steps, the formation of 3-methoxyphenyllithium (8b) and its
addition to ketone 9, are quite exothermic in batch reactions
with worst case temperature rise (Tad) of 61.9 and 133 °C,
respectively. Very low reaction temperatures are required
due to the instability of 3-methoxyphenyllithium intermedi-
ate. We concluded then that a continuous reaction may be a
good alternative to batch synthesis to improve the reaction
yield and to minimize the safety concerns of the reaction
sequence. Consequently, we built a benchtop continuous
reactor utilizing HPLC pumps and tubing to carry out this
reaction in a continuous mode. However, we were unsuc-
cessful due to the inefficient heat transfer in this system.
Recently, with the availability of the microreactor system,
we revisited this reaction and reexamined the continuous
process using unsubstituted cyclohexanone as a model
compound.17
First, we carried out the reaction on a small scale in batch
mode at -10 and -65 °C and obtained 32 and 80% yield
of the expected tertiary alcohol 11, respectively. Clearly, low
temperature is preferred due to the instability of lithium
intermediate 8b at high temperatures. The CYTOS system
offers two low-temperature ranges, -20 °C with a conven-
tional chiller, and -65 °C with a more powerful chiller. With
two successive microreactor systems, we would be able to
carry out both reaction steps in a continuous mode with
precise control. However, with only one system available
and a regular chiller, we chose to study the metal-halogen
exchange step on the microreactor at -14 °C with 17 s
residence time, and then react the resulting lithium interme-
diate 8b with cyclohexanone in batch mode at -40 °C. We
achieved 87% yield and a throughput of 54 g of product per
hour.
carried out in the presence of BF3‚Et2O in ether at -25 °C.19
In addition to using ether as solvent, such reaction conditions
involving the use of ethyl diazoacetate pose safety concerns
for large-scale synthesis without the proper safety and
engineering controls to mitigate both the chemical and
operational hazards associated with the use of this reagent.20
We investigated this reaction, in a batch mode, on 70-mg
scale without any significant safety concerns. Several
variables were studied including solvent, reaction tem-
perature, stoichiometry, and order of reagent addition to
eliminate the by-products and increase the yield and purity.
We found the solvent and reaction temperature are critical
for the success of the reaction. Increasing the reaction
temperature resulted in more by-products. An 81% yield was
obtained when 0.2 equiv of BF3‚Et2O was added at <15 °C
to a mixture of ethyl diazoacetate (13) and N-Boc-4-
piperidone (12) in 1:4 ratio of CH2Cl2/MTBE solvent
mixture.
However, reaction calorimetry results using the Mettler
RC-1 reaction calorimeter show that the addition of the
BF3‚Et2O solution to a mixture of N-Boc-4-piperidone (12)
and ethyl diazoacetate (13) is very exothermic (Figure 3).
The heat release rate with this mode of addition is not feed-
controlled, and an initiation period is observed. The reaction
is very sluggish, and the majority of the reaction occurs en
mass once 60% of the material has been added. Also, it is at
this point that the majority of the gas release takes place.
On the basis of the calculated worst case temperature rise
(Tad ) 45.63 °C), the reaction would quickly heat up and
approach the solvent reflux temperature, depending on the
starting reaction temperature, possibly spewing the contents
of the reactor if loss of cooling or stirring occurred with full
accidental mischarge of all the BF3‚Et2O. Even using the
RC-1, the reaction temperature was difficult to control as
the reaction quickly progressed. Although the Tad result
showed that the decomposition temperature of ethyl diazo-
acetate would not be reached, a chemical reaction hazard
still exists for this addition and mode of operation due to
evolution of large amounts of nitrogen gas as the reaction
proceeds, which could result in over-pressurization of the
reaction vessel. Scaling this reaction to kilogram scales safely
in a conventional reactor is not recommended because of
these reasons.
Reactions Involving Hazardous Reagents. In process
research, we often encounter reactions involving potentially
hazardous reagents such diazo compounds, azides, etc. We
often have to devote time and resources to find suitable
experimental and engineering designs to safely scale-up these
reactions or to design safer alternative syntheses.18 Since the
actual reaction volumes in a microreactor are very small,
the safety concerns are minimized. This has been proven
when we studied the following ring-expansion reaction (eq
4) on the microreactor system.
However, using the microreactor, we were able to reduce
the time-consuming process research to find optimal safe
conditions. Applying the conditions established on the 70-
mg scale reaction in batch mode, without any further
optimization, the reaction proceeded smoothly and safely
with precise control on the microreactor system to form the
desired product in 89% yield. This is a rapid reaction with
only 1.8 min residence time. Under these conditions, we
achieved a throughput of 91 g per hour. This reaction
The reaction of N-Boc-4-piperidone (12) with ethyl
diazoacetate (13) gives 90% crude yield of product 14 when
(16) Unpublished results.
(17) (a) Wu, T.; Xiong, H.; Rieke, R. J. Org. Chem. 1990, 55, 5045. (b) Thurkauf,
A.; Costa, B.; Yamaguchi, S.; Mattson, M.; Jacobson, A. J. Med. Chem.
1990, 33, 1452.
(18) (a) Clark, J.; Shah, A.; Peterson, J. Thermochim. Acta 2002, 392. (b) Perrault,
W.; Pearlman, B.; Godrej, D.; Jeganathan, A.; Yamagata, K.; Chen, J.; Lu,
C.; Herrinton, P.; Gadwood, R.; Chan, L.; Lyster, M.; Maloney, M.;
Moeslein, J.; Greene, M.; Barbachyn, M. Org. Process. Res. DeV. 2003, 7,
533.
(19) Roglans, A.; Marquet, J.; Moreno-Manas, M. Synth. Commun. 1992, 22,
1249.
(20) (a) Searle, N. Organic Syntheses; John Wiley and Sons: New York, 1962;
Collect. Vol. 4, p 426. (b) The use of ethyl diazoacetate on 340-kg
scale has been reported without safety data. Simpson, J.; Godfrey, J.;
Fox, R.; Kotnis, A.; Kacsur, D.; Hamm, J.; Totelben, M.; Rosso, V.;
Mueller, R.; Delaney, E.; Deshpande, R. Tetrahedron: Asymmetry 2003,
14, 1569.
458
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development