Application of Continuous Flow in Tazobactam Synthesis

Article abstract of DOI:10.1021/acs.oprd.1c00127

Tazobactam is a β-lactamase inhibitor. In this work, a combination of continuous flow and batch experiments for the synthesis of tazobactam has been developed. The first three steps and the preparation of the peroxyacetic acid are continuously carried out in the microreactors, which improves the procedure safety and efficiency. There is also a final step of the deprotection reaction in the microreactor, which can increase the yield and reduce the formation of impurities. Under optimized process conditions, the total yield of the target product reached 37.09% (30.93% in batch). The continuous flow method not only greatly reduces the reaction time but also significantly improves procedure safety and increases the yield.

Full text of DOI:10.1021/acs.oprd.1c00127

pubs.acs.org/OPRD
Article
Application of Continuous Flow in Tazobactam Synthesis
Shuhao Zhou, Yunting Xin, Jiasheng Wang, Chengjun Wu,* and Tiemin Sun*
Cite This: Org. Process Res. Dev. 2021, 25, 1648−1657
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Tazobactam is a β-lactamase inhibitor. In this work, a combination of continuous flow and batch experiments for the
synthesis of tazobactam has been developed. The first three steps and the preparation of the peroxyacetic acid are continuously
carried out in the microreactors, which improves the procedure safety and efficiency. There is also a final step of the deprotection
reaction in the microreactor, which can increase the yield and reduce the formation of impurities. Under optimized process
conditions, the total yield of the target product reached 37.09% (30.93% in batch). The continuous flow method not only greatly
reduces the reaction time but also significantly improves procedure safety and increases the yield.
KEYWORDS: continuous flow, tazobactam, synthesis, industrial process
In recent years, the application of flow chemistry technology
to the field of drug synthesis and production has become the
focus of much attention.4 The flow reactors have unique
characteristics that can improve the mixing effect, provide
precise temperature control, and greatly shorten the cycle of
process screening and process amplification. Compared with
traditional batch methods, the flow reactors can not only
improve the reaction performance but also improve safety.5−7
Due to the extremely fine pore size and high heat transfer
efficiency of the microchannels, some dangerous reactions in
batch experiments can be carried out safely, such as the
nitration reaction, fluorination reaction, azide reaction,
oxidation and reduction reaction, etc.8−12
If flow chemistry could be used, current tazobactam
synthesis procedures will still have room for improvement.
For example, the first three reactions use the same solvent; if
the solids in the reaction procedure can be avoided, the
continuous flow could be easily realized and production
efficiency can be improved. In the process of producing 5 from
4 in the original literature,13 40% peroxyacetic acid was the
oxidant that had a high yield close to 95% and overoxidized
product 5′ was hardly tested, but the use of peroxyacetic acid
was very dangerous in the industry. Therefore, a variety of
alternative and safer oxidation systems have been devel-
oped.14−16 However, they have not achieved such a high
reaction yield as 40% peroxyacetic acid was used. Meanwhile,
the content of overoxidized products 5′ was much high so that
the purification was difficult; it would increase the correspond-
ing cost and the postprocedure would be more complicated. As
we all know, the safety of using hazardous reagents in
microreactors can be greatly improved, so peroxyacetic acid
1. INTRODUCTION
Tazobactam is a β-lactamase inhibitor. It was first marketed in
the United States in 1992 in combination with piperacillin.
Although it has little antibacterial activity itself, but it can
inhibit the bacterial β-lactamase activity to prevent bacteria
from hydrolyzing β-lactam antibiotics.1 Due to its low toxicity
and strong activity, a variety of combinations of tazobactam
and β-lactam antibiotics have been developed and applied in
clinics widely.2
In our previous work, an industrialized route of tazobactam
reported in the literature has been optimized. As shown in
Scheme 1, key intermediate 5 was originally synthesized using
a three-step process from 6-aminopenicilanic acid (6-APA) as a
starting material. Potassium bromide and sulfuric acid were
used to take the place of the corrosive and toxic hydrobromic
acid or bromine. Next, peroxyacetic acid and benzophenone
hydrazine were used for esterification in most literature
studies,3 but an overoxidation product was produced and a
higher concentration of peroxyacetic acid (40%) was needed to
increase the yield, which posed extreme safety hazards for
industrial production (Scheme 2). Thus, diphenylmethanol
was used for esterification and hydrogen peroxide for the
oxidation reaction in our scheme, which increased the yield
and improved production safety. Next, 9 was produced by 5
through three steps that were similar to the literature studies.
Then, 1,2,3-triazole was introduced by an anion resin that can
be recycled; it can avoid using azide and acetylene and has
greater value in terms of economy and environmental
protection. Finally, the target product tazobactam was obtained
by traditional methods.
Each step has been further optimized in the batch
experiments with our efforts, and the total yield has been
increased from about 20% reported in the literature to 30.93%;
the production process can be carried out stably including the
yield, purity, safety, economy, and environmental protection,
and other issues have been optimized to the maximum. We
have scaled up the production process to industrial production
in factories.
