Resolution of rac-Praziquanamine with quinic acid

Published by Michael Wolfle on 11 May 2010 - 7:50am

 

 
Here is a procedure from a German Patent Application from 1975 which uses quinic acid for the racemic resolution of praziquanamine. The corresponding US patent (US 3,993,760, 1976) provides the translated procedure (page 5, column 7, "Example 1" lines 32-50) here. However, this procedure gives less information about the yield obtained.
 
The result of the first trial conducted under the conditions described in the patent procedure was disappointing (see Procedure 1 and Patent Procedure).  Methanolic solutions of Praziquanamine (1 eq.) and quinic acid (1.3 eq.) were combined and heated at reflux for 15 min. Upon cooling to room temperature, the precipitate which formed was filtered off. The filtrate was concentrated to a fifth of the volume and a second precipitate was collected. The amorphous precipitates contained a different enantiomeric excess of each enantiomers:
Precipitate 1: yield 72% (ee 11% (+)-PZA)*
Precipitate 2: yield 13% (ee 45% (-)-PZA)*
 
*) ee determined by optical rotation of the liberated amine from the diastereomeric salt (so far, all attempts to determinate ee by HPLC have been unsuccessful)

With small variations in the conditions the ee and yield of the praziquanamine resolution were improved (see Procedure 2). The methanolic solutions of amine and acid were combined in a sealed tube, heated to reflux and cooled to room temperature. The first precipitate was filtered off, the liquor was concentrated untill a second solid started to precipitate and then the mixture was stored for 10 h in the fridge at 6°C.
Precipitate 1: yield 59% (ee 48% (+)-PZA)*
Precipitate 2: yield 23% (ee 85% (-)-PZA)*

The current issues I am working on are finding a suitable method to determine the ee by using chromatography (HPLC, GC) and optimizing the resolution conditions.
 
Syncom BV found a resolution method for PZQamine by screening of a number of chiral acids and identified (-)-dianisoyl-L-tartaric as an excellent resolving agent for the (+)-enantiomere.

Procedure 1:
Praziquanamine (202 mg, 1.00 mmol) was dissolved in methanol (0.8 mL) and added to a warm solution of quinic acid (1.30 mmol, 250 mg) in methanol (4.2 mL). After a few seconds an amorphous precipitate formed. The stirred mixture was heated at reflux for 15 min and allowed to cool to room temperature. The amorphous precipitate was filtered off, the remaining solution was concentrated to about 0.8 mL and the resulting second amorphous precipitate was also filtered off. Both precipitates and the remaining filtrate were adjusted to pH 12 by adding a 2 N NaOH solution, extracted with DCM and dried over sodium sulfate. The concentrated residues were analyzed by optical rotation (and chiral HPLC).
 

*Calculation of the enantiomeric excess by optical rotation:
ee = ([α]obs/[α]max) x 100        
with
[α]D20 (-)-praziquanamine = -306° [1]

Procedure 2:
Praziquanamine (202 mg, 1.00 mmol) was dissolved in methanol (0.8 mL) and added to a warm solution of quinic acid (1.30 mmol, 250 mg) in methanol (4.2 mL) in a sealed tube. The stirred mixture was heated to reflux and cooled to room temperature. The amorphous precipitate was filtered off, the remaining solution was concentrated to about 1.5 mL when a solid started to precipitate. The mixture was stored in the fridge at 6°C for 10 h and then the second amorphous precipitate was filtered off. Both precipitates and the filtrate were adjusted to pH 12 by adding a 2 N NaOH solution, extracted with DCM and dried over sodium sulfate. The concentrated residues were analyzed by optical rotation (and chiral HPLC).
 

Patent Procedure (Translated from the original patent application, language: German, page 11-12): [1]

Example 1:
[…]
24.3 g (+/-)-4-oxo-1,2,3,6,7,11b-hexahydro-4H-pyrazino[2,1-a]isochinoline (m.p. 118-119°C […]) is dissolved in 100 mL methanol and added to a warm solution of 30 g quinic acid in 500 mL methanol. It is heated to reflux for 15 min and then cooled to 20°C. The obtained crystals are filtered off, the mother liquor is reduced to 100 mL and the precipitated crystals are separated. The quinic acidic salt of (-)-4-oxo-1,2,3,6,7,11b-hexahydro-4H-pyrazino[2,1-a]isochinoline is obtained, m.p 196°C. The salt is dissolved in water, the solution is made basic and extracted with chloroform. After drying and evaporation the left rotating free base is obtained, m.p. 120°C, [α]D20 = -306°.

