Bio-based N-alkyl-2-pyrrolidones by Pd-catalyzed reductive N-alkylation and decarboxylation of glutamic acid

Article abstract of DOI:10.1039/c7gc01829k

Environmental regulations boost the search for new safer and less toxic bio-based solvents to replace controversial high-boiling solvents such as N-methyl-2-pyrrolidone and N,N-dimethylformamide in the chemical industry. Recently, N-alkyl-2-pyrrolidones and 5-methyl-N-alkyl-2-pyrrolidones were proposed as attractive alternative solvents for many applications. Here, we report a bio-based two-step chemocatalytic system for the synthesis of a broad range of N-alkyl-2-pyrrolidones starting from glutamic acid and C₃-C₅ carbonyl compounds. In the first step N-mono-alkylated derivatives of glutamic acid were synthesized in high yields (>85%) by a mild and efficient Pd-catalyzed reductive N-alkylation. Subsequently, thermally induced lactamization to the corresponding N-alkylpyroglutamic acid followed by Pd-catalyzed decarboxylation at 250 °C under inert atmosphere resulted in N-alkyl-2-pyrrolidones. Hydrolytic degradation was partially counteracted by the neutralization of the N-alkylpyroglutamic acid substrate with a base, resulting in yields up to 82%. Finally, both reaction steps were successfully combined in a one-pot process using the same Pd/Al₂O₃ catalyst in different conditions of gas atmosphere and temperature.

Full text of DOI:10.1039/c7gc01829k

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: F. De Schouwer,
S. Adriaansen, L. Claes and D. E. De Vos, Green Chem., 2017, DOI: 10.1039/C7GC01829K.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Page 1 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Bio-based N-alkyl-2-pyrrolidones by Pd-catalyzed reductive N-alkylation
and decarboxylation of glutamic acid
Free De Schouwer,a Sander Adriaansen,a Laurens Claesa and Dirk E. De Vos*a
a Centre for Surface Chemistry and Catalysis, Department of Microbial and Molecular Systems, KU Leuven ꢀ University of
Leuven, Celestijnenlaan 200F, Post Box 2461, 3001 Heverlee, Belgium. Eꢀmail: dirk.devos@kuleuven.be
Abstract
Environmental regulations boost the search for new safer and less toxic bioꢀbased solvents to replace controversial highꢀ
boiling solvents such as Nꢀmethylꢀ2ꢀpyrrolidone and N,Nꢀdimethylformamide in the chemical industry. Recently, Nꢀalkylꢀ2ꢀ
pyrrolidones and 5ꢀmethylꢀNꢀalkylꢀ2ꢀpyrrolidones were proposed as attractive alternative solvents for many applications.
Here, we report a bioꢀbased twoꢀstep chemocatalytic system for the synthesis of a broad range of Nꢀalkylꢀ2ꢀpyrrolidones
starting from glutamic acid and C3ꢀC5 carbonyl compounds. In the first step Nꢀmonoꢀalkylated derivatives of glutamic acid
were synthesized in high yields (> 85%) by a mild and efficient Pdꢀcatalyzed reductive Nꢀalkylation. Subsequently, thermally
induced lactamization to the corresponding Nꢀalkylpyroglutamic acid followed by Pdꢀcatalyzed decarboxylation at 250 °C
under inert atmosphere resulted in Nꢀalkylꢀ2ꢀpyrrolidones. Hydrolytic degradation was partially counteracted by the
neutralization of the Nꢀalkylpyroglutamic acid substrate with a base, resulting in yields up to 82%. Finally, both reaction
steps were successfully combined in a oneꢀpot process using the same Pd/Al2O3 catalyst in different conditions of gas
atmosphere and temperature.
Introduction
Solvents have a very large contribution to the environmental impact of chemical processes.1,2
Common highꢀboiling solvents, in particular Nꢀmethylꢀ2ꢀpyrrolidone (NMP), Nꢀethylꢀ2ꢀ
pyrrolidone (NEP), N,Nꢀdimethylformamide (DMF) and N,Nꢀdimethylacetamide (DMAc),
have toxic and carcinogenic properties and were recently included in the REACH ’Candidate
list of Substances of Very High Concern for Authorisation’.3 These solvents should therefore
be replaced by safer, less toxic and preferably biodegradable alternatives.2,4 Since the
physicochemical properties and the toxicity of solvents are highly related to their molecular
structure, subtle modifications in the structure may eventually result in solvents which are
more sustainable and show a similar or even improved performance in typical applications.
Recently, Nꢀalkylꢀ2ꢀpyrrolidones and 5ꢀmethylꢀNꢀalkylꢀ2ꢀpyrrolidones were proposed as
attractive alternatives to NMP, NEP, DMF and DMAc. In fact, Nꢀbutylꢀ2ꢀpyrrolidone, Nꢀ
isobutylꢀ2ꢀpyrrolidone, Nꢀtertꢀbutylꢀ2ꢀpyrrolidone, Nꢀpentylꢀ2ꢀpyrrolidone, Nꢀ(methylꢀ
substituted butyl)ꢀ2ꢀpyrrolidones, 5ꢀmethylꢀNꢀpropylꢀ2ꢀpyrrolidone and 5ꢀmethylꢀN-butylꢀ2ꢀ
pyrrolidone have been identified as nonꢀreprotoxic, nonꢀmutagenic and biodegradable, and
behave similarly in terms of polarity, vapor pressure and viscosity.5,6 These compounds can
be applied as solvents, surfactants and precursors for polymers, agrochemicals and
pharmaceuticals.5–8 For instance, Nꢀbutylꢀ2ꢀpyrrolidone is already commercially available and
1

Green Chemistry
Page 2 of 22
View Article Online
DOI: 10.1039/C7GC01829K
is successfully applied as a solvent in industrially relevant coupling reactions and polymer
syntheses. Although the current production method is based on fossil feedstocks, these
alkylpyrrolidones can be produced from biomassꢀderived platform molecules,9,10 which can
alleviate the environmental impact of a certain process.
The bioꢀbased synthesis of 5ꢀmethylꢀNꢀalkylꢀ2ꢀpyrrolidones is based on the reductive
amination of levulinic acid – or its ester derivatives – with amines. The synthesis involves the
hydrogenation of a Schiff base intermediate to a secondary amine, followed by an
intramolecular condensation at high temperature to generate the lactam moiety. The reaction
has been studied with homogeneous Ruꢀ, Irꢀ and Inꢀbased catalysts,11–14 heterogeneous
catalysts such as Au,15 Ni, Pd,16 Pt, Ru, Rh and Ir supported on carbon or metal oxides7,8,17–22
or even in the absence of any metal catalyst,23 thereby using either H2 or formic acid as a
reducing agent. The reductive amination of levulinic acid can be performed in various
solvents, or even under solventꢀfree conditions with a heterogeneous PtꢀMoOx/TiO2 catalyst.24
Succinic acid and glutamic acid are potential bioꢀbased precursors to 2ꢀpyrrolidone and NMP.
For instance, succinic acid can be converted to succinimide in the presence of ammonia, and
further to NMP by methylation using methanol over a Rh/ZrO2/C catalyst.25,26 Alternatively,
enzymatic decarboxylation of glutamic acid produces γꢀaminobutyric acid, which can be
converted into either 2ꢀpyrrolidone by an intramolecular condensation at high temperature, or
even into NMP in the presence of methanol and a halideꢀbased catalyst.27 We demonstrated
earlier that 2ꢀpyrrolidone can also be obtained successfully by the decarboxylation of
(pyro)glutamic acid using a Pd/Al2O3 catalyst in aqueous media under inert conditions.28
However, Nꢀalkylation of succinimide or 2ꢀpyrrolidone with other alcohols than methanol is
still inefficient26,27 and rather underexplored. This inspired us to develop a multiꢀstep
chemocatalytic system for the bioꢀbased synthesis of a broad range of bioꢀbased C3-C5 Nꢀ
alkylꢀ2ꢀpyrrolidones from renewable glutamic acid and simple carbonyl compounds.
Although not all synthesized Nꢀalkylꢀ2ꢀpyrrolidones are indicated as nonꢀtoxic – more
specifically Nꢀpropylꢀ2ꢀpyrrolidone, Nꢀisopropylꢀ2ꢀpyrrolidone and Nꢀ2ꢀbutylꢀ2ꢀpyrrolidone6
ꢀ they were synthesized for comparison and completeness. This approach does not rely on
petroleum based ammonia or alkylamines, because nitrogen can be recycled from a renewable
resource. The system operates in an aqueous medium and consists of (i) the Pdꢀcatalyzed
reductive Nꢀalkylation of glutamic acid with an aldehyde or a ketone to a monoꢀalkylated
derivative, (ii) the lactamization to Nꢀalkylpyroglutamic acid and (iii) the Pdꢀcatalyzed
decarboxylation to Nꢀalkylꢀ2ꢀpyrrolidone (Scheme 1).
2

