Scale-Up Syntheses of Two Naturally Occurring Procyanidins: (-)-Epicatechin-(4β,8)-(+)-catechin and (-)-Epicatechin-3-0-galloyl- (4β, 8)-(-)-epicatechin-3-0-gallate

Article abstract of DOI:10.1021/op700031n

A scaleable process for the synthesis of two naturally occurring procyanidins, namely (-)-epicatechin-(4β,8)-(+)-catechin (1) and(-)-epiratechin-3-O-galloyl-(4β,8)-(-)-epicatechin-3-O-gallate (2), is described. The key steps were highlighted by improvements for the benzylation of (+)-catechin (3), stereo-selective reduction of the C-3 keto group of (2A)-5,7,3′,4′-tetrakis(benzyloxy)flavan-3-one (10), and coupling between 4-hydroxyethoxy-5,7,3′,4′-tetra-O-benzyl-(-)-epicatechin (11) and 5,7,3′,4′-tetra-O-benzyl-(+)-catechin (4) or 5,7,3′,4′-tetra-O-benzyl-(-)-epicatechin (6), respectively. The debenzylation performed in a biphasic system resulted in an improved yield and purity of the target compounds. The chemistry was scaled-up to produce multigram quantities of the title compounds (1 and 2) for various in vitro, ex vivo, and in vivo studies. Moreover, the scale-up process provided a detailed description for the preparation of multihundred to kilogram scale quantities of intermediates used in the synthesis of these two titled procyanidins.

Full text of DOI:10.1021/op700031n

Organic Process Research & Development 2007, 11, 422−430
Scale-Up Syntheses of Two Naturally Occurring Procyanidins:
(-)-Epicatechin-(4â,8)-(+)-catechin and
(-)-Epicatechin-3-O-galloyl-(4â,8)-(-)-epicatechin-3-O-gallate†
Pradeep K. Sharma,‡ Alexander Kolchinski,‡ He´le`ne A. Shea,§ Jayesh J. Nair,§ Yanni Gou,| Leo J. Romanczyk, Jr.,*, and
Harold H. Schmitz#
Chemical Process Research & DeVelopment, Catalytic SerVices, and Analytical DiVision, Johnson Matthey
Pharmaceutical Materials, Inc., 25 Patton Road, DeVens, Massachusetts 01434, U.S.A., Masterfoods USA, A MARS
Incorporated Company, 800 High Street, Hackettstown, New Jersey 07840, U.S.A., and MARS, Incorporated,
6885 Elm Street, McLean, Virginia 22101, U.S.A.
Abstract:
Our interest is focused mainly at the cancer and vascular
biology areas, where initial observations have shown different
procyanidins to evoke different activities.10-23 To unequivo-
cally determine the natural oligomeric procyanidins associ-
ated with these activities and to confirm the structures
assigned to these compounds, a synthesis program was
initiated.24-27 A series of oligomers of defined regio- and
stereochemistry were synthesized for comparison to oligo-
mers purified from cocoa extracts for structure-activity
A scaleable process for the synthesis of two naturally occurring
procyanidins, namely (-)-epicatechin-(4â,8)-(+)-catechin (1)
and (-)-epicatechin-3-O-galloyl-(4â,8)-(-)-epicatechin-3-O-gal-
late (2), is described. The key steps were highlighted by
improvements for the benzylation of (+)-catechin (3), stereo-
selective reduction of the C-3 keto group of (2R)-5,7,3′,4′-
tetrakis(benzyloxy)flavan-3-one (10), and coupling between
4-hydroxyethoxy-5,7,3′,4′-tetra-O-benzyl-(-)-epicatechin (11)
and 5,7,3′,4′-tetra-O-benzyl-(+)-catechin (4) or 5,7,3′,4′-tetra-
O-benzyl-(-)-epicatechin (6), respectively. The debenzylation
performed in a biphasic system resulted in an improved yield
and purity of the target compounds. The chemistry was scaled-
up to produce multigram quantities of the title compounds (1
and 2) for various in Witro, ex WiWo, and in WiWo studies.
Moreover, the scale-up process provided a detailed description
for the preparation of multihundred to kilogram scale quantities
of intermediates used in the synthesis of these two titled
procyanidins.
(3) Saito, M.; Hosoyama, H.; Ariga, T.; Kataoka, S.; Yamaji, N. J. Agric. Food.
Chem. 1998, 46, 1460.
(4) (a) Liviero, L.; Puglisis, P. P.; Morazzoni, P.; Bombardelli, E.
Fitoterapia 1994, 65, 203. (b) Constable, A.; Varga, N.; Richoz, J.;
Stadler, R. H. Mutagenesis 1996, 11, 189. (c) Inanami, O.; Watanabe,
Y.; Syuto, B.; Nakano, M.; Kuwabara, M. Free Radical Res. 1998,
29, 359.
(5) Inanami, O.; Watanabe, Y.; Syuto, B.; Nakano, M.; Kuwabara, M. Free
Radical Res. 1998, 29, 359.
(6) Hartisch, C.; Kolodziej, H.; Von Bruchhausen, F. Planta Med. 1977, 63,
106.
(7) Scalbert, A. Phytochemistry 1991, 30, 3875.
(8) Kameda, K.; Takau, T.; Okuda, H.; Kimura, Y.; Okuda, T.; Hatano, T.;
Agata, I.; Arichi, S. J. Nat. Prod. 1987, 50, 680.
(9) (a) So, F. V.; Guthrie, N.; Chambers, A. F.; Carroll, K. K. Cancer Lett.
1997, 112, 127 and references cited therein. (b) Kolodziej, H.; Haberland,
C.; Woerdenbag, H.; Konigs, A. W. T. Phytotherapy Res. 1995, 9, 410. (c)
Ramljak, D.; Romanczyk, L. J.; Metheny-Barlow, L. J.; Thompson, N.;
Knezevic, V.; Galaperin, M.; Ramesh, A.; Dickson, R. B. Mol. Cancer Ther.
2005, 4, 537.
(10) Romanczyk, L. J., Jr.; Hammerstone, J. F., Jr.; Buck, M. M. U.S. Patent
5,712,305, January 27, 1998.
(11) Romanczyk, L. J., Jr.; Hammerstone, J. F., Jr.; Buck, M. M.;
Post, L. S.; Cipolla, G. G.; McClelland, C. A.; Mundt, J. A.;
Schmitz, H. H. PCT Int. Appl. WO 9736597. Chem. Abstracts 1997, 127,
322787.
(12) Romanczyk, L. J., Jr.; Hammerstone, J. F., Jr.; Buck, M. M. U.S. Patent
5,554,645, September 10, 1996.