Received: April 14, 2021
Published: June 17, 2021
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657
© 2021 American Chemical Society
1648

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Scheme 1. Initial Synthesis Route to Tazobactam
Scheme 2. Synthesis of Intermediate 5 and Generation of Impurity
Figure 1. Semicontinuous flow process for the synthesis of intermediate 2.
was chosen in our scheme using continuous flow. Here, we
report the synthesis method of tazobactam by combining
continuous flow and batch experiments on the route used in
our industrial production.
batch experiment, but it will cause the precipitation of
potassium bromide, which will block the pipeline in the
microreactor. When ethanol was not added, there was no
potassium bromide solid precipitated, but intermediate 2 was
not soluble in this aqueous solution. It adheres to the pipe wall
as oil and results in low yield. When adhesion products in the
pipeline gradually increase, they eventually result in blockage
of the pipeline. Therefore, it is necessary to add a solvent to
2. RESULTS AND DISCUSSION
2.1. Semicontinuous Method of Step 1. The first step
of the reaction uses ethanol to solubilize intermediate 2 in the
1649
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Table 1. Investigation into Reaction Conditions of Step 1
entry total flow rate (mL/min) residence time (min) increased volume of water (mL) volume of dichloromethane (mL) temperature (°C) yield (%)
1
2
3
4
5
6
7
8
6
5
4
3
8
5
6
7.5
10
3.75
3
2.5
2.14
1.875
1.67
2
2
2
2
2
2
2
0
100
100
100
100
100
100
100
100
100
100
80
100
100
100
100
100
100
0
50
50
50
50
50
50
50
50
50
50
50
80
100
120
100
100
100
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
−5
5
10
30.2
46.2
40.8
39.2
62.1
68.7
71.0
75.3
73.0
68.6
77.2
70.2
80.1
85.5
82.8
73.5
88.2
68.0
10
12
14
16
18
15
15
15
15
15
15
15
15
9
10
11
12
13
14
15
16
17
18
2
Figure 2. Semicontinuous flow process for the synthesis of 4.
1650
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
dissolve intermediate 2 without causing the precipitation of
potassium bromide. Dichloromethane was chosen as the
solvent in our scheme to perform the two-phase reaction in
sulfuric acid so that intermediate 2 was soluble in the organic
phase. More importantly, dichloromethane can be used as a
solvent for the next reaction. It can be used continuously
without changing the solvent. The device was designed as
shown in Figure 1, including a feed system (three plunger
metering pumps), two T-shaped mixers (M1 and M2), two
microreactors (R1 and R2), a backpressure regulator (BPR),
and an external high- and low-temperature-integrated machine
connected to control different reaction temperatures. The role
of R1 was to mix and precool the reaction solution.
As shown in Figure 1, there were three streams in total. Two
plunger pumps were used to separately introduce solution S1
[6-APA (21.6 g, 0.1 mol) and potassium bromide (60.1 g, 0.5
mol) in 1.25 M H2SO4 (250 mL)] and solution S2 (CH2Cl2,
100 mL) into M1 for mixing and precooling in microreactor
R1 at 0 °C. The solution flow rates were 3.9 mL/min and 1.38
mL/min, respectively. The combined solution was mixed with
solution S3 (sodium nitrite aqueous solution, 3 mol/L), which
was introduced into M2 with a flow rate of 0.75 mL/min. The
resulting solution passed through microreactor R2 (the
residence time is 5 min; the temperature is 0 °C). After
running for 6 min, the system reached a steady state. The
output of the reactor was continuously collected in a flask for
10 min, and then, the liquid was separated. Finally,
intermediate 2 was obtained after the organic layer was dried
and the solvent was evaporated in vacuo. The purity and yield
were tested by high-performance liquid chromatography
(HPLC). The preliminary experiment yield was only 30%,
and then, the specific parameters of the reaction were
optimized in detail.
To find suitable reaction conditions, the effects of reaction
parameters were systematically studied (Table 1). Considering
that the total volume of the aqueous phase was reduced due to
the replacement of ethanol with dichloromethane, this would
inevitably lead to pH values of the aqueous phase being
changed, which in turn would result in the lower yield of the
reaction. Therefore, the appropriate amount of water was
needed for the reaction solution to maintain a proper pH
value. Water was added to the sodium nitrite solution, which
can increase the volume of the sodium nitrite solution and
improve the accuracy of the flow rate. It was found that adding
100 mL of water was more beneficial to the reaction after
experiments. The amount of dichloromethane was also
investigated. Although dichloromethane did not participate in
the reaction, but its amount would affect the total flow rate and
the mixing effect in the microreactors. We found that it was
more appropriate when dichloromethane is 100 mL. In the
investigation of residence time, when the total flow rate was 15
mL/min and the residence time was 2 min, the reaction yield
can be greatly improved. Finally, the reaction temperature was
investigated and 5 °C was selected as the optimum reaction
temperature.
time, improving efficiency, and reducing the consumption of
solvents.
2.2. Semicontinuous Method of Step 2. The initial plan
was to combine intermediate 2 with 1-(3-dimethylaminoprop-
yl)-3-ethylcarbodiimide hydrochloride (EDC), diphenylmetha-
nol, and 4-dimethylaminopyridine (DMAP) separately. They
were fed into the microreactors using four plunger metering
pumps directly, as shown in Figure 2a. After experiments,
intermediate 2 was not completely transformed regardless of
changes in the concentration, residence time, reaction
temperature, or parameters such as the flow rate and pressure.
