A highly sustainable route to pyrrolidone derivatives - Direct access to biosourced solvents

Article abstract of DOI:10.1039/c5gc00417a

Access to a series of 5-methylpyrrolidone derivatives is described directly using the biosourced levulinic acid in the absence of any additive, catalyst or solvent. The highly selective reaction proceeds with an E-factor as low as 0.2. Products are recovered in very good yields after a simple distillation.

Full text of DOI:10.1039/c5gc00417a

Green Chemistry
View Article Online
View Journal
COMMUNICATION
A highly sustainable route to pyrrolidone
derivatives – direct access to biosourced solvents†
Cite this: DOI: 10.1039/c5gc00417a
Received 20th February 2015,
Accepted 27th March 2015
A. Ledoux, L. Sandjong Kuigwa, E. Framery and B. Andrioletti*
DOI: 10.1039/c5gc00417a
www.rsc.org/greenchem
Access to a series of 5-methylpyrrolidone derivatives is described (EChA) and that is under discussion for restriction in its use.10
directly using the biosourced levulinic acid in the absence of any Of note is that NMP is a mild-volatile and thermally stable
additive, catalyst or solvent. The highly selective reaction proceeds chemical, widely used in industries as a solvent for appli-
with an E-factor as low as 0.2. Products are recovered in very good cations such as the synthesis of plastic and resin polymers,
yields after a simple distillation.
extraction of aromatics in oil processing, as a cleaning agent of
silicon wafers, etc. Thus, its production is estimated to be
200 000–250 000 tons per year.11 Accordingly, considering the
vital relevance of NMP in numerous industrial applications,
finding alternatives to NMP is a priority. Among the potential
alternatives for NMP, the use of fuel-based solvents such as
DMSO (dimethylsulfoxide), NEP (N-ethylpyrrolidone), DMF (di-
methylformamide) or DMAc (dimethylacetamide) was pro-
posed. However, they do not display performances comparable
to NMP and several of them are carcinogenic. Yet, new solvents
are still highly desired.
N-Substituted-5-MeP can be considered as the closest ana-
logue of NMP available from biomass. The synthesis of
5-methyl-pyrrolidones from LA was first reported using hetero-
geneous catalysis and dihydrogen as the reducing agent by
Shilling,12 Crook,13 and more recently by Manzer et al.14 Thus,
Manzer et al. describe the need for different transition-metals
such as Ni, Cu, Rh, Ru, Ir or Pt, grafted on silica, alumina or
carbon as potential heterogeneous catalysts. Later on, Huang
et al.9 reported a synthesis of N-alkyl-5-methylpyrrolidones
involving LA and primary amines using the [Ru]-catalyzed
decomposition of formic acid (FA). In addition to the use of an
expensive metal catalyst, air-sensitive electron-rich phosphine-
ligands such as P(t-Bu)3 or P(Cy)3 were also required for ensur-
ing a good conversion. Very recently, a procedure requiring
several equivalents of DMSO as an additive and a tertiary
amine was reported.8b The DMSO/tertiary amine system was
described as an activating combination for promoting the
decomposition of FA and was proposed as an alternative to
metal-based catalysis. However, this procedure requires critical
purification steps in order to remove all additives, solvents
as well as the unreacted starting material. Comparison of
E-factors15 for some of these reported procedures shows that
catalysis is preferable over the use of a large amount of addi-
tives in terms of wastes and environmental footprint (Table 1).
The replacement of toxic, non-renewable fossil solvents is one
of the main scientific challenges both in industry and acade-
mia. As an example, the massive use of solvent explains the
poor E-factor of the fine or pharmaceutical industry and some
of the most common solvents are now part of the REACH1
“Substances of Very High Concern” list (i.e. NMP, DMF, DMAc,
etc.).2 Thus, the development of environmentally benign
alternatives has become a priority.3 The use of platform mole-
cules available from biomass constitutes a very promising solu-
tion as it opens a new field of enormous potential impact for
the chemical supply chain. In addition, the development of
new bio-based solvents constitutes an opportunity for investi-
gating new routes for the development of chemicals displaying
better toxicological or ecotoxicological profiles.
Levulinic acid (LA) is among the most promising chemicals
produced from biomass feedstock.4 It is produced generally by
the acidic hydrolysis of carbohydrates.5 LA is a platform mole-
cule that can be used for numerous chemical applications
such as the synthesis of γ-valerolactone (GVL) already used as
a fuel additive or solvent.6 LA can also be converted to N-sub-
stituted-5-methyl-pyrrolidones (N-substituted-5-MeP) by reduc-
tive amination and cyclisation.7–9 N-substituted-5-MeP are
structurally related to the well-known N-methylpyrrolidone
(NMP) that was recently implemented in the “Substance of
Very High Concern” list by the European Chemical Agency
ICBMS-UMR 5246 Université Claude Bernard-Lyon 1, Equipe de CAtalyse, SYnthèse
et ENvironnement (CASYEN), Domaine Scientifique de la Doua-Bât. Curien/CPE, 2°
étage aile C, 43 Boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex, France.