(13) Romanczyk, L. J., Jr.; Hammerstone, J. F., Jr.; Buck, M. M.;
Post, L. S.; Cipolla, G. G.; McClelland, C. A.; Mundt, J. A.;
Schmitz, H. H. PCT Int. Appl. WO 9736,497. Chem. Abstracts 1997, 127,
330680.
(14) Kenny, T. P.; Keen, C. L.; Jones, P.; Kung, H. J.; Schmitz, H. H.; Gershwin,
M. E. Exp. Biol. Med. (Maywood) 2004, 229, 765.
(15) Engler, M. B.; Engler, M. M.; Chen, C. Y.; Malloy, M. J.;
Browne, A.; Chiu, E. Y.; Kwak, H. K.; Milbury, P.; Paul,
S. M.; Blumberg, J.; Mietus-Synder, M. L. J. Am. Coll. Nutr. 2004, 23,
197.
Introduction
Polyphenols are an important class of natural products.
Such compounds are widely distributed in nature and possess
a diverse range of biological activities. These include, but
are not limited to, the inhibition of HIV 1 replication in Vitro,1
reducing the risk of heart disease,2 suppressing ulcer forma-
tion,3 possessing antimutagenic,4 neuroprotective,5 anti-
inflammatory,6 antibacterial,7 and hypotensive8 properties,
and can inhibit the growth of cancer cells.9
* To whom correspondence should be addressed. E-mail: leo.romanczyk@
mss.effem.com. Telephone: (908) 892-1125. Fax: (908) 892-1196.
† Part of this work has been incorporated in patent application WO2007005248
and US2007004796(A1).
‡ Chemical Process Research & Development, Johnson Matthey Pharmaceuti-
cal Materials, Inc.
§ Catalytic Services, Johnson Matthey Pharmaceutical Materials, Inc.
| Analytical Division, Johnson Matthey Pharmaceutical Materials, Inc.
Masterfoods USA.
(16) Kenny, T. P.; Keen, C. L.; Jones, P.; Kung, H. J.; Schmitz, H. H.; Gershwin,
M. E. Exp. Biol. Med. (Maywood) 2004, 229, 255.
(17) Maron, D. J. Curr. Atheroscler. Rep. 2004, 6, 73.
(18) Fischer, N. D.; Hughes, M.; Gerhard-Herman, M.; Hollenberg, N. K. J.
Hypertersion 2003, 21, 2281.
(19) Pearson, D. A.; Paglieroni, H. G.; Rein, D.; Wun, T.; Schramm, D. D.;
Wang, J.; Holt, R. R.; Gosselin, R.; Schmitz, H. H.; Keen, C. L. Throm.
Res. 2004, 15, 191.
# MARS, Incorporated.
(1) Shahat, A. A.; Ismail, S. I.; Hammouda, F. M.; Azzam, S. A.; Lemiere, G.;
De Bruyne, T.; De Swaet, S.; Piteters, L.; Vlietinck, A. Phytomedicine 1998,
5, 133.
(2) (a) Aviram, M.; Fuhrman, B. Artherosclerosis 1998, 137, S45. (b)Yama-
koshi, J.; Kataoka, S.; Koga, T.; Ariga, T. Artherosclerosis 1999, 142, 139.
422
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development
10.1021/op700031n CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/26/2007

Scheme 1a
a Reagents and conditions: (i) BnBr, NaH, DMF.
Scheme 2a
Figure 1.
Figure 2.
a Reagents and conditions: (i) Dess-Martin periodinane reagent, wet CH2Cl2,
RT.
Scheme 3
relationships in various in Vitro, ex ViVo, and in ViVo models
of assessment. The results from these initial studies has
addressed some of the complex, synthetic challenges posed
by the proanthocyanidins (condensed or nonhydrolyzable
tannins) related to the difficulty in controlling the interflavan
regio- and stereochemistry, as well as the sensitivity of the
unprotected compounds to acid,28 alkali,29 and oxidizing30
environments. Through these efforts, 10-100 milligram
quantities of some of these compounds have been made to
confirm much of the in Vitro results elaborated by the natural
products.24,27 However, in order to fully develop the potential
clinical applications for this class of compounds and address
pharmacological and toxicological requirements, 100 g to
1000 g levels of material are essential for full investigations.
To this end, we report on the process scale-up conditions
required for the multigram synthesis of two naturally
occurring procyanidin dimers.
These conditions are applicable to pilot manufacture at
the kilogram level through the manipulation of the chemistry
accomplished earlier.24-27 The target dimers selected for
scale-up syntheses were 1 and 2 (Figure 1). To the best of
our knowledge, this effort represents the first large-scale
synthesis for this class of compounds.
The total synthesis of the target dimers, if developed,
would obviously require a lengthy synthetic sequence invol-
ving several stereoselective steps. The preparations of the
target dimers, however, can be greatly simplified by using
naturally occurring 3 and 5 as the starting materials
(Figure 2).
The literature describes a semisynthetic route (Scheme
1)24a that is based on naturally occurring 3, exclusively. This
route gives reliable results for 100 mg to 1 g scales but
requires several HPLC purifications and involves other
protocols, which are difficult to implement and/or have safety
issues at the industrial scale. Following the same sequence
of transformations we significantly modified the existing
procedures and developed a process applicable for the
preparation of the target dimers in kilogram quantities
sufficient for a variety of in Vitro, ex ViVo, and in ViVo studies.
Naturally occurring 3, which serves as a major building
block in this synthetic sequence, is available in kilogram
quantities from commercial sources. Incorporation of 5 as
the other starting material could have provided much easier
access to the midstream intermediate 6, rather than oxidation
and stereoselective reduction of 3 (Schemes 2 and 3). Our
lengthy outsourcing attempts, however, brought us to the
realization that pure 5 was not available in sufficient
quantities. Hence, similar to the literature route,24a we based
(20) Schramm, D. D.; Wang, J. F.; Holt, R. R.; Ensunsa, J. L.; Gonsalvas, J. L.;
Lazarus, S. A.; Schmitz, H. H.; German, J. B.; Keen, C. L. Am. J. Clin.
Nutr. 2001, 73, 36.
(21) Schramm, D. D.; Wang, J. F.; Schmitz, H. H.; Keen, C. L. FASEB J. 2000,
14, A762.
(22) Schroeter, H.; Heiss, C.; Blazer, J.; Kleinbongard, P.; Keen, C. L.;
Hollenberg, N. K.; Seis, H.; Kwick-Uribe, C.; Schmitz, H. H.; Kelm, M.
PNAS 2006, 103, 1024.
