Catalyst-free transformation of levulinic acid into pyrrolidinones with formic acid

Article abstract of DOI:10.1039/c3gc42125b

Levulinic acid (LA) is transformed into pyrrolidinones by formic acid in DMSO without a catalyst. Mechanistic studies suggest the involvement of an iminium intermediate and a rate-limiting hydride transfer step.

Full text of DOI:10.1039/c3gc42125b

Green Chemistry
View Article Online
View Journal
COMMUNICATION
Catalyst-free transformation of levulinic acid into
pyrrolidinones with formic acid†
Cite this: DOI: 10.1039/c3gc42125b
Yawen Wei,a Chao Wang,*a Xue Jiang,a Dong Xue,a Zhao-Tie Liua and
Received 13th October 2013,
Accepted 9th November 2013
Jianliang Xiao*a,b
DOI: 10.1039/c3gc42125b
www.rsc.org/greenchem
Levulinic acid (LA) is transformed into pyrrolidinones by formic
acid in DMSO without a catalyst. Mechanistic studies suggest the
involvement of an iminium intermediate and
hydride transfer step.
a rate-limiting
The use of renewable biomass to replace the limited and non-
renewable fossil fuels has been recognised as an ideal solution
for problems faced with fossil fuels.1 Carbohydrates constitute
the major part of biomass raw materials.1c,2 LA is one of the
platform chemicals derived from carbohydrates,1p and it can
be transformed into fuels through γ-valerolactone (GVL).3
Transforming LA into value added fine chemicals would also
be desirable. One such example is the synthesis of pyrrolidi-
nones from LA by reductive amination (RA). Heterogeneous
catalytic hydrogenation was first explored for this reaction by
Shilling,4 Crook5 and their co-workers, and later by Manzer,6
Cao7 and co-workers. Fu and co-workers reported the first
homogeneous catalytic system for RA of LA to produce pyr-
rolidinones.8 Recently, a mild and efficient iridium-catalysed
RA of LA system was disclosed by our group.9 In all the
examples reported so far, however, precious metal based cata-
lysts have to be used to effect this transformation. Herein, we
present an example of RA of LA under catalyst-free conditions
with formic acid as the reducing agent under mild conditions
(Scheme 1).
Scheme 1 Catalyst-free transformation of LA into pyrrolidinones with
formic acid.
reduction reactions under catalyst-free conditions is rare,
however. One famous example is the Leuckart–Wallach reac-
tion discovered early in 1885.11 The Leuckart reaction normally
requires high temperature and has poor selectivity. Thus, the
development of a mild and selective catalyst-free system for
the RA of LA would be desirable.
During our effort to develop an iron-catalysed RA of LA with
formic acid, we serendipitously discovered that LA could react
with benzylamine to afford 1a in DMSO in the presence of
formic acid without a catalyst (Scheme 2). 72% conversion of
LA into 1a was obtained in DMSO in 4 h at 100 °C. The conver-
sion of LA rose to 89% with the addition of one equivalent of
triethylamine, which can balance the acidity of the system (ESI
Fig. S1†). It is noteworthy that the temperature required for
this conversion is much lower than the typical Leuckart reac-
tion (mostly above 180 °C). Several solvents were then screened
with 5 equivalents of formic acid and 1 equivalent of triethyl-
amine at 100 °C for 4 h (ESI Table S1†). The results suggest
Formic acid is a by-product formed during the acidic dehy-
dration of carbohydrates into LA. It is also a common hydro-
gen source for transfer hydrogenation reactions.1e Reductive
transformation of LA using formic acid as hydrogen source
would be an ideal process, with the emitted carbon dioxide
being reabsorbed by plants.1u,3b–d,f,h,7–10 As indicated, catalysts
were used in all of these systems. The use of formic acid for
aKey Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education,
Department of Chemistry & Chemical Engineering, Shaanxi Normal University,
Xi’an, 710062, China. E-mail: c.wang@snnu.edu.cn; Tel: +86-29-81530840
bDepartment of Chemistry, University of Liverpool, Liverpool, L69 7ZD, UK.
