Author Topic: OTC piperidone --> fentanyl  (Read 2116 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
OTC piperidone --> fentanyl
« on: December 05, 2003, 09:59:00 PM »
OK bees, I'm gonna keep it short.

Aspartame --hydrolysis--> phenylalanine
phenylalanine --decarboxylation--> phenethylamine
phenethylamine + acetone + formaldehyde --mannich reaction--> phenethyl piperidone
NPP + aniline --red.alkylation-->  ANPP
ANPP + propionic anhydride --> fentanyl

So you read Drone 342's inspiring posts on OTC fentanyl analogs and thought "that's great!". But you don't like the idea of handling multigram quantities of tropane alkaloids (let alone extracting them), and it seems PPA has gotten rather hard to find these days (maybe the DEA knew what was coming?). Also, the exhaustive methylation and quat replacement weren't working too well. What's a poor drug warrior to do?

First, find a new source of the phenethyl chain, with no N-alkylation. Make it common, and make it uncontrollable.

Done. That would be our fake sugar, and the amino acids it contains. For a long time, hive bees have dreamed of turning phenylalanine into something fun, and from a new and unexpected direction comes their chance. The procedure for decarboxylating tryptophan on rhodium's page should be quite sufficient for the decarboxylation of phenylalanine to phenethylamine.

Second, find a method of making piperidone that doesn't need scary or suspicious chemicals. And I say:


 Phenethylpiperidone from acetone, formaldehyde, & phenethylamine, under gentle contitions--what could be sweeter?

This transformation is called a "double mannich reaction", and is commonly used in synthesizing naturally-occuring alkaloids--a famous, somewhat elaborated variation is how the tropine ring is made. Here are some refs that may be of interest. One of these first papers is actually written by Mannich himself, and deals directly with making piperidones from a primary amine and alkyl ketones. Unfortunately, I can't get them, due to their age and obscurity, but beilstein tells me they contain something of interest to the hive at large, so maybe a kind bee will retrieve and post them.

Arch. Pharm. (Weinheim) 255 269 (1917) 272 : diethyl ketone --> 3,5 dimethyl piperidone
Z. Naturforsh. B 1958 897

Archiv der Pharmazie 1992 625

Die Pharmazie 1998 442

Also, ref. to (these, which I do have)

Organic Reactions I, p.303

Synthesis 1973 p.703

Molecules 1996 p.233

for VERY comprehensive reviews of the Mannich reaction.

More papers with almost-but-not-quite versions of our reaction:

Biol. Pharm. Bull. 2003 188 : They make a piperidone out of two aldehydes and methyl ethyl ketone.

JCS perkin II 2001 2037 : They make a piperidone from pyridinaldehyde and "dialkyloxoglutarate", then perform _another_ double mannich to get a diazabicyclononanone.

So there. Yields on this sort of reaction can be expected to be in the range of 30 to 80 %, with our case being a relatively easy one, due to its low steric hinderance.

Conditions will be room temperature methanol, with a 1:1 phenethylamine to acetone ratio, and an excess of formaldehyde. Add a dash of HCl for flavor, and stir.

Be careful upon tasting ;)

Congratulations, you can now saturate the world heroin market with synthetic smack made from artificial sweetener, formaldehyde, common solvents, and aniline, for about the same trouble as it takes to cook up a batch of p2p meth. Speedballs, anyone? ;D

BTW, if you used MEK in the mannich reaction, you would get 3-methyl phenethylpiperidone, for those who like 3-methyl fentanyls. I don't know what the potency increase/decrease would be, but it might be worthwhile.


  • Guest
« Reply #1 on: December 05, 2003, 11:06:00 PM »
NPP + aniline --red.alkylation-->  ANPP

should bee reductive amination...

nings idea is very appealing, some brave bee should try it out in practice so that all the juicy details can bee posted...

Good work ning, you truly are one-man-show  ;)


  • Guest
« Reply #2 on: December 06, 2003, 12:20:00 AM »
A compliment from a fine bee! ;D brain is in beilstein :-[  Just got to ask it the right questions... :)

I thought that reductive amination was called reductive alkylation (i.e. of the N-variety) when the amine group was big in the case of, say, reducing a N-formyl to an n-methyl group or something like that.

Also, maybee this is an obvious modification, but if it's hard to find aniline, perhaps red. aminating the piperidone with ammonia, then mixing in bromo- or iodo-benzene would work. Alternately, nitration of benzene or perhaps toluene, then reduction with iron or zinc powder would do... 

I don't know much of anything about the SAR of fentanyls, but I would really appreciate it if rhodium or someone could say whether a hydroxy, methoxy, or methyl group ortho or para to the amine on the "top" phenyl group of the fentanyl would reduce its' effectiveness.

If not, we can use (my favorite) acetaminophen instead of aniline for the reductive amination, and it really will be OTC, in the narrowest sense of the word, since the propionic acid for the anhydride can bee made by haloform reaction on MEK.

I really wish to hear of some bee doing this experiment too. There's so much possible money in it(as well as chances to get offed), someone almost has to. Of course, whether they decided to tell the hive about it is a different matter...greed sucks :(


  • Guest
Another possibility in order to obtain a more...
« Reply #3 on: December 06, 2003, 12:46:00 AM »
Another possibility in order to obtain a more powerful product would bee to use good ol' amphetamine as precursor for N-(2-phenylpropyl)-piperidinone, wich could following usual path yield alpha-methyl-fentanyl...

Coule bee a nice "from-scratch" route, especially considering this route to p2p:

Post 464635 (missing)

(Lego: "P2P from bromobenzene/CuI/Acetylacetone", Methods Discourse)


  • Guest
Would work...
« Reply #4 on: December 06, 2003, 06:40:00 AM »
Seems like it would double the work, though. How much potency increase could one expect from adding the alpha-methyl?

An idea just hit. If you're looking for power, well...ning is not sure if it would work, but this mod just suggested itself. There should be power aplenty in...

the OTC Carfentanil modification!