(23) Ramljak, D.; Romanczyk, L. J.; Metheny-Barlow, L.; Thompson, N.;
Knezevic, V.; Galperin, M.; Ramesh, A.; Dickson, R. B. Mol. Cancer Ther.
2005, 4, 537.
(24) (a) Tu¨ckmantel, W.; Kozikowski, A. P.; Romanczyk, L. J., Jr. J.
Am. Chem. Soc. 1999, 121, 12073. (b) Goetz, G.; Fkyerat, A.; Metais,
N.; Kunz, M.; Tabacchi, R.; Pezet, R.; Pont, V. Phytochemistry 1999, 52,
759.
(25) Kozikowski, A. P.; Tu¨ckmantel, W.; George, C. J. Org. Chem. 2000, 65,
5371.
(26) Kozikowski, A. P.; Tu¨ckmantel, W.; Hu, Y. J. Org. Chem. 2001,
66, 1287.
(27) Kozikowski, A. P.; Tu¨ckmantel, W.; Bo¨ttcher, G.; Romanczyk, L. J., Jr. J.
Org. Chem. 2003, 68, 1641.
(28) Prieur, C.; Rigaud, J.; Cheynier, V.; Moutounet, M. Phytochemistry 1994,
36, 781.
(29) Ferreira, D.; Steynberh, J. P.; Burger, J. F. W.; Bezuidenhoudt, B. C. D. In
Phenolic Metabolism in Plants; Stafford, H. A., Ibrahim, R. K., Eds.;
Plenum: New York, 1992; pp 255-295.
(30) De Freitas, V. A. P.; Glories, Y.; Laguerre, M. J. Agric. Food Chem. 1998,
46, 376.
Vol. 11, No. 3, 2007 / Organic Process Research & Development
•
423

Table 1. Crystallization of 4
entry
solvent/conditions
yielda
purity of 4
1
2
3
4
5
6
7
EtOH/ -5° C
76%
78%
50%
63 (%AUC)
67 (%AUC)
76 (%AUC)
EtOH/EtOAc (8/2)/ RT
MTBE/ RT
CH2Cl2/ -20° C
PhMe/ RT
CF3Ph/ RT
ClCHdCCl2/ -20° C
82%
80%
46%
70 (% AUC)
70 (%AUC)
>97 (%AUC)
a All values are isolated yields.
allowed to stir at rt for 18 to 24 h. The solids were removed
through a celite pad, and the filtrate was diluted with EtOAc.
This was washed sequentially with dilute hydrochloric acid
and water to produce product, which was subsequently
purified by crystallization (see Table 1).
Figure 3.
our development work on 3 as the only flavan-3-ol starting
In order to avoid the use of silica gel chromatography or
a silica gel plug, variations in trituration or recrystallization
procedures of 4 were attempted. Table 1 shows the various
conditions employed.
In the most successful experiment (Table 1, entry 7),
4 was treated with hot trichloroethylene followed by
cooling to -20 °C for 24 h to afford product in 45-50%
yields having >97% (AUC) purity. These yields and
purities were obtained on a consistent basis for the synthesis
of 4.
material.
Results and Discussion
Most synthetic transformations required for the prepara-
tion of dimeric species from unprotected 3 are poorly
compatible with free phenolic functionalities. Other proper-
ties of free polyphenols, such as their air sensitivity and
complicated extractive behavior, also pose handling problems
for the phenolic intermediates. For this reason, protection
of the free phenolic functionalities before executing core
synthetic transformations, followed by the deprotection step,
proved to be the most fruitful in the synthesis of the
procyanidins.24 Benzylation was used as the most common
way to protect the phenolic functionalities of procyanidins
and related molecules.
1. Benzylation of 3. The original protocol for the
benzylation of 3 (Scheme 1) employed BnBr/NaH in DMF
to produce 4 in 20% yield.24a The major byproducts, 7, 8,
and 9 (Figure 3), were removed by column chromatography
followed by crystallization. These experimental conditions
also lead to partial penta-O-benzylation of 3.26 In addition,
NaH in DMF represents a significant safety hazard and is
known to have caused several incidents in industrial
settings.31-33
Multiple alternative benzylation protocols of 3 and 5 are
described in the literature.24b Other methods for benzylation,
which were attempted in our laboratory, involved the use of
various bases such as K2CO3, DBU, Et3N, Cs2CO3, NaHCO3,
Na2CO3, and Hu¨nig’s base in various solvents such as DMF,
acetone, EtOAc, water, and EtOAc/water with and without
phase transfer catalyst (nBu4NI) at RT, at moderate temper-
ature (40-45 °C), or at reflux. These experiments only
allowed for incremental yield increases and purity improve-
ments having C-benzylated compounds as major byproducts
(7, 8, and 9).
2. Oxidation of 4 to 10. Our efforts to obtain 5 in
large quantities from commercial sources were not successful.
In addition, compound 5 is considerably more expensive than
3. Thus, it was decided to use 4 as starting material for
6 by inverting the stereochemistry at the C-3 position
(Schemes 2 and 3) through an oxidation/stereoselective
reduction sequence. The use of Dess-Martin periodinane in
the oxidation of 4 to 10 is known to give reproducible yields
of 92% after column chromatography.24a However, the cost
of Dess-Martin periodinane reagent, and the potential
interruptions with its supply prompted us to look for a
replacement.
There are several published reports that claim low yield
for the oxidation of the C-3 alcohol to the ketone under the
following conditions; DMSO/Ac2O,34 CrO3 activated with
DMSO,35 PDC,36 NMO/catalytic tetrapropylammonium
peruthenate,24a or Oppenauer conditions (fluorenone,
KOCMe3).24a In our studies we also found these conditions
to be impractical as a mixture of products, and incomplete
reactions were observed. Our attempts to apply previously
untested Swern oxidation and sodium hypochlorite/TEMPO
or MnO2 for the preparation of 10 did not produce the desired
ketone. Rather a complex mixture of products was observed.
Even the application of the synthetically equivalent, stabilized
IBX did not produce the desired ketone. The lack of practical
alternatives resulted in our return to Dess-Martin periodi-
nane (Scheme 2). The isolation protocol was significantly
modified where the column chromatography step was
The best conditions, which were eventually scaled up,
involved slow addition of BnBr to a stirred suspension of 3
and K2CO3 in DMF at <30 °C. The suspension was then
(31) Buckley, J.; Webb, R. L.; Laird, T.; Ward. R. J. Chem. Eng. News 1982,
60, 5.
(32) DeWall, G. Chem. Eng. News 1982, 60, 5.
(33) Laird, T. Chem. Ind. 1986, 17, 134.
