131

CHAPTER-3

STUDIES TOWARDS NEW AND IMPROVED

SYNTHESIS OF VALACYCLOVIR

132

3.1 INTRODUCTION

Viruses [95a] are almost as versatile as bacteria in the range of

diseases they can cause. A virus is defined as various numbers of

submicroscopic parasites that can infect any animal. All viruses replicate

only within living cells and cannot usually survive for long time outside

the host cell. Most prominent mechanism of viruses entry to human body

is by the absorption to superficial cells. [95b-c].

Out of several viral families across the globe, one of the most

commonly existing viral families is represented by Herpes viruses [96, 97

a-c]. Table-3.1 gives a glimpse of the herpes viruses that infect humans.

Table 3.1: Herpes viruses that infect humans

Type Synonym

Human herpesvirus 1 (HHV-1) Herpes simplex virus 1 (HSV-1)

Human herpesvirus 2 (HHV-2) Herpes simplex virus 2 (HSV-2)

Human herpesvirus 3 (HHV-3) Varicella-Zoster virus (VZV)

Human herpesvirus 4 (HHV-4) Epstein-Barr virus (EBV),

Human herpesvirus 5 (HHV-5) Cytomegalovirus (CMV)

Human herpesvirus 6 (HHV-6) Roseolovirus

Human herpesvirus 7 (HHV-7) Roseolovirus

Human herpesvirus 8 (HHV-8) Kaposi's sarcoma-associated

herpesvirus (KSHV),

Significant progress has been achieved in the development of antiviral

compounds [98] and an increasing number of antiviral agents are now in

regular use (Table-3.2). Many are nucleoside analogues that interfere

with viral replication, but other targets have also been successfully

133

exploited [99]. Much research emphasis has been concentrated on agents

that act to inhibit the replication of HIV [100].

Table 3.2: Principal antiviral agents (other than anti-HIV agents) in

present use

Compound Indication Mode of action Route of

administration

Acyclovir Herpes simplex;

Varicella-zoster Nucleoside analogue

Oral; topical;

intravenous

Adefovir

dipivoxil Hepatitis B Nucleotide analoguea Oral

Amantadine Influenza A Uncoating of virus Oral

Cidofovir Cytomegalovirus Nucleoside analogue Intravenous

Entecavir Hepatitis B Nucleoside analogue Oral

Famciclovir Herpes simplex

Varicella-zoster Nucleoside analoguea Oral

Fomivirsen Cytomegalovirus Anti-sense oligonucleoside Intra-ocular

Foscarnet Cytomegalovirus DNA polymerase inhibitor Intravenous

Ganciclovir Cytomegalovirus Nucleoside analogue Intravenous

Interferon-I Chronic hepatitis Immunomodulation Subcutaneous

Lamivudine Chronic hepatitis Nucleoside analogue; reverse

transcriptase inhibitor Oral

Ribavirin Respiratory

syncytial virus

Chronic hepatitis

Nucleoside analogue Nebulizer

Oral

Valacyclovir Herpes simplex;

Varicella-zoster Nucleoside analoguea Oral

Valganciclovir Cytomegalovirus Nucleoside analoguea Oral

Oseltamivir Influenza Neuraminidase inhibitor Oral

Zanamivir Influenza Neuraminidase inhibitor Inhalation

a Pro-drug formulation.

134

By and large, an antiviral drug exerts its effect by inhibiting the Virus

by any of the following methods; (a) binding with the receptor molecule

(b) avoid the replication and (c) stopping the virus from releasing its

proteins [101].(d) by strengthening the immune system of the body

[102a].

Valacyclovir, 49 a most promising second generation nucleoside

analogue DNA polymerase inhibitor which rapidly converted to acyclovir,

45 which has demonstrated antiviral activity against HSV types 1 (HSV-

1) and 2 (HSV-2) and VZV both in cell culture and in vivo [102b]. The

chemical name of 49 is L-valine, 2-[(2-amino-1,6-dihydro-6-oxo-9H-

purin-9-yl)methoxy]ethylester mono hydrochloride salt. Valacyclovir (49)

is indicated for the treatment of; herpes zoster (shingles), suppression of

genital herpes in immunocompetent individuals, suppression of

recurrent genital herpes in HIV-infected [102c] individuals and is also

indicated for the treatment of cold sores (herpes labialis).

This chapter is dedicated to discuss the process research and

development activity of valacyclovir (49) carried out in our laboratory.

Before discussing the process development activity, a small section to

introduce the history of valacyclovir, its mechanism of action, and

detailed review on available synthetic processes from its birth to till date

is illustrated.

135

3.1.1 Mechanism of action

Valacyclovir (49) is well absorbed to release 45 into the bloodstream

[103]. Uptake of acyclovir is enhanced in herpes virus-infected cells,

presumably because of its rapid activation to the monophosphate form,

46 catalyzed by a herpes virus-encoded thymidine kinase (TK) [104].

(Step-1, Figure-3.1) Subsequent conversion of the monophosphate to the

active form, acyclovir triphosphate, 48 is accomplished by host-cell

enzymes. Compound 48 functions as a substrate for viral, but not

cellular, DNA polymerase, competing with deoxyguanosine triphosphate

for incorporation into the elongating chain [105]. The incorporation of

acyclovir triphosphate into the growing chain of viral DNA results in

chain termination, as acyclovir lacks the 3'-hydroxyl group necessary for

subsequent elongation (Step-2, Figure-3.1)

Structurally, acyclovir is acycloguanosine, an analogue of the purine

nucleoside, guanosine, in which the deoxyribose moiety has lost its cyclic

configuration. Acyclovir itself [106a] is inactive, in order to achieve its

antiviral effect, it must first be phosphorylated to the triphosphate form

within the infected cell [106b]. Although the second and third phosphate

groups are added by cellular enzymes, the initial phosphorylation step is

accomplished by a viral thymidine kinase, specified by herpes simplex

and varicella zoster viruses. The cellular form of this enzyme is much

less efficient in producing acyclovir monophosphate. This unique feature

is the basis for the selective toxicity of acyclovir, for two reasons: first, it

136

means that the active form of the drug is produced only in virally infected

cells; secondly, by the law of mass action. In biochemical assays,

acyclovir triphosphate inhibits replication of herpes viral DNA. [107].

Figure 3.1: Acyclovir: Mechanism of action (Step 1: activation)

Figure 3.1: Acyclovir: Mechanism of action (Step 2: incorporation

into growing DNA chain)

137

3.2 REVIEW OF LITERATURE

3.2.1 Development history of Valacyclovir and its pharmaceutical

properties

Among the several structural analogues of acyclovir [108-110],

valacyclovir, 49 is one of the successful candidates to be enjoyed as a

drug. The later is metabolized into the former after oral administration

similar to famciclovir [111], 51 a pro-drug of penciclovir, 50 [112]

(Figure-3.2). These two drugs are subsequently approved by the U.S Food

and Drug Administration (FDA) for the treatment of acute herpes zoster

and herpes simplex genitalis [113].