Take your aniline, and stir it with propionic acid. If I understand correctly, when dehydrated, the imine will spontaneously steal hydrogen from the OH group, changing it back into a ketone. This step must bee done first, so 2 out of 3 slots on the aniline's nitrogen are occupied. Of course, any more conventional method of N-acylation will work as well.

Next, add HCl to this to make it a salt. Can you see where we are going yet? This time, do not reductively aminate. Add the piperidone and instead, dehydrate gently, till it's all turned into a quaternary imine salt. Now, what is this? It seems to bee highly reactive, and could perhaps even bee called a mannich base...?

Now, add formaldehyde. Ning has NO idea if this will work, but has a funny feeling that given the high reactivity of formaldehyde, ANOTHER mannich-type reaction will occur, giving a -CH2OH group hanging off the piperidone ring, right underneath the top nitrogen. Right where it should be. Overalkylation is prevented, as there's just no more free slots to go around, and as this reaction effectively "eats" the imine double bond, there is no need to reduce anything!

Use KMnO4 or something to fully oxidize this cute little alcohol to a carboxylic acid, and esterify it with a 'lil bit of methanol and H2SO4. Purify, and treat with extreme care. Don't sneeze! ;D

Carfentanil, OTC, with no reductive amination required. Maybe. Possibly. Perhaps. Very speculative, ning thanks the hive collective for its patience. Will study up some more and report back with findings.

Ning out.


  • Guest
Sorry, my last post is somewhat misinformed.
« Reply #5 on: December 06, 2003, 09:28:00 PM »
Sorry, my last post is somewhat misinformed. However, it seems the general idea is on target. What you do is this:

react the piperidone with nitromethane, then add the aniline. They will eliminate one water and condense.

Now there are two possibilities here, and I'm going to bet both of them happen simultaneously:
1. the addition product dehydrates to a nitrostyrene (sound familiar, bees?), which then undergoes michael addition to the aniline, or
2. the addition product dehydrates with the aniline directly

Both of them give the same product. (ning knows nothing about stereochemistry)

Now for a little more hocus-pocus. Using basic KMnO4, the dangling nitromethyl is oxidized to the proper carboxylic acid form, and esterified with methanol. The propionic anhydride is added to the aniline's amine, and you have carfentanil.This will work.

Better yet, though I can't find it anywhere else (yet), my book said that boiling aqueous HCl will hydrate and remove the nitro as hydroxylamine HCl, leaving you with the desired carboxylic acid. This is apparently how hydroxylamine is made commercially. This would be even better than permanganate, IF it doesn't hydrolyze off the aniline too. Will it? I have no idea.

The coolest extension of this idea is this, though--when you rip off the nitro that way, it consumes one H2O. When you esterify the acid with methanol, it produces one H2O. Are you thinking what I'm thinking? A heterogroup transesterification!Maybe boiling methanolic HCl will directly exchange a nitro for a methyl carboxylate thingy?

Well, enough ranting. Here are some refs.

Nitro to carboxylate, KMnO4, 81% yield: TL 1999 4449

cyclic ketone to nitrostyrene with nitromethane, 71% : T 1996 9275
JCS 1934 608
Syth.Comm. 2000 2071
Monatsh. Chem. 2000 949

Finally, direct dehydration attachments

JCS 1948 951
Helv. Chim. Acta 1953 49
Bull.Soc.Chim.Fr. 1944, 41

In one of these, they directly react nitromethane, ketone, and secondary amine! I don't know what the yield is, though.

Guys, it doesn't get much easier than this! OTC Carfentanil! No reductions necesary! Can it bee for real? Damn, I love chemistry sometimes!


  • Guest
simple simplifications
« Reply #6 on: December 06, 2003, 09:43:00 PM »
react the piperidone with nitromethane, then add the aniline. They will eliminate one water and condense.

Now there are two possibilities here, and I'm going to bet both of them happen simultaneously:
1. the addition product dehydrates to a nitrostyrene (sound familiar, bees?), which then undergoes michael addition to the aniline, or
2. the addition product dehydrates with the aniline directly

Do you realize how much you are simplifying things here...?

For example, in the paper you quote - Tetrahedron 52(27), 9275-9288 (1996) - the above is a four-step procedure...


  • Guest
yeah, still have to read some more
« Reply #7 on: December 06, 2003, 10:30:00 PM »
I guess my meaning is that there an lots of mini-routes to go about doing the attachment. My personal feeling at this moment is that the dehydration to nitrostyrene followed by michael reaction with aniline would probably work best. (fewest steps, probably high-yielding) Also, the hive has a lot of collective expertise in the area of nitrostyrenes. I'll research this a bit more.

Sorry if I'm jumping the gun a bit here, but there seems to bee no really amazing, novel chem here, just a bunch of well known steps in a new order. Of course I don't expect a one pot synth from windshield wiper fluid, milk, and chewing gum, or something like that. I'm just proud of realizing the ring could bee monkeyed without grignards or reduction... ;D


Wait a minute! What 4 steps? You mean Scheme 1 on pg. 9276? Read it again, and you will see that we are only following that path from (3) to (4), the Knoevanagel condensation part. The rest has no relavance to this synthesis.

cyclohexanone to 1-nitromethyl cyclohexene-- 80%
cyclopetanone to nitromethycyclopentene --70%
other non cyclics had lower yields.

That is one step!


  • Guest
still a lot of variables
« Reply #8 on: December 07, 2003, 02:13:00 AM »
You are right about the one-step thing, it was just a little hard to follow your line of thought...

Thus, the Knoevenagel condensation between ketones 3a-h and nitromethane gave the unsaturated nitro compounds 4, which may have allylic nature (entries a,c,g,h), vinylic (entry d) or a mixture of both (entries b, d, e, f).

So, if 4-piperidone behaves as a cyclohexanone, you'll end up with 1-nitromethyl-cyclohexene. Will allylic nitroalkenes undergo michael addition (note that 1-nitromethyl-cyclohexene isn't a conjugated nitroalkene)? And, if it will, on what carbon will the aniline attach? in the 3-position or the 4-position of the cyclohexene?