(34) Miura, S.; Midorikawa, T.; Awata, N. Radioisotopes 1983, 32, 225.
(35) Nonaka, G.; Morimota, S.; Kinjo, J.; Nohara, T.; Nishioka, I. Chem. Pharm.
Bull. 1987, 35, 149.
(36) Zalikowski, J.; Gilbert, K.; Borden, W. J. Org. Chem. 1980, 45, 346.
424
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development

Table 2. Stereoselective reduction of 10 to 6 using Ru catalystsa
entry
catalyst
additive solvents
condition
yield
1
2
3
4
Cl2Ru(II)-(R)-BINAP (0.1 equiv)
(dppb)Ru(II)Cl2 (0.1 equiv)
Cl2Ru(II)-(R)-BINAP (0.1 equiv)
(dppb)Ru(II)Cl2 (0.1 equiv)
5% Ru/C 10 wt %
THF
THF
H2 (50 psi), RT, 7 h
H2 (50 psi), RT, 7 h
H2 (50 psi), 50 °C, 7 h
H2 (50 psi), 50 °C, 7 h
10 (100%)
10 (100%)
4 (3%) 6 (5%) 10 (69%)
10 (100%)
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
THF, MeOH
5
H2 (100 psi), 50 °C, 65 h 10 (100%)
6
7
8
9
10
11
12
13
14
15
16
Ru(II)-(R)-BINAP (0.1 equiv)
Ru(II)-(TsDPEN)b (0.011 equiv)
Ru(II)-(TsDPEN)b (0.011 equiv)
Ru(II)-dppbc (0.01 equiv)
Cs2CO3 (0.1 equiv), THF
KOH (0.1 equiv), THF/IPA
KOH (0.1 equiv), THF/IPA
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
Cs2CO3 (0.1 equiv), THF
H2 (200 psi), 50 °C, 65 h 4 (5%) 6 (82%) 10 (1.5%)
RT, 15 h
80 °C, 15 h
H2 (50 psi), 50 °C, 16 h
10 (100%)
4 (4%) 6 (19%) 10 (45%)
4 (33%) 6 (0%) 10 (59%)
Ru(II)-dppbc (0.01 equiv)
H2 (200 psi), 50 °C, 16 h 4 (3%) 6 (0.3%) 10 (56%)
Cl2Ru(II)-(R)-BINAP (0.01 equiv)
Cl2Ru(II)-(R)-BINAP (0.01 equiv)
Ru(II)-dppbc (0.01 equiv)
H2 (50 psi), 75 °C, 4 h
H2 (200 psi), 75 °C, 4 h
H2 (50 psi), 75 °C, 4 h
H2 (200 psi), 75 °C, 4 h
4 (3.4%) 6 (16%) 10 (42%)
4 (2.8%) 6 (25%) 10 (48%)
4 (8%) 6 (0.7%) 10 (38%)
4 (3.6%) 6 (1.4%) 10 (36%)
Ru(II)-dppbc (0.01 equiv)
Cl2Ru(II)-(R)-BINAP (0.01 equiv)
Cl2Ru(II)-(R)-BINAP-(2R)-(-)
-1,1-bis(4-methoxyphenyl)-3-
methyl-1,2-butadiene (0.1 equiv)
RuCl2(PPh3)2 (0.02 equiv)
RuCl2(PPh3)2 (0.02 equiv)
RuCl2(PPh3)2 (0.02 equiv)
RuCl2(PPh3)2 (0.02 equiv)
RuCl2(PPh3)2 (0.02 equiv)
RuCl2(PPh3)2 (0.02 equiv)
Ru(II)-rac-BINAP (0.1 equiv)
H2 (200 psi), 75 °C, 16 h 4 (6%) 6 (84%) 10 (5%)
H2 (200 psi), 75 °C, 16 h 4 (5%) 6 (83%) 10 (6%)
17
18
19
20
21
22
23
THF
H2 (200 psi), 75 °C, 16 h 4 (9%) 6 (8%) 10 (61%)
H2 (200 psi), 75 °C, 16 h 4 (17%) 6 (19%) 10 (29%)
Cs2CO3 (0.1 equiv), THF
IPA/THF
Cs2CO3 (0.1 equiv), IPA/THF reflux, 16 h
LiBr (6 equiv), THF
KOH (6 equiv), THF/IPA
THF
reflux, 16 h
4 (2%) 6 (0%) 10 (78%)
4 (1%) 6 (2%) 10 (10%)
4 (2%) 6 (2%) 10 (68%)
4 (2%) 6 (9%) 10 (0%)
reflux, 16 h
reflux, 16 h
H2 (200 psi), 75 °C, 16 h 4 (66%) 6 (27%) 10 (0%)
a All the reactions were performed at the 100 mg to 400 mg scale. The completion of the reaction was monitored by HPLC. bObtained in situ from [Ru(p-
c
cymene)Cl2]2+TsDPEN. Obtained in situ from [Cl2(COD)Ru(II)]polymer+dppb.
eliminated. Instead, the periodinane reduction products were
removed from the reaction mixture by crystallization from
methylene chloride. This modification resulted in good yield
(77-85%) and made isolation significantly less time-
consuming and labor-intensive. The oxidation of 4 to 10 was
performed at the 440-g scale under Dess-Martin37 condi-
tions. The crude product was directly purified by crystal-
lization to produce the desired ketone in high purity.
3. Synthesis of 6 from 10. The stereoselective reduction
at the C-3 position of 10 to 6 was reported in the literature
using L-Selectride and LiBr at -78 °C in THF.24a In addition
to the low temperature used for the reaction, the use of
L-Selectride and dry LiBr were identified as potential
problems for scale up because of reaction volume efficiency.
Therefore, our initial efforts were focused at finding alterna-
tive reagents and conditions for the stereoselective reduction.
First, we focused our efforts on a catalytic hydrogenation of
the C-3 ketone functionality. There is a great deal of literature
precedence for the stereoselective reduction of ketones to
alcohols using Ru based catalysts.38 Noyori’s BINAP-Ru (II)
catalysts are recognized as general and efficient catalysts for
the hydrogenation of ketones to alcohols.39,40 However, the
stereoselective reduction of aliphatic ketones remains a
difficult task since it requires the catalyst to differentiate
between alkyl groups. Nonetheless, our initial efforts were
based upon Ru based catalysts, and the results are sum-
marized in Table 2.
The use of various commercially available Ru catalysts
gave satisfactory results. The best catalyst was found to be
Ru(II)-(R)-BINAP (Entry 6, Table 2), which gave 6 in 82%
yield. The final reaction mixture also contained 1.5%
unreacted starting material, 5% of the undesired enantiomer,
and several small unidentified impurities.
