Efficient, metal-free production of succinic acid by oxidation of biomass-derived levulinic acid with hydrogen peroxide

Article abstract of DOI:10.1039/c5gc00098j

A practical, scalable, metal-free synthesis of succinic acid from the biomass-derived platform chemical levulinic acid is described. Treatment of levulinic acid with the inexpensive, simple oxidant hydrogen peroxide under the catalytic action of trifluoroacetic acid gives succinic acid in high yield and enables facile product isolation by simple distillation of the volatile catalyst and byproducts.

Full text of DOI:10.1039/c5gc00098j

Green Chemistry
View Article Online
View Journal
COMMUNICATION
Efficient, metal-free production of succinic acid
by oxidation of biomass-derived levulinic acid with
hydrogen peroxide†
Cite this: DOI: 10.1039/c5gc00098j
Received 16th January 2015,
Accepted 5th February 2015
DOI: 10.1039/c5gc00098j
Saikat Dutta, Linglin Wu and Mark Mascal*
www.rsc.org/greenchem
A practical, scalable, metal-free synthesis of succinic acid from the derived from lignocellulose hydrolysates. Although generally
biomass-derived platform chemical levulinic acid is described. good yields and productivity have been reported, challenges
Treatment of levulinic acid with the inexpensive, simple oxidant associated with downstream processing, including selectivity
hydrogen peroxide under the catalytic action of trifluoroacetic acid issues, the use of bases as neutralizing agents, and product
9
gives succinic acid in high yield and enables facile product isolation isolation complicate the overall economics of the process.
by simple distillation of the volatile catalyst and byproducts.
In principle, chemical-catalytic pathways offer much faster
and more scalable routes to SA from carbohydrates. Although
no practical access to SA directly from raw biomass has yet
been developed, approaches via furfural and levulinic acid,
both one step removed from biomass, have been described.
Thus, Choudhary et al. recently reported the oxidation of fur-
Introduction
fural, a derivative of hemicellulose, with H
2 2
O at 80 °C over
2
4 hours to give SA in up to 74% yield. However, the SA was
Succinic acid (SA) is an organic chemical of major commercial
potential. Although the current market for SA is limited by its
comparatively high price, it has been proposed as a feedstock
for a variety of high-volume commodity chemicals, including
contaminated with a maleic acid by-product and the reaction
is dependent on an ultimately degradable catalyst (Amberlyst-
1
0
15). Beyond this, the cost of the feedstock and long reaction
period give little advantage over fermentative routes. Related
–3 methods involving furfural and other furans using a range of
oxidants and catalysts generally give SA in lower yields and
1
,4-butanediol (BDO), gamma-butyrolactone (GBL), maleic
1
anhydride (MA), and tetrahydrofuran (THF), among others.
The recent description of biodegradable polypropylene succi-
nate in the form of a stereocomplex with properties compar-
1
1,12
selectivities, and are described in reviews.
Levulinic acid (LA) is a renewable feedstock of exceptional
promise. Unlike furfural, LA can be derived both from hemi-
cellulose and the major, cellulosic fraction of carbohydrates. It
can be produced in high yield via the acidic processing of
able to LDPE may also stimulate new commercial
4
applications. SA is conventionally sourced via the C
4
stream
of the light naphtha raffinate of petroleum, usually by hydro-
genation of MA or maleic acid, or oxidation of BDO, although
1
3
5
biomass,
and although this is practiced commercially
there are other approaches. Recently, however, a number of
only on a limited scale at present, economic projections have
indicated that the production costs of LA could fall as low as
$
the hydrolysis of the biomass derived platform molecule
5-(chloromethyl)furfural (CMF). As such, the potential of LA
to unlock key renewable markets is vast.
Remarkably, the conversion of LA to SA was first described
in a paper by Tollens as early as 1879. Nitric acid was
employed as the oxidant, resulting in a mixture of organic
acids, including SA, albeit in low yield. The first report of the
action of hydrogen peroxide on LA was published in 1934,
which described a reaction at 60 °C in the presence of a cupric
salt catalyst, again giving a mixture of carboxylic acids but only
companies have begun producing SA via fermentative pathways
with the goal of becoming competitive with petrochemical
routes, such that the global demand for SA has been predicted
to increase from the current <100 kT to >700 kT per annum by
0.04–$0.10 lb−
1 14
.
LA can also be accessed in high yield by
1
5
6
,7
2
020, representing a ca. $1B market.
