Mechanistic insights and safety evaluation of the Ritter reaction utilizing tert -butyl acetate as the tert -butyl cation source

Article abstract of DOI:10.1021/op3000042

The Ritter reaction utilizing tert-butyl acetate as the tert-butyl cation source was investigated by in situ FTIR and calorimetry under various reaction conditions. It was established that, when a batch mode reaction in acetic acid was performed, minimal isobutylene was evolved into the headspace, thereby avoiding pressurization of the reaction vessel. The safety of these conditions is due to the equilibrium of isobutylene and acetic acid with tert-butyl acetate. In addition, the observation of acetic anhydride provided insight into the role of acetic acid in the mechanistic pathway.

Full text of DOI:10.1021/op3000042

Safety/Environmental Report
pubs.acs.org/OPRD
Mechanistic Insights and Safety Evaluation of the Ritter Reaction
Utilizing tert-Butyl Acetate as the tert-Butyl Cation Source
,
†
,†
†
‡
†
Scott W. Roberts,* Stephen M. Shaw,* Jacqueline E. Milne, Dawn E. Cohen, Justin T. Tvetan,
†
†
Joseph Tomaskevitch, Jr., and Oliver R. Thiel
†
‡
Department of Chemical Process Research and Development, Department of Analytical Research and Development, Amgen Inc.,
One Amgen Center Drive, Thousand Oaks, California 91320-1799, United States
*
S Supporting Information
ABSTRACT: The Ritter reaction utilizing tert-butyl
acetate as the tert-butyl cation source was investigated by
in situ FTIR and calorimetry under various reaction
conditions. It was established that, when a batch mode
reaction in acetic acid was performed, minimal isobutylene
was evolved into the headspace, thereby avoiding
pressurization of the reaction vessel. The safety of these
conditions is due to the equilibrium of isobutylene and
acetic acid with tert-butyl acetate. In addition, the
observation of acetic anhydride provided insight into the
role of acetic acid in the mechanistic pathway.
Figure 1. Proposed isobutylene/acetic acid/tert-butyl acetate equili-
brium.
INTRODUCTION
■
The Ritter reaction provides a powerful method to construct N-
tert-butyl amides from nitriles. A variety of tert-butyl cation
sources have been employed in this transformation, including
1
1
solvent, and MSA. It was determined that reagent addition
rates and initial solvent composition played an important role
in the control of isobutylene emissions.
1
2
3
tert-butanol, isobutylene, tert-butyl acetate, and tert-butyl
ethers. Regardless of the tert-butyl cation source, isobutylene
gas is considered a likely intermediate. Isobutylene is a VOC
flammability class 4 gas, and uncontrolled emission events
have been reported in batch processing mode. In 2009, our
group reported Ritter reaction conditions utilizing sulfuric acid,
tert-butyl acetate, and glacial acetic acid as solvent, whereby the
acetic acid acts as a trap for isobutylene (Figure 1). The
reported conditions were safely and successfully scaled to a
multikilogram batch process following a thorough hazard
analysis.
Recently, Wirth and co-workers reported the development of
a Ritter reaction as a continuous process, citing concerns about
the safety of the Ritter reaction in batch mode. This claim
prompted us to disclose the details of the safety study
conducted on our process prior to scale-up, examining both
the safety profile and isobutylene emission in batch mode.
4
−6
The reactions described employed 4-methoxybenzonitrile as
a model substrate, and were conducted in a closed system/
Mettler-Toledo RC1e calorimeter and monitored using a
Mettler-Toledo ReactIR 4000 equipped with a DiComp fiber-
7
8
12
optic probe. Individual reference spectra were acquired and
used to identify species within the crude reaction mixture.
Trends were generated by integrating indicative IR peaks that
were selected by overlaying reference spectra with those of the
9
1
3
reaction mixture. Due to the functional complexity and
infrared superposition of several reaction species, isobutylene
concentration could not be determined in any of the attempted
1
experiments. H NMR spectra of the crude reaction mixture
were taken at the end of every experiment to confirm IR results.
In all experiments, full conversion to the desired N-(tert-butyl)-
10
4
-methoxybenzamide was observed after ageing overnight, with
no impact on conversion observed between conditions.