  • Guest
Pathway Illustrated
« Reply #9 on: December 08, 2003, 06:11:00 PM »
Well, got a lot of junk for y'all, but been having trouble uploading. Some good refs and info, dealing with the matter at hand. 4 pages of it, in fact, so stay tuned. But for now, satisfy your hunger with this illustration.

The route, illustrated for your viewing pleasure.

aspartame hydrolyzes to phenylalanine


aspartame ("OC(=O)CC(N)C(=O)NC(C(=O)OC)Cc1ccccc1>>NC(C(=O)O)Cc1ccccc1")

phenylalanine decarboxylates to phenethylamine


phenylalanine ("NC(C(=O)O)Cc1ccccc1>>NCCc3ccccc3")

phenethylamine cyclizes with acetone and formaldehyde to phenethylpiperidone, via Mannich reaction


phenethylamine,acetone,formaldehyde ("N(CCc3ccccc3).CC(=O)C.C=O.C=O>>c2(CCN(CCc3ccccc3)CC2)=O")

phenethylpiperidone is condensed with nitromethane to give a nitrostyrene


to nitrostyrene ("c2(CCN(CCc3ccccc3)CC2)=O.CN(=O)=O>>c2(CCN(CCc3ccccc3)CC2)=CN(=o)=o")

which is then attached to aniline via Michael addition


conjugate addition ("C2(CCN(CCc3ccccc3)CC2)=CN(=o)=o.Nc4ccccc4>>c1ccccc1NC2(CCN(CCc3ccccc3)CC2)CN(=o)=o")

The nitro group is then either oxidized or hydrolyzed to give a carboxylic acid


almost there ("c1ccccc1NC2(CCN(CCc3ccccc3)CC2)CN(=o)=o>>N(c1ccccc1)c2(CCN(CCc3ccccc3)CC2)C(=O)O")

Which is then esterified with methanol


closer... ("N(c1ccccc1)c2(CCN(CCc3ccccc3)CC2)C(=O)O>>N(c1ccccc1)c2(CCN(CCc3ccccc3)CC2)C(=O)OC")

And finally, propionic anhydride is added


carfentanyl ("N(c1ccccc1)c2(CCN(CCc3ccccc3)CC2)C(=O)OC>>CCC(=O)N(c1ccccc1)c2(CCN(CCc3ccccc3)CC2)C(=O)OC")

And there it is.

By the way, Rhodium: I have a really cool thesis that might answer some of your questions. Shortly the relevant parts will bee posted here.  :)


  • Guest
Check this out:
« Reply #10 on: December 09, 2003, 12:30:00 AM »
It's an article on various types of acrylate polymers. Look near the bottom for the part on Michael Addition Cure-In-Place Technology:,1846,1093,00.html

Looks almost exactly like our system. I like how the formation of tertiary amines is slower than that of secondary ones.

here is a clip:

"Acrylic esters can also be used as reactive diluents and modifiers in other cure-in-place coating technologies, such as two-component epoxy systems cured with polyamines. In these systems, the amine-curing agent is more of a co-reactant than an initiator or catalyst. Acrylate esters can react with an amine through a Michael Addition reaction, as illustrated in Figure 11a"


  • Guest
The Big Post
« Reply #11 on: December 10, 2003, 01:15:00 AM »
Here is the info, as promised:

First, a snippet from my ever-so-useful Chemistry of Organic Compounds, 1951, regarding the formation of the piperidone ring via Mannich reaction(Step 2a):
Pg. 705
    The beta-amino ketones are sufficiently stable to be isolated in the free state. They may be prepared by the addition of ammonia or primary or secondary amines to alpha,beta-unsaturated ketones (pg. 703).
    Substituted beta-amino ketones can be prepared by the Mannich reaction (p. 494). Thus dialkyl- and alkylarylamino ketones result from the condensation of secondary amines with formaldehyde and ketones.

R.CO.CH2.R' + H2C=O + HN.R2 --> R.CO.CH(R').CH2.N.R2

The amine usually is used in the form of the hydrochloride or the reaction may be carried out in glacial acetic acid. Aromatic aldehydes may be used instead of formaldehyde. to give aryl substituted products. If a primary amine is used instead of a secondary amine and the ketone contains a hydrogen on both alpha-carbon atoms, a cyclic compound results.


piperidone! ("CN.C=O.C=O.CC(=O)C>>CN1CCC(=O)CC1")

Yes! Exactly what I meant to say! And believe it or not, that is the exact graphic in the book! Any doubters? Find a copy for yourself, it's a really useful book.

Now for the condensation of cyclic ketones with nitromethane (Step 3a):

If we condense the nitromethane and piperidone first to a nitrostyrene, then perhaps we can perform a Michael addition to the conjugated double bond.

Synthetic Communications 2000, 2071 : Gel entrapped base catalyzed henry reaction : synthesis of conjugated nitroalkenes


cyclohexanone to nitrostyrene ("O=C1CCCCC1.CN(=O)=O>>C1CCCCC1=CN(=O)=O")

Yield was 56%, not so good, reaction time was 15 minutes, and base catalyst was NaOH gelled with agar. Other papers do better, like the tetrahedron paper mentioned earlier. Conjugated position is preferred for the double bond, because it is the lowest energy state. With molecular sieves or dean-stark trap to take up the water released, yields will surely be better.  As the making of nitrostyrenes is a common operation for bees, there should be lots of expertise around here about how to get high yields, etc.

Then we have the second part, the Michael addition of the aryl amine to the conjugated double bond:


Michael addition to PhN (" C1CC(CCN1C)=CN(=O)=O.Nc1ccccc1>>C1CC(CCN1C)(CN(=O)=O)Nc1ccccc1")

With addition of heat, and maybe a little basic catalyst, the amine adds to the nitrostyrene at the beta-position. In the old Chemistry of Organic Compounds book, it says on p. 703, referring to alpha-beta unsaturated carbonyl compounds:

"Ammonia, which does not form stable addition products with the carbonyl group, gives beta-amino ketones. The addition product of ammonia and mesityl oxide is called diacetonamine.

