Author Topic: A really wet reductive alkylation  (Read 9583 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
A really wet reductive alkylation
« on: July 04, 2002, 03:05:00 PM »
Freifelder stated a long time ago that phenylacetone + aq.methylamine -> N-methylimine which is hydrogenated to N-methylamphetamine. The wet enviroment doesn´t disturb the imine formation. Yet when bees here want to make MDMA they tend to choose anhydrous conditions. In particular in the borohydride reduction of the imine to amine.

I decided to challenge this:

10g 2-fluorophenylacetone (65mmol) was dissolved in 50ml toluene and to this was added a solution made of 10.6g (130mmol) ethylamine HCL and 5.25g NaOH in 50ml water. The mixture was vigorously stirred at room temp for 2 hours. The aqueous phase was then removed and the toluene phase was transferred to a 250ml rb flask containing 1.9g NaBH4 (50mmol), 25ml water and 15ml EtOH and the mixture was vigorously stirred for a further 2 hours at room temp. Diluted hydrochloric acid was then added dropwise until pH2 was reached, the phases was separated and the toluene phase was extracted twice with 20ml 5% HCL and then discarded. The combined aqueous phases was made strongly alkaline with 50% aq NaOH and extracted twice with 50ml toluene. The toluene extracts was dried over MgSO4 and stripped of solvent in a rotovap. The residual yellow oil (nasty smell) was dissolved in 50ml EtOAc and dry HCL in IPA added until pH4 was reached. The white crystals was isolated and dried to constant weight.

Yield 8.9g (40.9mmol,62.9%) N-ethyl-1-(2-fluorophenyl)-2-aminopropane hydrochloride

Longer reaction times and 4-5 eq ethylamine should give better yields.

So many things to do, so little time.... :P


  • Guest
Your detailed writeups are great, Barium!
« Reply #1 on: July 04, 2002, 04:02:00 PM »
I think it has been mentioned earlier by Sunlight that the reaction does not need to be strictly anhydrous when methylamine is used (40% straight from the bottle), but the yield is a little on the low side like yours - around 60%. It is very important to use anhydrous conditions if ammonia is used as the amine though, the unsubstituted imine is just too unstable in an aqueous environment.

Strange substrates you choose - but I assume that you are an intelligent guy who wants to work for our cause by developing methods for us to use without ever having to own a controlled substance, putting yourself at risk, correct?

Entropy just isn't what it used to be.


  • Guest
Yeah, I´m just developing methods.
« Reply #2 on: July 04, 2002, 04:47:00 PM »
Yeah, I´m just developing methods. Which kind of substituted substrates anyone then wants to use is a matter of choice to every bee. ;)
By doing it this way I can work really open ordering everything I need through the major companies without questions.


  • Guest
slight digression
« Reply #3 on: July 04, 2002, 08:53:00 PM »
say for example someone wanted to reductivly aminate Phenylacetylcarbinol to PPA, and used ammonium acetate as opposed to ammonia? Or conversly methylamine (acetic salt)
Would this improve the stability of the shiff's base and inhibit imine polymerization?
Another modifer would be to dry the shiff's base substrate with drying agent prior to reduction, will the shiff's base prepared from ammonim acetate reduce under the aluminum amalgum reduction scheme?
Sorry about all the dumb questions.


  • Guest
reductive amination
« Reply #4 on: July 05, 2002, 05:09:00 AM »
Al/Hg is a very low yielding method for producing primary amines from ketones, not recommended.

I don't see how using ammonium acetate instead of ammonia in the amination would help imine formation, only drying agents like molecular sieves (or other) is really effective here, and even then the yields aren't stellar.

However, ammonium formate and Pd/C will probably make PPA in good yield from l-PAC, see the new paper at

for details. It will perhaps even produce norpseudoephedrine (threo-PPA) selectively.

Entropy just isn't what it used to be.


  • Guest
nice reference
« Reply #5 on: July 05, 2002, 06:10:00 AM »
Thanks that's good article.
It's interesting to note that they speak of this hemiaminal intermediate pathwat as some "new" discovery.
There are pretty old papers detailing this very same intermediate in reductive amination reactions.
Maybe they speak of this specific substrate?


  • Guest
A long night is over....
« Reply #6 on: July 05, 2002, 09:27:00 AM »
I decided to try a number of variations of the previous reaction. Since I have not yet learned to use all the functions of this place properly I cannot upload any documents. So I have to write this one.