Conditions A. The conditions using sulfuric acid and tert-
butyl acetate as solvent were performed by the addition of neat
sulfuric acid over a 30 min period to a solution of 4-
methoxybenzonitrile in tert-butyl acetate at 45 °C (eq 1).
After the acid addition was complete, the reaction was
monitored over a period of 22 h. At the end of the acid
RESULTS AND DISCUSSION
■
Our primary focus was to develop a robust and scalable Ritter
reaction. During our initial optimization, a number of
conditions were identified that could perform the desired
transformation. This report focuses on the hazard evaluation of
three main conditions which were considered: (Conditions A)
3a
Special Issue: Safety of Chemical Processes 12
Received: January 4, 2012
tert-butyl acetate (neat), and sulfuric acid; (Conditions B)
tert-butyl acetate, toluene as solvent, and methanesulfonic acid
(MSA); and (Conditions C) tert-butyl acetate, acetic acid as
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/op3000042 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Safety/Environmental Report
Table 2. Calorimetry data from various Ritter reaction
conditions
a
addition PATR
heat of reaction
(kJ/mol)
conditions
solvent
acid
time
(°C)
A
A
B
C
C
tert-BuOAc
tert-BuOAc
toluene
H SO4
30 min
<1 min
4 h
72
47
I8
−103.0
−63.0
−24.4
−56.2
−67.8
−48.9
2
H SO4
2
addition, the reaction had produced 22 L/kg of isobutylene
MSA
MSA
MSA
MSA
(
3.5% of the total isobutylene available based upon full
14
acetic acid
acetic acid
acetic acid
40 min
<1 min
30 min
53
47
37
degradation of tert-butyl acetate, 13−22 psi; Table 1).
Controlled venting of the reaction mixture at 45 °C resulted
in a minor visible release of gas into the headspace, with no
change in reaction volume.
b
C
a
RC-1 used to calculate ATR from the measured heat of reaction; not
the true adiabatic temperature rise. Thus, “predicted adiabatic
temperature rise” (PATR) is reported. Conditions: 2.0 equiv t-
BuOAc; 3.0 equiv MSA, 2.0 vol acetic acid; 20 °C (no nitrile present).
It should be noted that a 72 °C predicted adiabatic
b
temperature rise (PATR) was calculated during the 30 min
1
5
acid addition (ΔH = −103.0 kJ/mol; Table 2). The heat of
rxn
reaction is believed to include a combination of: (a) the
intrinsic heat of reaction, (b) heat of acid mixing, (c) heat of
dosing, (d) the heat of tert-butyl acetate decomposition into
isobutylene and acetic acid, and (e) heat loss from off-gas/
evaporation of isobutylene into the headspace.
1
6
During the course of the reaction, a decrease in tert-butyl
acetate concentration and an increase in acetic acid
concentration was observed by in situ IR. N-(tert-butyl)-4-
methoxybenzamide increased in concentration throughout the
acid addition, and reached a “steady-state” concentration at the
end of addition. Accelerating rate calorimetry (ARC) analysis of
these conditions revealed three exotherms: (i) 30−80 °C (due
to heat of reaction), (ii) 90−100 °C, and (iii) 145−150 °C
1
7
(
Figure 2). After quickly increasing to 3.2 bar, the pressure
decreased during the initial exotherm, possibly due to
isobutylene equilibration back into solution as tert-butyl acetate.
1
H NMR analysis of the bomb contents after the first exotherm
Figure 2. ARC analysis (red = temperature, blue = pressure).
Conditions: 1.0 equiv of 4-methoxybenzonitrile, 1.0 equiv of sulfuric
acid, 3.8 vol of tert-butyl acetate; components were added together at
ambient temperature and heated in the ARC.
showed desired product, the second exotherm showed 4-
methoxybenzamide present with partial decomposition, and the
third exotherm was not investigated as it was far removed from
the process temperature.