[1]    “Pyrazinoisochinolin-Derivat“, R. Polke, F. Loebich, J. Seubert, H. Thomas and P. Andrews, German Patent Application (1975) DE 2,331,713.

 

Comments

Michael Wolfle's picture

The measurement of the enantiomeric composition of PZQamine is still a problem. Analytical chiral HPLC methods so far we used wasn’t successful. We have tested two chiral columns to separate the enantiomers of Praziquanamine: Chiralcel OD-H and AD-H. With the first column no separation of the enantiomers were obtained (one broad peak). The latter one showed a separation, but due to a large tailing (no baseline separation) of the amine the ee couldn’t be determined.

Attempts to analyze the enantiomeric ratio by chiral GC (Varian CP-Chirasil-Dex CB) were also unsuccessful - no signals of the compounds were monitored.

Determination of the enantiomeric excess (optical purity) by polarimetry has limitations, because of possible deviations depends on the concentration of the enantiomers (and enantiomer composition).[1] This non-linearity of optical purity versus enantiomeric excess is also known as the Horeau-effect.[2] However, polarometry allows a crude evaluation of the enantiomeric excess of the amine and also the precipitated diastereomeric salt which can be analyzed by polarimetry without liberation of the amine. In this case the optical rotation of the diastereomeric salt is the sum of the molar rotations [M]D20 of it component ions:[3]

[M]salt = [M] (quinic acid) + [M] (praziquanamine)

with
[M]D20= [α]D20*M/1000     (M: molecular weight)

and
[α]D20 quinic acid = -43.2° (c=1, H2O); M = 192.2 g/mol
[M]D20 quinic acid = -8.30° mol/L

and
[α]D20 (-)-praziquanamine = -306° (c=1, H2O); M = 202.4 g/mol
[M]D20 (-)-praziquanamine = -61.9° mol/L
 
 

Methods:

Chiral HPLC:

column Chiralcel AD-H, eluent: Hexane/IPA/TEA 90:10:0.1 (isocratic), UV detector 254 nm, flow rate 1 mL min-1, retention time: (-)-PZA: 39.2 min, (+)-PZA: 42.9 min, strong tailing (about 8 min).

Column Chiralcel OD-H, eluent: Hexane/IPA/TEA 80:20:0.1 (isocratic), UV detector 254 nm, flow rate 1 mL min-1, retention time: 20.7 min, broad signal for both enantiomeres, no separation.

Chiral GC:
column Varian CP Chirasil-Dex CB (25m x 0.25 mm, 0.25 µm) isotherm, 200°C, 300 min, no signal monitored.
 
 
[1]  “Do the Terms “% ee” and “% de” Make Sense as Expressions of Stereoisomer Composition or Stereoselectivity?” R. E. Gawley, J. Org. Chem. 2006, 71, 2411-2416.
 
[2]  "Interactions d'enantiomeres en solution ; influence sur le pouvoir rotatoire : Purete optique et purete enantiomerique." A.Horeau, Tetrahedron Lett. 1969, 3121-3124.
 
[3]    “CRC Handbook of the Optical Resolution via Diastereometric Salt Formation”, D. Kozma, CRC press, 2002, p. 108.

quintus's picture

 What other chiral acids have you tried for the resolution?
Quinic acid seems to be working, not well, but working. With this you may need to use very slow cooling, perhaps with re-heating and cooling. Try other similar acids. I would go for the tartaric series or mandelic acid.
Also you may have a polymorph problem preventing efficient crystallisation. Get a powder x-ray done to see if you have polymorphs.
Use the melting point for the first approximation of success, also the salt from the first crystallisation you should re-crystallise, the solubility should go down as the ee% goes up.

quintus's picture

 There are 2 papers worth reading for this problem;
Kellog R. M. etal Chem. Eur. J. 2005, 11, 5619 doi: 10.1002/chem.200500440
and Kellog R. M. Angew. Chem. Int ed Engl. 1998, 37, 2349

ndt228's picture

Regarding your problems with the above:
1) The earlier Syncom contribution on this thread stated that Chiralcel OJ-H, Chiralpak IA or AS-H were all effective at resolving PZQ Amine
2) If you can't access these columns (I know how much they cost) then you might improve performance on your AD-H column by (a) switching from TEA to Et2NH as amine modifier.  This should sharpen your peaks/reduce tailing   (b) increasing the EtOH conent of the eluent to reduce the run-time.    (c) trialling other combinations of alcohol (eg IPA, n-PrOH) and alkane (eg) Heptane, iso-octane. 
Regards, Nick