    Rxn  A   B   C   D   E
     1   Et   1  0.5  1   52%
     2   Et   2  0.5  1   55%
     3   Et   1  1    1   56%
     4   Et   2  1    3   73%
     5   Me  1  0.5  1   55%
     6   Me  2  0.5  1   56%
     7   Me  1  1    1   49%
     8   Me  2  1    3   78%

A. RNH2 R=
B. Time for imine formation in hours
C. Eq. NaBH4 to substrate
D. Imine reduction time
E. Yield of amine HCl
The substrate was 2-fluorophenylacetone in all reactions.

The volumes of solvents were the same as in the original reaction in this thread, as were the work up.

It seems that the critical part is the imine reduction time.


  • Guest
Manic Street Scientists
« Reply #7 on: July 05, 2002, 01:13:00 PM »
What? You ran 25 reactions yesterday? How many mmol each, and how did you analyze them? HPLC and then something with graph area calculation?

The methylamine aminations, where they performed exactly the same way using MeNH2.HCl?

Entropy just isn't what it used to be.


  • Guest
Not that many..
« Reply #8 on: July 05, 2002, 02:50:00 PM »
it was just 11 reactions altogether. 9 reductive alkylations and two other reductions.

I used 65mol 2-fluorophenylacetone for each reaction. I have not done a HPLC on them yet. The yield is just from the first crystallisation. I´ll recrystallise them later and run a HPLC on all of them.


  • Guest
got P2P?
« Reply #9 on: July 05, 2002, 05:23:00 PM »
Got P2P? ;)   you used 65 MOL?/rxn?  mmol, maybe?


  • Guest
I have always assumed based on what I've been ...
« Reply #10 on: July 05, 2002, 11:39:00 PM »
I have always assumed based on what I've been told about NaBH4 that the reason for using anhydrous conditions with the NaBH4 reductive amination is that if you don't use anhydrous conditions then the extra H2O molecules will destroy the free hydride ions supplied by the NaBH4 before those hydride ions ever have a chance to reduce the respective imine to its corresponding amine, and yields will suffer. 

LaBTop was the one responsible for popularizing this method here as I recall, and he stressed using anhydrous conditions to get the 99% yield that he claimed, so maybe he will have something to add to this thread.


  • Guest
No the problem is reduction of the ketone
« Reply #11 on: July 06, 2002, 01:54:00 AM »
To an alcohol, acidic or basic aqueous solutions will speed degradation of the hydride.
NaBH4 is a strong reducing. agent, that's why NaBHCN is used instead. If the substrate is anhydrous then equilibrium is to  the right on Imine/hemiaminal RXN and the ketone isn't around to be reduced.


  • Guest
More borohydride.
« Reply #12 on: July 06, 2002, 05:19:00 AM »
C. Eq. NaBH4 to substrate
I am guessing that you really are specifying the molar ratio not an equivalence ratio.

If you would use more than 1 mole of NaBH4 per mole of P2Pmethylimine you could get 80-85% yield and could reduce the amount of time allowed for imine formation and reduction.


  • Guest
Aurelius: Of course it was mmol.
« Reply #13 on: July 06, 2002, 09:44:00 AM »
Aurelius: Of course it was mmol. Sorry but I was kind of tired when I wrote that.

neuromodulator: acid aqueous solutions destroy borohydrides really quick. But slightly alkaline aqueous solutions are quite stable. Some companies actually sell aqueous alkaline sodium borohydride solutions in huge quantities.

Spisshak: cyanoborohydride won´t touch the ketone at all at a specific pH, that´s the very reason for using it. It´s called selectivity.

Terbium: Yes molar ratio. Sorry again..
What I really wanted to see with those variations were when the yield would go up. Not having to use 5 mol amine/mol ketone or 4 mol borohydride/mol imine. With 2 mol amine/mol ketone and 1 mol borohydride/mol ketone it was the imine reduction time that was critical.  


  • Guest
Barium: Yes, that makes perfect sense in that ...
« Reply #14 on: July 06, 2002, 05:07:00 PM »
Barium:  Yes, that makes perfect sense in that NaBH4 dissociates into an equilibrium mixture of Na+, BH3, BH4-, and H- so alkalinity will of course not destroy the hydride ions in that case.

However, to get the absolute maximum yield in the reductive amination of a ketone the pH must be exactly 7.0 based on the studies I dug up years ago.  As long as the NaBH4 is only slightly alkaline, I don't guess it is going to make any kind of a significant difference, though, which is your point I guess.