Next, we examined a faster sulfuric acid addition rate at lower
reaction temperature (20 °C, <1 min), followed by heating to
observed, most probably due to the isobutylene gas being
18
4
5 °C. The acid was manually added in one portion, and the
driven into the headspace. In conclusion, a higher overall
operating pressure and evolution of isobutylene gas was
observed for these conditions.
reactor was immediately sealed. Consequently, pressure due to
compression was not captured. Visible gas generation was
immediately evident, and 10 psi pressure was generated after
the acid addition, equilibrating to 7 psi. After thermal
equilibration at 20 °C, the reactor was heated to 45 °C, and
a pressure increase to 21 psi was observed which equilibrated to
Conditions B. In 2005, conditions minimizing isobutylene
emissions were reported by Diaz and co-workers for the tert-
butoxycarbonyl (Boc) deprotection of amines on pilot-plant
19
scale. Their protocol involved polymerization of isobutylene
by employing MSA/toluene as an external/internal scrubbing
system. The second set of reaction conditions studied
employed toluene as solvent and was performed by adding
neat tert-butyl acetate over 3 h into a solution of 4-
methoxybenzonitrile, MSA, and toluene at 45 °C (eq 2). It
was postulated that the MSA/toluene conditions may create an
“internal” isobutylene scrubbing system similar to that reported
by Diaz and co-workers.
1
2
5 psi after 22 h (ΔH = −63.0 kJ/mol, 47 °C PATR; Table
rxn
). Also noteworthy was a steep decrease in tert-butyl acetate
observed by in situ IR immediately following the acid charge.
Interestingly, the tert-butyl acetate concentration subsequently
increased upon equilibration at 20 °C, presumably due to the
sequestering action of acetic acid by reacting with isobutylene
to form tert-butyl acetate in solution. When the temperature
was raised to 45 °C, a small decrease in tert-butyl acetate was
Table 1. Initial, compression, and steady-state pressures produced from various Ritter reaction conditions
addition
time
compression pressure
(psi)
initial pressure
steady-state pressure
isobutylene gas evolution
% of total isobutylene
a
a
b
c
conditions
solvent
(psi)
(psi)
(L/kg)
available
A
B
C
t-BuOAc
toluene
30 min
4 h
0.2
17
9
22
13
3
13
15
22
17
1
3.5
1.0
0.3
acetic acid
40 min
1
a
b
Pressure corrected for compression (observed pressure − compression pressure). Liter of isobutylene per kg of 4-methoxybenzonitrile starting
(g)
c
material. Based upon the full degradation of t-BuOAc.
B
dx.doi.org/10.1021/op3000042 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Safety/Environmental Report
psi), and the heat of reaction was equal to −48.9 kJ/mol (Table
2
). Compared to the heat of reaction observed for the Ritter
reaction using these conditions, the heat associated with MSA/
acetic acid mixing and tert-butyl acetate decomposition
represents a significant factor in the total heat of reaction.
ARC analysis identified three exotherms: (i) 30−55 °C (due
to heat of reaction), (ii) 75−79 °C (due to loss of the tert-butyl
group), and (iii) 135−150 °C (not investigated, far removed
The addition of tert-butyl acetate produced 17 L/kg
isobutylene (1.0% of the total available isobutylene, 13−15
psi; Table 1). After the reaction aged overnight, the reactor was
slowly vented at 45 °C, and a vigorous off-gassing event ensued.
The reactor was resealed and cooled to 20 °C. At ambient
temperature, the reactor was slowly vented and did not produce
any visible off-gassing. However, upon reheating the reactor to
22
from the process temperature; Figure 5). It should be noted
that the exotherm due to the heat of reaction under acetic acid
conditions was approximately half of the exotherm observed
when employing the tert-butyl acetate conditions (Figure 2 vs
Figure 5). Furthermore, no pressure increase was observed for
the system until the boiling point of AcOH had been reached,
which differed from the tert-butyl acetate conditions as well.
Overall, it was considered that the decrease in pressure during
the reaction provided additional evidence of the improved
safety of the acetic acid conditions.
While performing the reactions in a sealed vessel was used to
understand the isobutylene equilibrium of the system, utilizing
a sealed vessel on large scale was not considered practical.