Afterward, the order of feeding was changed; intermediate 2
was mixed with EDC and DMAP first and then the confluent
reaction occurred with diphenylmethanol, but the results were
still not as expected, as shown in Figure 2b. Finally, we
designed the process illustrated in Figure 2c. After changing
the order of reactants again, the reaction results were much
improved. Then, this model as the basis for further
experimental optimization was used.
The device designed includes a feeder system (three plunger
metering pumps), two T-shaped mixers (M1 and M2), two
microreactors (R1 and R2), and a backpressure regulator
(BPR), as shown in Figure 2c.
There were three streams in total too. Two plunger pumps
were used to separately introduce solution S4 [intermediate 2
(19.58 g, 0.07 mol) and EDC (13.4 g, 0.07 mol) in CH2Cl2
(300 mL)] and solution S5 [diphenylmethanol (12.88 g, 0.07
mol) in CH2Cl2 (100 mL)] into M1 and R1 for mixing. The
solution flow rates were 6 and 2 mL/min, respectively. The
combined solution was mixed with solution S5 [DMAP (0.4 g,
0.0032 mol) in CH2Cl2 (100 mL)], which was introduced into
M2 at a flow rate of 2 mL/min. Also, the resulting solution
passed through microreactor R2 (the residence time is 2 min;
the temperature is 25 °C). After running for 6 min, the system
reached a steady state. The output of the reactor was
continuously collected in a flask for 10 min, and then, the
collected mixture was added in 100 mL of water and separated.
Finally, intermediate 4 was obtained after the organic layer was
dried and the solvent was evaporated in vacuo. The purity and
yield were tested by HPLC. It was exciting to obtain excellent
results with a yield of 92.4% by the semicontinuous method.
To find suitable reaction conditions, the effects of residence
time and reaction temperature were systematically studied
(Table 2). When the residence time was shortened to 1 min,
Table 2. Investigation into Reaction Conditions of Step 2
residence time
(min)
temperature
pressure
(MPa)
yield
(%)
entry
(°C)
1
2
3
4
5
6
7
8
3
2
1
0.5
1
1
1
1
1
1
1
1
2
25
25
25
25
40
40
50
60
10
0
0
0
0
88.5
94.0
98.2
90.7
74.2
72.6
61.0
42.3
90.5
84.1
80.4
65.0
74.2
76.6
0
0.2
0.4
0.4
0
0
0
0
0
0
After a series of optimizations of the reaction conditions, the
reaction yield in this step reached 88.2%, but the yield in
conventional batch experiments could reach 90%. So, no better
results were achieved in the microreactors than batch
experiments. Although the yield was slightly reduced, it was
still very valuable for research. More importantly, the reaction
can continuously flow with the subsequent reactions, saving
9
10
11
12
13
14
0
−10
−20
−20
−20
3
1651
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 3. Semicontinuous flow process for the synthesis of 5.
the yield increased to 98%. The shortened time was more
favorable to improving production efficiency. In the inves-
tigation of the reaction temperature, when the reaction
temperature increased up to 60 °C, the yield decreased
sharply because higher temperatures may have some impacts
on the stability of the reactants. The decrease of yield was
found when the temperature decreased to −20 °C, which
indicated that the lower temperature was not conducive to the
rapid progress of the reaction so that the residence time was
extended in an attempt to increase the yield at this temperature
but the result did not reach our expectations. When the
temperature was between 10 and 30 °C, the reaction yield was
relatively constant. Room temperature was chosen as the
reaction temperature without heating or cooling, which is
green and environmentally friendly.
2.3. Semicontinuous Method of Step 3. Peroxyacetic
acid was initially selected as the oxidant in the synthesis of
tazobactam reported in the literature. There were numerous
advantages such as high yield, few impurities, and high
efficiency. However, because of its accidents in the industry,
the use of 40% peroxyacetic acid was strictly restricted and a
series of safeguards were required. After that, a variety of safer
oxidation systems have been developed but none of them was
better than peroxyacetic acid in efficiency and yield. Because of
the small inner diameter of the pipe, larger contact area, high
heat transfer efficiency, and continuous flow characteristics, the
safety of peroxyacetic acid can be greatly improved. Here,
peroxyacetic acid was chosen as the oxidant in the micro-
reactor.
Table 3. Investigation into Reaction Conditions of Step 3
residence time
(min)
temperature
peroxyacetic acid
equivalent
yield
(%)
entry
(°C)
1
2
3
4
5
6
7
8
9
3
2
1
0.5
1
1
1
1
1
2
2
2
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.0
1.2
1.3
87.0
90.2
93.1
85.4
90.5
94.5
74.8
84.6
96.8
96.0
2
10
25
40
25
25
25
10
1
time selected was 1 min, and 1.2 equiv of peroxyacetic acid
would be better.
2.4. Semicontinuous Method of Preparing Peroxy-
acetic Acid. The danger of peroxyacetic acid was reduced in
the microreactor, but its preparation, storage, and trans-
portation had the same danger. If peroxyacetic acid could be
prepared and used in situ in the microreactor, the above-
mentioned dangerous problems could be greatly reduced,
which was similar to the previous device, as shown in Figure 4.
