268591 Production of Levulinic Acid From Cellulose by Hydrothermal Decomposition Combined with Aqueous Phase Dehydration with a Solid Acid Catalyst

Wednesday, October 31, 2012: 1:10 PM
315 (Convention Center )
Ronen Weingarten, Yong Tae Kim, Geoffrey A. Tompsett, Wm. Curtis Conner Jr. and George W. Huber, Department of Chemical Engineering, University of Massachusetts Amherst, Amherst, MA

Levulinic acid is a versatile renewable platform molecule which can be used as the basis for the production of various high-volume chemicals and fuels with numerous potential industrial applications.1, 2  The increased relevance of levulinic acid is in part due to its potential capability to serve as a biobased chemical intermediate to produce fuels via conventional petrochemical technology.3, 4  Today commercially, levulinic acid is produced with homogeneous acid catalysts, such as sulfuric acid.5, 6  The sulfuric acid increases the cost of producing levulinic acid because it is expensive to purchase, has to be neutralized and is difficult to separate from levulinic acid.  This study introduces a new process to produce levulinic acid from cellulose at high concentrations without the use of a homogeneous acid catalyst.  The process consists of 2 reaction steps: (1) non-catalytic hydrothermal decomposition of cellulose at moderate temperatures (190-270 °C) to produce organic water-soluble compounds including glucose and HMF; (2) water-soluble compounds are further reacted with a solid acid catalyst at relatively low temperatures (160 °C) to produce levulinic acid and formic acid.  Unreacted cellulose can be recycled back to the first reactor for further decomposition.  The cellulose hydrothermally decomposes at high initial cellulose concentrations of 29 wt% while maintaining high selectivity towards water-soluble compounds, which are levulinic acid precursors. The final result is a fairly pure aqueous stream of levulinic acid and formic acid at relatively high concentrations.  The maximum obtainable yield of levulinic acid in this process is 28% of the theoretical.

Amberlyst 70 was chosen as a solid acid catalyst for the conversion of water soluble organics into HMF, levulinic acid and formic acid in the second step of this process.  In this respect a key effort must be placed towards the design of improved hydrothermally stable solid acid catalysts displaying high activity and selectivity to HMF and levulinic acid comparable to their homogeneous counterparts.  Accordingly, a series of well characterized solid acid metal phosphate catalysts have been prepared for the two step dehydration/rehydration reaction to produce levulinic acid from glucose.  The catalysts include three zirconium phosphates and two tin phosphates with different ratios of metal to phosphorous. The total amount of acid sites and the concentration of Brønsted sites in these catalysts have been characterized with TPD using gas phase NH3 and isopropylamine respectively.  The results obtained for the catalytic activity and selectivity have been compared with that of homogeneous catalysts, including HCl as a Brønsted acid and Yb(OTf)3 as a Lewis acid.  The catalyst selectivity is a function of the Brønsted to Lewis acid site ratio for both the heterogeneous and homogeneous reactions.  Brønsted acid sites increase levulinic acid selectivity while Lewis acid sites increase catalytic activity but lead to a higher formation of degradation products (humins).  Overall, this comprehensive study lays the grounds for further optimization to produce levulinic acid from cellulose without using homogeneous acid catalysts.


  1. J. J. Bozell, L. Moens, D. C. Elliott, Y. Wang, G. G. Neuenscwander, S. W. Fitzpatrick, R. J. Bilski and J. L. Jarnefeld, Resources, Conservation and Recycling, 2000, 28, 227-239.
  2. P. Gullón, A. Romaní, C. Vila, G. Garrote and J. C. Parajó, Biofuels, Bioproducts and Biorefining, 2011.
  3. J.-P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus, L. Clarke and H. Gosselink, Angewandte Chemie International Edition, 2010, 49, 4479-4483.
  4. J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and J. A. Dumesic, Science, 2010, 327, 1110-1114.
  5. US Pat., 4897497, 1990.
  6. US Pat., 5608105, 1997.

Extended Abstract: File Not Uploaded