Conditions C were investigated under a low flow of nitrogen
gas to determine any possible changes in the safety profile and/
or calorimetry. For this experiment, the heat of reaction was
4
5 °C, 9 L/kg isobutylene (0.5% of the total available
isobutylene, 8 psi) was generated, and an additional off-gassing
event was observed upon controlled venting. The total volume
loss from venting was approximately 10%. Consequently, the
bulk polymerization and internal scrubbing of isobutylene
appears disrupted and not operational under these reaction
conditions. The PATR was the lowest investigated (18 °C,
20
ΔH = −24.4 kJ/mol, Table 2). In situ IR analysis displayed
rxn
an increase in tert-butyl acetate concentration as the addition
progressed and then declined in concentration after the
addition ceased and ageing continued, presumably due to
equilibration of the system. In conclusion, these conditions
produced higher operating pressure, and visible off-gassing of
isobutylene was observed while venting at 45 °C.
Conditions C. The third set of conditions using acetic acid
as solvent were performed by adding MSA diluted in acetic acid
over 30 min to a reactor containing a slurry of 4-
methoxybenzonitrile, tert-butyl acetate, and acetic acid at an
internal temperature of 20 °C (eq 3 and Figure 3).
−
57 kJ/mol, and the PATR was 49 °C. Thus, no change in
calorimetry (or visible generation of gas) was observed as
compared to the sealed experiments. Furthermore, multikilo-
gram manufacturing utilizing conditions C and a nitrogen
23
sweep of the reactor headspace produced excellent results.
An overall summary of the safety attributes (closed system)
for the conditions investigated is listed below:
(
1) Conditions A involved the addition of sulfuric acid to
tert-butyl acetate (solvent). These conditions showed the
highest PATR and the highest evolution of isobutylene.
2) Conditions B involved the addition of tert-butyl acetate
to a solution MSA and toluene. These conditions showed
the lowest PATR; conversely, opening of the sealed
vessel led to a visible off-gas event.
(
Minimal isobutylene was produced for the process (1 L/kg,
.3% of the total available isobutylene, 1−3 psi; Table 1). The
(
3) Conditions C involved the addition of an MSA/acetic
acid mixture to a tert-butyl acetate/acetic acid mixture.
These conditions showed the second-highest PATR
observed and provided the lowest evolution of
isobutylene gas. Off-gas events were not observed, even
while stressing the conditions at a higher temperature.
0
effect of temperature on the isobutylene buffering capacity of
acetic acid was investigated by heating the crude reaction
mixture. Heating the crude reaction mixture from 20 to 45 °C
did not produce any additional pressure, and controlled venting
of the reactor did not produce noticeable off-gassing or changes
in the IR. Upon further heating from 45 to 55 °C, the pressure
did not increase, and no visible off-gassing or changes in the IR
were observed. A 53 °C PATR was observed for the process
MECHANISTIC INSIGHT
■
Our use of the Ritter reaction for API manufacturing included a
crystallization of the desired tert-butyl amide directly out of the
crude reaction mixture. On the basis of the literature, two
different mechanisms have been proposed, involving two
distinct intermediates whose proposed existence was not fully
understood. The generally accepted mechanism for the Ritter
reaction involving isobutylene in the presence of sulfuric acid is
the activation of nitrile 3 by the tert-butyl group 4 to generate
the corresponding nitrilium ion 5, followed by formation of
iminosulfate 6 (Scheme 1). Hydrolysis of iminosulfate 6
(
ΔH = −56.2 kJ/mol; Table 2). Thermal conversion of 65%
rxn
was observed at the end of the addition (35% accumulation,
Figure 4). Thermal conversion was lower than the IR
conversion (IR conversion at end of addition = 85%), and
the difference can be rationalized by the cumulative nature of
factors controlling the heat of reaction described previously.
Interestingly, faster MSA addition times at 20 °C (<1 min) did
not affect the pressure (≤2 psi during the entire process, ΔH
=
rxn
−67.8 kJ/mol, 47 °C PATR; Table 2). In an attempt to
2
determine the heat of reaction contribution from mixing MSA
with acetic acid and tert-butyl acetate decomposition into acetic
acid and isobutylene gas, the procedure was repeated by adding
a premixed solution of MSA/acetic acid into a solution of tert-
butyl acetate/acetic acid at 20 °C (30 min addition time, nitrile
omitted). Minimal pressure was produced upon addition (1
generates amide 7.