Scheme 2 Catalyst-free RA of LA. Reaction conditions: LA (1 mmol),
benzylamine (2 mmol), HCOOH (5 mmol), Et3N (1 mmol), DMSO (3 mL),
100 °C, 4 h, 89% conversion.
E-mail: j.xiao@liv.ac.uk
†Electronic supplementary information (ESI) available: Experimental details.
See DOI: 10.1039/c3gc42125b
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Communication
Green Chemistry
Table 2 RA of LA with aromatic aminesa
that the sulfinyl group is crucial for the reaction to proceed. In
tetramethylene sulfoxide, 63% conversion was observed;
however, all the other solvents, such as H2O, MeOH, DCM,
DMF, toluene, and CH3CN, afforded less than 25% of
conversion.
Various aliphatic amines were then reacted with LA to test
the generality of this protocol. The results are summarised
in Table 1.
Entry
R
Product
Yieldb (%)
Good to excellent isolated yields were obtained for benzylic
amines with differing substituents on the phenyl ring (Table 1,
entries 1–11). Activities were lower for 4-CF3, 3-F, and 3,4-Cl
substituted substrates (Table 1, entries 6, 8, 11). The non-
benzylic amine, phenylethylamine, also reacted well (Table 1,
entry 12). Varying the chain length of the aliphatic amines did
not affect the activities very much (Table 1, entries 13–17).
Branched aliphatic amines afforded similar yields (Table 1,
entries 18–20). A cyclic amine also worked well (Table 1, entry
21). Good yield was given by a chiral amine, albeit with a low
diastereoisomeric ratio (Table 1, entry 22).
1
2
3
4
5
6
7
8
4-OMe
4-Me
3,4-Me
3,4-OMe
4-NH2
3-OMe
4-H
4-F
4-Cl
4-Br
4-OCF3
2a
2b
2c
2d
2f
2e
2g
2h
2i
70
44
75
71
55
26
34
28
13
8
9
10
11
2j
2k
23
a Reaction conditions: LA (1 mmol), amine (3 mmol), HCOOH
(5 mmol), Et3N (1 mmol), DMSO (3 mL), 100 °C. b Isolated yield.
The success enjoyed by aliphatic amines could not be
shared with aromatic amines, however. Only 50% conversion
was obtained when using p-anisidine at the above optimal con-
ditions. By increasing the amount of amine to 3 equivalents
and adding p-anisidine and formic acid through 4 portions,
the conversion of LA could rise to 79% and a 70% isolated
yield of the product was obtained. However, acceptable yields
were only obtained for electron-rich aromatic amines (Table 2,
entries 1–5). Other aromatic amines tested including aniline
and 3-methoxylaniline all failed to give more than 50% yield.
In our previous study of the same reaction using an iridium
catalyst, evidence points to the reaction proceeding via the
hydrogenation of an in situ formed iminium ion followed by
spontaneous cyclisation.9 A similar pathway appears to be
operating for the transformation in question, except with the
reducing hydride coming from the formate instead of a metal
complex. Thus, when LA was reacted with benzylamine in the
presence of DCOOH, the 1H NMR of the isolated product
showed that the deuterium atom exclusively goes to the C5
carbon, supporting the involvement of an iminium intermedi-
ate (ESI Scheme S1†). In previous studies, a number of reduci-
ble intermediates were suggested for the Leuckart reaction,
including in particular imines and enamines.11c,12 The reac-
tion shown in Scheme S1† rules out the possibility of an
enamine intermediate.