E-mail: bruno.andrioletti@univ-lyon1.fr
†Electronic supplementary information (ESI) available: Detailed synthetic pro-
cedure and spectral characterization of the synthesized pyrrolidones. See DOI:
10.1039/c5gc00417a
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Communication
Green Chemistry
Table 1 Comparison of the E-factors from common procedures for the
synthesis of 5-methyl-N-cyclohexylpyrrolidone
E-factor (kg waste
per kg product)
Entry
Promoter
Reference
1
2
3
Ruthenium catalysis
Activating agents
Heterogeneous platinum
catalysis
90
199
66
9
8
7
4
None
0.2
This work
Although the recent developments allow the synthesis of a
wide range of N-substituted-5MeP derivatives in reasonable to
very good yields, the use of metal-catalysts, activating agents
and solvents as well as purification steps still constitute strong
limitations in terms of costs and environmental footprint.
Herein, we report a new and simple process for the synthesis
of N-substituted-5-MeP derivatives which overcomes most of
these above-mentioned limitations, hence affording a very
advantageous E-factor value.16
Scheme 1 Leuckart–Wallach mechanism and formation of the form-
amide byproduct.
Within the frame of our efforts to develop an environ-
mentally friendly, industrially applicable synthesis of the 5-methyl-
N-propylpyrrolidone (5Me-NPP) 2a, we re-investigated the Ru-
catalyzed approach by Huang et al.9 Aiming at developing an
air and moisture tolerant catalyst, we considered first the com-
bination of [Ru(p-cymene)Cl2]2 and (o-tolyl)3P as a catalytic
system for promoting the reductive amination of LA by N-propyl-
amine in the presence of FA (Table 2).
Scheme 2 Limitation of the formamide formation under autogenous
pressure.
Interestingly, under argon at normal pressure and in the
absence of a solvent, we discovered that decreasing strongly
the catalyst loading did not influence drastically the outcome
of the reaction. Indeed, 2a was still formed in decent amounts
even in the presence of 0.05 mol% catalyst (Table 2, entries
1–3). More interestingly, when the reaction was carried out in
the absence of a catalyst17, 2a was still obtained in 56% yield
along with the formamide 3a and H2O (Table 2, entry 4). Actu-
ally, this result is consistent with the well-known Leuckart–
Wallach (LW) type mechanism (Scheme 1).18
tion, initially formed in situ from LA and propylamine. The
synthesis of formamides through this process is known to be
reversible in the presence of water.19 However, increasing the
reaction time had no effect on the formation of 2a or on the
hydrolysis of 3a. Interestingly, this limitation was overcome by
performing the reaction in an autoclave (Scheme 2).
Indeed, in an isochoric system, the production of CO2 and
H2 from the decomposition of FA allowed production of the
desired pyrrolidone 2a as the only product. A maximum
pressure of 25 bars after 4.5 h was obtained. Interestingly, a
pressure monitor can be used for determining the progress of
the reaction assuming that CO2 follows the ideal gas law (H2 is
consumed for the reduction of the imine function). Indeed the
quantity of CO2 released at full conversion corresponds to the
theoretical pressure calculated from the decomposition of FA
(Fig. 1). As predicted, 2a was obtained quantitatively in 4.2 h at
25 bar for a temperature set at 160 °C20 (for details of calcu-
lations see the ESI†). Using these conditions, 2a was isolated
as an almost pure compound. A simple distillation afforded
pure 2a as a colourless liquid in 84% isolated yield.
As depicted in Scheme 1, formic acid is formally converted
to CO2 and “H2” through the hydrogenation of the imine func-
Table 2 Effect of the catalyst loadinga
Entry
[Ru] mol%
Conversionb (%)
2a : 3a
Furthermore, by setting parameters according to the ideal
gas law, the reaction could be performed at higher pressure
and temperature, thus shortening the reaction time.21 (see the
ESI†).
Various N-substituted-5MeP were synthesised from LA, FA
and several amines (Table 3). Preferentially, we chose aliphatic
1
2
3
4
1
86
72
58
56
1 : 0
4 : 1
6 : 4
0.5
0.05
0
11 : 9
a The ratio LA/FA/1a is 1 : 1 : 1. b Determined by 1H NMR.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Communication
corresponding pyrrolidone (5MeNtBP was not reported even in
the presence of a catalyst or activating agents),8,9 but interestingly
1f afforded 2f in 88% isolated yield (Table 3, entries 5 and 6).