Me2.C=CH.CO.CH3 + NH3 --> Me2.C(NH2).CH.CO.CH3

Phorone (p. 205) adds two moles of ammonia to give triacetonediamine,  which on heating cyclizes to triacetonamine.


Michael addition I ("CC(C)=CC(=O)C=C(C)C>>CC(C)(N)CC(=O)CC(C)(N)C")


Cylclization II ("CC(C)(N)CC(=O)CC(C)(N)C>>N1C(C)(C)CC(=O)CC1(C)C")

hey, isn't that TEMPO?

Anyway, this is very suggestive, and the addition issue is no problem for us; since we are not using a carbonyl compound, there's no way the aniline will attach onto the nitro group.

JOC 1958, 729 says "ammonia added readily to I to form tris(2-sulfamylethyl)amine"


conj. alkene adding to ammonia (" C=C-S(=O)(=O)-N.C=C-S(=O)(=O)-N.C=C-S(=O)(=O)-N.N>>N(CC-S(=O)(=O)-N)(CC-S(=O)(=O)-N)CC-S(=O)(=O)-N")

Basically, they just let it sit with ammonia overnight. No yield given, but you get the idea it was quantitative.

JACS 1950, 3298 says "The substances included were all synthesized by reaction of the appropriate amine with methyl acrylate giving excellent yields in most cases"


methyl acrylate adds to primary and secondary amines (" c1ccccc1CNCCC(=O)OC.C=CC(=O)OC>>c1ccccc1CN(CCC(=O)OC)CCC(=O)OC")

this one got 80% yield, but the yields are sort of funny. When the amine was in a saturated heterocyle, yields were from 90 to 100% (!). When the amine was primary, or secondary with a really short second group, the yields sucked (37%, 44%). One factor is that these two compounds were stated to have "intractable hydrochloride salts", indicating that they may have gotten those low yields due to trouble with crystallization.

In McMurry 4th, on page 915 (The Michael Reaction), we can see a table listing Michael acceptors and donors. One of them is nitroethylene. (C=CHNO2) Apparently, nitrostyrene is an even better Michael acceptor than an acrylate, because the nitro group spreads charge better than a carbonyl, giving a stronger acidity to its alpha carbons. The nitro form of a nitroalkane has a pKa around 9, while its aci form has a pKa from 2 to 6. (wow!) This makes it very easy to attack.

Furthermore, since the Michael addition is just a special class of nucleophilic addition to a conjugated bond, let's turn to page 750, where we will see, in the chapter on Conjugate Nucleophilic Addition to a,b-Unsaturated Carbonyl Groups, this:

"Conjugate Addition of Amines
   Primary and secondary amines add to a,b-unsaturated carbonyl compounds to yield beta-amino ketones and aldehydes. Reaction occurs rapidly under mild conditions, and yields are good. Note that, if only one equivalent of amine is used, the conjugate addition product is obtained to the complete exclusion of the direct addition product."


Conjugate addn. I ("CC(=O)C=C.CCNCC>>CC(=O)CCN(CC)CC")

3-Buten-2-one + Diethylamine --EtOH--> 4-N,N-Diethylamino-2-butanone (92%)


Conjugate addn. II ("O=C1C=CCCC1.CN>>O=C1CC(CCC1)NC")

2-Cyclohexanone + Methylamine --EtOH--> 3-(N-methylamino)cyclohexanone

Note that the second one is cyclic, like our carfentanyl nitrostyrene.

Also, while browsing the net, I found this in a thesis by James K Murray Jr, for Drexel University.
It is


( I quote, from page 135:

"A one-pot, three step sequence of nitroaldol formation, dehydration, and Michael addition is often used to avoid potential problems with the nitroalkenes, which are often prone to polymerization or decomposition. As with the nitroaldol reaction, Michael additions of nitroalkenes and nitroalkane anions find their major synthetic utility in the numerous transformations that the Michael adducts can undergo, primarily involving the nitro group. Various oxygen, sulfur, nitrogen, and phosphorus nucleophiles can be employed in conjugate additions to nitroalkenes (eqn. 2.31).


eqn.2.31 ("CC(C)=C(C)N(=O)=O.N(C)C>>CC(C)(C(C)N(=O)=O)N(C)C")

These reactions are operationally simple: the nitroalkene and nucleophile are mixed in the presence of base and the conjugate addition products are formed. (...) The same bases that are typically employed in the nitroaldol reaction, alkali metal hydroxides or alkoxides and tertiary amines, are also suitable for Michael additions to nitroalkenes."

However, (pg. 140)
"The direct addition of nitrogen nucleophiles to nitroalkenes is not often a synthetically useful reaction, since the reaction tends to be reversible favoring the nitroalkene. To incorporate nitrogen nucleophiles, an addition/elimination sequence is generally used, as shown in Equation 2.41 for the preparation of 361.


eqn 2.41 ("c1ccccc1SC=CN(=O)=O.N2CCCC2>>C2CCCN2C=CN(=O)=O")

THF, RT 2hr"

I don't know what he is talking about, since in his figure, a conjugate addition is clearly not being performed. At the very least, this is very confusing language, especially in light of the VERY NEXT FIGURE:

"Chiral nitrogen nucleophiles, common sources of asymmetric induction, can be added to nitroalkenes, followed by rapid reduction of the nitro group with samarium(II) iodide, as a simple method for the synthesis of non-racemic 1,2-diamines such as 364 (Equation 2.42)


eqn 2.42 (" O=N(=O)C1=CCCCC1.OCC2CCCN2>>OCC3CCCN3C4CCCCC4N(=O)=O")

CH2Cl2, RT 30 min, 95%
And the reduction is in a different step. Now it seems very unlikely that that little OH makes all the difference between the reaction not working at all, and giving almost quantitative yield in 30 minutes at room temperature, so I'm going to assume that, for some reason, in the first case he was trying to add without disturbing the double bond, and in the second case he was doing a direct addition.

