N-Methylation

Fmoc-N-Methylation

applicable for non-basic side chains


Overall Reaction
overallsm.png

This method describes the methylation of Fmoc protected amino acids. Aliphatic sidechains are well tolerated and the reaction scales nicely to the 10g scale.[1] Acid stable sidechain protecting groups can be employed for amino acids containing nucleophillic sidechains,[2] however, installing the methyl in sequence using the mitsunobu in-sequence methylation is often preferred. A variety of other methods exist however none have been found in our hands to have the same scalability and ease of work up as the oxazolidinone route.[3]


Step 1

Oxazolidinone Formation
step1sm.png
Reaction Protocol
  1. Prepare an oil bath set to 140oC
  2. Weigh out a 1:1 mass ratio of paraformaldehyde to amino acid (e.g. if one is suing 5g Fmoc-Leu-OH, 5g of paraformaldehyde would be required)
  3. Add Fmoc-AA and paraformaldehyde to a round bottom flask with a stirbar and add ~75mL of toluene per 5g of Fmoc-AA
  4. Place the charged round bottom flask in the oil bath and equip with a reflux condenser and let the reaction mixture stir while the oil bath comes to temp.
  • NOTE: the paraformaldehyde will not fully dissolve and will remain as a white solid in solution.
  1. Add 0.2eq of Camphorsulfonic Acid to the reaction and allow the system to stir at reflux for >2hrs
  • NOTE: during this time the paraformadelyde will decompose and repolymerize on the flask and condensor walls, this is normal.
  • PRO-TIP: putting a little tinfoil jacket over the round bottom portion will aid in the reflux and resulting cleanup.
  1. After at least 2hrs, remove the reflux condesor and allow the system to condense to ~15mL per 5g of Fmoc-AA
  2. Remove the roundbottom from the oil bath and allow it to cool to room temp.
Work Up
  1. Add ~25mL of EtOAC per 5g of Fmoc-AA and filter through a cotton plug to remove residual paraformaldehyde
  2. Wash with 2x Saturated NaHCO3 (shake the shit out of it)
  3. Wash with 1x brine
  4. dry over your favorite drying agent (pros use NaSO4 )
  5. rotovap to a viscous oil or occasionally white solid.

Example Reaction Table
Name g MW mmol eq d mL
Fmoc-Leu-OH 5 353.41 14.14787357 1 n/a n/a
paraformaldehyde 5 n/a n/a n/a n/a n/a
CSA 0.657310206 232.3 2.829574715 0.2 n/a n/a

Step 2

Reductive Ring Opening
step2sm.png
Reaction Protocol
  1. Add ~6-10mL of DCM per 5g of Fmoc-AA and a stir bar to the round bottom flask and try to suspend the oil as best you can
  2. Carefully add an equal amount of TFA and allow the solution to stir for ~5min at Room Temperature
  3. add 3eq of Triethylsilane to the reaction mixture and allow to stir overnight.
  • PRO-TIP: A septa with a flacid balloon can be placed on the flask to prevent evaporative loss but allow for gas evolution.
Work Up
  1. Blow down the reaction mixture to remove the bulk of the TFA, DCM and excess TES. It should form a clear goopy oil.
  2. Bring the resulting oil back up in EtOAC and wash 2x with H2O (shake vigorously but not shit shaker status to avoid emulsion)
  3. Wash 1x with brine
  4. dry over your favorite drying agent (pros use NaSO4 )
  5. rotovap to a viscous oil or occasionally white solid.
  6. Sometimes precipitation from Hexanes:EtOAc mixtures can work (normally around 80% Hexanes) or blowing down from a hexanes rich mixture.
  • Some have had luck with putting the precipitation mixture on dry ice and then blowing down. YMMV.

Example Reaction Table
Name g MW mmol eq d mL
Oxazole 5 365.43 13.68251101 1 n/a n/a
CSA 4.773007142 116.28 41.04753304 3 0.728 6.556328492

Downloads


Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License