Because of its higher viscosity, n-octylamine 1g also displayed
a lower reactivity (conv. 86%). However, a remarkable 80%
yield was obtained after purification. Of note is that due to
their higher boiling points, 2g and 2h were isolated by liquid–
liquid extraction using EtOAc and a saturated NH4Cl solution
(Table 3, entries 7 and 8). However, 2h was isolated in a mod-
erate 49% yield because of its partial solubility in water.
Conclusions
In conclusion, we have developed a very efficient and environ-
mentally friendly procedure to synthesize and produce N-sub-
stituted-5-methylpyrrolidone derivatives. Advantageously, our
methodology does not require any metal catalyst, additive or
special care (anhydrous or oxygen free conditions) for
affording the expected pyrrolidones in a very efficient way. In
addition, monitoring the evolution of the pressure over time
allows an easy follow up of the reaction. Interestingly, as mod-
erate pressures are used, this methodology could certainly be
developed at the industrial scale without any significant cost
increase. Accordingly, a series of N-substituted pyrrolidones
was prepared. These unprecedented results clearly demon-
strate the efficiency of our approach that allows the synthesis
of potential NMP solvent substitutes from the biosourced levu-
linic acid with an exceptional E-factor of 0.2.
Fig. 1 Evolution of the temperature and self-generated pressure for the
synthesis of 2a. The ratio LA/FA/1a is 1 : 1 : 1. The maximal pressure of 25
bar corresponds to 100% conversion of FA at 160 °C.
Table 3 Scope and limitation of the reactiona
Conversionb
(%)
Yieldc
(%)
Entry
Amine R=
Product
General experimental procedure
1
2
3
4
n-Propyl (1a)
n-Butyl (1b)
i-Propyl (1c)
i-Butyl (1d)
t-Butyl (1e)
Cyclohexyl (1f)
n-Octyl (1g)
Benzyl (1h)
>99
>99
>99
>99
—
89
86
5MeNPP (2a)
5MeNBP (2b)
5MeNiPP (2c)
5MeNiBP (2d)
5MeNtBP (2e)
5MeNChP (2f)
5MeNOP (2g)
5MeNBzP (2h)
84
82
80
86
—
88
80
49
An autoclave equipped with a manometer, a safety valve and a
thermometer was charged with levulinic acid (1 equiv.). Next,
the amine (1 equiv.) was slowly added using a syringe followed
by formic acid (1 equiv., exothermic reaction). When all
reagents were added, the reactor was sealed and the reaction
runs for 3–10 h at 160–200 °C upon monitoring the pressure
and the temperature. Then, the reaction was allowed to cool to
room temperature before the reactor was degassed and slowly
opened. The crude pyrrolidone was collected as an orange to
brown liquid. Depending on the nature of the amine, the
crude product was washed with a dilute aqueous solution of
5
6
7d
8d
>99
a Conditions: 60–150 mmol scale, ratio LA/FA/1a–h: 1 : 1 : 1.
b Determined by 1H NMR. c Isolated yield after distillation. d Isolated
yield after extraction.
amines having a low boiling point and viscosity in order to Na2CO3, and extracted with AcOEt. The separated organic layer
obtain potential NMP alternatives for solvent applications. was washed with brine, and dried over anhydrous MgSO4, fil-
Using our optimized conditions, aliphatic (iso)propyl- and tered and concentrated under reduced pressure to afford the
(iso)butylamines 1a–d reacted with full conversion (Table 3, desired compounds. In the case of the volatile amine (bp <
entries 1–4). The resulting pyrrolidones 2a–d were purified by 100 °C), a distillation under pressure was carried out and
distillation under reduced pressure and isolated in 80–86% afforded a clear oil.
yield. The difference between the conversion and the yield is
attributable to some loss during the distillation process that
was realized on a laboratory scale (60–150 mmol). Cyclohexyl-
and t-butylamines 1e and 1f displayed different reactivities
Acknowledgements
depending on the steric hindrance of the amine. Indeed, using B.A. deeply acknowledges the French National Agency (ANR
our mild conditions, the t-butylamine did not afford the ANR-09-CP2D-12 for financial support.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Communication
Green Chemistry
9 Y.-B. Huang, J.-J. Dai, X.-J. Deng, Y.-C. Qu, Q.-X. Guo and
Y. Fu, ChemSusChem, 2011, 4, 1578–1581.
Notes and references
1 REACH = Registration, Evaluation and Authorization of 10 See http://echa.europa.eu/en/information-on-chemicals.
Chemicals Regulation.