The biological route to succinic acid is centered around
1
6
native overproducer bacteria and genetically engineered
8
E. coli. Carbon sources are typically sugars, which may also be
1
7
Department of Chemistry, University of California Davis, 1 Shields Avenue, Davis,
CA 95616, USA. E-mail: mjmascal@ucdavis.edu
†
Electronic supplementary information (ESI) available: Description of materials
used, synthetic procedures, and NMR spectra for products appearing in this
work. See DOI: 10.1039/c5gc00098j
1
8
trace SA. A 1954 patent reported the gas phase oxidation of
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Communication
Green Chemistry
2
LA with O and a vanadium catalyst at 375 °C, wherein a
1
9
maximum yield of 83% was claimed. This approach might
have been of preparative interest were not the conditions so
severe.
The current emphasis on green chemical production has
led to a renewed interest in the conversion of LA to SA, and a
flurry of recent publications describing this reaction has
appeared. Thus, a 2012 patent describes the heating of LA with
2
nitric acid–NaNO at 40 °C for 4 h to give mixtures of SA and
2
0
oxalic acid, the former in up to 52% yield. Liu et al. reported
the application of a Mn(III) catalyst in the oxidation of methyl
2 2
Scheme 2 Oxidation of LA to acetic and formic acids with H O . RA =
retro aldol.
levulinate at 90 °C under 5 bar of O
2
to give a mixture of
dimethyl succinate, malonate, and oxalate esters, along with
2
1
related acetal derivatives. The maximum yield of succinate (50%), formic acid (24%), and methanol (17%) (estimated
1
was 52% in a 20 hour reaction. Podolean et al. employed Ru(III) yields by H NMR integration). Considering the reaction in
functionalized silica-coated magnetic nanoparticles under Scheme 1, methanol is an expected byproduct, and the strong
1
0 bar O at 150 °C for 6 hours in the conversion of LA to SA, oxidizing conditions also lead to the conversion of LA to acetic
2
where catalyst recycling experiments demonstrated good re- acid. This chemistry is precedented – in fact, a recent patent
usability (×3) at conversions of 54–58% and a 4% loading of uses LA as a feedstock for producing acetic acid with a range
2
2
24
ruthenium. Finally, an interesting reaction was reported by of oxidants, including H
2
O
2
2
.
Formic acid is also seen in
4
Caretto and Perosa that involved simple heating of LA in some cases as a byproduct. The observation of acetic acid
a dimethylcarbonate–base mixture at 200 °C for 4 h to give can be explained as shown in Scheme 2 by invoking the
dimethyl succinate among a range of other products in up to alternative migration product in the Baeyer–Villiger oxidation,
2
3
2
0% yield.
i.e. 3-hydroxypropanoic acid (HPA), as an intermediate. The
In all of the above cases, the conversion of LA to SA involves conversion of LA to HPA co-produces a molecule of acetic acid
either modest yields, poor selectivity, severe conditions, and/or on hydrolysis of the initially formed ester. HPA can then
potentially foulable catalysts. In the anticipation that LA is undergo a retro aldol cleavage to give acetic acid and form-
poised to become a low-cost feedstock with the potential to aldehyde, the latter ultimately being oxidized to formic acid.
supply value to range of chemical markets, we were motivated
While the acetic acid, formic acid, and methanol are vola-
to explore metal-free catalytic approaches to the oxidation of tile and can easily be removed from the reaction mixture, sep-
LA to valuable chemicals such as SA. Below, we report the aration of the SA product from aq. sulfuric acid is difficult,
efficient conversion of LA into SA using an inexpensive, simple and the recycle of sulfuric acid is costly. A remarkably simple
oxidant under gentle conditions.
solution to these issues presented itself in the form of trifluoro-
2
5
acetic acid (TFA) which, with pK
sufficiently acidic to catalyze the reaction. Thus, when a
mixture of LA and 30% aqueous H in TFA was heated at
0 °C, the starting material was consumed within 2 hours and
the result was a 62% yield of SA, alongside 43% acetic acid,
a
of ca. 0, was found to be
Results and discussion
O
2 2
9
At the outset, we had identified hydrogen peroxide as a
reagent of great and hitherto under-exploited potential for the
oxidation of LA due to its well known Baeyer–Villiger type
mechanism of action (Scheme 1).