In order to gain further insight into the reaction mechan-
ism, we studied the kinetic isotope effect by measuring the
initial rates of the reaction of LA with p-anisidine and 4-F-
aniline using HCOOH and DCOOH. We observed kH/kD = 2.9
for p-asinidine and 4.0 for 4-F-aniline, respectively, which
suggest that the transfer of hydride from formic acid to the
imine or iminium intermediate is involved in the rate-limiting
step (Scheme 3).13
Table 1 RA of LA with aliphatic aminesa
Entry
R
Product
Time (h)
Yieldb (%)
1
2
3
4
5
6
7
8
Benzyl
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
1q
1r
12
12
12
12
12
15
12
12
12
12
12
12
12
12
12
12
12
15
15
12
12
12
87
89
91
89
88
76
89
74
91
81
72
86
86
90
93
93
4-OMe-benzyl
4-Me-benzyl
4-F-benzyl
4-Cl-benzyl
4-CF3-benzyl
3-OMe-benzyl
3-F-benzyl
2-OMe-benzyl
2-F-benzyl
3,4-Cl-benzyl
1-Phenylethyl
1-Butyl
1-Pentyl
1-Hexyl
1-Octyl
1-Dodecyl
Isopentyl
9
10
11
12
13
14
15
16
17
18
19
20
21
22
The kinetic isotope data appear at odds with the results in
Table 2, however. In fact, a “Hammett plot” for different sub-
stituents para to the amino group of aromatic amines based
on the yields reveals a surprisingly linear relationship, with ρ
being −1.4 (Fig. 1),14 suggesting a positive charge, rather than
a negative one, being developed at the transition state of the
rate limiting step, which is inconsistent with hydride transfer
being involved. One explanation for these somehow conflicting
results is the rate-limiting hydride transfer step being preceded
88
77
87
91
3-Pentyl
1s
1t
1u
1v
2-Ethylhexyl
Cyclohexyl
(R)-1-Phenylethyl
85
88 (1.5 : 1)c
a Reaction conditions: LA (1 mmol), amine (2 mmol), HCOOH
(5 mmol), Et3N (1 mmol), DMSO (3 mL), 100 °C. b Isolated yield.
c Numbers in brackets refers to dr.
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Communication
K3 × K4 could render kobs larger with amines bearing more elec-
tron-donating substituents, leading to a negative ρ.
The reason why DMSO is so effective a solvent is unclear.
One possible explanation is found in its high solvent basicity
(0.647 vs. 0.613 for DMF, 0.286 for CH3CN, 0.128 for toluene
and 0.025 for water),15 which would be expected to promote
the nucleophilic attack by the amine at the ketone and by the
formate at the iminium ion. Another possible explanation is
Scheme 3 Kinetic isotope effect experiments for RA of LA. The reac-
tions were carried out in NMR tubes and the initial rates (based on <10%
conversions) were measured by 1H NMR with 1,3,5-trimethoxybenzene
as an internal standard.
+
from the abnormal behaviour of pKa values of RNH3 . The pKa
of a substance normally increases, when switching the solvent
from water to DMSO. However, this is not the case for
+ 16a
+
RNH3 .
RNH3 has smaller or similar pKa values in DMSO
+
than in water. For example, PhNH3 has a pKa of 4.6 in water
and 3.6 in DMSO and Et3NH+ has a pKa of 10.8 in water and
9.0 in DMSO.16 For formic acid, its pKa in water is 3.77 and it
is expected to increase in DMSO, as judged by that of acetic
acid, whose pKa increases from 4.76 to 12.6 on going from
water to DMSO.16 Thus the pKa of aromatic or aliphatic amine
salts could be similar or smaller than that of formic acid in
DMSO, ensuring that a large amount of free amines are
present in DMSO to participate in the imine formation step.
In conclusion, a catalyst-free system for RA of LA has
been developed. The solvent DMSO is critical for the high reac-
tivity observed. Mechanistic studies suggest that the rate-limit-
ing step for the reaction is hydride transfer from formic acid to
the iminium intermediate. The method described here is an
extension of the classic Leuckart–Wallach reaction and pro-
vides a practical, alternative way to convert LA to value-added
chemicals.
We are grateful for the financial support of the National
Science Foundation of China (21103102), Program for Chang-
jiang Scholars and Innovative Research Team in University
(IRT 1070), the Cheung Kong Scholar Programme, the Funda-
mental Research Funds for the Central Universities
(GK200902009), the Program for Key Science and Technology
Innovative Team of Shaanxi Province (2012KCT-21 and
2013KCT-17) and the Distinguished Doctoral Research Found
from Shaanxi Normal University (S2011YB02).
