The Journal of Organic Chemistry
Featured Article
also investigated. Reactions were analyzed by HPLC. For spectroscopic
characterization, a 5 mL sample was collected under optimum
conditions, and the product was precipitated with ice−water. The
solid product was collected by filtration, washed with water, and
vacuum-dried to afford 5′,3′-bis(methanesulfonyl)-2.2′-anhydro-5-
methyluridine (4) as a white solid (0.97 g, 97% yield, mp 237.8−
238.3 °C). 1H NMR (400 MHz, DMSO-d6): δ 7.83 (s, 1H), 6.44 (d, J =
5.8, 1H), 5.65 (d, J = 5.7, 1H), 5.49 (s, 1H), 4.75−4.65 (m, 1H), 4.37−
4.28 (m, 1H), 4.20−4.08 (m, 1H), 3.44 (s, 3H), 3.15 (s, 3H), 1.81 (s,
3H). 13C{1H} NMR (100 MHz, DMSO): δ 171.7, 159.4, 132.3, 117.7,
90.4, 86.3, 82.3, 81.5, 68.1, 38.1, 37.3, 13.9. FTIR (cm−1) v: 3101.1,
3011.8, 2998.4, 2936.5, 1674.5, 1624.8, 1572.8, 1561.4, 1354.8, 1172.3,
996.6, 882.5.
Procedure 3: Preparation of Amberlite IRA 400-OBz.
Amberlite IRA 400-Cl ion-exchange resin (14−52 mesh) was treated
with 20% aqueous sodium benzoate at room temperature and stirred for
6 h. The resultant Amberlite IRA 400-OBz ion-exchange resin (loading
= 1.8 mmolg−1) was collected by filtration and washed with water
followed by methanol and oven-dried at 40 °C. The success was
confirmed by FTIR by the presence of Bz-carbonyl group on the resin.
The used ion-exchange resin was regenerated by repeating the above
process and reused. FTIR (cm−1) v: 3355.7, 1595.2, 1554.5, 1476.2,
1371.9, 888.7, 827.7, 720.8.
The exchange capacity of the Amberlite IRA 400-OBz resin was
determined by treating Amberlite IRA 400-OBz (0.2 g) with sodium
chloride (1 M, 30 mL) at room temperature and stirred for 2 h. The
resin was subsequently removed by filtration. The amount of sodium
benzoate in the filtrate was then titrated with HCl (0.01 M) using
methyl orange as indicator. The exchange capacity or loading of the
polymer supported nucleophile was determined to be 1.8 mmol g−1 of
OBz−.
Procedure 4: Synthesis of 5′-Benzoyl-3′-methanesulfonyl-
2.2′-anhydro-5-methyluridine (5) from Compound 4.12,44
Compound 4 (1 M) in DMF was pumped through a heated Amberlite
IRA 400-OBz (prepared in procedure 3) packed column reactor
(Omnifit EZ column 10 mm/100 mm, 3.6 g of resin = 4.7 mL bed
volume, 1.8 mmolg−1 exchange capacity) (Figure S3). The effect of
residence time and temperature was investigated for reaction
optimization. Reactions were analyzed by HPLC.
Procedure 5: Synthesis of 5′-benzoyl-3′-methanesulfonyl-
2.2′-anhydro-5-methyluridine (5) from Trimesylate 3.12,44
Trimesylate 3 (1 M, 1 equiv) in DMF was pumped through a heated
Amberlite IRA 400-OBz (prepared in section 1.4) packed column
reactor (Omnifit EZ column 10 mm/100 mm, 3.6 g of resin = 4.7 mL
bed volume, 1.8 mmol g−1 exchange capacity) held at 120 °C (Figure
S4). Reactions were analyzed by HPLC. For spectroscopic character-
ization, a 5 mL sample was collected under optimum conditions, and
the product was precipitated with ice−water. The solid product was
collected by filtration, washed with water, and oven-dried at 80 °C to
afford 5′-benzoyl-3′-methanesulfonyl-2.2′-anhydro-5-methyluridine
(5) as a white solid (2.07 g, 98% yield, mp 238−239.7 °C). 1H NMR
(400 MHz, DMSO-d6): δ 7.98−7.75 (m, 3H), 7.72−7.58 (m, 1H),
7.58−7.41 (m, 2H), 6.47 (d, J = 7.5, 1H), 5.79−5.53 (m, 2H), 4.80 (s,
1H), 4.43−4.11 (m, 2H), 3.46 (s, 3H), 1.76 (s, 3H). 13C{1H} NMR
(100 MHz, DMSO): δ 171.7, 165.6, 159.3, 134.1, 132.4, 129.7, 129.2,
117.8, 90.4, 86.3, 82.4, 81.6, 63.0, 38.0, 13.9. FTIR (cm−1) v: 3004.7,
2995.8, 2924.3, 1719.9, 1639.7, 1562.4, 1480.2, 1457.6, 1346.6, 1269.0,
1256.3, 1175.4, 1119.7, 1084.9, 1017.1, 978.1, 819.9, 713.4
Procedure 6: Synthesis of 5′-Benzoyl-3′α-methanesulfonyl-
2′α-bromothymidine (6).12,44 Compound 5 (1 M, 1 equiv) in DMF
was treated with either AcBr (1 equiv) in MeCN or HBr (33% wt % in
AcOH) (1 equiv) in MeCN in a Uniqsis 2 mL chip reactor fitted with a
Zaiput 5 bar back pressure regulator (Figure S5). The effect of
temperature, residence time and brominating agent concentration was
investigated for reaction optimization. Reactions were analyzed by
HPLC. For spectroscopic characterization, a 5 mL sample was collected
from a reaction in which AcBr was used and quenched with aqueous
NaOH. MeCN was removed in vacuo, and the product was precipitated