So that's what I have on the condensation and conjugate addition steps, and though not overwhelmingly convincing, it's very encouraging.

Finally, here's another snip from my venerable Chem. of Org. Cmpds. book on obviating the permanganate oxidation of nitro-ANPP to carfentanyl-minus-methyl-ester. I think it's called the nef reaction:

Pg. 256
5. Acid Hydrolysis.
(a) Hydrolysis of the nitro form of a primary nitro compound.
    When primary nitro compounds are boiled with concentrated aqueous hydrochloric acid, carboxylic acids and hydroxylamine hydrochloride are formed.

R.CH2.NO2 + HCl + H2O --> R.COOH + HONH3.Cl

By this reaction, which involves an oxidation of the methylene group and reduction of the nitro group, carboxylic acids may be prepared from hydrocarbons. The price of hydroxylamine, which was produced by the reduction of nitrous acid and isolated by way of acetoxime (p.207), has been reduced greatly because of the above process.
(b) Hydrolysis of the aci form of a primary or secondary nitro compound.
    If a primary or secondary nitro compound first is converted to the salt of the aci form by alkali and then hydrolyzed by 25% sulfuric acid, aldehydes and ketones are produced with the evolution of nitrous oxide (Nef reaction, p. 354).

2 R2.C=NOONa + 2 H2SO4 --> 2 R2.C=O + N2O +2 NaHSO4 + H2O

Well, that's how it stands right now. After all this searching and reading and printing, I am further convinced that the proposed reaction is both reasonable and will work well in practice. Hope this long, boring drone will be of some use to inquiring minds. :)  Wherever did our dear Drone go off to? :(


  • Guest
Some modifications
« Reply #12 on: December 16, 2003, 12:41:00 AM »
This one goes out to the silent masses out there. Yes, you know who you are. Ning, having been faced with some harsh, but true, criticism of his eeevil opioid synth, is back with revisions to make the impossible...practical. We hope.
First the route:

Reflux the nitrostyrene obtained by knoevenagel condensation of nitromethane and piperidone in concentrated aqueous HBr, to give a bromo-carboxylic acid. It would be a lachrymator, but it surely is too heavy to evaporate well.


The first magic... ("c1ccccc1CC-N1CCC(CC1)=CN(=O)=O>>c1ccccc1CC-N1CCC(CC1)(Br)-C(=O)O")

JCS Perkin I 1981, 2520
JCS 1911, 1514
JCS 1931, 952
And best of all, (if I could get it!!!)
Collect. Czech. Chem. Commun. 1983, 2952

Then a small esterification step, MeOH/H2SO4:


Methylation ("c1ccccc1CC-N1CCC(CC1)(Br)-C(=O)O.OC>>c1ccccc1CC-N1CCC(CC1)(Br)-C(=O)OC")

Now, you can attach the ANILIDE you made previously, without fear of hydrolysis! Concentrated methanolic NaOH comes to mind, perhaps with a bit of PTC thrown in for good measure.


The second magic... ("c1ccccc1CC-N1CCC(CC1)(Br)-C(=O)OC.c1ccccc1NC(=O)CC>>c1ccccc1CC-N1CCC(CC1)(N(c1ccccc1)C(=O)CC)-C(=O)OC")

To see this done, see JMC 1970, 559 where yield was 70-94% for an iodide attaching to a substituted acetanilide.
Justus Leibig's Annalen 1889, 321
JCS 1947, 1486
Russian Journal of Organic Chemistry 1998, 494 (528)

(actually, the refs for the previous 2 steps might bee reversed. Sorry if this causes trouble, will edit as soon as I'm sure)

So, why do we do this in such a backwards manner?
So we can dispense with the propionic anhydride:


No propionic anhydride! ("c1ccccc1N.CCC(=O)O>>c1ccccc1NC(=O)CC")

I have refs for this too, but not right now. It's relatively easy, you just reflux aniline and propionic acid. Water is evolved. If you like, you can drive it all the way to completion by using a column set to 100 C at the top.

And finally, how to get aniline, OTC.


KMnO4 oxidation of toluene ("c1ccccc1C>>c1ccccc1C(=O)O")


Ammonium Benzoate heated > 100 C ("c1ccccc1C(=O)[O-].[N+]>>c1ccccc1C(=O)N")


NaOCl hoffmann rearrangement ("c1ccccc1C(=O)N>>c1ccccc1N")

The permanganate oxidation of toluene is covered well. If refs wanted, I'll get some.

The heating dehydration of carboxylic acid ammonium salt to give amide is also well covered. In fact, this synth relies on it to give the anilide as well.

See Org. Syn CV2, pg. 44 "4-aminoveratrole", where they do exactly that hoffmann rearrangement of amide to an aniline, with NaOH/NaOCl.  Yield 80%

What this means is that from toluene, or from benzoic acid/sodium benzoate (common food preservatives) and household ammonia, you can make aniline.
From methyl ethyl ketone and bleach, you can make propionic acid.
From aniline and propionic acid, you can have propionanilide. And with the piperidone thoroughly beaten on in previous posts, from fake sugar, formaldehyde and acetone, plus a little methanol, racing fuel and NaBr from the pool supply, you're set with carfentanil. By performing the addition/hydrolysis simultaneously and before the addition, we should be able to dodge the "delicate amide doesn't like acid" issue.

Ning feels like the gods and demons of drug chemistry want this synth to happen--there's always another way around the obstacles nature and man throw up before us. That's part of the thrill, isn't it?

Thanks guys, for teaching ning about "hoffmann rearrangement" :)


  • Guest
The second magic...
« Reply #13 on: December 17, 2003, 12:17:00 AM »
The 'second magic' seems to bee to good to bee true.

N-Alkylation of amides is difficult. The paper you quote (J. Med. Chem., 1970, 13(3), 599-601) uses simple alkyl- or benzyliodides in 50% molar excess.