Page 3 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Scheme 1 Synthesis of N-alkyl-2-pyrrolidones from glutamic acid and aldehydes/ketones: (i) Pd-catalyzed reductive N-
alkylation of glutamic acid to N-alkylglutamic acid, (ii) intramolecular condensation to N-alkylpyroglutamic acid at elevated
temperature and (iii) Pd-catalyzed decarboxylation to N-alkyl-2-pyrrolidone under inert atmosphere.
The first step typically proceeds by Pd/Cꢀcatalyzed hydrogenation under mild conditions29–31
and will be optimized for various aldehydes and ketones in terms of carbonyl
compound/glutamic acid molar ratio, H2 pressure, temperature and pH, thereby aiming at high
yields of the monoꢀalkylated product. The second step is accomplished by heating at > 120 °C
and is usually combined with the last step, which is catalyzed by Pd/Al2O3 and proceeds at
250 °C under inert N2 atmosphere.28 Parameters will be tuned to obtain the Nꢀalkylꢀ2ꢀ
pyrrolidones in high yield. The main focus will be on Nꢀbutylꢀ2ꢀpyrrolidone due to its high
relevance.5,6 Finally, a oneꢀpot process is designed, in which the reductive Nꢀalkylation and
the decarboxylation are facilitated by the same Pd/Al2O3 catalyst in a consecutive fashion.
Experimental
Catalyst synthesis
Supported metal catalysts, either commercial (Pd/C, Ru/C, Rh/C and Pd/BaSO4) or selfꢀ
prepared (Pd/Al2O3 (acidic), Pd/Al2O3 (basic) and Pd/SiO2), generally have a 5 wt% metal
loading. Nonꢀcommercial catalyst were synthesized by incipient wetness impregnation of
commercial supports with an aqueous solution of Pd(NH3)4Cl2·H2O. Impregnated supports
were dried at 60 °C overnight. The dry materials were granulated to particles with size 250ꢀ
500 µm, subsequently calcined at 500 °C (30 min, 2 °C minꢀ1, 100 ml minꢀ1 O2) and reduced
at 300 °C (1 h, 2 °C minꢀ1, 100 ml minꢀ1 H2) in a quartz Uꢀtube.
Reductive N-alkylation of amino acids and lactamization
Reactions were performed in a 50 ml highꢀpressure Parr batch reactor (Series 5500; Type SSꢀ
316; Model 4848 temperature controller).
3

Green Chemistry
Page 4 of 22
View Article Online
DOI: 10.1039/C7GC01829K
N-Butylamino acids. The reactor was charged with glutamic acid, monosodium glutamate or
glutamine (3.4 mmol), butyraldehyde (3.4 – 28 mmol), 5 wt% Pd/C (2.5 mol% Pd) and water
(20 ml). The pH was eventually adjusted to 5ꢀ6 with H3PO4. The reactor was sealed, purged 3
times with N2, 3 times with H2 and finally pressurized with 50 bar H2. The reactor was kept at
50 °C while stirring vigorously (830 rpm). After 6 h, the pressure was released and the solid
catalyst was removed by centrifugation. Evaporation of water, butanol (originating from the
hydrogenation of butyraldehyde) and unreacted butyraldehyde was achieved at 60 °C under
vacuum, and the solid residue was analyzed with 1HꢀNMR spectroscopy.
Other Nꢀalkylamino acids and monosodium Nꢀalkylglutamates were prepared via analogous
procedures, described in detail in the supporting information (ESI). Upscaled monosodium Nꢀ
alkylglutamates were used for subsequent lactamization to synthesize Nꢀalkylpyroglutamic
acids.
N-Alkylpyroglutamic acid. The crude monosodium Nꢀalkylglutamate was dissolved in water
(15 ml) and heated at 210 °C for 4 h under inert N2 atmosphere to ensure complete
conversion. Next, monosodium Nꢀalkylpyroglutamate was converted into the free acid form
by passing the aqueous solution through an ionꢀexchange column filled with the Amberlyst®
15 resin (hydrogen form), followed by purification with activated charcoal (SigmaꢀAldrich,
161551). Finally, water was evaporated at 100 °C under vacuum and the product was dried at
60 °C overnight and analyzed with 1HꢀNMR spectroscopy.
Decarboxylation of N-alkyl(pyro)glutamic acids
Decarboxylation reactions were performed in a 50 ml highꢀpressure Parr batch reactor (5500;
Type SSꢀ316; Model 4848 temperature controller) equipped with a Teflon liner. In a standard
reaction, the reactor was charged with Nꢀalkyl(pyro)glutamic acid (2 mmol), the catalyst
(metal/substrate = 4 mol%) and water (20 ml). The reactor was sealed, purged 6 times with N2
while stirring and finally pressurized with 6 bar N2. To study the effect of H2 on the
decarboxylation, the reactor was purged 3 times with N2, 3 times with H2 and finally
pressurized with the desired H2 pressure. Next, the reactor was heated at 250 °C and stirred
vigorously (830 rpm). After 6 h, heating was stopped and the reactor was cooled down to
room temperature in an ice bath. The pressure was released and the solid catalyst was
removed by centrifugation. The crude reaction mixture was analyzed with 1HꢀNMR
spectroscopy. This procedure was also applied to study the stability of Nꢀalkylꢀ2ꢀpyrrolidones.
4

