Forum: Reddit, TheeHive
Author: isoquinoline
Date: 26.07.2023
Mescaline synthesis success & Henry reaction optimization, isoquinoline, 26.07.2023
SYNTHESIS AND OPTIMIZATION OF 3,4,5-TRIMETHOXY-β-NITROSTYRENE The Henry reaction, while extremely versatile, performs vastly different for different substituted benzaldehydes. 3,4,5-trimethoxybenzaldehyde is notoriously reluctant to condense with nitromethane. For this reason, I set out to try to optimize the reaction conditions as best as I could.
For the production of β-nitrostyrenes, my solvent of choice is invariably glacial acetic acid (GAA). For each mmol of starting aldehyde, I use 0.5 mL of GAA (so if you start with 50 mmol aldehyde, dissolve it into 25 mL GAA). 3,4,5-trimethoxybenzaldehyde dissolves readily into this volume with gentle heating and stirring.
Since I had a decent amount of nitromethane, I decided to use anywhere between 2-2.5x the molar amount of starting aldehyde.
EXPERIMENT 1: 52 mmol 3,4,5-TMBA; 26 mL GAA; 2 mol. eq. nitromethane (6.3 g); 0.3 mol. eq. ethanolamine
The reaction mixture was heated at 80C for 1 hour. The mixture was then brought to RT and diluted with 100 mL cold dH2O. The crystals were then filtered, washed with a saturated solution of KHSO3, and recrystallized from methanol. YIELD: 4.0 g (32% of theory)
From this experiment it was clear the reaction either needed more time or more catalyst. Bisulfite washing revealed that the majority of the starting aldehyde was left unreacted. I decided to increase the amount of catalyst and slightly increase the amount of nitromethane.
EXPERIMENT 2: 10 mmol 3,4,5-TMBA; 5 mL GAA; 2.4 mol. eq. nitromethane (1.47 g); 1.5 mol. eq. ethanolamine
The reaction mixture was heated at 80C for 40 minutes. Afterwards, the reaction was worked up as exactly described in experiment 1. YIELD (PRIOR TO RECRYSTALLIZATION): 2.4 g (100% of theory); AFTER ReX: 2.3 g (96% of theory)
Now we're getting somewhere! Now let's scale it up and see if we can reduce the amount of nitromethane
EXPERIMENT 3: 63.9 mmol 3,4,5-TMBA; 32 mL GAA; 2 mol. eq. nitromethane (7.8 g); 1.5 mol. eq. ethanolamine
Reaction conditions and workup exactly as before (including recrystallization). This time, product actually began precipitating out of solution by the 25 minute mark. Perhaps slightly more GAA should be used to avoid this. YIELD: 13.8 g (90% of theory)
So, this time the yields dropped a tad. Perhaps sticking with 2.4x nitromethane is worthwhile for this reaction.
The long-short of this optimization process is: use a lot more catalyst! Most Henry reactions I've run used 0.15x. Using ten times the usual amount seems to go a long way for 3,4,5-trimethoxybenzaldehyde. Any additional insights for further optimization of the Henry would be greatly appreciated in the comments!
REDUCTION OF 3,4,5-TRIMETHOXY-β-NITROSTYRENE to 3,4,5-TRIMETHOXYPHENETHYLAMINE
In total I gathered 20.2 grams of 3,4,5-trimethoxy-β-nitrostyrene and, after recrystallizing the entire mass of crystals, ended up with beautiful yellow shards weighing 19.0 grams in total (mp 120-121C). The reduction process was split into two days, starting with 9.5 g for each.
PROCEDURE
To a clean, dry 500 mL flat-bottom two-necked flask there was added 80 mL IPA and 40 mL dH2O. With good stirring, 11.3 g NaBH4 was dissolved in the solution, and the flask placed into an ice bath. Slowly over the course of an hour or so, 9.5 g 3,4,5-trimethoxy-β-nitrostyrene was added. The flask was fitted with a thermometer and the temperature was kept below 20C during the addition process.
Once the last bit of nitrostyrene was added, the ice bath was removed and the mix kept to stir until all of the yellow color had faded. At this point, the mixture was brought up to 60C and held for 30 minutes. Then, 2 mL of a 2M CuCl2 solution was added to the mix, causing violent frothing and a strong exotherm. The mixture was brought up to reflux at 80C for 30 minutes and then left to cool.
The mixture was filtered through celite and basified with 50 mL 25% NaOH solution. The alcohol layer on top was decanted, and the bottom aqueous layer was extracted once with IPA. The IPA extracts were pooled and dried over 3A molecular sieves for 60 minutes. The IPA was distilled off leaving a black oily residue, which was diluted with cold acetone. To this mixture then was added conc. HCl dissolved in acetone dropwise until the solution was made mildly acidic and no more precipitate formed. The mixture was then placed in the freezer overnight and filtered in the morning. After washing the crystals thoroughly with acetone there was recovered 4.6 g mescaline HCl (47% of theory).
This procedure was repeated exactly and I recovered the EXACT same amount of mescaline HCl, leaving me with 9.2 g of positively glittering, slightly off-white crystals of mescaline HCl. I'm honestly unsure as to why the yields were a bit lower than usual for this reduction, though I think it may be a combination of the CuCl2 reduction not going as well as it typically does (it was considerably less frothy.. perhaps I needed a bit more heat) and mechanical losses (I was extremely clumsy during the workup).
All in all I am extremely happy with the results. The 9.2 g was added to 1.5 g of mescaline HCl I recovered from a previous procedure leaving me with 10.7 g in total. Now I am considering recrystallizing the entire mass, but I am not entirely sure the best way to do this. I was thinking of dissolving everything into boiling dH2O and slowly adding acetone or IPA until crystals just begin to crash out. If anyone has any experience with recrystallizing mescaline HCl let me know!
Cheers.