Alternatively, the use of Al(OiPr)3 with 2-propanol in
refluxing PhMe under Meerwein-Ponndorf-Verley reduc-
tion conditions41 resulted in preferential reduction to afford
6. The ratio between desired diastereomer 6 and unwanted
4 in the crude reaction mixture was found to be 89:7 (%
AUC). The diastereomeric purity of 6 was significantly
upgraded either by crystallization (most conveniently from
partially evaporated reaction mixture) or by trituration
with methanol. The combination of these two purification
techniques allowed us to improve the diasteriomeric ratios
(6/4) to 650:1. Compound 6 was obtained in high optical
and chemical purity (>99% AUC) and in 80-85% isolated
yield.
(37) Dess-Martin periodinane (DMP) used in the process was purchased from
commercial sources. The literature indicates that the intermediate [1-hy-
droxy-1, 2-benziodoxol-3(1H)-one] is explosive under excessive heating
(>200 °C) or impact [J.B. Plumb & D.J. Harper, Chem. Eng. News 1990,
July 16, 3]. In our experience, the oxidation of the secondary alcohol was
found to be slow when DMP was used in dry methylene chloride and
required several hours for the completion of the reaction. However, when
wet methylene chloride [S.D. Meyer & S.L. Schreiber, J. Org. Chem. 1994,
59, 7549] was used, the reaction required only 2 to 3 h for completion and
a very mild exotherm was observed (∼5 to 8 °C) at the 440-g scale reaction.
We have also observed that the rate of the addition of wet methylene chloride
controls the exotherm and rate of the reaction.
4. Synthesis of 11. The literature describes a procedure
to produce 11 in 40% yield by treating 6 with ethylene glycol
(38) Tang, W.; Zhang, X. Chem. ReV. 2003, 103, 3029 and references cited
therein.
(39) Miyashita, A.; Takaya, H.; Souchi, T.; Noyori, R. Tetrahedron 1984, 40,
1245.
(40) Ohta, T.; Takaya, H.; Noyori, R. Inorg. Chem. 1988, 27, 566.
(41) Wilds, A. L. Org. React. 1944, 2, 178.
Vol. 11, No. 3, 2007 / Organic Process Research & Development
•
425

Scheme 4a
of the undesired higher oligomers. To suppress the formation
of the higher oligomers, the concentration of 4 was increased
(from 2.6 to 4 equiv) so that at any given time excess 4 was
present to react with the electrophile. As a result, the
treatment of 1 equiv of 11 with 4 equiv of 4 improved the
ratio of 12 to higher oligomers from 2.6:1 to 4:1. Compound
12 was isolated from silica gel chromatography in 72-76%
yield with >96% (AUC) purity.
a Reagents and conditions: (i) Ethylene glycol, DDQ, DMAP, CH2Cl2.
Scheme 5a
6. Synthesis of 1. Our initial attempts to prepare
compound 1 under the reported conditions24 led to the desired
product at 88% (AUC) purity where the presence of many
partially benzylated intermediates necessitated purification
by preparative HPLC. The use of reversed phase preparative
HPLC for the isolation of 1 would be tedious and not
economical for any large-scale production. We also ob-
served that during the workup, the amount of byproducts
increased, which indicated the product not to be stable under
such conditions. To improve the yield and purity criteria for
the debenzylation of 1, we examined pressure, solvent,
catalyst, and time of reaction conditions listed in Table 3.
Our evaluations showed that altering the pressure from
1 to 15 psig did not greatly affect the reaction yield (entries
1-4). However, the total amount of time the reaction
was allowed to progress greatly affected the yield (entries
3-4) because the dimer 1 would cleave into monomer
units. Changing the catalyst from 20% Pd(OH)2/C to either
Pd black or 5% Pd/C did not give favorable results (entries
5-8). Using EtOAc rather than THF and MeOH in the
reaction mixtures greatly increased the yield and purity of
the final product (entries 9-10, 13-16). We also observed
that the reaction could not be done exclusively in H2O since
the substrate was insoluble in H2O (entry 11). If the reaction
was done exclusively in EtOAc, the reaction rate was very
fast, but the final yield and purity of the product were <90%
(entry 12). Based on the above results and having an
understanding of the solubility of 12 (soluble in organic
solvent and insoluble in water) and the desired product 1
(soluble in water and partially soluble in organic solvent), it
was decided to run the reaction in a biphasic solution so
that as the product formed, it would transfer to the aqueous
layer (Scheme 6). Thus, the product can be isolated from
the aqueous layer, and most of the intermediates/byproducts
are removed by simple liquid/liquid extraction.
a Reagents and conditions: (i) Bentonite clay K-10, CH2Cl2, 0 °C; (ii) column
chromatography.
Scheme 6a
a Reagents and conditions: (i) 20% Pd(OH)2/C (50% H2O wet), EtOAc/H2O
(1/3, v/v), RT, 15 psi, 3-4 h.
in DCM with DDQ and DMAP (Scheme 4).24a The procedure
required repeated column chromatographies to obtain the
desired purity.
Our attempt to replace DDQ with less expensive and safer
reagents such as IBX or chloranil did not produce even a
trace of the desired compound. Thus, it was decided to use
DDQ as an oxidizing agent but modify the isolation
conditions. Decreasing the concentration of ethyl acetate
inthe heptane-ethyl acetate eluent allowed us to replace a
two-step column chromatography method with a single
column chromatography step. Additional purification of 11
was achieved by crystallization from heptane and ethyl
acetate. Compound 11 was obtained in good yield (>70%)
and purity (>99% AUC).
The final solvent conditions using EtOAc/H2O gave the
desired product in quantitative yield and 98% purity.
7. Synthesis of 13. The aforementioned conditions for
the coupling reaction between 11 and 4 in the presence of
Bentonite K-10 were also applied to the coupling reaction
between 11 and 6 (Scheme 7). Most of the unreacted 6 was
removed from the evaporated reaction mixture by crystal-
lization from ethyl acetate.
Final separation was achieved by column chromatography
on silica gel. The desired product 13 was isolated in 72-
78% yield with >98% purity.
8. Synthesis of 14. The galloylation of 13 using tri-O-
benzyloxy galloyl chloride in the presence of DMAP in
C5H5N produced 14 in high yield (Scheme 8). Trituration
of the reaction mixture with CH2Cl2 at RT caused precipita-
5. Synthesis of 12. The coupling reaction between 8 and
4 was performed by using Bentonite clay K-10 at 0 °C as
reported by Kozikowski et al.27 (Scheme 5). In addition to 4
and 12, higher oligomers were also detected in the reaction
mixture.
