An improved procedure for the reduction of 2,4-dinitrobenzaldehyde to 2,4-diaminobenzaldehyde with iron and acetic acid under dose-controlled conditions

Full text of DOI:10.1021/op0100089

Organic Process Research & Development 2001, 5, 539−541
An Improved Procedure for the Reduction of 2,4-Dinitrobenzaldehyde to
2,4-Diaminobenzaldehyde with Iron and Acetic Acid under Dose-Controlled
Conditions
David C. Whritenour,* Steven J. Brenek, and Norma J. Tom
Pfizer, Inc., Global Research and DeVelopment, Chemical Research and DeVelopment, Eastern Point Road,
Groton, Connecticut 06340
To fulfill a need for gram to kilogram quantities of 2,4-
diaminobenzaldehyde (2) a safe and operationally simple
procedure to reduce 2,4-dinitrobenzaldehyde (1) to 2,4-
diaminobenzaldehyde (2) was developed.The reduction of
1 to 2 with iron powder and aqueous HCl in ethanol at 95
Figure 1.
°C has been described.1 Due to the highly energetic nature
of 1 as shown by DSC2 (Figure 2) and its impact sensitivity,3
a more thorough study was undertaken to adapt this process
reduction within 1 h. How oxygen acts to activate the iron
surface and allow the reaction to occur is currently unclear.
To obtain a reproducible procedure for reactions under a
nitrogen atmosphere, the iron powder was activated by the
for safe scale-up.
Attempts to scale the published procedure1 gave tarry
solids that were difficult to granulate and contained signifi-
addition of 1 equiv of acetic acid to the iron/water mixture,
cant quantities of iron. Chromatography was required to
prior to the addition of a solution of 1. The use of 4 equiv
purify the material, and the isolated yields were typically
of iron and 6.5 equiv of acetic acid per nitro group was the
less than 50%. Further investigations led to a procedure
preferred stoichiometry.
employing an acetic acid/ethyl acetate/water mixture and a
The reaction energy can be effectively controlled by the
reaction temperature below 50 °C. High temperatures were
rate of addition of 1, as a solution in ethyl acetate/acetic
acid, to a slurry of iron powder in water. Ideally 1 was added
not required, and the product was cleaner. It was noted that
the decomposition/polymerization of 2 is catalyzed by strong
acid. Subsequent product isolations were conducted in base
(aqueous NaOH)-washed glassware. An extractive work-up
with ethyl acetate and treatment of the extracts with activated
carbon (Darco G60) (10% w/w relative to 1) were effective
at removing the iron salts from the crude product. After
partial concentration, the desired product could then be
precipitated into hexanes to give good quality material
(>95% by HPLC, NMR) in good yields (70-80%). This
isolation protocol also minimizes any polymerization due to
the presence of the diamine and aldehyde functionality.
In attempts to optimize the loading level of reagents, initial
difficulties were encountered in reproducing our own results.
It was observed that the reduction could be accomplished
with as little as 2.5 equiv (mol/mol) of iron and 3.5 equiv of
acetic acid for each nitro group if it was run open to the air.
(Literature stoichiometry4 is 3 equiv of iron and 6 equiv of
acid per nitro group.) In a second experiment, under nitrogen,
no reaction was observed even with efficient stirring at reflux
overnight. When the reactor was opened to the air and the
mixture transferred to a second container, the mixture self-
heated and proceeded quite quickly to give almost complete
at such a rate as to maintain the temperature below 50 °C.
Under these conditions, the kinetics are fast enough to
maintain dose rate control of the reaction exotherm and limit
the accumulation of 1 in the reaction mixture. Therefore,
there is a lower probability of thermal runaway, thus reducing
the overall risk of the process. An online monitoring tool,
such as in situ FTIR, is recommended to ensure that 1 does
not accumulate.
DSC analysis of the product 2 shows no apparent energy
release between ambient temperature and 350 °C. Further-
more, DSC analysis of samples of the reaction mixture
throughout the addition also show no energy release. TLC
analysis of these samples showed only 2 and led to the
conclusion that under these reaction conditions there is no
accumulation of 1 and the reduction intermediates.
Heat Flow Calorimetry Results and Discussion
Reaction calorimetry was performed using a Mettler-
Toledo RC-1e reaction calorimeter equipped with a 1-L
SV01 reactor. The reactor was initially charged with iron
powder and water held isothermally at 50 °C. Automated
doses (linear) of 0.5 equiv of acetic acid, then a solution of
1 in ethyl acetate and acetic acid were completed. Heat flow
was monitored to give Figure 3 and the accompanying table.
Figure 3 illustrates the heat flow associated with each dose
in W/L throughout this process and clearly shows, along with
the TLC and DSC profiling noted previously, that the
reaction is dose rate-controlled (2-h addition) and there is
(1) Merlic, Craig A.; Motamed, Soheil; Quinn, Bruce J. Org. Chem. 1995, 60,
3365-3369.
(2) DSC Result of isolated solid: Exothermic decomposition near 218 °C
liberating 2669 J/g of energy. DSC Result of 1:1 w/w in ethyl acetate:
Exothermic decomposition near 219 °C liberating 477 J/g of energy.
(3) BAM Fallhammer Test: Impact sensitive above 56 J.
(4) Owsley, D. C.; Bloomfield, J. J. Synthesis 1977, 118-120.
10.1021/op0100089 CCC: $20.00 © 2001 American Chemical Society and The Royal Society of Chemistry
Published on Web 08/07/2001
Vol. 5, No. 5, 2001 / Organic Process Research & Development
•
539

