Synthesis of primary amides by lipase-catalyzed amidation of carboxylic acids
with ammonium salts in an organic solvent
Mike J. J. Litjens, Adrie J. J. Straathof,* Jaap A. Jongejan and Joseph J. Heijnen
Delft University of Technology, Kluyver Laboratory for Biotechnology, Julianalaan 67, 2628 BC Delft, The
Netherlands. E-mail: straathof@stm.tudelft.nl
Received (in Liverpool, UK) 30th March 1999, Accepted 12th May 1999
The synthesis of butyramide and oleamide, by Candida
antarctica lipase B-catalyzed amidation of the carboxylic
acids, in an organic solvent with ammonium bicarbonate or
ammonium carbamate as a source of ammonia results in
good yields, making prior activation of the acids un-
necessary.
The experimental results are very encouraging, yields at
equilibrium for the amide (after 17 days) were found to be
80–90% when using ammonium bicarbonate and 90–100%
when using ammonium carbamate as the source of ammonia
(Fig. 1). In addition the model predictions agree reasonably well
with the experimental results. No butyramide was formed in the
absence of the enzyme, thus the amidation is enzyme-catalyzed.
Fig. 1 shows that the amidation proceeds better with ammonium
carbamate than with ammonium bicarbonate as the source of
ammonia. Model calculations show that this results from the
combined effect of a higher ammonia concentration and a lower
water concentration. Ammonium bicarbonate dissolves as one
equivalent of ammonia, water, and carbon dioxide, whereas
ammonium carbamate dissolves as two equivalents of ammonia
and one equivalent of carbon dioxide. It is obvious that an
increased concentration of ammonia improves the yield at
equilibrium, and that an increased concentration of water, which
is a coproduct of the amidation, decreases the yield at
equilibrium.
With ammonium carbamate as the source of ammonia, the
yields obtained in the present experiments approach the
optimized yields of the two-step process by De Zoete et al.2 for
the synthesis of octanamide and oleamide from the carboxylic
acids via the esters. The yield of our direct enzymatic amidation
may be improved by increasing the ammonia concentration.
According to the model calculations4 the ammonia concentra-
tion is below 30 mmol l21 under the experimental conditions
employed. Up to 230 mmol l21 of ammonia may be used by
saturating the reaction medium with gaseous ammonia.4 The
removal of reaction water, for example by molecular sieves,
may further improve the yield.
Primary amides are important derivatives of several types of
carboxylic acids, such as fatty acids and amino acids. In
addition, enantioselective amidation of chiral acids may be used
as an alternative to esterification in kinetic resolution processes.
However, direct reaction of carboxylic acids with ammonia
requires extreme conditions (200 °C, 7 bar anhydrous ammo-
nia).1 This may lead to the formation of byproducts in the
production of heat sensitive amides such as oleamide. A
selective reaction under mild conditions is therefore desirable.
For amidation of aliphatic acids with ammonia, lipases have
been considered as the catalyst, but they have been rejected
because the formation of a carboxylate anion would lead to
precipitation of the ammonium carboxylate rather than forma-
tion of the acyl–enzyme complex and subsequent reaction with
ammonia.2,3 For this reason amides are usually synthesized
from a neutral and activated form of the acid, such as the acid
chloride (which takes away the need for a catalyst) or an ester.
For example, De Zoete et al.2 developed a one-pot procedure for
the esterification and subsequent ammoniolysis using Candida
antarctica lipase B (CALB) to catalyze both steps. In many
cases it would be attractive if such an activation step could be
avoided.
A recent thermodynamic study on the hydrolysis and
ammoniolysis of the ester butyl butyrate indicates that the
unactivated butyric acid should react efficiently with ammonia
under mild conditions in anhydrous methyl isobutyl ketone
(MIBK).4 We decided to verify these thermodynamic predic-
tions experimentally.
To estimate the productivity of CALB in our one-step
procedure as compared to the two-step procedure by De Zoete
et al.,2 an experiment was performed with oleic acid as the
substrate. The experimental conditions were similar to those
mentioned for butyric acid, except that the temperature was
Here we demonstrate that primary amides can indeed be
formed in good yields under mild reaction conditions, by direct
amidation of the carboxylic acids (Scheme 1). The amidation of
butyric acid to form butyramide and water serves as a model
reaction, using CALB as the biocatalyst. The reaction is
performed in MIBK at 25 °C, and ammonium salts that partly
dissolve during the course of the reaction are used as a
convenient source of ammonia. The experiments were carried
out in 33 ml closed glass vessels containing 25 ml dry MIBK
with various butyric acid concentrations (concentrations are
mentioned in Fig. 1). Immobilized CALB (15 mg) (Novozym
435, a kind gift of NOVO Nordisk, with a catalytic activity of
approximately 11000 PLU g21 preparation) and solid ammo-
nium bicarbonate or ammonium carbamate (ca. 2.5 mmol) were
added, and the suspension was stirred. Samples were taken
through a septum to prevent the escape of gas during sampling,
and were immediately centrifuged and analyzed for butyric acid
and butyramide by GC.
Fig. 1 Yields of butyramide at equilibrium for various initial butyric acid
concentrations with solid ammonium bicarbonate (8, ––) or solid
ammonium carbamate (2, ···) as the source of ammonia. Markers are
measured data, lines are predictions using the model presented in ref. 4.
lipase
O
O
R
C
+
NH3
R
C
+ H2O
OH
NH2
Scheme 1
Chem. Commun., 1999, 1255–1256
1255