Initial observations from TLC and HPLC analysis re-
vealed that the low yield of 12 was caused by the formation
426
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development

Table 3. Conditions for the debenzylation of 12
Scheme 7a
temp time pressure
entry
1
solvent(s)
(°C) (h)
(psig)
catalyst yield
THF/MeOH/
H2O
25
25
25
25
1.5
3
1-2
20% Pd
(OH)2/C
18%
88%
86%
48%
30 wt %
20% Pd
2
3
4
THF/MeOH/
H2O
1-2
15
(OH)2/C
30 wt %
20% Pd
THF/MeOH/
H2O
2
(OH)2/C
30 wt %
20% Pd
THF/MeOH/
H2O
5
15
(OH)2/C
30 wt %
Pd Black
60 wt %
Pd Black
60 wt %
Pd Black
5 wt %
5
6
7
8
9
THF/MeOH/
H2O
THF/MeOH/
H2O
THF/MeOH/
H2O
THF/MeOH/
H2O
25
25
25
25
1.5
3
1-2
1-2
1-2
15
84%
90%
0%
1.5
1
a Reagents and conditions: (i) Bentonite clay K-10, CH2Cl2, 0 °C; (ii) column
chromatography.
5% Pd/C
120 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
20% Pd
(OH)2/C
30 wt %
22%
88%
Scheme 8a
EtOAc/MeOH/ 25
H2O
1
10
10 EtOAc/MeOH/ 25
H2O
1
1
1
1
4
4
4
15
15
15
15
15
15
15
96%
0%
11 H2O
25
25
25
25
25
25
12 EtOAc
82%
2%
13 EtOAc/H2O
14 EtOAc/H2O
15 EtOAc/H2O
16 EtOAc/H2O
86%
96%
98%
a Reactions and conditions: (i) Galloyl acid, (COCl)2, Et3N, CH2Cl2, 0 °C;
(ii) silica gel plug.
Scheme 9a
tion of 3,4,5-tri-O-benzyl gallic acid, which was removed
by filtration. The product was then purified by passage
through a silica gel plug. This reaction was efficiently
performed several times in the laboratory setting.
9. Synthesis of 2. The hydrogenation of 14 to produce 2
was performed under similar conditions as those previously
described for the synthesis of 1, except for slight modifica-
tions in the workup (Scheme 9). These are outlined in the
Experimental Section.
a Reagents and conditions: (i) 20% Pd(OH)2/C (50% H2O wet), EtOAc/H2O
(1/3, v/v), RT, 15 psi, 3-4 h.
Summary
In summary, the scale-up synthesis of compounds 1 and
2 has been described with several improvements. The
benzylation of 3 was achieved using K2CO3 as a base in
DMF, and the desired product was isolated after crystalliza-
tion. Compounds 10 and 11 were isolated in good yield after
improving the reaction and isolation conditions. The conver-
sion of compound 10 to 6 was achieved by stereoselective
reduction with Al(iOPr)3 and IPA and had excellent selectiv-
ity. The desired compound was isolated after trituration with
methanol. The debenzylation conditions (to obtain 1 and 2)
were optimized using a combination of water and ethyl
acetate as a solvent in 3/1 ratio (v/v). The desired compounds
were isolated in good yield and high purity after the
extractive workup, thus avoiding either the chromatographic
or HPLC purification.
Vol. 11, No. 3, 2007 / Organic Process Research & Development
•
427

Experimental Section
mixture showed complete consumption of the starting
material. Saturated NaHCO3 (4 L) solution was slowly added,
followed by a 10% aqueous solution of sodium thiosulfate
pentahydrate (161.5 g /1.6 L in water). The white precipitate
formed upon quenching was suction filtered. The filtrate was
transferred to a separatory funnel. The organic layer was
separated, and the aqueous phase was extracted with CH2-
Cl2 (650 mL). The combined organic phase was dried over
MgSO4 (Note: During the drying process, the reduced DMP
reagent precipitates). Once the precipitation process was
completed, the reaction mixture was suction filtered, and
the solvent was removed under vacuum. The residue was
triturated with methanol (600 mL) for 1 h at RT, filtered
(filtrate saved), dissolved in boiling dichloromethane
(800 mL), and diluted with methanol (1.5 L). Precipitation
was observed when the reaction mixture cooled to RT.
After stirring for 18 h at RT and subsequently at ice
bath temperature for 1 h, the solids were suction filtered
and washed with methanol (4 × 100 mL). The re
sulting pale pink precipitate was dried in Vacuo to
afford 10. Yield ) 336 g, 76.6%. HPLC purity ) 94%
(AUC).
5,7,3′,4′-Tetra-O-benzyl-(-)-epicatechin (6). A 22-L,
three-necked round-bottom flask equipped with a heating
mantle, an overhead stirrer, a thermometer, and a distillation
unit was charged with 10 (1263 g, 1.95 mol, 1 equiv), PhMe
(10 L), Al(OiPr)3 (796.6 g, 3.9 mol, 2 equiv), and IPA (5 L)
with agitation. A slightly yellow turbid solution was obtained
after stirring at RT for ∼30 min (Note: A mild endotherm
was obserVed as the internal temperature dropped to 14.3
°C). The suspension was heated to reflux with continuous
stirring. As the internal temperature increased, the suspension
became a clear yellow solution. As the internal temperature
reached g82 °C (Note: It took about 1.75 to 2 h at this
scale to reach this temperature), the distillation of the solvent
(acetone and IPA) began and was collected in a 2 L round-
bottom flask. After collecting about 1400 mL of distillate,
HPLC analysis of the reaction mixture indicated the presence
of unreacted starting material. Additional IPA (500 mL) was
added to the reaction mixture, and the distillation was
continued. After an additional 2.2 L of distillate were
collected, another sample was submitted for HPLC analysis.
(Note: HPLC results indicated the consumption of the
starting material.) The reaction mixture was cooled to RT
(IT ) 18.8 °C). To the reaction mixture was slowly added
10% aq. H2SO4 (v/v, 3 L) with stirring. The IT rose to
47.2 °C (Note: Initial addition of aqueous H2SO4 resulted
in a gel, which dissolVed as more acid was added to the
reaction mixture with good agitation. At the end of the
addition, a clear biphasic solution was obtained). The
mixture was allowed to cool to RT. Once the IT reached
RT, the reaction mixture was transferred to a separatory
funnel, and the organic layer was separated. The organic layer
was washed with 10% aqueous H2SO4 (v/v, 1 × 3 L). The
combined aqueous layers were washed with toluene (1 × 3
L). The organic layers were combined, washed with 20%
aq. NaCl (w/v, 1 × 2.5 L), dried over Na2SO4 (750 g), and
filtered. The solvent was removed under vacuum keeping
General. Solvents and reagents were obtained from
commercial sources and were used without any purification.
