Methanol Chemistry on Transition
Metal and Alloy Surfaces: Optimizing the Anode of Fuel
Cell Catalysts
Global demand for methanol in 1999
was nearly nine billion gallons. Among other applications,
methanol is used to produce formaldehyde, MTBE, and
acetic acid. A small - but steadily increasing
- niche for this chemical is in fuel supply to steam
reformers and to direct methanol fuel cells (DMFC's).
DMFC's hold out the promise of direct conversion of
methanol to electricity, a process that can be as much
as four times more efficient than combustion in internal
combustion engines. Before DMFC's will be commercially
viable, however, a number of technological problems
must be solved. Chief among these is the need
to develop an electrooxidation catalyst that will both
produce a high power density and be impervious to CO
poisoning over time. Platinum and platinum-ruthenium
catalysts have been found to be effective in this chemistry,
but improved catalysts are still needed. Theoretical
tools can assist in this process both by providing fundamental
understanding of methanol behavior on currently existing
catalysts and by suggesting new alloys that may lead
to improved catalyst performance.
We are currently performing a systematic
study of methanol partial oxidation on Pt(111).
We have fully analyzed the methanol decomposition pathway
that proceeds through hydroxyl O-H scission. To
see a movie of the proposed rate-limiting step
in this process (O-H bond scission in methanol to produce
the methoxy radical), together with the ensuing step
in the reaction pathway (C-H scission of methoxy), click
here.
We are in the process of evaluating
an alternative pathway for methanol decomposition on
Pt(111), the scission of a methanol C-H bond to produce
atomic hydrogen and hydroxymethylene. Preliminary
results indicate that this pathway could be energetically
competitive with O-H bond scission.
While methanol usage in fuel cells
is our primary area of study, we are also interested
in the methanol synthesis process. Industrially,
methanol is often produced over Cu/ZnO/alumina.
Although this process is quite valuable, much debate
remains over both the active phase of the catalyst and
the exact mechanism of methanol synthesis on these catalysts.
We have studied the partial oxidation of methanol (the
microscopic reverse process of methanol synthesis from
CO) on Cu(111) in an effort to gain fundamental insight
into this reaction. To
see a movie of the proposed rate-limiting step (C-H
bond scission in the methoxy radical), click here.
For more information on this project,
please refer to the following publications:
A First-Principles Study of Methanol
Decomposition on Pt(111), J. Greeley, M. Mavrikakis,
Journal of the American Chemical Society 124, 7193 (2002)
[ Journal
] [ Abstract
] [ HTML
Full Text ] [ PDF
]
Methanol Decomposition on Cu(111):
A DFT Study, J. Greeley, M. Mavrikakis, Journal of Catalysis,
208, 291 (2002)
[ Journal
] [ Abstract
]
Competitive Paths for Methanol Decomposition
on Pt(111), J. Greeley, M. Mavrikakis, Journal of the
American Chemical Society 126, 3910 (2004).
[ Journal
] [ Abstract
] [ PDF
]
Jeff Greeley
Research Assistant
Department of Chemical Engineering
University of Wisconsin at Madison
back
to top
|
Ab-Initio Study of
Methanol Synthesis from CO2 Hydrogenation
on Cu(111)
Methanol is a very import chemical
compound that is used in the synthesis of other chemicals,
such as Formaldehyde, Formic Acid, and MTBE. Although
the use of MTBE fades out in the near future there are
other applications emerging. Direct Methanol Fuel Cells
(DMFC) are an interesting alternative to H2
powered fuel cells and may become a major energy source.
Currently, Methanol is industrially
synthesized from CO/CO2/H2 mixtures
(syngas) at high pressures (40-120 bar) on Cu/ZnO/Al2O3
catalysts and temperatures between 200-300 °C. It
has been proposed that the main source of carbon in
methanol is CO2 and not CO. However, copper
is also a very good catalyst for the water gas shift
reaction, which is able to oxidize CO with H2O
to CO2. The goal of this project is to determine
a feasible reaction mechanism that explains the formation
of methanol from CO2. Therefore we use Density
Functional Theory (DFT) and investigate the thermodynamic
and kinetic properties of intermediate elementary steps
on the surface. From this data we can make conclusions
about the stability of reaction intermediates and kinetic
limitations of possible reaction pathways.
