Large-Scale Stereoselective Synthesis of 1,3-Oxathiolane Nucleoside, Lamivudine, via ZrCl4-Mediated N-Glycosylation

Article abstract of DOI:10.1021/acs.oprd.9b00414

A stereoselective large-scale synthetic process is described to produce 1,3-oxathiolane nucleoside, lamivudine. A mild, inexpensive, and readily available zirconium (IV) chloride (ZrCl₄) catalyst acts as a substrate activator for the key N-glycosylation step at room temperature. An optimum of 0.5 equiv of ZrCl₄ is required, which gives encouraging results with respect to chemical efficiency and stereoselectivity. The focus of this work was to develop a new Lewis acid catalyst for N-glycosylation reaction that permits mild and selective synthesis of lamivudine at a large scale. It allowed preferential formation of a single isomer of nucleoside out of four possible stereoisomers, starting from the corresponding 1,3-oxathiolane acetate substrate (racemic and/or diastereomeric mixture of isomers). The thermal behavior for the critical N-glycosylation step was also studied by differential scanning calorimetry and reaction calorimetry techniques.

Full text of DOI:10.1021/acs.oprd.9b00414

Subscriber access provided by La Trobe University Library
Full Paper
Large-Scale Stereoselective Synthesis of 1, 3-Oxathiolane
4
Nucleoside, Lamivudine via ZrCl-Mediated N-Glycosylation
Umesh P. Aher, Dhananjai Srivastava, HARISHCHANDRA
JADHAV, Girij Pal Singh, Jayashree B.S., and Gautham G. Shenoy
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00414 • Publication Date (Web): 19 Feb 2020
Downloaded from pubs.acs.org on February 22, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Large-Scale Stereoselective Synthesis of 1, 3-
Oxathiolane Nucleoside, Lamivudine via ZrCl4-
Mediated N-Glycosylation
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
*
†
†
†
†
Umesh P. Aher ’ , Dhananjai Srivastava , Harishchandra S. Jadhav , Girij P. Singh ,
§
§
Jayashree B. S and Gautham G. Shenoy
†Chemical Research Department, Lupin Research Park, Lupin Limited, 46A /47A, Village
Nande, Taluka Mulshi, Pune-412115, Maharashtra, India
§Department of Pharmaceutical Chemistry, Manipal Academy of Higher Education (MAHE),
Manipal-576104, Karnataka, India
KEYWORDS: Lamivudine (3TC), N-glycosylation, Stereoselectivity, ZrCl4
ABSTRACT: A stereoselective large-scale synthetic process is described to produce 1, 3 –
oxathiolane nucleoside, lamivudine (3TC). A mild, inexpensive and readily available zirconium
(IV) chloride (ZrCl ) catalyst acts as a substrate activator for the key N-glycosylation step at room
4
temperature. An optimum 0.5 equiv of ZrCl is required which gives encouraging results with
4
respect to chemical efficiency and stereoselectivity. The focus of this work was to develop a new
Lewis acid (LA) catalyst for N-glycosylation reaction that permits mild and selective synthesis of
lamivudine at large scale. It allowed preferential formation of single isomer of nucleoside out of
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
four possible stereoisomers starting from the corresponding 1, 3-oxathiolane acetate substrate
(racemic and/or diastereomeric mixture of isomers). The thermal behavior for the critical N-
glycosylation step was also studied by differential scanning calorimetry (DSC) and reaction
calorimetry (RC1) techniques.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TM
INTRODUCTION: Lamivudine (3TC , 1) (Figure 1.) is a first-generation FDA certified
nucleoside reverse transcriptase inhibitor (NRTI) drug class used to treat HIV-1 infection as well
as chronic hepatitis B virus (HBV) in the 1990s.1
Proprietary names where Lamivudine is used as
O
N
NH2
1
Active pharmaceutical ingradient
N
®
O
Epivir
®
Combivir
Trizivir
Epzicom
Triumeq
HO
S
®
®
1
®
Lamivudine (3TC)
®
Temixys
Figure 1. 1, 3-oxathiolane nucleoside, lamivudine (3TC) is a currently approved antiviral agent.
In spite of the fact that, lamivudine was endorsed over two decades back, it still remains as a key
segment of first-line treatment for HIV on account of its virological efficacy and capacity to be
2
combined with further antiretroviral agents in customary and new combination therapies. It has
four stereoisomers however, investigators first reported racemic mixture of lamivudine. It is also
3
known as (±)-BCH-189 to treat HIV-1. BCH-189 enantiomers were further separated out by a
method of chiral high pressure liquid chromatography and found to be resistant to both types 1 and
2 of human immunodeficiency viruses. However, the (-)-enantiomer (3TC) has been discovered
substantially less cytotoxic than the (+)-enantiomer.4a, b, c, d The increasing demand for lamivudine
(3TC), 1 has one of the reasons for the progress in a wide variety of synthetic approaches that
ACS Paragon Plus Environment
2

Page 3 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
enable its production on a large scale with high enantiomeric purity. Several methodologies for
the manufacture of 1 have been described, however, identification as well as implementation of a
scalable synthetic sequence becomes necessary and the same outlined here in this work.
The (-)-enantiomer (3TC) holding unnatural L-β-configuration is the first FDA-approved
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
d
nucleoside analogue. The construction of many such nucleoside analogues requires the selective
glycosylation reaction that is desirable to unite appropriate furanose sugar derivative and
nucleobase by forming of C–N glycosidic bonds while maintaining selectivity. Generally, stereo
control and/or activation aided by a suitable catalyst or chiral auxiliary portion are obligatory
features of any stereoselective glycosylation. Catalytic forms of this reaction are preferable
because they require significantly smaller quantities of costly chirality sources. Highly β-selective
N-glycosylation reactions has been also reported in literature for the synthesis of lamivudine. The
key glycosylation step in lamivudine synthesis where, the enantiomerically pure oxathiolane
acetate coupled with silylated cytosine using several Lewis acids (LAs) such as SnCl , TiCl , TMS
4
4
triflate (TMSOTf), silanes [triethylsilane (Et SiH) or polymethylhydrosiloxane (PMHS)] and
3
5
a,b,c
iodine (I ), pyridinium triflate and trimethylsilyl iodide (TMSI) is described.
However, some
2
of these methods are still associated with shortcomings based on their toxicity, flammable nature,
being expensive and uneasy while handling reagents that limited their practical application in large
scale in pharmaceutical manufacturing. Therefore, there is a need for the development of a safe,
efficient, inexpensive and industrially viable catalyst for glycosylation method. In the present
study, we have established an efficient method for glycosylation through activation of oxathiolane
acetate utilizing 0.5 equiv of ZrCl . The reaction completes within short time for large scale
4
production with high yield at room temperature. The versatileness of this approach is evident from
the fact that even the facile glycosylation takes place without isolation of enantiomerically pure
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
oxathiolane substrate that is superior over previous methods.5a, b, c. The oxathiolane acetate taken
in situ in solution gives stereo selective single nucleoside isomer out of four possible
stereoisomers.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION: The crucial disconnection common to stereoselective synthetic
methods for synthesis of lamivudine 1 is illustrated in Scheme 1 and relies upon a glycosylation
approach to link the two suitably substituted 1, 3-oxathiolane derivatives (2a or 2b or 2c) and
cytosine.
Scheme 1. Retrosynthetic approaches for synthesis of 1
O
O
N
NH2
O
OAc
O
HN
Cytosine
S
2
b
This work
0.5 equiv of ZrCl4
O
O
O
N
NH2
O
O
O
OH
HO
N
O
OAc
O
S
S
S
1
2c
2a
Lamivudine (3TC)
Among the methods available for the industrial manufacturing of lamivudine, the procedure
6
reported by Whitehead and co-workers is the mostly chosen till date. This procedure involves the
chiral auxiliary approach where l-menthyl bearing chirally pure oxathiolane lactol 2a undergoes
in situ chlorination reaction with thionyl chloride in presence of catalytic N, N-dimethylformamide
(DMF) in dichloromethane solvent (-Cl replaces –OH). Then, silylated cytosine replaces –Cl
without any efforts in an S 2-like nucleophilic substitution to produce β-isomer predominantly. It
N
ACS Paragon Plus Environment
4