All the reactions were performed in glass reactors. Different
HPLC methods were used for the benzylated and phenolic
compounds using standard HPLC equipment with PDA
detection and data systems. Separations were performed with
a Phenomenex Synergi 4 µ Fusion-RP 80 Å (150 mm × 4.6
mm) column and Chiralpak AD-RH (150 mm × 4.6 mm)
column. The mobile phase for isocratic and gradient HPLC
separations was prepared using 0.01% TFA in water and
0.01% TFA in acetonitrile. All the compounds were moni-
tored against reference materials including the starting
materials.
5,7,3′,4′-Tetra-O-benzyl-(+)-catechin (4). A dry 12 L,
three-necked round-bottom flask equipped with a mechanical
stirrer, a dropping funnel, a N2 inlet, and an internal
temperature probe was charged with 3 (400 g, 1.38 mol, 1
equiv) and DMF (4 L, 1 g/10 mL, 10 vol). To this solution
was slowly added K2CO3 (1430.5 g, 10.38 mol, 7.5 equiv)
with stirring. The suspension was allowed to stir at RT for
0.5 h. To this was slowly added BnBr (1180.2 g, 6.9 mol, 5
equiv) via the addition funnel (Note: A mild exotherm was
obserVed as the internal temperature rose to 30.6 °C from
21.6 °C). It took about 4.5 h to complete the addition of
benzyl bromide at this scale. The suspension was then
allowed to stir at RT for 18 h. The consumption of the
starting material was monitored by TLC (30% EtOAc/
heptane, v/v). After complete consumption of the starting
material, the reaction mixture was suction filtered through a
pad of celite (500 g) to remove K2CO3. The celite pad was
washed with EtOAc (3 × 1 L, 3 × 500 mL). The combined
filtrates were sequentially washed with 10% aqueous HCl
(2 × 1.5 L), H2O (2 × 1 L), and 30% aqueous NaCl (1 ×
2 L). The organic layer was dried over anhydrous MgSO4
(300-g) and filtered, and the solvent was removed under
vacuum to afford an off-white to light yellow colored
semisolid. The semisolid was chased with heptane (2 × 500
mL). Trichloroethylene (2 L, 1 g/5 mL based on the starting
material, 3) was added to the solid and heated at reflux until
a clear orange to red solution was obtained. The solution
was first allowed to cool to RT with agitation, and then it
was further cooled to -20 to -26 °C in the freezer for 56
h. The solids obtained were suction filtered and washed with
cold trichloroethylene (-20 °C, 2 × 500 mL) and cold
heptane (-20 °C, 1 × 500 mL). The solids were dried under
high vacuum at 50-55 °C for 18 h to produce 4 as an off-
white to white solid. Yield ) 412 g, 46%. HPLC purity )
98% (AUC).
(2R)-5,7,3′,4′-Tetrakis(benzyloxy)flavan-3-one (10). To
a solution of 4 (440 g, 0.675 mol, 1 equiv) in dichlo-
romethane (3.2 L) was added at once with stirring at
RT Dess-Martin periodinane reagent (315.4 g, 0.74 mol,
1.1 equiv) in one portion. CH2Cl2 saturated with water
(242 mL) was added dropwise during 1.5 h. The internal
temperature (IT) gradually increased from 16 to 24 °C. The
IT reached its maximum in approximately 50 min then
gradually decreased to 20 °C. HPLC analysis of the reaction
428
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development

the bath temperature e45 °C to produce a light yellow
semisolid. The solids were triturated with methanol (1 g/8
mL, 8 L) at RT for 19 h. The solids were suction filtered,
washed with methanol (5 × 500 mL), and dried under high
vacuum at 40-45 °C for 20 h to give 6. Yield ) 1018.18 g,
80.3%. HPLC purity ) 99% (AUC).
completion of the reaction was monitored by HPLC and TLC
(EtOAc /CHCl3/heptane: 1/14/14, v/v/v). The reaction
mixture was filtered through a pad of celite to remove
bentonite K-10. The celite pad was washed with EtOAc (2
× 20 mL). The filtrates were combined, and the solvents
were removed under vacuum to afford a foamy solid. Most
of the unreacted 6 was isolated by dissolving the foamy solid
with boiling EtOAc (750 mL) followed by cooling the
solution to RT with agitation. The reaction mixture was then
purified by silica gel chromatography using ethyl acetate and
heptane as an eluent to afford 13. Yield ) 93 g, 60.8%.
HPLC purity ) 98.5% (AUC).
5,7,3′,4′-Tetra-O-benzyl-3-O-(3,4,5-tri-O-benzylgalloyl)-
(-)-epicatechin-(4â,8)-[5,7,3′,4′-tetra-O-benzyl-3-O-(3,4,5-
tri-O-benzylgalloyl)-(-)-epicatechin] (14). To a suspension
of tri-O-benzyl gallic acid (206.96 g, 470 mol, 5 equiv) and
DMF (1.62 mL, catalytic amount) in anhydrous CH2Cl2 (3.27
L) was slowly added oxalyl chloride (65.33 g, 44.9 mL, 515
mol, 5.5 equiv) at RT with agitation under N2. The reaction
mixture was stirred for 1 h at RT. Additional oxalyl chloride
(4 mL) and DMF (0.5 mL) were added to the reaction
mixture. After 1.5 h, the reaction mixture was concentrated
under vaccum. The residue was chased with PhMe (2 × 500
mL). To this was then added a solution of 13 (122 g, 93.8
mol, 1 equiv) in dry pyridine (2.72L), and the reaction
mixture was stirred at RT for 90 h. H2O (135 mL) was added
to the reaction mixture, and the stirring was continued for
an additional 4 h at RT. The reaction mixture was diluted
with CH2Cl2 (6 L) and 25% aq HCl (600 mL), respectively.
The organic layer was separated, and the aqueous layer was
extracted with CH2Cl2 (1 × 2 L). The combined organic
layers were washed with brine (750 mL), dried over Na2-
SO4 (some of the tri-O-benzyl gallic acid was precipitated
during the storage at RT), and was filtered. The solvent was
removed in Vacuo to afford a semisolid, which was purified
by a silica gel plug (2.5 kg) using CH2Cl2 as an eluent. The
crude product was further purified on a silica gel column
using heptane/CHCl3 /EtOAc, 14 /14/1 (v/v/v) to produce
14 as an off-white foamy solid. Yield ) 161 g, 80%. HPLC
purity ) 96% (AUC).