Page 5 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Product analysis and identification
The synthesis of Nꢀalkyl(pyro)glutamates by reductive Nꢀalkylation of (pyro)glutamic acid
1
was evaluated in terms of yield and selectivity by HꢀNMR spectroscopy. To that end, 6 mg
of the solid product was dissolved in 600 µl D2O, or 300 µl of the crude reaction mixture was
1
mixed with 300 µl of D2O, and transferred to an NMR tube. HꢀNMR spectra were recorded
on a Bruker Avance 400 MHz spectrometer equipped with a BBO 5 mm atma probe and a
sample case. When water was present, the broad signal around δ = 4.7 ppm was suppressed by
applying a zgpr pulse program: p1 9.75 ꢁs; plw1 15W; plw9 5.7ꢀ05W; o1P on the resonance
signal of water, determined and selected automatically; d1 5 s.
Reaction mixtures obtained by decarboxylation of Nꢀalkyl(pyro)glutamic acids were also
1
analyzed by HꢀNMR spectroscopy. The conversion and product yield were determined by
using DMSO as an external standard. NMR samples were prepared by mixing 300 µl of the
1
crude reaction mixture with DMSO in 300 µl D2O. HꢀNMR spectra were recorded on the
same spectrometer using an adapted zgpr pulse program where d1 was extended to 25 s.
Prolonged relaxation times were required for accurate quantitative analysis. Product
identification was also performed using gas chromatography coupled to a mass spectrometer
(GCꢀMS), using an Agilent 6890 GC, provided with an HPꢀ5 ms column and coupled to a
5973 MSD mass spectrometer. Data are provided in the Supporting Information (ESI).
Results and discussion
Reductive N-alkylation of glutamic acid
Reductive Nꢀalkylation offers a mild and efficient strategy to synthesize a broad range of Nꢀ
alkylated derivatives of glutamic acid by using various aldehydes and ketones as precursors
for the alkyl substituent. The reaction proceeds through a Schiff base intermediate, generated
by the condensation of glutamic acid and the carbonyl compound, which can be reduced by
metalꢀcatalyzed hydrogenation,30–32 borohydride reagents33–35 or
a
Leuckartꢀtype
reaction.23,34,36 In this study, we will focus on the metalꢀcatalyzed reductive Nꢀalkylation of
glutamic acid, because this approach will generate only water as a byꢀproduct, which is
preferred in terms of green chemistry (Scheme 2).
5

Green Chemistry
Page 6 of 22
View Article Online
DOI: 10.1039/C7GC01829K
Scheme 2 Pd-Catalyzed reductive N-alkylation of glutamic acid using aldehydes or ketones.
For instance, N,Nꢀdimethylglutamic acid can be produced in high yield by the Pdꢀcatalyzed
reductive Nꢀmethylation in water using an excess of formaldehyde.29 However, the aim here is
to direct the selectivity towards monoꢀalkylation rather than to diꢀalkylation, since otherwise
the subsequent lactamization to Nꢀalkylpyroglutamic acid will be inhibited. Although the
extent of Nꢀalkylation is generally guided by the steric hindrance of the amino acid side chain
and by the bulkiness of the alkyl substituent(s) in the carbonyl compound (except for methyl
and ethyl groups),34 other parameters such as reaction time, aldehyde/ketone concentration
and temperature may have an influence as well.31 Therefore, in an initial experiment the
synthesis of Nꢀbutylglutamic acid was attempted under similar conditions as for the reductive
Nꢀmethylation of glutamic acid.29 A 1:1ꢀmixture of butyraldehyde and glutamic acid in water
was contacted with 5 wt% Pd/C at 50 °C and 50 bar H2 (Figure 1). Under these mild
conditions the selectivity towards Nꢀbutylglutamic acid was > 99%, but the yield was only
12% after 6 h and did not increase at longer reaction times. Higher yields were obtained by
increasing the butyraldehyde/glutamic acid molar ratio; an optimum was reached around
6.5:1, where Nꢀbutylglutamic acid was obtained in 75% yield (> 99% selectivity). At even
higher butyraldehyde concentrations, however, phase separation was observed as a result of
solubility issues. Under these conditions, the conversion of glutamic acid dropped to 56% and
besides Nꢀbutylglutamic acid also N,Nꢀdibutylglutamic acid was obtained, in respectively
42% and 14% yield.
6

Page 7 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Figure 1 Reductive Nꢀbutylation of glutamic acid at various butyraldehyde concentrations. Conditions: glutamic acid (0.17
M
in water, 20 ml), 5 wt% Pd/C (2.5 mol% Pd), H2 (50 bar), 50 °C, 6 h. Legend: Nꢀbutylglutamic acid (ꢀ), N,Nꢀdibutylglutamic
acid (ꢁ). The conversion always corresponds to the sum of Nꢀbutylglutamic acid and N,Nꢀdibutylglutamic acid.
Next, the influence of pH variation was studied by comparing the reductive Nꢀbutylation of
glutamic acid, glutamine and monosodium glutamate (Figure 2). Indeed, a 0.17
M aqueous
solution containing glutamine or monosodium glutamate has a pH of 5.3 and 6.9 respectively,
whereas the pH of a saturated solution of glutamic acid in water is about 3.0. The monoꢀ
alkylated product was obtained in 91% yield (> 99% selectivity) from glutamine. For
monosodium glutamate the yield was slightly reduced to 86%, because the diꢀalkylated
product was also present in 14% yield. These observations are partially consistent with the
effect of the pH on Schiff base condensation, which proceeds optimally around pH 5. Slightly
acidic conditions are essential to increase the electrophilicity of the carbonyl group in
butyraldehyde, while rendering the amine functionality in glutamic acid sufficiently
nucleophilic, and also to facilitate the elimination of water in the end.37 In rather neutral
conditions, as for monosodium glutamate, the secondary amine in the monoꢀalkylated product
is also susceptible to reductive Nꢀalkylation, resulting in a decreased selectivity. When the
aqueous solution containing monosodium glutamate was acidified to pH 5.7, the yield of
monosodium Nꢀbutylglutamate could be improved to 92% (96% selectivity). Moreover, under
these optimized conditions the butyraldehyde/monosodium glutamate molar ratio could be
reduced to 1.9:1. The strong difference in Nꢀalkylation susceptibility between monosodium
glutamate and glutamic acid – especially at relatively low aldehyde concentrations – can also
be tuned in terms of reaction time and H2 pressure (vide infra). Finally, the synthesis of
monosodium Nꢀbutylglutamate was scaled up, starting from 2.5 g monosodium glutamate in
the same reaction volume. The butyraldehyde/monosodium glutamate ratio was slightly
7

Green Chemistry
Page 8 of 22
View Article Online
DOI: 10.1039/C7GC01829K
adjusted to maximize the yield of the monoꢀalkylated product: 94% at a 1.5:1 ratio after
adjusting the pH to 5.7 with H3PO4 and 87% at a 1.4:1 ratio without any pH adjustment
(Figures S9ꢀS10). The latter conditions were selected to facilitate workꢀup procedures and
further decarboxylation of these substrates.
Figure 2 Reductive Nꢀbutylation of glutamic acid, glutamine and monosodium glutamate (MSG). Conditions: amino acid
(0.17
M in water, 20 ml), butyraldehyde (5 equiv.), 5 wt% Pd/C (2.5 mol% Pd), H2 (50 bar), 50 °C, 6 h. Legend: Nꢀ
butylamino acid (blue) and N,Nꢀdibutylamino acid (red). The conversion always corresponds to the sum of Nꢀbutylamino
acid and N,Nꢀdibutylamino acid.
The scope of the reductive Nꢀalkylation of monosodium glutamate was extended towards
other C3ꢀC5 aldehydes and ketones (Scheme 2); the optimized results are shown in Table 1.
The reactivity of aldehydes (entries 1, 3, 5, 6, 8 and 9) was much higher than that of ketones
(entries 2, 4 and 7), which can be attributed to the steric hindrance and the reduced
electrophilic character of the carbonyl group in ketones. Nevertheless, the monoꢀalkylated
products could be obtained in > 85% yield by increasing the ketone/monosodium glutamate
molar ratio to ≥ 5:1 and performing the reaction at 70 °C in the presence of acetic acid,
eventually at longer reaction times (entries 2 and 4; Figure S15). However, the reductive Nꢀ
alkylation with 2ꢀpentanone was impeded not only because of its low reactivity, but also by
its low solubility in water, resulting in phase separation (entry 7). In that case the Pd/C
catalyst prefers to reside in the organic phase, thereby favoring the hydrogenation of 2ꢀ
pentanone. In general, the selectivity towards monoꢀalkylation was > 99% for sterically
hindered carbonyl compounds like ketones (entries 2 and 4) and αꢀbranched aldehydes
(entries 5 and 9), whereas a mixture of monoꢀ and diꢀalkylated derivatives of glutamic acid
was obtained for linear and βꢀbranched aldehydes (entries 1, 3, 6 and 8).
8