In 1966, Glikmans, Torck, Hellin, and Coussemant proposed
an alternative mechanism utilizing the original Ritter reaction
24
conditions (sulfuric acid, acetic acid as solvent). This
mechanism (Scheme 2) differed from the Ritter mechanism
by invoking the participation of acetic acid. On the basis of the
C
dx.doi.org/10.1021/op3000042 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Safety/Environmental Report
Figure 3. IR and calorimetry profiles (relative time axis cropped). Conditions C: 1.0 equiv of 4-methoxybenzonitrile, 2.0 equiv of tert-butyl acetate (1
vol acetic acid); 3.0 equiv of MSA (1.0 vol acetic acid); 20 °C, 30 min addition time.
Figure 5. ARC analysis (temperature = red, pressure = blue).
Conditions: 1.0 equiv of 4-methoxybenzonitrile, 2.0 equiv of tert-butyl
acetate; 3.0 equiv of MSA, 2.0 vol of acetic acid, components added
together at ambient temperature and heated in ARC.
Figure 4. Heat flow vs time (time axis cropped). Conditions C: 1.0
equiv of 4-methoxybenzonitrile, 2.0 equiv of tert-butyl acetate (1 vol
acetic acid); 3.0 equiv of MSA (1.0 vol acetic acid); 20 °C, 30 min
21
addition time. .
which is immediately converted to amide 7. These results
corroborate the previous mechanistic studies reported by Jiang
and Prasad, in which the quenching of a Ritter reaction with
detection of acetic anhydride in the crude reaction mixture by
GC analysis, they argued that acetic acid reacts with nitrilium
ion 5 to generate acyl imidate 8, then another equivalent of
acetic acid reacts with acyl imidate 8 to generate amide 7 and
acetic anhydride (9).
18
18
3b
Na OH failed to produce O-enriched product.
CONCLUSIONS
■
Under reaction conditions C, acetic anhydride was detected
Three conditions for the Ritter reaction were investigated using
tert-butyl acetate as the cation source. Employing tert-butyl
acetate as the solvent and sulfuric acid created 22 L/kg of
isobutylene. The “all-in” addition of sulfuric acid resulted in
visible gas generation. Utilizing toluene as solvent and MSA
created 17 L/kg isobutylene. Venting of the closed system at 45
2
5
by in situ IR. Additional control experiments proved acetic
anhydride was not generated from background reactions of
26
MSA with acetic acid or tert-butyl acetate. Importantly, the
acyl imidate 8 was not detected in any of the conditions
investigated and is believed to be a short-lived intermediate
D
dx.doi.org/10.1021/op3000042 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Safety/Environmental Report
Scheme 1. Ritter’s originally proposed mechanism
nitrile (45 g, 0.34 mol, 1.0 equiv) and tert-butyl acetate (171
mL, 3.8 vol) to form a light-brown slurry. After thermal
equilibration at 20 °C, an RC-1 heat flow calibration was
performed (preaddition). Sulfuric acid (18 mL, 0.34 mol, 1.0
equiv) was added in one portion, and the reactor was
immediately sealed. The solution initially turned orange,
followed by a dark greenish brown (CAUTION: Gas bubbles
observed during the initial exotherm!). After addition was
complete, the reactor was heated to 45 °C and aged overnight
(
CAUTION: Gas bubbles observed during heating!). Following
the overnight age, the reactor vent was opened to observe for
isobutylene off-gassing. The reactor was cooled to 20 °C, an
RC-1 heat flow calibration was performed, and the crude
reaction mixture was sampled for NMR analysis.
Ritter Reaction using Toluene, tert-Butyl Acetate, and
MSA (3 h addition). To a 1 L MP06 closed-system RC-1
calorimeter equipped with mechanical stirring, diaphragm
pump, and DiComp fiber-optic probe (attached to a Mettler-
Toledo ReactIR 4000) was charged 4-methoxybenzonitrile (40
g, 0.30 mol, 1.0 equiv), toluene (144 mL, 3.6 vol), and MSA
Scheme 2. Glikmans et al. proposed mechanism generating
acetic anhydride
(
19.5 mL, 0.90 mol, 3.0 equiv) to form a light-brown solution.
