension of 1.1 g of lithium aluminum hydride in 120 ml of anhydrous tetrahydrofuran, 10.8 g (0.03 mol) of crude (±)-3-phenyl-N-formylmorphinan in 50 ml of tetrahydrofuran was added dropwise. After the mixture had been refluxed for 3 hrs, it was cooled to room temperature and ethyl acetate followed by water were added dropwise. The resulting suspension was dried, filtered and the filtrate concentrated in vacuo to give a dark brown oil which was distilled, bp 170°-180° (0.05 mm Hg) to afford 7.6 g (73%) of crude (±)-3-phenoxy-N-methylmorphinan.
This crude morphinan, 7.6 g (0.02 mole) on treatment with oxalic add (2.4 g) in ether (40 ml) afforded 6.8 g of crude oxalate mp 110°-120°. Several recrystallizations from acetonitrile gave 3.6 g (37%) of pure (±)-3-phenoxy-N-methylmorphinan oxalate, mp 144°-146°.

[Megatherium]

All these refs. are by Grewe, which is kind of a pioneer in the morphinan synthesis  :
Chem. Ber. (1948) vol 48 p 289
Ann. (1949) vol 564 p 161
Angew. Chem. (1947) vol 59 p 194

And, J. Chem. Rev. (1977) p 77 contains a detailed discussion of the Grewe cyclization.

[Rhodium]

Your last ref was a bit off, but I found the review anyway:

Benzomorphans: synthesis, stereochemistry reactions, and spectroscopic characterizations
David C. Palmer and Michael J. Strauss

Chem. Rev. 77, 1-42 (1977)

(https://www.thevespiary.org/rhodium/Rhodium/pdf/benzomorphan.review.pdf)

Abstract
The problems of prolonged physical pain and mental anguish, the efficacy of opiates in ameliorating these experiences, and the consequent difficulties of physical dependence and addiction associated with opiate use have been the subject of extensive study by chemists, biochemists, pharmacologists, and physicians for many decades. The problems of opiate addiction in the United States and the associated social trauma have served, in part, to intensify efforts directed toward understanding the basic mechanisms of the action of these drugs and to intensify the search for a better analgesic which has no harmful side effects and does not induce physical dependence. The detailed mechanisms of action are still in large part unknown and the ideal analgesic has not yet been found, but substantial progress is being made.

[Pimpo]

First, a late thank you to Assholium, that patent is instructive !

Interest in this subject seems to have declined recently, but I am still following this up, I found the following in Helv. Chim. Acta 34 (1951), p. 2211-2217, '271. Oxy-morphinane' by O. Schnider and A. Grüssner:

The article also mentions the preparation of (-)-3-hydroxy-N-allyl-morphinan - this is Levallorphan, isn't it? -, which cannot easily be prepared from the racemate. I will omit this.

The resolution of racemorphan is described on p. 2213, here my translation:

(-) and (+) 3-Hydroxy-N-methyl-morphinane

22.8 g Racemic 3-hydroxy-N-methyl-morphinane is dissolved with 13.2 g D-tartaric acid in 300 ml water and the filtrated solution is allowed to stand at RT for 24 h after adding seed crystals of the levorotary D-tartaric salt [i.e. the corresponding salt of the title compound!?]. The crystals are filtered off with suction and washed twice with 5 ml water and then dried. Yield 12.1 g.

To the mother liquor is added dextrorotary D-tartaric salt and it is stirred for 4 - 5 h, the crystalline mass is filtered off with suction and washed twice with 5 ml water and then dried. Yield 10.2 g.

By concentrating, adding seed crystals and letting stand practically quantitative yields of the (-) and (+) tartrate are obtained.



The article has photos (microscope) of the (-) and (+) tartrates:

(-) looks like rectangular plates
(+) looks like very fine needles that are tied together in the middle in bunches. Kind of like a bursting star or so.

P.S.: Erratum: The article reported earlier by me was found in Helv. Chim. Acta 33 (1950) not in 23 (1950), which of course doesn't exist.