with ice−water. The solid product was collected by filtration and
washed with water and oven-dried at 50 °C to afford 5′-benzoyl-3′α-
CONCLUSION
■
We successfully synthesized stavudine (d4T, 1) by a continuous
flow process accomplishing six chemical transformations over
five continuous flow reactors from an affordable starting material
(5-methyluridine). Single step continuous flow synthesis was
demonstrated with an average of 97% yield, 21.4 g/h throughput
per step and a total of 15.5 min residence time over five
individual steps. We postulate that the total residence time of the
multistep flow system can be reduced in more flexible flow
systems where the individual reactor volumes can be altered to
match the single step conditions better. Furthermore, we
demonstrated an elegant multistep continuous flow synthesis of
1 in 87% total yield with a total residence time of 19.9 min and a
117 mg/h from a 0.1 M starting material without intermediate
purification and isolation. The total residence time is better than
the reported procedures (13.5−28 h).12,14 Unlike in the single-
step procedure, a slight excess of reagents was necessary in some
of the multistep continuous flow steps. However, inline workup
procedure could be incorporated in flow to neutralize excess
base before carrying further workup processes offline.
Continuous flow technology has an important role to play in
ensuring rapid and local production of important medicines on
demand to maintain the health and welfare of the society.
EXPERIMENTAL PROCEDURES
■
General Information. Chemicals were supplied by Sigma-Aldrich,
Merck and Industrial Analytical and used as received. Anhydrous
solvents were sourced from Sigma-Aldrich. Nuclear magnetic
resonance (NMR) spectra were recorded at room temperature as
solutions in deuterated dimethyl sulfoxide (DMSO-d6). A Bruker
Avance-400 spectrometer (400 MHz) was used to record the spectra,
and the chemical shifts are reported in parts per million (ppm) with
coupling constants in hertz (Hz). Infrared spectra were recorded from
4000 to 500 cm−1 using a Bruker spectrometer, and peaks (υmax) are
reported in wavenumbers (cm−1). High-performance liquid chroma-
tography (HPLC) data was obtained using Agilent 1100 with a UV
detector. HPLC analysis was performed on ACE Generix 5 C18(2)
column (150 mm × 4.6 mm i.d) at ambient temperature using an
isocratic system. The mobile phase consisted of 30% water and 70%
MeCN. The sample injection volume was 1 μL, eluted at a flow rate of 1
mL/min, and detected at 254 nm with a run time of 6 min.
Procedure 1: Synthesis of 2′,3′,5′-Tris(methanesulfonyl)-5-
methyluridine (3).12,44 5-Methyluridine (2) (0.1 M, 1 equiv) in
DMF (32 equiv) premixed with an appropriate amine base in DMF/
DCM or DMF/chloroform was treated with MsCl in DCM or
chloroform in a 1 mL PTFE coil reactor (0.8 mm i.d.) (Figure S1). The
PTFE reactor was sonicated where necessary to prevent reactor
clogging. The effects of bases, temperature, solvents, and residence time
were investigated for reaction optimization. Reactions were analyzed by
HPLC. For spectroscopic characterization, a 10 mL sample was
collected under optimum conditions and DCM or chloroform was
removed in vacuo. The product was precipitated with ice-diluted
aqueous NH4Cl. The solid product was collected by filtration and
washed with water and vacuum-dried to afford 2′,3′,5′-tris-
(methanesulfonyl)-5-methyluridine (3) as an off-white solid (0.24 g,
97% yield, mp 77.3−78.2 °C). 1H NMR (400 MHz, DMSO-d6) δ 11.55
(s, 1H), 7.58 (s, 1H), 5.97 (d, J = 4.5, 1H), 5.68−5.51 (m, 1H), 5.44−
5.24 (m, 1H), 4.56−4.43 (m, 2H), 3.38 (s, 3H), 3.36 (s, 3H), 3.26 (s,
3H), 1.80 (s, 3H). 13C{1H} NMR (100 MHz, DMSO) δ 164.2, 150.9,
136.9, 110.5, 88.4, 78.9, 76.7, 74.4, 38.4, 37.3, 12.5. FTIR (cm−1) v:
3027.3, 2939.6, 1689.0, 1470.1, 1348.6, 1334.7, 1272.2, 1171.5, 1068.8,
967.5, 946.7, 835.7, 583.7, 522.2.
Procedure 2: Synthesis of 5′,3′-bis(methanesulfonyl)-2.2′-
anhydro-5-methyluridine (4).12,14,44 Trimesylate 3 (1 M, 1 equiv)
in DMF was treated with DBU (1 equiv) in a Uniqsis 2 mL chip reactor
(Figure S2). The effect of temperature and residence time was
investigated for reaction optimization. The use of alternative bases was
G
J. Org. Chem. XXXX, XXX, XXX−XXX