Page 9 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Table 1 Reductive Nꢀalkylation of monosodium glutamate using various aldehydes and ketones at high H2 pressure (Scheme
2)a
Entry Substituent
(R)
Aldehyde/ketone
(equiv.)
Temperature Time Conversion Alkylation (%)
(°C)
RT
70
(h)
6
(%)
88
Mono
85
Di
3
1
2b
nꢀPropyl
Isopropyl
nꢀButyl
Propionaldehyde (2)
Acetone (5)
6
97
97
0
3
Butyraldehyde (1.4)
2ꢀButanone (6)
RT
70
6
95
87
8
4b
5
2ꢀButyl
24
6
86
86
0
Isobutyl
nꢀPentyl
2ꢀPentyl
Isopentyl
Isobutyraldehyde (3)
Valeraldehyde (2.8)
2ꢀPentanone (5)
Isovaleraldehyde (2.3)
RT
RT
70
> 99
> 99
0
6
7b
16
24
16
16
94
86
8
n.d.c
99
n.d.c
89
n.d.c
10
0
8
RT
9
a
2ꢀMethylbutyl 2ꢀMethylbutyraldehyde (3) RT
Conditions: monosodium glutamate (0.74
99
99
b
M
in water, 20 ml), 5 wt% Pd/C (2 mol% Pd), H2 (60 bar). The reaction was
performed in the presence of acetic acid (0.5 equiv.). c n.d.: not determined.
Previous experiments were performed under 50 bar H2 to increase the hydrogen solubility in
the aqueous medium and to boost the hydrogenation of the Schiff base intermediate; about 20
bar H2 was consumed during the reaction. However, hydrogenation of the carbonyl compound
to the corresponding alcohol was always observed as a competitive side reaction under these
conditions. The initial hydrogenation selectivity, reflecting the ratio Nꢀalkylated
products/alcohol, is likely to be different at lower H2 pressures. Therefore, the reductive Nꢀ
alkylation was studied for a 2:1ꢀmixture of butyraldehyde and glutamic acid at various H2
pressures (Figure 3). At 50 bar H2 the reaction was nearly complete within 10 min, but the
yield of Nꢀbutylglutamic acid was only 36%, which corresponds to a hydrogenation
selectivity of only 16%. However, after 2 h at 7.5 bar H2 the hydrogenation selectivity
increased to 76%; the conversion was 87% and Nꢀbutylglutamic acid and N,Nꢀdibutylglutamic
acid were obtained in 78% and 9% yield respectively. A further decrease in H2 pressure to 2.5
bar improved the hydrogenation selectivity up to 96%, despite a lower hydrogenation activity
and hence a higher tendency towards diꢀalkylation. The selectivity for the monoꢀalkylated
product could be increased at lower butyraldehyde concentrations, e.g. from 87% to 93% –
both at 80% conversion – by using a 1.5:1ꢀmixture of butyraldehyde and glutamic acid, but
full conversion of glutamic acid could not be reached. Both an excess of aldehyde and a mild
H2 pressure are thus essential to achieve a high conversion.
9

Green Chemistry
Page 10 of 22
View Article Online
DOI: 10.1039/C7GC01829K
Figure 3 Reductive Nꢀbutylation of glutamic acid at various H2 pressures. Conditions: glutamic acid (0.17
M
in water, 20 ml),
butyraldehyde (2 equiv.), 5 wt% Pd/C (2 mol% Pd), RT. H2 was supplied continuously in order to keep the pressure constant.
Legend: reaction at 2.5 bar (yellow), 7.5 bar (blue), 15 bar (orange), 30 bar (grey) and 50 bar (green).
Using the combination of an optimized H2 pressure (7.5 bar) and butyraldehyde/glutamic acid
molar ratio (2:1), glutamine and monosodium glutamate were also subjected to the reductive
Nꢀbutylation (Figure 4). Glutamine was again the most reactive compound, mainly as a result
of favorable pH conditions for Schiff base formation. In contrast to the observations after 6 h
at 50 bar H2 (Figure 2), the Nꢀalkylation of monosodium glutamate at lower pressure
proceeded more slowly than for glutamic acid. However, the hydrogenation selectivity after 2
h was only 76% for glutamic acid compared to 88% for monosodium glutamate and 92% for
glutamine (Figure 4). These observations demonstrate that the pH affects besides Schiff base
formation also the hydrogenation of butyraldehyde. This competitive side reaction seems to
be favored in rather acidic conditions and is even more pronounced at higher H2 pressures.
For instance, when the reductive Nꢀalkylation of monosodium glutamate and glutamic acid
with butyraldehyde was performed at 50 bar H2, the hydrogenation selectivity decreased to
58% and 16% respectively. Consequently, when the reaction was performed with an excess of
butyraldehyde, monosodium Nꢀbutylglutamate could be produced in 86% yield within 1 h,
whereas for glutamic acid the reaction ceased already after 10 min and Nꢀbutylglutamic acid
was obtained in only 38% yield (Figure S16).
10

Page 11 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Figure 4 Reductive Nꢀbutylation of glutamic acid, monosodium glutamate and glutamine. Conditions: substrate (0.17
M in
water, 20 ml), butyraldehyde (2 equiv.), 5 wt% Pd/C (2.5 mol% Pd), RT, H2 (7.5 bar). Legend: glutamic acid (orange),
monosodium glutamate (green) and glutamine (blue).
The scope in the reductive Nꢀalkylation of glutamic acid was also extended towards other C3ꢀ
C5 aldehydes and ketones and optimized at 7.5 bar H2 (Table 2). Similar trends were observed
as for monosodium glutamate at 50 bar H2 (Table 1). First, the lower reactivity of ketones
could be compensated by performing the reaction at 70 °C, but the addition of acetic acid was
not required. In this way, Nꢀisopropylglutamic acid and Nꢀ(2ꢀbutyl)glutamic acid were
obtained in > 99% yield after 20 h and 48 h respectively (Table 2, entries 2 and 4). Reaction
times were very long because Schiff base formation was slowed down under acidic
conditions. Secondly, a higher aldehyde/glutamic acid molar ratio had to be applied for
propionaldehyde than for butyraldehyde (entries 1 and 3), because the hydrogenation of both
the Schiff base and the aldehyde are affected by steric effects. In a 2:1ꢀmixture of the
aldehyde and glutamic acid, propionaldehyde and butyraldehyde were hydrogenated with
35% and 22% selectivity respectively. Thirdly, the reductive Nꢀalkylation of glutamic acid
was not possible with C5 aldehydes and ketones. As a result of the low solubility of both
glutamic acid and C5 carbonyl compounds in water, the mixture separates into three phases:
insoluble glutamic acid, an aqueous phase containing a small amount of glutamic acid and the
aldehyde/ketone, and an organic phase. For monosodium glutamate these solubility issues
were only observed for C5 ketones (Table 1, entry 7). Such issues were not encountered with
the other, lower molecular weight carbonyl compounds.
11