The reactor was heated to 45 °C, and after thermal
equilibration, an RC-1 heat flow calibration was performed
(
preaddition). tert-Butyl acetate (403 mL, 3.0 mol, 10.0 equiv)
was charged by diaphragm pump over 3 h. After addition was
complete, the reaction mixture was held to allow the heat flow
to return to baseline, and an RC-1 heat flow calibration was
performed. The reaction mixture was aged overnight at 45 °C.
After the overnight age, the reactor vent was opened to observe
for isobutylene off-gassing (CAUTION: Violent off-gas event
observed!). The reactor was cooled to 20 °C, and the reactor
vent was opened to observe for isobutylene off-gassing. The
reactor was reheated to 45 °C, and the reactor vent was again
opened to observe for isobutylene off-gassing (CAUTION: Gas
bubbles observed during venting!). The reactor was cooled to 20
°C and sampled for NMR analysis.
Ritter Reaction using Acetic Acid, tert-Butyl Acetate,
and MSA (40 min addition). To a 1 L MP06 closed-system
RC-1 calorimeter equipped with mechanical stirring, diaphragm
pump, and DiComp fiber-optic probe (attached to a Mettler-
Toledo ReactIR 4000) was charged 4-methoxybenzonitrile
(100 g, 0.75 mol, 1.0 equiv), acetic acid (100 mL, 1.0 vol), tert-
butyl acetate (202 mL, 1.5 mol, 2.0 equiv) to form a light
brown slurry. After thermal equilibration at 20 °C, an RC-1
heat flow calibration was performed (preaddition). A premixed
solution of acetic acid (100 mL, 1.0 vol) and MSA (146 mL,
2.25 mol, 3.0 equiv) was charged by diaphragm pump over 40
min. After addition was complete, the reaction mixture was held
to allow the heat flow to return to baseline, and a RC-1 heat
flow calibration was performed. The reaction mixture was aged
overnight at 20 °C. After the overnight age, the reactor vent
was opened to observe for isobutylene off-gassing. The reactor
was cooled to 20 °C and sampled for NMR analysis.
°
C produced a visible off-gas event. However, using acetic acid
as solvent and MSA, very little isobutylene was produced (1 L/
kg). Moreover, even the “all-in” addition of MSA and venting at
elevated temperatures did not produce an off-gas event. The
equilibrium between acetic acid and tert-butyl acetate is
believed to buffer and control isobutylene emissions, resulting
in lower operating pressures. These conditions have been
conducted on large-scale in batch-processing mode.
EXPERIMENTAL SECTION
■
Ritter Reaction using tert-Butyl Acetate (neat) and
Sulfuric Acid (30 min addition time). To a 1 L MP06
closed-system RC-1 calorimeter equipped with mechanical
stirring, diaphragm pump, and DiComp fiber-optic probe
(
attached to a Mettler-Toledo ReactIR 4000) was charged 4-
methoxybenzonitrile (45 g, 0.34 mol, 1.0 equiv) and tert-butyl
acetate (171 mL, 3.8 vol) to form a light-brown slurry. The
reactor was heated to 45 °C and the slurry dissolved to produce
a light-brown, homogeneous solution. After thermal equilibra-
tion, an RC-1 heat flow calibration was performed (preaddi-
tion). Sulfuric acid (18 mL, 0.34 mol, 1.0 equiv) was charged by
diaphragm pump over 33 min. After addition was complete, the
reaction mixture was held to allow the heat flow to return to
baseline, and an RC-1 heat flow calibration was performed. The
reaction mixture was aged overnight at 45 °C. After the
overnight age, the reactor vent was opened to observe for
isobutylene off-gassing. The reactor was cooled to 20 °C and
sampled for NMR analysis. In all experiments, the analytical
Ritter Reaction using Acetic Acid, tert-Butyl Acetate,
and MSA (“All-In” Addition). To a 1 L MP06 closed-system
RC-1 calorimeter equipped with mechanical stirring, diaphragm
pump, and DiComp fiber-optic probe (attached to a Mettler-
Toledo ReactIR 4000) was charged 4-methoxybenzonitrile (45
g, 0.34 mol, 1.0 equiv), acetic acid (90 mL, 2.0 vol), tert-butyl
acetate (91 mL, 0.67 mol, 2.0 equiv) to form a light-brown
slurry. After thermal equilibration at 20 °C, an RC-1 heat flow
calibration was performed (preaddition). MSA (66 mL, 1.02
mol, 3.0 equiv) was added in one portion, and the reactor was
9
data was in agreement with previously published data.
