Development of an Efficient Process for the Decomposition of the Borate Complexes Formed during the Large-Scale Synthesis of (S)-1,2,4-Butanetriol

Article abstract of DOI:10.1021/op400271k

An improved multikilogram-scale process for the production of (S)-1,2,4-butanetriol has been developed. This process involves the efficient removal of residual boric acid and the decomposition of the borate complexes formed during the reduction of (-)-dimethyl malate with sodium borohydride by methanolysis using a circular distillation-coupled hydrolysis apparatus.

Full text of DOI:10.1021/op400271k

Article
pubs.acs.org/OPRD
Development of an Efficient Process for the Decomposition of the
Borate Complexes Formed during the Large-Scale Synthesis of
(S)‑1,2,4-Butanetriol
Qian Liu, Fang-Jun Xiong, Qiu-Qin He,* and Fen-Er Chen*
Department of Chemistry, Fudan University, Handan Road 220, Shanghai 200433, China
ABSTRACT: An improved multikilogram-scale process for the production of (S)-1,2,4-butanetriol has been developed. This
process involves the efficient removal of residual boric acid and the decomposition of the borate complexes formed during the
reduction of (−)-dimethyl malate with sodium borohydride by methanolysis using a circular distillation-coupled hydrolysis
apparatus.
INTRODUCTION
Scheme 1. Reduction of 7 with NaBH
Compound 1
To Give the Target
4
■
(
S)-1,2,4-Butanetriol (1) is a useful building block for the
synthesis of HMG-CoA reductase inhibitors such as lovastatin
2), rosuvastatin (3), and atorvastatin (4) as well as the
synthesis of antineoplastic natural products such as acutiphycin
(
1
,2
(
5) and didehydroacutiphycin (6) (Figure 1).
convenient, and scalable method for the synthesis of 1 involving
the use of circular distillation-coupled hydrolysis for the
removal of residual boric acid as well as the cleavage of the
borate complexes formed during the reduction of 7 with
NaBH4.
RESULTS AND DISCUSSION
At the beginning of our own investigation, we repeated the
procedure for the reduction of 7 using NaBH and attempted
■
4
to purify the crude triol product 1 by distillation in the same
4
way as Izawa et al. Not surprisingly, this procedure did not
provide any of the desired product and resulted only in the
formation of decomposition byproducts. The failure of this
process was attributed to the formation of boric acid during the
reduction of 7 with NaBH , which could readily chelate with 1
4
5
to form borate complexes (Scheme 2). With this in mind, the
Scheme 2. Complexation Reaction of Boric Acid with Multi-
Hydroxylated Compounds
Figure 1. Structures of several HMG-CoA reductase inhibitors and
antineoplastic natural products.
Although a variety of different methods for the synthesis of 1
3
from chiral starting materials have been reported, the chiron
route starting from L-malic acid is currently used as an industrial
development of a successful process for the production of high-
quality 1 would require the effective removal of any residual
boric acid as well as the cleavage of the borate complexes prior
to the purification of the crude product by direct distillation.
Several different methods of purification have been
developed for the removal of borate complexes and boric
process for synthesis of 1, with the reduction of (−)-dimethyl
malate (7) with sodium borohydride (NaBH ) being the final
4
4
step in the process (Scheme 1). Izawa et al. recently attempted
to produce high-purity 1 via distillation of the crude product
resulting from the reduction of 7. Unfortunately, however, their
efforts in this regard were unsuccessful, as only the decomposed
byproduct was detected. This problem has therefore severely
limited the production and subsequent application of this
material. Herein we report the development of a high-yielding,
Received: September 30, 2013
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/op400271k | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
acid from the crude triol 1, including the use of
3
d
3b
chromatography and ion-exchange resins. However, the
use of chromatography on the industrial scale is impractical and
expensive, whereas ion-exchange resins such as the glucos-
amine-specific borate adsorption resin IRA-743 or XE-583 can
lose their capacity after only a few cycles. Because of these
limitations, it is clear that there is an urgent need for the
development of an alternative manufacturing-scale procedure
for the removal of the borate-based impurities that are formed
during the synthesis of 1.