Green Chemistry
Page 12 of 22
View Article Online
DOI: 10.1039/C7GC01829K
Table 2 Reductive Nꢀalkylation of glutamic acid using various aldehydes and ketones at constant low H2 pressure (Scheme
2)a
Entry Substituent Aldehyde/ketone
Temperature Time Conversion Alkylation (%)
(R)
(equiv.)
Propionaldehyde (3) RT
(°C)
(h)
3
(%)
96
Mono
90
Di
6
1
2
3
4
5
nꢀPropyl
Isoproyl
nꢀButyl
2ꢀButyl
Isobutyl
Acetone (3)
70
RT
70
20
6
> 99
98
> 99
91
0
Butyraldehyde (2)
2ꢀButanone (3)
7
48
6
> 99
> 99
> 99
> 99
0
Isobutyraldehyde (4) RT
0
a Conditions: glutamic acid (0.17
M in water, 20 ml), 5 wt% Pd/C (2.5 mol% Pd), H2 (7.5 bar).
Decarboxylation of N-alkylglutamic acids
Decarboxylation is well described for the conversion of biomass constituents like fatty acids
to dieselꢀtype hydrocarbons in organic media. The reaction is catalyzed by noble metals,
mainly Pd and Pt supported on carbon and inorganic oxides, and proceeds under inert
atmosphere or in the presence of a minor H2 pressure to ensure catalyst stability.38–40
Recently, the metalꢀcatalyzed decarboxylation has also been reported for arylcarboxylic
acids,41 itaconic acid42,43 and amino acids28,44 in aqueous media. For instance, we identified
Pd/Al2O3 as a selective catalyst in the decarboxylation of (pyro)glutamic acid at 250 °C and
40ꢀ50 bar N2 and steam, resulting in 70% yield of 2ꢀpyrrolidone at 83% conversion after 6 h.
Therefore, in an initial attempt monosodium Nꢀbutylglutamate was exposed to the same
conditions. However, after 6 h the yield of Nꢀbutylꢀ2ꢀpyrrolidone was only 35% at 65%
conversion. The product mixture also contained 2ꢀpyrrolidone and propionic acid in 12% and
15% yield respectively, together with traces of γꢀbutyrolactone, γꢀhydroxybutyric acid,
butyric acid and methylamine. The decreased reaction rate compared to that of pyroglutamic
acid may be explained by the higher pH of an aqueous solution containing monosodium Nꢀ
butylglutamate. In fact, when the decarboxylation of unsubstituted pyroglutamic acid was
performed at pH 7, the conversion and selectivity to 2ꢀpyrrolidone decreased to 56% and 45%
after 6 h.28 However, when the decarboxylation of monosodium Nꢀbutylglutamate was
performed under acidic conditions, by adding a minor amount of H3PO4, the conversion was
only 5% after 17 h, which suggests that the catalytic activity is also influenced by inorganic
constituents in the reaction medium. Therefore, a pretreatment step was designed to remove
inorganic salts: monosodium Nꢀalkylglutamates were converted to Nꢀalkylpyroglutamic acids
by heating the aqueous solution at 210 °C, followed by ionꢀexchange chromatography using
an Amberlyst® 15 resin in the hydrogen form.
12

Page 13 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
Next, the Nꢀalkylpyroglutamic acids were contacted with Pd/Al2O3 under the same conditions
as monosodium Nꢀbutylglutamate (Table 3). The conversion of all these substrates was
increased to > 80% after 6 h, which clearly demonstrates the advantage of the pretreatment
step. Moreover, when monosodium glutamate was converted in the same way to pyroglutamic
acid, the decarboxylation resulted in 83% yield of 2ꢀpyrrolidone at 98% conversion after 4 h
(entry 1), which is a substantial improvement compared to the use of commercial
pyroglutamic acid.28 However, the decarboxylation of Nꢀalkylated pyroglutamic acids (entries
2ꢀ9) proceeded with lower selectivity, probably as a result of the consecutive degradation of
Nꢀalkylꢀ2ꢀpyrrolidones. The pathways towards the observed degradation products are
proposed in Scheme 3, based on kinetic experiments (Figure S17) and earlier work on the
degradation of 2ꢀpyrrolidone under these conditions.28
13

Green Chemistry
Page 14 of 22
View Article Online
DOI: 10.1039/C7GC01829K
Table 3 Pdꢀcatalyzed decarboxylation of pyroglutamic and various Nꢀalkylpyroglutamic acids in watera
Entry Product
Conversion Yield Yield
Yield
Yield
(%)
(%)
2-pyrrolidone propionic acid othersb
(%)
(%)
(%)
1c
2
3
4
5
6
7
8
9
98
83
54
61
63
58
66
57
57
64
/
13
2
99
16
15
15
16
9
13
17
13
19
11
15
15
7
16
6
99
> 99
99
9
6
> 99
90
13
3
15
11
8
88
5
82
3
b
a Conditions: Nꢀalkylpyroglutamic acid (0.1
M in water, 20 ml), 4.8 wt% Pd/Al2O3 (4 mol% Pd), N2 (6 bar), 250 °C, 6 h.
c
Others: γꢀbutyrolactone, γꢀhydroxybutyric acid, butyric acid and nonꢀidentified, lower molecular weight compounds.
Pyroglutamic acid was prepared from monosodium glutamate and the decarboxylation was performed during 4 h.
Degradation of an Nꢀalkylꢀ2ꢀpyrrolidone typically starts by hydrolysis to a 4ꢀ
(alkylamino)butyric acid (ii). Under inert atmosphere, this intermediate can be
dehydrogenated by the Pd catalyst on either side of the secondary amine (iii and vii), resulting
in Schiff bases which are readily hydrolyzed to an aldehyde/ketone and a primary amine such
as γꢀaminobutyric acid (iv), or a primary alkylamine (viii). The aldehydes are probably further
degraded by decarbonylation (vi), whereas ketones are expected to be more stable under these
1
conditions. Evidence was provided by HꢀNMR analysis of the reaction mixtures obtained by
decarboxylation of several Nꢀalkylpyroglutamic acids (Table 3): acetone and 2ꢀbutanone were
identified as side products from Nꢀisopropylpyroglutamic acid and Nꢀ(2ꢀbutyl)pyroglutamic
14