Figure 3.2: Second generation antiherpes drugs

Since, we are primarily interested in valacyclovir (49), systematic

review of literature on various synthetic methodologies adopted by

various researchers has been detailed here.

138

3.2.2 Review of reported synthetic approaches

The synthesis of valacyclovir (49) reported till date, mainly starts from

acyclovir (45) by its condensation with protected valine derivatives. The

philosophy of synthetic routes to 49 is briefly described in the following

section.

3.2.2.1 Beauchamp et al. approach: The first synthesis [114] for 49

involves condensation of acyclovir (45) with N-Cbz-L-valine (52) in the

presence of 1,3-Dicyclohexylcarbodiimide (DCC) in the presence of

catalytic amount of dimethylaminopyridine (DMAP) in DMF followed by

hydrogenolysis of resulting N-Cbz protected valacyclovir (53) using 5%

Pd/C catalyst as depicted in Scheme-3.1.

Scheme 3.1: Beauchamp et al. synthetic approach to Valacyclovir.

Similar kind of work reported by Wang and co-workers [114b] involves

condensation of methoxy (54a) or nitro (54b) substituted N-Cbz-L-valine

as protected aminoacid as depicted in Scheme-3.2. Once the

condensation is achieved, hydrogenolysis of the resulting intermediates

using Pd/C in formic acid resulted 49.

139

Scheme 3.2: Synthesis of Valacyclovir by Wang and co-workers.

3.2.2.2 Jackson et al. approach: Jackson et al. [115] described an

improved process for the preparation of 49 by condensing acyclovir (45)

with N-carboxyanhydride of Z-valine (56) in the presence of DMAP in a

single step with 97.7% yield and 95% purity (Scheme-3.3).

Scheme 3.3: Jackson’s synthetic approach.

The process may be performed in a single step without the need for

protection and subsequent deprotection and is thus much simpler and

quicker to perform than the previously reported procedures.

3.2.2.3 Montoro et al. approach: Montoro et al. [116] developed a new

route by sequentially converting the hydroxy group in the side chain of

acyclovir (45) to a good leaving group Z (Z= tosylate, or mesylate, or

140

chloride), and reacting the obtained protected acyclovir (59) with alkaline

salt of protected valine 60 to provide the N-substituted valacyclovir (61).

Hydrolysis of 61 using HCl provided valacyclovir HCl (49) with 78% yield

(Scheme-3.4).

Scheme 3.4: Montoro et al. synthetic approach

Incorporation of protection-deprotection, more number of reaction stages

in Montoro‟s synthetic route made this process less attractive for

commercial production.

141

3.2.2.4 Etinger Marina Yu et al. approach: The process reported by

Etinger et al. [117] involves the condensation of acyclovir (45) with N-

Boc-L-valine (62) in the presence of N-(3-dimethylaminopropyl)-N-

ethylcarbodiimide as a coupling agent to get the N-BOC protected

valacyclovir (63) with 97% yield and 98% purity. The N-BOC group was

then removed by hydrolysis using 12N HCl furnishing Valacyclovir

hydrochloride with 92% yield and 98% purity by HPLC. (Scheme-3.5)

Scheme 3.5: Etinger et al. synthetic approach

3.2.3 Summary of literature review

Careful examination of the reported approaches reveal that they follow

common approach in designing the synthesis, by protecting amino group

in L-valine with different protecting groups followed by removing the

protected groups at the end. Beauchamp‟s first synthetic approach to

valacyclovir appears to be promising, but covers insufficient information

on control of process related impurities specifically the enantiomer (D-

isomer).D-Isomer impurity is possible due to the carryover from L-Valine

and or racemisation during reaction conditions. Montoro‟s synthetic

142

route though economical compared to Beauchamp‟s route but involves

multiple stages. Etinger et al approach also involves protection-

deprotection stages though simple compared to both Beauchamp‟s and

Montoro‟s synthetic approaches. Jackson‟s approach is a simple, safe,

quick and economical process for the preparation of 49 and is

appreciable compared to the other approaches. But, it is silent on the

formation of isomeric impurity; moreover, the bulky protecting groups

used in almost all the known synthetic routes may cause steric

hindrance and resulting to incomplete formation of the required L-valine

ester. The protection-deprotection steps further increase the risk of

racemization leading to formation of other isomer in the product. In view

of above limitations associated with reported procedures, we felt a need

for not only a simple, impurity free, cost effective and scalable process,

but also to have an IP advantage by means of freedom to operate to enter

into the market.

3.3. OBJECTIVE OF THE PRESENT WORK

With the above background, we adopted two approaches to achieve

the goal. First approach is the identification of alternative route of

synthesis and second one being the improvement of the original route to

achieve 49 by circumventing the said disadvantages. Both the

approaches have been described separately in two sections as below.

143

Section 1. Development of an alternative, facile, and efficient synthesis to

49

Section 2. To provide an improved process for 49 by circumventing the

challenges associated with the reported process.

3.4 RESULTS & DISCUSSION

3.4.1 Section-1: Studies towards new synthesis of Valacyclovir

To circumvent the problem of steric hindrance and racemization, we

thought to use a small size versatile synthetic precursor of amino group

such as azide functionality. The azido group being an excellent masked

amino equivalent is relatively stable and can be readily reduced to the

desired amino function when required. This protocol led to the synthesis

of valacyclovir, 49 from acyclovir, 45 and the results are presented here.

Our synthesis of valacyclovir [118] started with reaction of imidazole,

64 with sodium azide and sulfuryl chloride to furnish hydrochloride salt

of Imidazole-1-sulfonyl azide (65), which is then treated with L-valine

(57) to give α-azido acid derivative of L-valine (66) as shown in Scheme-

3.6. Compound 65 is proved to be equal to triflylazide in its ability to act

as diazo-donor in the conversion of α-amino acid to α-azido acid. The

azido acid (66) was characterized by proton NMR which showed the

presence of acidic proton at 11.5 ppm and by the absence of amino

protons of valine moiety. Further, the structure of 66 was confirmed by

144

the presence of IR absorption frequency peak at 2110 cm-1 corresponds

to azide moiety.

Scheme 3.6: Preparation of 2(S)-azido-3-methylbutyric acid (66).

The 2(S)-azido-3-methylbutyric acid (66) obtained is then condensed

with acyclovir (45) in presence of DCC and DMAP to obtain a novel

intermediate, 9-[(2-(2S-Azido-3-methylbutyroxy) ethoxy) methyl] guanine

(67) as shown in Scheme 3.7. The structure of 67 was confirmed by

HRMS, which shown the mass value of 325 against the calculated value

of 325 that corresponds to mass value of 67. The IR frequencies at 2099

cm-1 and 1732 cm-1 indicated the formation of desired azido ester (67).