This leads to several problems:
1) Using your alkylbromide in excess would drastically lower the overall yield
2) Iodides are much better leaving groups than bromides
3) To N-alkylate an amide one have to use strong bases (in the paper it is KOH in acetone). This will cleave your methyl ester at least to some percentage.
4) Using strong bases in the presence of alkylbromides will lead to some to extend to the elimination product.


  • Guest
Right, of course
« Reply #14 on: December 17, 2003, 11:37:00 PM »
I know it's hard, and maybe not even practical, but thought was worth a try. Ultimately, it might just bee necesary to use propionic anhydride. But I wanted to do a literature survey to see what there was to come up with.

More will be posted here as well, promise. :)


  • Guest
OK, some facts to back the hocus-pocus
« Reply #15 on: December 18, 2003, 10:44:00 PM »
carfentanil synth V2 refs:

<F1> : Penultimate assembly, formation of propionanilide


step F1 ("c1ccccc1N.OC(=O)CC>>c1ccccc1NC(=O)CC")

Journal of the American Chemical Society 1940, 3523. New Compounds.


The propionamides were obtained by refluxing the amine and propionic acid for several hours.

Journal of the Chemical Society 1898, 33. Decomposition of Camphoric Acid by Fusion with Potash or Soda.

The fraction 135-150 C contains propionic acid, CH3.CH2.COOH.--
After repeated very careful fractionation with a column, an acid was isolated which distilled constantly at 140 C, and was evidently propionic acid. The anilide of this acid was prepared by heating it with aniline and recrystallizing the product from light petroleum (bp 100-120); the white plates thus obtained were almost insoluble in water, more soluble in light petroleum, and very soluble in alcohol and ether. The substance melted at 103-104 C. The melting point of propionanilide as given by Sestini (Zeitschrift fur Chemie, 1871, 35) is 92 C, and by Kelbe (Ber., 1883, 16, 1200) as 105 C. A specimen of the anilide which we prepared from pure propionic acid and pure aniline melted at 103-104 C, and was identical with the anilide obtained by us as described above.

Journal of the Chemical Society 1908, 1033. Melting Points of the Anilides, etc. of Fatty Acids.

Propionic acid--
Amide: melts 79
Anilide: melts 105
p-toluidide: melts 123

The anilides, p-toluidides, and a-napthalides described in the present paper were prepared by the following method. A mixture of 1 to 3 grams of the fatty acid and the equivalent amount of amine were heated in a sealed tube to 160-190 C for eight to twelve hours. In no case was any pressure observed in the tube after cooling. The product of the reaction, generally a solid or an oil which soon solidified, was purified by recrystallization from aqueous alcohol, or, in the case of the derivatives of the higher fatty acids, absolute alcohol. In certain cases, when the product tended to be oily, it was found advisable to first spread the mass on a porous tile and leave for 24 hours before recrystallization. The yields varied between 30 and 80 per cent, depending largely on the degree of purity of the fatty acid employed. The anilides and p-toluidides were pure white or faintly yellow; the a-napthalides generally had a pink color, not unlike that which a-napthylamine acquires on contact with the air. On boiling in alcoholic solution with animal charcoal, all traces of pink colour could be removed without, however, affecting the melting point. When pure, the a-napthalides were completely free from the objectionable odor of the base.

Merck Index 8th ed. 1968

Aniline melts at -6 and boils at 184.
Propionic acid melts at -21 and boils at 141.

Suggestively, aniline acetate spontaneously converts to acetanilide with age.

<F2> : Final assembly, N-alkylation of propionanilide


Step F2 ("c1ccccc1CCN(CC2)CCC2(Br)C(=O)OC.N(C(=O)CC)c3ccccc3>>c1ccccc1CCN(CC2)CCC2(N(C(=O)CC)c3ccccc3)C(=O)OC")

Journal of Medicinal Chemistry 1970, 539. Felder, Pitre, Fumagalli, Lorenzotti. Radioopaque Contrast Media. XVIII. Derivatives of 2-(3-amino-2,4,6-triiodophenyl)alkanoic Acids.


Yield 87% ("OC(=O)Cc1c(I)cc(I)c(c1I)NC(=O)CCC.ICCCC>>OC(=O)Cc1c(I)cc(I)c(c1I)N(C(=O)CCC)CCCC")


Yield 95% ("OC(=O)Cc1c(I)cc(I)c(c1I)NC(=O)CCC.ICc1ccccc1>>OC(=O)Cc1c(I)cc(I)c(c1I)N(C(=O)CCC)Cc2ccccc2")

2-(3-Alkylacylamino-2,4,6-triiodophenyl)alkanoic Acids:

A solution of 0.045 mol of alkyl iodide in 2.5 ml of acetone was added for 30 minutes to a solution of 0.03 mol of 2-(3-acylamino-2,4,6-triiodophenyl)alkanoic acid and 0.12 mol of KOH in 35 ml of H2O. The mixture was stirred for 4 hr at 35 C and then poured into 200 ml of ice water and extracted twice with ether (30 ml). The crude product, obtained by precipitation with 18% HCl, was purified further by reprecipitation, extraction with boiling ethyl acetate, or recrystallization from a suitable solvent.

Journal of Organic Chemistry 1992, 1864. Kawabata, Minami, Hiyama. Stereoselective Synthesis of b-Lactams by Oxidative Coupling of Dianions of Acyclic Tertiary Amides.


Yield 95% ("COc1ccc(cc1)NC(=O)CCC.BrCC(=O)OC(C)(C)C>>COc1ccc(cc1)N(C(=O)CCC)CC(=O)OC(C)(C)C")

The solid (~9 mmol) dissolved in dichloromethane (20 ml) was treated with tert-butyl bromoacetate (3.08 ml, 18 mmol) and then with benzyltriethylammonium bromide (0.25 g, 0.92 mmol). To the resulting mixture was added 50% aq NaOH (1.0 ml) at 0 C. After being stirred at room temperature for 17 h, the mixture was poured into saturated NH4Cl solution and extracted with ethyl acetate. The organic phase was washed with water, dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by silica gel column chromatography.