Page 15 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
acid respectively. Aldehydes were not detected for the other Nꢀalkylpyroglutamic acids in
Table 3, but substantial amounts of propionic acid and 2ꢀpyrrolidone were present in the
reaction medium. These compounds are produced by decarbonylation of 4ꢀoxobutyric acid
(ix) and by condensation of γꢀaminobutyric acid (v), respectively. γꢀHydroxybutyric acid, γꢀ
butyrolactone, butyric acid and methylamine were identified as minor side products, which is
in accordance with our previous work.28
Scheme 3 Possible reaction pathways towards the observed side products in the decarboxylation of Nꢀalkylpyroglutamic
acids: (i) decarboxylation to Nꢀalkylꢀ2ꢀpyrrolidone, (ii) hydrolysis to 4ꢀ(alkylamino)butyric acid, (iii) dehydrogenation to 4ꢀ
(alkylideneamino)butyric acid, (iv) hydrolysis to γꢀaminobutyric acid and an aldehyde/ketone, (v) intramolecular
condensation to 2ꢀpyrrolidone, (vi) decarbonylation to an alkane, (vii) dehydrogenation to 4ꢀ(alkylimino)butyric acid, (viii)
hydrolysis to 4ꢀoxobutyric acid and a primary alkylamine, (ix) decarbonylation to propionic acid, (x) hydrogenation to γꢀ
hydroxybutyric acid, (xi) intramolecular condensation to γꢀbutyrolactone, (xii) CꢀO hydrogenolysis to butyric acid, (xiii)
dehydrogenation to an imine and (xiv) hydrolysis to an aldehyde/ketone.
NꢀAlkylation of pyroglutamic acid can affect both the hydrolysis and the subsequent
dehydrogenation as a result of steric and electronic effects. On the one hand, the electrophilic
character of the carbonyl group in the lactam ring is reduced by an electronꢀdonating alkyl
substituent, which should result in a decreased hydrolysis rate. On the other hand, bulky alkyl
substituents may obstruct the nucleophilic attack of water (less hydrolysis) and the adsorption
of the substrate onto the catalyst surface (less dehydrogenation). The latter was clearly
15

Green Chemistry
Page 16 of 22
View Article Online
DOI: 10.1039/C7GC01829K
demonstrated for pyroglutamic acid derivatives with a βꢀmethylated alkyl substituent: Nꢀ
isobutylpyroglutamic acid and Nꢀ(2ꢀmethylbutyl)pyroglutamic acid were decarboxylated to
the corresponding Nꢀalkylꢀ2ꢀpyrrolidones with 66% and 78% selectivity respectively (Table
3, entries 6 and 9), which is a much better result than for the pyroglutamic acids bearing linear
alkyl substituents (entries 2, 4 and 7). NꢀAlkylꢀ2ꢀpyrrolidones are more stable compared to 2ꢀ
pyrrolidone under typical reaction conditions (Table 4). However, the stability of Nꢀalkylꢀ2ꢀ
pyrrolidones does not correlate with the selectivity to these compounds in the decarboxylation
of corresponding Nꢀalkylpyroglutamic acids. This observation might rather be explained in
terms of the reactivity of the substrate, which decreases in the order pyroglutamic acid > C3 ≈
C4 > C5 Nꢀalkylpyroglutamic acid. For instance, the conversion of pyroglutamic acid was
nearly complete within 4 h, whereas a C5 Nꢀalkylpyroglutamic acid did not even reach full
conversion after 6 h. An intermediate reactivity was observed for C3 and C4 Nꢀ
alkylpyroglutamic acids. Hydrolysis of Nꢀalkylꢀ2ꢀpyrrolidones is also dependent on the pH of
the reaction medium. At the onset of the reaction, the medium is highly acidic (pH 2ꢀ3) and
hence hydrolytic degradation will proceed significantly faster. The acidity will decrease
during the reaction (to a pH of about 5) as a result of decarboxylation and generation of basic
side products, and hence the selectivity tends to improve as the reaction proceeds. The
selectivity to Nꢀalkylꢀ2ꢀpyrrolidones in the decarboxylation of Nꢀalkylpyroglutamic acids
therefore depends on how fast the pH of the reaction medium increases, which in turn
depends on the reactivity of the substrate (Figure S4). As a result, one may expect a lower
selectivity for Nꢀalkylpyroglutamic acids compared to pyroglutamic acid.
Table 4 Stability of 2ꢀpyrrolidone and Nꢀbutylꢀ2ꢀpyrrolidone under typical conditions for Pdꢀcatalyzed decarboxylation of Nꢀ
alkylpyroglutamic acids in watera
Entry Substrate
Conversion Yield
Yield
Yield
(%)
2-pyrrolidone (%) propionic acid (%) othersb (%)
1
2
41
/
33
12
8
0
20
8
a
b
Conditions: substrate (0.1
M in water, 20 ml), 4.8 wt% Pd/Al2O3 (4 mol% Pd), N2 (6 bar), 250 °C, 6 h. Others: γꢀ
butyrolactone, γꢀhydroxybutyric acid, butyric acid and nonꢀidentified, lower molecular weight compounds.
When the pH was adjusted by the addition of 0.1
M NH3 to the reaction medium, the yield of
Nꢀbutylꢀ2ꢀpyrrolidone increased from 63% to 70% at 96% conversion (Table 5, entries 1 and
3). A similar effect was observed for Nꢀisobutylpyroglutamic acid, with eventually 82% yield
16

Page 17 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
of Nꢀisobutylꢀ2ꢀpyrrolidone at 97% conversion (entry 6). These results support the hypothesis
that the reaction medium is actually too acidic at the onset of the decarboxylation, and that
degradation will be more pronounced for Nꢀalkylpyroglutamic acids with lower reactivity.
Neutralization of Nꢀalkyl(pyro)glutamic acids by ammonia results in a substantial
improvement of the overall yield and selectivity. Adding an excess of base, however,
decreases the decarboxylation rate and even facilitates hydrolytic degradation of Nꢀalkylꢀ2ꢀ
pyrrolidones under basic conditions. This effect was demonstrated previously by the addition
of NaOH or using a catalyst with a basic support in the decarboxylation of pyroglutamic
acid.28
Table 5 Variation of pH in the Pdꢀcatalyzed decarboxylation of Nꢀbutylpyroglutamic acid and Nꢀisobutylpyroglutamic acid
in watera
Entry Substrate Additive
Conversion Yield
Yield
Yield
Yield
(%)
N(i)BP (%) 2-pyrrolidone propionic acid othersb (%)
(%)
15
9
(%)
13
11
14
15
8
1
2
3
4
5
NBPG
NiBPG
NBPG
NBPG
NiBPG
/
> 99
> 99
96
63
66
70
68
82
8
/
13
4
0.1
M
NH3
8
0.15
0.1
M
NH3 93
10
4
0
M
NH3 97
3
a Conditions: substrate (0.1
M in water, 20 ml), additive, 4.8 wt% Pd/Al2O3 (4 mol% Pd), N2 (6 bar), 250 °C, 6 h. NBPG: Nꢀ
butylpyroglutamic acid; NBP: Nꢀbutylꢀ2ꢀpyrrolidone; NiBPG: Nꢀisobutylpyroglutamic acid; NiBP: Nꢀisobutylꢀ2ꢀpyrrolidone.
b Others: γꢀbutyrolactone, γꢀhydroxybutyric acid, butyric acid and nonꢀidentified, lower molecular weight compounds.
Catalyst screening
Besides the pH, also the catalyst might affect the selectivity to Nꢀalkylꢀ2ꢀpyrrolidones.
Supports with strong acidic or basic properties might induce undesired hydrolysis of Nꢀalkylꢀ
2ꢀpyrrolidone,28 whereas the metal may be involved in further dehydrogenation. Therefore,
selected Pdꢀcatalysts with neutral, slightly basic and acidic supports and also other
commercial metal catalysts were screened on their activity in the degradation of Nꢀbutylꢀ2ꢀ
pyrrolidone at 250 °C (Table 6). Pd is the most promising metal, because Rh strongly
promoted the decarbonylation to propionic acid and Ru appeared to be a highly active
reforming catalyst (entries 1ꢀ3). Among the Pdꢀbased materials, the degradation of Nꢀbutylꢀ2ꢀ
pyrrolidone was less pronounced for catalysts with an inorganic support (entries 4ꢀ7 vs entry
1). However, the rather neutral Pd/SiO2 and Pd/BaSO4 catalysts showed a very low activity in
the decarboxylation of Nꢀbutylpyroglutamic acid (entries 10 and 11), which can be attributed
to a low Pd dispersion.28 In fact, HAADFꢀSTEM imaging was performed in previous
17