Further, the presence of azomethane proton as a singlet at δ value of 7.8

and protons of two methyl groups of isopropyl moiety as two doublets at

δ values at 0.80 and 0.89 respectively and the absence of alcohol proton

of acyclovir moiety and acid protons of azido acid moiety in the proton

NMR confirmed the structure of 67 (Figure-3.13b). Finally, the catalytic

reduction of 67 with raney nickel yielded the valacyclovir freebase. Mass

spectrum showed protonated molecular ion at m/z 325. IR Spectrum

145

displayed absence of azide stretching frequency. Further, carbonyl

stretching at 1732 cm-1 along with significant maxima at 3327, 3110 and

1H NMR displayed singlet for azomethane proton at δ 7.84, amido proton

at δ 10.81 and isopropyl protons in the range of δ 0.80-0.90. The

freebase is then treated with aqueous hydrochloric acid to form

Valacyclovir hydrochloride (49) with an overall yield of 67.3% as depicted

in Scheme-3.7.

Scheme 3.7: New synthesis for Valacyclovir HCl (49) using azido L-

valine (66).

Comparison of obtained sample with Innovator sample [113] by HPLC

revealed that the principle peak obtained using Innovator product

matches with the product obtained by our route as described below. The

analytical results by HPLC [119] were depicted in Figure-3.3, 3.4 & 3.5

which revealed that the product obtained is matching with the innovator

samples and confirms the feasibility of the product as per the synthetic

methodology followed as depicted in Scheme-3.7.

146

AU

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Minutes

5.30 5.40 5.50 5.60 5.70 5.80 5.90 6.00 6.10 6.20 6.30 6.40 6.50 6.60 6.70 6.80

Valtrax

(Innovator)

Valacyclovir

(Inhouse)

Figure 3.3: Overlaid chromatogram of innovator tablet (Valtrex) with new route Valacyclovir sample.

AU

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1.30

1.40

1.50

nm

200.00 210.00 220.00 230.00 240.00 250.00 260.00 270.00 280.00 290.00 300.00 310.00 320.00 330.00 340.00 350.00 360.00 370.00 380.00 390.00

252.5

Valtrax

Figure 3.4 UV spectra of Valtrex-innovator tablet.

AU

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

nm

200.00 210.00 220.00 230.00 240.00 250.00 260.00 270.00 280.00 290.00 300.00 310.00 320.00 330.00 340.00 350.00 360.00 370.00 380.00 390.00

252.5

Valacyclovir (Inhouse)

Figure 3.5 UV spectra of Valacyclovir (49)-new route sample

147

Having established the feasibility of the route, selected at research

phase, we have taken up optimization of each step to establish the

robustness and ruggedness for process for making 49. The stage wise

optimization details of the route are discussed below in detail.

3.4.1.1 Optimization studies for coupling reaction:

Coupling experiment performed using DCC [120] in DMF for the

synthesis of 67 during feasibility was discouraging because of tedious

process of removing byproduct DCU from the product and also low yield.

During optimization we have explored the coupling reaction using

various acid activating agents such as acid chlorides, mixed anhydride

and using various dehydrating agents such as, EDC [121], MeSO2Cl-Et3N

and N,N‟-carbonyldiimidazole [122] (CDI). To our surprise, reaction did

not proceed with acid chloride prepared using thionylchloride or

oxalylchloride. Reaction using methane sulfonyl chloride as acid

activating agent gave poor yield and quality. Though reaction is going for

completion using EDC, CDI, and DEAD, yields and quality obtained were

not encouraging as shown in Table-3.3

Table 3.3: Selected esterification attempts for the synthesis of 67.

entry reagent catalyst solvent reaction

time (h)

reaction

temp (°C)

yield

(%)

1

EDC

DMAP DCM 6 28 62

2 HOBT DMSO 5* 28 55

3 Et3N,DMAP DMF 4* 28 59

4

DCC

DMAP DCM 4* 28 69

5 DMAP DMSO 2 28 72

148

6

DCC

DMAP DMF 1 28 82

7 HOBT DMF 2.5 28 77

8 Et3N DMAP DMF 15** 28 45

9 MeSO2Cl DMAP MeCN 4* 45 59

10 MeSO2Cl Et3N MeCN 3* 45 62

11 CDI --- DMSO 24* 80 58

12 CDI --- THF 8 65 66

13 DEAD,

TPP --- DMF 20* 60 45

14 DEAD,

TPP --- THF 20* 65 70

*poor reaction mass quality less than 70%; **incomplete reaction less than 85%

Above results showed that coupling using DCC was the preferable

approach for further optimization.

Yield improvement using acyclovir variants:

After feasibility studies of esterification using DCC, we further wanted

to excavate possible improvements for specific coupling of acyclovir (45)

and azido L-valine (66). We presumed that by modifying the substrate,

acyclovir, chemically by protecting amine group at C-2 position with

various labile nucleophiles to alter the solubility profile there by to

achieve enhancement in the reaction yields. Once the coupling is done

successfully, deprotection can be done easily to get desired amine group

by simple chemical modifications (Scheme-3.8).

149

Reaction conditions: (a) 2(S)-azido-3-methylbutyric acid, DCC, DMAP, DMF, 10-15 °C

(b) 69a: HCl, 50 ºC; 69b: NaOH, 50 ºC; 69c: AcOH, 50 ºC.

Scheme 3.8: Synthesis of α-azidoesters from acyclovir variants

Compound 68a was prepared by treating 45 with

dimethylformamidodimethylacetal under neat conditions or using solvent

medium at above 70 0C followed by isolation at ambient temperature to

give DMF protected acyclovir (68a) with 98% purity by HPLC. Compound

68a was characterized using different spectroscopic techniques. N2-

acetyl acyclovir, 68b was prepared by the partial hydrolysis of N2,9-

diacetyl acyclovir and 68c was synthesized using trityl chloride. The

structures of 68a and 68b were characterized by different spectroscopic

techniques. As envisaged, the derivatives 68a-68c had shown dramatic

improvement in the solubility profile as compared to the parent

compound 45 in DMF which led to better yields during esterification. The

measured solubilities of acyclovir variants in DMF in the increasing order

150

are Acyclovir (150 volumes), Acetyl acyclovir, 68b (80 volumes), DMF

acyclovir, 68a (75 volumes) and trityl acyclovir, 68c (5 volumes) as

captured in Figure-3.6.

8285

92 95

150

80 75

5

0

50

100

150

200

250

Acy Ac-Acy DMF-Acy Tr-Acy

Comparison of solubilities and yields

% yield Solubility in DMF (No. of times)

Figure 3.6: Comparison of solubility of 45, 68a-c and yields during

condensation

It was observed that trityl protected acyclovir resulted in enhanced

yield (95%) over other protections or without protection when same

procedure was employed for esterification. Even though protected

acyclovir derivatives showed improvement in yield due to increased

solubility, cycle time and number of stage will increase for the synthesis

of Valacyclovir HCl. Further, trityl protected derivative (68c) generates

huge amount of by-product, trityl alcohol, on deprotection and adds to

cost and generates huge effluent made this option relatively less

attractive. Thus, we focused our attention on the optimization of DCC

catalyzed esterification using 45 and Azido acid, 66 [123].