1
4
5% formic acid, and 9% HPA (estimated yields by H NMR
integration). The initially formed monomethyl ester of SA
trans-esterifies with TFA to give methyl trifluoroacetate (45%),
which is captured in a cold trap. The volatile part of the reac-
tion mixture thus consists of TFA methyl ester (bp 43 °C), TFA
Initial experiments with H
2 2
O in aqueous sulfuric acid
showed promise, giving a mixture of SA (48%), acetic acid
(
bp 72 °C), the TFA–water azeotrope (79 wt% TFA, bp 105 °C)
and finally acetic acid (bp 118 °C). The residual, white solid
mass could be triturated with ether, in which SA is largely
insoluble, to give pure SA (60% isolated yield). A scaled up
reaction starting with 10.0 g of LA provided 6.0 g of SA (59%).
The triturate consists of HPA, a small amount of SA, and a
mixture of unidentified, minor products.
The management of the TFA would be an important aspect
of this process from an applied perspective. Taking the larger
scale reaction (processing 10.0 g LA feedstock) as an example,
the total 50 mL of 30% H O used is capable of delivering a
2
2
2
maximum of 48 g H O at the completion of the reaction.
Scheme 1 Oxidation of LA to SA with H
2
O
2
.
The 200 mL of TFA used corresponds to 298 g, of which 180 g
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Communication
will combine with the 48 g of H
2
O to form 228 g of the azeo-
2 C. Delhomme, D. Weuster-Botz and F. E. Kühn, Green
Chem., 2009, 11, 13.
3 I. Bechthold, K. Bretz, S. Kabasci, R. Kopitzky and
A. Springer, Chem. Eng. Technol., 2008, 31, 647.
4 (a) J. M. Longo, A. M. DiCiccio and G. W. Coates, J. Am.
Chem. Soc., 2014, 136, 15897; (b) S. K. Ritter, Chem. Eng.
News, 2014, 92(46), 5.
5 B. Cornils and P. Lappe, Ullmann’s Encyclopedia of Indus-
trial Chemistry, Electronic Release, Wiley-VCH, Weinheim,
2000, DOI: 10.1002/14356007.a08_523.pub2.
trope, with a remainder of 118 g TFA. While we used 30% aq.
in this work, 50% H is generally available, with indus-
H
2
O
2
2
O
2
trial concentrations up to 70%. For the 50% and 70% grades,
the delivery of less water with the same quantity of H O
would result in the formation of 121 g and 76 g of the azeo-
trope, respectively, from which only 25.5 and 16 g of water
would need to be removed. Recycling of TFA is accomplished
by dehydration of the azeotrope by membrane pervaporation.
We have performed the above reaction using 50% H
no significant variation in outcome.
2
2
2
6
2 2
O with
6 Global Bio Succinic Acid Market (Applications and
Geography) - Size, Share, Trends, Analysis, Research, Future
Demand, Scope and Forecast, 2013–2020, Allied Market
Research, 2014 (http://www.alliedmarketresearch.com).
Conclusions
7
WP 8.1. Determination of market potential for selected
platform chemicals. Itaconic acid, succinic acid, 2,5-furan-
dicarboxylic acid, 2012, Weastra, S.R.O. 2012 (http://www.
weastra.com).
In summary, we have described an efficient, one-step, non-fer-
mentative, metal-free process for the conversion of biomass-
derived levulinic acid into succinic acid under mild conditions
and in short reaction times. The mass balance of the reaction
consists mainly of acetic and formic acids, which are them-
selves useful commodity chemicals. TFA is an unfoulable
organic acid catalyst which can be fully recycled, involving the
minimal removal of co-produced water if industrial strength
solutions of H O are used. This simple, practical method
8
9
K. K. Cheng, X.-B. Zhao, J. Zeng and J.-A. Zhang, Biofuels,
Bioprod. Biorefin., 2012, 6, 302.
J. Yi, S. Choi, M.-S. Han, J. W. Lee and S. Y. Lee, Bio-
processing Technologies in Biorefinery for Sustainable Pro-
duction of Fuels, Chemicals, and Polymers, ed. S.-T. Yang,
H. A. El-Enshasy and N. Thongchul, John Wiley & Sons,
Inc., 1st edn, 2013, pp. 317–330.
2
2
paves the way for the commercial production of renewable SA
and its many useful derivatives.
1
1
0 H. Choudhary, S. Nishimura and K. Ebitani, Chem. Lett.,
012, 41, 409.
1 L. A. Badovskaya and L. V. Povarova, Chem. Heterocycl.
Compd., 2009, 45, 1023.