Page 5 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
is extensively reported that, the β-selectivity could be because of intermediate oxonium ion
formation which gets stabilized by neighboring l-menthyl ester group via anchimeric assistance.6
The method requires extremely efficient crystallization procedure with dynamic kinetic resolution
for achieving synthesis of chirally pure oxathiolane lactol 2a, which in sequence could be used, to
produce lamivudine 1. This approach represents a convenient alternative to previous
methodologies5c, d for lamivudine synthesis. Previous methodologies uses highly unstable and
expensive reagents and requires in a significant quantity and also mandatory requirement of
corresponding chirally pure oxathiolane lactol 2a or its corresponding acetate 2c. Isolation of
oxathiolane acetate (2c) (single stereoisomer) from a mixture of isomers (racemic or
diastereomeric) by chemical resolution or crystallization or any enzymatic techniques goes in hand
with the leaving of 50 per cent (racemic) or extra (diastereomeric mixtures) of the undesired
isomers of substrate. Thus, it is somewhat difficult to synthesize oxathiolane acetate 2c in pure
form with quantitative yields. Access to an enantiopure 2c was explored based on the approach
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
suggested by Whitehead and co-workers. The method involves the reaction of commercially
available l-menthyl glyoxylate monohydrate with 1, 4-dithiane 2, 5-diol in toluene and further
treatment of crude mixture with triethylamine in n-hexane solvent. Literature procedure reported6
early formation of four stereoisomers while, the subsequent addition of catalytic amount of
triethylamine in n-hexane is preferred to induce rapid interconversion between the stereoisomers.
Exact repetition of this methodology on large scale is difficult and not robust. Furthermore,
acetylation of pure oxathiolane lactol 2a with Ac O in pyridine as a base could lead to
2
interconversion to form all the four isomers again. Recently developed strategy by Mandala et al.5b
for preparation of this key intermediate used sodium bicarbonate as a base and acetic anhydride in
acetonitrile afforded mixture of diastereomers (93%), crystallization with solution containing
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
small amount of triethylamine at - 20°C for 36 h required to get desired pure oxathiolane acetate
2
c (46%) (Scheme 2) which is time consuming and critical to maintain such low temperature at
large scale.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As a part of efforts focused on developing large-scale manufacturing route to lamivudine, it was
decided first to synthesize of oxathiolane lactol 2a as per literature procedure (Scheme 2).6
Further, this lactol was converted into an acetate 2b (racemic and/or diastereomeric mixture of
isomers) which was achieved in one step using the reported procedure for a related compound.5d,
7
The approach for the preparation of 2b, shown in Scheme 2. Acetylation of oxathiolane lactol 2a
with acetic anhydride and pyridine as a base afforded a mixture of isomers for oxathiolane acetate
2b.
Scheme 2. Synthesis of oxathiolane acetate 2a, 2b and 2c
Acetic acid ,
OH
Toluene,
O
110-111°C
O
S
OH
OH
O
S
OH
O
O
TEA, n-heptane
S
OH
80%
L-Menthyl Glyoxalate
Monohydrate
1,4-dithiane
-2,5-diol
2a (mixture of isomers)
n-hexane, TEA,
20°C, 36 h
Acetic anhydride,
Pyridine
O
-
O
O
O
O
O
OAc
Dichloromethane
OAc
Literature yield5b
S
80%
S
46%
2c
2
b (mixture of isomers)
Attempts to isolate chirally pure isomer of oxathiolane acetate 2c were unsuccessful as there was
a possibility of formation of four lactol isomers of 2a in solution which were readily acetylated in
situ to form four corresponding racemic and/or diastereomeric mixture of isomers. Further, it was
difficult to isolate and analyze this pure oxathiolane acetate by chiral HPLC. A specific HPLC
ACS Paragon Plus Environment
6

Page 7 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
7
method is also known to determine its isomeric purity. It is known from the literature that, the
reaction is very much base sensitive and to avoid base mediated lactol interconversion, as a
precautionary measure, the reaction sequence was kept as 2a to be added in the mixture of
dichloromethane and acetic anhydride (3.0 equiv) at cooled condition (acidic media) and thereafter
minimum mole of pyridine (0.9 equiv) was added to complete the reaction. Further, the reaction
mixture was simply worked up by quenching with water and extracted in dichloromethane
followed by distillation afforded the residue. The residue was further treated with n-heptane
solvent by heating at 60-70°C for 1 h and cooled to room temperature. The solid precipitate filtered
at 5-10°C afforded 80% yield and was of HPLC purity showing 92% purity with 8% amounting
to the formation of undesired diastereomer for 2b. Later, the oxathiolane acetate 2b was taken in
solution form as such without isolation for further glycosylation step at large scale.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Further, we started our investigation on the glycosylation step which is quite crucial to the
overall synthetic approach using starting material oxathiolane acetate 2b as such without any
further crystallization or classical resolution. Also, preliminary requirement for glycosylation was
a silylation of cytosine completely, which usually described in acetonitrile or chlorinated solvents.
In this case, a mixture of bis (trimethylsilyl) acetamide (BSA) as a robust silylating agent and
acetonitrile solvent in appeared crucial due to lesser reactivity of cytosine. A screening of solvent
was also simultaneously performed, and it was found that BSA was suitable for silylation of
cytosine in presence of acetonitrile. Acetonitrile further allowed for better reactivity and solubility
that permits the silylation at room temperature in a few minutes. However, use of expensive BSA
and acetonitrile would increase the overall process cost at large scale. Silylation with hexamethyl
8
disilazane (HMDS) is also most commonly used method with acid catalysis. The addition of
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
catalytic amount of trimethylsilyl chloride (TMSCl) has been known for silylation of cytosine,
hence, this reagent (HMDS with catalytic TMSCl) was used in our method as shown in Scheme 4.
The primary step of each glycosylation reaction is to activate the leaving group, which usually
occurs through its interaction with an activator/promoter or, a catalyst more rarely. Reactions of
anomeric mixtures of oxathiolane acetate 2b with silylated cytosine and practically any ordinary
LA caused to the formation of inseparable mixture of N-glycosylated anomers. Since only β-
isomer demonstrates the advantageous biological activity, any efficient methodologies required
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
a
the use of a stereoselective nitrogen glycosylation. Tse et al. reported a practical and efficient
preparation of a therapeutically essential nucleoside (3TC), based on a novel diastereoselective N-
glycosylation procedure. It is noteworthy to mention that, the high stereo selectivity of the N-
glycosylation reaction involving chirally pure oxathiolane acetate 2c was noticed only when TMSI
was applied as the promotor. They used highly expensive and non-stable (even though more
effective) TMSI reagent for N-glycosylation reaction starting with oxathiolane acetate 2c having
(2R, 5R)-configuration (Scheme 1) that gave β-isomer in the ratio β: α = 23:1. This β-selectivity
could be because of stabilization of oxonium ion by neighboring l-menthyl ester group by means
5
a
of anchimeric assistance. More recently, Caso et al. reported the use of silanes (PMHS or Et SiH)
3
in combination with iodine (I ) as novel N-glycosylation reagents for the synthesis of 1, 3-
2
oxathiolane nucleosides lamivudine. These systems (acts as sources of anhydrous HI) were
invented to behave as substrate activators and promoters for N-glycosylation reactions. Also, the
4
type of the cytosine protective group at the N position further linked to the outcome of the stereo
chemical reaction. But in this case nucleobase protection was an additional requirement for the
5
b
improvement in the α/β ratio of a nucleoside product. Mandala et al. also described the use of
pyridinium triflate as a novel N-glycosylation reagent first time at high temperature in acetonitrile
ACS Paragon Plus Environment
8

Page 9 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
solvent. However, in all the cases (TMSI, Silane/I and pyridinium triflate-Mediated N-
2
glycosylation reaction) the particular separation of the isomers of 2b was mentioned as compulsory
for ensuring the target nucleosides with high optical purity.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5c
In 1991, W. B Choi et al. reported a highly stereoselective base-linking reaction that operated
through in situ formation of a complex between an appropriate cyclic precursor and a suitable LA
(Scheme 3). The use of stannic (IV) chloride (2 equiv) and dichloromethane as a solvent at ambient
temperature condition resulted in the exclusive formation of the β-nucleoside product for
lamivudine. This recommends a potentially general approach to the regulation of the
stereochemistry in N-glycosylation. The N-glycosylation reaction stereoselectivity could be
rationalized based on a preferential interaction between heteroatom and LA.
Scheme 3. Concept of in situ complexation of LAs for controlling nucleoside stereochemistry5c
RO
RO
Pathway
A
RO
O
Silylated Base
Lewis acid
OAc
O
O
X
X
X
Base
Pathway
B
L
L
L
L
M
RO
RO
RO
RO
Base
O
ß-attack
O
O
O
OAc
OAc
X
X
X
X
Cl
L
M
L
M
L
L
L
L
L
syn
anti
Where R = tert-butyldiphenylsilyl, X = S or O, M = Sn, L= Cl
The use of TMSOTf and TMSI, the exclusive purpose of which was to produce an oxonium ion,
should adhere to pathway A and results in no stereo controls. However, in examples where LA
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
could form precomplex to a heteroatom of a ring (i.e., pathway B), selectivity in diastereofacial
manner could be accomplished by minimizing destabilization of 1, 2-steric interactions with the
aid of complexation anti to the protected hydroxy methyl substituent. At any event, this
complexation ought to considerably restrict the methodology of the silylated base moiety to the α-
face. The possibility of the associated metal that delivers one chloride ligand to the α-face of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
carbonium ion could then undergo S 2 attack and resulted the formation of the β-N-nucleoside.
N
In the context of our interest in new procedures of N-glycosylation, the investigation for more
competent, stereoselective, and cost-effective alternatives to the above mentioned strategies for
large scale synthesis of lamivudine caused a great concern to scrutinize the synthetic ability of the
N-glycosylation reagents on the basis of above in situ complexation concept of any non-explored
LA. Hence, our investigation have been directed towards the search of an appropriate catalyst
including hitherto non-explored LA for the said glycosylation which not only gives high chemical
yields but also high stereo specificity. Ability of LA to complex with a lone pair of electrons is
different for different acids. Hence, one could expect different chemical prospective. Although, in
the literature number of LAs like SnCl , TiCl , Cu-Triflate, TiCl (OiPr), TMSI has been used,
4
4
3
they are not effective and none of them gave desirable results either in specificity or chemical
yield. Each LA unlike Brønsted acid gives entirely different course of reaction9a, b. Contrary to
Brønsted acid, where pKa is the essential factor for its efficacy, amongst the LA, the ability of
complexing with a host molecule and extent of positive charge generation in the acceptor atom
(e.g. complexing with carbonyl oxygen) determines their reaction feature however, it is not
predictable. Hence, identifying an appropriate and ideal LA for a specific set of reaction itself
becomes a complex phenomenon involving skill and requires extensive experimentation. This
would be evident from the facts observed during our experiments using a number of LAs such as
ACS Paragon Plus Environment
1
0

Page 11 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
SnCl , TiCl , Cu-Triflate, TMSI, Sc(SO CF ) , BF .OEt for the said glycosylation reaction with
4
4
3
3 3
3
2
more than 1 mole, and surprisingly none of them gave any useful conversion. Moreover, these LAs
suffer from one or more shortcomings such as toxicity, cost and managing complexity or hazards,
non-feasibility towards different substituted or functionalized acylation reactants and the process
complications, primarily reaction mixture extraction difficulty. Surprisingly, use of tetravalent
zirconium salt such as zirconium (IV) chloride satisfies most of the requirements. Zirconium (IV)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
chloride (ZrCl ) explored as a catalyst at a temperature, preferably between 25-30°C, to obtain a
4
nucleoside product with cis-(-)-configuration stereoselectively almost in quantitative yield
(Scheme 4). Thus, in the course of our investigation in the development of stereoselective
glycosylation reaction for the efficient synthesis of lamivudine, it was discovered that the
zirconium (IV) chloride, among various transition metal chlorides, is the reagent of choice for the
glycosylation reaction.
Scheme 4. ZrCl -Mediated N-Glycosylation reaction
4
O
O
N
NH2
O
OH
O
HN
cytosine
S
2a
(
mixture of isomers)
HMDS,
TMSCl,
Acetic anhydride,
Pyridine,
120-130°C
Dichloromethane
NHTMS
Cl
Cl
Cl
N
Zr
Cl , TMSO
O
O
O
N
N
NH2
O
O
OAc
N
O
O
Dichloromethane, rt
S
S
2b
3
(mixture of isomers)
ACS Paragon Plus Environment
1
1

Organic Process Research & Development
Page 12 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
It is evident from the Table 1 (provided in entry 6), that shows that, ZrCl gives the correct
4
specificity and optimum chemical yield.
Table 1. Experimental conditions for optimization of different activators for N-glycosylation
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
entry activators
equiv solvent (5
volumes)
temp
(°C)
time
(h)
Yield
(%)
β / α ratio
1
2
Cu-triflate
TMSI
1.0
1.5
dichloromethane rt
24
18
*
*
dichloromethane 35-40
51
94 / 6
3
4
5
6
TiCl4
TiCl4
SnCl4
ZrCl4
1.1
1.1
1.1
0.5
acetonitrile
rt
12
12
12
12
*
*
dichloromethane 35-40
dichloromethane rt
dichloromethane rt
30
*
95.69/4.31
*
60%
98.15/1.85
7
8
9
1
ZrCl4
0.5
0.5
1.0
1.0
acetonitrile
rt
12
12
12
12
*
*
ZrCl4
dichloromethane rt
dichloromethane rt
dichloromethane rt
**
*
**
*
BF .OEt
3
2
0
Sc-triflate
*
*
All these reactions were performed with 2b, with silylated cytosine using different activators.
β / α ratio analyzed by HPLC method.
a
*product conversion is less than 5% hence could not isolated them and could not be analyzed.
**No reaction where, premixing of donor (oxathiolane acetate, 2b) and acceptor (silylated
cytosine) followed by ZrCl addition was performed.
4
Zirconium (IV) chloride, otherwise referred as zirconium tetrachloride (ZrCl ) is a high melting
4
solid and used as a (weak) LA in organic synthesis.9a, b In contrast with molecular TiCl , powdered
4
ZrCl comprises a polymeric structure in which every Zr element is octahedrally coordinated. This
4
discrepancy in structure is liable for the striking differences in their properties. TiCl is distillable,
4
but zirconium (IV) chloride is a solid powder with a high liquifying point. Most zirconium
compounds are relatively less toxic, easily operated, cost effective and can withstand small amount
ACS Paragon Plus Environment
1
2

Page 13 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
of moisture. Zirconium (IV) compounds have high coordinating ability that offers them with an
adequate LA behavior. Zr4+ has a high value of charge to size ratio (Z2/r = 22.22 e2/m10), relative
to the majority of main group elements and lighter and heavier transition metal ions like Li+ (Z2/r
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
=
1.35 e2/m10), Bi3+ (Z2/r = 8.82 e2/m10), In3+ (Z2/r = 11.39 e2/m10), Sc3+ (Z2/r = 12.33
e2/m10), Fe3+ (Z2/r = 13.85 e2/m10), V3+ (Z2/r = 14.06 e2/m10) and Al3+ (Z2/r = 16.98 e2/m10)
1
0
and is comparatively softer hard acid . This fundamental chemical characteristic of zirconium
(IV) chloride has made it an encouraging catalyst with strong acid nature, easy to handle, and lower
scale of toxicity. In addition, its relatively high-level abundance and therefore lower cost offer
advantageous use in glycosylation.
ZrCl -catalyzed efficient Ferrier glycosylation procedure stated the mild reaction conditions and
4
shorter reaction time in case of pseudoglycals synthesis and higher stereoselectivity and yields for
synthesizing the 2,3-unsaturated glycosylated compounds.11a,b The complete stereocontrol
approach at anomeric center of 1-thio-D-glucopyranose and higher yields of corresponding
condensed compounds were also noticed for the preparation of O-glycosyl disaccharides using
1
2
ZrCl as a LA. Zirconium (IV) chloride is an ideal LA since it is stable, an efficient, low-cost,
4
and an appropriate catalyst. Most of the glycosylation reactions are achieved by pre-mixing donor
1
3
and acceptor at the same time and further by adding the promoter for donor activation. This
strategy did not work in case when, ZrCl was used as promotor. Thus, in the present work, we
4
have used alternate strategy which preactivates glycosyl donor initially in the absence of acceptor
i.e. preactivation of oxathiolane acetate 2b with ZrCl and then treated with silylated cytosine for
4
glycosylation reaction.
Investigations were undertaken with different moles of ZrCl per mole of substrate, e.g. 0.1
4
equiv, 0.3 equiv, 0.5 equiv and 1.0 equiv for glycosylation reaction, but best yield was obtained
ACS Paragon Plus Environment
1
3

Organic Process Research & Development
Page 14 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with 0.5 equiv which is less than the mole equivalent of other catalyst used so far for the
glycosylation reaction. The mole wise experimental studies of ZrCl are comparatively shown in
4
Table 2. Nucleobase variations also gave positive results in case of 5-Fluorocytosine (Table 2.
entry -5).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Optimization of mole ratio zirconium (IV) chloride as an activator for N-glycosylation
a
b
entry
ZrCl4
equiv)
solvent (5
volumes)
temperature time (h)
(°C)
yield (%) β / α
(
1
2
3
4
0.1
0.3
0.5
1.0
0.5
dichloromethane
dichloromethane
dichloromethane
dichloromethane
dichloromethane
rt
rt
rt
rt
rt
12
12
12
12
12
29
47
60
58
65
98.53/1.47
97.50/2.50
98.15/1.85
98.45/1.55
98.26/1.76
5c
a
Isolated yield, b β / α ratio analyzed by HPLC method, silylated 5-fluorocytosine as a
c
nucleobase is used instead of silylated cytosine.
The notable characteristics of this catalytic glycosylation procedure are associated with lesser
reaction times, mild conditions, quantitative yields, better conversions, clean reaction profiles,
tolerability of different functional groups, easy isolation of product and operational simplicity.
5
c
On the basis of earlier studies in the literature , a reasonable mechanism for the stereoselective
N-glycosylation using ZrCl is proposed, as shown in Scheme 5. ZrCl could form a precomplex
4
4
to oxathiolane ring heteroatom sulphur (intermediate I). The stereofacial selectivity could be
accomplished by minimizing the destabilization of 1, 2-steric interactions with the aid of
complexing ZrCl anti to the l-menthyl ester functionality. Additionally, intermediate II could be
4
generated where the Zr metal gives one of its ligand (chloride) to the α-face while removing acyl
group as shown in intermediate I. The subsequent α-chloro derivative II could then follows S 2-
N
like reaction path with silylated cytosine, constructing the glycosidic C-N bond and results the
ACS Paragon Plus Environment
1
4

Page 15 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
formation of β-selective nucleoside intermediate III. Finally, deprotection of trimethylsilyl (TMS)
group with HCl and removing ZrCl catalyst from intermediate III produces 3. This precomplex
4
formation should significantly inhibit the approach of the silylated cytosine to the α-face, at the
very least.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 5. Plausible reaction mechanism for stereoselective N-glycosylation using ZrCl4
NH2
N
O
O
O
N
O
O
O
Cl
Zr
Cl
OAc
O
TMSCl
Cl
Cl
S
S
3
2b
HCl
Cl
Zr
Cl
Zr
anti
NHTMS
N
anti
Cl
Cl
Cl
Cl
Cl
O
Cl
O
O
N
O
O
OAc
O
O
S+
III
S+
I
silylated
cytosine
Cl Cl
Zr
anti
TMSOAc
NHTMS
N
OAc
Cl
O
O
Cl
N
OTMS
O
S+
II
ß-attack
Surprisingly, a chiral auxiliary l-menthyl group is needed to be used in executing the process
however, the use of optically pure intermediate 2c and high temperature condition in the
ACS Paragon Plus Environment
1
5

Organic Process Research & Development
Page 16 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
glycosylation step is not a necessary limitation of the present work. The present process is
demonstrated for preparing lamivudine at an industrial scale.
Reaction calorimetry evaluation for N-glycosylation step. When considering the migration
of a reaction procedure from lab scale to large scale, it is necessary to identify how much heat gets
released/dissipated. At a low scale level, the heat produced may not trigger a major concern,
however when scaling up, that heat could slowly get build up and be extremely unsafe in turn
could result runaway reactions. Since the use of zirconium (IV) chloride as a LA to be performed
at large scale, thorough assessment of that heat exchange, stability of the N-glycosylation reaction,
its quenching and work up procedure also must be carried out. Safety investigations to determine
the onset of degradation points were performed using differential scanning calorimetry (DSC) and
reaction calorimetry. The thermal behavior during the preparation of compound 3 from 2b solution
was studied in a reaction calorimeter.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As per protocol of a 50 gm batch input of oxathiolane acetate (2b), dichloromethane (250 ml)
and 20.2 g of ZrCl was charged in single portion while temperature of the reaction mass was
4
maintained at 25°C and stirred for 15 mins. Then, the reaction mass temperature was raised to
3
0°C. To this 234 g of silylated cytosine reaction mass was charged at 30°C. Sample was removed
to assess the decomposition of reaction mixture by DSC. The reaction mass was maintained at
0°C for ½ h. Further, the temperature was raised to 35°C and maintained for 4 h. Then, 200 ml
3
of methanol was added to the reaction mass at 35°C and continued maintaining for ½ h and the
decomposition of reaction mixture was checked by DSC by removing samples at different
operating parameters.
The above-mentioned protocol was subjected to the reaction calorimeter evaluation utilizing
Mettler make SV 01 RC1 reaction calorimeter. The heat changing trends indicated by reaction
ACS Paragon Plus Environment
1
6

Page 17 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
calorimetry (RC1) during charging of ZrCl followed by silylated cytosine reaction mass and
4
quenching with methanol are shown in Figure 2. The evolution of heat of reaction was calculated
using Q flow (exchange from jacket), Q accumulation (of reaction mass), Q dosing (heat inputs by
dosed reactants) and Q loss (to surrounding from reactor). Initialization parameters and plots of
various parameters, specific heat determination and heat of reaction are given in Table 3.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Reaction calorimetric measurement of the N-glycosylation reaction. Tr: reaction
temperature (green peak), mr: mass of reaction (blue peak), Qr: heat of reaction (red peak)
Table 3. Reaction calorimetric data for the N-nlycosylation reaction (addition of ZrCl and
4
cytosine reaction mass and quenching the reaction with methanol)
unit
time
reacti process
reaction heat of
specific heat ΔTad MTS
operation (min) on
tempera- enthalpy reaction per capacity Cpr
R
(°C)
reagents
addition
mass ture (°C) ΔH (kJ) kg reaction
(kg)
at start kJ/
mass (kJ/kg) (kg. K)
(°C)
ZrCl4
26
0.319 25
6.415
20.1 1.82002
11.05 36.05
ACS Paragon Plus Environment
1
7

Organic Process Research & Development
Page 18 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cytosine
mass
235
43
0.339 30
0.573 35
8.045
23.6
50.4
1.82009
1.82082
13.03 33.03
27.71 62.71
methanol
28.922
Heat of reaction: As shown in Table 3. the heat evolved for zirconium (IV) chloride addition
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
was 6.415 kJ (exotherm) for 0.319 kg of initial reaction mass i.e. 20.1 kJ/kg of initial reaction
mass. This could lead to adiabatic temperature rise (ΔT ) of 11.05°C. The overall heat evolved
ad
for cytosine reaction mixture addition was 8.045 kJ (exotherm) for 0.339 kg of initial reaction
mass i.e. 23.6 kJ/kg of initial reaction mass. This could lead to ΔT of 13.03°C. The heat evolved
ad
for methanol addition was 28.922 kJ (exotherm) for 0.573 kg of initial reaction mass i.e. 50.4 kJ/kg
of initial reaction mass. This could lead to ΔT of 27.71°C.
ad
DSC findings for the reaction mixtures tested at various reaction stages are shown in Table 4.
DSC of the reaction mixture samples taken after ZrCl addition, cytosine reaction mass addition
4
and quenching reaction mass with methanol (before and after) indicated that their exothermic
decomposition points were not much differ from the exothermic decomposition points of starting
materials and product formed. It was in the range 215 to 274°C.
Table 4. DSC results at different operating parameters for the reaction mixtures.
Sample
endotherm onset exotherm onseta heat of evolution
°C)
(
-
-
(°C)
J/g
2
2
b in dichloromethane
231.9
215.5
257.8
274.4
262.0
255.0
32.0
33.1
107.9
33.5
130.6
89.6
b in dichloromethane + ZrCl4
cytosine reaction mass
234.8
cytosine solution addition
before quenching reaction mass
-
216
-
after quenching reaction mass
a
In the Supporting Information, plots illustrating the DSC results are provided
ACS Paragon Plus Environment
1
8

Page 19 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
As depicted in Table 3, the maximum temperature of synthetic reaction (MTSR) during addition
of ZrCl , cytosine reaction mass and methanol charging, calculated as summation of the process
4
temperature and adiabatic temperature rise (ΔT ), were 36.05°C, 33.03°C and 62.7°C
ad
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
respectively. The onset of decomposition temperature recorded above 215°C as calculated by DSC
data of reaction mixtures at various stages, shows that the reaction is safe even though cooling
system fails, since the MTSR values are much lower than the onset decomposition temperature.
Secondary reactions, such as decomposition with exotherm, can only occur when the MTSR value
is greater than their onset temperature. Based on heat evolution data maximum heat flow calculated
as depicted in below table 5.
Table 5. Maximum heat flow results
unit operation
reaction mass
0.323 kg
maximun heat flow (Qmax
)
ZrCl (half addition)
21W/0.323 kg
4
Cytosine mass addition (initial phase)
0.379 kg
42W/0.379 kg
rd
Methanol addition (1/3 portion)
0.632 kg
74W/0.632 kg
Based on the heat evolution data, reaction during methanol addition was instantaneous. Maximum
heat flow (Qmax) observed during methanol addition was 74 W/0.632 kg of reaction mass. This was
rd
observed when almost 1/3 of methanol was added.
The equipment setup available for a common-purpose reactor could be used in the manufacturing
process as per the RC1 trend data. The reaction was scaled up to 174 kg of input in a 3000 L glass
line reactor with an adequate jacketed cooling system and outlet connected to the standard
scrubber. The addition of zirconium (IV) chloride started at 25.5°C and temperature raised up to
3
2.0° (time taken for addition was 35 min) recorded by means of a digital temperature indicator,
verifying the reflection as in the RC1 experiment and addition of silylated cytosine reaction mass
ACS Paragon Plus Environment
1
9

Organic Process Research & Development
Page 20 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
showed temperature rise from 25.0° to 31.0°C (time taken for addition was 60 min) and addition
of methanol to the reaction mass after completion of reaction recorded the temperature rise as
27.2°C to 40.1°C.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 6. Synthesis of lamivudine
1
) NaBH , K HPO
4
Ethanol, Water
O
4
2
O
N
O
N
NH2
NH2
OH
HO
OH
O
N
O
N
O
S
2
) (S)-(-)-BINOL ,
Methanol
S
4
3
O
N
Ethyl acetate : Water
rt
NH2
O
N
HO
S
1
The nucleoside analogue of cis configuration 3 obtained from the glycosylation reaction of the
compound 2b with cytosine thereof was reduced to give a specific stereoisomer lamivudine 1.
Hydride reducing agents like lithium aluminium hydride, lithium borohydride or sodium
borohydride are found suitable for such reduction reactions. According to previous reports, lithium
4
a
aluminium hydride provides the subsequent nucleoside 1 after stirring 1 h at 0°C, however, gave
low yield owing to the degradation during the reaction. It is also reported that stereo integrity is
1
3
retained utilizing sodium borohydride as the reducing agent in the presence of a buffer like
dipotassium hydrogen phosphate. Therefore, this method was used for reduction of l-menthyl ester
group to provide the target lamivudine 1 (Scheme 6). After reduction using sodium borohydride
and neutralization with aqueous HCl, a practical problem cropped up as lamivudine exhibits high
solubility in aqueous media, making it difficult to isolate effectively. The side products salts such
as NaCl, KCl and boric acid were in dissolved form along with lamivudine in the aqueous layer.
In polar solvents, lamivudine is well isolated from solution by forming a salt with poor solubility.
ACS Paragon Plus Environment
2
0

Page 21 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
If required, the water-insoluble salt could then be converted to the free base, or to a various salt
1
4
thereof by usual methods. As per report salicylate salt is particularly suitable for this purpose,
but salicylate salts do not allow to enrich the purity. In the present work, we used our earlier
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
15
reported (S)-BINOL Co-crystal strategy as lamivudine obtained according to the process was
having purity about 98%, i.e. there is 98% ee and 2% of its cis- (+)-isomer. Therefore, enantiomeric
purity was further enriched using (S)-BINOL which gave 99.8 % ee. Lamivudine-(S)-BINOL
complex compound, 4 obtained was found water insoluble at lower temperature therefore the
inorganic salts were removed by simple water slurry to the 4. It was further easily got separated
from (S)-BINOL with extraction procedure in water and ethyl acetate at room temperature. (S)-
BINOL reagent was later completely recovered and recycled from ethyl acetate. Lamivudine thus
obtained in aqueous layer was isolated by distillation and recrystallized from denatured spirit and
acetone mixture in highly pure form.
This stereoselective method for the synthesis of 1, 3-oxathiolane nucleoside analogue using
16
zirconium (IV) chloride (ZrCl ) as an activator in N-glycosylation step was found to be very
4
effective.
CONCLUSION:
The present work aimed in developing a large scale, operationally convenient and scalable
synthetic process for preparation of lamivudine 1 reliably reproducible on, at least 33.3 kg scale.
The key N-glycosylation step resulted in a stereoselective β-anomer of lamivudine using ZrCl4
starting even with chirally impure substrate where, previous methods fails to attain this. A
plausible mechanism was also proposed and discussed for this stereoselective N-glycosylation
reaction. The safety of process for the N-glycosylation step was evaluated in detail using DSC and
RC1 studies. Furthermore, a problematic isolation of water soluble lamivudine free from inorganic
ACS Paragon Plus Environment
2
1

Organic Process Research & Development
Page 22 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
salts was achieved. The synthesis, carried out at kilogram scale, provided a material that satisfied
all regulatory requirements.
EXPERIMENTAL SECTION:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
General Remarks: All reagents and solvents purchased commercially and used as received
without further purification. Melting points were recorded using an automated melting point
apparatus. Specific rotations were measured using a JASCO P-1030. Metrohm instrument is used
to determine moisture content by the Karl Fischer titration method. RC1 and DSC studies were
carried out at Intertek India Pvt. Ltd, Mumbai: Mettler make SV 01 800 ml capacity all glass
reactor equipped with a flow agitator type propeller, jacketed with controlled temperature
circulator, condenser, addition funnel and probes.
HPLC analysis was performed on SHIMADZU LC-2010 or UltiMate 3000. HPLC system was
1
13
equipped with a chiral column (Chiracel OJH, 4.6mm × 250 mm, 5μm). H and C NMR spectra
were measured in CD OD or CDCl on a Bruker spectrometer (AVANCE-III HD) at frequencies
3
3
of 500 and 125 MHz, respectively. Proton chemical shifts (δ) are measured in ppm with respect to
tetramethylsilane (TMS) (δ) 0 ppm or to residual protons in the solvent as internal standard. Mass
spectra were recorded with the ACQUITY UPLC PDA detector utilizing electron spray ionization
(ESI).
(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl 5-hydroxy-1, 3-oxathiolane-2-carboxylate
(2a) : A mixture of toluene 500 mL, l-menthyl glyoxylate monohydrate ( 100 g, 0.434 mol) and
acetic acid (10.5 g, 0.174 mol) was heated to reflux at about 110-111°C. The reaction heating
maintained for 3 h to remove water azeotropically. After observing the complete removal of water,
cooled the reaction mixture up to 90°C. Distilled off the solvent under vacuum not more than 90°C
till about 300 mL solvent remained in the flask. Further cooled the reaction mass to 40-50°C and
charged 1, 4-dithiane 2, 5-diol (33.8 g, 0.222 mol). The reaction mixture was heated further for 2
ACS Paragon Plus Environment
2
2

Page 23 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
h at 85-90°C. The reaction progress monitored by HPLC method. After completion of the reaction,
mixture was filtered through celite bed to remove any traces of undissolved particles. The filtrate
was then cooled to 0 to 5°C. To this filtrate, a solution of trimethylamine (6 mL) and n-heptane
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(600 mL) was added slowly at 0 to 5°C. The resulting slurry was stirred for 6 h at 0 to 5°C. The
solid suspension then filtered through Buchner and sucked dried under vacuum. The wet cake was
rinsed with precooled n-heptane (200 mL). The solid material was dried in a vaccum oven for 16
1
h at 45°C under vacuum to provide 100 g of a white solid 2a. H NMR (500 MHz, CDCl ): δ 6.02
3
(d, J=4.5 Hz, 1H), 5.92 (bs, 1H), 5.56-5.58 (d, J=5.5 Hz, 2H,), 4.70-4.77 (td, J=5.5 Hz, 3H) , 3.29-
3
.33 ( m, 3H ), 3.14-3.17 (dd, J = 14, J =3, 1H), 3.05-3.15 (d, J=13.5, 1H), 1.94-2.05 (m, 4H),
1
2
1.68-1.72 (bd, J=2.5Hz, 4H), 1.41-1.52 (m, 4H), 1.02-1.08 (m, 4H), 0.91-0.93 (m, 14H), 0.75-0.80
13
(
m, 6H) ppm. C NMR (125 MHz,CDCl ): δ 172.4, 172.2, 169.2, 103.2, 103.1, 101.2, 101.1, 80.2
3
,
80.1, 78.0, 77.8, 76.9, 76.7, 75.9, 75.8, 47.0, 46.8, 46.7, 40.6, 40.3, 40.1, 39.9, 38.4, 34.1, 34.0,
31.4, 31.3, 26.2, 26.1, 26.0, 23.3, 23.2, 21.9, 20.7, 16.2, 16.1 ppm. ESI-MS m/z: calcd for:
C H O S: 288.1, found [M + NH ]: 306.0; Loss on drying at 60°C under vacuum was found to
1
4
24
4
4
be 0.3% w/w. Specific optical rotation -57° (10 mg/mL in methanol at 20°C).
1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl 5-acetoxy-1, 3-oxathiolane-2-carboxylate
2b): To a cooled solution of the dichloromethane 250 mL and acetic anhydride (50 mL, 0.52 mol),
a (50 g, 0.17 mol) was added at 5-10 °C. The solution was further cooled to 0°C and pyridine
12.5 mL, 0.15 mol) was added dropwise within 60 min and stirred for 1 h. The progress of reaction
(
(
2
(
monitored by thin layer chromatography. The reaction mixture was quenched into water (250 mL
and phases were separated. The organic phase washed with water (250 mL). The desired product
is in the organic layer. The aqueous layer was extracted again one more time. The combined
organic phase was distilled under vacuum to get oily residue. To the oily residue n-heptane (500
ACS Paragon Plus Environment
2
3

Organic Process Research & Development
Page 24 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mL) was added. The mixture was heated at temperature 65-70°C for 1 h. The reaction mixture
further cooled to room temperature and then at 0-5°C. The solid suspension was held for 30 min
at same temperature. The solid suspension then filtered through Buchner and sucked dried under
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
vacuum. The wet cake was rinsed with cold n-heptane (75 ml). The solid material was dried for
1
1
6 h in a vacuum oven at 40 °C under vacuum to give 45 g of a white solid 2b. H NMR (500
MHz, CDCl ): δ 6.57-6.72 (d, J=5 Hz, 1H), 5.55-5.56 (s, 1H), 4.63-4.70 (td, J = 13.5, J =5.5Hz,
3
1
2
1
H), 3.36-3.40 (dd, J =14.5Hz, J =5.1Hz, 1H), 3.09-3.21(d, J=15, 1H), 2.04 (s, 3H),1.86-1.95 (m,
1 2
2
H), 1.61-1.64 (bd, J=14 Hz, 2H),1.43 (bd, J=4.5Hz, 1H), 1.32-1.38 (bt, J=14.5Hz,1H), 0.91-1.03
1
3
(
m, 2H), 0.79-0.88 (m, 7H), 0.69-70 (d, J=8.5Hz, 3H) ppm. C NMR (125 MHz, CDCl ): δ 169.9,
3
169.4,168.58, 168.41, 99.55, 98.97, 80.09, 79.7, 79.5, 75.8,75.73,75.6, 46.8, 46.7, 46.6, 40.4, 39.9,
3
:
7.4, 37.0, 36.8, 33.9, 31.1, 25.9, 25.8, 25.6, 23.0, 22.9, 21.8, 20.9, 20.6, 20.5, 16.0 ppm. ESI-MS
-
+
+
+
m/z calcd for: C H O S :330.2 Found: ES [M + NH ]: 348.2, and ES [Dimer M +Na ]:683.2,
1
6
26
5
4
Specific optical rotation was –62° (c, 0.51, CHCl ) and mp was found to be 105.2°C
3
(1R, 2S, 5R)-2-isopropyl-3-methylcyclohexyl (2R, 5S)-5-(4-amino-2-oxopyrimidin-1(2H)-
yl)-1, 3-oxathiolane-2-carboxylate (3): Part a) :Dichloromethane (870L) having water content
not more than 0.07% w/w was charged into the glass line reactor under N atmosphere. Acetic
2
anhydride (174 L, 1823.35 mol) was added to the reactor at room temperature. The reaction
mixture was further cooled to 0 to 10°C and 2a (174 kg, 603.2 mol) was charged and stirred for
1
0 mins. The reaction mixture shows temperature decrease after addition of 2a (i.e. endothermic
reaction). Pyridine (43.5 L, 561.15 mol) was slowly added to the above reaction mixture at 0 to
0°C. The temperature of reaction was allowed to raise up to 10 to 15°C and maintained for 2 h.
1
The reaction progress was monitored by HPLC method to control the content of 2a below 2.0 %
limit. After completion of reaction, the reaction mixture was quenched with water (870 L) and
ACS Paragon Plus Environment
2
4

Page 25 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
stirred at 10 to 25°C for 3-4 h. Further, settled the reaction mass and phases were separated. The
required product is in the organic phase. The organic phase was then charged into the reactor and
added 10% aq. acetic acid solution (870 L). The reaction mixture was stirred, settled and phases
were separated. The organic phase was again charged into the reactor and added 10% sodium
chloride solution (870 L). The reaction mixture was stirred and settled for 30 min and phases were
again separated. The water content of organic phase was analyzed and should be not more than
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
.2% w/w, if it was more then solvent had to be distilled out azeotropically and controlled the
water content well within the limit not more than 0.2% w/w. This organic layer containing product
oxathiolane acetate 2b.
Part b): Silylation of Cytosine procedure: Meanwhile, hexamethyldisilazane (402 L), cytosine
(100.67 kg, 906.12 mol) and trimethylchlorosilane (50 L) were charged into the separate stainless
steel reactor under nitrogen atmosphere. This reaction mixture was heated and maintained to 120-
130°C till a clear solution obtained. Distilled off the excess solvent under vacuum below 120°C to
get white residue. Vacuum was released carefully with nitrogen gas to avoid any moisture contact
to this reaction mass. Dichloromethane (348 L) was added to the reaction mass and stirred for 30
min. This silylated cytosine mass was kept under nitrogen atmosphere for further use.
Zirconium (IV) chloride (70.3 kg, 301.7 mol) was added into the organic phase obtained in Part
a), and stirred under N atmosphere for 15-30 mins at 20-30°C. After the stirring was completed,
2
the silylated cytosine mass obtained in part b) was charged within 30 to 40 mins at 20 to 40°C by
applying continuous cooling to the reactor to avoid exothermic reaction. After completion of
addition, the reaction mixture was stirred and maintained for 12 h at room temperature. The
progress of reaction was monitored by HPLC method to control the content of 2b below 4.0 %
limit. The reaction mass was then treated with methanol (690 L) by applying cooling and
ACS Paragon Plus Environment
2
5

Organic Process Research & Development
Page 26 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
temperature control (between 20° to 40°C) to avoid exothermic reaction. The reaction solvent was
then distilled out under vacuum below temperature 60°C till stirrable volume remains in the
reactor. Methanol (174 L) was again charged to the reaction mass and co-distilled to remove traces
dichloromethane. Further, after cooling the reaction mass, cyclohexane (348 L) and 10% aq. HCl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(1218 L) was added. The resulting solid was stirred for 30 to 60 min at 20 to 30°C. The solid
suspension was then centrifuged and kept under spinning till complete expulsion of traces mother
liquor was achieved. The wet cake was then spray washed with cyclohexane (174 L) in centrifuge
under spinning till expulsion of mother liquor took place. The wet cake was again spray washed
with methanol and water (1:1) mixture (870 L) in centrifuge under spinning till expulsion of
mother liquor. The solid wet cake was then removed from centrifuge and charged into the reactor
containing water (1740 L). The reaction mixture was further treated with trimethylamine (52.2L)
and pH was recorded not less than 8. The slurry mass stirred for 30 min and then centrifuged.
Centrifuge was kept under spinning till complete expulsion of traces mother liquor. The wet cake
again spray washed with water (174 L) in centrifuge under spinning till completed expulsion of
mother liquor. The solid wet cake was then removed from centrifuge and charged into the reactor
containing water (1740 L). The slurry mass stirred for 30 min and then centrifuged. Centrifuge
was kept under spinning till the removal of traces mother liquor. The wet cake again spray washed
with water (174 L) in centrifuge under spinning till completed expulsion of mother liquor. The wet
cake was finally removed from the centrifuge and analyzed for cytosine content and related
substances by HPLC method. The solid wet cake was dried for 16 to 24 h in a vacuum oven at
1
6
0°C under vacuum to provide 100.5 Kg of 3. H NMR (500 MHz, CD OD): δ 8.32-8.34 (d, J=7.5
3
Hz, 1H), 6.42-6.44 (dd, J =6Hz, J =5 Hz, 1H), 5.93-5.94 (d, J=7.5Hz, 1H), 5.63 (s, 1H,),4.78-4.84
1
2
(td, J =10.8 Hz, J =4.1Hz, 1H), 3.56-3.59 (dd, J =12.2Hz, J =4.7Hz, 1H,), 3.18-3.21 (dd, J =12
1
2
1
2
1
ACS Paragon Plus Environment
2
6

Page 27 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Hz, J =6Hz, 1H), 1.97-2.07 (m, 2H,), 1.73-1.77 (bd, J=12 Hz, 2H), 1.45-1.56 (m, 2H), 1.06-1.17
2
1
3
(
m, 2H) ppm; 0.92-0.97(q, J=7 Hz, 7H) , 0.80-0.81(d, J=7 Hz, 3H) ppm. C NMR (125 MHz,
CD OD): δ 171.1, 167.7, 157.9, 142.6, 95.9, 91.2, 79.8, 77.5, 49.4, 41.8, 36.9, 35.2, 32.7, 27.2,
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-
2
4.2, 22.3, 21.0, 16.4 ppm. ESI-MS : m/z calcd for: C H N O S:381.17 Found: ES [M -1]: 380.2,
1
8
27
3
4
+
and ES [Dimer M +1]:763.2; HPLC purity: 98.4% with major by-product is sum of impurities at
RRT about 1.01 & 1.02 observed 1.6%; chiral HPLC purity: 97.4% ; Loss on drying content: 2.6%.
4
-amino-1-[(2R, 5S)-(2-hydroxymethyl)-1,3-oxathiolan-5-yl]-1,2-dihydropyrimidin-2-one
(S)-BINOL co-crystal (4) : To the stainless steel reactor, water (225 L) was charged and cooled
to 15-20°C followed by di-potassium hydrogen phosphate (205.5 kg, 1162.45 mol) and stirred for
15-20 min at 15-30°C. 3 (150 kg, 393.19 mol) and denatured spirit (1125 L) were added to the
reactor and stirred for 15-30 mins. The reaction mixture was further cooled to 10-15°C. A solution
of NaBH was prepared by dissolving 45 kg of NaBH in the precooled (20-25°C) water (450 L)
4
4
containing 4.5 L of 25% w/w aq. NaOH solution. NaBH solution was then added to the reaction
4
mass slowly over the period of 2 h at the temperature at 10 to 15°C. Reaction progress monitored
by HPLC to control the content of 3 below 0.5 % limit. After completion of the reaction, the mass
of reaction was settled, and phases were separated. The lower aqueous phase was discarded, and
upper phase was treated with 50% aq. HCl solution (approx. 48 L) and pH was achieved about 4
to 5. Furthermore, pH was adjusted 6.8 to 7.2 using 10% aq. NaOH solution (35 L). The reaction
solvent was then distilled out under vacuum below temperature 50°C till a stirrable volume
remains in the reactor. Further, water (375 L) was added to the reaction mixture and reaction mass
heated up to 55 to 60°C for 30 mins. The aqueous reaction mass further cooled to 20-30°C and
washed with dichloromethane 3 times with 300 L each (By product l-menthol was extracted in this
layer). The aqueous reaction mass was finally distilled out under vacuum completely and degassed
ACS Paragon Plus Environment
2
7

Organic Process Research & Development
Page 28 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
for 1 h at the temperature below 50°C. The reaction mixture co-distilled with denatured spirit (300
L) for two times and further denatured spirit (300 L) was charged into the reactor. The water
content of this reaction mass was controlled below 2.0% limit. Further quantity of denatured spirit
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(1500 L) was charged to the reaction mass and heated to achieve temperature level at 70 to 75°C
and maintained for 60 mins. The undissolved inorganic solid was hot filtered through sparkler
filter at temp 55-60°C and the bed was washed with denatured spirit (75 L). The filtrate was
charged into the reactor and the solvent was distilled out under vacuum at temperature below 50°C
up to stirrable volume. The reaction mass was further cooled at 20 to 30°C. Methanol (1200 L)
and (S)-BINOL (112.5 kg, 392.9 mol) were added to the reaction mass. The reaction mass then
heated to get a clear solution at 60-65°C. The reaction mass further treated with activated charcoal
(
7.5 kg) and hot filtered through celite bed with sparkler filter followed by bed wash with methanol
150 L). The filtrate was charged into the reactor and the solvent was distilled out under vacuum
(
at temperature below 50°C till 300 L solvent remained in the reactor. The reaction mass cooled at
0 to 30°C. The reaction further cooled at 0 to 5°C and stirred for 60 to 120 mins. The solid
2
suspension was filtered through centrifuge. Centrifuge was kept under spinning till the removal of
traces mother liquor. The wet cake again spray washed with precooled methanol (75 L) in
centrifuge under spinning till completed expulsion of mother liquor. The wet cake was removed
from the centrifuge and analyzed for enantiomer and diastereomer content by HPLC method with
enantiomer content limit not more than 0.25% and diastereomer content limit not more than 0.5%.
The solid wet cake was dried for 4 to 8 h in a vacuum oven at 45°C under vacuum to provide
1
1
43.06 Kg of 4. H NMR (500 MHz, CD OD):δ 8.06-8.08 (d, J=7.5 Hz, 1H), 7.83-7.89(dd, J =21
3
1
Hz, J =8.5 Hz, 4H),7.30-7.31(d, J=8.5Hz, 2H), 7.24-7.27 (t, J=1Hz, 2H,), 7.16-7.19 (t, J=8.5Hz,
2
2H), 7.02-7.04 ( d, J=9Hz, 2H), 6.29-6.31 (t, J=5.5Hz,1H,), 5.88-5.90 (d, J=7.5 Hz, 1H,), 5.27-
ACS Paragon Plus Environment
2
8

Page 29 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
5
.29 (t, J=4.0 Hz, 1H), 3.93-3.94 (dd, J =12.5Hz, J =3.5Hz, 1H), 3.88-3.89(dd, J =12.5Hz,
1
2
1
J =4.5Hz, 1H), 3.51-3.54 (dd, J =12Hz, J =5.5Hz, 1H), 3.12-3.15 (dd, J =11.7Hz, J =4.2Hz, 1H)
2
1
2
1
2
1
3
ppm. C NMR (125 MHz, CD OD): δ 167.7, 157.9, 154.1, 142.7, 135.8, 130.5, 130.4, 129.0,
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
27.1, 125.8, 123.8, 119.2, 116.1,95.6, 88.8, 87.9, 64.03, 38.4 ppm. ESI-MS : m/z calcd for
+
-
C H N O S: Found: ES [Dimer M +1]:459.0 and for (S)-BINOL ; ES [Dimer M -1] :285.0,
2
8
25
3
5
Chiral HPLC purity: 99.60%, enantiomer content 0.03%, diastereomer content by HPLC method
was 0.42%, Loss on drying content: 0.4% w/w.
4
-amino-1-((2R, 5S)-2-(hydroxymethyl)-1, 3-oxathiolan-5-yl) pyrimidin-2(1H)-one (1) : To
the stainless steel reactor, water (525 L) was charged and cooled to 2 to 8°C followed by 4 (105
kg, 203.86 mol) and stirred for 60 to 120 mins. The solid suspension was filtered through
centrifuge. Centrifuge was kept under spinning till the removal of traces mother liquor. The wet
cake again spray washed with precooled water (100 L) in centrifuge under spinning till the
expulsion of mother liquor was completed. The wet cake was removed from the centrifuge and
charged into the reactor containing water (525L). Ethyl acetate (525 L) was then charged and
heated with the reaction mixture at 40-45°C. The reaction mixture was then stirred to get a clear
solution. The reaction mixture was further cooled and settled at 25 to 30°C. The phases were
separated, and aqueous layer washed again with ethyl acetate (525 L). (S)-BINOL was recovered
from ethyl acetate layer by distillation solvent (97%). Aqueous layer analyzed for (S)-BINOL
content by HPLC to control its limit below 50 ppm. Aqueous layer further transferred to reactor
through micron filter and the solvent was distilled out under vacuum at below 50°C till a thick
mass was obtained. Denatured spirit (92 L) was added to the reactor and stirred for 15-30 min.
Further, water content of the reaction mass was checked which found between 10 to 18% w/w.
The reaction mass was then heated the reaction mass at 55 to 75°C to get a clear solution.
ACS Paragon Plus Environment
2
9