Lars Grabow
Research Assistant
Department of Chemical and Biological Engineering
University of Wisconsin-Madison
back
to top
|
Adsorption and Dissociation
of Dioxygen on Transition Metal and Alloy Surfaces:
Optimizing the Cathode of Fuel Cell Catalysts
Oxidation processes catalyzed
heterogeneously by transition metals have tremendous
importance in the chemical, energy, pharmaceutical,
and microelectronics industries. Examples range from
the production of such important chemicals as methanol
and ethylene epoxide to the removal of volatile organics
in industrial emission. Oxygen is usually the preferred
oxidant for its availability (air) and environmental
friendliness. Therefore, an understanding of the surface
behavior and surface chemistry of O2 is essential
to improving existing processes and designing more effective
catalysts. Furthermore, fuel cell technologies could
be a major beneficiary. The reduction of oxygen is the
primary reaction that takes place on the cathode of
many types of fuel cells, and the activation of O2
is most likely the Rate Limiting Step for the entire
fuel cell operation.
Of the many reaction possibilities
open to the O2 molecule, one that happens
perhaps most frequently is its dissociation into two
oxygen atoms - which are much more reactive than the
molecule itself. The dissociation of the oxygen molecule
on transition metal surfaces is therefore of considerable
research and practical interest. The dissociation of
O2 is known to take place via different mechanisms
on different metal surfaces. On some surfaces the oxygen
molecule directly dissociates upon adsorption. On others,
the molecule adsorbs initially intact into a precursor
state. Subsequent thermal equilibration with the surface
determines whether the O2 precursor desorbs
or dissociates. Molecularly adsorbed O2 species
have indeed been detected experimentally on various
transition metal surfaces at low temperature (e.g.,
Pt(111), Pd(111), Ni(111), Cu(111), Cu(110), Ag(110),
and Ir(111)). Their presence, even though elusive under
real reaction conditions, is shown to impart particular
characteristics to the dissociation of O2
on these surfaces.
We have used self-consistent, periodic
DFT calculations to study O and O2 adsorption
and O2 dissociation on several transition
metal surfaces, including Ir(111), Cu(111) and Cu(211),
and Au(111) and Au(211). Our goal is to ascertain the
geometry and energetics of possible precursors on these
surfaces, identify possible dissociation pathways, and
determine the dissociation activation energies. We are
currently studying the interaction of O and O2
with various alloys of Cu, Au, Pt and other elements.
The primary emphasis will be on the optimization of
the cathode catalyst for low-temperature PEM fuel cells.
For more information on this project,
please refer to the following publications:
Adsorption and dissociation of O2on
Pt-Co and Pt-Fe alloys, Y. Xu, A. Ruban, M. Mavrikakis,
Journal of the American Chemical Society 126, 4717 (2004).
[ Journal
] [ Abstract
] [ PDF
]
Adsorption and Dissociation of O2
on Gold Surfaces: Effect of Steps and Strain, Y. Xu
and M. Mavrikakis, Journal of Physical Chemistry B 107
(2003) 9298.
[ Journal
] [ Abstract
] [ PDF
]
The adsorption and dissociation of
O2 molecular precursors on Cu: The effect
of steps, Y. Xu and M. Mavrikakis, Surface Science,
538 (2003) 219.
[ Journal
] [ Abstract
] [ PDF
]
Universality in heterogeneous catalysis,
J.K. Nørskov, T. Bligaard, A. Logadottir, S.
Bahn, L.B. Hansen, M. Bollinger, H. Bengaard, B. Hammer,
Z. Sljivancanin, M. Mavrikakis, Y. Xu, S. Dahl, and
C.J.H. Jacobsen, Journal of Catalysis, 209 (2002) 275.
[ Journal
] [ Abstract
]
Adsorption and dissociation of O2
on Ir(111), Y. Xu and M. Mavrikakis, Journal of Chemical
Physics, 116 (2002) 10846.
[ Journal
] [ Abstract
] [ HTML
Full Text ] [ PDF
]
Adsorption and dissociation of O2
on Cu(111): Thermochemistry, reaction barrier and the
effect of strain, Y. Xu and M. Mavrikakis, Surface Science,
494 (2001) 131.
[ Abstract
] [ HTML
Full Text ] [ PDF
Full Text ] [ Journal
]
Ye Xu
Research Assistant
Department of Chemical Engineering
University of Wisconsin - Madison
back
to top
|
Surface Chemistry of
Gallium Nitride (GaN)
Group III-nitrides possess direct band
gaps ranging from 2.0-6.3eV and band gap engineering
could lead to devices emitting bright light covering
the entire spectrum from green to UV. However, until
a decade ago p-type doping of GaN could not be achieved
and efficiently working devices, which require p-n junctions,
could not be produced. This doping problem was overcome
so that the first highly efficient light emitting devices
based on GaN could be built and are commercially available
today. For example, many traffic lights use GaN.
The next major problem in the development
of high-quality GaN based devices concerns the material
quality. Sapphire is the most suitable substrate on
which growth of GaN can be performed. However, sapphire
has a lattice mismatch of 13% with respect to GaN. Therefore,
it is no surprise that attempts to grow GaN directly
on sapphire resulted in a huge number of defects, which
penetrate the GaN epilayer (threading defects). Various
techniques exist to overcome the defects introduced
by lattice mismatch. This includes lateral epitaxial
overgrowth (LEO) where a masking layer is included between
the seeding material and the grown films. Small windows
in the masking layer allows the alignment of the grown
film with the seeding layer. Dislocations are then confined
to the regions above windows, which improve the quality
of the film above the masked regions. Control of the
lateral:vertical growth ratio is important as a reduction
in the window size allows for the production of larger
high-quality regions.
Various dopants and surfactants can
be introduced to alter the lateral:vertical growth ratio
of LEO grown GaN. Studies using density functional theory
is ideal for insight into the mechanism by which these
alter the growth ratio as growth rates are directly
related to the surface chemistry. In connection with
Dr
T. F. Kuech's experimental efforts, we are currently
investigating the effect of Sb and Bi on the adsorption
and diffusuion of Ga and N on various facets of GaN,
so that we can improve the control of defect-free GaN
single crystals.
To
see a movie of the rotated isosurfaces of the electronic
density of Bi on a GaN (11 2 0) surface, click here.
For more information on this project,
please refer to the following publications:
Influence of Bi impurity as a surfactant
during the growth of GaN by metalorganic vapor phase
epitaxy, L. Zhang, H. F. Tang, J. Schieke, M. Mavrikakis,
and T. F. Kuech, Journal of Crystal Growth 242, 302
(2002)
[ Journal
] [ Abstract
] [ HTML
Full Text ] [ PDF
]
The addition of Sb as a surfactant
to GaN growth by metal organic vapor phase epitaxy,
with: L. Zhang, H. F. Tang, J. Schieke, M. Mavrikakis,
and T. F. Kuech, Journal of Applied Physics 92, 2304
(2002)
[ Journal
] [ Abstract
] [ HTML
Full Text ] [ PDF
]
Jaco Schieke
Research Assistant
Department of Chemical Engineering
University of Wisconsin - Madison
back
to top
|
Study of Active Sites for CO adsorption
on Methanol Synthesis Catalyst
The synthesis of methanol on Cu/ZnO
catalysts is a reaction widely performed in industry,
and it has been the subject of numerous research investigations
over the past several decades. Research has attempted
to discern both the detailed structure of the active
site of the methanol synthesis catalysts and the relationship
between this structure and the activity of the methanol
synthesis reaction.
Among several unsettled aspects of
the methanol synthesis mechanism, the nature of the
active intermediate (either CO or CO2) for
hydrogenation to methanol has been heavily disputed.
We are currently studying several aspects of MeOH synthesis
catalyst by using Density Functional Theory (DFT) Calculations.
Experimental studies have shown a downward shift of
50cm -1 of CO on the catalyst with respect to highly
reduced catalyst. Our aim is to explain this downward
shift of frequency and hence determine the active site
for CO adsorption.
Among a variety of models we are exploring
the possibilities of strain effects, effect of Cu or
and adspecies on the substrate Cu(111) surface. Similarly
we are also considering the effect of Cu and Zn being
alloyed together in bulk phase or on the surface alone.
It is interesting to see the modification of surface
properties of the surface by alloying. Alloying
also increases the number of distinct sites possible
on the surface of the catalyst. For example for the
Cu3Zn alloy which we are studying, there
are 11 different sites where CO may be adsorbed. The
geometry of Cu3Zn(111) as well as the sites
at which adsorption of CO may take place may be viewed
below.
 |
|
| The
geometry of Cu3Zn(111) (left) and the
sites at which adsorption of CO may take place
(right).
|
For more information on this project,
please refer to the following publications:
CO Vibrational Frequencies on Methanol
Synthesis Catalysts: a DFT study, J. Greeley, A. Gokhale,
J. Kreuser, H. Topsoe, N-Y. Topsoe, J. A. Dumesic, M.
Mavrikakis, Journal of Catalysis 213, 63 (2003)
[ Journal
] [ Abstract
] [ HTML
Full Text ] [ PDF
]
Jeffrey Greeley and Amit Gokhale
Research Assistants
Department of Chemical Engineering
University of Wisconsin - Madison
back
to top
|
Surface Chemistry of Copper
Tin alloys
Copper catalysts are widely used in
industry for a variety of Chemical Reactions like Water-Gas
Shift reaction, methanol synthesis and hydrogenation
reactions. Studies on Pt-Sn alloys have shown a drastic
change in the catalytic properties of Pt. Similar observations
have also been noted when Sn was added to Ru or Rh based
catalyst.
Carbon monoxide in general does not
adsorb on Sn. Thus the rate of coking, and hence
the deactivation of the catalyst, is likely to be drastically
reduced by the addition of Sn to it. Sn is also likely
to reduce the heat of adsorption of CO on the metal
it is alloyed with. Hence an alloy of Cu and tin would
seemingly work well for water gas shift reaction as
compared to Cu alone. With this motive, we are currently
studying the adsorption of different species like O,
H, OH, N, CO and common poisons like S and C on Cu6Sn5
using periodic self-consistent Density Functional Theory
(DFT) calculations.
Cu6Sn5 itself
has a NiAs type of structure. Hence a slab of CuSn may
be either Cu terminated or Sn terminated. These slabs
may be viewed below.
Cu terminated slab (left) and Sn
terminated slab (right).
In collaboration with Dr J. A. Dumesic
we are currently pursuing the characterization of of
Cu-Sn alloy catalysts with a variety of techniques including
Mössbauer and XRD. Our goal is to use experiment
and theory to correlate electronic structure with alloy
composition, and design the next generation of improved
catalysts for a variety of important applications.
For more information on this project,
please refer to the following publication:
Effect of Sn on the Reactivity of Cu
Surfaces; A. A. Gokhale, G. W. Huber, J. A. Dumesic,
M. Mavrikakis, Journal of Physical Chemistry B, 108,
14062 (2004)
[ Journal
] [ Abstract
] [ PDF
]
Amit Gokhale
Research Assistant
Department of Chemical Engineering
University of Wisconsin - Madison
back
to top
|
Low Temperature Water
Gas Shift Reaction
The water gas shift reaction (WGSR),
CO + H2O 
CO2 + H2 is an industrially important
route to H2 production and plays an important
role in many current technologies such as methanol synthesis,
methanol steam reforming, ammonia synthesis, coal gasification,
as well as fuel cell technology. Using quantum chemical
methods such as Density Functional Theory (DFT), I have
studied this reaction extensively on a series of ten
late transition metals. This study has helped us to
determine the characteristics that define a good WGS
catalyst, and hence allowed us to screen for potentially
active WGS catalysts.
Since Cu-based catalysts are used in
industry for WGSR I have calculated detailed minimum
energy paths for all the elementary steps on a Cu(111)
surface using DFT and built a robust microkinetic model
for WGSR on copper catalysts using the DFT-generated
parameters. This model conclusively proves that the
carboxyl mechanism is the dominant mechanism on this
surface and points out that the formate species is essentially
a spectator on the catalytic surface under typical WGSR
conditions.
Amit Gokhale
Research Assistant
Department of Chemical Engineering
University of Wisconsin - Madison
back
to top
|
Low Temperature Water
Gas Shift Reaction on Gold
Water gas shift (WGS) is the reaction
of water and carbon monoxide to produce hydrogen and
carbon dioxide (CO + H2O 
CO2 + H2), and is an important
step in the production of H2 via the steam
reforming of hydrocarbons. Hydrogen can be used as a
fuel in fuel cells for power generation in a variety
of applications and in the production of ammonia. WGS
also plays a role in the overall chemistry of hydrocarbon
oxidation reactions and auto exhaust combustion catalysis.
WGS is an exothermic reaction ( H
= -40.6 KJ/mol). Depending on conditions, various catalysts
are used to accelerate this reaction. For example, the
high-temperature shift reaction is performed at 590-720
K using a catalyst based on iron-oxide while the low-temperature
shift reaction is performed at 470-520K using a (Cu,
Zn, Al)-based catalyst. Recent work has suggested that
supported gold catalysts are very active in the WGS
reaction at low temperatures; even more than the CuO/ZnO
catalyst currently used industrially. Along with their
promising technical performance, the relatively low
and stable price of gold compared with the cost of platinum
group metals adds to the incentives for exploring gold
catalyst technologies.
Various mechanisms have been proposed
for this reaction, the most popular one being the redox
mechanism. We are using periodic, self-consistent density
functional theory (DFT) calculations to study the reaction
mechanism of WGS on Au single crystal surfaces. The
thermochemistry of relevant reaction intermediates and
the activation barriers of the various elementary steps
of the proposed alternative WGS reaction mechanisms
are determined and analyzed. Furthermore, the vibrational
frequencies of the reaction intermediates and of the
transition states are calculated which could then be
used to calculate the pre-exponential factors for each
elementary step. All this first principles derived information
is used to build a micro-kinetic model for bridging
the pressure and temperature gap and develop improved
WGS catalysts.
Using state-of-the-art theoretical
tools, we hope to gain insights with respect to several
issues related to the enhanced activity of WGS reaction
over Au catalysts that still remain controversial. Key
mechanistic aspects of WGS are examined using appropriate
experimental methods.
Shampa Kandoi
Research Assistant
Department of Chemical and Biological Engineering
University of Wisconsin-Madison
back
to top
|
Preferential Oxidation of CO
in the Presence of Hydrogen
As the energy sector moves towards
hydrogen, hydrogen fuelled proton exchange membrane
(PEM) fuel cells have been recognized as a promising
energy technology for replacement of internal combustion
engines in automobiles and other transportation systems.
However, an important problem for the usage of PEM fuel
cells is their sensitivity to low levels (ppm) of CO
because CO easily poisons PEM fuel cell anode. A promising
method to remove trace amounts of CO from H2
supplied to the anode is preferential oxidation of CO
(PROX) in the presence of excess H2 .
An effective PROX catalyst should meet
several important requirements: (i) show high activity
for CO oxidation (ii) oxidize CO much easier than hydrogen
(iii) function at temperatures similar to that of PEM
fuel cells and (iv) be resistant towards deactivation
by carbon di-oxide and water present in the feed. A
good PROX catalyst should be able to selectively oxidize
10,000 ppm of CO to concentrations of less than 5 ppm
at low temperatures, without decreasing the H2
content of the fuel gas.
The PROX reaction has been studied
experimentally on supported monometallic as well as
bimetallic catalysts extensively. Platinum, gold and
copper based catalysts, like Pt/-Al2O3
, Pt/A-zeolite, Au/-Fe2O3 , Cu
mixed with ceria oxide have been shown to be promising
PROX catalysts. In this study we have used a combination
of DFT and microkinetic modeling to explain the superior
low-temperature performance of Au and Cu catalysts,
as compared to Pt catalysts. We have employed periodic,
self-consistent density functional theory (DFT) calculations
to address competitive CO oxidation (CO + O
CO2) and hydrogen oxidation (H + O
OH, H + OH
H2O, OH + OH
H2O + O) on Au(111), Cu(111) and Pt(111).
The geometry and thermochemistry of various reactive
intermediates are studied, as well as reaction pathways
and activation energies for the corresponding elementary
steps. A rigorous barrier analysis scheme is implemented
for the appropriate decomposition of the activation
energy barrier of each elementary step into its major
components. Finally, using the DFT-derived parameters
for surface reaction steps and handbook values of gas
phase reaction thermochemistry, we construct a simple
micro-kinetic model to predict trends in PROX selectivity
on Au, Cu and Pt surfaces as a function of temperature.
For more information on this project,
please refer to the following publication:
Why Au and Cu are More Selective than
Pt for Preferential Oxidation of CO at low Temperature;
S.Kandoi, A.A. Gokhale, L.C. Grabow, J.A. Dumesic, and
M. Mavrikakis, Catalysis Letters 93, 93 (2004).
[ Journal
] [ Abstract
]
Shampa Kandoi, Amit Gokhale and Lars Grabow
Research Assistants
Department of Chemical and Biological Engineering
University of Wisconsin-Madison
back
to top
|
Fischer Tropsch Synthesis
Fischer Tropsch Synthesis (FTS) affords
a direct route to produce clean hydrocarbon liquid fuel
from syngas (a mixture of CO and H2). Traditionally
coal or natural gas have been used as feedstocks but
there is growing promise that biomass derived H2
could be utilized. Co and Fe are the conventional FTS
catalysts but it has been recently found that Pt promoters
significantly improve the performance of Fe catalysts.
We seek to explain this Pt-promotion using ab initio
techniques based on Density Functional Theory (DFT)
and microkinetic modeling. A comprehensive model would
lead to a better understanding of this important industrial
process especially in the light of similar models for
pure Co and Fe which have been previously developed
in our group.
Collaborators: Prof. C.
H. Bartholomew
(Brigham Young University)
Rahul Nabar
Research Assistant
Department of Chemical and Biological Engineering
University of Wisconsin-Madison
|
|