2
Experimental section
TFA-catalyzed oxidation of levulinic acid with hydrogen
peroxide
12 A. Takagaki, S. Nishimura and K. Ebitani, Catal. Surv. Asia,
2012, 16, 164.
1
3 D. J. Hayes, S. Fitzpatrick, M. H. B. Hayes and J. R. H. Ross,
The biofine process – production of Levulinic acid,
furfural, and formic acid from Lignocellulosic feedstocks,
in Biorefineries-Industrial Processes and Products: Status Quo
and Future Directions, ed. B. Kamm, P. R. Gruber and
M. Kamm, Wiley-VCH, Verlag GmbH, Weinheim,
Germany, 2005, DOI: 10.1002/9783527619849.ch7; See also:
B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Ind. Eng.
Chem. Res., 2007, 46, 1696.
Levulinic acid (2.00 g, 17.2 mmol) was dissolved in TFA
2 2
(40 mL) and 30% aq. H O (2.0 mL) was carefully added. The
flask was mounted with a water-cooled condenser and −78 °C
volatiles trap, and the colorless mixture was placed in an oil
bath at 90 °C and stirred for 20 min. Additional 30% aq. H
2 2
O
(8.0 mL) was added portionwise at a rate of 2 mL every 20 min.
The reaction was allowed to stir a further 20 min after the final
addition, at which point the LA had been completely con-
1
sumed as indicated by H NMR analysis. The mixture was
1
4 J. J. Bozell, L. Moens, D. C. Elliott, Y. Wang,
G. G. Neuenschwander, S. W. Fitzpatrick, R. J. Bilski and
J. L. Jarnefeld, Resour., Conserv. Recycl., 2000, 28, 227.
5 M. Mascal and E. B. Nikitin, Green Chem., 2010, 12, 370.
6 For example, cellulosic gasoline: M. Mascal, S. Dutta and
I. Gandarias, Angew. Chem., Int. Ed., 2014, 53, 1854.
7 B. Tollens, Chem. Ber., 1879, 12, 334.
cooled to room temperature and a measured amount of 1,4-
1
dioxane was added as an internal standard. The H NMR spec-
trum was measured and the yields were estimated as follows:
succinic acid (62%), acetic acid (43%), 3-hydroxypropanoic
acid (9%), and formic acid (45%). Methyl trifluoroacetate
1
1
(45%) was obtained in the cold trap. The volatiles were evapor-
1
1
ated to give a white solid (1.70 g) which was triturated with
8 A. P. Ponsford and I. Smedley-Maclean, Biochem. J., 1934,
Et O (2 × 2 mL) to give pure succinic acid (1.22 g, 60%). Experi-
2
28, 892.
mental details of the scale-up of this reaction and the sulfuric
acid-catalyzed oxidation of levulinic acid are given in the ESI.†
1
2
9 A. P. Dunlop and S. Smith, US Patent, 2676186, 1954.
0 D. S. Van Es, F. Van der Klis and J. Van Haveren, PCT WO,
2012/044168 A1, 2012.
2
2
1 J. Liu, Z. Du, T. Lu and J. Xu, ChemSusChem, 2013, 6, 2255.
2 I. Podolean, V. Kuncser, N. Gheorghe, D. Macovei,
V. I. Parvulescu and S. M. Coman, Green Chem., 2013, 15,
3077.
Notes and references
1
A. Cukalovic and C. V. Stevens, Biofuels, Bioprod. Biorefin.,
008, 2, 505.
2
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Communication
Green Chemistry
2
3 A. Caretto and A. Perosa, ACS Sustainable Chem. Eng., 2013, 26 Reviews: (a) S.-L. Wee, C.-T. Tye and S. Bhatia, Sep. Purif.
1
, 989.
Technol., 2008, 63, 500; (b) P. D. Chapman, T. Oliveira,
A. G. Livingston and K. Li, J. Membr. Sci., 2008, 318, 5;
(c) T. C. Bowen, R. D. Noble and J. L. Falconer, J. Membr.
Sci., 2004, 245, 1.
2
2
4 Y. Fu and J. Li, PCT Int. Appl., WO 2013159322 A1, 2013.
5 M. Namaziana, M. Zakery, M. R. Noorbalab and
M. L. Coote, Chem. Phys. Lett., 2008, 451, 163.
Green Chem.
This journal is © The Royal Society of Chemistry 2015