All Blogs | Help | Support | About
2nd December 2010 @ 05:35
Scale-up of Synthesis of (-)-Dibenzoyl-L-tartaric acid (MW10-1-30)

See also:
See Synthesis of (+)-Dibenzoyl-D-tartaric acid (MW10-5)
and Synthesis of (+)-Dibenzoyl-D-tartaric acid - alternative route (MW10-6)

Reaction Scheme


Hazard and Risk Assessment:
HRAF MW10.pdf


Start time: 4:00 PM 02/12/2010
End time: 8:00 PM 02/12/2010

Procedure: [1]
A mixture L-(+)-tartaric acid (27.0 g, 180 mmol, M.W. 150.1 g/mol) and benzoyl chloride (88.6 g, 73 mL, 630 mmol; M.W. 140.6 g/mol, 1.21 g/mL) was heated to 130°C for 4 h. After which time a pale yellow solid was formed. The mixture was cooled to room temperature, washed with cold Et2O and recrystallized from toluene (400 mL).

Yield of the anhydride: 48.3 g (141 mmol, 79%) colorless, crysalline solid
[Anhydride, C18H12O7 M.W. = 340.3 g/mol]
[Dibenzoyl-tartaric acid, M.W. = 358.3 g/mol]
[Dibenzoyl-tartaric acid * 2 i-PrOH, C24H30O10 M.W. = 478.5 g/mol]

Hydrolysis of the anhydride:
Start time: 3:00 PM 03/12/2010
End time: 5:00 PM 03/12/2010

The anhydride (48.3 g, 141 mmol) was dissolved in a mixture of acetone (200 mL) and water (20 mL) and heated to reflux for 2 h. Water (200 mL) was added and aceton was removed under reduced pressure. After further addition of water (200 mL) the mixture was heated refluxed and cooled to 0°C for 30 min. The water was removed by filtration and the colorless solid residue was freeze-dried (alternative method: dissolving solid in ethyl acetate and dry over Na2SO4).

The solid was recrystallized from i-PrOH/hexane:
1. crystallization ~ 50 g from i-PrOH/hexane (~ 500 mL 1:1): yield 28.3 g (59.0 mmol, 33%)
2. crystallization 33 g from i-PrOH/hexane (~ 400 mL 1:2): yield 24.4 g (51.0 mmol, 28%), remain 8.5 g
3. crystallization (8.5 g) from i-PrOH/hexane (~ 50 mL 1:2): yield 6.22 g (13.0 mmol, 7%)

-> Yield 58.9 g (123 mmol, 68%) (-)-dibenzoyl-L-tartaric acid * 2 i-PrOH
[M.W.(C24H30O10) = 478.5 g/mol]

Analytical data:
m.p. 98-100°C (lit. 95–98°C for DBTA * H2O [3])
1H NMR (DMSO-d6, 200 MHz): δ = 1.03 (d, J=6 Hz, 12H), 3.78 (sep, J=6 Hz, 2H), 5.88 (s, 2H), 7.57-7.64 (m, 4H), 7.70-7.74 (m, 2H), 8.00-8.04 (m, 4H), 14.00 (bs, 2H). Data: 1H NMR MW10-7.pdf
13C NMR (DMSO-d6, 50.3 MHz): δ = 25.5 (4C), 62.2 (2C), 71.6 (2C), 128.6 (2C), 129.1 (4C), 129.5 (4C), 134.2 (2C), 164.8 (2C), 167.3 (2C). Data: 13C NMR MW10-6
IR (neat): nu = 3465 cm-1, 2975, 1730, 1237, 1098, 938, 706.Data: MW10-1-30 acid.pdf -> no OH-peak for i-PrOH Data: IR MW10-5.pdf
[α]D20 = -85.0° (c=1, EtOH). ([α]D20 = 119.4° (c=1.055, MeOH))
MS (ESI (-)) m/z (%): 357 (100) [M-H]-.
HRMS (ESI (-)) Calcd. for [ C18H13O8]-: 357.0616, found: 357.0616.
HRMS (ESI (+)) Calcd. for [C18H14O8+Na]+: 381.0581, found: 381.0579.
C24H30O10 (478.5 g/mol): calc. C 60.24%, H 6.32%; found: C 60.27%, H 6.33% -> dibenzoyltartaric acid * 2 i-PrOH

More spectra and data:
  • Synthesis of (+)-Dibenzoyl-D-tartaric acid - alternative route (MW10-6)
  • Synthesis of (+)-Dibenzoyl-D-tartaric acid (MW10-5)
  • Synthesis of (-)-Dibenzoyl-L-tartaric acid (MW10-1-30)

    Results:
    The procedure for the synthesis of (-)-dibenzoyl-L-tartaric acid is easy to scale-up also the purification / separation from the side product benzoic acid is not a problem. However the product couldn't be crystallized from water as (-)-dibenzoyl-L-tartaric acid monohydrate (neccessary?) therefore it was recrystallized from iso-propanol/hexane.
    Yield could be improved but ordering the chemical from a chemical supplier is more economic... (AU$ 16 per kg (-)-dibenzoyl-L-tartaric acid monohydrate for an order volume of 1000 kg)


    References:
  • [1] “The Reduction of Various Sugar Acids to Glycitols with Lithium Aluminum Hydride”, R. K. Ness, H. G. Fletcher and C. S. Hudson, J. Am. Chem Soc. 1951, 73, 10, 4759-4761.

    [2] “Relative Stereoselectivity of the Reactions of (2R,3R)- and (2R,3S)-Dibenzoyloxysuccinic Anhydrides with Chiral Amines and Alcohols”, K. H. Bell, Aust. J. Chem. 1987, 40, 399-404. doi:10.1071/CH9870399.

    [3] "Convenient and Inexpensive Synthesis of (1R,2R)- trans-1-Amino-6-nitroindan-2-ol", S. Kozhushkov, D. Yufit and A. de Meijere, Adv. Synth. Catal. 2005, 347, 255– 265. (for analytical data references)

    [4] "Novel processes for the preparation or (R)-alpha-(2,3-dimethoxyphenyl)-1-[2-(4-fluorophenyl)ethyl]-4-piperidinemethanol", Daugs et al. US Patent Application (2002), US 2002/0151717A1. [Example 53, p. 42]
    Attached Files
    1H NMR MW10-7.pdf
    1H NMR MW10-7 JCAMP-DX.dx