As shown in Figure 3, two plunger pumps were used to
separately introduce solution S7 [intermediate 4 (28.2 g) in
CH2Cl2 (300 mL)] and solution S8 (20% peroxyacetic acid)
into a T-shaped mixer (M) at the flow rates of 18.3 mL/ min
and 1.7 mL/min, respectively. Then, the solution passed
through microreactor R (the residence time was 1 min; the
reaction temperature was 25 °C). After running for 6 min, the
system reached a steady state. The output of the reactor was
continuously collected in a flask for 10 min, and then, the
collected mixture was washed with saturated sodium bisulfite
and water. Finally, intermediate 5 was obtained after the
organic layer was dried and the solvent was evaporated in
vacuo. The purity and yield were tested by HPLC. It was
exciting to get better yields than in batch experiments. Then,
some reaction conditions were investigated including the
reaction temperature and residence time, as shown in Table 3.
The results showed that the reaction temperature hardly
affected the yield so that the same temperature as the previous
step was chosen, which was 25 °C too; no additional
temperature adjustment was required, which was convenient
for continuous operation and saving energy. The residence
Figure 4. Semicontinuous flow process for the synthesis of
peroxyacetic acid.
Two plunger pumps were used to introduce solution S9 (50%
H2O2) and solution S10 (acetic acid with catalytic amounts of
H2SO4 and EDTA) into a T-shaped mixer (M) and then
passed through microreactor R. After investigation of reaction
conditions, 20% content of peroxyacetic acid was obtained
when the residence time was 10 min and reaction temperature
was 55 °C, as shown in Table 4 (peroxyacetic acid
concentration measured by the titration method). Now, we
developed a continuous process of preparation of 20%
peroxyacetic acid, which can be carried out simultaneously
with the oxidation reaction and continuously flow into the
microreactors. The preparation and use of 20% peroxyacetic
acid all took place in microreactors, which can avoid the
explosion risk that may occur in the preparation, storage, and
transportation in batch experiments. What is more, this
1652
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
into microreactor R4 to prepare peroxyacetic acid in situ.
Peroxyacetic acid later directly flowed into mixer M6, and it
was mixed with the reaction solution flowing out of
microreactor R3 to undergo the oxidation reaction in
microreactor R6. Finally, it was derived through the back-
pressure regulator (BPR). Then, the reaction solution was in
turn washed with the sodium bisulfite solution, sodium
bicarbonate solution, and saturated sodium chloride solution.
Finally, intermediate 5 was obtained after drying and distilling
off the solvent.
Table 4. Investigation into Reaction Conditions for
Preparing Peroxyacetic Acid
temperature
residence time
(min)
peroxyacetic acid
content (%)
entry
(°C)
EDTA
1
2
3
4
5
6
7
8
9
25
25
25
25
35
45
55
55
65
5
8
−
−
−
−
−
−
−
+
8.4
10.8
16.5
15.0
18.7
19
18.5
20.6
20.0
10
13.3
10
10
10
10
10
Since the reaction conditions of the relevant steps have been
investigated separately in early experiments, here, we only
optimized the adaptation conditions when they are jointly
performed. The conditions for the first step were the same as
the optimal conditions in the semicontinuous process. It
should be noted that the concentration of intermediate 2 was
0.84 mol/L after the first step, while the required
concentration of intermediate 2 was 0.23 mol/L in the second
step. To achieve synergy, we tried to decrease the
concentration of intermediate 2 in the first step or increase
the concentration of other reactants in the second and third
steps to match the high concentration of intermediate 2, but
no better results were obtained after many attempts. So, we
decided to add some CH2Cl2 into the solution of intermediate
2 after the first step to maintain the concentration at 0.23 mol/
L, which can be better carried out in the subsequent reactions.
About 265 mL of CH2Cl2 was added per 100 mL of solution of
intermediate 2 from the first step after calculations. It should
increase the flow rates in the second and third steps because of
the larger volume of the intermediate 2 solution so that it can
avoid the accumulation of intermediate 2. Thus, we increased
the related flow rates in the second and third steps and made
them suitable and sustainable. Considering that the reaction
temperature was 55 °C during the preparation of 20%
peroxyacetic acid in R4, the oxidation reaction temperature
of step 3 was 25 °C in R6. We set a reactor device behind the
microreactor R4 to precool the solution to prevent the
+
method can not only reduce the production risk but also
improve efficiency greatly.
2.5. Fully Continuous Flow Process of These Four
Reactions. We tried to develop a fully continuous flow
process to perform these four steps. As shown in Figure 5, the
four-step reactions were connected in sequence according to
the semicontinuous method obtained in the previous experi-
ments. First, three plunger pumps were separately used to
introduce solutions S1, S2, and S3 through two T-shaped
mixers (M1 and M2) and flowed into the microreactor (R2)
for reaction. The reaction temperature was 5 °C, and the
residence time was 2 min. It flowed into the separatory funnel
through the pressure regulator (BPR). Now, the solution of
intermediate 2 was obtained after separating the dichloro-
methane layer and washing it with saturated sodium chloride.