(-)-Epicatechin-(4â,8)-(+)-catechin (1). To a 6 L pres-
sure bottle was added 20% Pd(OH)2/C (50% wet, 48 g,
60 wt %). To this a solution of 12 (80 g, 0.062 mol) in
EtOAc (HPLC grade, 300 mL) was added followed by H2O
(HPLC grade, 900 mL). The bottle was sealed and purged
with N2 (15 psi) three times and then with H2 at 15 psi (three
times). The reactor was pressurized with hydrogen (15 psi),
and stirring was started. After stirring for 3 h at RT, the
reactor was vented and purged with N2 (three times).
The reaction mixture was filtered through a cartridge
(Millipore, Opticap 4′′, 0.22 µm) directly into a separatory
funnel containing hexane (300 mL). The aqueous layer was
separated. The vessel and cartridge were washed with H2O
(2 × 1 L). Each time the aqueous layer was separated and
combined with the other aqueous layers. The aqueous layer
was frozen and lyophilized to afford 1 as a pale yellow
solid. Yield ) 35.1 g, quantitative. HPLC purity ) 96%
(AUC).
5,7,3′,4′-Tetra-O-benzyl-4-(2-hydroxyethoxy)-(-)-epi-
catechin (11). To a stirred solution of 6 (219 g, 0.35 mol,
1.0 equiv) in CH2Cl2 (2.2 L) was added anhydrous ethylene
glycol (112.5 mL, 2.0 mol, 5.7 equiv) and DDQ (100 g, 0.44
mol, 1.26 equiv) at RT. The color of the reaction mixture
turned green and then almost black. After the reaction
mixture was stirred at RT for 2 h, a solution of DMAP (87
g, 0.71 mol, 2.0 equiv) in CH2Cl2 (500 mL) was added, and
the reaction mixture was further stirred for an additional 10
min. Silica gel (1 kg) was added, and the reaction mixture
was dried under vacuum for 20 h at RT. The dry silica gel
was placed on top of a 400 g silica gel column, and the
product was eluted with heptane-ethyl acetate (2/1, v/v, 14
L). Fractions containing pure product (TLC and HPLC
analysis) were combined, and the solvent was removed under
vacuum. The residue was dissolved in boiling ethyl acetate
(200 mL) and upon cooling diluted with heptane (200 mL).
Crystallization began after a few minutes. The resulting
suspension was vigorously stirred overnight at RT for 18 h.
The solid was filtered, washed with heptane, and dried in
Vacuo to afford 11 as an off-white solid. Yield ) 171 g,
71.5%. HPLC purity ) 99% (AUC).
5,7,3′,4′-Tetra-O-benzyl-(-)-epicatechin-(4â,8)-[5,7,3′,4′-
tetra-O-benzyl-(+)-catechin] (12). To an ice cold (IT
<5 °C) suspension of 4 (1189.7 g, 1.83 mol, 4.46 equiv)
and Bentonite K-10 (574 g) in CH2Cl2 (10 L) under N2
atmosphere was slowly added a solution of 11 (290 g, 0.41
mol, 1.0 equiv) in CH2Cl2 (1 L) with stirring at such a rate
that the internal temperature was maintained at <6 °C
throughout the addition (Note: It took ∼1.5 h at this scale
for the addition). The reaction mixture was stirred at this
temperature for an additional 1 h (the IT rose to 10 °C).
The clay was suction filtered through a pad of celite, and
the celite was washed with CH2Cl2 (2 L). The filtrates were
combined, and the solvent was removed under vacuum to
produce an off-white solid. The solid was analyzed by HPLC
indicating a mixture of 4 (excess), the desired dimer 12, and
higher oligomers. Most of the unreacted 4 was crystallized
from ethyl acetate. The remaining reaction mixture was
purified by silica gel chromatography using ethyl acetate and
heptane as an eluent to give 12. Yield ) 323.4 g, 61%. HPLC
purity ) 98.7% (AUC).
5,7,3′,4′-Tetra-O-benzyl-(-)-epicatechin-(4â,8)-[5,7,3′,4′-
tetra-O-benzyl-(-)-epicatechin] (13). A suspension of 6
(306 g, 0.47 mol, 4.0 equiv) and bentonite K-10 (165 g) in
CH2Cl2 (3650 mL) was cooled in an ice bath under N2
atmosphere. To this was slowly added a solution of 11 (83.6
g, 0.12 mol, 1.0 equiv) in CH2Cl2 (150 mL) with stirring
while keeping the IT <5 °C throughout the addition (Note:
It took ∼2.5 h for the addition at this scale). The reaction
mixture was stirred for an additional 1 h at ice-bath
temperature (The internal temperature rose to ∼10 °C). The
Vol. 11, No. 3, 2007 / Organic Process Research & Development
•
429

(-)-Epicatechin-3-O-galloyl-(4â,8)-(-)-epicatechin-3-
O-gallate (2). In a 6 L hydrogenation glass apparatus was
added a solution of 14 (41.6 g, 0.466 mol, 96% purity)
in EtOAc (228 mL) to a suspension of 20% Pd(OH)2/C (60
wt % wet, 25 g) in water (HPLC grade, 685 mL) at RT with
stirring. The reaction vessel was purged with N2 (twice)
followed by H2, and the reaction was allowed to stir at RT
in the presence of 15 psi H2 for 4 h. After the reaction
mixture was purged with N2, it was filtered through a filter
cartridge (Millipore, Opticap, 4′′, 0.22 µm). The cartridge
was washed with H2O (300 mL), EtOAc (1 L), and warm
water (25-30 °C, 1 L). The combined washes were
transferred to a 6 L separatory funnel, and hexane (300 mL)
was added (Note: the organic layer turns cloudy. Also,
addition of hexane helps in the separation of the layers).
The aqueous layer was separated. The cartridge was again
washed with EtOAc (1 L) and warm water (25-30 °C, 1
L). Again the aqueous layer was separated. The organic
layers were combined and washed with H2O (1 L). The
aqueous layers were frozen and lyophilized for 72 h to
afford 2 as an off-white solid. Yield ) 15.6 g, 91%. HPLC
purity ) 98.5% (AUC).
Acknowledgment
The authors are grateful to Drs. Heather Taft, Daniel
Coughlin, and Richard Lombardy for their support, encour-
agement, and valuable discussions during the course of this
work.
Supporting Information Available
HPLC characterization data and NMR spectra assignments
for 1, 2, 4, 6, 10, 11, 12, 13, and 14. This material is available
free of charge via the Internet at http://pubs.acs.org.
Received for review February 9, 2007.
OP700031N
430
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development