It is well-known that boric acid can undergo methanolysis to
6
give trimethyl borate, which can be readily removed from the
crude product mixture by codistillation with methanol. It was
envisaged that the borate complexes would undergo the reverse
reaction shown in Scheme 2 following the removal of boric
acid, giving the free triol 1 and boric acid; the latter of these
two products would go on to react with methanol and be
evaporated, leaving only triol 1.
It was estimated that at least 160 volumes of methanol would
be required to allow for the complete removal of the residual
boric acid and borate complexes from the crude product,
suggesting that the cleavage of the borate complexes was an
equilibrium reaction that was slower than the corresponding
complex-forming process (Scheme 3). On the basis of this
Scheme 3. Equilibrium between the Borate Complexes and
Triol 1
Figure 2. Apparatus for the circular distillation-coupled hydrolysis
process for the decomposition of the borate complexes.
requirement for a high loading of methanol in a single run, the
inclusion of some process for recycling the methanol would be
necessary. It is well-known that trimethyl borate is moisture-
and base-sensitive, so the decision was made to quench the
vaporized trimethyl borate with sodium hydroxide solution to
give methanol, which could be recycled using our newly
designed circular distillation-coupled hydrolysis apparatus
1
,2,4-butanetriol (1) without any of the decomposition
byproduct.
CONCLUSION
■
An efficient, facile, and economical circular distillation process
for the purification of (S)-1,2,4-butanetriol (1) has been
developed. This method allows for the cleavage of the borate
complexes to afford the free triol product as well as the
simultaneous removal of any residual boric acid and the
recycling of the reaction solvent. Furthermore, this process
could be applied universally for the removal of residual boric
acid and borate complexes, especially for highly water-soluble
and thermally sensitive products.
(
Figure 2).
As shown in Figure 2, the apparatus consists of two
condensers, a packed column, a sample point, and two reactors.
The reactors themselves are a glass-lined reactor (A) and a 316
stainless steel reactor (B). Reactor A is used to contain a
methanol solution of the crude reduction mixture containing
triol 1, its borate complexes, and the residual boric acid,
whereas reactor B is equipped with a packed distillation column
and charged with methanol, water, and sodium hydroxide. In
practice, both of the reactors are heated, and the boric acid in
reactor A is converted to trimethyl borate before being
codistilled with methanol into reactor B. As the concentration
of boric acid in the crude product mixture is reduced, the borate
complexes gradually decompose to generate the free triol and
more boric acid. In reactor B, the trimethyl borate derived from
the distillation in reactor A is immediately saponified by sodium
hydroxide, and the resulting methanol is evaporated and
transferred back to reactor A. In this way, an effective circular
methanol flow is established, enabling the recycling of the
solvent. In addition, the terminal point can be readily
determined by checking the combustion until the green flame
caused by trimethyl borate disappears. Finally, the solution in
reactor A is concentrated and distilled to give high-purity (S)-
EXPERIMENTAL SECTION
General. All of the reagents used in the current study were
■
purchased from commercial suppliers and used without further
1
13
purification. the H and C NMR spectra were recorded at 400
MHz on Bruker instruments in D O, and the chemical shift
values are reported as parts per million relative to TMS, which
was used an internal standard. Specific rotations were
determined using a PerkinElmer polarimeter (model 341-LC).
(S)-1,2,4-Butanetriol (1). A 200 L glass-lined reactor was
2
charged with tetrahydrofuran (100 L) and NaBH (7.2 kg, 180
4
mol), and the resulting mixture was cooled to −5 °C with
stirring. (−)-Dimethyl malate (24.8 kg, 153 mol) was then
added through a piston-type metering pump at a rate of 6 kg/h
whilst the temperature was maintained in the range of −5 to 0
°C. Upon completion of the addition, the mixture was stirred
B
dx.doi.org/10.1021/op400271k | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
for 10 h in the same temperature range. Upon completion of
the reaction (as determined by TLC), 36% (w/w) hydrochloric
acid (19 kg, 187 mol) was slowly charged to the stirred mixture
whilst maintaining the temperature in the range of −5 to 0 °C.