Green Chemistry
Page 18 of 22
View Article Online
DOI: 10.1039/C7GC01829K
research28 on the Pd/Al2O3 (acidic) and Pd/SiO2 catalysts in order to verify if particle size
might influence reaction selectivity or conversion. The number average particle diameter of
Pd clusters on Pd/SiO2 (8.8 nm) was clearly larger than on Pd/Al2O3 (acidic) (3.7 nm) and this
could explain the substantially lower conversion obtained with Pd/SiO2 (ESI). Higher
activities were observed for Pd supported on slightly acidic and basic Al2O3 (entries 8 and 9)
and > 80% of Nꢀbutylꢀ2ꢀpyrrolidone was preserved after 6 h at 250 °C in the presence of these
materials (entries 4 and 5), while this was only 56% with Pd/C (entry 1). Therefore Pd/Al2O3
is the most active and selective catalyst in the decarboxylation of Nꢀbutylpyroglutamic acid,
which is in accordance with earlier work on pyroglutamic acid; the slightly acidic and basic
properties of respectively both Al2O3 supports minimize lactam hydrolysis, hence, side
product formation.28
Table 6 Catalyst screening on the stability of Nꢀbutylꢀ2ꢀpyrrolidone and the metalꢀcatalyzed decarboxylation of Nꢀ
butylpyroglutamic acid in watera
Entry Substrateb Catalystc Conversion Yield
Yield
Yield
Yield
(%)
NBP (%)
2-pyrrolidone propionic acid othersd (%)
(%)
19
< 1
9
(%)
7
1
2
3
4
NBP
NBP
NBP
NBP
Pd/C
44
/
/
/
/
18
> 99
8
Ru/C
> 99
56
< 1
39
12
Rh/C
Pd/Al2O3
(acidic)
Pd/Al2O3
(basic)
Pd/SiO2
20
8
0
5
NBP
17
18
/
5
10
2
6
7
8
NBP
/
9
4
5
3
9
NBP
Pd/BaSO4 15
/
9
3
NBPG
Pd/Al2O3
(acidic)
Pd/Al2O3
(basic)
> 99
63
15
13
9
NBPG
88
32
59
16
13
0
10
NBPG
NBPG
Pd/SiO2
19
17
5
3
3
3
5
6
11
Pd/BaSO4 29
a
b
Conditions: substrate (0.1
M in water, 20 ml), catalyst (4 mol% metal), N2 (6 bar), 250 °C, 6 h. NBP: Nꢀbutylꢀ2ꢀ
pyrrolidone; NBPG: Nꢀbutylpyroglutamic acid. c Commercially obtained catalysts: 5 wt% Pd/C, 5 wt% Ru/C, 5 wt% Rh/C
d
and 5 wt% Pd/BaSO4; selfꢀprepared catalysts: 4.8 wt% Pd/Al2O3 (acidic), 4.7 wt% Pd/Al2O3 (basic), and 5.1 wt% Pd/SiO2
Others: γꢀbutyrolactone, γꢀhydroxybutyric acid, butyric acid and nonꢀidentified, lower molecular weight compounds.
One-pot synthesis
In a final approach the reductive Nꢀalkylation of glutamic acid, the thermal cyclization to Nꢀ
alkylpyroglutamic acid and the decarboxylation to Nꢀalkylꢀ2ꢀpyrrolidones were combined in a
18

Page 19 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
oneꢀpot process; both steps were performed in a consecutive fashion by simply changing the
gas atmosphere and temperature. Pd/Al2O3 was selected over Pd/C as a catalyst for both steps,
because the latter appeared to be less performant in the decarboxylation step (Table 6). In a
preliminary experiment the catalytic activity of Pd/Al2O3 in the reductive Nꢀalkylation of a
2:1ꢀmixture of butyraldehyde and glutamic acid was evaluated at room temperature under 7.5
bar H2. After 6 h, Nꢀbutylglutamic acid and N,Nꢀdibutylglutamic acid were obtained in 91%
and 9% yield, which is almost the same result as for Pd/C (Table 2, entry 3). Consequently,
the catalyst support does not play an important role in the reductive Nꢀalkylation. Next, the
catalyst was removed, H2O, butanol and unreacted butyraldehyde were evaporated and the
crude solid product containing Nꢀbutylated glutamic acids was exposed to the conditions for
decarboxylation with fresh Pd/Al2O3. NꢀButylꢀ2ꢀpyrrolidone was obtained in 65% yield after
6 h, together with 2ꢀpyrrolidone and propionic acid in 18% and 10% yield at 94% conversion.
Again, this result is consistent with the decarboxylation of Nꢀbutylpyroglutamic acid derived
from monosodium glutamate (Table 3, entry 4). Interesting to note is that the Al2O3 might
also catalyze the intramolecular cyclocondensation to Nꢀbutylpyroglutamic acid,45 though at
250 °C the contribution will be relatively small compared to the spontaneous
cyclocondensation.46 Finally, in the oneꢀpot reaction the reductive Nꢀalkylation of glutamic
acid was performed with 2 equivalents of butyraldehyde at room temperature under 7.5 bar H2
for 6 h, and then the gas atmosphere was changed to 6 bar N2 and the temperature was
increased to 250 °C. By prolonging the decarboxylation reaction time to 16 h, Nꢀbutylꢀ2ꢀ
pyrrolidone was obtained in 65% yield, together with 15% yield of 2ꢀpyrrolidone and 10%
yield of propionic acid at 92% conversion.
Conclusions
A broad range of bioꢀbased Nꢀalkylpyrrolidones was successfully synthesized in a multiꢀstep
chemocatalytic approach using glutamic acid and C3ꢀC5 aldehydes and ketones. The Pdꢀ
catalyzed reductive Nꢀalkylation of renewable glutamic acid proceeds under mild conditions.
Both a low H2 pressure and a rather neutral pH are the key elements to obtain the monoꢀ
alkylated product in high yield. The intramolecular condensation to the corresponding Nꢀ
alkylpyroglutamic acid proceeds readily at high temperature in the absence of catalyst.
Pd/Al2O3 was identified as the most active and selective catalyst for decarboxylation of Nꢀ
alkylpyroglutamic acids at 250 °C under inert atmosphere. The product selectivity in the last
step was mainly determined by the consecutive hydrolysis of Nꢀalkylꢀ2ꢀpyrrolidone under
acidic conditions, and therefore also by the intrinsic reactivity of the substrate. The yield
19