Ritter Reaction using tert-Butyl Acetate (neat) and
Sulfuric Acid (“all-in” addition). To a 1 L MP06 closed-
system RC-1 calorimeter equipped with mechanical stirring,
diaphragm pump, and DiComp fiber-optic probe (attached to a
Mettler-Toledo ReactIR 4000) was charged 4-methoxybenzo-
E
dx.doi.org/10.1021/op3000042 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Safety/Environmental Report
immediately sealed. After addition was complete, the reaction
mixture was held to allow the heat flow to return to baseline,
and an RC-1 heat flow calibration was performed. The reaction
mixture was aged overnight at 20 °C. In the morning, the
mixture was heated to 45 °C, and the vent was opened to
observe for isobutylene off-gassing. The reactor was further
heated to 55 °C, and the reactor vent was opened for observing
isobutylene off-gassing The reactor was cooled to 20 °C and
sampled for NMR analysis.
(9) Baum, J. C.; Milne, J. E.; Murry, J. A.; Thiel, O. R. J. Org. Chem.
2
009, 74, 2207−2209.
10) (a) Brandt, J. C.; Elmore, S. C.; Robinson, R. I.; Wirth, T. Synlett
010, 20, 3099−3013. (b) “Thermodynamic runaways” during batch
processing was specifically cited as a concern.
11) Sulfuric acid was initially reported with these conditions (see
(
2
(
reference 9). However, MSA was selected for future development due
to low water content and improved impurity profile. Details will be
disclosed in a separate publication.
(12) An MP06, 1 L, 6 bar pressure-rated calorimeter was utilized for
all studies. Solutions were charged into the closed system using a
diaphragm pump.
ASSOCIATED CONTENT
■
(
13) See Supporting Information for reference peak assignment and
IR spectra.
14) The calculation of pressure from the addition of a liquid into a
*
S
Supporting Information
General experimental methods, reference peak assignments,
control experiments, and closed system calculations. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(
closed system is based on the compression of the headspace inside the
system. In our experiments, the headspace was air (headspace not
purged prior to each experiment); thus, the Ideal Gas Law can be used
for the calculation of pressure; see Supporting Information for
calculations. For the purposes of comparison between conditions,
isobutylene was assumed to be an ideal gas. Isobutylene evolution was
calculated on the basis of closed system pressure and does not include
amount lost during venting.
AUTHOR INFORMATION
■
Corresponding Author
scottr01@amgen.com; stshaw@amgen.com.
*
(
15) It should be noted that the heat of reaction for the two examples
Notes
listed for conditions A should be similar. We believe that the heat of
reaction for the “all-in” condition is lower due to the lower reaction
temperature (25 °C vs 45 °C for the dosed conditions), the possible
loss of isobutylene through the open port during the addition (an
evaporative effect), and a reaction that was potentially incomplete
(mixture was heated to 45 °C to ensure reaction completion).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Mr. Seth Huggins for assistance related to the IR
spectrometry experiments, Mr. Mike Ronk for mass spectrom-
etry analysis, and Dr. Matthew Bio for valuable discussions
regarding the preparation of this manuscript. Ms. Jean Baum is
acknowledged for assistance in development of the MSA/tert-
butyl acetate/acetic acid conditions in support of the drug
candidate program. Mrs. Jiemin Bao is acknowledged for
development of the KF method for the determination of water
content in strong acids. Dr. Margaret Faul, Dr. Robert Larsen,
Dr. Jerry Murry, and Mr. Greg Sukay are gratefully acknowl-
edged for support of this project.
(
16) RC-1 data was adjusted accordingly to remove the contribution
from heat of dosing.
17) All ARC experiments were conducted utilizing magnetic stirring.