Next, three plunger pumps were utilized to introduce solutions
S4, S5, and S6 separately. They all flowed into microreactor
R3. The reaction temperature was 25 °C, and the residence
time was 1 min. Thus, the solution of intermediate 3 was
obtained. The intermediate 3 solution was directly flowed into
microreactor R6 without being separated. At the same time,
two plunger pumps were used to introduce solutions S7 and S8
Figure 5. Fully continuous flow process of these four reactions.
1653
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 6. Semicontinuous Flow Process for the Synthesis of 1.
peroxyacetic acid from bringing more heat into microreactor
R6. Now, all reaction conditions and reactant concentrations
have been specified, as shown in Figure 5. Excitingly, we got
intermediate 5 with higher yield and purity than batch
experiments. What is more, it was continuously suitable and
efficient for industrial production without the amplification
effect.
3. CONCLUSIONS
A stable and effective synthetic method for the preparation of
tazobactam in batch experiments combined with a continuous
flow method has been developed. Compared with all
traditional batch experiments, it has improved procedure
safety and efficiency and cut back side reactions. The total
yield was 37.09%, and the purity was 99.85%. The use of the
continuous flow strategy provides a good solution for the
synthesis of tazobactam, and it has been applied to industrial
production.
2.6. Semicontinuous Method of Step 9. The reaction
temperature was 55 °C and the reaction time was 4 h in batch
experiments in this step. However, the temperature and
reaction time were detrimental to the stability of the β-lactam
ring. High temperatures and long reaction times may cause the
β-lactam ring to be destroyed. As a result, the yield and purity
were lower, which will affect the final product quality.
Therefore, we hope to be able to reduce the adverse effects
and increase the yield and purity by flow chemistry. As
illustrated in Figure 6, a plunger pump is used to introduce the
m-cresol solution of intermediate 12 into the microreactor (we
have experimentally verified that intermediate 12 hardly reacts
in m-cresol solution at room temperature). The conditions
such as the flow rate, residence time, and reaction temperature
were systematically studied, as provided in Table 5. When the
flow rate was 20 mL/min, the residence time was 1 min, and
the reaction temperature was 75 °C, better yield and purity
were obtained.
4. EXPERIMENTAL SECTION
1
4.1. General Information. In this paper, H NMR and
13C NMR spectra were measured using a Bruker 600 MHz
AVIII HD nuclear magnetic resonance spectrometer. TMS was
the internal standard. The mass spectra of the compounds
were all measured with an Agilent 1000 four-link liquid-mass
spectrometer. High-performance liquid chromatography was
measured with an Agilent 1100 high-performance liquid
chromatograph. The infrared spectrum was measured with a
Bruker IFS 55. The optical rotation was measured with an
Agilent fully automatic polarimeter. Continuous flow instru-
ments include microreactors (Himile Mechanical Science and
Technology (Shandong) Co., Ltd.), pumps (Shanghai Sanotac
Scientific Instruments Co., Ltd.), and a high- and low-
temperature circulating device (Gongyi Yuhua Instrument
Co., Ltd.). Other reaction instruments include an FC204
analytical balance (Shanghai Precision Balance Instrument
Factory), a DF-101S collector-type constant temperature
heating magnetic stirrer (Gongyi Yuhua Instrument Co.,
Ltd.), and an EYELA N-1001 rotary evaporator (EYELA,
Germany). The reagents used in the experiments were all
commercially available chemicals or of analytical grade.
4.2. 6α-Bromopenicillin-3-α-carboxylic acid (2). 6-APA
(21.6 g) and potassium bromide (60.1 g) were added to 1.25
mol/L sulfuric acid (250 mL) to dissolve them at 0 °C, as
solution S1. Dichloromethane (100 mL) was S2. Sodium
nitrite (10.6 g) was dissolved in 250 mL of water as solution
S3. Three plunger pumps were used to introduce the three
solutions, respectively, into microreactor R2 after passing
through two T-shaped mixers M1 and M2 and a microreactor
R1 to precool; the flow rates were 6.25, 2.5, and 6.25 mL/min,
respectively. The reaction temperature was 5 °C, and the
residence time was 2 min. After the reaction, the solution was
collected at the discharge port (placed in an ice bath to control
temperature), the liquid was separated, the organic phase was
washed with saturated brine, and then dichloromethane was
added to the solution to prepare for the next reaction; 265 mL
of dichloromethane was added per 100 mL of solution. ESI-
MS: 279.4 (M − H+).
Table 5. Investigation into Reaction Conditions of Step 9
flow rate
(mL/min)
residence time
(min)
temperature
yield
(%)
entry
(°C)
1
2
3
4
5
6
7
12
14
16
18
20
22
20
1.54
1.43
1.25
1.10
1.00
0.90
1.00
75
75
75
75
75
75
85
82.6
84.6
83.8
84.0
85.6
81.0
78.3
An improved synthetic route to tazobactam is shown in
Scheme 3, and a comparison of process mass intensity (PMI)
breakdown for initial and improved synthetic routes is shown
in Figure 7. Compared to the initial synthetic route, application
of the continual flow method is safer and efficient; it proceeds
with a higher overall yield (36% increase) and a lower PMI
(7% reduction).