11 A. L. Harreus in “2-pyrrolidones”, Ullmann’s Encyclopedia of
Industrial Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA,
2000, ch. 2.
2 See for instance F. Roschangar, R. A. Sheldon and
C. H. Senanayakea, Green Chem., 2015, 17, 752–768.
3 Y. Gu and F. Jérôme, Chem. Soc. Rev., 2013, 42, 9550–9570.
12 W. L. Shilling, US Pat, 32355562, 1996.
4 (a) B. V. Timokhin, V. A. Baransky and G. D. Eliseeva, Russ. 13 L. R. Crook, B. A. Jansen, K. E. Spencer and D. H. Watson,
Chem. Rev., 1999, 68, 73–84; (b) J. J. Bozell, L. Moens, GB Patent, 1036694, 1996.
D. C. Elliott, Y. Wang, G. G. Neuenscwander, S. W. Fitzpatrick, 14 (a) L. E. Manzer and F. E. Herkes, US Pat, 2004192933,
R. J. Bilski and J. L. Jarnefeld, Resour. Conserv. Recycl.,
2000, 28, 227–239; (c) J. J. Bozell and G. R. Peteresen, Green
Chem., 2010, 12, 538–554.
2003; (b) L. E. Manzer, US Patent, 6743819, 2004;
(c) L. E. Manzer, WO Patent, 2004084633, 2004;
(d) L. E. Manzer, US Patent, 6841520, 2005.
5 (a) L. Deng, J. Li, D.-M. Lai, Y. Fu and Q.-X. Guo, Angew. 15 (a) R. A. Sheldon, Chem. Ind., 1992, 903–906; (b) R. A. Sheldon,
Chem., Int. Ed., 2009, 48, 6529–6532; (b) P. Gallezot, Chem.
Soc. Rev., 2012, 41, 1538–1558; (c) J. J. Thomas and
http://www.sheldon.nl/roger/efactor.html; (c) R. A. Sheldon,
Green Chem., 2007, 9, 1273–1283.
G. R. Barile, Biomass Wastes, 1985, 8, 1461–1494; 16 See also B. Andrioletti, A. Ledoux, L. Sandjong Kuigwa and
(d) M. Kitano, F. Tanimoto and M. Okabayashi, Chem. Econ. E. Framery, FR Patent, 3003571, 2014.
Eng. Rev., 1975, 7, 25–29; (e) W. A. Farone and J. E. Cuzens, 17 Reactions were repeated in new, metal-free glassware and
WO, 9810986, 1998; (f) S. Fitzpatrick, WO, 8910362, 1990.
S. W. Fitzpatrick, WO, 9640609, 1997.
6 (a) H. Heeres, R. Handana, D. Chunai, C. Borromeus Ras-
reactors.
18 V. J. Webers and W. F. Bruce, J. Am. Chem. Soc., 1948, 70,
1422–1424.
rendra, B. Girisuta and H. Jan Heeres, Green Chem., 2009, 19 (a) S. Antonczak, M. F. Ruiz-Mpez and J. L. Rivail, J. Am.
11, 1247–1255; (b) G. W. Huber, S. Iborra and A. Corma,
Chem. Rev., 2006, 106, 4044–4098; (c) Y.-S. Yoon, H. Khil
Shin and B.-S. Kwak, Catal. Commun., 2002, 3, 349–355;
Chem. Soc., 1994, 116, 3912–3921; (b) L. Gorb, A. Asensio,
I. Tuñón and M. F. Ruiz-López, Chem. – Eur. J., 2005, 11,
6743–6753.
(d) M. Mehdi, V. Fábos, R. Tuba, A. Bodor, L. T. Mika and 20 Reaction conditions for a maximum pressure of 25 bar:
I. T. Horváth, Top. Catal., 2008, 48, 49–54.
7 A. S. Touchy, S. M. A. Hakim Siddiki, K. Kon and
K. Shimizu, ACS Catal., 2014, 4, 3045–3050.
total volume 100 mL, FA 60 mmol, LA/FA/1a: 1 : 1 : 1, tem-
perature 160 °C (for details, see the ESI†).
21 Reaction conditions for
a
maximum pressure of
8 (a) Y. Wei, C. Wang, X. Jiang, D. Xue, J. Lia and J. Xiao, Chem.
Commun., 2013, 49, 5408–5410; (b) Y. Wei, C. Wang, X. Jiang,
D. Xue, Z.-T. Liu and J. Xiao, Green Chem., 2014, 16, 1093–1096.
64 bar: total volume 100 mL, FA 120 mmol,
LA/FA/1a: 1 : 1 : 1, temperature 200 °C (for details, see
the ESI†).
Green Chem.
This journal is © The Royal Society of Chemistry 2015