Green Chemistry
Page 20 of 22
View Article Online
DOI: 10.1039/C7GC01829K
could be improved by neutralization of Nꢀalkylpyroglutamic acid with ammonia, resulting in
70% yield of Nꢀbutylꢀ2ꢀpyrrolidone and 82% yield of Nꢀisobutylꢀ2ꢀpyrrolidone. Monosodium
glutamate can also be used as substrate, but the alkylated derivative should be purified from
inorganic salts to achieve high catalytic activity in de the decarboxylation step. Finally, a twoꢀ
step oneꢀpot process was developed, in which the reductive Nꢀalkylation and decarboxylation
occur in a consecutive fashion; Nꢀbutylꢀ2ꢀpyrrolidone was obtained in 65% yield after 16 h.
Reductive Nꢀalkylation and decarboxylation of glutamic acid using simple carbonyl
compounds offers a new route towards bioꢀbased 2ꢀpyrrolidone derivatives compared to
conventional synthesis methods, and allows to recycle nitrogen from a renewable resource.
Acknowledgements
F. D. S. is grateful to the Agency for Innovation by Science and Technology in Flanders
(IWTꢀVlaanderen) for a doctoral fellowship. L. C. thanks Flanders Innovation &
Entrepreneurship for a postꢀdoctoral grant. D. E. D. V. is grateful to KU Leuven for longꢀterm
structural Methusalem funding, to Belspo (IAPꢀPAI 7/05) and to FWO for research project
funding. We also thank Karel Duerinckx for assistance with NMR measurements and Birgit
Claes for assistance with ICPꢀAES measurements.
References
1
D. Prat, A. Wells, J. Hayler, H. Sneddon, C. R. Mcelroy, S. Abouꢀshehada and P. J. Dunn, Green Chem.,
2016, 18, 288–296.
2
J. H. Clark, T. J. Farmer, A. J. Hunt and J. Sherwood, Int. J. Mol. Sci., 2015, 16, 17101–17159.
ECHA, REACH ‘Candidate List of substances of very high concern for Authorisation’, 2017.
Y. Gu and F. Jérôme, Chem Soc Rev, 2013, 42, 9550–9570.
3
4
5
J. Sherwood, H. L. Parker, K. Moonen, T. J. Farmer and A. J. Hunt, Green Chem., 2016, 18, 3990–3996.
WO 2013107822A1, 2013.
6
7
US 6818593 B2, 2004.
8
US 6855731 B2, 2005.
9
P. T. Anastas and M. M. Kirchhoff, Acc. Chem. Res., 2002, 35, 686–694.
J. J. Bozell and G. R. Petersen, 2010, 12.
10
11
12
13
14
15
Y. Wei, C. Wang, X. Jiang, D. Xue, J. Li and J. Xiao, Chem. Commun., 2013, 49, 5408–5410.
C. Ortizꢀcervantes, M. Floresꢀalamo and J. J. García, Tetrahedron Lett., 2016, 57, 766–771.
Y. Huang, J. Dai, X. Deng, Y. Qu, Q. Guo and Y. Fu, ChemSusChem, 2011, 4, 1578–1581.
Y. Ogiwara, T. Uchiyama and N. Sakai, Angew. Chemie - Int. Ed., 2016, 55, 1864–1867.
X. Du, L. He, S. Zhao, Y. Liu, Y. Cao, H. He and K. Fan, Angew. Chemie - Int. Ed., 2011, 50, 7815–
20

Page 21 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01829K
7819.
J. Zhang, B. Xie, L. Wang, X. Yi and C. Wang, ChemCatChem, 2016, 8, 1–8.
16
17
J. D. Vidal, M. J. Climent, P. Concepcion, A. Corma, S. Iborra and M. J. Sabater, ACS Catal., 2015, 5,
5812–5821.
18
19
20
21
22
23
24
25
26
27
J. D. Vidal, M. J. Climent, A. Corma, P. Concepcion and S. Iborra, ChemSusChem, 2016, 10, 119–128.
US 2006/0247444 A1, 2006.
US 7129362B2, 2006.
US 4560760, 1985.
US 3235562, 1966.
A. Ledoux, L. S. Kuigwa, E. Framery and B. Andrioletti, Green Chem., 2015, 17, 3251–3254.
A. S. Touchy, S. M. A. H. Siddiki, K. Kon and K. Shimizu, ACS Catal., 2014, 4, 3045–3050.
US 4814464, 1989.
US 6670483, 2003.
T. M. Lammens, M. C. R. Franssen, E. L. Scott and J. P. M. Sanders, Green Chem., 2010, 12, 1430–
1436.
28
F. De Schouwer, L. Claes, N. Claes, S. Bals, J. Degrève and D. E. De Vos, Green Chem., 2015, 17,
2263–2270.
29
30
31
32
33
F. De Schouwer, T. Cuypers, L. Claes and D. E. De Vos, Green Chem., 2017, 19, 1866–1876.
W. R. Bowman, J. Chem. Soc., 1950, 1342–1345.
W. R. Bowman, J. Chem. Soc., 1950, 1346–1349.
W. R. Bowman and D. R. Coghlan, Tetrahedron, 1997, 53, 15787–15798.
E. W. Baxter and A. B. Reitz, Reductive aminations of carbonyl compounds with borohydride and
borane reducing agents, 2002.
34
35
36
37
38
L. Aurelio, R. T. C. Brownlee and A. B. Hughes, Chem. Rev., 2004, 104, 5823–5846.
G. Verardo, P. Geatti, E. Pol and A. G. Giumanini, Can. J. Chem., 2002, 80, 779–788.
H. T. Clarke, H. B. Gillespie and S. Z. Weisshaus, J. Am. Chem. Soc., 1933, 55, 4571–4587.
F. A. Carey, Organic Chemistry 4th edition, 2002.
M. Snåre, I. Kubičková, P. MäkiꢀArvela, K. Eränen and D. Y. Murzin, Ind. Eng. Chem. Res., 2006, 45,
5708–5715.
39
40
J. P. Ford, J. G. Immer and H. H. Lamb, Top Catal, 2012, 55, 175–184.
R. W. Gosselink, S. A. W. Hollak, S. Chang, J. Van Haveren, K. P. De Jong, J. H. Bitter and D. S. Van
Es, ChemSusChem, 2013, 6, 1576–1594.
41
42
S. Matsubara, Y. Yokota and K. Oshima, Org. Lett., 2004, 6, 2071–2073.
J. Le Nôtre, S. C. M. Witteꢀvan Dijk, J. van Haveren, E. L. Scott and J. P. M. Sanders, ChemSusChem,
2014, 7, 2712–2720.
43
44
J. Verduyckt and D. E. De Vos, Chem. Sci., 2017, 8, 2616–2620.
J. Verduyckt, M. Van Hoof, F. De Schouwer, M. Wolberg, M. Kurttepeli, P. Eloy, E. M. Gaigneaux, S.
Bals, C. E. A. Kirschhock and D. E. De Vos, ACS Catal., 2016, 6, 7303–7310.
45
A. BladéꢀFont, Tetrahedron Lett., 1980, 21, 2443–2446.
21

Green Chemistry
Page 22 of 22
View Article Online
DOI: 10.1039/C7GC01829K
46
Y. Teng, E. L. Scott and J. P. M. Sanders, J. Chem. Technol. Biotechnol., 2012, 87, 1458–1465.
22