However, because only four steps in the whole route were
converted to continuous flow, the advantages of continuous
flow cannot be fully reflected in the PMI and total yield.
Therefore, we have compared the continuous flow method and
batch method of the first three steps in productivity and
space−time yield (STY) (Table 6). Thanks to these improve-
ments, the productivity and STY of the whole route were
estimated to increase by 20% and 80%, respectively, on a
manufacturing scale.
4.3. 6α-Bromopenicillin-3α-carboxylic Acid Sulfoxide
Diphenylmethyl Ester (5). EDC (13.4 g) was added to the
total volume of 300 mL solution of the previous step, and it
was introduced into microreactor R3 by a pump; the flow rate
was 24 mL/min. Diphenylmethanol (12.90 g) was dissolved in
dichloromethane (100 mL) and introduced into microreactor
1654
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Scheme 3. Improved Synthetic Route to Tazobactam
Figure 7. Comparison of PMI breakdown for initial and improved synthetic routes.
R3 by a pump at a flow rate of 8 mL/min. DMAP (0.4 g) was
dissolved in dichloromethane (100 mL) and introduced into
microreactor R3 by a pump at a flow rate of 8 mL/min. The
residence time was 1 min, and the reaction temperature was 25
°C. After the reaction, the solution was directly inputted into
microreactor R6 without separation. At the same time, 50%
hydrogen peroxide was introduced into microreactor R4 by a
pump at a flow rate of 1.85 mL/min. Acetic acid that was
added with sulfuric acid and EDTA was introduced into
microreactor R4 by a pump at a flow rate of 1.85 mL/min. The
residence time was 10 min. The reaction temperature was 55
Table 6. Comparison of the First Three Steps in Batch and
Continuous Flow Experiments
batch experiment
74.6%
continual flow experiment
attribute yield (%)
PMI
productivity
(mol/day)
STY (kg/(dm3 d ay)
83.8%
96.64
4
77.1
8
0.07
19
1655
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
°C. Finally, the reaction produced 20% peroxyacetic acid and
then it flowed into microreactor R6 and underwent an
oxidation reaction with intermediate 3 inputted by the
previous pump. The residence time was 1 min, and the
reaction temperature was 25 °C. The reaction solution was
received at the discharge port and washed with saturated
sodium bisulfite, saturated sodium bicarbonate, water, and
saturated sodium chloride. A yellow oil (30.3 g) with a yield of
83.8% (based on 6-APA) was obtained after the organic layer
was dried and evaporated in vacuo. ESI-MS: 485.3 (M + Na+).
HPLC analysis: 98.63%.
Yield: 61.59% (based on intermediate 6), HPLC analysis:
98.25%.
4.8. 2β-(1,2,3-triazol-1-yl) Methyl-2α-methyl-6,6-dihy-
dropenicillin-3α-carboxylic Acid Diphenylmethyl 1,1-
Dioxide (12). Intermediate 11 (36 g), dichloromethane (216
mL), and acetic acid (63 g) were added to a flask and cooled to
10 °C; then, potassium permanganate (28.8 g) was added, and
the reaction was kept at 25 °C for 6 h. After the reaction
completed, water (80 g) was added and the temperature was
kept between 5 and 10 °C. Hydrogen peroxide (50%) was
added dropwise at a controlled temperature of 5−10 °C until
the reaction solution turned white and stirred for 1 h.
Dichloromethane was evaporated in vacuo, then ethyl acetate
(50 mL) was added and stirred for 1 h at 0−5 °C, and
intermediate 12 was obtained after filtration. Yield: 92%, ESI-
MS: 465.22 (M − H+), 467.23 (M + H+)
4.9. Tazobactam (1). Intermediate 12 (30 g) was
dissolved in m-cresol (200 mL) at 25 °C. It was pumped
into the microreactor at a flow rate of 20 mL/min, the reaction
temperature was 75 °C, and the retention time was 1 min. The
reaction solution was received at the discharge port; ethyl
acetate (660 mL) was added and extracted twice with saturated
sodium bicarbonate solution (200 mL); and washed the water
layer using methyl isobutyl ketone (30 mL), ethyl acetate (30
mL), dichloromethane (30 mL). Activated carbon (3 g) was
added to the water layer to decolorize and then filtered. After
filtration, the filtrate was cooled to 0−5 °C and 15%
hydrochloric acid was added to adjust the pH value to 1.0−
1.5 slowly, it was allowed to stand for 3 h and then filtered.
Finally, white solid tazobactam (16.53 g) was obtained after
drying. Yield: 85.6%. m.p. 197−200 °C. ESI-MS: 301.12 (M +
H+). IR (KBr), v/cm−1: 3431.7 (−OH), 1799.7 (−CO),
1713.5 (−CO), 1328.5 (as, −SO2), 1140.9 (as, −SO2),
790.6 (−CO). 1H NMR (600 MHz, DMSO-d6) δ = 8.09 (s,
1H), 7.79 (s, 1H), 5.24 (d, J = 15.3, 1H), 5.18 (d, J = 4.1, 1H),
4.93 (d, J = 15.3 Hz, 1H), 4.79 (s, 1H), 3.71 (dd, J = 16.5, 4.4
Hz, 1H), 3.33 (d, J = 16.4, 1H), 1.35 (s, 3H). 13C NMR (151
MHz, DMSO-d6) δ = 171.3, 167.8, 133.3, 126.8, 64.5, 61.9,
59.8, 49.8, 37.8, 15.7. HPLC analysis: 99.85%.
4.4. 6,6-Dihydropenicillin-3α-carboxylic Acid Sulfox-
ide Diphenylmethyl Ester (6). Intermediate 5 (32 g) and
ammonium acetate solution (NH4OAc, 21.696 g; H2O, 173.44
mL) were added to tetrahydrofuran (170 mL). Zinc powder
(14.72 g) was added in batches, and the raw materials
completely disappeared after 1 h of reaction at room
temperature. Tetrahydrofuran was distilled off, and ethyl
acetate (200 mL) was added for extraction. The extract was
washed with water to near-neutral and then with saturated
sodium chloride solution (100 mL) and dried over anhydrous
sodium sulfate. After filtration, the solvent was distilled off
from the filtrate, cyclohexane−ethyl acetate (8: 1, 90 mL) was
added at room temperature, and the mixture was stirred for 1.5
h. The white solid was filtered to obtain the product (24.2 g)
1
with a yield of 91.2%. ESI-MS: 382.18 (M − H+). H NMR
(600 MHz, CDCl3): δ 7.38−7.30 (m, 10H), 6.99 (s, 1H), 4.92
(dd, J = 3.8, 2.8 Hz, 1H), 4.64 (s, 1H), 3.35 (dd, J = 3.3, 1.4
Hz, 2H), 1.69 (s, 3H), 0.94 (s, 3H). HPLC analysis: 98.99%.
4.5. 3-Methyl-[2-oxo-4-(2-benzothiazole dithio)-1-
azacyclobutyl]-3-butenoic acid (8). Intermediate 6 (27
g) and 2-mercaptobenzothiazole (11.78 g) were added to
toluene (450 mL) and refluxed for 80 min. Water was removed
with a water separator. The solvent was distilled in vacuo at
65−70 °C, and the next reaction was directly performed. ESI-
MS: 555.4 (M + Na+).
4.6. 2β-Bromomethyl-2α-methylpenicillin-3α-carbox-
ylic Acid Diphenylmethyl Ester (9). All of the products
from the previous step were added to dichloromethane (360
mL) and cooled to −15 to −20 °C; then, copper bromide (18
g) was added and the reaction was maintained at −15 to −20
°C for 8 h. Diatomaceous earth (4 g) was added, stirred for 15
min, and filtered. The filtrate was washed twice with 1%
sodium bicarbonate solution (200 mL) and then washed twice
with water (200 mL) and saturated sodium chloride solution
(100 mL). Then, it was dried over anhydrous sodium sulfate.
After filtration, the solvent was distilled off and the next
reaction was directly carried out.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.oprd.1c00127.
Mass spectrum of 6α-bromopenicillin-3-α-carboxylic
acid (2); mass spectrum and HPLC spectrum of 6α-
bromopenicillin-3α-carboxylic acid sulfoxide diphenyl-
1
methyl ester (5); mass spectrum, H NMR spectrum
4.7. 2β-(1,2,3-Triazol-1-yl) methyl-2α-methyl-6,6-di-
hydropenicillin-3α-carboxylic Acid Diphenylmethyl
Ester (11). Acetone (300 mL) was added to the product in
the previous step, and then, 1,2,3-triazole (112.5 g) and water
(40 g) were added. The temperature was kept at −15 °C, resin
sulfonic acid (47.5 g) was added, and then the reaction was
maintained at −15 °C for 20 h. When the reaction completed,
the organic phase was filtered and evaporated in vacuo at 40−
50 °C. Then, dichloromethane (200 mL) was added and
separated, and the organic layer was washed with water (200
mL) and saturated sodium chloride (50 mL). Intermediate 11
was obtained after the organic layer was dried and evaporated
in vacuo. Then, the crude product was recrystallized from an
ethyl acetate−n-hexane solution to give pure intermediate 11.
(600 MHz, CDCl3), and HPLC spectrum of 6,6-
dihydropenicillin-3α-carboxylic acid sulfoxide diphenyl-
methyl ester (6); mass spectrum of 3-methyl-[2-oxo-4-
(2-benzothiazole dithio)-1-azacyclobutyl]-3-butenoic
acid (8); HPLC spectrum of 2β-(1,2,3-triazol-1-yl)
methyl-2α-methyl-6,6-dihydropenicillin-3α-carboxylic
acid diphenylmethyl ester (11); mass spectrum of 2β-
(1,2,3-triazol-1-yl) methyl-2α-methyl-6,6-dihydropeni-
cillin-3α-carboxylic acid diphenylmethyl 1,1-dioxide
1
(12); mass spectrum, H NMR spectrum (600 MHz,
DMSO-d6), 13C NMR spectrum (151 MHz, DMSO-d6),
HPLC spectrum, and infrared spectrum of tazobactam
(1); and calculations of productivity, space−time yield
(STY), and process mass intensity (PMI) (PDF)
1656
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657

Organic Process Research & Development
pubs.acs.org/OPRD
Article
and 3β-Azido-3α-methylcepham-1, 1-Dioxides. Synthesis 1986, 1986,
292−296.