(
ARC analysis of the reagents/solvent (nitrile absent) did not show
exothermic activity up to 150 °C. See Supporting Information for ARC
analysis.
(
18) We note that large temperature fluctuations could possibly
influence the observed IR trends due to the thermal dependence of
infrared absorptions.
(19) (a) Dias, E. L.; Hettenbach, K. W.; am Ende, D. J. Org. Process
Res. Dev. 2005, 9, 39−44. (b) It should be noted that the presence of
methanol was shown to terminate the scrubbing ability of the system.
REFERENCES
■
(
20) We believe the heat of reaction is lower for these conditions,
(
1) tert-Butyl alcohols: (a) Callens, E.; Burton, A. J.; Barrett, A. G. M.
when compared to the MSA/acetic acid conditions, because there is
no heat of mixing between MSA and acetic acid.
Tetrahedron Lett. 2006, 47, 8699−8701. (b) Tamaddon, F.; Khoobi,
M.; Keshavarz, E. Tetrahedron Lett. 2007, 48, 3643−3646.
(
21) The heat spike postaddition is an artifact of the dosing method,
rather than an actual kinetic effect.
22) Bomb contents were analyzed by H NMR after exothermic
(
c) Martínez, A. G.; Alvarez, R. M.; Vilar, E. T.; Fraile, A. G.;
Hanack, M.; Subramanian, L. R. Tetrahedron Lett. 1989, 30, 581−582.
2) Isobutylene gas: Ritter, J. J.; Minieri, P. P. J. Am. Chem. Soc. 1948,
0, 4045−4048.
3) tert-Butyl acetate: (a) Reddy, K. L. Tetrahedron Lett. 2003, 44,
453−1455. (b) Jiang, X.; Lee, G. T.; Villhauer, E. B.; Prasad, K.;
Prashad, M. Org. Process Res. Dev. 2010, 14, 883−889.
4) tert-Butyl ethers: (a) Reference 3a. (b) Bonse, G.; Blank, H. U.
DE 3002203, 1981.
5) For seminal research, see: (a) Reference 2. (b) Benson, F. R.;
1
(
(
7
(
1
activity. ARC analysis of the reagents/solvent showed a 1 °C exotherm
at 130 °C; however, it was not of the same magnitude of the real
system. See Supporting Information for ARC analysis.
(
23) Conditions utilized for manufacturing were slightly modified
from reference 9 (equivalents, solvent volume). Details will be
disclosed in a separate publication.
(
(
24) Glikmans, G.; Torck, B.; Hellin, M.; Coussemant, F. Bull. Soc.
Chim. Fr. 1966, 4, 1376−1386.
25) (a) The acetic anhydride peak assignment was confirmed by
(
Ritter, J. J. J. Am. Chem. Soc. 1949, 71, 4128−4129. (c) Plaut, H.;
Ritter, J. J. J. Am. Chem. Soc. 1951, 73, 4076−4077. (d) Ritter, J. J.;
Yonkers, N. Y. U.S. Patent 2,573,673, 1951.
(
charging acetic anhydride into the crude reaction mixture. (b) The use
of MSA resulted in lower water content. Adding 1.0 equiv water to
conditions C resulted in no detection of acetic anhydride by IR.
(
6) For reviews, see: (a) Krimen, L. I.; Cota, D. L. Org. React. 1969,
7, 213−325. (b) Bishop, R. In Comprehensive Organic Synthesis; Trost,
B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 6, pp
61−300. (c) Kurti, L.; Czako, B. Strategic Applications of Named
Reactions in Organic Synthesis; Elsevier Academic Press: New York,
005; pp 382−383.
7) “Volatile Organic Compounds” per definition CFR Part 51.100
s). Isobutylene boiling point = 6.9 °C, flash point = −80 °C (closed
cup).
1
(26) Further processing and examination of IR data, including
derivitization and principal component analysis failed to identify acyl-
imidate intermediates of type 8 in all experiments.
2
̈
́
2
(
(
(
8) Chang, S.-J. Org. Process Res. Dev. 1999, 3, 232−234 and
references cited therein.
F
dx.doi.org/10.1021/op3000042 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX