(58e) Hierarchical Nickel Incorporated USY Zeolites for Hydrodeoxygenation of Lignin-Derived Pyrolysis Oil Model Compounds | AIChE

(58e) Hierarchical Nickel Incorporated USY Zeolites for Hydrodeoxygenation of Lignin-Derived Pyrolysis Oil Model Compounds

Authors 

Gamliel, D. P. - Presenter, University of Connecticut
Valla, J. A., University of Connecticut

Pyrolysis
of biomass and agricultural wastes is a readily achieved method for production of
sustainable fuels and platform chemicals. While pyrolysis oil
yields of up to 75 wt.% of the original biomass have been demonstrated,[1]
raw pyrolysis oil is not suitable for combustion fuel applications due to poor
stability and low heating value. Hydrodeoxygenation (HDO) of the
oil fraction is a common method for removal of pyrolysis oil oxygen, which is
eliminated in the form of water, thus preserving atomic carbon in the pyrolysis
oil. Traditional HDO catalysts include NiMo, CoMo and precious metals, however
the former require sulfidation and the latter are costly. Here, we propose
Ultra-stable Y (USY) zeolite supported Ni catalysts for liquid-phase HDO of
anisole, a common product formed from lignin pyrolysis.

Traditional
zeolite catalysts are highly microporous, and as a result may limit diffusion
of reactants and products to or from catalyst acid sites. Moreover, the low
external surface area may result in increased micropore blockage after Ni
impregnation. A variety of mesoporous USY zeolites were created using top-down alkaline
desilication (DS), desilication with a pore-directing agent (DS-PDA)[2]
and surfactant-mediated desilication (SA)[3].
As shown in Figure 1(A), DS formed a broad pore-size distribution with very
little micropore volume, whereas DS-PDA preserved some microporosity and
produced a narrower range of mesopores. The SA method preserved about 90% of
the micropore volume and produced a narrow range of mesopore size at about 30
Ǻ. Material acidity and crystallinity decreased after all treatments, and
a clear correlation between material crystallinity, acidity and micropore
volume was observed. Figure 1(B) shows a visualization of the types of pores
formed from each treatment. Alkaline treatment strength and PDA concentration
were varied in order to determine the effect of each variable on zeolite
properties. 

Ni
was impregnated by the incipient wetness method, and NiO crystal size was
approximately 30 nm, determined by application of the Scherrer equation to
X-ray diffraction (XRD) patterns. After reduction, the Ni crystal size
decreased significantly on the more microporous materials, but remained high on
the larger-pore materials. Larger crystal sizes and greater interaction with
the DS and DS-PDA zeolites was observed from temperature programmed reduction
(TPR).

Each
material was evaluated as a catalyst for the liquid-phase HDO of anisole at 200
°C and 750 psi. Overall reaction rate increased in the order
Ni-SA-USY<Ni-USY<Ni-DS-USY<Ni-DS-USY-PDA, shown in Figure 1(C). A
simplified kinetic model of the anisole reaction pathways was proposed, and
kinetic parameters were fit to the data and normalized to the number of Ni and
acid sites. Analysis of the fitted kinetic parameters showed that the larger
mesoporous zeolites enhanced access of reactant anisole to the Ni sites for
more efficient hydrogenation of the aromatic ring, whereas acid-catalyzed
deoxygenation of the hydrogenated intermediate was about an order of magnitude
faster. The Ni-DS-USY-PDA was the most effective catalyst overall, as
hydrogenation of the aromatic ring was shown to be the rate determining step in
this reaction system.       


Figure
1.
(A)
BJH pore size distribution of mesoporous USY materials. (B) Visualization of
each mesoporous USY material, white space indicates pore area. (C) Reaction
rates for anisole hydrodeoxygenation (bars correspond to left axis), in
addition to final anisole conversion and mass balance (corresponding to right
axis).

 

 

 

References

[1]      O. Onay,
Fuel 86 (2007) 1452–1460.

[2]      D.
Verboekend, G. Vile, J. Perez-Ramirez, Cryst. Growth Des. 12 (2012) 3123–3132.

[3]      J.
García-Martínez, M. Johnson, J. Valla, K. Li, J.Y. Ying, Catal. Sci. Technol. 2
(2012) 987.