  • Guest
« Reply #15 on: May 18, 2003, 07:07:00 AM »
If the yields are that good with a wet borohydride reductive alkylation, I imagine it would work-out great with NaBH(OAc)3!


  • Guest
« Reply #16 on: June 10, 2003, 03:06:00 PM »
Unfortunately triacetoxyborohydrides are very sensitive to water. So wet conditions can't be used with them. But under dry conditions they are great selective reducing agents.


  • Guest
sodium triacetoxyborohydride
« Reply #17 on: June 10, 2003, 03:33:00 PM »
I cannot seem to find a decent preparation of sodium triacetoxyborohydride, where it is made sure that the product doesn't contain any excess HOAc - could anybody help me find such a prep? Commercial sodium triacetoxyborohydride is a little too steep...


  • Guest
Prep of Sodium Triacetoxyborohydride
« Reply #18 on: June 10, 2003, 04:56:00 PM »
I use this method to make my own sodium triacetoxyborohydride:

To a suspension of 100 mmol sodium borohydride in 50 ml dry toluene is added 300 mmol GAA dropwise during good stirring. The temperature is kept below 30°C with cooling. When the gas evolution ceases, about 20 minutes, the suspended sodium triacetoxyborohydride is isolated by filtration1 and washed once with freshly distilled EtOAc. The filtercake is dried to constant weight under vaccum. The yield is almost quantitative if dry ingredients are used.

1. If one is accurate in measuring out the reagent (no excess HOAc) the toluene suspension can be freed from toluene by distillation under reduced pressure (bp. about 40°C @ 80 mbar). The result is a good quality sodium triacetoxyborohydride.

Commercial manufacturers use THF instead of toluene but I don't see any reason why since the product is as good if toluene is chosen.


  • Guest
PTC - imine reduction
« Reply #19 on: June 16, 2003, 02:48:00 PM »
It seems like there is no great benefit from anhydrous conditions anymore, unless one wants to squeeze out the highest possible yield.

1-(2,4-dimethoxyphenyl)-2-propanone, 250 mmol
Methylamine HCl, 375 mmol
NaOH, 375 mmol
Sodium borohydride, 290 mmol
Aliquat 336, 5 mol%

To a solution of the ketone in 100 ml toluene was added a solution of 25,3 g methylamine hydrochloride in 30 ml water followed by dropwise addition of 15 g NaOH dissolved in 40 ml water while stirring the mixture violently. The addition of NaOH was allowed to take five minutes so the liberated methylamine had time to react with the ketone instead of being pushed out of solution. When the addition was complete the mixture was stirred for another hour at room temperature. The aqueous layer was then separated and discharged.

To the toluene solution of the imine was added 4,8 g (12 mmol, roughly 5 mol%) Aliquat 336 followed by a solution of 11 g sodium borohydride in 35 ml water containing 25 mg NaOH1 in one portion. The mixture was violently stirred2 for one hour at 35° by slight warming. The residual borohydride was destroyed by addition of diluted acetic acid until gas evolution ceased and the aqeous layer separated and discharged. The toluene solution was extracted with 3x100 ml 10% HCl and the extracts combined, washed with 50 ml toluene and saturated with NaCl.
The aqueous solution was then made alkaline with NaOH until pH 14 was reached and extracted with 2x50 ml toluene. The combined toluene extractions was washed once with water and once with brine, and then dried over MgSO4. The solvent was removed at 40° @ 80 mbar leaving a slightly yellow oil weighing 48 g (92%). The hydrochloride was made by dissolving the free base in 250 ml EtOAc and adding 5N HCl/IPA until pH 4 was reached. Sparkly crystals formed immediately and was removed by filtration and dried to constant weight. Yield 45g 2,4-dimethoxy-N-methylamphetamine hydrochloride (2,4-DMMA)

1. NaOH is added to avoid the decomposition of sodium borohydride in water.
2. The stirring should powerful enough to create a emulsion. This allows the surface area of the two phases to be as large as possible, which means better contact and short reaction time.

I'm sure about 0,5 molar equivalents sodium borohydride is sufficient. The large amount used here was actually due to a miscalculation on my side.  :-[
My firm belief is that the in situ liberation of methylamine is more effective than using a premade 40% aqueous solution. The addition of NaOH to MeAm HCl, using the concentrations above, makes a nearly saturated solution of NaCl at the same time. This kicks out the methylamine from the aqueous phase and forces it into contact with the ketone. If a premade 40% solution is used the amine is much more prone to stay in where it is, thus less of the target imine is formed.