<B3> : Hydrolysis and bromination, Nitromethylenepiperidine to 4-halo-piperidylacetic acid


Step B3 ("c1ccccc1CCN(CC2)CCC2=CN(=O)=O>>c1ccccc1CCN(CC2)CCC2(Br)C(=O)O")

Journal of the Chemical Society 1911, 1513. Optically Active Derivatives of 1-methylcyclohexylidene4-acetic acid.


Yield unknown ("CC1CCC(CC1)=CC(=O)O>>CC1CCC(CC1)(Br)CC(=O)O")

4-Bromo-1-methylcyclohexyl acetic acid:
A reaction which resembles the above reduction in that it should convert the centroasymmetric optically active acid into a saturated substance which is potentially optically inactive, is involved in the addition of hydrogen bromide to the 1-methylcylohexylidene-4-acetic acids. When the unsaturated dl-acid is mixed with fuming hydrobromic acid (saturated at 0 C) it dissolves, but in a short time an oil, which rapidly crystallizes, separates on the surface. The crystalline mass is washed with water, left in contact with porous earthenware until quite dry, and then recrystallized from a little formic acid, in which it is very soluble, and from which it separates, usually in plates, but sometimes in hard, brilliant prisms.
    Similar experiments were made on the action of hydrobromic acid on d- and l-1-methylcyclohexylidene-4-acetic acids, and resulted in both cases in the production of an optically inactive bromo-acid identical in all respects with that obtained from the externally compensated unsaturated acid.
   We have previously prepared this substance by the action of fuming hydrobromic acid on 4-hydroxymethylcyclohexyl-4-acetic acid (Trans., 1908, 93, 1082); Wallach (Annalen, 1907, 353, 312) made the same substance by treating the 1-methylcyclohexylidene-D3-acetic acid of Markwald and Meth with hydrobromic acid.

   Finely divided 4-bromo 1-methyl cylohexyl 4-acetic acid dissolves readily in sodium carbonate solution, but the liquid gradually clouds and an oil separates; the change occurs rapidly when the solution is warmed at 40 C. The oil is extracted with ether, the ethereal extract carefully dried and evaporated, and the residue distilled; the whole quantity passes over at 122 C. [...] There can be no doubt that this hydrocarbon is 4-methylene 1-methyl cyclohexane, and identified with the compound which Wallach obtained (Annalen, 1906, 347, 345; 1909, 365, 267) by the slow distillation both of 1-methyl cyclohexylidene 1-acetic acid, and of 1-methyl D3-cyclohexene 1-acetic acid, intramolecular change taking place in the latter case.


  • Guest
And a paper on the subject...
« Reply #16 on: December 18, 2003, 11:19:00 PM »
Wherein they perform simultaneous hydrolysis and chlorination of nitro-alkenes.

Journal of the Chemical Society 1947, 1485. Heath, Rose. Aliphatic Nitro-compounds. Part IV. Reaction of a-Nitro-olefins with Hydrochloric Acid.

   a-Nitro-olefins react with anhydrous hydrogen chloride in ether giving 1:2-dichloronitroso compounds  which rearrange, if and alpha-hydrogen atom is present, into 1:2-dichloro-oximes. These on hydrolysis with water afford hydroxylamine hydrochloride and an a-hydroxy- or a-chloro-carboxylic acid according to the extent of the hydrolysis. Reaction of the nitro-olefins containing an a-hydrogen atom with concentrated aqueous hydrochloric acid gives similar products, but the intermediate dichloro-oximes are not isolated. It is suggested that the initial step is a 1:4 addition of hydrogen chloride to the nitro-olefin.

   The literature on the action of hydrochloric acid on a-nitro-olefins is scanty. Haitinger (Monatsh., 1881, 2, 287; Wien Akad. Ber., 1878, 77, 428; A, 1879, 700) reported that 1-nitro-2-methylprop-1-ene, treated with hydrogen chloride at 20 C or with boiling concentrated hydrochloric acid, gave hydroxylamine hydrochloride, carbon dioxide, and ammonia; a hydroxy-acid, mp. 65 C, was also isolated but not characterized. Our experience, below, suggests that this was probably impure a-hydroxy isobutyric acid (mp. 79-80 C). Priebs (Annalen, 1884, 225, 319) obtained phenylchloroacetic acid by treatment of b-nitrostyrene with fuming hydrochloric acid at 100 C.
   A study of the action of hydrochloric acid on a-nitro-olefins was undertaken as part of a general exploration of the chemistry of the aliphatic nitro-compounds. In the case of nitro-olefins with an a-hydrogen atom (R2C=CH.NO2), treatment with ethereal HCl gave hydroxylamine hydrochloride and an a-chloro or a-hydroxy acid (cf. Priebs, loc. cit).
   The possibility that the first stage in the reaction was 1:2 addition of HCl to the double bond was excluded when it was found that 2-nitro isopropyl chloride, Me.CHCl-CH.NO2, was not affected by ethereal HCl under the conditions which converted 1-nitro prop-1-ene, Me.CH=CH.NO2, into a-chloropropionic acid and hydroxylamine HCl. It therefore follows that the first addition is 1:4 to give R2CCl.C=N(OH)->O, which, as the aci-form of a primary nitro-paraffin, breaks down via the hydroxamic acid to hydroxylamine HCl and an a-chloro or a-hydroxy acid (depending on the lability of the chlorine atom). The mechanism of conversion of the aci-primary nitro-paraffin into hydroxamic acid is still obscure but is considered to involve the following steps (cf. Yale, Chem. Reviews, 1943, 33, 226):

R.CH2.NO2 --> R.CH=N(->O)OH --> R.CHCl.N(->O)HOH --(-H2O)--> R.CHCl.NO --> R.CCl=N.OH R.C(OH)=N.OH --(+H2O)--> R.COOH + NH2OH

   In this mechanism, an important stage is the rearrangement of the 1-chloro nitroso compound to an oxime, which requires a hydrogen atom on the carbon carrying the nitroso group. From secondary nitro-paraffins, in which this hydrogen is not available, the product expected would therefore be a 1-chloro nitroso compound or its breakdown products. In agreement with this we have found that treatment of nitro-olefins of the type RR1C=CR2.NO2 with ethereal HCl gives deep blue highly lachrymatory liquids. Although it has not been possible to obtain these analytically pure on account of the closeness of their boiling points to those of the parent nitro-olefins, there seems little doubt that these are the dichloro nitroso compounds.