(14) Deshpande, P. B.; Udayampalayam, P. S.; Karale, S. N.;
Kannappan, J.; Gnanaprakasam, A. A Process for the Preparation of
Benzyl-2-oxo-4-(heteroaryl) dithio-alpha-isoprenyl-1-azetidineazetate
Derivatives. WO2004039776A22004.
(15) Xu, W.; Li, Y.; Zhang, Q.; Zhu, H. A new approach to the
synthesis of tazobactam using an organosilver compound. Synthesis
2005, 2005, 442−446.
(16) Buynak, J.; Chen, H. Inhibitors of Serine and Metallo-β-
Lactamases. WO2003087105A12003.
AUTHOR INFORMATION
■
Corresponding Authors
Chengjun Wu − Key Laboratory of Structure-Based Drug
Design and Discovery, Shenyang Pharmaceutical University,
Ministry of Education, Shenyang 110016, P. R. China;
orcid.org/0000-0001-5785-8065; Email: chengjunspu@
163.com
Tiemin Sun − Key Laboratory of Structure-Based Drug Design
and Discovery, Shenyang Pharmaceutical University, Ministry
of Education, Shenyang 110016, P. R. China; orcid.org/
0000-0002-9981-0243; Email: suntiemin@126.com
Authors
Shuhao Zhou − Key Laboratory of Structure-Based Drug
Design and Discovery, Shenyang Pharmaceutical University,
Ministry of Education, Shenyang 110016, P. R. China
Yunting Xin − Key Laboratory of Structure-Based Drug
Design and Discovery, Shenyang Pharmaceutical University,
Ministry of Education, Shenyang 110016, P. R. China
Jiasheng Wang − Key Laboratory of Structure-Based Drug
Design and Discovery, Shenyang Pharmaceutical University,
Ministry of Education, Shenyang 110016, P. R. China
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.oprd.1c00127
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Kadima, T. A.; Weiner, J. H. Mechanism of suppression of
piperacillin resistance in enterobacteria by tazobactam. Antimicrob.
Agents Chemother. 1997, 41, 2177−2183.
(2) Perry, C. M.; Markham, A. Piperacillin/Tazobactam an updated
review of its use in the treatment of bacterial infections. Drugs 1999,
57, 805−809.
(3) Zhou, Y.; Wu, C.; Ma, H.; Chen, J.; Sun, T. Precise Preparation
of a High-Purity Key Intermediate of Tazobactam. Org. Process Res.
Dev. 2020, 24, 2898−2905.
(4) Mason, B. P.; Price, K. E.; Steinbacher, J. L.; Bogdan Andrew, R.;
Mcquade, D. T. Greener Approaches to Organic Synthesis Using
Microreactor Technology. Chem. Rev., 2007, 107, 2300−2318.
(5) Malet-Sanz, L.; Susanne, F. Continuous Flow Synthesis. A
Pharma Perspective. J. Med. Chem. 2012, 55, 4062−4098.
(6) Porta, R.; Benaglia, M.; Puglisi, A. Flow Chemistry: Recent
Developments in the Synthesis of Pharmaceutical Products. Org.
Process Res. Dev. 2016, 20, 2−25.
(7) Pastre, J. C.; Browne, D. L.; Ley, S. V. Flow chemistry syntheses
of natural products. Chem. Soc. Rev. 2013, 42, 8849−8869.
(8) Baumann, M.; Moody, T. S.; Smyth, M.; Wharry, S. A
Perspective on Continuous Flow Chemistry in the Pharmaceutical
Industry. Org. Process Res. Dev. 2020, 24, 1802−1813.
(9) Plutschack, M. B.; Pieber, B.; Gilmore, K.; Seeberger, P. H. The
Hitchhiker’s Guide to Flow Chemistry. Chem. Rev. 2017, 117,
11796−11893.
(10) Kawaguchi, T.; Miyata, H.; Ataka, K.; Mae, K.; Yoshida, J.-i.
Room Temperature Swern Oxidations by Using a Microscale Flow
System. Angew. Chem. Int. Ed. 2005, 44, 2413−2416.
(11) Meenen, E. V.; Moonen, K.; Acke, D.; Stevens, C. V.
Straightforward continuous synthesis of α-aminophosphonates under
microreactor conditions. Arkivoc 2006, 2006, 31−45.
(12) Movsisyan, M.; Delbeke, E.; Berton, J.; Battilocchio, C.; Ley, S.
V.; Stevens, C. V. Taming hazardous chemistry by continuous flow
technology. Chem. Soc. Rev. 2016, 45, 4892−4928.
(13) Micetich, R. G.; Maiti, S. N.; Spevak, P.; Tanaka, M.; Yamazaki,
T.; Ogawa, K. Synthesis of 2β-Azidomethylpenicillin-1, 1-Dioxides
1657
https://doi.org/10.1021/acs.oprd.1c00127
Org. Process Res. Dev. 2021, 25, 1648−1657