Figure 2.
Figure 3.
no accumulation of reaction intermediates. A slower rate of
addition would reduce the cooling load needed to maintain
isothermal reaction conditions at 50 °C.
acetic acid/ethyl acetate (800 mL). A portion of the aldehyde
solution (5 mL) was added to the iron mixture, which led to
a dissipation of the frothing. The reaction mixture was
warmed to 35 °C with heat from a steam bath. The steam
was turned off, and the dinitrobenzaldehyde solution was
added at a rate to maintain the temperature below 50 °C.
The addition was complete after 6 h. The reaction mixture
was diluted with water (1 L), and Celite (100 g) was added.
The reaction mixture was stirred an additional 3 h while the
temperature dropped to 25 °C. The solids were filtered off,
and the organic layer was separated. The aqueous phase was
extracted with ethyl acetate (3 × 400 mL). The combined
Experimental Section
2,4-Diaminobenzaldehyde, 2. To a nitrogen purged 5-L,
4-neck flask fitted with a condenser, mechanical stirrer,
addition funnel, and temperature probe was added 325 mesh
iron dust (220 g, 3.9 mol), water (800 mL), and acetic acid
(5 mL). Over the next hour, some frothing occurred and the
temperature rose to 28 °C. In a separate container, 2,4-
dinitrobenzaldehyde (97 g, 0.49 mol) was dissolved in 1:1
540
•
Vol. 5, No. 5, 2001 / Organic Process Research & Development

extracts were used to wash the solids from the initial
filtration. The filtered organics were washed with water (400
mL) and saturated aqueous NaHCO3 (3 × 400 mL). The
organics were dried over anhydrous MgSO4 and Darco G-60
(10 g). After filtration, the organics were concentrated in
vacuo to a slurry and diluted with 1 L of hexanes. The
precipitated solids were collected by suction filtration and
dried in air to give 2,4-diaminobenzaldehyde (48 g, 71%)
as a light yellow solid.
Mp 151 °C. 1H MNR (acetone-d6) δ 5.48 (br s, 2H), 5.94
(d, 1H J ) 1.9 Hz), 6.08 (dd, 1H, J ) 2.0, 8.6 Hz), 6.75 (br
s, 2H), 7.20 (d, 1H, J ) 8.6 Hz), 9.51 (s, 1H). 13C NMR
(acetone-d6) 189.8, 155.0, 153.4, 137.8, 111.6, 104.8, 96.9.
Anal. Calcd for C7H8N2O: C, 61.75; H, 5.92; N, 20.58;
Found C, 61.69; H, 5.98; N, 20.39.
Received for review January 31, 2001.
OP0100089
Vol. 5, No. 5, 2001 / Organic Process Research & Development
•
541