RR1C=CR2.NO2 + HCl --> [ RR1CCl.CR2=N(->O)OH] + HCl --> [ RR1CCl.CClR2.NH(->O)OH ] - H2O --> RR1CCl.CClR2.NO

   Attempts to break down these dichloro nitroso compounds by further treatment with HCl gave a trace of hydroxylamine HCl as the only product identified.


- Action of Anhydrous HCl on Nitroethylene.
37 g nitroethylene was added dropwise to 300 cc stirred saturated HCl below 0 C. Stirring was continued at 0 C for 4 hours and at 20 C for a further 12 hours. After removal of the ether under reduced pressure ab-dichloro acetaldoxime remained as a colorless mobile liquid which began to decompose immediately. The crude product was stirred with 300 cc water at 20 C for 16 hr; the mixture, which was homogeneous, still gave a positive test for a hydroxamic acid, and was refluxed for 30 min (test negative) and evaporated to dryness under reduced pressure. The residue was extracted with ether, giving 39 g of  chloracetic acid (82%), mp 62-63 C after recrystallization from chloroform, and 31.5 g hydroxylamine HCl (90%) as the ether-insoluble product.

- Action of Concentrated HCl on 2-nitroethanol
10 g nitroethanol was heated with 80 cc concentrated HCl in a sealed tube at 140 C for 12 h. The solution was evaporated and separated by ether into 6.2 g chloroacetic acid (60%) and 5.3 g hydroxylamine HCl (70%).

- Action of Anhydrous HCl on 1-nitro prop-1-ene
This was carried out using 51 g 1-nitro prop-1-ene and 250 cc ethereal HCl as described above for nitroethylene. The ab-dichloropropaldoxime (80%) was distilled, bp 78-80 C/12 mm., but began to decompose immediately after distillation. A 37 g portion was hydrolyzed by refluxing it in 100 cc water for 30 min. After evaporation the residue was separated by ether into 8.0 g lactic acid (33%), 0.7 g a-chloro propionic acid (2%), and 10.4 g hydroxylamine HCl (60%).

- Action of Concentrated HCl on 1-nitro prop-1-ene and 2-nitro isopropyl alcohol
20 g nitropropene was stirred with 100 cc concentrated HCl at 20 C for 24 h, then at 40 C for 2 h. After working up in the usual way, 12 g hydroxylamine HCl (75%), 6 g lactic acid (30%), and  0.5 g a-chloro propionic acid (2%) were obtained. Similarly, 10.5 g nitroisopropanol heated with conc. HCl in a sealed tube at 140 C for 2 h gave 5.3 g hydroxylamine HCl (75%) and 5.6 g a-chloro propionic acid (54%)

- Action of Anhydrous HCl on 2-nitro prop-1-ene
26.1 g nitropropene was added dropwise with stirring to 200 cc dry ether saturated at 0 C with anhydrous HCl; a deep blue color rapidly developed. Stirring was continued at 0 C  and 1.8 g precipitated hydroxylamine HCl removed.  Separated water was absorbed by the addition of anhydrous MgSO4, and the solution distilled, giving 20 g 1,2-dichloro 2-nitroso propane (45%), bp 48-50 C/60 mm., as a deep blue, powerfully lachrymatory liquid. It was not possible to obtain this material analytically pure owing to difficulties of separating small amounts of unchanged 2-nitro prop-1-ene (bp 45-48 C/60 mm).

- Action of Anhydrous HCl on 2-nitro but-2-ene
20.2 g nitrobutene was treated as described above for nitropropene. 11 g 2,3-dichloro 2-nitroso butane (35%) was obtained as a deep blue lachrymatory liquid. After hydrolysis with water at 20 C for 40 h only 0.1 g hydroxylamine HCl could be identified.

- Action of Concentrated HCl on 2-nitro but-2-ene
20 g nitrobutene was stirred at 20 C for 40 h with 100 cc concentrated HCl. The deep blue color which developed initially disappeared in 30 min. Evaporation to dryness under reduced pressure gave only 4 g hydroxylamine HCl (52%).

- Action of Anhydrous HCl on 1-nitro 2-methyl prop-1-ene
60 g nitroisobutene was treated with 400 cc saturated ethereal HCl as described above for nitroethylene. 75 g ab-dichloro isobutaldoxime (80%) was obtained as a colorless liquid which began to decompose immediately after distillation. Hydrolysis of 24 g of this dichloro-oxime with 250 cc boiling water for 16 h gave 9.5 g hydroxylamine HCl (90%) and 4.8 g a-hydroxybutyric acid (30%). Hydrolysis of the dichloro oxime with concentrated HCl for 1.5 h gave a 50% yield of a-hydroxyisobutyric acid.

- Action of Concentrated HCl on 1-nitro 2-methyl prop-1-ene
20.2 g nitroisobutylene was added dropwise with vigorous stirring to 100 cc concentrated HCl at 15-20 C. Stirring was continued for 16 h, and the mixture then boiled for 15 min and worked up in the usual way, giving 12 g hydroxylamine HCl (86%) and 11.3 g a-hydroxyisobutyric acid (54%).

- Action of Anhydrous HCl on 1-nitro 2-chloro propane
1-nitro 2-chloro propane was treated with anhydrous ethereal HCl under the conditions described for 1-nitro prop-1-ene (see above). On distillation unchanged 1-nitro 2-chloro propane was recovered quantitatively.