151

3.4.1.2 Synthetic studies on azide reduction

During the feasibility studies towards the reduction of azide

intermediate,69 our attempts to achieve amine using triphenylphosphine

in methanol or toluene, [124] and borontrifluoride-etherate in THF using

iodine catalyst [125] were not encouraging because of acyclovir formation

due to undesired hydrolysis. Staudinger reaction [126] conditions using

TPP in pyridine medium at room temperature resulted in 80% of desired

amine along with the hydrolyzed product 45 as a major impurity. In

order to reduce the formation of acyclovir during reduction, reaction

conditions like base, solvent medium and temperature were optimized

which showed no improvement in the product quality as shown in Table-

3.4.

Table 3.4: Hydrogenation of azide (67) using chemical reduction

S. No. Reagent Solvent Reaction

temp (°C)

Reaction

time (h)

Yield

(%)

RM

Purity (%)

Product

Purity (%)

1 TPP Py RT 5 75 85 95

2 TPP Py 10 12 80 90 95

3 TPP

Toluene,

Py (10

equimoles)

RT 10 -- Undesired

reaction --

By considering these experimental outcomes, it was decided to

investigate azide reduction methodologies involving metal catalysts [127].

Hydrogenation reaction carried out using Platinum [128] as catalyst in a

mixture of methanol and water as a reaction medium at 26-28 °C for 8 h

followed by isolation to yield 49 resulted only 70% yield and hydrolyzed

152

product acyclovir, 45 formation was minimized. Lithium aluminium

hydride gave only 65% yield when followed similar procedure as per prior

art [129]. Finally, according to our feasibility studies we focused to

optimize the reduction reactions using Raney Nickel and Palladium on

charcoal [130] and results are captured in Table-3.5. It was clearly

observed that Raney-Ni would be the best choice for the azide reduction

using ethanol as the reaction medium which resulted almost 82% of

yield.

Table 3.5: Experimental studies on azide reduction using metal

catalysts

Based on complete evaluation of studies on azide reduction following

can be deduced. Though, chemical reduction was achieved through

Staudinger reaction, it led to a potential impurity due to the undesired

hydrolysis occurred by the usage of basic condition for the reaction

which led to poor amine yield. Moreover, it demands additional

purification for the elimination of the hydrolyzed product, acyclovir, 45.

These disadvantages were overcome by metal catalytic reductions under

entry reactant catalyst solvent reaction

temp (°c)

reaction

time (h)

yield

(%)

1

H2 Gas

10% Pd/C

Methanol 60-65 12 58

2 Ethanol 70-75 15 61

3 acetic acid,

water 60-65 12 55

4 Pd, AcOH Methanol 60-65 10 64

5 PtO2 Water,

methanol 60-65 8 65

6 Ammonium

formate Pd Methanol 60-65 24 65

7 --- Raney Ni Ethanol 70-75 7 82

8 LAH --- Ether 26-28 3 20

153

mild conditions. Therefore, in the current context, catalytic reduction

using Raney Nickel is more preferable over chemical reduction to achieve

optimum yield and quality of final product, 49.

Finally, valacyclovir salt was prepared using aqueous hydrochloric

acid at pH 2.5-3.0. The reaction mixture was treated with charcoal to

remove color impurities and the purity of the product is found to be

98.33% by HPLC and chiral purity is 97.06%.

3.4.1.3 Process safety assessment

Having completed the feasibility studies for the new route, our next

aim was to ensure safe scale up of the process. With the scale up there

is always an increased liability from accidents, culminate to injury of

personal, loss of equipment and delay in key deliverables. One of the

anticipated key challenges for our route of synthesis is handling of sodium

azide and azide intermediates. Hitherto there is no literature precedence

for safety studies during preparation of 66. Therefore it is imperative to

perform reaction safety assessment using reaction calorimeter (RC) [131],

to understand the thermal events associated with each step. The

preliminary dust explosion studies to find out an engineering solution

during scale up were also helpful in concluding the safety assessment.

The key outcomes are to be essentially used for designing of plant

equipment, services and operations to impart scalability for the optimized

process.

154

Investigations using reaction calorimeter: We have investigated the

thermal behavior during preparation of Azido acid (66) from imidazole,

64 using reaction calorimeter and summary of observations is outlined

in this section. Reaction Calorimeter (Mettler make, Model- SV 01 800m1

capacity, all glass reactor equipped with a propeller type up flow agitator,

jacketed with controlled temperature circulator, addition funnel,

condenser and the probes) was used for performing the safety studies.

A suspension of sodium azide (26 g) in toluene (260 ml) was cooled to 0 °C

and charged sulfuryl chloride (32.4mL) to it. The temperature was raised to 25

°C and maintained mass for l6 hrs at 25°C- 30°C. The above solution was

charged into RC reactor, cooled to 0 °C and added imidazole (49 g) to the

reactor in portion wise. Temperature of the mass was raised to 25 °C and

continued stirring at 25 °C for 3hrs. Reaction mass was then cooled to 0°C, 5%

Na2CO3 solution (26O mL) was added, raised the temperature to 25 °C and

continued stirring for 15 min. Toluene and aqueous layer were separated,

toluene layer was washed with water (200 mL X 2), further with 5% brine

solution (200 mL) and again charged to the RC reactor. Ice-cooled HCI solution

(25mL conc. HCl in l00mL of water) was added to the solution and maintained

the temperature at 0 °C for 15mins before separation of layers. Meanwhile

water (275mL) was taken in a separate RB flask, cooled to 0 °C and added

K2C03 (70.8 g) followed by L-Valine (27.4 g) and CuSO4. 5H20 (50mg) at 0 °C.

This Valine complex was then charged to the empty RC reactor and stirred at 0

°C. To the above solution charged the aqueous layer (collected after HCI

addition) and maintained reaction temperature at 0 °C and continued stirring

155

for 1 hr before the reaction mixture was drained-off. The brief details are

summarized in Figure-3.7.

Figure 3.7: Step wise observations of reaction calorimetric studies

Summary of investigations: Our investigations with Reaction calorimeter

studies have revealed that, out of all the reaction steps “Imidazole addition” is

one of the first key step which can lead to an adiabatic temperature rise of

about 32 °C with an overall evolved heat of 261.2KJ/Kg of Imidazole. Second

key step is aqueous layer containing Imidazole-1-sulfonyl azide as

hydrochloride salt (65) addition to L-Valine complex which can lead to an

156

adiabatic temperature rise of about 10°C with an overall evolved heat of 236.7

KJ/Kg of imidazole.

Precussion Studies: Physical operations like drying, milling, sieving and

packing are commonly employed operations post isolation of an

intermediate and/or an active pharmaceutical ingredient in

manufacturing. The impact sensitivity of a product is easily measured to

understand the sensitivity of a product to shock and therefore exposable,

if it disintegrates with a hammer upon its contact to shock energy under

given test conditions. The process of this measurement is known to be

precussion test or fall hammer test [132].

General test procedure involves; Place the test sample (100mg

sample) enclosed in aluminum foil between the upper and lower parts of

the stamp and let the drop (Drop weight 5Kgs) hammer fall ( drop height

80-100cms). Generally, this test is performed in presence of two persons

in a dark room, to observe the reactions keenly. When a test sample

burns or produces sparks or smoke, it is construed as positive [133].

Three specimen samples of 67 from various experiments were subjected

to “fall hammer test” and found that it is “negative”. It implies that α-

azido ester (67) is not shock sensitive and can be handled in a

pharmaceutical manufacturing environment with normal powder

handling conditions without any safety hazard.

157

Based on safety assessment studies, we are pleased to find that we

could achieve our goal by maintaining sufficient cooling requirements

and we safely were able to test the process robustness under pilot plant

and successfully prepared multi kilogram amounts of 66 and challenges

anticipated in handling azide intermediates have been well addressed

during our safety assessment studies. Therefore, reaction calorimeter

proved to be an invaluable tool in the development of a safer procedure to

prepare 66.

3.4.2 Section-2: Studies towards an improved synthesis of

Valacyclovir (49)

Though we have a new route as described above, the synthesis of

valacyclovir (49) reported by Beauchamp and co-workers involving

condensation of acyclovir (45) and N-carbobenzyloxy-L-valine (52, N-Cbz-

L-valine) followed by de-protection of resulted N-Cbz protected

valacyclovir (53) using palladium catalyst with 55% yield (Scheme-3.9)

was taken for improvement. We have developed an improved process

[134] making use of the foresaid reported process.

158

Scheme 3.9: Synthetic scheme of Valacyclovir (49)

After critical evaluation of the scheme, following criticalities were

identified while practicing Beauchamp‟s approach and are listed below;

1. 3-4% of enantiomer (D-Isomer, 70) was formed [135] along with N-Cbz

valacyclovir (53).

2. Metal impurities palladium and aluminium carried from

palladium/alumina reagent and could not be removed to the

desirable levels by the conventional purifications.

3. Column chromatography and/or repeated crystallizations were

required to meet the desired purity.

4. Process was not economic to practice at industrial scale

159

Having understood the limitations of the reported process, we

attempted to improve the same by considering some of parameters to

address the issues and the process was described in detail below.

When condensation of 45 and 52 was performed in DMF using DCC

in the presence of catalytic amount of DMAP at about 60 °C, 3-4% of

unwanted D-isomer, 70 was observed along with the product, 53. The

formation of D-isomer is a function of reaction temperature. The results

of the controlled experiments were shown in Table-3.6. As the reaction

temperature is raised, impurity was gradually raised and reached to the

level of 3.3% at 25-30°C (Table-3.6, entry 8). Thus, the optimized reaction

temperature was fixed to -5 to 0 °C, wherein 70 was controlled to a level

of 1% in the reaction mass, and hence the temperature is one of the

critical parameter to control this impurity.

Table 3.6: Effect of reaction temperature on racemisaton

entry reaction

temperature (°C)

compound

70 (%)

01 -10-5 1.0

02 -5-0 1.0

03 0-5 1.1

04 5-10 1.0

05 10-15 1.5

06 15-20 1.8

07 20-25 2.7

08 25-30 3.3

160

However, the content of this impurity was increased significantly

during the scale up at the above set temperature. For example, in 50 kg

batch this impurity (70) was raised to 2.6-3% (Table-3.7).

Table 3.7: Variation in 70 content at different scales at -5 to 0 °C

entry scale compound

70 (%)

01 lab scale (upto 100 g) 1.0-1.8

02 kilo lab scale (upto 25 kg) 1.5-2.2

03 plant scale (upto 50 kg) 2.6-3.0

Further examination of the experiment revealed that compound 53

underwent racemization while distilling the DMF at 85 °C over a period of

10-12 h during the work-up of the reaction as shown in Table-3.8. From

this experiment it is understood that, though it is controlled in the

reaction part, distillation of DMF at higher temperature found to be the

culprit for the same at scale.

Table 3.8: Study on thermal sensitivity of N-Cbz Valacyclovir, 53

entry distillation condition compound

70 (%)

01 before distillation 1.1

02 distillation at 85 °C for 12 h 2.6

03 distillation at 85 °C for 30 h 3.2

Thus, the appropriate solvent system was sought to purify the

compound 53 in order to reduce the level of 70. Among the several

solvents explored, acetone-water and methanol-water solvent

161

combination were shown a better result in reducing the 70 content from

3.5% to 2.0%, which is an acceptable limit for this impurity in the

product as per European Pharmacopeia (Ph.Eur) monograph [138a].

Thus, aqueous acetone (entry nos. 01 & 02) was found to be the best

solvent system for the purification of N-Cbz valacyclovir (53) as shown in

Table-3.9 and was selected for purification.

Table 3.9: Solvent screening for the purification of N-Cbz

Valacyclovir (53)

entry solvent (s)

composition solvent ratio

solvent quantity

in volumes

70 (%)

01 acetone + water 2:1 18 2.0

02 acetone + water 2:1 18 2.1

03 acetone + water 4:1 15 2.3

04 acetone + water 20:1 42 2.3

05 methanol + water 2:1 30 2.3

06 Isopropanol - 8 2.4

07 methanol + water 2:1 30 2.5

08 1,4-dioxane - 25 2.5

09 ethyl acetate + water 2.5:1 7 2.5

10 acetone + water 2.5:1 7 2.6

11 Methanol - 5 2.7

12 methanol + water 3:1 20 2.7

13 DMF + water 0.2:1 24 3.0

14 Acetonitrile - 50 3.0

In the reported process for deprotection of Cbz group, N-Cbz

valacyclovir (53) was subjected to catalytic hydrogenation in a mixture of

162

methanol, THF and aqueous HCl. In our process for the deprotection,

catalytic hydrogenation reaction was conducted in DMF using palladium

supported on alumina. Filtration of the catalyst through celite followed

by usual work-up yielded crude valacyclovir (49) in 92% yield with 98.5%

purity. However, this isolated product contains corresponding

enantiomer (70) to an extent of 3.5-4.5%. As aqueous acetone solvent

used for the washability of 70 could not eliminate/reduce the content of

D-isomer of Valacyclovir, 71 it was decided to study the purification

procedure of Valacyclovir, 49 to achieve more enatiomeric purity. Hence,

various solvents were screened to remove D-isomer (71) from the

compound 49. Acetonitrile-water combination (entries 13 and 14) was

found to be ideal in reducing the 71 content from 4.2% to 2.7% as shown

in Table-3.10.

Table 3.10: Solvent screening for the purification of Valacyclovir

(49)

entry solvent (s)

composition

solvent

ratio

solvent qty. in

vol.

71

(%)

01 Methanol -- 10 4.57

02 Methanol -- 10 4.16

03 DMF -- 5 3.59

04 1,4-dioxane -- 10 3.47

05 tert-butanol -- 10 3.44

06 Dichloromethane -- 10 3.44

07 THF -- 10 3.44

08 methanol + water 12:3 12.5 3.22

09 methanol + water 11.25:3 19 3.00

163

10 methanol + water 7.5:3 14 2.98

11 ethanol + water 10:3 13 2.88

12 acetonitrile + water 1.2:3 7 2.88

13 acetonitrile + water 0.9:3 13 2.78

14 acetonitrile + water 0.91:3 13 2.74

After addressing the D-isomer issue in both the stages, we have

focused on the reduction of the content of heavy metals (palladium and

aluminum) present in 49. Despite hyflow bed filtration, these two metals

were found in compound 49 at a level of 60-90 ppm. Furthermore,

activated charcoal and activated clay were of little use in removing these

metallic impurities. In this context, we have used various resins and T-

63† resin, 72 (entry 2) was found to be very efficient in reducing the

heavy metal content to a level of less than 10 ppm [136] using 10% load

as shown in Table-3.11.

Table 3.11: Screening studies to remove metal impurities

entry solid support solvent resin load

(% w/w)

Al content

(ppm)

Pd content

(ppm)

01 EDTA water 20 -- 16.19

02 T-63† resin (72) water 10 4.2 1.48

03 T -63† resin water 50 4.6 3.51

04 T-63† resin water 100 6.3 0.70

05 CH-97† resin water 20 12.6 95.20

06 CH-97† resin water 50 10.7 23.94

07 CH-97† resin water 100 8.9 18.21

† manufactured by THERMAX LTD., India.

164

Besides the control of D-isomer content and heavy metal impurities, it

is also important to control individual impurity level [137] below 0.05%

(by HPLC) in the active compound (49). Thus, substantially pure

valacyclovir hydrochloride [138] was obtained from the technical grade

material by following the crystallization procedures. For instance,

technical grade material is dissolved in DMF at ambient temperature

and the product was crystallized by addition of IPA as an anti-solvent to

get pure 49. Further, if it is required to improve the purity of product,

the same purification procedure can be repeated as described in the

experimental section. All the impurities were checked after

crystallizations by HPLC and found that they are very well within the

specified limit. Product quality at various stages is depicted in Table-

3.12.

Table 3.12: Stage wise purity and D-isomer details

compound

53

compound

53 after

purification

tech

grade

API (49)

purified

API (49)

Purity (%) 98.5-99.0 99.0-99.5 98.3-

98.8 99.3-99.6

D-isomer %) 1.5-3.5 1.0-2.1 1.5-3.8 1.0-2.5

The content of impurities present in the purified valacyclovir (49) was

tabulated in Table-3.13.

165

Table 3.13: Content of impurities by HPLC in purified 49

Impurity Entry-1 Entry-2 Entry-3 Entry-4 EP44a

limit (%)

In-house

limit (%)

guanine† 0.003 0.003 0.003 0.003 2.00 0.15

acyclovir† 0.23 0.25 0.26 0.46 2.00 1.00

alanine† ND ND 0.01 0.01 0.10 0.10

O-acetyl† ND ND ND 0.002 0.20 0.15

N-formyl† 0.02 0.02 0.03 0.04 1.50 0.15

isoleucine† ND ND ND 0.009 0.10 0.10

N-Cbz valacyclovir † ND ND 0.008 ND 0.20 0.15

N-Cbz-L-Valine (52) ND ND ND ND NA 0.05

single maximum

impurity 0.02 0.01 0.01 0.0008 0.10 0.05

total content of

impurities 0.28 0.29 0.33 0.53 5.00 2.00

D-isomer (71) 2.3 2.5 2.6 2.8 3.00 3.00

† PHARMEUROPA Vol.18, No.2, April 2006

166

3.5 CONCLUSIONS

We achieved our goal in two aspects. Firstly, a new and concise

synthesis for Valacyclovir, 49 was developed by the azide precursor. The

azide chemistry is explored for the first time in the synthesis of

valacyclovir. Challenges anticipated in handling azide intermediates have

been well addressed during our safety assessment studies. We have seen

how the reaction calorimeter is a common tool to investigate chemical

reaction kinetics, to determine required data for chemical process safety

and to access fundamental information about phase changes. We have

presented that the integrated calorimetric approach substantiates that

the process parameters can be envisaged to ensure successful scale up of

optimized process at industrial level. The approach is extremely

attractive and also opens a new pathway to prepare other guanine

containing acyclic nucleosides.

Secondly, a facile and efficient process was developed for Valacyclovir

by controlling the related substances including chiral impurity following

the well known process by Beckump. Furthermore, process related heavy

metal impurities were effectively controlled to acceptable limits using

inexpensive and commercially available resin.

167

3.6 EXPERIMENTAL SECTION

General Procedures: All starting materials were commercial products.

The solvents and reagents were used without any purification. Melting

points (mp) were recorded with Buchi melting point B-540 instrument

and are uncorrected. IR spectra were recorded in the solid state as a KBr

dispersion using a Perkin- Elmer FT-IR spectrophotometer and only

diagnostic and/or intense peaks are reported. PNMR spectra were

recorded in CDCl3 and DMSO-d6 with Varian Mercury Plus 400 MHz

instrument. The chemical shifts are reported in δ ppm relative to TMS.

Multiplicity is indicated by one or more of the following: s (singlet), d

(doublet), t (triplet), q (quartet), m (multiplet), br (broad); Analysis of

metals was done by the use of inductively coupled plasma-optical

emission spectroscopy (ICP-OES). ICP-OES analysis is accurate to parts

per billion (ppb) to parts per trillion (ppt) range and applicable to most of

the metals in pharmaceuticals.

3.6.1 Preparation of N2-DMF acyclovir 68a: A mixture of acyclovir (50

g, 0.222 mol) and N,N-dimethylformamido dimethylacetal (500 mL, 5 vol.)

was maintained at reflux for 90 minutes. The reaction mixture was

cooled to 25 °C and water was charged for solid separation. The

separated solid was filtered and dried at 50 °C to give the desired product

68a (45 g, 98 % by HPLC).

168

3.6.2 Preparation of N2-Acetyl acyclovir 68b: N2,N9-Diacetylacyclovir

(25 g, 0.081 mol) was taken in methanol (125 mL) at ambient

temperature. TMS chloride (25 g, 0.231 mol) was then added (for

selective deprotection) to the mixture at 25 °C and maintained at the

same temperature for an hour. After completion of the reaction by TLC,

the separated solid was filtered followed by methanol washings. The solid

was dried at 60 °C to get the desired product 68b (20 g, 96.4% by HPLC).

3.6.3 Preparation of N2-Trityl acyclovir 68c: Acyclovir (25 g, 0.111

mol) was taken in DMF (250 mL) along with TEA (59.4 g, 0.588 mol). The

mixture was heated to 50 °C followed by drop wise addition of trityl

chloride solution in DMF (250 mL) at 50 °C (80.5 g, 0.289 mol). Reaction

mixture was maintained for completion of the reaction at 50 °C. Reaction

mass was cooled to room temperature and unwanted solids were filtered.

Add water (500 mL) to the filtrate and the separated solid was filtered.

The obtained filtrate was triturated with water (1000 mL). The reaction

mixture was aged for 1 hr and the solid was filtered and dried at 65 °C

under reduced pressure to get the desired mono N-trityl acyclovir 68c

(22 g, 98% by HPLC); IR (νmax, cm-1) 3468, 3101, 2931, 1686, 1632,

1579, 1269, 1120, 790; 1H NMR (DMSO-d6, 400 MHz) δ: 10.64 (s, 1 H),

7.73 (s, 1 H), 7.67 (s, 1 H), 7.3 (m, 15 H), 4.83 (s, 2 H), 4.51 (t, J = 5.6

Hz, 1 H), 3.20 (q, J = 5.6 Hz, 2 H), 2.92 (t, J = 4.8 Hz, 2 H).

169

3.6.4 Preparation of 69a: L-Valine azide (60%, 13.5 g, 0.056 mol) was

taken along with DMF (150 mL) and cooled to 10-15 °C. DCC (15.4 g,

0.074 mol) was charged under nitrogen and the mixture was aged for 30

minutes at 10-15 °C. DMF-acyclovir, 68a (10 g, 0.037 mol) was charged

followed by DMAP (0.686 g, 0.0056 mol) at the same temperature.

Reaction mass was aged for 2 h and solid was separated from the

reaction mass by filtration. Clear filtrate was subjected to cooling to 10-

15 °C followed by drop wise addition of water (750 mL) for separation of

solid. Separated solid was filtered and dried at 60 °C to give desired

product 69a (13.6 g, 90% by HPLC); IR (νmax, cm-1) 3085, 3044, 2102,

1741, 1663, 1633, 1113; 1H NMR (DMSO-d6, 400 MHz) δ: 11.33 (s, 1 H),

8.57 (s, 1 H), 7.93 (s, 1 H), 5.45 (s, 2 H), 4.32 (m, 1 H), 4.23 (m, 1 H),

4.10 (d, J = 4.8 Hz, 1 H), 3.16 (s, 3 H), 3.03 (s, 3 H), 2.03 (m, J = 6.8 Hz,

1 H), 0.88 (d, J = 7.2 Hz, 3 H), 0.80 (d, J = 6.8 Hz, 3 H).

3.6.5 Preparation of 69b: L-Valine azide (60%, 6.7 g, 0.028 mol) was

taken along with DMF (50 mL) under nitrogen atmosphere and cooled to

10-15 °C. DCC (11.25 g, 0.054 mol) was charged and the mixture was

aged for 30 minutes at 10-15 °C. A pre cooled (10-15 °C) mixture of N2-

acetyl acyclovir, 68b (5 g, 0.019 mol), DMAP (2.5 g, 0.020 mol) and DMF

(75 mL) was added to the azide solution at 10-15 °C. After completion of

the reaction by TLC, by-product was removed by filtration and the filtrate

was triturated with water (625 mL) at 10-15 °C. Separated solid was

170

filtered and dried at 70 °C to get the desired compound 69b (3 g, 97.2%

by HPLC); IR (νmax, cm-1) 3238, 3167, 3085, 2107, 1743, 1688, 1667,

1264, 1102; 1H NMR (DMSO-d6, 400 MHz) δ: 12.0 (s, 1 H), 11.78 (s, 1

H), 8.13 (s, 1 H), 5.48 (s, 2 H), 4.31 (m, 1 H), 4.21 (m, 1 H), 4.08 (d, J =

5.2 Hz, 1 H), 2.01 (m, J = 5.2 Hz, 1 H), 2.18 (s, 3 H), 0.88 (d, J = 6.8 Hz,

3 H), 0.79 (d, J = 6.8 Hz, 3 H).

3.6.6 Preparation of 69c: L-Valine azide (60%, 7.67 g, 0.032 mol) was

taken along with DMF (150 mL) and cooled to 10-15 °C. DCC (8.8 g,

0.042 mol) was charged under nitrogen and the mixture was aged for 30

minutes at 10-15 °C. N-Trityl-acyclovir, 68c (10 g, 0.021 mol) was

charged followed by DMAP (0.4 g, 0.0033 mol) at the same temperature.

Reaction mass was aged for 1 h and solid was separated from the

reaction mass by filtration. Clear filtrate was subjected to cooling to 10-

15 °C followed by drop wise addition of water (750 mL) for separation of

solid. Separated solid was filtered and dried at 60 °C to give desired

product, 69c (16.5 g, 90% by HPLC); IR (νmax, cm-1) 3326, 3117, 3050,

2928, 2102, 1743, 1706, 1625, 1090, 737, 699; 1H NMR (DMSO-d6, 400

MHz) δ: 10.66 (s, 1 H), 7.7 (s, 1H), 7.65 (s, 1 H), 7.21 (m, 15 H), 4.84 (s, 2

H), 4.05 (d, J = 5.2 Hz, 1 H), 4.01 (m, 1 H), 3.86 (m, 1 H), 3.77 (m, 2 H),

2.0 (m, 1 H), 0.88 (m, J = 6.8, 2 H), 0.81 (m, J = 6.8, 2 H).

171

3.6.7 Preparation of 9-[(2-(2S-Azido-3-methylbutyroxy) ethoxy)

methyl] guanine, 67: L-Valine azide (66, 90%, 21.3 g, 0.13 moles) was

taken along with DMF (400 mL) under nitrogen atmosphere and cooled to

10-15 °C. DCC was charged and the mixture was aged for 30 minutes at

10-15 °C. Acyclovir (20 g, 0.089 moles) was charged followed by DMAP

(1.62 g, 0.013 moles) at the same temperature. Reaction mass was aged

for 45 minutes and solid was separated from the reaction mass by

filtration. Clear filtrate was subjected to cooling to 0-5°C followed by drop

wise addition of water (1350 mL) for separation of solid. Separated solid

was filtered and dried at 60 °C to give desired product, 67 (25.5 g, 82%)

which was confirmed by spectral data and purity by HPLC was found to

be 92.81%.

3.6.8 Preparation of valacyclovir hydrochloride 49: Compound 67 (20

g) was taken in ethanol (300 mL) along with raney nickel (4 g) and

maintained at 50-60 °C for about 7 hours. After reaction was completed,

reaction mass was cooled to ambient temperature and unwanted solid

was separated by filtration. Filtrate pH was adjusted to 2.5-3.0 using

aqueous hydrochloric acid (10%, 35 mL). The homogeneous reaction was

treated with charcoal at 50-60 °C and then charcoal was separated by

hyflow filtration. The filtrate was concentrated under reduced pressure at

50°C. The obtained residue was treated with acetone (100 mL) and the

172

product, 49 (16.9 g, 82%) was separated by filtration followed by drying

at 50 °C and 49 is confirmed by spectral data.

3.6.9 Preparation of Valacyclovir (49)

(a) Preparation of Cbz-protected Valacyclovir (53): Cbz-L-Valine (52,

83.6 g, 0.332 mol) was dissolved in DMF (350 mL) and the solution was

cooled to -5 °C. A solution of DCC (68.6 g, 0.333 mol) in DMF (150 mL)

was added below 0 °C. After aging for 20 minutes, acyclovir (45, 50 g,

0.222 mol) and DMAP (4 g, 0.032 mol) were charged and reaction

mixture was stirred at -5-0 °C for about six hours. Dicyclohexylurea was

filtered, 80% of the solvent was removed by distillation and the remaining

solution was diluted with water (300 mL). Precipitated compound 53 was

filtered at ambient temperature and recrystallised from methanol (88.5 g,

87%). Purity by HPLC 99.3%, mp 157-159 °C; IR (KBr) cm-1 3311, 2931,

2855, 1726, 1630, 1606, 1536, 1390, 1183, 1128, 1101, 1041; 1H NMR

(CDCl3, 200 MHz) δ: 10.8 (br, 1H), 7.9 (br, 1H), 7.7 (s, 1H), 7.22 (m, 5H),

5.38 (s, 2H), 5.1 (s, 2H), 4.25 (m, 1Ha), 4.35 (m, 1Hb), 3.75 (d, 1H), 3.7

(m, 2H), 2.15 (m, 1H), 0.91 (d, 3H), 0.88 (d, 3H).

Purification of Cbz-protected Valacyclovir (53): Cbz Protected

valacyclovir (53, 25 g) having 3.1% of D-isomer, 70 was dissolved in a

mixture of acetone (300 mL) and water (75 mL) at reflux temperature and

cooled to ambient temperature. The mixture was stirred for few hours

173

after diluting with additional water (75 mL) and filtered to provide

compound 53 (21.2 g, 84.8%, 2.1% 70 by HPLC).

(b) Deprotection of Cbz group of 53: Cbz Protected valacyclovir (53, 5

g, 0.011 mol) and dry 5% Pd on alumina mixture (0.5 g) were taken in

DMF (50 mL) in a hydrogenator vessel. Hydrogen pressure of 4 kg/cm2

was applied at about 30 °C for the completion of the reaction. 70% of the

solvent was removed by distillation under vacuum below 80 °C and the

resultant concentrated solution was cooled to 10 °C. pH was adjusted to

3.0-4.0 using aqueous HCl at 10 °C, diluted with water (12.5 mL) and

catalyst was removed by filtration through celite at ambient temperature.

Filtrate was saturated with acetone (225 mL), precipitated valacyclovir

HCl (49) was filtered and dried under suction (4.7 g, 98.5% pure by

HPLC).

3.6.10 Purification of Valacyclovir HCl (49) to remove 71:

Valacyclovir HCl (49, 25 g) having 3.5% of 71 was dissolved in 25%

aqueous acetonitrile (250 mL) at 70 °C. The mixture was cooled to 30 °C

and diluted slowly with acetonitrile (75 mL) at 30 °C under stirring. The

separated solid was filtered and dried under vacuum to yield compound

49 (19.5 g, 75%, 2.6% 71 by HPLC).

3.6.11 Procedure for removal of heavy metal impurities: A

suspension of valacyclovir hydrochloride (49, 25 g) in water (50 mL) was

174

heated to 65 °C with simultaneous stirring. To the obtained solution, T-

63 resin (procured from Thermax Ltd.) was added and the mixture was

stirred for 20 minutes. The suspension was filtered through 0.45 m

filter paper and washed with water (25 mL). To the filtrate, acetone (500

mL) was added slowly. The obtained suspension was stirred for 1 h and

the precipitate was filtered and dried to yield compound 49 (24 g, 96%,

both Pd and Al content less than 10 ppm).

3.6.12 Purification of Valacyclovir HCl (49): Valacyclovir HCl (49, 50

g) wet material obtained from the above procedure was taken in DMF

(230 mL) and maintained for 7 h at 30 °C. Isopropanol (115 mL) was

added to the above mixture over a period of an hour and isolated product

was filtered at ambient temperature, washed with isopropanol and dried

under vacuum at 60 °C. Dried material was subjected to same

purification process to provide pure 49 which was further converted into

dihydrate (49, 20.7 g, 99.66% purity by HPLC) having individual

impurity below 0.05%, mp decomposed with foaming at 178 °C (lit.

Valacyclovir HCl monohydrate mp 150 °C, decomposes with foaming at

195 °C); IR (KBr) cm-1 3327, 3196, 2967, 1732, 1631, 1101, 1037; 1H

NMR (DMSO-d6, 200 MHz) δ: 10.9 (br, 1H), 8.5 (br, 3H), 7.83 (s,1H), 6.6

(s, 2H), 5.38 (s, 2H), 4.40 (m, 1Hb), 4.28 (m, 1Ha), 3.82 (d, 1H), 3.74 (m,

2H), 2.13 (m, 1H), 0.91 (d, 3H), 0.88 (d, 3H).

175

3.7 APPENDIX

Spectral data of active pharmaceutical ingredient, novel intermediates

and impurities are mentioned as follows.

Figure 3.8a: Mass Spectrum of Compound 49

Figure 3.8b:

1H NMR Spectrum (DMSO-d6, 400 MHz) of Compound 49

176

Figure 3.8c: D2O- NMR Spectrum (DMSO-d6, 400 MHz) of Compound 49

Figure 3.8d:

13C NMR Spectrum (DMSO-d6, 400 MHz) of Compound 49

177

Figure 3.9a: HRMS of Compound 68c

Figure 3.9b:

1H NMR spectrum (DMSO-d6, 400 MHz) of Compound 68c

178

Figure 3.9c:

13C NMR spectrum (DMSO-d6, 200 MHz) of Compound 68c

Figure 3.10a: HRMS of Compound 69a

179

Figure 3.10b:

1H NMR (DMSO-d6, 400 MHz) Spectrum of Compound 69a

Figure 3.10c:

13C NMR (DMSO-d6, 200 MHz) Spectrum of Compound 69a

180

Figure 3.11a: HRMS of Compound 69b

Figure 3.11b:

1H NMR Spectrum (DMSO-d6, 400 MHz) of Compound 69b

181

Figure 3.11c: 13

C NMR Spectrum (DMSO-d6, 200 MHz) of Compound 69b

Figure 3.12a: HRMS of Compound 69c

182

Figure 3.12b: 1H NMR Spectrum (DMSO-d6, 400 MHz) of Compound 69c

Figure 3.12c: 13

C NMR Spectrum (DMSO-d6, 400 MHz) of Compound 69c

183

Figure 3.13a: HRMS Spectrum of Compound 67

Figure 3.13b:

1H NMR Spectrum (DMSO-d6, 400 MHz)) of Compound 67

184

Figure 3.13c: 13

C NMR Spectrum (DMSO-d6, 200 MHz) of Compound 67