Fig. 1 “Hammett plot” for substituents para to the amino group of aro-
matic amines based on yields in 12 h.
by some complex equilibria as illustrated in Scheme 4. The
kinetic isotope effect mainly results from the rate difference
caused by using HCOOH vs. DCOOH in the hydride transfer
step defined by the rate constant k, whilst the “Hammett plot”
is to some degree a reflection of the overall rate constant kobs
,
which is likely to be a compound of k and the equilibrium con-
stants K1–K4. Although k is expected to increase with electron-
deficient amines, the equilibrium constants K3 × K4 may
become larger with more electro-rich ones. The increase in
References
1 (a) J. R. Rostrup-Nielsen, Science, 2005, 308, 1421–1422;
(b) G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006,
106, 4044–4098; (c) M. Schlaf, Dalton Trans., 2006, 4645–
4653; (d) J. N. Chheda, G. W. Huber and J. A. Dumesic,
Angew. Chem., Int. Ed., 2007, 46, 7164–7183; (e) A. Corma,
S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–2502;
(f) P. Gallezot, Catal. Today, 2007, 121, 76–91;
(g) L. D. Schmidt and P. J. Dauenhauer, Nature, 2007, 447,
914–915; (h) P. Gallezot, ChemSusChem, 2008, 1, 734–737;
(i) J. J. Bozell, Science, 2010, 329, 522–523; ( j) J. J. Bozell
and G. R. Petersen, Green Chem., 2010, 12, 539–554;
(k) U. Biermann, U. Bornscheuer, M. A. R. Meier,
J. O. Metzger and H. J. Schäfer, Angew. Chem., Int. Ed.,
2011, 50, 3854–3871; (l) D. J. Braden, C. A. Henao,
Scheme 4 Possible equilibria preceding the hydride transfer step,
which are likely to complicate the electronic effects of amines on the
overall reaction rate.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Communication
Green Chemistry
J. Heltzel, C. C. Maravelias and J. A. Dumesic, Green Chem.,
2011, 13, 1755–1765; (m) A. A. Rosatella, S. P. Simeonov,
R. F. M. Frade and C. A. M. Afonso, Green Chem., 2011, 13,
754–793; (n) S. Van de Vyver, J. Geboers, P. A. Jacobs and
B. F. Sels, ChemCatChem, 2011, 3, 82–94; (o) S. Chikkali
and S. Mecking, Angew. Chem., Int. Ed., 2012, 51, 5802–
5808; (p) P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538–1558;
(q) J. C. Serrano-Ruiz, A. Pineda, A. M. Balu, R. Luque,
J. M. Campelo, A. A. Romero and J. M. Ramos-Fernández,
Catal. Today, 2012, 195, 162–168; (r) G. W. Huber,
J. N. Chheda, C. J. Barrett and J. A. Dumesic, Science, 2005,
308, 1446–1450; (s) J. Q. Bond, D. M. Alonso, D. Wang,
R. M. West and J. A. Dumesic, Science, 2010, 327, 1110–
1114; (t) J. P. Lange, R. Price, P. M. Ayoub, J. Louis,
L. Petrus, L. Clarke and H. Gosselink, Angew. Chem., Int.
Ed., 2010, 49, 4479–4483; (u) T. Thananatthanachon and
Catal. Lett., 2012, 142, 779–787; (k) S. G. Wettstein,
J. Q. Bond, D. M. Alonso, H. N. Pham, A. K. Datye and
J. A. Dumesic, Appl. Catal., B, 2012, 117–118, 321–329.
4 W. L. Shilling, US Pat, 32355562, 1996.
5 L. R. Crook, B. A. Jansen, K. E. Spencer and D. H. Watson,
GB Pat, 1036694, 1996.
6 (a) L. E. Manzer, US Pat, 6743819, 2004; (b) L. E. Manzer
and F. E. Herkes, US Pat, 2004192933, 2003;
(c) L. E. Manzer, WO Pat, 2004084633, 2004.
7 X. L. Du, L. He, S. Zhao, Y. M. Liu, Y. Cao, H. Y. He
and K. N. Fan, Angew. Chem., Int. Ed., 2011, 50, 7815–
7819.
8 Y. B. Huang, J. J. Dai, X. J. Deng, Y. C. Qu, Q. X. Guo and
Y. Fu, ChemSusChem, 2011, 4, 1578–1581.
9 Y. W. Wei, C. Wang, X. Jiang, D. Xue, J. Li and J. L. Xiao,
Chem. Commun., 2013, 49, 5408–5410.