The resulting mixture was stirred for 1 h at same temperature
before being filtered. The filter cake was then washed with
tetrahydrofuran (3 × 5 L), and the combined filtrate and
washings were concentrated under vacuum to give the crude
product as a viscous liquid containing a small quantity of white
solid. The crude product was then placed in a 200 L glass-lined
reactor (A), followed by methanol (80 L). Methanol (48 L),
water (30 L), and NaOH (2 kg, 50 mol) were then added
sequentially to a separate reactor (B). Both reactors were
heated to 85 °C, and the azeotropic mixture of methanol and
trimethyl borate that was generated in reactor A was transferred
to reactor B, where the trimethyl borate was hydrolyzed to give
methanol, which was evaporated and condensed back into
reactor A. The reaction was monitored by combustion until the
green flame of trimethyl borate disappeared (ca. 10 h), at which
point the solution in reactor A was sequentially concentrated
and distilled (181−184 °C/20 mmHg) to give (S)-1,2,4-
(6) Schlesinger, H. I.; Brown, H. C.; Mayfield, D. L.; Gilbreath, J. R. J.
Am. Chem. Soc. 1953, 75 (1), 213−215.
butanetriol (12.6 kg, 76% yield for two steps) as a light-yellow
1
oil. H NMR (D O): δ 1.57 (1H, m), 1.66 (1H, m), 3.40 (1H,
2
13
m), 3.51 (1H, m), 3.63 (2H, m), 3.73 (1H, m). C NMR
D O): δ 34.74, 58.29, 65.54, 68.85. [α] = −25.2 (c = 1,
2
0
(
2
D
3a
20
MeOH) (lit [α]D = −25.5°).
AUTHOR INFORMATION
Corresponding Authors
F.-E.C.: E-mail: rfchen@fudan.edu.cn. Telephone: +86-21-
5643811.
Q.-Q.H.: E-mail: qqhe@fudan.edu.cn. Telephone: +86-21-
5643811.
■
*
6
*
6
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Dr. Fr-Chen for his continued support of
this research project.
REFERENCES
■
(
1) (a) Clive, D. L. J.; Murthy, K. S. K.; Wee, A. G. H.; Prasad, J. S.;
Da Silva, G. V. J.; Majewski, M.; Anderson, P. C.; Evans, C. F.;
Haugen, R. D. J. Am. Chem. Soc. 1990, 112 (8), 3018−3028. (b) Kim,
J. S.; Seok, M. G. KR20100087931, 2010.
(
2) Smith, A. B.; Chen, S. S. Y.; Nelson, F. C.; Reichert, J. M.;
Salvatore, B. A. J. Am. Chem. Soc. 1995, 117 (48), 12013−12014.
3) (a) Tandon, V. K.; Van Leusen, A. M.; Wynberg, H. J. Org. Chem.
(
1
983, 48 (16), 2767−2769. (b) Furukawa, Y.; Ohtaka, S.
WO2003106392 A1, 2003. (c) Shibata, T.; Iino, K.; Sugimura, Y.
Heterocycles 1986, 24 (5), 1331−1346. (d) Pawlak, J.; Nakanishi, K.;
Iwashita, T.; Borowski, E. J. Org. Chem. 1987, 52 (13), 2896−2901.
(
(
e) Sun, X. L.; Wu, Y. L. Chin. J. Org. Chem. 2002, 22 (7), 501−503.
f) MacNeil, P. A.; Roberts, N. K.; Bosnich, B. J. Am. Chem. Soc. 1981,
1
03 (9), 2273−2280. (g) Hayashi, H.; Nakanishi, K.; Brandon, C.;
Marmur, J. J. Am. Chem. Soc. 1973, 95 (26), 8749−8757.
4) Honda, Y.; Katayama, S.; Kojima, M.; Suzuki, T.; Kishibata, N.;
Izawa, K. Org. Biomol. Chem. 2004, 2 (14), 2061−2070.
5) (a) Pizer, R.; Ricatto, P. J. Inorg. Chem. 1994, 33 (11), 2402−
406. (b) Miyazaki, Y.; Yoshimura, K.; Miura, Y.; Sakashita, H.;
(
(
2
Ishimaru, K. Polyhedron 2003, 22 (6), 909−916. (c) Rietjens, M.;
Steenbergen, P. A. Eur. J. Inorg. Chem. 2005, 2005 (6), 1162−1174.
(d) Chiappe, C.; Signori, F.; Valentini, G.; Marchetti, L.; Pomelli, C.
S.; Bellina, F. J. Phys. Chem. B 2010, 114 (15), 5082−5088.
C
dx.doi.org/10.1021/op400271k | Org. Process Res. Dev. XXXX, XXX, XXX−XXX