T. B. Rauchfuss, Angew. Chem., Int. Ed., 2010, 49, 6616– 10 T. Thananatthanachon and T. B. Rauchfuss, ChemSusChem,
6618; (v) H. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, 2010, 3, 1139–1141.
Science, 2007, 316, 1597–1600; (w) G. Yong, Y. Zhang and 11 (a) R. Leuckart, Ber. Dtsch. Chem. Ges., 1885, 18, 2341–2344;
J. Y. Ying, Angew. Chem., Int. Ed., 2008, 47, 9345–9348;
(x) J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009,
131, 1979–1985.
2 F. W. Lichtenthaler and S. Mondel, Pure Appl. Chem., 1997,
69, 1853–1866.
(b) O. Wallach, Ber. Dtsch. Chem. Ges., 1891, 24, 3992;
(c) H. W. Gibson, Chem. Rev., 1969, 69, 673–692;
(d) S. C. Lee and S. B. Park, Chem. Commun., 2007, 3714–
3716; (e) D. O’Connor, A. Lauria, S. P. Bondi and S. Saba,
Tetrahedron Lett., 2011, 52, 129–132.
3 (a) H. Mehdi, V. Fábos, R. Tuba, A. Bodor, L. Mika and 12 (a) V. J. Webers and W. F. Bruce, J. Am. Chem. Soc., 1948,
I. Horváth, Top. Catal., 2008, 48, 49–54; (b) L. Deng, J. Li,
D. M. Lai, Y. Fu and Q. X. Guo, Angew. Chem., Int. Ed., 2009,
48, 6529–6532; (c) H. Heeres, R. Handana, D. Chunai,
C. Borromeus Rasrendra, B. Girisuta and H. Jan Heeres,
Green Chem., 2009, 11, 1247–1255; (d) L. Deng, Y. Zhao,
J. Li, Y. Fu, B. Liao and Q. X. Guo, ChemSusChem, 2010, 3,
1172–1175; (e) F. M. A. Geilen, B. Engendahl, A. Harwardt,
W. Marquardt, J. Klankermayer and W. Leitner, Angew.
70, 1422–1424; (b) P. L. DeBenneville and J. H. Macartney,
J. Am. Chem. Soc., 1950, 72, 3073–3075; (c) N. J. Leonard
and R. R. Sauers, J. Am. Chem. Soc., 1957, 79, 6210–6214;
(d) A. García Martínez, E. Teso Vilar, A. García Fraile,
P. Martínez Ruiz, R. Macías San Antonio and
M. P. Martínez Alcázar, Tetrahedron: Asymmetry, 1999, 10,
1499–1505; (e) F. Barba, J. Recio and B. Batanero, Tetra-
hedron Lett., 2013, 54, 1835–1838.
Chem., Int. Ed., 2010, 49, 5510–5514; (f) D. Kopetzki and 13 P. I. Awachie and V. C. Agwada, Tetrahedron, 1990, 46,
M. Antonietti, Green Chem., 2010, 12, 656–660; 1899–1910.
(g) J. C. Serrano-Ruiz, D. Wang and J. A. Dumesic, Green 14 Hammett plot relates rate constants with substrate elec-
Chem., 2010, 12, 574–577; (h) X. L. Du, Q. Y. Bi, Y. M. Liu,
Y. Cao and K. N. Fan, ChemSusChem, 2011, 4, 1838–1843;
tronic properties. Product yields are used in Fig. 1 with the
assumption that they reflect the reaction rates.
(i) F. M. A. Geilen, B. Engendahl, M. Hölscher, 15 J. Catalan, J. Phys. Chem. B, 2009, 113, 5951–5960.
J. Klankermayer and W. Leitner, J. Am. Chem. Soc., 2011, 16 (a) F. G. Bordwell, Acc. Chem. Res., 1988, 21, 456–463;
133, 14349–14358; ( j) A. Hengne, N. Biradar and C. Rode,
(b) http://evans.harvard.edu/pdf/evans_pKa_table.pdf
Green Chem.
This journal is © The Royal Society of Chemistry 2014